+
+unsigned int crc32(unsigned char const *p, unsigned int len)
+{
+ int i;
+ unsigned int crc = 0;
+ while (len--) {
+ crc ^= *p++;
+ for (i = 0; i < 8; i++)
+ crc = (crc >> 1) ^ ((crc & 1) ? 0xedb88320 : 0);
+ }
+ return crc;
+}
+
+int main(int argc, char **argv) {
+ unsigned int result;
+ if (argc != 2) {
+ printf("no string passed as argument\n");
+ return -1;
+ }
+ result = crc32(argv[1], strlen(argv[1]));
+ printf("0x%x\n", result);
+ return 0;
+}
diff --git a/Documentation/pcmcia/driver-changes.txt b/Documentation/pcmcia/driver-changes.txt
new file mode 100644
index 000000000000..9c315ab48a02
--- /dev/null
+++ b/Documentation/pcmcia/driver-changes.txt
@@ -0,0 +1,51 @@
+This file details changes in 2.6 which affect PCMCIA card driver authors:
+
+* in-kernel device<->driver matching
+ PCMCIA devices and their correct drivers can now be matched in
+ kernelspace. See 'devicetable.txt' for details.
+
+* Device model integration (as of 2.6.11)
+ A struct pcmcia_device is registered with the device model core,
+ and can be used (e.g. for SET_NETDEV_DEV) by using
+ handle_to_dev(client_handle_t * handle).
+
+* Convert internal I/O port addresses to unsigned long (as of 2.6.11)
+ ioaddr_t should be replaced by kio_addr_t in PCMCIA card drivers.
+
+* irq_mask and irq_list parameters (as of 2.6.11)
+ The irq_mask and irq_list parameters should no longer be used in
+ PCMCIA card drivers. Instead, it is the job of the PCMCIA core to
+ determine which IRQ should be used. Therefore, link->irq.IRQInfo2
+ is ignored.
+
+* client->PendingEvents is gone (as of 2.6.11)
+ client->PendingEvents is no longer available.
+
+* client->Attributes are gone (as of 2.6.11)
+ client->Attributes is unused, therefore it is removed from all
+ PCMCIA card drivers
+
+* core functions no longer available (as of 2.6.11)
+ The following functions have been removed from the kernel source
+ because they are unused by all in-kernel drivers, and no external
+ driver was reported to rely on them:
+ pcmcia_get_first_region()
+ pcmcia_get_next_region()
+ pcmcia_modify_window()
+ pcmcia_set_event_mask()
+ pcmcia_get_first_window()
+ pcmcia_get_next_window()
+
+* device list iteration upon module removal (as of 2.6.10)
+ It is no longer necessary to iterate on the driver's internal
+ client list and call the ->detach() function upon module removal.
+
+* Resource management. (as of 2.6.8)
+ Although the PCMCIA subsystem will allocate resources for cards,
+ it no longer marks these resources busy. This means that driver
+ authors are now responsible for claiming your resources as per
+ other drivers in Linux. You should use request_region() to mark
+ your IO regions in-use, and request_mem_region() to mark your
+ memory regions in-use. The name argument should be a pointer to
+ your driver name. Eg, for pcnet_cs, name should point to the
+ string "pcnet_cs".
diff --git a/Documentation/serial/driver b/Documentation/serial/driver
index e9c0178cd202..ac7eabbf662a 100644
--- a/Documentation/serial/driver
+++ b/Documentation/serial/driver
@@ -107,8 +107,8 @@ hardware.
indicate that the signal is permanently active. If RI is
not available, the signal should not be indicated as active.
- Locking: none.
- Interrupts: caller dependent.
+ Locking: port->lock taken.
+ Interrupts: locally disabled.
This call must not sleep
stop_tx(port,tty_stop)
diff --git a/Documentation/video4linux/API.html b/Documentation/video4linux/API.html
index 4b3d8f640a4a..441407b12a9f 100644
--- a/Documentation/video4linux/API.html
+++ b/Documentation/video4linux/API.html
@@ -1,399 +1,16 @@
-
-Video4Linux Kernel API Reference v0.1:19990430
-
-
-
-
-
-Devices
-Video4Linux provides the following sets of device files. These live on the
-character device formerly known as "/dev/bttv". /dev/bttv should be a
-symlink to /dev/video0 for most people.
-
-
-| Device Name | Minor Range | Function |
-
|---|
| /dev/video | 0-63 | Video Capture Interface |
-
| /dev/radio | 64-127 | AM/FM Radio Devices |
-
| /dev/vtx | 192-223 | Teletext Interface Chips |
-
| /dev/vbi | 224-239 | Raw VBI Data (Intercast/teletext) |
-
-
-Video4Linux programs open and scan the devices to find what they are looking
-for. Capability queries define what each interface supports. The
-described API is only defined for video capture cards. The relevant subset
-applies to radio cards. Teletext interfaces talk the existing VTX API.
-
-
Capability Query Ioctl
-The VIDIOCGCAP ioctl call is used to obtain the capability
-information for a video device. The struct video_capability object
-passed to the ioctl is completed and returned. It contains the following
-information
-
-
-| name[32] | Canonical name for this interface |
-
| type | Type of interface |
-
| channels | Number of radio/tv channels if appropriate |
-
| audios | Number of audio devices if appropriate |
-
| maxwidth | Maximum capture width in pixels |
-
| maxheight | Maximum capture height in pixels |
-
| minwidth | Minimum capture width in pixels |
-
| minheight | Minimum capture height in pixels |
-
-
-The type field lists the capability flags for the device. These are
-as follows
-
-
-| Name | Description |
-
|---|
| VID_TYPE_CAPTURE | Can capture to memory |
-
| VID_TYPE_TUNER | Has a tuner of some form |
-
| VID_TYPE_TELETEXT | Has teletext capability |
-
| VID_TYPE_OVERLAY | Can overlay its image onto the frame buffer |
-
| VID_TYPE_CHROMAKEY | Overlay is Chromakeyed |
-
| VID_TYPE_CLIPPING | Overlay clipping is supported |
-
| VID_TYPE_FRAMERAM | Overlay overwrites frame buffer memory |
-
| VID_TYPE_SCALES | The hardware supports image scaling |
-
| VID_TYPE_MONOCHROME | Image capture is grey scale only |
-
| VID_TYPE_SUBCAPTURE | Capture can be of only part of the image |
-
-
-The minimum and maximum sizes listed for a capture device do not imply all
-that all height/width ratios or sizes within the range are possible. A
-request to set a size will be honoured by the largest available capture
-size whose capture is no large than the requested rectangle in either
-direction. For example the quickcam has 3 fixed settings.
-
-
Frame Buffer
-Capture cards that drop data directly onto the frame buffer must be told the
-base address of the frame buffer, its size and organisation. This is a
-privileged ioctl and one that eventually X itself should set.
-
-The VIDIOCSFBUF ioctl sets the frame buffer parameters for a capture
-card. If the card does not do direct writes to the frame buffer then this
-ioctl will be unsupported. The VIDIOCGFBUF ioctl returns the
-currently used parameters. The structure used in both cases is a
-struct video_buffer.
-
-
-| void *base | Base physical address of the buffer |
-
| int height | Height of the frame buffer |
-
| int width | Width of the frame buffer |
-
| int depth | Depth of the frame buffer |
-
| int bytesperline | Number of bytes of memory between the start of two adjacent lines |
-
-
-Note that these values reflect the physical layout of the frame buffer.
-The visible area may be smaller. In fact under XFree86 this is commonly the
-case. XFree86 DGA can provide the parameters required to set up this ioctl.
-Setting the base address to NULL indicates there is no physical frame buffer
-access.
-
-
Capture Windows
-The capture area is described by a struct video_window. This defines
-a capture area and the clipping information if relevant. The
-VIDIOCGWIN ioctl recovers the current settings and the
-VIDIOCSWIN sets new values. A successful call to VIDIOCSWIN
-indicates that a suitable set of parameters have been chosen. They do not
-indicate that exactly what was requested was granted. The program should
-call VIDIOCGWIN to check if the nearest match was suitable. The
-struct video_window contains the following fields.
-
-
-| x | The X co-ordinate specified in X windows format. |
-
| y | The Y co-ordinate specified in X windows format. |
-
| width | The width of the image capture. |
-
| height | The height of the image capture. |
-
| chromakey | A host order RGB32 value for the chroma key. |
-
| flags | Additional capture flags. |
-
| clips | A list of clipping rectangles. (Set only) |
-
| clipcount | The number of clipping rectangles. (Set only) |
-
-
-Clipping rectangles are passed as an array. Each clip consists of the following
-fields available to the user.
-
-
-| x | X co-ordinate of rectangle to skip |
-
| y | Y co-ordinate of rectangle to skip |
-
| width | Width of rectangle to skip |
-
| height | Height of rectangle to skip |
-
-
-Merely setting the window does not enable capturing. Overlay capturing
-(i.e. PCI-PCI transfer to the frame buffer of the video card)
-is activated by passing the VIDIOCCAPTURE ioctl a value of 1, and
-disabled by passing it a value of 0.
-
-Some capture devices can capture a subfield of the image they actually see.
-This is indicated when VIDEO_TYPE_SUBCAPTURE is defined.
-The video_capture describes the time and special subfields to capture.
-The video_capture structure contains the following fields.
-
-
-| x | X co-ordinate of source rectangle to grab |
-
| y | Y co-ordinate of source rectangle to grab |
-
| width | Width of source rectangle to grab |
-
| height | Height of source rectangle to grab |
-
| decimation | Decimation to apply |
-
| flags | Flag settings for grabbing |
-
-The available flags are
-
-
-| Name | Description |
-
|---|
| VIDEO_CAPTURE_ODD | Capture only odd frames |
-
| VIDEO_CAPTURE_EVEN | Capture only even frames |
-
-
-
Video Sources
-Each video4linux video or audio device captures from one or more
-source channels. Each channel can be queries with the
-VDIOCGCHAN ioctl call. Before invoking this function the caller
-must set the channel field to the channel that is being queried. On return
-the struct video_channel is filled in with information about the
-nature of the channel itself.
-
-The VIDIOCSCHAN ioctl takes an integer argument and switches the
-capture to this input. It is not defined whether parameters such as colour
-settings or tuning are maintained across a channel switch. The caller should
-maintain settings as desired for each channel. (This is reasonable as
-different video inputs may have different properties).
-
-The struct video_channel consists of the following
-
-
-| channel | The channel number |
-
| name | The input name - preferably reflecting the label
-on the card input itself |
-
| tuners | Number of tuners for this input |
-
| flags | Properties the tuner has |
-
| type | Input type (if known) |
-
| norm | The norm for this channel |
-
-
-The flags defined are
-
-
-| VIDEO_VC_TUNER | Channel has tuners. |
-
| VIDEO_VC_AUDIO | Channel has audio. |
-
| VIDEO_VC_NORM | Channel has norm setting. |
-
-
-The types defined are
-
-
-| VIDEO_TYPE_TV | The input is a TV input. |
-
| VIDEO_TYPE_CAMERA | The input is a camera. |
-
-
-
Image Properties
-The image properties of the picture can be queried with the VIDIOCGPICT
-ioctl which fills in a struct video_picture. The VIDIOCSPICT
-ioctl allows values to be changed. All values except for the palette type
-are scaled between 0-65535.
-
-The struct video_picture consists of the following fields
-
-
-| brightness | Picture brightness |
-
| hue | Picture hue (colour only) |
-
| colour | Picture colour (colour only) |
-
| contrast | Picture contrast |
-
| whiteness | The whiteness (greyscale only) |
-
| depth | The capture depth (may need to match the frame buffer depth) |
-
| palette | Reports the palette that should be used for this image |
-
-
-The following palettes are defined
-
-
-| VIDEO_PALETTE_GREY | Linear intensity grey scale (255 is brightest). |
-
| VIDEO_PALETTE_HI240 | The BT848 8bit colour cube. |
-
| VIDEO_PALETTE_RGB565 | RGB565 packed into 16 bit words. |
-
| VIDEO_PALETTE_RGB555 | RGV555 packed into 16 bit words, top bit undefined. |
-
| VIDEO_PALETTE_RGB24 | RGB888 packed into 24bit words. |
-
| VIDEO_PALETTE_RGB32 | RGB888 packed into the low 3 bytes of 32bit words. The top 8bits are undefined. |
-
| VIDEO_PALETTE_YUV422 | Video style YUV422 - 8bits packed 4bits Y 2bits U 2bits V |
-
| VIDEO_PALETTE_YUYV | Describe me |
-
| VIDEO_PALETTE_UYVY | Describe me |
-
| VIDEO_PALETTE_YUV420 | YUV420 capture |
-
| VIDEO_PALETTE_YUV411 | YUV411 capture |
-
| VIDEO_PALETTE_RAW | RAW capture (BT848) |
-
| VIDEO_PALETTE_YUV422P | YUV 4:2:2 Planar |
-
| VIDEO_PALETTE_YUV411P | YUV 4:1:1 Planar |
-
-
-
Tuning
-Each video input channel can have one or more tuners associated with it. Many
-devices will not have tuners. TV cards and radio cards will have one or more
-tuners attached.
-
-Tuners are described by a struct video_tuner which can be obtained by
-the VIDIOCGTUNER ioctl. Fill in the tuner number in the structure
-then pass the structure to the ioctl to have the data filled in. The
-tuner can be switched using VIDIOCSTUNER which takes an integer argument
-giving the tuner to use. A struct tuner has the following fields
-
-
-| tuner | Number of the tuner |
-
| name | Canonical name for this tuner (eg FM/AM/TV) |
-
| rangelow | Lowest tunable frequency |
-
| rangehigh | Highest tunable frequency |
-
| flags | Flags describing the tuner |
-
| mode | The video signal mode if relevant |
-
| signal | Signal strength if known - between 0-65535 |
-
-
-The following flags exist
-
-
-| VIDEO_TUNER_PAL | PAL tuning is supported |
-
| VIDEO_TUNER_NTSC | NTSC tuning is supported |
-
| VIDEO_TUNER_SECAM | SECAM tuning is supported |
-
| VIDEO_TUNER_LOW | Frequency is in a lower range |
-
| VIDEO_TUNER_NORM | The norm for this tuner is settable |
-
| VIDEO_TUNER_STEREO_ON | The tuner is seeing stereo audio |
-
| VIDEO_TUNER_RDS_ON | The tuner is seeing a RDS datastream |
-
| VIDEO_TUNER_MBS_ON | The tuner is seeing a MBS datastream |
-
-
-The following modes are defined
-
-
-| VIDEO_MODE_PAL | The tuner is in PAL mode |
-
| VIDEO_MODE_NTSC | The tuner is in NTSC mode |
-
| VIDEO_MODE_SECAM | The tuner is in SECAM mode |
-
| VIDEO_MODE_AUTO | The tuner auto switches, or mode does not apply |
-
-
-Tuning frequencies are an unsigned 32bit value in 1/16th MHz or if the
-VIDEO_TUNER_LOW flag is set they are in 1/16th KHz. The current
-frequency is obtained as an unsigned long via the VIDIOCGFREQ ioctl and
-set by the VIDIOCSFREQ ioctl.
-
-
Audio
-TV and Radio devices have one or more audio inputs that may be selected.
-The audio properties are queried by passing a struct video_audio to VIDIOCGAUDIO ioctl. The
-VIDIOCSAUDIO ioctl sets audio properties.
-
-The structure contains the following fields
-
-
-| audio | The channel number |
-
| volume | The volume level |
-
| bass | The bass level |
-
| treble | The treble level |
-
| flags | Flags describing the audio channel |
-
| name | Canonical name for the audio input |
-
| mode | The mode the audio input is in |
-
| balance | The left/right balance |
-
| step | Actual step used by the hardware |
-
-
-The following flags are defined
-
-
-| VIDEO_AUDIO_MUTE | The audio is muted |
-
| VIDEO_AUDIO_MUTABLE | Audio muting is supported |
-
| VIDEO_AUDIO_VOLUME | The volume is controllable |
-
| VIDEO_AUDIO_BASS | The bass is controllable |
-
| VIDEO_AUDIO_TREBLE | The treble is controllable |
-
| VIDEO_AUDIO_BALANCE | The balance is controllable |
-
-
-The following decoding modes are defined
-
-
-| VIDEO_SOUND_MONO | Mono signal |
-
| VIDEO_SOUND_STEREO | Stereo signal (NICAM for TV) |
-
| VIDEO_SOUND_LANG1 | European TV alternate language 1 |
-
| VIDEO_SOUND_LANG2 | European TV alternate language 2 |
-
-
-
Reading Images
-Each call to the read syscall returns the next available image
-from the device. It is up to the caller to set format and size (using
-the VIDIOCSPICT and VIDIOCSWIN ioctls) and then to pass a suitable
-size buffer and length to the function. Not all devices will support
-read operations.
-
-A second way to handle image capture is via the mmap interface if supported.
-To use the mmap interface a user first sets the desired image size and depth
-properties. Next the VIDIOCGMBUF ioctl is issued. This reports the size
-of buffer to mmap and the offset within the buffer for each frame. The
-number of frames supported is device dependent and may only be one.
-
-The video_mbuf structure contains the following fields
-
-
-| size | The number of bytes to map |
-
| frames | The number of frames |
-
| offsets | The offset of each frame |
-
-
-Once the mmap has been made the VIDIOCMCAPTURE ioctl starts the
-capture to a frame using the format and image size specified in the
-video_mmap (which should match or be below the initial query size).
-When the VIDIOCMCAPTURE ioctl returns the frame is not
-captured yet, the driver just instructed the hardware to start the
-capture. The application has to use the VIDIOCSYNC ioctl to wait
-until the capture of a frame is finished. VIDIOCSYNC takes the frame
-number you want to wait for as argument.
-
-It is allowed to call VIDIOCMCAPTURE multiple times (with different
-frame numbers in video_mmap->frame of course) and thus have multiple
-outstanding capture requests. A simple way do to double-buffering
-using this feature looks like this:
-
-/* setup everything */
-VIDIOCMCAPTURE(0)
-while (whatever) {
- VIDIOCMCAPTURE(1)
- VIDIOCSYNC(0)
- /* process frame 0 while the hardware captures frame 1 */
- VIDIOCMCAPTURE(0)
- VIDIOCSYNC(1)
- /* process frame 1 while the hardware captures frame 0 */
-}
-
-Note that you are not limited to only two frames. The API
-allows up to 32 frames, the VIDIOCGMBUF ioctl returns the number of
-frames the driver granted. Thus it is possible to build deeper queues
-to avoid loosing frames on load peaks.
-
-While capturing to memory the driver will make a "best effort" attempt
-to capture to screen as well if requested. This normally means all
-frames that "miss" memory mapped capture will go to the display.
-
-A final ioctl exists to allow a device to obtain related devices if a
-driver has multiple components (for example video0 may not be associated
-with vbi0 which would cause an intercast display program to make a bad
-mistake). The VIDIOCGUNIT ioctl reports the unit numbers of the associated
-devices if any exist. The video_unit structure has the following fields.
-
-
-| video | Video capture device |
-
| vbi | VBI capture device |
-
| radio | Radio device |
-
| audio | Audio mixer |
-
| teletext | Teletext device |
-
-
-
RDS Datastreams
-For radio devices that support it, it is possible to receive Radio Data
-System (RDS) data by means of a read() on the device. The data is packed in
-groups of three, as follows:
-
-| First Octet | Least Significant Byte of RDS Block |
-| Second Octet | Most Significant Byte of RDS Block
- |
| Third Octet | Bit 7: | Error bit. Indicates that
-an uncorrectable error occurred during reception of this block. |
-| | Bit 6: | Corrected bit. Indicates that
-an error was corrected for this data block. |
-| | Bits 5-3: | Received Offset. Indicates the
-offset received by the sync system. |
-| | Bits 2-0: | Offset Name. Indicates the
-offset applied to this data. |
-
-
-
+V4L API
+Video For Linux APIs
+
diff --git a/Documentation/video4linux/CARDLIST.cx88 b/Documentation/video4linux/CARDLIST.cx88
index 216f705495cc..4377aa11f567 100644
--- a/Documentation/video4linux/CARDLIST.cx88
+++ b/Documentation/video4linux/CARDLIST.cx88
@@ -13,17 +13,17 @@ card=11 - Prolink PlayTV PVR
card=12 - ASUS PVR-416
card=13 - MSI TV-@nywhere
card=14 - KWorld/VStream XPert DVB-T
-card=15 - DVICO FusionHDTV DVB-T1
+card=15 - DViCO FusionHDTV DVB-T1
card=16 - KWorld LTV883RF
-card=17 - DViCO - FusionHDTV 3 Gold
+card=17 - DViCO FusionHDTV 3 Gold-Q
card=18 - Hauppauge Nova-T DVB-T
card=19 - Conexant DVB-T reference design
card=20 - Provideo PV259
-card=21 - DVICO FusionHDTV DVB-T Plus
+card=21 - DViCO FusionHDTV DVB-T Plus
card=22 - digitalnow DNTV Live! DVB-T
card=23 - pcHDTV HD3000 HDTV
card=24 - Hauppauge WinTV 28xxx (Roslyn) models
card=25 - Digital-Logic MICROSPACE Entertainment Center (MEC)
card=26 - IODATA GV/BCTV7E
card=27 - PixelView PlayTV Ultra Pro (Stereo)
-card=28 - DViCO - FusionHDTV 3 Gold-T
+card=28 - DViCO FusionHDTV 3 Gold-T
diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134
index d5ed95d28500..735e8ba02d9f 100644
--- a/Documentation/video4linux/CARDLIST.saa7134
+++ b/Documentation/video4linux/CARDLIST.saa7134
@@ -54,3 +54,9 @@
55 -> LifeView FlyDVB-T DUO [5168:0306]
56 -> Avermedia AVerTV 307 [1461:a70a]
57 -> Avermedia AVerTV GO 007 FM [1461:f31f]
+ 58 -> ADS Tech Instant TV (saa7135) [1421:0350,1421:0370]
+ 59 -> Kworld/Tevion V-Stream Xpert TV PVR7134
+ 60 -> Typhoon DVB-T Duo Digital/Analog Cardbus
+ 61 -> Philips TOUGH DVB-T reference design
+ 62 -> Compro VideoMate TV Gold+II
+ 63 -> Kworld Xpert TV PVR7134
diff --git a/Documentation/video4linux/CARDLIST.tuner b/Documentation/video4linux/CARDLIST.tuner
index aeb8df8ce890..e78020f68b2e 100644
--- a/Documentation/video4linux/CARDLIST.tuner
+++ b/Documentation/video4linux/CARDLIST.tuner
@@ -59,3 +59,6 @@ tuner=57 - Philips FQ1236A MK4
tuner=58 - Ymec TVision TVF-8531MF
tuner=59 - Ymec TVision TVF-5533MF
tuner=60 - Thomson DDT 7611 (ATSC/NTSC)
+tuner=61 - Tena TNF9533-D/IF
+tuner=62 - Philips TEA5767HN FM Radio
+tuner=63 - Philips FMD1216ME MK3 Hybrid Tuner
diff --git a/Documentation/video4linux/README.saa7134 b/Documentation/video4linux/README.saa7134
index 1a446c65365e..1f788e498eff 100644
--- a/Documentation/video4linux/README.saa7134
+++ b/Documentation/video4linux/README.saa7134
@@ -57,6 +57,15 @@ Cards can use either of these two crystals (xtal):
- 24.576MHz -> .audio_clock=0x200000
(xtal * .audio_clock = 51539600)
+Some details about 30/34/35:
+
+ - saa7130 - low-price chip, doesn't have mute, that is why all those
+ cards should have .mute field defined in their tuner structure.
+
+ - saa7134 - usual chip
+
+ - saa7133/35 - saa7135 is probably a marketing decision, since all those
+ chips identifies itself as 33 on pci.
Credits
=======
diff --git a/MAINTAINERS b/MAINTAINERS
index a0b0d595d17c..19a9a1c53037 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -512,11 +512,11 @@ W: http://linuxppc64.org
S: Supported
BTTV VIDEO4LINUX DRIVER
-P: Gerd Knorr
-M: kraxel@bytesex.org
+P: Mauro Carvalho Chehab
+M: mchehab@brturbo.com.br
L: video4linux-list@redhat.com
-W: http://bytesex.org/bttv/
-S: Orphan
+W: http://linuxtv.org
+S: Maintained
BUSLOGIC SCSI DRIVER
P: Leonard N. Zubkoff
@@ -1149,7 +1149,7 @@ S: Maintained
INFINIBAND SUBSYSTEM
P: Roland Dreier
-M: roland@topspin.com
+M: rolandd@cisco.com
P: Sean Hefty
M: mshefty@ichips.intel.com
P: Hal Rosenstock
@@ -2625,10 +2625,11 @@ W: http://rio500.sourceforge.net
S: Maintained
VIDEO FOR LINUX
-P: Gerd Knorr
-M: kraxel@bytesex.org
+P: Mauro Carvalho Chehab
+M: mchehab@brturbo.com.br
L: video4linux-list@redhat.com
-S: Orphan
+W: http://linuxtv.org
+S: Maintained
W1 DALLAS'S 1-WIRE BUS
P: Evgeniy Polyakov
diff --git a/Makefile b/Makefile
index 1fdace757e15..77aab7bdde01 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
VERSION = 2
PATCHLEVEL = 6
-SUBLEVEL = 12
-EXTRAVERSION =
+SUBLEVEL = 13
+EXTRAVERSION =-rc1
NAME=Woozy Numbat
# *DOCUMENTATION*
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index c8d94dcd8ef7..620f2ca94ed2 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -361,6 +361,11 @@ config NO_IDLE_HZ
Alternatively, if you want dynamic tick automatically enabled
during boot, pass "dyntick=enable" via the kernel command string.
+ Please note that dynamic tick may affect the accuracy of
+ timekeeping on some platforms depending on the implementation.
+ Currently at least OMAP platform is known to have accurate
+ timekeeping with dynamic tick.
+
config ARCH_DISCONTIGMEM_ENABLE
bool
default (ARCH_LH7A40X && !LH7A40X_CONTIGMEM)
diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c
index 4c38bd8bc298..b713c44c6fb4 100644
--- a/arch/arm/kernel/armksyms.c
+++ b/arch/arm/kernel/armksyms.c
@@ -30,9 +30,6 @@ extern void __lshrdi3(void);
extern void __modsi3(void);
extern void __muldi3(void);
extern void __ucmpdi2(void);
-extern void __udivdi3(void);
-extern void __umoddi3(void);
-extern void __udivmoddi4(void);
extern void __udivsi3(void);
extern void __umodsi3(void);
extern void __do_div64(void);
@@ -134,9 +131,6 @@ EXPORT_SYMBOL(__lshrdi3);
EXPORT_SYMBOL(__modsi3);
EXPORT_SYMBOL(__muldi3);
EXPORT_SYMBOL(__ucmpdi2);
-EXPORT_SYMBOL(__udivdi3);
-EXPORT_SYMBOL(__umoddi3);
-EXPORT_SYMBOL(__udivmoddi4);
EXPORT_SYMBOL(__udivsi3);
EXPORT_SYMBOL(__umodsi3);
EXPORT_SYMBOL(__do_div64);
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 8f146a4b4752..bbea636ff687 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -32,6 +32,7 @@
#include
#include
#include
+#include
extern const char *processor_modes[];
extern void setup_mm_for_reboot(char mode);
@@ -85,8 +86,10 @@ EXPORT_SYMBOL(pm_power_off);
void default_idle(void)
{
local_irq_disable();
- if (!need_resched() && !hlt_counter)
+ if (!need_resched() && !hlt_counter) {
+ timer_dyn_reprogram();
arch_idle();
+ }
local_irq_enable();
}
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 8cf733daa800..35b7273cfdb4 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -359,7 +359,8 @@ void cpu_init(void)
"I" (offsetof(struct stack, abt[0])),
"I" (PSR_F_BIT | PSR_I_BIT | UND_MODE),
"I" (offsetof(struct stack, und[0])),
- "I" (PSR_F_BIT | PSR_I_BIT | SVC_MODE));
+ "I" (PSR_F_BIT | PSR_I_BIT | SVC_MODE)
+ : "r14");
}
static struct machine_desc * __init setup_machine(unsigned int nr)
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 34892758f098..a931409c8fe4 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -502,3 +502,126 @@ int __init setup_profiling_timer(unsigned int multiplier)
{
return -EINVAL;
}
+
+static int
+on_each_cpu_mask(void (*func)(void *), void *info, int retry, int wait,
+ cpumask_t mask)
+{
+ int ret = 0;
+
+ preempt_disable();
+
+ ret = smp_call_function_on_cpu(func, info, retry, wait, mask);
+ if (cpu_isset(smp_processor_id(), mask))
+ func(info);
+
+ preempt_enable();
+
+ return ret;
+}
+
+/**********************************************************************/
+
+/*
+ * TLB operations
+ */
+struct tlb_args {
+ struct vm_area_struct *ta_vma;
+ unsigned long ta_start;
+ unsigned long ta_end;
+};
+
+static inline void ipi_flush_tlb_all(void *ignored)
+{
+ local_flush_tlb_all();
+}
+
+static inline void ipi_flush_tlb_mm(void *arg)
+{
+ struct mm_struct *mm = (struct mm_struct *)arg;
+
+ local_flush_tlb_mm(mm);
+}
+
+static inline void ipi_flush_tlb_page(void *arg)
+{
+ struct tlb_args *ta = (struct tlb_args *)arg;
+
+ local_flush_tlb_page(ta->ta_vma, ta->ta_start);
+}
+
+static inline void ipi_flush_tlb_kernel_page(void *arg)
+{
+ struct tlb_args *ta = (struct tlb_args *)arg;
+
+ local_flush_tlb_kernel_page(ta->ta_start);
+}
+
+static inline void ipi_flush_tlb_range(void *arg)
+{
+ struct tlb_args *ta = (struct tlb_args *)arg;
+
+ local_flush_tlb_range(ta->ta_vma, ta->ta_start, ta->ta_end);
+}
+
+static inline void ipi_flush_tlb_kernel_range(void *arg)
+{
+ struct tlb_args *ta = (struct tlb_args *)arg;
+
+ local_flush_tlb_kernel_range(ta->ta_start, ta->ta_end);
+}
+
+void flush_tlb_all(void)
+{
+ on_each_cpu(ipi_flush_tlb_all, NULL, 1, 1);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+ cpumask_t mask = mm->cpu_vm_mask;
+
+ on_each_cpu_mask(ipi_flush_tlb_mm, mm, 1, 1, mask);
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr)
+{
+ cpumask_t mask = vma->vm_mm->cpu_vm_mask;
+ struct tlb_args ta;
+
+ ta.ta_vma = vma;
+ ta.ta_start = uaddr;
+
+ on_each_cpu_mask(ipi_flush_tlb_page, &ta, 1, 1, mask);
+}
+
+void flush_tlb_kernel_page(unsigned long kaddr)
+{
+ struct tlb_args ta;
+
+ ta.ta_start = kaddr;
+
+ on_each_cpu(ipi_flush_tlb_kernel_page, &ta, 1, 1);
+}
+
+void flush_tlb_range(struct vm_area_struct *vma,
+ unsigned long start, unsigned long end)
+{
+ cpumask_t mask = vma->vm_mm->cpu_vm_mask;
+ struct tlb_args ta;
+
+ ta.ta_vma = vma;
+ ta.ta_start = start;
+ ta.ta_end = end;
+
+ on_each_cpu_mask(ipi_flush_tlb_range, &ta, 1, 1, mask);
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+ struct tlb_args ta;
+
+ ta.ta_start = start;
+ ta.ta_end = end;
+
+ on_each_cpu(ipi_flush_tlb_kernel_range, &ta, 1, 1);
+}
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index 06054c9ba074..1b7fcd50c3e2 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -424,15 +424,19 @@ static int timer_dyn_tick_disable(void)
return ret;
}
+/*
+ * Reprogram the system timer for at least the calculated time interval.
+ * This function should be called from the idle thread with IRQs disabled,
+ * immediately before sleeping.
+ */
void timer_dyn_reprogram(void)
{
struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
- unsigned long flags;
- write_seqlock_irqsave(&xtime_lock, flags);
+ write_seqlock(&xtime_lock);
if (dyn_tick->state & DYN_TICK_ENABLED)
dyn_tick->reprogram(next_timer_interrupt() - jiffies);
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_sequnlock(&xtime_lock);
}
static ssize_t timer_show_dyn_tick(struct sys_device *dev, char *buf)
diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile
index c0e65833ffc4..8725d63e4219 100644
--- a/arch/arm/lib/Makefile
+++ b/arch/arm/lib/Makefile
@@ -11,7 +11,7 @@ lib-y := backtrace.o changebit.o csumipv6.o csumpartial.o \
strnlen_user.o strchr.o strrchr.o testchangebit.o \
testclearbit.o testsetbit.o uaccess.o getuser.o \
putuser.o ashldi3.o ashrdi3.o lshrdi3.o muldi3.o \
- ucmpdi2.o udivdi3.o lib1funcs.o div64.o \
+ ucmpdi2.o lib1funcs.o div64.o \
io-readsb.o io-writesb.o io-readsl.o io-writesl.o
ifeq ($(CONFIG_CPU_32v3),y)
diff --git a/arch/arm/lib/longlong.h b/arch/arm/lib/longlong.h
deleted file mode 100644
index 90ae647e4d76..000000000000
--- a/arch/arm/lib/longlong.h
+++ /dev/null
@@ -1,183 +0,0 @@
-/* longlong.h -- based on code from gcc-2.95.3
-
- definitions for mixed size 32/64 bit arithmetic.
- Copyright (C) 1991, 92, 94, 95, 96, 1997, 1998 Free Software Foundation, Inc.
-
- This definition file is free software; you can redistribute it
- and/or modify it under the terms of the GNU General Public
- License as published by the Free Software Foundation; either
- version 2, or (at your option) any later version.
-
- This definition file is distributed in the hope that it will be
- useful, but WITHOUT ANY WARRANTY; without even the implied
- warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
- See the GNU General Public License for more details.
-
- You should have received a copy of the GNU General Public License
- along with this program; if not, write to the Free Software
- Foundation, Inc., 59 Temple Place - Suite 330,
- Boston, MA 02111-1307, USA. */
-
-/* Borrowed from GCC 2.95.3, I Molton 29/07/01 */
-
-#ifndef SI_TYPE_SIZE
-#define SI_TYPE_SIZE 32
-#endif
-
-#define __BITS4 (SI_TYPE_SIZE / 4)
-#define __ll_B (1L << (SI_TYPE_SIZE / 2))
-#define __ll_lowpart(t) ((u32) (t) % __ll_B)
-#define __ll_highpart(t) ((u32) (t) / __ll_B)
-
-/* Define auxiliary asm macros.
-
- 1) umul_ppmm(high_prod, low_prod, multipler, multiplicand)
- multiplies two u32 integers MULTIPLER and MULTIPLICAND,
- and generates a two-part u32 product in HIGH_PROD and
- LOW_PROD.
-
- 2) __umulsidi3(a,b) multiplies two u32 integers A and B,
- and returns a u64 product. This is just a variant of umul_ppmm.
-
- 3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
- denominator) divides a two-word unsigned integer, composed by the
- integers HIGH_NUMERATOR and LOW_NUMERATOR, by DENOMINATOR and
- places the quotient in QUOTIENT and the remainder in REMAINDER.
- HIGH_NUMERATOR must be less than DENOMINATOR for correct operation.
- If, in addition, the most significant bit of DENOMINATOR must be 1,
- then the pre-processor symbol UDIV_NEEDS_NORMALIZATION is defined to 1.
-
- 4) sdiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
- denominator). Like udiv_qrnnd but the numbers are signed. The
- quotient is rounded towards 0.
-
- 5) count_leading_zeros(count, x) counts the number of zero-bits from
- the msb to the first non-zero bit. This is the number of steps X
- needs to be shifted left to set the msb. Undefined for X == 0.
-
- 6) add_ssaaaa(high_sum, low_sum, high_addend_1, low_addend_1,
- high_addend_2, low_addend_2) adds two two-word unsigned integers,
- composed by HIGH_ADDEND_1 and LOW_ADDEND_1, and HIGH_ADDEND_2 and
- LOW_ADDEND_2 respectively. The result is placed in HIGH_SUM and
- LOW_SUM. Overflow (i.e. carry out) is not stored anywhere, and is
- lost.
-
- 7) sub_ddmmss(high_difference, low_difference, high_minuend,
- low_minuend, high_subtrahend, low_subtrahend) subtracts two
- two-word unsigned integers, composed by HIGH_MINUEND_1 and
- LOW_MINUEND_1, and HIGH_SUBTRAHEND_2 and LOW_SUBTRAHEND_2
- respectively. The result is placed in HIGH_DIFFERENCE and
- LOW_DIFFERENCE. Overflow (i.e. carry out) is not stored anywhere,
- and is lost.
-
- If any of these macros are left undefined for a particular CPU,
- C macros are used. */
-
-#if defined (__arm__)
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
- __asm__ ("adds %1, %4, %5 \n\
- adc %0, %2, %3" \
- : "=r" ((u32) (sh)), \
- "=&r" ((u32) (sl)) \
- : "%r" ((u32) (ah)), \
- "rI" ((u32) (bh)), \
- "%r" ((u32) (al)), \
- "rI" ((u32) (bl)))
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
- __asm__ ("subs %1, %4, %5 \n\
- sbc %0, %2, %3" \
- : "=r" ((u32) (sh)), \
- "=&r" ((u32) (sl)) \
- : "r" ((u32) (ah)), \
- "rI" ((u32) (bh)), \
- "r" ((u32) (al)), \
- "rI" ((u32) (bl)))
-#define umul_ppmm(xh, xl, a, b) \
-{register u32 __t0, __t1, __t2; \
- __asm__ ("%@ Inlined umul_ppmm \n\
- mov %2, %5, lsr #16 \n\
- mov %0, %6, lsr #16 \n\
- bic %3, %5, %2, lsl #16 \n\
- bic %4, %6, %0, lsl #16 \n\
- mul %1, %3, %4 \n\
- mul %4, %2, %4 \n\
- mul %3, %0, %3 \n\
- mul %0, %2, %0 \n\
- adds %3, %4, %3 \n\
- addcs %0, %0, #65536 \n\
- adds %1, %1, %3, lsl #16 \n\
- adc %0, %0, %3, lsr #16" \
- : "=&r" ((u32) (xh)), \
- "=r" ((u32) (xl)), \
- "=&r" (__t0), "=&r" (__t1), "=r" (__t2) \
- : "r" ((u32) (a)), \
- "r" ((u32) (b)));}
-#define UMUL_TIME 20
-#define UDIV_TIME 100
-#endif /* __arm__ */
-
-#define __umulsidi3(u, v) \
- ({DIunion __w; \
- umul_ppmm (__w.s.high, __w.s.low, u, v); \
- __w.ll; })
-
-#define __udiv_qrnnd_c(q, r, n1, n0, d) \
- do { \
- u32 __d1, __d0, __q1, __q0; \
- u32 __r1, __r0, __m; \
- __d1 = __ll_highpart (d); \
- __d0 = __ll_lowpart (d); \
- \
- __r1 = (n1) % __d1; \
- __q1 = (n1) / __d1; \
- __m = (u32) __q1 * __d0; \
- __r1 = __r1 * __ll_B | __ll_highpart (n0); \
- if (__r1 < __m) \
- { \
- __q1--, __r1 += (d); \
- if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
- if (__r1 < __m) \
- __q1--, __r1 += (d); \
- } \
- __r1 -= __m; \
- \
- __r0 = __r1 % __d1; \
- __q0 = __r1 / __d1; \
- __m = (u32) __q0 * __d0; \
- __r0 = __r0 * __ll_B | __ll_lowpart (n0); \
- if (__r0 < __m) \
- { \
- __q0--, __r0 += (d); \
- if (__r0 >= (d)) \
- if (__r0 < __m) \
- __q0--, __r0 += (d); \
- } \
- __r0 -= __m; \
- \
- (q) = (u32) __q1 * __ll_B | __q0; \
- (r) = __r0; \
- } while (0)
-
-#define UDIV_NEEDS_NORMALIZATION 1
-#define udiv_qrnnd __udiv_qrnnd_c
-
-#define count_leading_zeros(count, x) \
- do { \
- u32 __xr = (x); \
- u32 __a; \
- \
- if (SI_TYPE_SIZE <= 32) \
- { \
- __a = __xr < ((u32)1<<2*__BITS4) \
- ? (__xr < ((u32)1<<__BITS4) ? 0 : __BITS4) \
- : (__xr < ((u32)1<<3*__BITS4) ? 2*__BITS4 : 3*__BITS4); \
- } \
- else \
- { \
- for (__a = SI_TYPE_SIZE - 8; __a > 0; __a -= 8) \
- if (((__xr >> __a) & 0xff) != 0) \
- break; \
- } \
- \
- (count) = SI_TYPE_SIZE - (__clz_tab[__xr >> __a] + __a); \
- } while (0)
diff --git a/arch/arm/lib/udivdi3.c b/arch/arm/lib/udivdi3.c
deleted file mode 100644
index e343be4c6642..000000000000
--- a/arch/arm/lib/udivdi3.c
+++ /dev/null
@@ -1,222 +0,0 @@
-/* More subroutines needed by GCC output code on some machines. */
-/* Compile this one with gcc. */
-/* Copyright (C) 1989, 92-98, 1999 Free Software Foundation, Inc.
-
-This file is part of GNU CC.
-
-GNU CC is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2, or (at your option)
-any later version.
-
-GNU CC is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with GNU CC; see the file COPYING. If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA. */
-
-/* As a special exception, if you link this library with other files,
- some of which are compiled with GCC, to produce an executable,
- this library does not by itself cause the resulting executable
- to be covered by the GNU General Public License.
- This exception does not however invalidate any other reasons why
- the executable file might be covered by the GNU General Public License.
- */
-/* support functions required by the kernel. based on code from gcc-2.95.3 */
-/* I Molton 29/07/01 */
-
-#include "gcclib.h"
-#include "longlong.h"
-
-static const u8 __clz_tab[] = {
- 0, 1, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 5,
- 5, 5, 5, 5, 5, 5, 5, 5,
- 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
- 6, 6, 6, 6, 6, 6, 6, 6,
- 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
- 7, 7, 7, 7, 7, 7, 7, 7,
- 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
- 7, 7, 7, 7, 7, 7, 7, 7,
- 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
- 8, 8, 8, 8, 8, 8, 8, 8,
- 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
- 8, 8, 8, 8, 8, 8, 8, 8,
- 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
- 8, 8, 8, 8, 8, 8, 8, 8,
- 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
- 8, 8, 8, 8, 8, 8, 8, 8,
-};
-
-u64 __udivmoddi4(u64 n, u64 d, u64 * rp)
-{
- DIunion ww;
- DIunion nn, dd;
- DIunion rr;
- u32 d0, d1, n0, n1, n2;
- u32 q0, q1;
- u32 b, bm;
-
- nn.ll = n;
- dd.ll = d;
-
- d0 = dd.s.low;
- d1 = dd.s.high;
- n0 = nn.s.low;
- n1 = nn.s.high;
-
- if (d1 == 0) {
- if (d0 > n1) {
- /* 0q = nn / 0D */
-
- count_leading_zeros(bm, d0);
-
- if (bm != 0) {
- /* Normalize, i.e. make the most significant bit of the
- denominator set. */
-
- d0 = d0 << bm;
- n1 = (n1 << bm) | (n0 >> (SI_TYPE_SIZE - bm));
- n0 = n0 << bm;
- }
-
- udiv_qrnnd(q0, n0, n1, n0, d0);
- q1 = 0;
-
- /* Remainder in n0 >> bm. */
- } else {
- /* qq = NN / 0d */
-
- if (d0 == 0)
- d0 = 1 / d0; /* Divide intentionally by zero. */
-
- count_leading_zeros(bm, d0);
-
- if (bm == 0) {
- /* From (n1 >= d0) /\ (the most significant bit of d0 is set),
- conclude (the most significant bit of n1 is set) /\ (the
- leading quotient digit q1 = 1).
-
- This special case is necessary, not an optimization.
- (Shifts counts of SI_TYPE_SIZE are undefined.) */
-
- n1 -= d0;
- q1 = 1;
- } else {
- /* Normalize. */
-
- b = SI_TYPE_SIZE - bm;
-
- d0 = d0 << bm;
- n2 = n1 >> b;
- n1 = (n1 << bm) | (n0 >> b);
- n0 = n0 << bm;
-
- udiv_qrnnd(q1, n1, n2, n1, d0);
- }
-
- /* n1 != d0... */
-
- udiv_qrnnd(q0, n0, n1, n0, d0);
-
- /* Remainder in n0 >> bm. */
- }
-
- if (rp != 0) {
- rr.s.low = n0 >> bm;
- rr.s.high = 0;
- *rp = rr.ll;
- }
- } else {
- if (d1 > n1) {
- /* 00 = nn / DD */
-
- q0 = 0;
- q1 = 0;
-
- /* Remainder in n1n0. */
- if (rp != 0) {
- rr.s.low = n0;
- rr.s.high = n1;
- *rp = rr.ll;
- }
- } else {
- /* 0q = NN / dd */
-
- count_leading_zeros(bm, d1);
- if (bm == 0) {
- /* From (n1 >= d1) /\ (the most significant bit of d1 is set),
- conclude (the most significant bit of n1 is set) /\ (the
- quotient digit q0 = 0 or 1).
-
- This special case is necessary, not an optimization. */
-
- /* The condition on the next line takes advantage of that
- n1 >= d1 (true due to program flow). */
- if (n1 > d1 || n0 >= d0) {
- q0 = 1;
- sub_ddmmss(n1, n0, n1, n0, d1, d0);
- } else
- q0 = 0;
-
- q1 = 0;
-
- if (rp != 0) {
- rr.s.low = n0;
- rr.s.high = n1;
- *rp = rr.ll;
- }
- } else {
- u32 m1, m0;
- /* Normalize. */
-
- b = SI_TYPE_SIZE - bm;
-
- d1 = (d1 << bm) | (d0 >> b);
- d0 = d0 << bm;
- n2 = n1 >> b;
- n1 = (n1 << bm) | (n0 >> b);
- n0 = n0 << bm;
-
- udiv_qrnnd(q0, n1, n2, n1, d1);
- umul_ppmm(m1, m0, q0, d0);
-
- if (m1 > n1 || (m1 == n1 && m0 > n0)) {
- q0--;
- sub_ddmmss(m1, m0, m1, m0, d1, d0);
- }
-
- q1 = 0;
-
- /* Remainder in (n1n0 - m1m0) >> bm. */
- if (rp != 0) {
- sub_ddmmss(n1, n0, n1, n0, m1, m0);
- rr.s.low = (n1 << b) | (n0 >> bm);
- rr.s.high = n1 >> bm;
- *rp = rr.ll;
- }
- }
- }
- }
-
- ww.s.low = q0;
- ww.s.high = q1;
- return ww.ll;
-}
-
-u64 __udivdi3(u64 n, u64 d)
-{
- return __udivmoddi4(n, d, (u64 *) 0);
-}
-
-u64 __umoddi3(u64 u, u64 v)
-{
- u64 w;
-
- (void)__udivmoddi4(u, v, &w);
-
- return w;
-}
diff --git a/arch/arm/mach-aaec2000/Makefile.boot b/arch/arm/mach-aaec2000/Makefile.boot
new file mode 100644
index 000000000000..8f5a8b7c53c7
--- /dev/null
+++ b/arch/arm/mach-aaec2000/Makefile.boot
@@ -0,0 +1 @@
+ zreladdr-y := 0xf0008000
diff --git a/arch/arm/mach-aaec2000/core.c b/arch/arm/mach-aaec2000/core.c
index fc145b3768fa..aece0cd4f0a3 100644
--- a/arch/arm/mach-aaec2000/core.c
+++ b/arch/arm/mach-aaec2000/core.c
@@ -128,8 +128,8 @@ aaec2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction aaec2000_timer_irq = {
.name = "AAEC-2000 Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = aaec2000_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = aaec2000_timer_interrupt,
};
static void __init aaec2000_timer_init(void)
diff --git a/arch/arm/mach-clps711x/time.c b/arch/arm/mach-clps711x/time.c
index 383d4e0c6e35..1a23f0dcd4b8 100644
--- a/arch/arm/mach-clps711x/time.c
+++ b/arch/arm/mach-clps711x/time.c
@@ -57,8 +57,8 @@ p720t_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction clps711x_timer_irq = {
.name = "CLPS711x Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = p720t_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = p720t_timer_interrupt,
};
static void __init clps711x_timer_init(void)
diff --git a/arch/arm/mach-clps7500/core.c b/arch/arm/mach-clps7500/core.c
index 0bc7da488612..90e85f434f6f 100644
--- a/arch/arm/mach-clps7500/core.c
+++ b/arch/arm/mach-clps7500/core.c
@@ -298,8 +298,8 @@ clps7500_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction clps7500_timer_irq = {
.name = "CLPS7500 Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = clps7500_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = clps7500_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-ebsa110/core.c b/arch/arm/mach-ebsa110/core.c
index ef362d44949d..86ffdbb5626e 100644
--- a/arch/arm/mach-ebsa110/core.c
+++ b/arch/arm/mach-ebsa110/core.c
@@ -173,8 +173,8 @@ ebsa110_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction ebsa110_timer_irq = {
.name = "EBSA110 Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = ebsa110_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = ebsa110_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-epxa10db/time.c b/arch/arm/mach-epxa10db/time.c
index 1b991f3cc3c6..4b1084dde8dd 100644
--- a/arch/arm/mach-epxa10db/time.c
+++ b/arch/arm/mach-epxa10db/time.c
@@ -56,8 +56,8 @@ epxa10db_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction epxa10db_timer_irq = {
.name = "Excalibur Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = epxa10db_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = epxa10db_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-footbridge/dc21285-timer.c b/arch/arm/mach-footbridge/dc21285-timer.c
index da5b9b7623ca..14a62d6008fe 100644
--- a/arch/arm/mach-footbridge/dc21285-timer.c
+++ b/arch/arm/mach-footbridge/dc21285-timer.c
@@ -43,7 +43,7 @@ timer1_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction footbridge_timer_irq = {
.name = "Timer1 timer tick",
.handler = timer1_interrupt,
- .flags = SA_INTERRUPT,
+ .flags = SA_INTERRUPT | SA_TIMER,
};
/*
diff --git a/arch/arm/mach-footbridge/isa-timer.c b/arch/arm/mach-footbridge/isa-timer.c
index a4fefa0bb5a1..c1d74f7ab669 100644
--- a/arch/arm/mach-footbridge/isa-timer.c
+++ b/arch/arm/mach-footbridge/isa-timer.c
@@ -72,7 +72,7 @@ isa_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction isa_timer_irq = {
.name = "ISA timer tick",
.handler = isa_timer_interrupt,
- .flags = SA_INTERRUPT,
+ .flags = SA_INTERRUPT | SA_TIMER,
};
static void __init isa_timer_init(void)
diff --git a/arch/arm/mach-h720x/cpu-h7201.c b/arch/arm/mach-h720x/cpu-h7201.c
index 743656881ed6..af9e4a5d5ea7 100644
--- a/arch/arm/mach-h720x/cpu-h7201.c
+++ b/arch/arm/mach-h720x/cpu-h7201.c
@@ -41,8 +41,8 @@ h7201_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction h7201_timer_irq = {
.name = "h7201 Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = h7201_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = h7201_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-h720x/cpu-h7202.c b/arch/arm/mach-h720x/cpu-h7202.c
index 21b8fb6122cd..593b6a2a30e1 100644
--- a/arch/arm/mach-h720x/cpu-h7202.c
+++ b/arch/arm/mach-h720x/cpu-h7202.c
@@ -171,8 +171,8 @@ static struct irqchip h7202_timerx_chip = {
static struct irqaction h7202_timer_irq = {
.name = "h7202 Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = h7202_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = h7202_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-imx/time.c b/arch/arm/mach-imx/time.c
index 11f1e56c36bc..ea805bfa5e54 100644
--- a/arch/arm/mach-imx/time.c
+++ b/arch/arm/mach-imx/time.c
@@ -72,8 +72,8 @@ imx_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction imx_timer_irq = {
.name = "i.MX Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = imx_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = imx_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c
index bd1e5e3c9d34..dacbf504dae2 100644
--- a/arch/arm/mach-integrator/core.c
+++ b/arch/arm/mach-integrator/core.c
@@ -20,6 +20,7 @@
#include
#include
#include
+#include
#include
#include
#include
@@ -156,16 +157,6 @@ EXPORT_SYMBOL(cm_control);
#define TICKS2USECS(x) ((x) / TICKS_PER_uSEC)
#endif
-/*
- * What does it look like?
- */
-typedef struct TimerStruct {
- unsigned long TimerLoad;
- unsigned long TimerValue;
- unsigned long TimerControl;
- unsigned long TimerClear;
-} TimerStruct_t;
-
static unsigned long timer_reload;
/*
@@ -174,7 +165,6 @@ static unsigned long timer_reload;
*/
unsigned long integrator_gettimeoffset(void)
{
- volatile TimerStruct_t *timer1 = (TimerStruct_t *)TIMER1_VA_BASE;
unsigned long ticks1, ticks2, status;
/*
@@ -183,11 +173,11 @@ unsigned long integrator_gettimeoffset(void)
* an interrupt. We get around this by ensuring that the
* counter has not reloaded between our two reads.
*/
- ticks2 = timer1->TimerValue & 0xffff;
+ ticks2 = readl(TIMER1_VA_BASE + TIMER_VALUE) & 0xffff;
do {
ticks1 = ticks2;
status = __raw_readl(VA_IC_BASE + IRQ_RAW_STATUS);
- ticks2 = timer1->TimerValue & 0xffff;
+ ticks2 = readl(TIMER1_VA_BASE + TIMER_VALUE) & 0xffff;
} while (ticks2 > ticks1);
/*
@@ -213,14 +203,12 @@ unsigned long integrator_gettimeoffset(void)
static irqreturn_t
integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
{
- volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
-
write_seqlock(&xtime_lock);
/*
* clear the interrupt
*/
- timer1->TimerClear = 1;
+ writel(1, TIMER1_VA_BASE + TIMER_INTCLR);
/*
* the clock tick routines are only processed on the
@@ -247,8 +235,8 @@ integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction integrator_timer_irq = {
.name = "Integrator Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = integrator_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = integrator_timer_interrupt,
};
/*
@@ -256,32 +244,29 @@ static struct irqaction integrator_timer_irq = {
*/
void __init integrator_time_init(unsigned long reload, unsigned int ctrl)
{
- volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
- volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
- volatile TimerStruct_t *timer2 = (volatile TimerStruct_t *)TIMER2_VA_BASE;
- unsigned int timer_ctrl = 0x80 | 0x40; /* periodic */
+ unsigned int timer_ctrl = TIMER_CTRL_ENABLE | TIMER_CTRL_PERIODIC;
timer_reload = reload;
timer_ctrl |= ctrl;
if (timer_reload > 0x100000) {
timer_reload >>= 8;
- timer_ctrl |= 0x08; /* /256 */
+ timer_ctrl |= TIMER_CTRL_DIV256;
} else if (timer_reload > 0x010000) {
timer_reload >>= 4;
- timer_ctrl |= 0x04; /* /16 */
+ timer_ctrl |= TIMER_CTRL_DIV16;
}
/*
* Initialise to a known state (all timers off)
*/
- timer0->TimerControl = 0;
- timer1->TimerControl = 0;
- timer2->TimerControl = 0;
+ writel(0, TIMER0_VA_BASE + TIMER_CTRL);
+ writel(0, TIMER1_VA_BASE + TIMER_CTRL);
+ writel(0, TIMER2_VA_BASE + TIMER_CTRL);
- timer1->TimerLoad = timer_reload;
- timer1->TimerValue = timer_reload;
- timer1->TimerControl = timer_ctrl;
+ writel(timer_reload, TIMER1_VA_BASE + TIMER_LOAD);
+ writel(timer_reload, TIMER1_VA_BASE + TIMER_VALUE);
+ writel(timer_ctrl, TIMER1_VA_BASE + TIMER_CTRL);
/*
* Make irqs happen for the system timer
diff --git a/arch/arm/mach-iop3xx/iop321-time.c b/arch/arm/mach-iop3xx/iop321-time.c
index 9b7dd64d1b8f..d53af1669502 100644
--- a/arch/arm/mach-iop3xx/iop321-time.c
+++ b/arch/arm/mach-iop3xx/iop321-time.c
@@ -86,7 +86,7 @@ iop321_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction iop321_timer_irq = {
.name = "IOP321 Timer Tick",
.handler = iop321_timer_interrupt,
- .flags = SA_INTERRUPT
+ .flags = SA_INTERRUPT | SA_TIMER,
};
static void __init iop321_timer_init(void)
diff --git a/arch/arm/mach-iop3xx/iop331-time.c b/arch/arm/mach-iop3xx/iop331-time.c
index e01696769263..1a6d9d661e4b 100644
--- a/arch/arm/mach-iop3xx/iop331-time.c
+++ b/arch/arm/mach-iop3xx/iop331-time.c
@@ -83,7 +83,7 @@ iop331_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction iop331_timer_irq = {
.name = "IOP331 Timer Tick",
.handler = iop331_timer_interrupt,
- .flags = SA_INTERRUPT
+ .flags = SA_INTERRUPT | SA_TIMER,
};
static void __init iop331_timer_init(void)
diff --git a/arch/arm/mach-ixp2000/core.c b/arch/arm/mach-ixp2000/core.c
index 0ee34acb8d7b..4b9d841e04c1 100644
--- a/arch/arm/mach-ixp2000/core.c
+++ b/arch/arm/mach-ixp2000/core.c
@@ -102,6 +102,11 @@ static struct map_desc ixp2000_io_desc[] __initdata = {
.physical = IXP2000_PCI_CSR_PHYS_BASE,
.length = IXP2000_PCI_CSR_SIZE,
.type = MT_DEVICE
+ }, {
+ .virtual = IXP2000_MSF_VIRT_BASE,
+ .physical = IXP2000_MSF_PHYS_BASE,
+ .length = IXP2000_MSF_SIZE,
+ .type = MT_DEVICE
}, {
.virtual = IXP2000_PCI_IO_VIRT_BASE,
.physical = IXP2000_PCI_IO_PHYS_BASE,
@@ -194,8 +199,8 @@ static int ixp2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction ixp2000_timer_irq = {
.name = "IXP2000 Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = ixp2000_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = ixp2000_timer_interrupt,
};
void __init ixp2000_init_time(unsigned long tick_rate)
diff --git a/arch/arm/mach-ixp2000/ixdp2800.c b/arch/arm/mach-ixp2000/ixdp2800.c
index aec13c7108a9..468a4bbfb724 100644
--- a/arch/arm/mach-ixp2000/ixdp2800.c
+++ b/arch/arm/mach-ixp2000/ixdp2800.c
@@ -42,12 +42,6 @@
#include
#include
-
-void ixdp2400_init_irq(void)
-{
- ixdp2x00_init_irq(IXDP2800_CPLD_INT_STAT, IXDP2800_CPLD_INT_MASK, IXDP2400_NR_IRQS);
-}
-
/*************************************************************************
* IXDP2800 timer tick
*************************************************************************/
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index f39e8408488f..04490a9f8f6e 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -298,8 +298,8 @@ static irqreturn_t ixp4xx_timer_interrupt(int irq, void *dev_id, struct pt_regs
static struct irqaction ixp4xx_timer_irq = {
.name = "IXP4xx Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = ixp4xx_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = ixp4xx_timer_interrupt,
};
static void __init ixp4xx_timer_init(void)
diff --git a/arch/arm/mach-lh7a40x/time.c b/arch/arm/mach-lh7a40x/time.c
index 51e1c814b400..be377e331f25 100644
--- a/arch/arm/mach-lh7a40x/time.c
+++ b/arch/arm/mach-lh7a40x/time.c
@@ -53,8 +53,8 @@ lh7a40x_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction lh7a40x_timer_irq = {
.name = "LHA740x Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = lh7a40x_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = lh7a40x_timer_interrupt,
};
static void __init lh7a40x_timer_init(void)
diff --git a/arch/arm/mach-omap/pm.c b/arch/arm/mach-omap/pm.c
index 00fac155df2a..6b03ccdc1e92 100644
--- a/arch/arm/mach-omap/pm.c
+++ b/arch/arm/mach-omap/pm.c
@@ -41,7 +41,9 @@
#include
#include
+#include
#include
+
#include
#include
#include
@@ -80,13 +82,13 @@ void omap_pm_idle(void)
return;
}
mask32 = omap_readl(ARM_SYSST);
- local_fiq_enable();
- local_irq_enable();
-#if defined(CONFIG_OMAP_32K_TIMER) && defined(CONFIG_NO_IDLE_HZ)
- /* Override timer to use VST for the next cycle */
- omap_32k_timer_next_vst_interrupt();
-#endif
+ /*
+ * Since an interrupt may set up a timer, we don't want to
+ * reprogram the hardware timer with interrupts enabled.
+ * Re-enable interrupts only after returning from idle.
+ */
+ timer_dyn_reprogram();
if ((mask32 & DSP_IDLE) == 0) {
__asm__ volatile ("mcr p15, 0, r0, c7, c0, 4");
@@ -102,6 +104,8 @@ void omap_pm_idle(void)
func_ptr();
}
+ local_fiq_enable();
+ local_irq_enable();
}
/*
diff --git a/arch/arm/mach-omap/time.c b/arch/arm/mach-omap/time.c
index 4205fdcb632c..dd34e9f4c413 100644
--- a/arch/arm/mach-omap/time.c
+++ b/arch/arm/mach-omap/time.c
@@ -4,7 +4,7 @@
* OMAP Timers
*
* Copyright (C) 2004 Nokia Corporation
- * Partial timer rewrite and additional VST timer support by
+ * Partial timer rewrite and additional dynamic tick timer support by
* Tony Lindgen and
* Tuukka Tikkanen
*
@@ -188,8 +188,8 @@ static irqreturn_t omap_mpu_timer_interrupt(int irq, void *dev_id,
static struct irqaction omap_mpu_timer_irq = {
.name = "mpu timer",
- .flags = SA_INTERRUPT,
- .handler = omap_mpu_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = omap_mpu_timer_interrupt,
};
static unsigned long omap_mpu_timer1_overflows;
@@ -203,7 +203,7 @@ static irqreturn_t omap_mpu_timer1_interrupt(int irq, void *dev_id,
static struct irqaction omap_mpu_timer1_irq = {
.name = "mpu timer1 overflow",
.flags = SA_INTERRUPT,
- .handler = omap_mpu_timer1_interrupt
+ .handler = omap_mpu_timer1_interrupt,
};
static __init void omap_init_mpu_timer(void)
@@ -261,7 +261,6 @@ unsigned long long sched_clock(void)
* so with HZ = 100, TVR = 327.68.
*/
#define OMAP_32K_TIMER_TICK_PERIOD ((32768 / HZ) - 1)
-#define MAX_SKIP_JIFFIES 25
#define TIMER_32K_SYNCHRONIZED 0xfffbc410
#define JIFFIES_TO_HW_TICKS(nr_jiffies, clock_rate) \
@@ -347,14 +346,55 @@ static irqreturn_t omap_32k_timer_interrupt(int irq, void *dev_id,
return IRQ_HANDLED;
}
+#ifdef CONFIG_NO_IDLE_HZ
+/*
+ * Programs the next timer interrupt needed. Called when dynamic tick is
+ * enabled, and to reprogram the ticks to skip from pm_idle. Note that
+ * we can keep the timer continuous, and don't need to set it to run in
+ * one-shot mode. This is because the timer will get reprogrammed again
+ * after next interrupt.
+ */
+void omap_32k_timer_reprogram(unsigned long next_tick)
+{
+ omap_32k_timer_start(JIFFIES_TO_HW_TICKS(next_tick, 32768) + 1);
+}
+
+static struct irqaction omap_32k_timer_irq;
+extern struct timer_update_handler timer_update;
+
+static int omap_32k_timer_enable_dyn_tick(void)
+{
+ /* No need to reprogram timer, just use the next interrupt */
+ return 0;
+}
+
+static int omap_32k_timer_disable_dyn_tick(void)
+{
+ omap_32k_timer_start(OMAP_32K_TIMER_TICK_PERIOD);
+ return 0;
+}
+
+static struct dyn_tick_timer omap_dyn_tick_timer = {
+ .enable = omap_32k_timer_enable_dyn_tick,
+ .disable = omap_32k_timer_disable_dyn_tick,
+ .reprogram = omap_32k_timer_reprogram,
+ .handler = omap_32k_timer_interrupt,
+};
+#endif /* CONFIG_NO_IDLE_HZ */
+
static struct irqaction omap_32k_timer_irq = {
.name = "32KHz timer",
- .flags = SA_INTERRUPT,
- .handler = omap_32k_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = omap_32k_timer_interrupt,
};
static __init void omap_init_32k_timer(void)
{
+
+#ifdef CONFIG_NO_IDLE_HZ
+ omap_timer.dyn_tick = &omap_dyn_tick_timer;
+#endif
+
setup_irq(INT_OS_TIMER, &omap_32k_timer_irq);
omap_timer.offset = omap_32k_timer_gettimeoffset;
omap_32k_last_tick = omap_32k_sync_timer_read();
diff --git a/arch/arm/mach-omap/usb.c b/arch/arm/mach-omap/usb.c
index 6e805d451d0e..7f37857b1a28 100644
--- a/arch/arm/mach-omap/usb.c
+++ b/arch/arm/mach-omap/usb.c
@@ -288,8 +288,8 @@ static void usb_release(struct device *dev)
static struct resource udc_resources[] = {
/* order is significant! */
{ /* registers */
- .start = IO_ADDRESS(UDC_BASE),
- .end = IO_ADDRESS(UDC_BASE + 0xff),
+ .start = UDC_BASE,
+ .end = UDC_BASE + 0xff,
.flags = IORESOURCE_MEM,
}, { /* general IRQ */
.start = IH2_BASE + 20,
@@ -355,8 +355,8 @@ static struct platform_device ohci_device = {
static struct resource otg_resources[] = {
/* order is significant! */
{
- .start = IO_ADDRESS(OTG_BASE),
- .end = IO_ADDRESS(OTG_BASE + 0xff),
+ .start = OTG_BASE,
+ .end = OTG_BASE + 0xff,
.flags = IORESOURCE_MEM,
}, {
.start = IH2_BASE + 8,
diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c
index 473fb6173f72..6e5202154f91 100644
--- a/arch/arm/mach-pxa/time.c
+++ b/arch/arm/mach-pxa/time.c
@@ -105,8 +105,8 @@ pxa_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction pxa_timer_irq = {
.name = "PXA Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = pxa_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = pxa_timer_interrupt,
};
static void __init pxa_timer_init(void)
diff --git a/arch/arm/mach-s3c2410/Kconfig b/arch/arm/mach-s3c2410/Kconfig
index 534df0c6c770..d4d03d0daaec 100644
--- a/arch/arm/mach-s3c2410/Kconfig
+++ b/arch/arm/mach-s3c2410/Kconfig
@@ -154,6 +154,11 @@ config S3C2410_PM_CHECK_CHUNKSIZE
the CRC data block will take more memory, but wil identify any
faults with better precision.
+config PM_SIMTEC
+ bool
+ depends on PM && (ARCH_BAST || MACH_VR1000)
+ default y
+
config S3C2410_LOWLEVEL_UART_PORT
int "S3C2410 UART to use for low-level messages"
default 0
diff --git a/arch/arm/mach-s3c2410/Makefile b/arch/arm/mach-s3c2410/Makefile
index 7c379aad5d62..f99b689e4392 100644
--- a/arch/arm/mach-s3c2410/Makefile
+++ b/arch/arm/mach-s3c2410/Makefile
@@ -18,6 +18,7 @@ obj-$(CONFIG_S3C2410_DMA) += dma.o
# Power Management support
obj-$(CONFIG_PM) += pm.o sleep.o
+obj-$(CONFIG_PM_SIMTEC) += pm-simtec.o
# S3C2440 support
diff --git a/arch/arm/mach-s3c2410/devs.c b/arch/arm/mach-s3c2410/devs.c
index 64792f678668..4664bd11adc1 100644
--- a/arch/arm/mach-s3c2410/devs.c
+++ b/arch/arm/mach-s3c2410/devs.c
@@ -96,8 +96,8 @@ struct platform_device s3c_device_lcd = {
.num_resources = ARRAY_SIZE(s3c_lcd_resource),
.resource = s3c_lcd_resource,
.dev = {
- .dma_mask = &s3c_device_lcd_dmamask,
- .coherent_dma_mask = 0xffffffffUL
+ .dma_mask = &s3c_device_lcd_dmamask,
+ .coherent_dma_mask = 0xffffffffUL
}
};
diff --git a/arch/arm/mach-s3c2410/irq.c b/arch/arm/mach-s3c2410/irq.c
index b668c48f4399..cf9f46d88061 100644
--- a/arch/arm/mach-s3c2410/irq.c
+++ b/arch/arm/mach-s3c2410/irq.c
@@ -40,8 +40,11 @@
* 04-Nov-2004 Ben Dooks
* Fix standard IRQ wake for EINT0..4 and RTC
*
- * 22-Feb-2004 Ben Dooks
+ * 22-Feb-2005 Ben Dooks
* Fixed edge-triggering on ADC IRQ
+ *
+ * 28-Jun-2005 Ben Dooks
+ * Mark IRQ_LCD valid
*/
#include
@@ -366,7 +369,6 @@ static struct irqchip s3c_irq_eint0t4 = {
#define INTMSK_UART1 (1UL << (IRQ_UART1 - IRQ_EINT0))
#define INTMSK_UART2 (1UL << (IRQ_UART2 - IRQ_EINT0))
#define INTMSK_ADCPARENT (1UL << (IRQ_ADCPARENT - IRQ_EINT0))
-#define INTMSK_LCD (1UL << (IRQ_LCD - IRQ_EINT0))
static inline void
s3c_irqsub_mask(unsigned int irqno, unsigned int parentbit,
@@ -716,7 +718,6 @@ void __init s3c24xx_init_irq(void)
case IRQ_UART0:
case IRQ_UART1:
case IRQ_UART2:
- case IRQ_LCD:
case IRQ_ADCPARENT:
set_irq_chip(irqno, &s3c_irq_level_chip);
set_irq_handler(irqno, do_level_IRQ);
diff --git a/arch/arm/mach-s3c2410/mach-bast.c b/arch/arm/mach-s3c2410/mach-bast.c
index f3e970039b65..549bcb1f32c0 100644
--- a/arch/arm/mach-s3c2410/mach-bast.c
+++ b/arch/arm/mach-s3c2410/mach-bast.c
@@ -27,6 +27,7 @@
* 10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA
* 14-Mar-2006 BJD Updated for __iomem changes
* 22-Jun-2006 BJD Added DM9000 platform information
+ * 28-Jun-2006 BJD Moved pm functionality out to common code
*/
#include
@@ -67,7 +68,6 @@
#include "devs.h"
#include "cpu.h"
#include "usb-simtec.h"
-#include "pm.h"
#define COPYRIGHT ", (c) 2004-2005 Simtec Electronics"
@@ -405,44 +405,13 @@ void __init bast_map_io(void)
usb_simtec_init();
}
-void __init bast_init_irq(void)
-{
- s3c24xx_init_irq();
-}
-
-#ifdef CONFIG_PM
-
-/* bast_init_machine
- *
- * enable the power management functions for the EB2410ITX
-*/
-
-static __init void bast_init_machine(void)
-{
- unsigned long gstatus4;
-
- printk(KERN_INFO "BAST Power Manangement" COPYRIGHT "\n");
-
- gstatus4 = (__raw_readl(S3C2410_BANKCON7) & 0x3) << 30;
- gstatus4 |= (__raw_readl(S3C2410_BANKCON6) & 0x3) << 28;
- gstatus4 |= (__raw_readl(S3C2410_BANKSIZE) & S3C2410_BANKSIZE_MASK);
-
- __raw_writel(gstatus4, S3C2410_GSTATUS4);
-
- s3c2410_pm_init();
-}
-
-#else
-#define bast_init_machine NULL
-#endif
-
MACHINE_START(BAST, "Simtec-BAST")
MAINTAINER("Ben Dooks ")
BOOT_MEM(S3C2410_SDRAM_PA, S3C2410_PA_UART, (u32)S3C24XX_VA_UART)
BOOT_PARAMS(S3C2410_SDRAM_PA + 0x100)
- MAPIO(bast_map_io)
- INITIRQ(bast_init_irq)
- .init_machine = bast_init_machine,
+
+ .map_io = bast_map_io,
+ .init_irq = s3c24xx_init_irq,
.timer = &s3c24xx_timer,
MACHINE_END
diff --git a/arch/arm/mach-s3c2410/mach-vr1000.c b/arch/arm/mach-s3c2410/mach-vr1000.c
index 76be074944a0..1db2855e3e56 100644
--- a/arch/arm/mach-s3c2410/mach-vr1000.c
+++ b/arch/arm/mach-s3c2410/mach-vr1000.c
@@ -371,16 +371,12 @@ void __init vr1000_map_io(void)
usb_simtec_init();
}
-void __init vr1000_init_irq(void)
-{
- s3c24xx_init_irq();
-}
MACHINE_START(VR1000, "Thorcom-VR1000")
MAINTAINER("Ben Dooks ")
BOOT_MEM(S3C2410_SDRAM_PA, S3C2410_PA_UART, (u32)S3C24XX_VA_UART)
BOOT_PARAMS(S3C2410_SDRAM_PA + 0x100)
- MAPIO(vr1000_map_io)
- INITIRQ(vr1000_init_irq)
+ .map_io = vr1000_map_io,
+ .init_irq = s3c24xx_init_irq,
.timer = &s3c24xx_timer,
MACHINE_END
diff --git a/arch/arm/mach-s3c2410/pm-simtec.c b/arch/arm/mach-s3c2410/pm-simtec.c
new file mode 100644
index 000000000000..2cb798832223
--- /dev/null
+++ b/arch/arm/mach-s3c2410/pm-simtec.c
@@ -0,0 +1,65 @@
+/* linux/arch/arm/mach-s3c2410/pm-simtec.c
+ *
+ * Copyright (c) 2004 Simtec Electronics
+ * Ben Dooks
+ *
+ * http://armlinux.simtec.co.uk/
+ *
+ * Power Management helpers for Simtec S3C24XX implementations
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+
+#include
+#include
+
+#include
+#include
+
+#include
+#include
+#include
+#include
+
+#include
+
+#include "pm.h"
+
+#define COPYRIGHT ", (c) 2005 Simtec Electronics"
+
+/* pm_simtec_init
+ *
+ * enable the power management functions
+*/
+
+static __init int pm_simtec_init(void)
+{
+ unsigned long gstatus4;
+
+ /* check which machine we are running on */
+
+ if (!machine_is_bast() && !machine_is_vr1000())
+ return 0;
+
+ printk(KERN_INFO "Simtec Board Power Manangement" COPYRIGHT "\n");
+
+ gstatus4 = (__raw_readl(S3C2410_BANKCON7) & 0x3) << 30;
+ gstatus4 |= (__raw_readl(S3C2410_BANKCON6) & 0x3) << 28;
+ gstatus4 |= (__raw_readl(S3C2410_BANKSIZE) & S3C2410_BANKSIZE_MASK);
+
+ __raw_writel(gstatus4, S3C2410_GSTATUS4);
+
+ return s3c2410_pm_init();
+}
+
+arch_initcall(pm_simtec_init);
diff --git a/arch/arm/mach-s3c2410/time.c b/arch/arm/mach-s3c2410/time.c
index 179f0e031af4..765a3a9ae032 100644
--- a/arch/arm/mach-s3c2410/time.c
+++ b/arch/arm/mach-s3c2410/time.c
@@ -137,8 +137,8 @@ s3c2410_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction s3c2410_timer_irq = {
.name = "S3C2410 Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = s3c2410_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = s3c2410_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-sa1100/h3600.c b/arch/arm/mach-sa1100/h3600.c
index 84c86543501a..65dbe991426d 100644
--- a/arch/arm/mach-sa1100/h3600.c
+++ b/arch/arm/mach-sa1100/h3600.c
@@ -727,7 +727,7 @@ static void h3800_IRQ_demux(unsigned int irq, struct irqdesc *desc, struct pt_re
static struct irqaction h3800_irq = {
.name = "h3800_asic",
.handler = h3800_IRQ_demux,
- .flags = SA_INTERRUPT,
+ .flags = SA_INTERRUPT | SA_TIMER,
};
u32 kpio_int_shadow = 0;
diff --git a/arch/arm/mach-sa1100/time.c b/arch/arm/mach-sa1100/time.c
index 19b0c0fd6377..0eeb3616ffea 100644
--- a/arch/arm/mach-sa1100/time.c
+++ b/arch/arm/mach-sa1100/time.c
@@ -99,8 +99,8 @@ sa1100_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction sa1100_timer_irq = {
.name = "SA11xx Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = sa1100_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = sa1100_timer_interrupt,
};
static void __init sa1100_timer_init(void)
diff --git a/arch/arm/mach-shark/core.c b/arch/arm/mach-shark/core.c
index a9bc5d0dbd85..aa0e2f6e02f6 100644
--- a/arch/arm/mach-shark/core.c
+++ b/arch/arm/mach-shark/core.c
@@ -84,8 +84,8 @@ shark_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
static struct irqaction shark_timer_irq = {
.name = "Shark Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = shark_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = shark_timer_interrupt,
};
/*
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 6a7cbea5e098..f01c0f8a2bb3 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -33,6 +33,7 @@
#include
#include
#include
+#include
#include
#include
@@ -788,38 +789,25 @@ void __init versatile_init(void)
*/
#define TIMER_INTERVAL (TICKS_PER_uSEC * mSEC_10)
#if TIMER_INTERVAL >= 0x100000
-#define TIMER_RELOAD (TIMER_INTERVAL >> 8) /* Divide by 256 */
-#define TIMER_CTRL 0x88 /* Enable, Clock / 256 */
+#define TIMER_RELOAD (TIMER_INTERVAL >> 8)
+#define TIMER_DIVISOR (TIMER_CTRL_DIV256)
#define TICKS2USECS(x) (256 * (x) / TICKS_PER_uSEC)
#elif TIMER_INTERVAL >= 0x10000
#define TIMER_RELOAD (TIMER_INTERVAL >> 4) /* Divide by 16 */
-#define TIMER_CTRL 0x84 /* Enable, Clock / 16 */
+#define TIMER_DIVISOR (TIMER_CTRL_DIV16)
#define TICKS2USECS(x) (16 * (x) / TICKS_PER_uSEC)
#else
#define TIMER_RELOAD (TIMER_INTERVAL)
-#define TIMER_CTRL 0x80 /* Enable */
+#define TIMER_DIVISOR (TIMER_CTRL_DIV1)
#define TICKS2USECS(x) ((x) / TICKS_PER_uSEC)
#endif
-#define TIMER_CTRL_IE (1 << 5) /* Interrupt Enable */
-
-/*
- * What does it look like?
- */
-typedef struct TimerStruct {
- unsigned long TimerLoad;
- unsigned long TimerValue;
- unsigned long TimerControl;
- unsigned long TimerClear;
-} TimerStruct_t;
-
/*
* Returns number of ms since last clock interrupt. Note that interrupts
* will have been disabled by do_gettimeoffset()
*/
static unsigned long versatile_gettimeoffset(void)
{
- volatile TimerStruct_t *timer0 = (TimerStruct_t *)TIMER0_VA_BASE;
unsigned long ticks1, ticks2, status;
/*
@@ -828,11 +816,11 @@ static unsigned long versatile_gettimeoffset(void)
* an interrupt. We get around this by ensuring that the
* counter has not reloaded between our two reads.
*/
- ticks2 = timer0->TimerValue & 0xffff;
+ ticks2 = readl(TIMER0_VA_BASE + TIMER_VALUE) & 0xffff;
do {
ticks1 = ticks2;
status = __raw_readl(VA_IC_BASE + VIC_IRQ_RAW_STATUS);
- ticks2 = timer0->TimerValue & 0xffff;
+ ticks2 = readl(TIMER0_VA_BASE + TIMER_VALUE) & 0xffff;
} while (ticks2 > ticks1);
/*
@@ -859,12 +847,10 @@ static unsigned long versatile_gettimeoffset(void)
*/
static irqreturn_t versatile_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
{
- volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
-
write_seqlock(&xtime_lock);
// ...clear the interrupt
- timer0->TimerClear = 1;
+ writel(1, TIMER0_VA_BASE + TIMER_INTCLR);
timer_tick(regs);
@@ -875,8 +861,8 @@ static irqreturn_t versatile_timer_interrupt(int irq, void *dev_id, struct pt_re
static struct irqaction versatile_timer_irq = {
.name = "Versatile Timer Tick",
- .flags = SA_INTERRUPT,
- .handler = versatile_timer_interrupt
+ .flags = SA_INTERRUPT | SA_TIMER,
+ .handler = versatile_timer_interrupt,
};
/*
@@ -884,31 +870,32 @@ static struct irqaction versatile_timer_irq = {
*/
static void __init versatile_timer_init(void)
{
- volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
- volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
- volatile TimerStruct_t *timer2 = (volatile TimerStruct_t *)TIMER2_VA_BASE;
- volatile TimerStruct_t *timer3 = (volatile TimerStruct_t *)TIMER3_VA_BASE;
+ u32 val;
/*
* set clock frequency:
* VERSATILE_REFCLK is 32KHz
* VERSATILE_TIMCLK is 1MHz
*/
- *(volatile unsigned int *)IO_ADDRESS(VERSATILE_SCTL_BASE) |=
- ((VERSATILE_TIMCLK << VERSATILE_TIMER1_EnSel) | (VERSATILE_TIMCLK << VERSATILE_TIMER2_EnSel) |
- (VERSATILE_TIMCLK << VERSATILE_TIMER3_EnSel) | (VERSATILE_TIMCLK << VERSATILE_TIMER4_EnSel));
+ val = readl(IO_ADDRESS(VERSATILE_SCTL_BASE));
+ writel((VERSATILE_TIMCLK << VERSATILE_TIMER1_EnSel) |
+ (VERSATILE_TIMCLK << VERSATILE_TIMER2_EnSel) |
+ (VERSATILE_TIMCLK << VERSATILE_TIMER3_EnSel) |
+ (VERSATILE_TIMCLK << VERSATILE_TIMER4_EnSel) | val,
+ IO_ADDRESS(VERSATILE_SCTL_BASE));
/*
* Initialise to a known state (all timers off)
*/
- timer0->TimerControl = 0;
- timer1->TimerControl = 0;
- timer2->TimerControl = 0;
- timer3->TimerControl = 0;
+ writel(0, TIMER0_VA_BASE + TIMER_CTRL);
+ writel(0, TIMER1_VA_BASE + TIMER_CTRL);
+ writel(0, TIMER2_VA_BASE + TIMER_CTRL);
+ writel(0, TIMER3_VA_BASE + TIMER_CTRL);
- timer0->TimerLoad = TIMER_RELOAD;
- timer0->TimerValue = TIMER_RELOAD;
- timer0->TimerControl = TIMER_CTRL | 0x40 | TIMER_CTRL_IE; /* periodic + IE */
+ writel(TIMER_RELOAD, TIMER0_VA_BASE + TIMER_LOAD);
+ writel(TIMER_RELOAD, TIMER0_VA_BASE + TIMER_VALUE);
+ writel(TIMER_DIVISOR | TIMER_CTRL_ENABLE | TIMER_CTRL_PERIODIC |
+ TIMER_CTRL_IE, TIMER0_VA_BASE + TIMER_CTRL);
/*
* Make irqs happen for the system timer
diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index c08710b1ff02..edffa47a4b2a 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -437,7 +437,7 @@ void __init paging_init(struct meminfo *mi, struct machine_desc *mdesc)
memtable_init(mi);
if (mdesc->map_io)
mdesc->map_io();
- flush_tlb_all();
+ local_flush_tlb_all();
/*
* initialise the zones within each node
@@ -522,6 +522,69 @@ static inline void free_area(unsigned long addr, unsigned long end, char *s)
printk(KERN_INFO "Freeing %s memory: %dK\n", s, size);
}
+static inline void
+free_memmap(int node, unsigned long start_pfn, unsigned long end_pfn)
+{
+ struct page *start_pg, *end_pg;
+ unsigned long pg, pgend;
+
+ /*
+ * Convert start_pfn/end_pfn to a struct page pointer.
+ */
+ start_pg = pfn_to_page(start_pfn);
+ end_pg = pfn_to_page(end_pfn);
+
+ /*
+ * Convert to physical addresses, and
+ * round start upwards and end downwards.
+ */
+ pg = PAGE_ALIGN(__pa(start_pg));
+ pgend = __pa(end_pg) & PAGE_MASK;
+
+ /*
+ * If there are free pages between these,
+ * free the section of the memmap array.
+ */
+ if (pg < pgend)
+ free_bootmem_node(NODE_DATA(node), pg, pgend - pg);
+}
+
+/*
+ * The mem_map array can get very big. Free the unused area of the memory map.
+ */
+static void __init free_unused_memmap_node(int node, struct meminfo *mi)
+{
+ unsigned long bank_start, prev_bank_end = 0;
+ unsigned int i;
+
+ /*
+ * [FIXME] This relies on each bank being in address order. This
+ * may not be the case, especially if the user has provided the
+ * information on the command line.
+ */
+ for (i = 0; i < mi->nr_banks; i++) {
+ if (mi->bank[i].size == 0 || mi->bank[i].node != node)
+ continue;
+
+ bank_start = mi->bank[i].start >> PAGE_SHIFT;
+ if (bank_start < prev_bank_end) {
+ printk(KERN_ERR "MEM: unordered memory banks. "
+ "Not freeing memmap.\n");
+ break;
+ }
+
+ /*
+ * If we had a previous bank, and there is a space
+ * between the current bank and the previous, free it.
+ */
+ if (prev_bank_end && prev_bank_end != bank_start)
+ free_memmap(node, prev_bank_end, bank_start);
+
+ prev_bank_end = (mi->bank[i].start +
+ mi->bank[i].size) >> PAGE_SHIFT;
+ }
+}
+
/*
* mem_init() marks the free areas in the mem_map and tells us how much
* memory is free. This is done after various parts of the system have
@@ -540,16 +603,12 @@ void __init mem_init(void)
max_mapnr = virt_to_page(high_memory) - mem_map;
#endif
- /*
- * We may have non-contiguous memory.
- */
- if (meminfo.nr_banks != 1)
- create_memmap_holes(&meminfo);
-
/* this will put all unused low memory onto the freelists */
for_each_online_node(node) {
pg_data_t *pgdat = NODE_DATA(node);
+ free_unused_memmap_node(node, &meminfo);
+
if (pgdat->node_spanned_pages != 0)
totalram_pages += free_all_bootmem_node(pgdat);
}
diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c
index 2c2b93d77d43..c3bd503b43a2 100644
--- a/arch/arm/mm/mm-armv.c
+++ b/arch/arm/mm/mm-armv.c
@@ -169,7 +169,14 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
memzero(new_pgd, FIRST_KERNEL_PGD_NR * sizeof(pgd_t));
+ /*
+ * Copy over the kernel and IO PGD entries
+ */
init_pgd = pgd_offset_k(0);
+ memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
+ (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
+
+ clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t));
if (!vectors_high()) {
/*
@@ -198,14 +205,6 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
spin_unlock(&mm->page_table_lock);
}
- /*
- * Copy over the kernel and IO PGD entries
- */
- memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
- (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
-
- clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t));
-
return new_pgd;
no_pte:
@@ -683,7 +682,7 @@ void __init memtable_init(struct meminfo *mi)
}
flush_cache_all();
- flush_tlb_all();
+ local_flush_tlb_all();
top_pmd = pmd_off_k(0xffff0000);
}
@@ -698,75 +697,3 @@ void __init iotable_init(struct map_desc *io_desc, int nr)
for (i = 0; i < nr; i++)
create_mapping(io_desc + i);
}
-
-static inline void
-free_memmap(int node, unsigned long start_pfn, unsigned long end_pfn)
-{
- struct page *start_pg, *end_pg;
- unsigned long pg, pgend;
-
- /*
- * Convert start_pfn/end_pfn to a struct page pointer.
- */
- start_pg = pfn_to_page(start_pfn);
- end_pg = pfn_to_page(end_pfn);
-
- /*
- * Convert to physical addresses, and
- * round start upwards and end downwards.
- */
- pg = PAGE_ALIGN(__pa(start_pg));
- pgend = __pa(end_pg) & PAGE_MASK;
-
- /*
- * If there are free pages between these,
- * free the section of the memmap array.
- */
- if (pg < pgend)
- free_bootmem_node(NODE_DATA(node), pg, pgend - pg);
-}
-
-static inline void free_unused_memmap_node(int node, struct meminfo *mi)
-{
- unsigned long bank_start, prev_bank_end = 0;
- unsigned int i;
-
- /*
- * [FIXME] This relies on each bank being in address order. This
- * may not be the case, especially if the user has provided the
- * information on the command line.
- */
- for (i = 0; i < mi->nr_banks; i++) {
- if (mi->bank[i].size == 0 || mi->bank[i].node != node)
- continue;
-
- bank_start = mi->bank[i].start >> PAGE_SHIFT;
- if (bank_start < prev_bank_end) {
- printk(KERN_ERR "MEM: unordered memory banks. "
- "Not freeing memmap.\n");
- break;
- }
-
- /*
- * If we had a previous bank, and there is a space
- * between the current bank and the previous, free it.
- */
- if (prev_bank_end && prev_bank_end != bank_start)
- free_memmap(node, prev_bank_end, bank_start);
-
- prev_bank_end = PAGE_ALIGN(mi->bank[i].start +
- mi->bank[i].size) >> PAGE_SHIFT;
- }
-}
-
-/*
- * The mem_map array can get very big. Free
- * the unused area of the memory map.
- */
-void __init create_memmap_holes(struct meminfo *mi)
-{
- int node;
-
- for_each_online_node(node)
- free_unused_memmap_node(node, mi);
-}
diff --git a/arch/arm/oprofile/Makefile b/arch/arm/oprofile/Makefile
index ba1a6e9f2b28..8ffb523e6c77 100644
--- a/arch/arm/oprofile/Makefile
+++ b/arch/arm/oprofile/Makefile
@@ -6,6 +6,6 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \
oprofilefs.o oprofile_stats.o \
timer_int.o )
-oprofile-y := $(DRIVER_OBJS) init.o
+oprofile-y := $(DRIVER_OBJS) init.o backtrace.o
oprofile-$(CONFIG_CPU_XSCALE) += common.o op_model_xscale.o
diff --git a/arch/arm/oprofile/backtrace.c b/arch/arm/oprofile/backtrace.c
new file mode 100644
index 000000000000..ec58d3e2eb8b
--- /dev/null
+++ b/arch/arm/oprofile/backtrace.c
@@ -0,0 +1,144 @@
+/*
+ * Arm specific backtracing code for oprofile
+ *
+ * Copyright 2005 Openedhand Ltd.
+ *
+ * Author: Richard Purdie
+ *
+ * Based on i386 oprofile backtrace code by John Levon, David Smith
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include
+#include
+#include
+#include
+#include
+
+
+/*
+ * The registers we're interested in are at the end of the variable
+ * length saved register structure. The fp points at the end of this
+ * structure so the address of this struct is:
+ * (struct frame_tail *)(xxx->fp)-1
+ */
+struct frame_tail {
+ struct frame_tail *fp;
+ unsigned long sp;
+ unsigned long lr;
+} __attribute__((packed));
+
+
+#ifdef CONFIG_FRAME_POINTER
+static struct frame_tail* kernel_backtrace(struct frame_tail *tail)
+{
+ oprofile_add_trace(tail->lr);
+
+ /* frame pointers should strictly progress back up the stack
+ * (towards higher addresses) */
+ if (tail >= tail->fp)
+ return NULL;
+
+ return tail->fp-1;
+}
+#endif
+
+static struct frame_tail* user_backtrace(struct frame_tail *tail)
+{
+ struct frame_tail buftail;
+
+ /* hardware pte might not be valid due to dirty/accessed bit emulation
+ * so we use copy_from_user and benefit from exception fixups */
+ if (copy_from_user(&buftail, tail, sizeof(struct frame_tail)))
+ return NULL;
+
+ oprofile_add_trace(buftail.lr);
+
+ /* frame pointers should strictly progress back up the stack
+ * (towards higher addresses) */
+ if (tail >= buftail.fp)
+ return NULL;
+
+ return buftail.fp-1;
+}
+
+/* Compare two addresses and see if they're on the same page */
+#define CMP_ADDR_EQUAL(x,y,offset) ((((unsigned long) x) >> PAGE_SHIFT) \
+ == ((((unsigned long) y) + offset) >> PAGE_SHIFT))
+
+/* check that the page(s) containing the frame tail are present */
+static int pages_present(struct frame_tail *tail)
+{
+ struct mm_struct * mm = current->mm;
+
+ if (!check_user_page_readable(mm, (unsigned long)tail))
+ return 0;
+
+ if (CMP_ADDR_EQUAL(tail, tail, 8))
+ return 1;
+
+ if (!check_user_page_readable(mm, ((unsigned long)tail) + 8))
+ return 0;
+
+ return 1;
+}
+
+/*
+ * | | /\ Higher addresses
+ * | |
+ * --------------- stack base (address of current_thread_info)
+ * | thread info |
+ * . .
+ * | stack |
+ * --------------- saved regs->ARM_fp value if valid (frame_tail address)
+ * . .
+ * --------------- struct pt_regs stored on stack (struct pt_regs *)
+ * | |
+ * . .
+ * | |
+ * --------------- %esp
+ * | |
+ * | | \/ Lower addresses
+ *
+ * Thus, &pt_regs <-> stack base restricts the valid(ish) fp values
+ */
+static int valid_kernel_stack(struct frame_tail *tail, struct pt_regs *regs)
+{
+ unsigned long tailaddr = (unsigned long)tail;
+ unsigned long stack = (unsigned long)regs;
+ unsigned long stack_base = (stack & ~(THREAD_SIZE - 1)) + THREAD_SIZE;
+
+ return (tailaddr > stack) && (tailaddr < stack_base);
+}
+
+void arm_backtrace(struct pt_regs const *regs, unsigned int depth)
+{
+ struct frame_tail *tail;
+ unsigned long last_address = 0;
+
+ tail = ((struct frame_tail *) regs->ARM_fp) - 1;
+
+ if (!user_mode(regs)) {
+
+#ifdef CONFIG_FRAME_POINTER
+ while (depth-- && tail && valid_kernel_stack(tail, regs)) {
+ tail = kernel_backtrace(tail);
+ }
+#endif
+ return;
+ }
+
+ while (depth-- && tail && !((unsigned long) tail & 3)) {
+ if ((!CMP_ADDR_EQUAL(last_address, tail, 0)
+ || !CMP_ADDR_EQUAL(last_address, tail, 8))
+ && !pages_present(tail))
+ return;
+ last_address = (unsigned long) tail;
+ tail = user_backtrace(tail);
+ }
+}
+
diff --git a/arch/arm/oprofile/init.c b/arch/arm/oprofile/init.c
index cce3d3015eb7..d315a3a86c86 100644
--- a/arch/arm/oprofile/init.c
+++ b/arch/arm/oprofile/init.c
@@ -20,6 +20,8 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
ret = pmu_init(ops, &op_xscale_spec);
#endif
+ ops->backtrace = arm_backtrace;
+
return ret;
}
diff --git a/arch/arm/oprofile/op_arm_model.h b/arch/arm/oprofile/op_arm_model.h
index 2d4caf4781ad..2148d07484b7 100644
--- a/arch/arm/oprofile/op_arm_model.h
+++ b/arch/arm/oprofile/op_arm_model.h
@@ -24,6 +24,8 @@ struct op_arm_model_spec {
extern struct op_arm_model_spec op_xscale_spec;
#endif
+extern void arm_backtrace(struct pt_regs * const regs, unsigned int depth);
+
extern int __init pmu_init(struct oprofile_operations *ops, struct op_arm_model_spec *spec);
extern void pmu_exit(void);
#endif /* OP_ARM_MODEL_H */
diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types
index 30c1dfbb052f..6d3a79e5fef8 100644
--- a/arch/arm/tools/mach-types
+++ b/arch/arm/tools/mach-types
@@ -6,7 +6,7 @@
# To add an entry into this database, please see Documentation/arm/README,
# or contact rmk@arm.linux.org.uk
#
-# Last update: Thu Mar 24 14:34:50 2005
+# Last update: Thu Jun 23 20:19:33 2005
#
# machine_is_xxx CONFIG_xxxx MACH_TYPE_xxx number
#
@@ -243,7 +243,7 @@ yoho ARCH_YOHO YOHO 231
jasper ARCH_JASPER JASPER 232
dsc25 ARCH_DSC25 DSC25 233
omap_innovator MACH_OMAP_INNOVATOR OMAP_INNOVATOR 234
-ramses ARCH_RAMSES RAMSES 235
+mnci ARCH_RAMSES RAMSES 235
s28x ARCH_S28X S28X 236
mport3 ARCH_MPORT3 MPORT3 237
pxa_eagle250 ARCH_PXA_EAGLE250 PXA_EAGLE250 238
@@ -323,7 +323,7 @@ nimbra29x ARCH_NIMBRA29X NIMBRA29X 311
nimbra210 ARCH_NIMBRA210 NIMBRA210 312
hhp_d95xx ARCH_HHP_D95XX HHP_D95XX 313
labarm ARCH_LABARM LABARM 314
-m825xx ARCH_M825XX M825XX 315
+comcerto ARCH_M825XX M825XX 315
m7100 SA1100_M7100 M7100 316
nipc2 ARCH_NIPC2 NIPC2 317
fu7202 ARCH_FU7202 FU7202 318
@@ -724,3 +724,66 @@ lpc22xx MACH_LPC22XX LPC22XX 715
omap_comet3 MACH_COMET3 COMET3 716
omap_comet4 MACH_COMET4 COMET4 717
csb625 MACH_CSB625 CSB625 718
+fortunet2 MACH_FORTUNET2 FORTUNET2 719
+s5h2200 MACH_S5H2200 S5H2200 720
+optorm920 MACH_OPTORM920 OPTORM920 721
+adsbitsyxb MACH_ADSBITSYXB ADSBITSYXB 722
+adssphere MACH_ADSSPHERE ADSSPHERE 723
+adsportal MACH_ADSPORTAL ADSPORTAL 724
+ln2410sbc MACH_LN2410SBC LN2410SBC 725
+cb3rufc MACH_CB3RUFC CB3RUFC 726
+mp2usb MACH_MP2USB MP2USB 727
+ntnp425c MACH_NTNP425C NTNP425C 728
+colibri MACH_COLIBRI COLIBRI 729
+pcm7220 MACH_PCM7220 PCM7220 730
+gateway7001 MACH_GATEWAY7001 GATEWAY7001 731
+pcm027 MACH_PCM027 PCM027 732
+cmpxa MACH_CMPXA CMPXA 733
+anubis MACH_ANUBIS ANUBIS 734
+ite8152 MACH_ITE8152 ITE8152 735
+lpc3xxx MACH_LPC3XXX LPC3XXX 736
+puppeteer MACH_PUPPETEER PUPPETEER 737
+vt001 MACH_MACH_VADATECH MACH_VADATECH 738
+e570 MACH_E570 E570 739
+x50 MACH_X50 X50 740
+recon MACH_RECON RECON 741
+xboardgp8 MACH_XBOARDGP8 XBOARDGP8 742
+fpic2 MACH_FPIC2 FPIC2 743
+akita MACH_AKITA AKITA 744
+a81 MACH_A81 A81 745
+svm_sc25x MACH_SVM_SC25X SVM_SC25X 746
+vt020 MACH_VADATECH020 VADATECH020 747
+tli MACH_TLI TLI 748
+edb9315lc MACH_EDB9315LC EDB9315LC 749
+passec MACH_PASSEC PASSEC 750
+ds_tiger MACH_DS_TIGER DS_TIGER 751
+e310 MACH_E310 E310 752
+e330 MACH_E330 E330 753
+rt3000 MACH_RT3000 RT3000 754
+nokia770 MACH_NOKIA770 NOKIA770 755
+pnx0106 MACH_PNX0106 PNX0106 756
+hx21xx MACH_HX21XX HX21XX 757
+faraday MACH_FARADAY FARADAY 758
+sbc9312 MACH_SBC9312 SBC9312 759
+batman MACH_BATMAN BATMAN 760
+jpd201 MACH_JPD201 JPD201 761
+mipsa MACH_MIPSA MIPSA 762
+kacom MACH_KACOM KACOM 763
+swarcocpu MACH_SWARCOCPU SWARCOCPU 764
+swarcodsl MACH_SWARCODSL SWARCODSL 765
+blueangel MACH_BLUEANGEL BLUEANGEL 766
+hairygrama MACH_HAIRYGRAMA HAIRYGRAMA 767
+banff MACH_BANFF BANFF 768
+carmeva MACH_CARMEVA CARMEVA 769
+sam255 MACH_SAM255 SAM255 770
+ppm10 MACH_PPM10 PPM10 771
+edb9315a MACH_EDB9315A EDB9315A 772
+sunset MACH_SUNSET SUNSET 773
+stargate2 MACH_STARGATE2 STARGATE2 774
+intelmote2 MACH_INTELMOTE2 INTELMOTE2 775
+trizeps4 MACH_TRIZEPS4 TRIZEPS4 776
+mainstone2 MACH_MAINSTONE2 MAINSTONE2 777
+ez_ixp42x MACH_EZ_IXP42X EZ_IXP42X 778
+tapwave_zodiac MACH_TAPWAVE_ZODIAC TAPWAVE_ZODIAC 779
+universalmeter MACH_UNIVERSALMETER UNIVERSALMETER 780
+hicoarm9 MACH_HICOARM9 HICOARM9 781
diff --git a/arch/arm/vfp/vfp.h b/arch/arm/vfp/vfp.h
index 55a02bc994a3..4b97950984e9 100644
--- a/arch/arm/vfp/vfp.h
+++ b/arch/arm/vfp/vfp.h
@@ -117,7 +117,13 @@ static inline u64 vfp_estimate_div128to64(u64 nh, u64 nl, u64 m)
if (nh >= m)
return ~0ULL;
mh = m >> 32;
- z = (mh << 32 <= nh) ? 0xffffffff00000000ULL : (nh / mh) << 32;
+ if (mh << 32 <= nh) {
+ z = 0xffffffff00000000ULL;
+ } else {
+ z = nh;
+ do_div(z, mh);
+ z <<= 32;
+ }
mul64to128(&termh, &terml, m, z);
sub128(&remh, &reml, nh, nl, termh, terml);
ml = m << 32;
@@ -126,7 +132,12 @@ static inline u64 vfp_estimate_div128to64(u64 nh, u64 nl, u64 m)
add128(&remh, &reml, remh, reml, mh, ml);
}
remh = (remh << 32) | (reml >> 32);
- z |= (mh << 32 <= remh) ? 0xffffffff : remh / mh;
+ if (mh << 32 <= remh) {
+ z |= 0xffffffff;
+ } else {
+ do_div(remh, mh);
+ z |= remh;
+ }
return z;
}
diff --git a/arch/arm/vfp/vfpdouble.c b/arch/arm/vfp/vfpdouble.c
index fa3053e84db5..b801cd66b6ea 100644
--- a/arch/arm/vfp/vfpdouble.c
+++ b/arch/arm/vfp/vfpdouble.c
@@ -32,6 +32,8 @@
*/
#include
#include
+
+#include
#include
#include
diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index 3aeedd2afc70..22f3da4e0829 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -89,7 +89,7 @@ void vfp_raise_sigfpe(unsigned int sicode, struct pt_regs *regs)
current->thread.error_code = 0;
current->thread.trap_no = 6;
- force_sig_info(SIGFPE, &info, current);
+ send_sig_info(SIGFPE, &info, current);
}
static void vfp_panic(char *reason)
diff --git a/arch/arm/vfp/vfpsingle.c b/arch/arm/vfp/vfpsingle.c
index 6849fe35cb2e..14dd696ddeb1 100644
--- a/arch/arm/vfp/vfpsingle.c
+++ b/arch/arm/vfp/vfpsingle.c
@@ -32,6 +32,8 @@
*/
#include
#include
+
+#include
#include
#include
@@ -303,7 +305,11 @@ u32 vfp_estimate_sqrt_significand(u32 exponent, u32 significand)
if (z <= a)
return (s32)a >> 1;
}
- return (u32)(((u64)a << 31) / z) + (z >> 1);
+ {
+ u64 v = (u64)a << 31;
+ do_div(v, z);
+ return v + (z >> 1);
+ }
}
static u32 vfp_single_fsqrt(int sd, int unused, s32 m, u32 fpscr)
@@ -1107,7 +1113,11 @@ static u32 vfp_single_fdiv(int sd, int sn, s32 m, u32 fpscr)
vsn.significand >>= 1;
vsd.exponent++;
}
- vsd.significand = ((u64)vsn.significand << 32) / vsm.significand;
+ {
+ u64 significand = (u64)vsn.significand << 32;
+ do_div(significand, vsm.significand);
+ vsd.significand = significand;
+ }
if ((vsd.significand & 0x3f) == 0)
vsd.significand |= ((u64)vsm.significand * vsd.significand != (u64)vsn.significand << 32);
diff --git a/arch/i386/boot/tools/build.c b/arch/i386/boot/tools/build.c
index 4a17956512e1..6835f6d47c31 100644
--- a/arch/i386/boot/tools/build.c
+++ b/arch/i386/boot/tools/build.c
@@ -70,7 +70,8 @@ void usage(void)
int main(int argc, char ** argv)
{
- unsigned int i, c, sz, setup_sectors;
+ unsigned int i, sz, setup_sectors;
+ int c;
u32 sys_size;
byte major_root, minor_root;
struct stat sb;
diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 9f63ae0f404b..b7808a89d945 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -159,9 +159,15 @@ char *__acpi_map_table(unsigned long phys, unsigned long size)
#endif
#ifdef CONFIG_PCI_MMCONFIG
-static int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size)
+/* The physical address of the MMCONFIG aperture. Set from ACPI tables. */
+struct acpi_table_mcfg_config *pci_mmcfg_config;
+int pci_mmcfg_config_num;
+
+int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size)
{
struct acpi_table_mcfg *mcfg;
+ unsigned long i;
+ int config_size;
if (!phys_addr || !size)
return -EINVAL;
@@ -172,18 +178,38 @@ static int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size)
return -ENODEV;
}
- if (mcfg->base_reserved) {
- printk(KERN_ERR PREFIX "MMCONFIG not in low 4GB of memory\n");
+ /* how many config structures do we have */
+ pci_mmcfg_config_num = 0;
+ i = size - sizeof(struct acpi_table_mcfg);
+ while (i >= sizeof(struct acpi_table_mcfg_config)) {
+ ++pci_mmcfg_config_num;
+ i -= sizeof(struct acpi_table_mcfg_config);
+ };
+ if (pci_mmcfg_config_num == 0) {
+ printk(KERN_ERR PREFIX "MMCONFIG has no entries\n");
return -ENODEV;
}
- pci_mmcfg_base_addr = mcfg->base_address;
+ config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config);
+ pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL);
+ if (!pci_mmcfg_config) {
+ printk(KERN_WARNING PREFIX
+ "No memory for MCFG config tables\n");
+ return -ENOMEM;
+ }
+
+ memcpy(pci_mmcfg_config, &mcfg->config, config_size);
+ for (i = 0; i < pci_mmcfg_config_num; ++i) {
+ if (mcfg->config[i].base_reserved) {
+ printk(KERN_ERR PREFIX
+ "MMCONFIG not in low 4GB of memory\n");
+ return -ENODEV;
+ }
+ }
return 0;
}
-#else
-#define acpi_parse_mcfg NULL
-#endif /* !CONFIG_PCI_MMCONFIG */
+#endif /* CONFIG_PCI_MMCONFIG */
#ifdef CONFIG_X86_LOCAL_APIC
static int __init
@@ -507,6 +533,22 @@ acpi_unmap_lsapic(int cpu)
EXPORT_SYMBOL(acpi_unmap_lsapic);
#endif /* CONFIG_ACPI_HOTPLUG_CPU */
+int
+acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base)
+{
+ /* TBD */
+ return -EINVAL;
+}
+EXPORT_SYMBOL(acpi_register_ioapic);
+
+int
+acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base)
+{
+ /* TBD */
+ return -EINVAL;
+}
+EXPORT_SYMBOL(acpi_unregister_ioapic);
+
static unsigned long __init
acpi_scan_rsdp (
unsigned long start,
@@ -1123,7 +1165,6 @@ int __init acpi_boot_init(void)
acpi_process_madt();
acpi_table_parse(ACPI_HPET, acpi_parse_hpet);
- acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg);
return 0;
}
diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 3762f6b35ab2..fc8b17521761 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -127,48 +127,23 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
regs->eip = (unsigned long)&p->ainsn.insn;
}
-struct task_struct *arch_get_kprobe_task(void *ptr)
-{
- return ((struct thread_info *) (((unsigned long) ptr) &
- (~(THREAD_SIZE -1))))->task;
-}
-
void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
{
unsigned long *sara = (unsigned long *)®s->esp;
- struct kretprobe_instance *ri;
- static void *orig_ret_addr;
+ struct kretprobe_instance *ri;
- /*
- * Save the return address when the return probe hits
- * the first time, and use it to populate the (krprobe
- * instance)->ret_addr for subsequent return probes at
- * the same addrress since stack address would have
- * the kretprobe_trampoline by then.
- */
- if (((void*) *sara) != kretprobe_trampoline)
- orig_ret_addr = (void*) *sara;
+ if ((ri = get_free_rp_inst(rp)) != NULL) {
+ ri->rp = rp;
+ ri->task = current;
+ ri->ret_addr = (kprobe_opcode_t *) *sara;
- if ((ri = get_free_rp_inst(rp)) != NULL) {
- ri->rp = rp;
- ri->stack_addr = sara;
- ri->ret_addr = orig_ret_addr;
- add_rp_inst(ri);
/* Replace the return addr with trampoline addr */
*sara = (unsigned long) &kretprobe_trampoline;
- } else {
- rp->nmissed++;
- }
-}
-void arch_kprobe_flush_task(struct task_struct *tk)
-{
- struct kretprobe_instance *ri;
- while ((ri = get_rp_inst_tsk(tk)) != NULL) {
- *((unsigned long *)(ri->stack_addr)) =
- (unsigned long) ri->ret_addr;
- recycle_rp_inst(ri);
- }
+ add_rp_inst(ri);
+ } else {
+ rp->nmissed++;
+ }
}
/*
@@ -286,36 +261,59 @@ no_kprobe:
*/
int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
{
- struct task_struct *tsk;
- struct kretprobe_instance *ri;
- struct hlist_head *head;
- struct hlist_node *node;
- unsigned long *sara = ((unsigned long *) ®s->esp) - 1;
+ struct kretprobe_instance *ri = NULL;
+ struct hlist_head *head;
+ struct hlist_node *node, *tmp;
+ unsigned long orig_ret_address = 0;
+ unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
- tsk = arch_get_kprobe_task(sara);
- head = kretprobe_inst_table_head(tsk);
+ head = kretprobe_inst_table_head(current);
- hlist_for_each_entry(ri, node, head, hlist) {
- if (ri->stack_addr == sara && ri->rp) {
- if (ri->rp->handler)
- ri->rp->handler(ri, regs);
- }
- }
- return 0;
-}
+ /*
+ * It is possible to have multiple instances associated with a given
+ * task either because an multiple functions in the call path
+ * have a return probe installed on them, and/or more then one return
+ * return probe was registered for a target function.
+ *
+ * We can handle this because:
+ * - instances are always inserted at the head of the list
+ * - when multiple return probes are registered for the same
+ * function, the first instance's ret_addr will point to the
+ * real return address, and all the rest will point to
+ * kretprobe_trampoline
+ */
+ hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+ if (ri->task != current)
+ /* another task is sharing our hash bucket */
+ continue;
-void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
- unsigned long flags)
-{
- struct kretprobe_instance *ri;
- /* RA already popped */
- unsigned long *sara = ((unsigned long *)®s->esp) - 1;
+ if (ri->rp && ri->rp->handler)
+ ri->rp->handler(ri, regs);
- while ((ri = get_rp_inst(sara))) {
- regs->eip = (unsigned long)ri->ret_addr;
+ orig_ret_address = (unsigned long)ri->ret_addr;
recycle_rp_inst(ri);
+
+ if (orig_ret_address != trampoline_address)
+ /*
+ * This is the real return address. Any other
+ * instances associated with this task are for
+ * other calls deeper on the call stack
+ */
+ break;
}
- regs->eflags &= ~TF_MASK;
+
+ BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+ regs->eip = orig_ret_address;
+
+ unlock_kprobes();
+ preempt_enable_no_resched();
+
+ /*
+ * By returning a non-zero value, we are telling
+ * kprobe_handler() that we have handled unlocking
+ * and re-enabling preemption.
+ */
+ return 1;
}
/*
@@ -403,8 +401,7 @@ static inline int post_kprobe_handler(struct pt_regs *regs)
current_kprobe->post_handler(current_kprobe, regs, 0);
}
- if (current_kprobe->post_handler != trampoline_post_handler)
- resume_execution(current_kprobe, regs);
+ resume_execution(current_kprobe, regs);
regs->eflags |= kprobe_saved_eflags;
/*Restore back the original saved kprobes variables and continue. */
@@ -534,3 +531,13 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
}
return 0;
}
+
+static struct kprobe trampoline_p = {
+ .addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+ .pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+ return register_kprobe(&trampoline_p);
+}
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 5f8cfa6b7940..ba243a4cc119 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -616,6 +616,33 @@ handle_io_bitmap(struct thread_struct *next, struct tss_struct *tss)
tss->io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY;
}
+/*
+ * This function selects if the context switch from prev to next
+ * has to tweak the TSC disable bit in the cr4.
+ */
+static inline void disable_tsc(struct task_struct *prev_p,
+ struct task_struct *next_p)
+{
+ struct thread_info *prev, *next;
+
+ /*
+ * gcc should eliminate the ->thread_info dereference if
+ * has_secure_computing returns 0 at compile time (SECCOMP=n).
+ */
+ prev = prev_p->thread_info;
+ next = next_p->thread_info;
+
+ if (has_secure_computing(prev) || has_secure_computing(next)) {
+ /* slow path here */
+ if (has_secure_computing(prev) &&
+ !has_secure_computing(next)) {
+ write_cr4(read_cr4() & ~X86_CR4_TSD);
+ } else if (!has_secure_computing(prev) &&
+ has_secure_computing(next))
+ write_cr4(read_cr4() | X86_CR4_TSD);
+ }
+}
+
/*
* switch_to(x,yn) should switch tasks from x to y.
*
@@ -695,6 +722,8 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr))
handle_io_bitmap(next, tss);
+ disable_tsc(prev_p, next_p);
+
return prev_p;
}
diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index 442a6e937b19..3db9a04aec6e 100644
--- a/arch/i386/kernel/syscall_table.S
+++ b/arch/i386/kernel/syscall_table.S
@@ -289,3 +289,5 @@ ENTRY(sys_call_table)
.long sys_add_key
.long sys_request_key
.long sys_keyctl
+ .long sys_ioprio_set
+ .long sys_ioprio_get /* 290 */
diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c
index 720975e1af50..87325263cd4f 100644
--- a/arch/i386/pci/common.c
+++ b/arch/i386/pci/common.c
@@ -25,7 +25,8 @@ unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 |
int pci_routeirq;
int pcibios_last_bus = -1;
-struct pci_bus *pci_root_bus = NULL;
+unsigned long pirq_table_addr;
+struct pci_bus *pci_root_bus;
struct pci_raw_ops *raw_pci_ops;
static int pci_read(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 *value)
@@ -133,7 +134,7 @@ struct pci_bus * __devinit pcibios_scan_root(int busnum)
printk("PCI: Probing PCI hardware (bus %02x)\n", busnum);
- return pci_scan_bus(busnum, &pci_root_ops, NULL);
+ return pci_scan_bus_parented(NULL, busnum, &pci_root_ops, NULL);
}
extern u8 pci_cache_line_size;
@@ -188,6 +189,9 @@ char * __devinit pcibios_setup(char *str)
} else if (!strcmp(str, "biosirq")) {
pci_probe |= PCI_BIOS_IRQ_SCAN;
return NULL;
+ } else if (!strncmp(str, "pirqaddr=", 9)) {
+ pirq_table_addr = simple_strtoul(str+9, NULL, 0);
+ return NULL;
}
#endif
#ifdef CONFIG_PCI_DIRECT
diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c
index 83458f81e661..78ca1ecbb907 100644
--- a/arch/i386/pci/irq.c
+++ b/arch/i386/pci/irq.c
@@ -57,6 +57,35 @@ struct irq_router_handler {
int (*pcibios_enable_irq)(struct pci_dev *dev) = NULL;
+/*
+ * Check passed address for the PCI IRQ Routing Table signature
+ * and perform checksum verification.
+ */
+
+static inline struct irq_routing_table * pirq_check_routing_table(u8 *addr)
+{
+ struct irq_routing_table *rt;
+ int i;
+ u8 sum;
+
+ rt = (struct irq_routing_table *) addr;
+ if (rt->signature != PIRQ_SIGNATURE ||
+ rt->version != PIRQ_VERSION ||
+ rt->size % 16 ||
+ rt->size < sizeof(struct irq_routing_table))
+ return NULL;
+ sum = 0;
+ for (i=0; i < rt->size; i++)
+ sum += addr[i];
+ if (!sum) {
+ DBG("PCI: Interrupt Routing Table found at 0x%p\n", rt);
+ return rt;
+ }
+ return NULL;
+}
+
+
+
/*
* Search 0xf0000 -- 0xfffff for the PCI IRQ Routing Table.
*/
@@ -65,23 +94,17 @@ static struct irq_routing_table * __init pirq_find_routing_table(void)
{
u8 *addr;
struct irq_routing_table *rt;
- int i;
- u8 sum;
- for(addr = (u8 *) __va(0xf0000); addr < (u8 *) __va(0x100000); addr += 16) {
- rt = (struct irq_routing_table *) addr;
- if (rt->signature != PIRQ_SIGNATURE ||
- rt->version != PIRQ_VERSION ||
- rt->size % 16 ||
- rt->size < sizeof(struct irq_routing_table))
- continue;
- sum = 0;
- for(i=0; isize; i++)
- sum += addr[i];
- if (!sum) {
- DBG("PCI: Interrupt Routing Table found at 0x%p\n", rt);
+ if (pirq_table_addr) {
+ rt = pirq_check_routing_table((u8 *) __va(pirq_table_addr));
+ if (rt)
+ return rt;
+ printk(KERN_WARNING "PCI: PIRQ table NOT found at pirqaddr\n");
+ }
+ for(addr = (u8 *) __va(0xf0000); addr < (u8 *) __va(0x100000); addr += 16) {
+ rt = pirq_check_routing_table(addr);
+ if (rt)
return rt;
- }
}
return NULL;
}
diff --git a/arch/i386/pci/legacy.c b/arch/i386/pci/legacy.c
index 1492e3753869..149a9588c256 100644
--- a/arch/i386/pci/legacy.c
+++ b/arch/i386/pci/legacy.c
@@ -45,6 +45,8 @@ static int __init pci_legacy_init(void)
printk("PCI: Probing PCI hardware\n");
pci_root_bus = pcibios_scan_root(0);
+ if (pci_root_bus)
+ pci_bus_add_devices(pci_root_bus);
pcibios_fixup_peer_bridges();
diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 021a50aa51f4..60f0e7a1162a 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -11,11 +11,9 @@
#include
#include
+#include
#include "pci.h"
-/* The physical address of the MMCONFIG aperture. Set from ACPI tables. */
-u32 pci_mmcfg_base_addr;
-
#define mmcfg_virt_addr ((void __iomem *) fix_to_virt(FIX_PCIE_MCFG))
/* The base address of the last MMCONFIG device accessed */
@@ -24,10 +22,31 @@ static u32 mmcfg_last_accessed_device;
/*
* Functions for accessing PCI configuration space with MMCONFIG accesses
*/
-
-static inline void pci_exp_set_dev_base(int bus, int devfn)
+static u32 get_base_addr(unsigned int seg, int bus)
{
- u32 dev_base = pci_mmcfg_base_addr | (bus << 20) | (devfn << 12);
+ int cfg_num = -1;
+ struct acpi_table_mcfg_config *cfg;
+
+ while (1) {
+ ++cfg_num;
+ if (cfg_num >= pci_mmcfg_config_num) {
+ /* something bad is going on, no cfg table is found. */
+ /* so we fall back to the old way we used to do this */
+ /* and just rely on the first entry to be correct. */
+ return pci_mmcfg_config[0].base_address;
+ }
+ cfg = &pci_mmcfg_config[cfg_num];
+ if (cfg->pci_segment_group_number != seg)
+ continue;
+ if ((cfg->start_bus_number <= bus) &&
+ (cfg->end_bus_number >= bus))
+ return cfg->base_address;
+ }
+}
+
+static inline void pci_exp_set_dev_base(unsigned int seg, int bus, int devfn)
+{
+ u32 dev_base = get_base_addr(seg, bus) | (bus << 20) | (devfn << 12);
if (dev_base != mmcfg_last_accessed_device) {
mmcfg_last_accessed_device = dev_base;
set_fixmap_nocache(FIX_PCIE_MCFG, dev_base);
@@ -44,7 +63,7 @@ static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
spin_lock_irqsave(&pci_config_lock, flags);
- pci_exp_set_dev_base(bus, devfn);
+ pci_exp_set_dev_base(seg, bus, devfn);
switch (len) {
case 1:
@@ -73,7 +92,7 @@ static int pci_mmcfg_write(unsigned int seg, unsigned int bus,
spin_lock_irqsave(&pci_config_lock, flags);
- pci_exp_set_dev_base(bus, devfn);
+ pci_exp_set_dev_base(seg, bus, devfn);
switch (len) {
case 1:
@@ -101,7 +120,11 @@ static int __init pci_mmcfg_init(void)
{
if ((pci_probe & PCI_PROBE_MMCONF) == 0)
goto out;
- if (!pci_mmcfg_base_addr)
+
+ acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg);
+ if ((pci_mmcfg_config_num == 0) ||
+ (pci_mmcfg_config == NULL) ||
+ (pci_mmcfg_config[0].base_address == 0))
goto out;
/* Kludge for now. Don't use mmconfig on AMD systems because
diff --git a/arch/i386/pci/numa.c b/arch/i386/pci/numa.c
index 9e3695461899..adbe17a38f6f 100644
--- a/arch/i386/pci/numa.c
+++ b/arch/i386/pci/numa.c
@@ -115,6 +115,8 @@ static int __init pci_numa_init(void)
return 0;
pci_root_bus = pcibios_scan_root(0);
+ if (pci_root_bus)
+ pci_bus_add_devices(pci_root_bus);
if (num_online_nodes() > 1)
for_each_online_node(quad) {
if (quad == 0)
diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h
index a8fc80ca69f3..a80f0f55ff51 100644
--- a/arch/i386/pci/pci.h
+++ b/arch/i386/pci/pci.h
@@ -27,6 +27,7 @@
#define PCI_ASSIGN_ALL_BUSSES 0x4000
extern unsigned int pci_probe;
+extern unsigned long pirq_table_addr;
/* pci-i386.c */
diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index 487d2e36b0a6..c05613980300 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -99,7 +99,7 @@ CONFIG_ACPI_DEALLOCATE_IRQ=y
# Firmware Drivers
#
CONFIG_EFI_VARS=y
-# CONFIG_EFI_PCDP is not set
+CONFIG_EFI_PCDP=y
CONFIG_BINFMT_ELF=y
# CONFIG_BINFMT_MISC is not set
@@ -650,7 +650,7 @@ CONFIG_MMTIMER=y
#
# Console display driver support
#
-# CONFIG_VGA_CONSOLE is not set
+CONFIG_VGA_CONSOLE=y
CONFIG_DUMMY_CONSOLE=y
#
diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig
index 47f45341ac62..73454eee26f1 100644
--- a/arch/ia64/configs/tiger_defconfig
+++ b/arch/ia64/configs/tiger_defconfig
@@ -1,7 +1,7 @@
#
# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-20050621
-# Tue Jun 21 14:03:24 2005
+# Linux kernel version: 2.6.13-rc1-20050629
+# Wed Jun 29 15:28:12 2005
#
#
@@ -80,18 +80,29 @@ CONFIG_MCKINLEY=y
# CONFIG_IA64_PAGE_SIZE_8KB is not set
CONFIG_IA64_PAGE_SIZE_16KB=y
# CONFIG_IA64_PAGE_SIZE_64KB is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
CONFIG_IA64_L1_CACHE_SHIFT=7
# CONFIG_NUMA is not set
CONFIG_VIRTUAL_MEM_MAP=y
CONFIG_HOLES_IN_ZONE=y
CONFIG_IA64_CYCLONE=y
CONFIG_IOSAPIC=y
+# CONFIG_IA64_SGI_SN_XP is not set
CONFIG_FORCE_MAX_ZONEORDER=18
CONFIG_SMP=y
CONFIG_NR_CPUS=4
CONFIG_HOTPLUG_CPU=y
# CONFIG_SCHED_SMT is not set
# CONFIG_PREEMPT is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
CONFIG_HAVE_DEC_LOCK=y
CONFIG_IA32_SUPPORT=y
CONFIG_COMPAT=y
@@ -257,6 +268,7 @@ CONFIG_BLK_DEV_CMD64X=y
# CONFIG_BLK_DEV_HPT366 is not set
# CONFIG_BLK_DEV_SC1200 is not set
CONFIG_BLK_DEV_PIIX=y
+# CONFIG_BLK_DEV_IT821X is not set
# CONFIG_BLK_DEV_NS87415 is not set
# CONFIG_BLK_DEV_PDC202XX_OLD is not set
# CONFIG_BLK_DEV_PDC202XX_NEW is not set
@@ -395,6 +407,7 @@ CONFIG_UNIX=y
CONFIG_INET=y
CONFIG_IP_MULTICAST=y
# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
# CONFIG_IP_PNP is not set
# CONFIG_NET_IPIP is not set
# CONFIG_NET_IPGRE is not set
@@ -407,6 +420,8 @@ CONFIG_SYN_COOKIES=y
# CONFIG_INET_TUNNEL is not set
CONFIG_IP_TCPDIAG=y
# CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
# CONFIG_IPV6 is not set
# CONFIG_NETFILTER is not set
@@ -598,9 +613,7 @@ CONFIG_GAMEPORT=m
# CONFIG_GAMEPORT_NS558 is not set
# CONFIG_GAMEPORT_L4 is not set
# CONFIG_GAMEPORT_EMU10K1 is not set
-# CONFIG_GAMEPORT_VORTEX is not set
# CONFIG_GAMEPORT_FM801 is not set
-# CONFIG_GAMEPORT_CS461X is not set
#
# Character devices
@@ -629,7 +642,6 @@ CONFIG_SERIAL_8250_NR_UARTS=6
CONFIG_SERIAL_8250_EXTENDED=y
CONFIG_SERIAL_8250_SHARE_IRQ=y
# CONFIG_SERIAL_8250_DETECT_IRQ is not set
-# CONFIG_SERIAL_8250_MULTIPORT is not set
# CONFIG_SERIAL_8250_RSA is not set
#
@@ -743,6 +755,7 @@ CONFIG_USB_DEVICEFS=y
CONFIG_USB_EHCI_HCD=m
# CONFIG_USB_EHCI_SPLIT_ISO is not set
# CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
CONFIG_USB_OHCI_HCD=m
# CONFIG_USB_OHCI_BIG_ENDIAN is not set
CONFIG_USB_OHCI_LITTLE_ENDIAN=y
@@ -779,9 +792,11 @@ CONFIG_USB_HIDINPUT=y
# CONFIG_USB_HIDDEV is not set
# CONFIG_USB_AIPTEK is not set
# CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
# CONFIG_USB_KBTAB is not set
# CONFIG_USB_POWERMATE is not set
# CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
# CONFIG_USB_EGALAX is not set
# CONFIG_USB_XPAD is not set
# CONFIG_USB_ATI_REMOTE is not set
@@ -838,7 +853,7 @@ CONFIG_USB_HIDINPUT=y
# CONFIG_USB_TEST is not set
#
-# USB ATM/DSL drivers
+# USB DSL modem support
#
#
@@ -856,6 +871,10 @@ CONFIG_USB_HIDINPUT=y
#
# CONFIG_INFINIBAND is not set
+#
+# SN Devices
+#
+
#
# File systems
#
@@ -863,6 +882,7 @@ CONFIG_EXT2_FS=y
CONFIG_EXT2_FS_XATTR=y
CONFIG_EXT2_FS_POSIX_ACL=y
CONFIG_EXT2_FS_SECURITY=y
+# CONFIG_EXT2_FS_XIP is not set
CONFIG_EXT3_FS=y
CONFIG_EXT3_FS_XATTR=y
CONFIG_EXT3_FS_POSIX_ACL=y
@@ -922,7 +942,6 @@ CONFIG_NTFS_FS=m
CONFIG_PROC_FS=y
CONFIG_PROC_KCORE=y
CONFIG_SYSFS=y
-# CONFIG_DEVFS_FS is not set
# CONFIG_DEVPTS_FS_XATTR is not set
CONFIG_TMPFS=y
CONFIG_TMPFS_XATTR=y
@@ -953,15 +972,18 @@ CONFIG_RAMFS=y
#
CONFIG_NFS_FS=m
CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
CONFIG_NFS_V4=y
CONFIG_NFS_DIRECTIO=y
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
CONFIG_NFSD_V4=y
CONFIG_NFSD_TCP=y
CONFIG_LOCKD=m
CONFIG_LOCKD_V4=y
CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
CONFIG_SUNRPC=m
CONFIG_SUNRPC_GSS=m
CONFIG_RPCSEC_GSS_KRB5=m
@@ -1069,6 +1091,7 @@ CONFIG_LOG_BUF_SHIFT=20
# CONFIG_DEBUG_KOBJECT is not set
# CONFIG_DEBUG_INFO is not set
# CONFIG_DEBUG_FS is not set
+# CONFIG_KPROBES is not set
CONFIG_IA64_GRANULE_16MB=y
# CONFIG_IA64_GRANULE_64MB is not set
# CONFIG_IA64_PRINT_HAZARDS is not set
@@ -1090,7 +1113,7 @@ CONFIG_CRYPTO=y
# CONFIG_CRYPTO_HMAC is not set
# CONFIG_CRYPTO_NULL is not set
# CONFIG_CRYPTO_MD4 is not set
-CONFIG_CRYPTO_MD5=m
+CONFIG_CRYPTO_MD5=y
# CONFIG_CRYPTO_SHA1 is not set
# CONFIG_CRYPTO_SHA256 is not set
# CONFIG_CRYPTO_SHA512 is not set
diff --git a/arch/ia64/configs/zx1_defconfig b/arch/ia64/configs/zx1_defconfig
index 21d6f9bab5e9..b7755e4436d2 100644
--- a/arch/ia64/configs/zx1_defconfig
+++ b/arch/ia64/configs/zx1_defconfig
@@ -1,7 +1,7 @@
#
# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.10
-# Wed Dec 29 09:05:48 2004
+# Linux kernel version: 2.6.13-rc1-20050629
+# Wed Jun 29 15:31:11 2005
#
#
@@ -12,6 +12,7 @@ CONFIG_EXPERIMENTAL=y
CONFIG_BROKEN=y
CONFIG_BROKEN_ON_SMP=y
CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
#
# General setup
@@ -24,23 +25,26 @@ CONFIG_BSD_PROCESS_ACCT=y
# CONFIG_BSD_PROCESS_ACCT_V3 is not set
CONFIG_SYSCTL=y
# CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=17
CONFIG_HOTPLUG=y
CONFIG_KOBJECT_UEVENT=y
# CONFIG_IKCONFIG is not set
+# CONFIG_CPUSETS is not set
# CONFIG_EMBEDDED is not set
CONFIG_KALLSYMS=y
# CONFIG_KALLSYMS_ALL is not set
# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
CONFIG_FUTEX=y
CONFIG_EPOLL=y
-# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
CONFIG_SHMEM=y
CONFIG_CC_ALIGN_FUNCTIONS=0
CONFIG_CC_ALIGN_LABELS=0
CONFIG_CC_ALIGN_LOOPS=0
CONFIG_CC_ALIGN_JUMPS=0
# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
#
# Loadable module support
@@ -59,12 +63,15 @@ CONFIG_IA64=y
CONFIG_64BIT=y
CONFIG_MMU=y
CONFIG_RWSEM_XCHGADD_ALGORITHM=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
CONFIG_TIME_INTERPOLATION=y
CONFIG_EFI=y
CONFIG_GENERIC_IOMAP=y
+CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
# CONFIG_IA64_GENERIC is not set
# CONFIG_IA64_DIG is not set
CONFIG_IA64_HP_ZX1=y
+# CONFIG_IA64_HP_ZX1_SWIOTLB is not set
# CONFIG_IA64_SGI_SN2 is not set
# CONFIG_IA64_HP_SIM is not set
# CONFIG_ITANIUM is not set
@@ -73,22 +80,36 @@ CONFIG_MCKINLEY=y
# CONFIG_IA64_PAGE_SIZE_8KB is not set
CONFIG_IA64_PAGE_SIZE_16KB=y
# CONFIG_IA64_PAGE_SIZE_64KB is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
CONFIG_IA64_L1_CACHE_SHIFT=7
# CONFIG_NUMA is not set
CONFIG_VIRTUAL_MEM_MAP=y
+CONFIG_HOLES_IN_ZONE=y
# CONFIG_IA64_CYCLONE is not set
CONFIG_IOSAPIC=y
+# CONFIG_IA64_SGI_SN_XP is not set
CONFIG_FORCE_MAX_ZONEORDER=18
CONFIG_SMP=y
CONFIG_NR_CPUS=16
# CONFIG_HOTPLUG_CPU is not set
+# CONFIG_SCHED_SMT is not set
# CONFIG_PREEMPT is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
CONFIG_HAVE_DEC_LOCK=y
CONFIG_IA32_SUPPORT=y
CONFIG_COMPAT=y
CONFIG_IA64_MCA_RECOVERY=y
CONFIG_PERFMON=y
CONFIG_IA64_PALINFO=y
+CONFIG_ACPI_DEALLOCATE_IRQ=y
#
# Firmware Drivers
@@ -120,6 +141,7 @@ CONFIG_ACPI_BUS=y
CONFIG_ACPI_POWER=y
CONFIG_ACPI_PCI=y
CONFIG_ACPI_SYSTEM=y
+# CONFIG_ACPI_CONTAINER is not set
#
# Bus options (PCI, PCMCIA)
@@ -129,6 +151,7 @@ CONFIG_PCI_DOMAINS=y
# CONFIG_PCI_MSI is not set
CONFIG_PCI_LEGACY_PROC=y
CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
#
# PCI Hotplug Support
@@ -138,7 +161,6 @@ CONFIG_HOTPLUG_PCI=y
CONFIG_HOTPLUG_PCI_ACPI=y
# CONFIG_HOTPLUG_PCI_ACPI_IBM is not set
# CONFIG_HOTPLUG_PCI_CPCI is not set
-# CONFIG_HOTPLUG_PCI_PCIE is not set
# CONFIG_HOTPLUG_PCI_SHPC is not set
#
@@ -146,10 +168,6 @@ CONFIG_HOTPLUG_PCI_ACPI=y
#
# CONFIG_PCCARD is not set
-#
-# PC-card bridges
-#
-
#
# Device Drivers
#
@@ -184,6 +202,7 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
# CONFIG_BLK_CPQ_CISS_DA is not set
# CONFIG_BLK_DEV_DAC960 is not set
# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
CONFIG_BLK_DEV_LOOP=y
# CONFIG_BLK_DEV_CRYPTOLOOP is not set
# CONFIG_BLK_DEV_NBD is not set
@@ -203,6 +222,7 @@ CONFIG_IOSCHED_NOOP=y
CONFIG_IOSCHED_AS=y
CONFIG_IOSCHED_DEADLINE=y
CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
#
# ATA/ATAPI/MFM/RLL support
@@ -246,6 +266,7 @@ CONFIG_BLK_DEV_CMD64X=y
# CONFIG_BLK_DEV_HPT366 is not set
# CONFIG_BLK_DEV_SC1200 is not set
# CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT821X is not set
# CONFIG_BLK_DEV_NS87415 is not set
# CONFIG_BLK_DEV_PDC202XX_OLD is not set
# CONFIG_BLK_DEV_PDC202XX_NEW is not set
@@ -275,6 +296,7 @@ CONFIG_CHR_DEV_OSST=y
CONFIG_BLK_DEV_SR=y
CONFIG_BLK_DEV_SR_VENDOR=y
CONFIG_CHR_DEV_SG=y
+# CONFIG_CHR_DEV_SCH is not set
#
# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
@@ -288,6 +310,7 @@ CONFIG_SCSI_LOGGING=y
#
CONFIG_SCSI_SPI_ATTRS=y
# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
#
# SCSI low-level drivers
@@ -303,13 +326,10 @@ CONFIG_SCSI_SPI_ATTRS=y
# CONFIG_MEGARAID_NEWGEN is not set
# CONFIG_MEGARAID_LEGACY is not set
# CONFIG_SCSI_SATA is not set
-# CONFIG_SCSI_BUSLOGIC is not set
# CONFIG_SCSI_CPQFCTS is not set
# CONFIG_SCSI_DMX3191D is not set
-# CONFIG_SCSI_EATA is not set
# CONFIG_SCSI_EATA_PIO is not set
# CONFIG_SCSI_FUTURE_DOMAIN is not set
-# CONFIG_SCSI_GDTH is not set
# CONFIG_SCSI_IPS is not set
# CONFIG_SCSI_INITIO is not set
# CONFIG_SCSI_INIA100 is not set
@@ -319,8 +339,6 @@ CONFIG_SCSI_SYM53C8XX_DEFAULT_TAGS=16
CONFIG_SCSI_SYM53C8XX_MAX_TAGS=64
# CONFIG_SCSI_SYM53C8XX_IOMAPPED is not set
# CONFIG_SCSI_IPR is not set
-# CONFIG_SCSI_PCI2000 is not set
-# CONFIG_SCSI_PCI2220I is not set
# CONFIG_SCSI_QLOGIC_ISP is not set
# CONFIG_SCSI_QLOGIC_FC is not set
CONFIG_SCSI_QLOGIC_1280=y
@@ -331,7 +349,7 @@ CONFIG_SCSI_QLA2XXX=y
# CONFIG_SCSI_QLA2300 is not set
# CONFIG_SCSI_QLA2322 is not set
# CONFIG_SCSI_QLA6312 is not set
-# CONFIG_SCSI_QLA6322 is not set
+# CONFIG_SCSI_LPFC is not set
# CONFIG_SCSI_DC395x is not set
# CONFIG_SCSI_DC390T is not set
# CONFIG_SCSI_DEBUG is not set
@@ -344,9 +362,9 @@ CONFIG_SCSI_QLA2XXX=y
#
# Fusion MPT device support
#
-CONFIG_FUSION=y
-CONFIG_FUSION_MAX_SGE=40
-# CONFIG_FUSION_CTL is not set
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
#
# IEEE 1394 (FireWire) support
@@ -368,12 +386,12 @@ CONFIG_NET=y
#
CONFIG_PACKET=y
# CONFIG_PACKET_MMAP is not set
-# CONFIG_NETLINK_DEV is not set
CONFIG_UNIX=y
# CONFIG_NET_KEY is not set
CONFIG_INET=y
CONFIG_IP_MULTICAST=y
# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
# CONFIG_IP_PNP is not set
# CONFIG_NET_IPIP is not set
# CONFIG_NET_IPGRE is not set
@@ -386,6 +404,8 @@ CONFIG_IP_MULTICAST=y
# CONFIG_INET_TUNNEL is not set
# CONFIG_IP_TCPDIAG is not set
# CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
#
# IP: Virtual Server Configuration
@@ -405,8 +425,6 @@ CONFIG_NETFILTER=y
CONFIG_IP_NF_ARPTABLES=y
# CONFIG_IP_NF_ARPFILTER is not set
# CONFIG_IP_NF_ARP_MANGLE is not set
-# CONFIG_IP_NF_COMPAT_IPCHAINS is not set
-# CONFIG_IP_NF_COMPAT_IPFWADM is not set
#
# SCTP Configuration (EXPERIMENTAL)
@@ -483,7 +501,6 @@ CONFIG_NET_PCI=y
# CONFIG_DGRS is not set
# CONFIG_EEPRO100 is not set
CONFIG_E100=y
-# CONFIG_E100_NAPI is not set
# CONFIG_FEALNX is not set
# CONFIG_NATSEMI is not set
# CONFIG_NE2K_PCI is not set
@@ -505,9 +522,11 @@ CONFIG_E1000=y
# CONFIG_HAMACHI is not set
# CONFIG_YELLOWFIN is not set
# CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
# CONFIG_SK98LIN is not set
# CONFIG_VIA_VELOCITY is not set
CONFIG_TIGON3=y
+# CONFIG_BNX2 is not set
#
# Ethernet (10000 Mbit)
@@ -564,18 +583,6 @@ CONFIG_INPUT_JOYDEV=y
CONFIG_INPUT_EVDEV=y
# CONFIG_INPUT_EVBUG is not set
-#
-# Input I/O drivers
-#
-# CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
-CONFIG_SERIO=y
-# CONFIG_SERIO_I8042 is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_PCIPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
#
# Input Device Drivers
#
@@ -585,6 +592,16 @@ CONFIG_SERIO=y
# CONFIG_INPUT_TOUCHSCREEN is not set
# CONFIG_INPUT_MISC is not set
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+# CONFIG_SERIO_I8042 is not set
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_PCIPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
#
# Character devices
#
@@ -603,7 +620,6 @@ CONFIG_SERIAL_8250_NR_UARTS=8
CONFIG_SERIAL_8250_EXTENDED=y
CONFIG_SERIAL_8250_SHARE_IRQ=y
# CONFIG_SERIAL_8250_DETECT_IRQ is not set
-# CONFIG_SERIAL_8250_MULTIPORT is not set
# CONFIG_SERIAL_8250_RSA is not set
#
@@ -611,6 +627,7 @@ CONFIG_SERIAL_8250_SHARE_IRQ=y
#
CONFIG_SERIAL_CORE=y
CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
CONFIG_UNIX98_PTYS=y
CONFIG_LEGACY_PTYS=y
CONFIG_LEGACY_PTY_COUNT=256
@@ -644,6 +661,12 @@ CONFIG_DRM_RADEON=y
# CONFIG_DRM_SIS is not set
# CONFIG_RAW_DRIVER is not set
# CONFIG_HPET is not set
+# CONFIG_HANGCHECK_TIMER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
#
# I2C support
@@ -668,6 +691,7 @@ CONFIG_I2C_ALGOPCF=y
# CONFIG_I2C_AMD8111 is not set
# CONFIG_I2C_I801 is not set
# CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
# CONFIG_I2C_ISA is not set
# CONFIG_I2C_NFORCE2 is not set
# CONFIG_I2C_PARPORT_LIGHT is not set
@@ -691,10 +715,14 @@ CONFIG_I2C_ALGOPCF=y
# CONFIG_SENSORS_ADM1025 is not set
# CONFIG_SENSORS_ADM1026 is not set
# CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
# CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 is not set
# CONFIG_SENSORS_DS1621 is not set
# CONFIG_SENSORS_FSCHER is not set
+# CONFIG_SENSORS_FSCPOS is not set
# CONFIG_SENSORS_GL518SM is not set
+# CONFIG_SENSORS_GL520SM is not set
# CONFIG_SENSORS_IT87 is not set
# CONFIG_SENSORS_LM63 is not set
# CONFIG_SENSORS_LM75 is not set
@@ -705,21 +733,29 @@ CONFIG_I2C_ALGOPCF=y
# CONFIG_SENSORS_LM85 is not set
# CONFIG_SENSORS_LM87 is not set
# CONFIG_SENSORS_LM90 is not set
+# CONFIG_SENSORS_LM92 is not set
# CONFIG_SENSORS_MAX1619 is not set
# CONFIG_SENSORS_PC87360 is not set
+# CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_SIS5595 is not set
# CONFIG_SENSORS_SMSC47M1 is not set
# CONFIG_SENSORS_VIA686A is not set
# CONFIG_SENSORS_W83781D is not set
# CONFIG_SENSORS_W83L785TS is not set
# CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
#
# Other I2C Chip support
#
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
# CONFIG_SENSORS_EEPROM is not set
# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
# CONFIG_SENSORS_PCF8591 is not set
# CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
# CONFIG_I2C_DEBUG_CORE is not set
# CONFIG_I2C_DEBUG_ALGO is not set
# CONFIG_I2C_DEBUG_BUS is not set
@@ -746,6 +782,7 @@ CONFIG_VIDEO_DEV=y
#
# Video Adapters
#
+# CONFIG_TUNER_MULTI_I2C is not set
# CONFIG_VIDEO_BT848 is not set
# CONFIG_VIDEO_CPIA is not set
# CONFIG_VIDEO_SAA5246A is not set
@@ -778,6 +815,11 @@ CONFIG_VIDEO_DEV=y
# Graphics support
#
CONFIG_FB=y
+CONFIG_FB_CFB_FILLRECT=y
+CONFIG_FB_CFB_COPYAREA=y
+CONFIG_FB_CFB_IMAGEBLIT=y
+CONFIG_FB_SOFT_CURSOR=y
+# CONFIG_FB_MACMODES is not set
CONFIG_FB_MODE_HELPERS=y
# CONFIG_FB_TILEBLITTING is not set
# CONFIG_FB_CIRRUS is not set
@@ -785,6 +827,7 @@ CONFIG_FB_MODE_HELPERS=y
# CONFIG_FB_CYBER2000 is not set
# CONFIG_FB_ASILIANT is not set
# CONFIG_FB_IMSTT is not set
+# CONFIG_FB_NVIDIA is not set
# CONFIG_FB_RIVA is not set
# CONFIG_FB_MATROX is not set
# CONFIG_FB_RADEON_OLD is not set
@@ -801,6 +844,7 @@ CONFIG_FB_RADEON_DEBUG=y
# CONFIG_FB_VOODOO1 is not set
# CONFIG_FB_TRIDENT is not set
# CONFIG_FB_PM3 is not set
+# CONFIG_FB_S1D13XXX is not set
# CONFIG_FB_VIRTUAL is not set
#
@@ -820,6 +864,7 @@ CONFIG_LOGO=y
# CONFIG_LOGO_LINUX_MONO is not set
# CONFIG_LOGO_LINUX_VGA16 is not set
CONFIG_LOGO_LINUX_CLUT224=y
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
#
# Sound
@@ -869,6 +914,8 @@ CONFIG_SND_AC97_CODEC=y
# CONFIG_SND_CS46XX is not set
# CONFIG_SND_CS4281 is not set
# CONFIG_SND_EMU10K1 is not set
+# CONFIG_SND_EMU10K1X is not set
+# CONFIG_SND_CA0106 is not set
# CONFIG_SND_KORG1212 is not set
# CONFIG_SND_MIXART is not set
# CONFIG_SND_NM256 is not set
@@ -876,6 +923,7 @@ CONFIG_SND_AC97_CODEC=y
# CONFIG_SND_RME96 is not set
# CONFIG_SND_RME9652 is not set
# CONFIG_SND_HDSP is not set
+# CONFIG_SND_HDSPM is not set
# CONFIG_SND_TRIDENT is not set
# CONFIG_SND_YMFPCI is not set
# CONFIG_SND_ALS4000 is not set
@@ -893,13 +941,14 @@ CONFIG_SND_FM801_TEA575X=y
# CONFIG_SND_INTEL8X0M is not set
# CONFIG_SND_SONICVIBES is not set
# CONFIG_SND_VIA82XX is not set
+# CONFIG_SND_VIA82XX_MODEM is not set
# CONFIG_SND_VX222 is not set
+# CONFIG_SND_HDA_INTEL is not set
#
# USB devices
#
# CONFIG_SND_USB_AUDIO is not set
-# CONFIG_SND_USB_USX2Y is not set
#
# Open Sound System
@@ -909,6 +958,8 @@ CONFIG_SND_FM801_TEA575X=y
#
# USB support
#
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
CONFIG_USB=y
# CONFIG_USB_DEBUG is not set
@@ -920,8 +971,6 @@ CONFIG_USB_BANDWIDTH=y
# CONFIG_USB_DYNAMIC_MINORS is not set
# CONFIG_USB_SUSPEND is not set
# CONFIG_USB_OTG is not set
-CONFIG_USB_ARCH_HAS_HCD=y
-CONFIG_USB_ARCH_HAS_OHCI=y
#
# USB Host Controller Drivers
@@ -929,7 +978,10 @@ CONFIG_USB_ARCH_HAS_OHCI=y
CONFIG_USB_EHCI_HCD=y
# CONFIG_USB_EHCI_SPLIT_ISO is not set
# CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
CONFIG_USB_OHCI_HCD=y
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
CONFIG_USB_UHCI_HCD=y
# CONFIG_USB_SL811_HCD is not set
@@ -947,12 +999,11 @@ CONFIG_USB_UHCI_HCD=y
#
CONFIG_USB_STORAGE=y
# CONFIG_USB_STORAGE_DEBUG is not set
-# CONFIG_USB_STORAGE_RW_DETECT is not set
# CONFIG_USB_STORAGE_DATAFAB is not set
# CONFIG_USB_STORAGE_FREECOM is not set
# CONFIG_USB_STORAGE_ISD200 is not set
# CONFIG_USB_STORAGE_DPCM is not set
-# CONFIG_USB_STORAGE_HP8200e is not set
+# CONFIG_USB_STORAGE_USBAT is not set
# CONFIG_USB_STORAGE_SDDR09 is not set
# CONFIG_USB_STORAGE_SDDR55 is not set
# CONFIG_USB_STORAGE_JUMPSHOT is not set
@@ -966,9 +1017,11 @@ CONFIG_USB_HIDINPUT=y
CONFIG_USB_HIDDEV=y
# CONFIG_USB_AIPTEK is not set
# CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
# CONFIG_USB_KBTAB is not set
# CONFIG_USB_POWERMATE is not set
# CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
# CONFIG_USB_EGALAX is not set
# CONFIG_USB_XPAD is not set
# CONFIG_USB_ATI_REMOTE is not set
@@ -978,7 +1031,6 @@ CONFIG_USB_HIDDEV=y
#
# CONFIG_USB_MDC800 is not set
# CONFIG_USB_MICROTEK is not set
-# CONFIG_USB_HPUSBSCSI is not set
#
# USB Multimedia devices
@@ -992,6 +1044,7 @@ CONFIG_USB_HIDDEV=y
# CONFIG_USB_SE401 is not set
# CONFIG_USB_SN9C102 is not set
# CONFIG_USB_STV680 is not set
+# CONFIG_USB_PWC is not set
#
# USB Network Adapters
@@ -1001,6 +1054,7 @@ CONFIG_USB_HIDDEV=y
# CONFIG_USB_PEGASUS is not set
# CONFIG_USB_RTL8150 is not set
# CONFIG_USB_USBNET is not set
+CONFIG_USB_MON=y
#
# USB port drivers
@@ -1016,7 +1070,6 @@ CONFIG_USB_HIDDEV=y
#
# CONFIG_USB_EMI62 is not set
# CONFIG_USB_EMI26 is not set
-# CONFIG_USB_TIGL is not set
# CONFIG_USB_AUERSWALD is not set
# CONFIG_USB_RIO500 is not set
# CONFIG_USB_LEGOTOWER is not set
@@ -1025,9 +1078,11 @@ CONFIG_USB_HIDDEV=y
# CONFIG_USB_CYTHERM is not set
# CONFIG_USB_PHIDGETKIT is not set
# CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_SISUSBVGA is not set
#
-# USB ATM/DSL drivers
+# USB DSL modem support
#
#
@@ -1040,6 +1095,15 @@ CONFIG_USB_HIDDEV=y
#
# CONFIG_MMC is not set
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
#
# File systems
#
@@ -1047,6 +1111,7 @@ CONFIG_EXT2_FS=y
CONFIG_EXT2_FS_XATTR=y
# CONFIG_EXT2_FS_POSIX_ACL is not set
# CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
CONFIG_EXT3_FS=y
CONFIG_EXT3_FS_XATTR=y
# CONFIG_EXT3_FS_POSIX_ACL is not set
@@ -1056,6 +1121,10 @@ CONFIG_JBD=y
CONFIG_FS_MBCACHE=y
# CONFIG_REISERFS_FS is not set
# CONFIG_JFS_FS is not set
+
+#
+# XFS support
+#
# CONFIG_XFS_FS is not set
# CONFIG_MINIX_FS is not set
# CONFIG_ROMFS_FS is not set
@@ -1089,7 +1158,6 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
CONFIG_PROC_FS=y
CONFIG_PROC_KCORE=y
CONFIG_SYSFS=y
-# CONFIG_DEVFS_FS is not set
# CONFIG_DEVPTS_FS_XATTR is not set
CONFIG_TMPFS=y
CONFIG_TMPFS_XATTR=y
@@ -1120,15 +1188,18 @@ CONFIG_RAMFS=y
#
CONFIG_NFS_FS=y
CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
CONFIG_NFS_V4=y
# CONFIG_NFS_DIRECTIO is not set
CONFIG_NFSD=y
CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
# CONFIG_NFSD_V4 is not set
# CONFIG_NFSD_TCP is not set
CONFIG_LOCKD=y
CONFIG_LOCKD_V4=y
CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
CONFIG_SUNRPC=y
CONFIG_SUNRPC_GSS=y
CONFIG_RPCSEC_GSS_KRB5=y
@@ -1209,6 +1280,8 @@ CONFIG_NLS_UTF8=y
# CONFIG_CRC_CCITT is not set
CONFIG_CRC32=y
# CONFIG_LIBCRC32C is not set
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
#
# Profiling support
@@ -1218,14 +1291,18 @@ CONFIG_CRC32=y
#
# Kernel hacking
#
+# CONFIG_PRINTK_TIME is not set
CONFIG_DEBUG_KERNEL=y
CONFIG_MAGIC_SYSRQ=y
+CONFIG_LOG_BUF_SHIFT=17
# CONFIG_SCHEDSTATS is not set
# CONFIG_DEBUG_SLAB is not set
# CONFIG_DEBUG_SPINLOCK is not set
# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
# CONFIG_DEBUG_KOBJECT is not set
# CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_FS is not set
+CONFIG_KPROBES=y
CONFIG_IA64_GRANULE_16MB=y
# CONFIG_IA64_GRANULE_64MB is not set
CONFIG_IA64_PRINT_HAZARDS=y
@@ -1252,6 +1329,7 @@ CONFIG_CRYPTO_MD5=y
# CONFIG_CRYPTO_SHA256 is not set
# CONFIG_CRYPTO_SHA512 is not set
# CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
CONFIG_CRYPTO_DES=y
# CONFIG_CRYPTO_BLOWFISH is not set
# CONFIG_CRYPTO_TWOFISH is not set
diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index b8db6e3e5e81..11957598a8b9 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -156,10 +156,13 @@
*/
#define DELAYED_RESOURCE_CNT 64
+#define PCI_DEVICE_ID_HP_SX2000_IOC 0x12ec
+
#define ZX1_IOC_ID ((PCI_DEVICE_ID_HP_ZX1_IOC << 16) | PCI_VENDOR_ID_HP)
#define ZX2_IOC_ID ((PCI_DEVICE_ID_HP_ZX2_IOC << 16) | PCI_VENDOR_ID_HP)
#define REO_IOC_ID ((PCI_DEVICE_ID_HP_REO_IOC << 16) | PCI_VENDOR_ID_HP)
#define SX1000_IOC_ID ((PCI_DEVICE_ID_HP_SX1000_IOC << 16) | PCI_VENDOR_ID_HP)
+#define SX2000_IOC_ID ((PCI_DEVICE_ID_HP_SX2000_IOC << 16) | PCI_VENDOR_ID_HP)
#define ZX1_IOC_OFFSET 0x1000 /* ACPI reports SBA, we want IOC */
@@ -1726,6 +1729,7 @@ static struct ioc_iommu ioc_iommu_info[] __initdata = {
{ ZX1_IOC_ID, "zx1", ioc_zx1_init },
{ ZX2_IOC_ID, "zx2", NULL },
{ SX1000_IOC_ID, "sx1000", NULL },
+ { SX2000_IOC_ID, "sx2000", NULL },
};
static struct ioc * __init
diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 786e70718ce4..7a8ae0f4b387 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -30,6 +30,7 @@
#include
#include
#include
+#include
#include
#include
@@ -149,12 +150,17 @@ static void receive_chars(struct tty_struct *tty, struct pt_regs *regs)
seen_esc = 2;
continue;
} else if ( seen_esc == 2 ) {
- if ( ch == 'P' ) show_state(); /* F1 key */
-#ifdef CONFIG_KDB
- if ( ch == 'S' )
- kdb(KDB_REASON_KEYBOARD, 0, (kdb_eframe_t) regs);
+ if ( ch == 'P' ) /* F1 */
+ show_state();
+#ifdef CONFIG_MAGIC_SYSRQ
+ if ( ch == 'S' ) { /* F4 */
+ do
+ ch = ia64_ssc(0, 0, 0, 0,
+ SSC_GETCHAR);
+ while (!ch);
+ handle_sysrq(ch, regs, NULL);
+ }
#endif
-
seen_esc = 0;
continue;
}
diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index 72dfd9e7de0f..cda06f88c66e 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -236,9 +236,7 @@ acpi_parse_iosapic (acpi_table_entry_header *header, const unsigned long end)
if (BAD_MADT_ENTRY(iosapic, end))
return -EINVAL;
- iosapic_init(iosapic->address, iosapic->global_irq_base);
-
- return 0;
+ return iosapic_init(iosapic->address, iosapic->global_irq_base);
}
@@ -772,7 +770,7 @@ EXPORT_SYMBOL(acpi_unmap_lsapic);
#ifdef CONFIG_ACPI_NUMA
-acpi_status __init
+acpi_status __devinit
acpi_map_iosapic (acpi_handle handle, u32 depth, void *context, void **ret)
{
struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
@@ -825,4 +823,28 @@ acpi_map_iosapic (acpi_handle handle, u32 depth, void *context, void **ret)
return AE_OK;
}
#endif /* CONFIG_NUMA */
+
+int
+acpi_register_ioapic (acpi_handle handle, u64 phys_addr, u32 gsi_base)
+{
+ int err;
+
+ if ((err = iosapic_init(phys_addr, gsi_base)))
+ return err;
+
+#if CONFIG_ACPI_NUMA
+ acpi_map_iosapic(handle, 0, NULL, NULL);
+#endif /* CONFIG_ACPI_NUMA */
+
+ return 0;
+}
+EXPORT_SYMBOL(acpi_register_ioapic);
+
+int
+acpi_unregister_ioapic (acpi_handle handle, u32 gsi_base)
+{
+ return iosapic_remove(gsi_base);
+}
+EXPORT_SYMBOL(acpi_unregister_ioapic);
+
#endif /* CONFIG_ACPI_BOOT */
diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index b1d5d3d5276c..69f88d561d62 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -470,18 +470,6 @@ ENTRY(load_switch_stack)
br.cond.sptk.many b7
END(load_switch_stack)
-GLOBAL_ENTRY(__ia64_syscall)
- .regstk 6,0,0,0
- mov r15=in5 // put syscall number in place
- break __BREAK_SYSCALL
- movl r2=errno
- cmp.eq p6,p7=-1,r10
- ;;
-(p6) st4 [r2]=r8
-(p6) mov r8=-1
- br.ret.sptk.many rp
-END(__ia64_syscall)
-
GLOBAL_ENTRY(execve)
mov r15=__NR_execve // put syscall number in place
break __BREAK_SYSCALL
@@ -637,7 +625,7 @@ END(ia64_ret_from_syscall)
* r8-r11: restored (syscall return value(s))
* r12: restored (user-level stack pointer)
* r13: restored (user-level thread pointer)
- * r14: cleared
+ * r14: set to __kernel_syscall_via_epc
* r15: restored (syscall #)
* r16-r17: cleared
* r18: user-level b6
@@ -658,7 +646,7 @@ END(ia64_ret_from_syscall)
* pr: restored (user-level pr)
* b0: restored (user-level rp)
* b6: restored
- * b7: cleared
+ * b7: set to __kernel_syscall_via_epc
* ar.unat: restored (user-level ar.unat)
* ar.pfs: restored (user-level ar.pfs)
* ar.rsc: restored (user-level ar.rsc)
@@ -704,72 +692,79 @@ ENTRY(ia64_leave_syscall)
;;
(p6) ld4 r31=[r18] // load current_thread_info()->flags
ld8 r19=[r2],PT(B6)-PT(LOADRS) // load ar.rsc value for "loadrs"
- mov b7=r0 // clear b7
+ nop.i 0
;;
- ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE) // load ar.bspstore (may be garbage)
+ mov r16=ar.bsp // M2 get existing backing store pointer
ld8 r18=[r2],PT(R9)-PT(B6) // load b6
(p6) and r15=TIF_WORK_MASK,r31 // any work other than TIF_SYSCALL_TRACE?
;;
- mov r16=ar.bsp // M2 get existing backing store pointer
+ ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE) // load ar.bspstore (may be garbage)
(p6) cmp4.ne.unc p6,p0=r15, r0 // any special work pending?
(p6) br.cond.spnt .work_pending_syscall
;;
// start restoring the state saved on the kernel stack (struct pt_regs):
ld8 r9=[r2],PT(CR_IPSR)-PT(R9)
ld8 r11=[r3],PT(CR_IIP)-PT(R11)
- mov f6=f0 // clear f6
+(pNonSys) break 0 // bug check: we shouldn't be here if pNonSys is TRUE!
;;
invala // M0|1 invalidate ALAT
- rsm psr.i | psr.ic // M2 initiate turning off of interrupt and interruption collection
- mov f9=f0 // clear f9
+ rsm psr.i | psr.ic // M2 turn off interrupts and interruption collection
+ cmp.eq p9,p0=r0,r0 // A set p9 to indicate that we should restore cr.ifs
- ld8 r29=[r2],16 // load cr.ipsr
- ld8 r28=[r3],16 // load cr.iip
- mov f8=f0 // clear f8
+ ld8 r29=[r2],16 // M0|1 load cr.ipsr
+ ld8 r28=[r3],16 // M0|1 load cr.iip
+ mov r22=r0 // A clear r22
;;
ld8 r30=[r2],16 // M0|1 load cr.ifs
ld8 r25=[r3],16 // M0|1 load ar.unat
- cmp.eq p9,p0=r0,r0 // set p9 to indicate that we should restore cr.ifs
- ;;
- ld8 r26=[r2],PT(B0)-PT(AR_PFS) // M0|1 load ar.pfs
-(pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled
- mov f10=f0 // clear f10
- ;;
- ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // load b0
- ld8 r27=[r3],PT(PR)-PT(AR_RSC) // load ar.rsc
- mov f11=f0 // clear f11
- ;;
- ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT) // load ar.rnat (may be garbage)
- ld8 r31=[r3],PT(R1)-PT(PR) // load predicates
(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
;;
- ld8 r20=[r2],PT(R12)-PT(AR_FPSR) // load ar.fpsr
- ld8.fill r1=[r3],16 // load r1
-(pUStk) mov r17=1
+ ld8 r26=[r2],PT(B0)-PT(AR_PFS) // M0|1 load ar.pfs
+(pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled
+ nop 0
;;
- srlz.d // M0 ensure interruption collection is off
- ld8.fill r13=[r3],16
- mov f7=f0 // clear f7
+ ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // M0|1 load b0
+ ld8 r27=[r3],PT(PR)-PT(AR_RSC) // M0|1 load ar.rsc
+ mov f6=f0 // F clear f6
;;
- ld8.fill r12=[r2] // restore r12 (sp)
- mov.m ar.ssd=r0 // M2 clear ar.ssd
- mov r22=r0 // clear r22
+ ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT) // M0|1 load ar.rnat (may be garbage)
+ ld8 r31=[r3],PT(R1)-PT(PR) // M0|1 load predicates
+ mov f7=f0 // F clear f7
+ ;;
+ ld8 r20=[r2],PT(R12)-PT(AR_FPSR) // M0|1 load ar.fpsr
+ ld8.fill r1=[r3],16 // M0|1 load r1
+(pUStk) mov r17=1 // A
+ ;;
+(pUStk) st1 [r14]=r17 // M2|3
+ ld8.fill r13=[r3],16 // M0|1
+ mov f8=f0 // F clear f8
+ ;;
+ ld8.fill r12=[r2] // M0|1 restore r12 (sp)
+ ld8.fill r15=[r3] // M0|1 restore r15
+ mov b6=r18 // I0 restore b6
- ld8.fill r15=[r3] // restore r15
-(pUStk) st1 [r14]=r17
- addl r3=THIS_CPU(ia64_phys_stacked_size_p8),r0
- ;;
-(pUStk) ld4 r17=[r3] // r17 = cpu_data->phys_stacked_size_p8
- mov.m ar.csd=r0 // M2 clear ar.csd
- mov b6=r18 // I0 restore b6
- ;;
- mov r14=r0 // clear r14
- shr.u r18=r19,16 // I0|1 get byte size of existing "dirty" partition
-(pKStk) br.cond.dpnt.many skip_rbs_switch
+ addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0 // A
+ mov f9=f0 // F clear f9
+(pKStk) br.cond.dpnt.many skip_rbs_switch // B
- mov.m ar.ccv=r0 // clear ar.ccv
-(pNonSys) br.cond.dpnt.many dont_preserve_current_frame
- br.cond.sptk.many rbs_switch
+ srlz.d // M0 ensure interruption collection is off (for cover)
+ shr.u r18=r19,16 // I0|1 get byte size of existing "dirty" partition
+ cover // B add current frame into dirty partition & set cr.ifs
+ ;;
+(pUStk) ld4 r17=[r17] // M0|1 r17 = cpu_data->phys_stacked_size_p8
+ mov r19=ar.bsp // M2 get new backing store pointer
+ mov f10=f0 // F clear f10
+
+ nop.m 0
+ movl r14=__kernel_syscall_via_epc // X
+ ;;
+ mov.m ar.csd=r0 // M2 clear ar.csd
+ mov.m ar.ccv=r0 // M2 clear ar.ccv
+ mov b7=r14 // I0 clear b7 (hint with __kernel_syscall_via_epc)
+
+ mov.m ar.ssd=r0 // M2 clear ar.ssd
+ mov f11=f0 // F clear f11
+ br.cond.sptk.many rbs_switch // B
END(ia64_leave_syscall)
#ifdef CONFIG_IA32_SUPPORT
@@ -885,7 +880,7 @@ GLOBAL_ENTRY(ia64_leave_kernel)
ldf.fill f7=[r2],PT(F11)-PT(F7)
ldf.fill f8=[r3],32
;;
- srlz.i // ensure interruption collection is off
+ srlz.d // ensure that inter. collection is off (VHPT is don't care, since text is pinned)
mov ar.ccv=r15
;;
ldf.fill f11=[r2]
@@ -945,11 +940,10 @@ GLOBAL_ENTRY(ia64_leave_kernel)
* NOTE: alloc, loadrs, and cover can't be predicated.
*/
(pNonSys) br.cond.dpnt dont_preserve_current_frame
-
-rbs_switch:
cover // add current frame into dirty partition and set cr.ifs
;;
mov r19=ar.bsp // get new backing store pointer
+rbs_switch:
sub r16=r16,r18 // krbs = old bsp - size of dirty partition
cmp.ne p9,p0=r0,r0 // clear p9 to skip restore of cr.ifs
;;
@@ -1024,14 +1018,14 @@ rse_clear_invalid:
mov loc5=0
mov loc6=0
mov loc7=0
-(pRecurse) br.call.sptk.few b0=rse_clear_invalid
+(pRecurse) br.call.dptk.few b0=rse_clear_invalid
;;
mov loc8=0
mov loc9=0
cmp.ne pReturn,p0=r0,in1 // if recursion count != 0, we need to do a br.ret
mov loc10=0
mov loc11=0
-(pReturn) br.ret.sptk.many b0
+(pReturn) br.ret.dptk.many b0
#endif /* !CONFIG_ITANIUM */
# undef pRecurse
# undef pReturn
@@ -1577,8 +1571,8 @@ sys_call_table:
data8 sys_add_key
data8 sys_request_key
data8 sys_keyctl
- data8 sys_ni_syscall
- data8 sys_ni_syscall // 1275
+ data8 sys_ioprio_set
+ data8 sys_ioprio_get // 1275
data8 sys_set_zone_reclaim
data8 sys_ni_syscall
data8 sys_ni_syscall
diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index 962b6c4e32b5..7d7684a369d3 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -531,93 +531,114 @@ GLOBAL_ENTRY(fsys_bubble_down)
.altrp b6
.body
/*
- * We get here for syscalls that don't have a lightweight handler. For those, we
- * need to bubble down into the kernel and that requires setting up a minimal
- * pt_regs structure, and initializing the CPU state more or less as if an
- * interruption had occurred. To make syscall-restarts work, we setup pt_regs
- * such that cr_iip points to the second instruction in syscall_via_break.
- * Decrementing the IP hence will restart the syscall via break and not
- * decrementing IP will return us to the caller, as usual. Note that we preserve
- * the value of psr.pp rather than initializing it from dcr.pp. This makes it
- * possible to distinguish fsyscall execution from other privileged execution.
+ * We get here for syscalls that don't have a lightweight
+ * handler. For those, we need to bubble down into the kernel
+ * and that requires setting up a minimal pt_regs structure,
+ * and initializing the CPU state more or less as if an
+ * interruption had occurred. To make syscall-restarts work,
+ * we setup pt_regs such that cr_iip points to the second
+ * instruction in syscall_via_break. Decrementing the IP
+ * hence will restart the syscall via break and not
+ * decrementing IP will return us to the caller, as usual.
+ * Note that we preserve the value of psr.pp rather than
+ * initializing it from dcr.pp. This makes it possible to
+ * distinguish fsyscall execution from other privileged
+ * execution.
*
* On entry:
- * - normal fsyscall handler register usage, except that we also have:
+ * - normal fsyscall handler register usage, except
+ * that we also have:
* - r18: address of syscall entry point
* - r21: ar.fpsr
* - r26: ar.pfs
* - r27: ar.rsc
* - r29: psr
+ *
+ * We used to clear some PSR bits here but that requires slow
+ * serialization. Fortuntely, that isn't really necessary.
+ * The rationale is as follows: we used to clear bits
+ * ~PSR_PRESERVED_BITS in PSR.L. Since
+ * PSR_PRESERVED_BITS==PSR.{UP,MFL,MFH,PK,DT,PP,SP,RT,IC}, we
+ * ended up clearing PSR.{BE,AC,I,DFL,DFH,DI,DB,SI,TB}.
+ * However,
+ *
+ * PSR.BE : already is turned off in __kernel_syscall_via_epc()
+ * PSR.AC : don't care (kernel normally turns PSR.AC on)
+ * PSR.I : already turned off by the time fsys_bubble_down gets
+ * invoked
+ * PSR.DFL: always 0 (kernel never turns it on)
+ * PSR.DFH: don't care --- kernel never touches f32-f127 on its own
+ * initiative
+ * PSR.DI : always 0 (kernel never turns it on)
+ * PSR.SI : always 0 (kernel never turns it on)
+ * PSR.DB : don't care --- kernel never enables kernel-level
+ * breakpoints
+ * PSR.TB : must be 0 already; if it wasn't zero on entry to
+ * __kernel_syscall_via_epc, the branch to fsys_bubble_down
+ * will trigger a taken branch; the taken-trap-handler then
+ * converts the syscall into a break-based system-call.
*/
-# define PSR_PRESERVED_BITS (IA64_PSR_UP | IA64_PSR_MFL | IA64_PSR_MFH | IA64_PSR_PK \
- | IA64_PSR_DT | IA64_PSR_PP | IA64_PSR_SP | IA64_PSR_RT \
- | IA64_PSR_IC)
/*
- * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc. The rest we have
- * to synthesize.
+ * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc.
+ * The rest we have to synthesize.
*/
-# define PSR_ONE_BITS ((3 << IA64_PSR_CPL0_BIT) | (0x1 << IA64_PSR_RI_BIT) \
+# define PSR_ONE_BITS ((3 << IA64_PSR_CPL0_BIT) \
+ | (0x1 << IA64_PSR_RI_BIT) \
| IA64_PSR_BN | IA64_PSR_I)
- invala
- movl r8=PSR_ONE_BITS
+ invala // M0|1
+ movl r14=ia64_ret_from_syscall // X
- mov r25=ar.unat // save ar.unat (5 cyc)
- movl r9=PSR_PRESERVED_BITS
+ nop.m 0
+ movl r28=__kernel_syscall_via_break // X create cr.iip
+ ;;
- mov ar.rsc=0 // set enforced lazy mode, pl 0, little-endian, loadrs=0
- movl r28=__kernel_syscall_via_break
+ mov r2=r16 // A get task addr to addl-addressable register
+ adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16 // A
+ mov r31=pr // I0 save pr (2 cyc)
;;
- mov r23=ar.bspstore // save ar.bspstore (12 cyc)
- mov r31=pr // save pr (2 cyc)
- mov r20=r1 // save caller's gp in r20
+ st1 [r16]=r0 // M2|3 clear current->thread.on_ustack flag
+ addl r22=IA64_RBS_OFFSET,r2 // A compute base of RBS
+ add r3=TI_FLAGS+IA64_TASK_SIZE,r2 // A
;;
- mov r2=r16 // copy current task addr to addl-addressable register
- and r9=r9,r29
- mov r19=b6 // save b6 (2 cyc)
+ ld4 r3=[r3] // M0|1 r3 = current_thread_info()->flags
+ lfetch.fault.excl.nt1 [r22] // M0|1 prefetch register backing-store
+ nop.i 0
;;
- mov psr.l=r9 // slam the door (17 cyc to srlz.i)
- or r29=r8,r29 // construct cr.ipsr value to save
- addl r22=IA64_RBS_OFFSET,r2 // compute base of RBS
+ mov ar.rsc=0 // M2 set enforced lazy mode, pl 0, LE, loadrs=0
+ nop.m 0
+ nop.i 0
;;
- // GAS reports a spurious RAW hazard on the read of ar.rnat because it thinks
- // we may be reading ar.itc after writing to psr.l. Avoid that message with
- // this directive:
- dv_serialize_data
- mov.m r24=ar.rnat // read ar.rnat (5 cyc lat)
- lfetch.fault.excl.nt1 [r22]
- adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r2
+ mov r23=ar.bspstore // M2 (12 cyc) save ar.bspstore
+ mov.m r24=ar.rnat // M2 (5 cyc) read ar.rnat (dual-issues!)
+ nop.i 0
+ ;;
+ mov ar.bspstore=r22 // M2 (6 cyc) switch to kernel RBS
+ movl r8=PSR_ONE_BITS // X
+ ;;
+ mov r25=ar.unat // M2 (5 cyc) save ar.unat
+ mov r19=b6 // I0 save b6 (2 cyc)
+ mov r20=r1 // A save caller's gp in r20
+ ;;
+ or r29=r8,r29 // A construct cr.ipsr value to save
+ mov b6=r18 // I0 copy syscall entry-point to b6 (7 cyc)
+ addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 // A compute base of memory stack
- // ensure previous insn group is issued before we stall for srlz.i:
+ mov r18=ar.bsp // M2 save (kernel) ar.bsp (12 cyc)
+ cmp.ne pKStk,pUStk=r0,r0 // A set pKStk <- 0, pUStk <- 1
+ br.call.sptk.many b7=ia64_syscall_setup // B
;;
- srlz.i // ensure new psr.l has been established
- /////////////////////////////////////////////////////////////////////////////
- ////////// from this point on, execution is not interruptible anymore
- /////////////////////////////////////////////////////////////////////////////
- addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 // compute base of memory stack
- cmp.ne pKStk,pUStk=r0,r0 // set pKStk <- 0, pUStk <- 1
+ mov ar.rsc=0x3 // M2 set eager mode, pl 0, LE, loadrs=0
+ mov rp=r14 // I0 set the real return addr
+ and r3=_TIF_SYSCALL_TRACEAUDIT,r3 // A
;;
- st1 [r16]=r0 // clear current->thread.on_ustack flag
- mov ar.bspstore=r22 // switch to kernel RBS
- mov b6=r18 // copy syscall entry-point to b6 (7 cyc)
- add r3=TI_FLAGS+IA64_TASK_SIZE,r2
- ;;
- ld4 r3=[r3] // r2 = current_thread_info()->flags
- mov r18=ar.bsp // save (kernel) ar.bsp (12 cyc)
- mov ar.rsc=0x3 // set eager mode, pl 0, little-endian, loadrs=0
- br.call.sptk.many b7=ia64_syscall_setup
- ;;
- ssm psr.i
- movl r2=ia64_ret_from_syscall
- ;;
- mov rp=r2 // set the real return addr
- and r3=_TIF_SYSCALL_TRACEAUDIT,r3
- ;;
- cmp.eq p8,p0=r3,r0
+ ssm psr.i // M2 we're on kernel stacks now, reenable irqs
+ cmp.eq p8,p0=r3,r0 // A
+(p10) br.cond.spnt.many ia64_ret_from_syscall // B return if bad call-frame or r15 is a NaT
-(p10) br.cond.spnt.many ia64_ret_from_syscall // p10==true means out registers are more than 8
-(p8) br.call.sptk.many b6=b6 // ignore this return addr
- br.cond.sptk ia64_trace_syscall
+ nop.m 0
+(p8) br.call.sptk.many b6=b6 // B (ignore return address)
+ br.cond.spnt ia64_trace_syscall // B
END(fsys_bubble_down)
.rodata
diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S
index facf75acdc85..86948ce63e43 100644
--- a/arch/ia64/kernel/gate.S
+++ b/arch/ia64/kernel/gate.S
@@ -72,38 +72,40 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
* bundle get executed. The remaining code must be safe even if
* they do not get executed.
*/
- adds r17=-1024,r15
- mov r10=0 // default to successful syscall execution
- epc
+ adds r17=-1024,r15 // A
+ mov r10=0 // A default to successful syscall execution
+ epc // B causes split-issue
}
;;
- rsm psr.be // note: on McKinley "rsm psr.be/srlz.d" is slightly faster than "rum psr.be"
- LOAD_FSYSCALL_TABLE(r14)
+ rsm psr.be | psr.i // M2 (5 cyc to srlz.d)
+ LOAD_FSYSCALL_TABLE(r14) // X
+ ;;
+ mov r16=IA64_KR(CURRENT) // M2 (12 cyc)
+ shladd r18=r17,3,r14 // A
+ mov r19=NR_syscalls-1 // A
+ ;;
+ lfetch [r18] // M0|1
+ mov r29=psr // M2 (12 cyc)
+ // If r17 is a NaT, p6 will be zero
+ cmp.geu p6,p7=r19,r17 // A (sysnr > 0 && sysnr < 1024+NR_syscalls)?
+ ;;
+ mov r21=ar.fpsr // M2 (12 cyc)
+ tnat.nz p10,p9=r15 // I0
+ mov.i r26=ar.pfs // I0 (would stall anyhow due to srlz.d...)
+ ;;
+ srlz.d // M0 (forces split-issue) ensure PSR.BE==0
+(p6) ld8 r18=[r18] // M0|1
+ nop.i 0
+ ;;
+ nop.m 0
+(p6) tbit.z.unc p8,p0=r18,0 // I0 (dual-issues with "mov b7=r18"!)
+ nop.i 0
+ ;;
+(p8) ssm psr.i
+(p6) mov b7=r18 // I0
+(p8) br.dptk.many b7 // B
- mov r16=IA64_KR(CURRENT) // 12 cycle read latency
- tnat.nz p10,p9=r15
- mov r19=NR_syscalls-1
- ;;
- shladd r18=r17,3,r14
-
- srlz.d
- cmp.ne p8,p0=r0,r0 // p8 <- FALSE
- /* Note: if r17 is a NaT, p6 will be set to zero. */
- cmp.geu p6,p7=r19,r17 // (syscall > 0 && syscall < 1024+NR_syscalls)?
- ;;
-(p6) ld8 r18=[r18]
- mov r21=ar.fpsr
- add r14=-8,r14 // r14 <- addr of fsys_bubble_down entry
- ;;
-(p6) mov b7=r18
-(p6) tbit.z p8,p0=r18,0
-(p8) br.dptk.many b7
-
-(p6) rsm psr.i
- mov r27=ar.rsc
- mov r26=ar.pfs
- ;;
- mov r29=psr // read psr (12 cyc load latency)
+ mov r27=ar.rsc // M2 (12 cyc)
/*
* brl.cond doesn't work as intended because the linker would convert this branch
* into a branch to a PLT. Perhaps there will be a way to avoid this with some
@@ -111,6 +113,8 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
* instead.
*/
#ifdef CONFIG_ITANIUM
+(p6) add r14=-8,r14 // r14 <- addr of fsys_bubble_down entry
+ ;;
(p6) ld8 r14=[r14] // r14 <- fsys_bubble_down
;;
(p6) mov b7=r14
@@ -118,7 +122,7 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
#else
BRL_COND_FSYS_BUBBLE_DOWN(p6)
#endif
-
+ ssm psr.i
mov r10=-1
(p10) mov r8=EINVAL
(p9) mov r8=ENOSYS
diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index 7bbf019c9867..01572814abe4 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -58,9 +58,6 @@ EXPORT_SYMBOL(__strlen_user);
EXPORT_SYMBOL(__strncpy_from_user);
EXPORT_SYMBOL(__strnlen_user);
-#include
-EXPORT_SYMBOL(__ia64_syscall);
-
/* from arch/ia64/lib */
extern void __divsi3(void);
extern void __udivsi3(void);
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index 88b014381df5..c170be095ccd 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -129,14 +129,13 @@ static struct iosapic {
char __iomem *addr; /* base address of IOSAPIC */
unsigned int gsi_base; /* first GSI assigned to this IOSAPIC */
unsigned short num_rte; /* number of RTE in this IOSAPIC */
+ int rtes_inuse; /* # of RTEs in use on this IOSAPIC */
#ifdef CONFIG_NUMA
unsigned short node; /* numa node association via pxm */
#endif
} iosapic_lists[NR_IOSAPICS];
-static int num_iosapic;
-
-static unsigned char pcat_compat __initdata; /* 8259 compatibility flag */
+static unsigned char pcat_compat __devinitdata; /* 8259 compatibility flag */
static int iosapic_kmalloc_ok;
static LIST_HEAD(free_rte_list);
@@ -149,7 +148,7 @@ find_iosapic (unsigned int gsi)
{
int i;
- for (i = 0; i < num_iosapic; i++) {
+ for (i = 0; i < NR_IOSAPICS; i++) {
if ((unsigned) (gsi - iosapic_lists[i].gsi_base) < iosapic_lists[i].num_rte)
return i;
}
@@ -598,6 +597,7 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery,
rte->refcnt++;
list_add_tail(&rte->rte_list, &iosapic_intr_info[vector].rtes);
iosapic_intr_info[vector].count++;
+ iosapic_lists[index].rtes_inuse++;
}
else if (vector_is_shared(vector)) {
struct iosapic_intr_info *info = &iosapic_intr_info[vector];
@@ -778,7 +778,7 @@ void
iosapic_unregister_intr (unsigned int gsi)
{
unsigned long flags;
- int irq, vector;
+ int irq, vector, index;
irq_desc_t *idesc;
u32 low32;
unsigned long trigger, polarity;
@@ -819,6 +819,9 @@ iosapic_unregister_intr (unsigned int gsi)
list_del(&rte->rte_list);
iosapic_intr_info[vector].count--;
iosapic_free_rte(rte);
+ index = find_iosapic(gsi);
+ iosapic_lists[index].rtes_inuse--;
+ WARN_ON(iosapic_lists[index].rtes_inuse < 0);
trigger = iosapic_intr_info[vector].trigger;
polarity = iosapic_intr_info[vector].polarity;
@@ -952,30 +955,86 @@ iosapic_system_init (int system_pcat_compat)
}
}
-void __init
+static inline int
+iosapic_alloc (void)
+{
+ int index;
+
+ for (index = 0; index < NR_IOSAPICS; index++)
+ if (!iosapic_lists[index].addr)
+ return index;
+
+ printk(KERN_WARNING "%s: failed to allocate iosapic\n", __FUNCTION__);
+ return -1;
+}
+
+static inline void
+iosapic_free (int index)
+{
+ memset(&iosapic_lists[index], 0, sizeof(iosapic_lists[0]));
+}
+
+static inline int
+iosapic_check_gsi_range (unsigned int gsi_base, unsigned int ver)
+{
+ int index;
+ unsigned int gsi_end, base, end;
+
+ /* check gsi range */
+ gsi_end = gsi_base + ((ver >> 16) & 0xff);
+ for (index = 0; index < NR_IOSAPICS; index++) {
+ if (!iosapic_lists[index].addr)
+ continue;
+
+ base = iosapic_lists[index].gsi_base;
+ end = base + iosapic_lists[index].num_rte - 1;
+
+ if (gsi_base < base && gsi_end < base)
+ continue;/* OK */
+
+ if (gsi_base > end && gsi_end > end)
+ continue; /* OK */
+
+ return -EBUSY;
+ }
+ return 0;
+}
+
+int __devinit
iosapic_init (unsigned long phys_addr, unsigned int gsi_base)
{
- int num_rte;
+ int num_rte, err, index;
unsigned int isa_irq, ver;
char __iomem *addr;
+ unsigned long flags;
- addr = ioremap(phys_addr, 0);
- ver = iosapic_version(addr);
+ spin_lock_irqsave(&iosapic_lock, flags);
+ {
+ addr = ioremap(phys_addr, 0);
+ ver = iosapic_version(addr);
- /*
- * The MAX_REDIR register holds the highest input pin
- * number (starting from 0).
- * We add 1 so that we can use it for number of pins (= RTEs)
- */
- num_rte = ((ver >> 16) & 0xff) + 1;
+ if ((err = iosapic_check_gsi_range(gsi_base, ver))) {
+ iounmap(addr);
+ spin_unlock_irqrestore(&iosapic_lock, flags);
+ return err;
+ }
- iosapic_lists[num_iosapic].addr = addr;
- iosapic_lists[num_iosapic].gsi_base = gsi_base;
- iosapic_lists[num_iosapic].num_rte = num_rte;
+ /*
+ * The MAX_REDIR register holds the highest input pin
+ * number (starting from 0).
+ * We add 1 so that we can use it for number of pins (= RTEs)
+ */
+ num_rte = ((ver >> 16) & 0xff) + 1;
+
+ index = iosapic_alloc();
+ iosapic_lists[index].addr = addr;
+ iosapic_lists[index].gsi_base = gsi_base;
+ iosapic_lists[index].num_rte = num_rte;
#ifdef CONFIG_NUMA
- iosapic_lists[num_iosapic].node = MAX_NUMNODES;
+ iosapic_lists[index].node = MAX_NUMNODES;
#endif
- num_iosapic++;
+ }
+ spin_unlock_irqrestore(&iosapic_lock, flags);
if ((gsi_base == 0) && pcat_compat) {
/*
@@ -986,10 +1045,43 @@ iosapic_init (unsigned long phys_addr, unsigned int gsi_base)
for (isa_irq = 0; isa_irq < 16; ++isa_irq)
iosapic_override_isa_irq(isa_irq, isa_irq, IOSAPIC_POL_HIGH, IOSAPIC_EDGE);
}
+ return 0;
}
+#ifdef CONFIG_HOTPLUG
+int
+iosapic_remove (unsigned int gsi_base)
+{
+ int index, err = 0;
+ unsigned long flags;
+
+ spin_lock_irqsave(&iosapic_lock, flags);
+ {
+ index = find_iosapic(gsi_base);
+ if (index < 0) {
+ printk(KERN_WARNING "%s: No IOSAPIC for GSI base %u\n",
+ __FUNCTION__, gsi_base);
+ goto out;
+ }
+
+ if (iosapic_lists[index].rtes_inuse) {
+ err = -EBUSY;
+ printk(KERN_WARNING "%s: IOSAPIC for GSI base %u is busy\n",
+ __FUNCTION__, gsi_base);
+ goto out;
+ }
+
+ iounmap(iosapic_lists[index].addr);
+ iosapic_free(index);
+ }
+ out:
+ spin_unlock_irqrestore(&iosapic_lock, flags);
+ return err;
+}
+#endif /* CONFIG_HOTPLUG */
+
#ifdef CONFIG_NUMA
-void __init
+void __devinit
map_iosapic_to_node(unsigned int gsi_base, int node)
{
int index;
diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index 2bc085a73e30..3bb3a13c4047 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -1,7 +1,7 @@
/*
* arch/ia64/kernel/ivt.S
*
- * Copyright (C) 1998-2001, 2003 Hewlett-Packard Co
+ * Copyright (C) 1998-2001, 2003, 2005 Hewlett-Packard Co
* Stephane Eranian
* David Mosberger
* Copyright (C) 2000, 2002-2003 Intel Co
@@ -692,82 +692,118 @@ ENTRY(break_fault)
* to prevent leaking bits from kernel to user level.
*/
DBG_FAULT(11)
- mov r16=IA64_KR(CURRENT) // r16 = current task; 12 cycle read lat.
- mov r17=cr.iim
- mov r18=__IA64_BREAK_SYSCALL
- mov r21=ar.fpsr
- mov r29=cr.ipsr
- mov r19=b6
- mov r25=ar.unat
- mov r27=ar.rsc
- mov r26=ar.pfs
- mov r28=cr.iip
- mov r31=pr // prepare to save predicates
- mov r20=r1
+ mov.m r16=IA64_KR(CURRENT) // M2 r16 <- current task (12 cyc)
+ mov r29=cr.ipsr // M2 (12 cyc)
+ mov r31=pr // I0 (2 cyc)
+
+ mov r17=cr.iim // M2 (2 cyc)
+ mov.m r27=ar.rsc // M2 (12 cyc)
+ mov r18=__IA64_BREAK_SYSCALL // A
+
+ mov.m ar.rsc=0 // M2
+ mov.m r21=ar.fpsr // M2 (12 cyc)
+ mov r19=b6 // I0 (2 cyc)
;;
+ mov.m r23=ar.bspstore // M2 (12 cyc)
+ mov.m r24=ar.rnat // M2 (5 cyc)
+ mov.i r26=ar.pfs // I0 (2 cyc)
+
+ invala // M0|1
+ nop.m 0 // M
+ mov r20=r1 // A save r1
+
+ nop.m 0
+ movl r30=sys_call_table // X
+
+ mov r28=cr.iip // M2 (2 cyc)
+ cmp.eq p0,p7=r18,r17 // I0 is this a system call?
+(p7) br.cond.spnt non_syscall // B no ->
+ //
+ // From this point on, we are definitely on the syscall-path
+ // and we can use (non-banked) scratch registers.
+ //
+///////////////////////////////////////////////////////////////////////
+ mov r1=r16 // A move task-pointer to "addl"-addressable reg
+ mov r2=r16 // A setup r2 for ia64_syscall_setup
+ add r9=TI_FLAGS+IA64_TASK_SIZE,r16 // A r9 = ¤t_thread_info()->flags
+
adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
- cmp.eq p0,p7=r18,r17 // is this a system call? (p7 <- false, if so)
-(p7) br.cond.spnt non_syscall
- ;;
- ld1 r17=[r16] // load current->thread.on_ustack flag
- st1 [r16]=r0 // clear current->thread.on_ustack flag
- add r1=-IA64_TASK_THREAD_ON_USTACK_OFFSET,r16 // set r1 for MINSTATE_START_SAVE_MIN_VIRT
- ;;
- invala
-
- /* adjust return address so we skip over the break instruction: */
-
- extr.u r8=r29,41,2 // extract ei field from cr.ipsr
- ;;
- cmp.eq p6,p7=2,r8 // isr.ei==2?
- mov r2=r1 // setup r2 for ia64_syscall_setup
- ;;
-(p6) mov r8=0 // clear ei to 0
-(p6) adds r28=16,r28 // switch cr.iip to next bundle cr.ipsr.ei wrapped
-(p7) adds r8=1,r8 // increment ei to next slot
- ;;
- cmp.eq pKStk,pUStk=r0,r17 // are we in kernel mode already?
- dep r29=r8,r29,41,2 // insert new ei into cr.ipsr
- ;;
-
- // switch from user to kernel RBS:
- MINSTATE_START_SAVE_MIN_VIRT
- br.call.sptk.many b7=ia64_syscall_setup
- ;;
- MINSTATE_END_SAVE_MIN_VIRT // switch to bank 1
- ssm psr.ic | PSR_DEFAULT_BITS
- ;;
- srlz.i // guarantee that interruption collection is on
+ adds r15=-1024,r15 // A subtract 1024 from syscall number
mov r3=NR_syscalls - 1
;;
-(p15) ssm psr.i // restore psr.i
- // p10==true means out registers are more than 8 or r15's Nat is true
-(p10) br.cond.spnt.many ia64_ret_from_syscall
- ;;
- movl r16=sys_call_table
+ ld1.bias r17=[r16] // M0|1 r17 = current->thread.on_ustack flag
+ ld4 r9=[r9] // M0|1 r9 = current_thread_info()->flags
+ extr.u r8=r29,41,2 // I0 extract ei field from cr.ipsr
- adds r15=-1024,r15 // r15 contains the syscall number---subtract 1024
- movl r2=ia64_ret_from_syscall
+ shladd r30=r15,3,r30 // A r30 = sys_call_table + 8*(syscall-1024)
+ addl r22=IA64_RBS_OFFSET,r1 // A compute base of RBS
+ cmp.leu p6,p7=r15,r3 // A syscall number in range?
;;
- shladd r20=r15,3,r16 // r20 = sys_call_table + 8*(syscall-1024)
- cmp.leu p6,p7=r15,r3 // (syscall > 0 && syscall < 1024 + NR_syscalls) ?
- mov rp=r2 // set the real return addr
- ;;
-(p6) ld8 r20=[r20] // load address of syscall entry point
-(p7) movl r20=sys_ni_syscall
- add r2=TI_FLAGS+IA64_TASK_SIZE,r13
+ lfetch.fault.excl.nt1 [r22] // M0|1 prefetch RBS
+(p6) ld8 r30=[r30] // M0|1 load address of syscall entry point
+ tnat.nz.or p7,p0=r15 // I0 is syscall nr a NaT?
+
+ mov.m ar.bspstore=r22 // M2 switch to kernel RBS
+ cmp.eq p8,p9=2,r8 // A isr.ei==2?
;;
- ld4 r2=[r2] // r2 = current_thread_info()->flags
+
+(p8) mov r8=0 // A clear ei to 0
+(p7) movl r30=sys_ni_syscall // X
+
+(p8) adds r28=16,r28 // A switch cr.iip to next bundle
+(p9) adds r8=1,r8 // A increment ei to next slot
+ nop.i 0
;;
- and r2=_TIF_SYSCALL_TRACEAUDIT,r2 // mask trace or audit
+
+ mov.m r25=ar.unat // M2 (5 cyc)
+ dep r29=r8,r29,41,2 // I0 insert new ei into cr.ipsr
+ adds r15=1024,r15 // A restore original syscall number
+ //
+ // If any of the above loads miss in L1D, we'll stall here until
+ // the data arrives.
+ //
+///////////////////////////////////////////////////////////////////////
+ st1 [r16]=r0 // M2|3 clear current->thread.on_ustack flag
+ mov b6=r30 // I0 setup syscall handler branch reg early
+ cmp.eq pKStk,pUStk=r0,r17 // A were we on kernel stacks already?
+
+ and r9=_TIF_SYSCALL_TRACEAUDIT,r9 // A mask trace or audit
+ mov r18=ar.bsp // M2 (12 cyc)
+(pKStk) br.cond.spnt .break_fixup // B we're already in kernel-mode -- fix up RBS
;;
- cmp.eq p8,p0=r2,r0
- mov b6=r20
+.back_from_break_fixup:
+(pUStk) addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r1 // A compute base of memory stack
+ cmp.eq p14,p0=r9,r0 // A are syscalls being traced/audited?
+ br.call.sptk.many b7=ia64_syscall_setup // B
+1:
+ mov ar.rsc=0x3 // M2 set eager mode, pl 0, LE, loadrs=0
+ nop 0
+ bsw.1 // B (6 cyc) regs are saved, switch to bank 1
;;
-(p8) br.call.sptk.many b6=b6 // ignore this return addr
- br.cond.sptk ia64_trace_syscall
+
+ ssm psr.ic | PSR_DEFAULT_BITS // M2 now it's safe to re-enable intr.-collection
+ movl r3=ia64_ret_from_syscall // X
+ ;;
+
+ srlz.i // M0 ensure interruption collection is on
+ mov rp=r3 // I0 set the real return addr
+(p10) br.cond.spnt.many ia64_ret_from_syscall // B return if bad call-frame or r15 is a NaT
+
+(p15) ssm psr.i // M2 restore psr.i
+(p14) br.call.sptk.many b6=b6 // B invoke syscall-handker (ignore return addr)
+ br.cond.spnt.many ia64_trace_syscall // B do syscall-tracing thingamagic
// NOT REACHED
+///////////////////////////////////////////////////////////////////////
+ // On entry, we optimistically assumed that we're coming from user-space.
+ // For the rare cases where a system-call is done from within the kernel,
+ // we fix things up at this point:
+.break_fixup:
+ add r1=-IA64_PT_REGS_SIZE,sp // A allocate space for pt_regs structure
+ mov ar.rnat=r24 // M2 restore kernel's AR.RNAT
+ ;;
+ mov ar.bspstore=r23 // M2 restore kernel's AR.BSPSTORE
+ br.cond.sptk .back_from_break_fixup
END(break_fault)
.org ia64_ivt+0x3000
@@ -842,8 +878,6 @@ END(interrupt)
* - r31: saved pr
* - b0: original contents (to be saved)
* On exit:
- * - executing on bank 1 registers
- * - psr.ic enabled, interrupts restored
* - p10: TRUE if syscall is invoked with more than 8 out
* registers or r15's Nat is true
* - r1: kernel's gp
@@ -851,8 +885,11 @@ END(interrupt)
* - r8: -EINVAL if p10 is true
* - r12: points to kernel stack
* - r13: points to current task
+ * - r14: preserved (same as on entry)
+ * - p13: preserved
* - p15: TRUE if interrupts need to be re-enabled
* - ar.fpsr: set to kernel settings
+ * - b6: preserved (same as on entry)
*/
GLOBAL_ENTRY(ia64_syscall_setup)
#if PT(B6) != 0
@@ -920,10 +957,10 @@ GLOBAL_ENTRY(ia64_syscall_setup)
(p13) mov in5=-1
;;
st8 [r16]=r21,PT(R8)-PT(AR_FPSR) // save ar.fpsr
- tnat.nz p14,p0=in6
+ tnat.nz p13,p0=in6
cmp.lt p10,p9=r11,r8 // frame size can't be more than local+8
;;
- stf8 [r16]=f1 // ensure pt_regs.r8 != 0 (see handle_syscall_error)
+ mov r8=1
(p9) tnat.nz p10,p0=r15
adds r12=-16,r1 // switch to kernel memory stack (with 16 bytes of scratch)
@@ -934,9 +971,9 @@ GLOBAL_ENTRY(ia64_syscall_setup)
mov r13=r2 // establish `current'
movl r1=__gp // establish kernel global pointer
;;
-(p14) mov in6=-1
+ st8 [r16]=r8 // ensure pt_regs.r8 != 0 (see handle_syscall_error)
+(p13) mov in6=-1
(p8) mov in7=-1
- nop.i 0
cmp.eq pSys,pNonSys=r0,r0 // set pSys=1, pNonSys=0
movl r17=FPSR_DEFAULT
@@ -1007,6 +1044,8 @@ END(dispatch_illegal_op_fault)
FAULT(17)
ENTRY(non_syscall)
+ mov ar.rsc=r27 // restore ar.rsc before SAVE_MIN_WITH_COVER
+ ;;
SAVE_MIN_WITH_COVER
// There is no particular reason for this code to be here, other than that
@@ -1204,6 +1243,25 @@ END(disabled_fp_reg)
// 0x5600 Entry 26 (size 16 bundles) Nat Consumption (11,23,37,50)
ENTRY(nat_consumption)
DBG_FAULT(26)
+
+ mov r16=cr.ipsr
+ mov r17=cr.isr
+ mov r31=pr // save PR
+ ;;
+ and r18=0xf,r17 // r18 = cr.ipsr.code{3:0}
+ tbit.z p6,p0=r17,IA64_ISR_NA_BIT
+ ;;
+ cmp.ne.or p6,p0=IA64_ISR_CODE_LFETCH,r18
+ dep r16=-1,r16,IA64_PSR_ED_BIT,1
+(p6) br.cond.spnt 1f // branch if (cr.ispr.na == 0 || cr.ipsr.code{3:0} != LFETCH)
+ ;;
+ mov cr.ipsr=r16 // set cr.ipsr.na
+ mov pr=r31,-1
+ ;;
+ rfi
+
+1: mov pr=r31,-1
+ ;;
FAULT(26)
END(nat_consumption)
diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 5978823d5c63..3aa3167edbec 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -34,6 +34,7 @@
#include
#include
+#include
extern void jprobe_inst_return(void);
@@ -263,13 +264,33 @@ static inline void get_kprobe_inst(bundle_t *bundle, uint slot,
}
}
+/* Returns non-zero if the addr is in the Interrupt Vector Table */
+static inline int in_ivt_functions(unsigned long addr)
+{
+ return (addr >= (unsigned long)__start_ivt_text
+ && addr < (unsigned long)__end_ivt_text);
+}
+
static int valid_kprobe_addr(int template, int slot, unsigned long addr)
{
if ((slot > 2) || ((bundle_encoding[template][1] == L) && slot > 1)) {
- printk(KERN_WARNING "Attempting to insert unaligned kprobe at 0x%lx\n",
- addr);
+ printk(KERN_WARNING "Attempting to insert unaligned kprobe "
+ "at 0x%lx\n", addr);
return -EINVAL;
}
+
+ if (in_ivt_functions(addr)) {
+ printk(KERN_WARNING "Kprobes can't be inserted inside "
+ "IVT functions at 0x%lx\n", addr);
+ return -EINVAL;
+ }
+
+ if (slot == 1 && bundle_encoding[template][1] != L) {
+ printk(KERN_WARNING "Inserting kprobes on slot #1 "
+ "is not supported\n");
+ return -EINVAL;
+ }
+
return 0;
}
@@ -290,6 +311,94 @@ static inline void set_current_kprobe(struct kprobe *p)
current_kprobe = p;
}
+static void kretprobe_trampoline(void)
+{
+}
+
+/*
+ * At this point the target function has been tricked into
+ * returning into our trampoline. Lookup the associated instance
+ * and then:
+ * - call the handler function
+ * - cleanup by marking the instance as unused
+ * - long jump back to the original return address
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+ struct kretprobe_instance *ri = NULL;
+ struct hlist_head *head;
+ struct hlist_node *node, *tmp;
+ unsigned long orig_ret_address = 0;
+ unsigned long trampoline_address =
+ ((struct fnptr *)kretprobe_trampoline)->ip;
+
+ head = kretprobe_inst_table_head(current);
+
+ /*
+ * It is possible to have multiple instances associated with a given
+ * task either because an multiple functions in the call path
+ * have a return probe installed on them, and/or more then one return
+ * return probe was registered for a target function.
+ *
+ * We can handle this because:
+ * - instances are always inserted at the head of the list
+ * - when multiple return probes are registered for the same
+ * function, the first instance's ret_addr will point to the
+ * real return address, and all the rest will point to
+ * kretprobe_trampoline
+ */
+ hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+ if (ri->task != current)
+ /* another task is sharing our hash bucket */
+ continue;
+
+ if (ri->rp && ri->rp->handler)
+ ri->rp->handler(ri, regs);
+
+ orig_ret_address = (unsigned long)ri->ret_addr;
+ recycle_rp_inst(ri);
+
+ if (orig_ret_address != trampoline_address)
+ /*
+ * This is the real return address. Any other
+ * instances associated with this task are for
+ * other calls deeper on the call stack
+ */
+ break;
+ }
+
+ BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+ regs->cr_iip = orig_ret_address;
+
+ unlock_kprobes();
+ preempt_enable_no_resched();
+
+ /*
+ * By returning a non-zero value, we are telling
+ * kprobe_handler() that we have handled unlocking
+ * and re-enabling preemption.
+ */
+ return 1;
+}
+
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+ struct kretprobe_instance *ri;
+
+ if ((ri = get_free_rp_inst(rp)) != NULL) {
+ ri->rp = rp;
+ ri->task = current;
+ ri->ret_addr = (kprobe_opcode_t *)regs->b0;
+
+ /* Replace the return addr with trampoline addr */
+ regs->b0 = ((struct fnptr *)kretprobe_trampoline)->ip;
+
+ add_rp_inst(ri);
+ } else {
+ rp->nmissed++;
+ }
+}
+
int arch_prepare_kprobe(struct kprobe *p)
{
unsigned long addr = (unsigned long) p->addr;
@@ -492,8 +601,8 @@ static int pre_kprobes_handler(struct die_args *args)
if (p->pre_handler && p->pre_handler(p, regs))
/*
* Our pre-handler is specifically requesting that we just
- * do a return. This is handling the case where the
- * pre-handler is really our special jprobe pre-handler.
+ * do a return. This is used for both the jprobe pre-handler
+ * and the kretprobe trampoline
*/
return 1;
@@ -599,3 +708,14 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
*regs = jprobe_saved_regs;
return 1;
}
+
+static struct kprobe trampoline_p = {
+ .pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+ trampoline_p.addr =
+ (kprobe_opcode_t *)((struct fnptr *)kretprobe_trampoline)->ip;
+ return register_kprobe(&trampoline_p);
+}
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index ebb71f3d6d19..6e35bff05d59 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -27,6 +27,7 @@
#include
#include
#include
+#include
#include
#include
@@ -707,6 +708,13 @@ kernel_thread_helper (int (*fn)(void *), void *arg)
void
flush_thread (void)
{
+ /*
+ * Remove function-return probe instances associated with this task
+ * and put them back on the free list. Do not insert an exit probe for
+ * this function, it will be disabled by kprobe_flush_task if you do.
+ */
+ kprobe_flush_task(current);
+
/* drop floating-point and debug-register state if it exists: */
current->thread.flags &= ~(IA64_THREAD_FPH_VALID | IA64_THREAD_DBG_VALID);
ia64_drop_fpu(current);
@@ -721,6 +729,14 @@ flush_thread (void)
void
exit_thread (void)
{
+
+ /*
+ * Remove function-return probe instances associated with this task
+ * and put them back on the free list. Do not insert an exit probe for
+ * this function, it will be disabled by kprobe_flush_task if you do.
+ */
+ kprobe_flush_task(current);
+
ia64_drop_fpu(current);
#ifdef CONFIG_PERFMON
/* if needed, stop monitoring and flush state to perfmon context */
diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index 6d57aebad485..bbb8bc7c0552 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -725,12 +725,32 @@ convert_to_non_syscall (struct task_struct *child, struct pt_regs *pt,
break;
}
+ /*
+ * Note: at the time of this call, the target task is blocked
+ * in notify_resume_user() and by clearling PRED_LEAVE_SYSCALL
+ * (aka, "pLvSys") we redirect execution from
+ * .work_pending_syscall_end to .work_processed_kernel.
+ */
unw_get_pr(&prev_info, &pr);
- pr &= ~(1UL << PRED_SYSCALL);
+ pr &= ~((1UL << PRED_SYSCALL) | (1UL << PRED_LEAVE_SYSCALL));
pr |= (1UL << PRED_NON_SYSCALL);
unw_set_pr(&prev_info, pr);
pt->cr_ifs = (1UL << 63) | cfm;
+ /*
+ * Clear the memory that is NOT written on syscall-entry to
+ * ensure we do not leak kernel-state to user when execution
+ * resumes.
+ */
+ pt->r2 = 0;
+ pt->r3 = 0;
+ pt->r14 = 0;
+ memset(&pt->r16, 0, 16*8); /* clear r16-r31 */
+ memset(&pt->f6, 0, 6*16); /* clear f6-f11 */
+ pt->b7 = 0;
+ pt->ar_ccv = 0;
+ pt->ar_csd = 0;
+ pt->ar_ssd = 0;
}
static int
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index d14692e0920a..2693e1522d7c 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -72,6 +72,8 @@ DEFINE_PER_CPU(unsigned long, ia64_phys_stacked_size_p8);
unsigned long ia64_cycles_per_usec;
struct ia64_boot_param *ia64_boot_param;
struct screen_info screen_info;
+unsigned long vga_console_iobase;
+unsigned long vga_console_membase;
unsigned long ia64_max_cacheline_size;
unsigned long ia64_iobase; /* virtual address for I/O accesses */
@@ -273,23 +275,25 @@ io_port_init (void)
static inline int __init
early_console_setup (char *cmdline)
{
+ int earlycons = 0;
+
#ifdef CONFIG_SERIAL_SGI_L1_CONSOLE
{
extern int sn_serial_console_early_setup(void);
if (!sn_serial_console_early_setup())
- return 0;
+ earlycons++;
}
#endif
#ifdef CONFIG_EFI_PCDP
if (!efi_setup_pcdp_console(cmdline))
- return 0;
+ earlycons++;
#endif
#ifdef CONFIG_SERIAL_8250_CONSOLE
if (!early_serial_console_init(cmdline))
- return 0;
+ earlycons++;
#endif
- return -1;
+ return (earlycons) ? 0 : -1;
}
static inline void
diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c
index b49d4ddaab93..0166a9847095 100644
--- a/arch/ia64/kernel/smp.c
+++ b/arch/ia64/kernel/smp.c
@@ -231,13 +231,16 @@ smp_flush_tlb_all (void)
void
smp_flush_tlb_mm (struct mm_struct *mm)
{
+ preempt_disable();
/* this happens for the common case of a single-threaded fork(): */
if (likely(mm == current->active_mm && atomic_read(&mm->mm_users) == 1))
{
local_finish_flush_tlb_mm(mm);
+ preempt_enable();
return;
}
+ preempt_enable();
/*
* We could optimize this further by using mm->cpu_vm_mask to track which CPUs
* have been running in the address space. It's not clear that this is worth the
diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
index b9f0db4c1b04..a676e79e0681 100644
--- a/arch/ia64/kernel/vmlinux.lds.S
+++ b/arch/ia64/kernel/vmlinux.lds.S
@@ -8,6 +8,11 @@
#define LOAD_OFFSET (KERNEL_START - KERNEL_TR_PAGE_SIZE)
#include
+#define IVT_TEXT \
+ VMLINUX_SYMBOL(__start_ivt_text) = .; \
+ *(.text.ivt) \
+ VMLINUX_SYMBOL(__end_ivt_text) = .;
+
OUTPUT_FORMAT("elf64-ia64-little")
OUTPUT_ARCH(ia64)
ENTRY(phys_start)
@@ -39,7 +44,7 @@ SECTIONS
.text : AT(ADDR(.text) - LOAD_OFFSET)
{
- *(.text.ivt)
+ IVT_TEXT
*(.text)
SCHED_TEXT
LOCK_TEXT
diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index e3fc4edea113..720a861f88be 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -312,7 +312,7 @@ pci_acpi_scan_root(struct acpi_device *device, int domain, int bus)
acpi_walk_resources(device->handle, METHOD_NAME__CRS, add_window,
&info);
- pbus = pci_scan_bus(bus, &pci_root_ops, controller);
+ pbus = pci_scan_bus_parented(NULL, bus, &pci_root_ops, controller);
if (pbus)
pcibios_setup_root_windows(pbus, controller);
@@ -373,6 +373,25 @@ void pcibios_bus_to_resource(struct pci_dev *dev,
res->end = region->end + offset;
}
+static int __devinit is_valid_resource(struct pci_dev *dev, int idx)
+{
+ unsigned int i, type_mask = IORESOURCE_IO | IORESOURCE_MEM;
+ struct resource *devr = &dev->resource[idx];
+
+ if (!dev->bus)
+ return 0;
+ for (i=0; ibus->resource[i];
+
+ if (!busr || ((busr->flags ^ devr->flags) & type_mask))
+ continue;
+ if ((devr->start) && (devr->start >= busr->start) &&
+ (devr->end <= busr->end))
+ return 1;
+ }
+ return 0;
+}
+
static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev)
{
struct pci_bus_region region;
@@ -386,7 +405,8 @@ static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev)
region.start = dev->resource[i].start;
region.end = dev->resource[i].end;
pcibios_bus_to_resource(dev, &dev->resource[i], ®ion);
- pci_claim_resource(dev, i);
+ if ((is_valid_resource(dev, i)))
+ pci_claim_resource(dev, i);
}
}
@@ -398,6 +418,10 @@ pcibios_fixup_bus (struct pci_bus *b)
{
struct pci_dev *dev;
+ if (b->self) {
+ pci_read_bridge_bases(b);
+ pcibios_fixup_device_resources(b->self);
+ }
list_for_each_entry(dev, &b->devices, bus_list)
pcibios_fixup_device_resources(dev);
@@ -418,18 +442,24 @@ pcibios_enable_resources (struct pci_dev *dev, int mask)
u16 cmd, old_cmd;
int idx;
struct resource *r;
+ unsigned long type_mask = IORESOURCE_IO | IORESOURCE_MEM;
if (!dev)
return -EINVAL;
pci_read_config_word(dev, PCI_COMMAND, &cmd);
old_cmd = cmd;
- for (idx=0; idx<6; idx++) {
+ for (idx=0; idxresource[idx];
+ if (!(r->flags & type_mask))
+ continue;
+ if ((idx == PCI_ROM_RESOURCE) &&
+ (!(r->flags & IORESOURCE_ROM_ENABLE)))
+ continue;
if (!r->start && r->end) {
printk(KERN_ERR
"PCI: Device %s not available because of resource collisions\n",
@@ -441,8 +471,6 @@ pcibios_enable_resources (struct pci_dev *dev, int mask)
if (r->flags & IORESOURCE_MEM)
cmd |= PCI_COMMAND_MEMORY;
}
- if (dev->resource[PCI_ROM_RESOURCE].start)
- cmd |= PCI_COMMAND_MEMORY;
if (cmd != old_cmd) {
printk("PCI: Enabling device %s (%04x -> %04x)\n", pci_name(dev), old_cmd, cmd);
pci_write_config_word(dev, PCI_COMMAND, cmd);
diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c
index 9e07f5463f21..783eb4323847 100644
--- a/arch/ia64/sn/kernel/io_init.c
+++ b/arch/ia64/sn/kernel/io_init.c
@@ -384,7 +384,7 @@ static int __init sn_pci_init(void)
extern void register_sn_procfs(void);
#endif
- if (!ia64_platform_is("sn2") || IS_RUNNING_ON_SIMULATOR())
+ if (!ia64_platform_is("sn2") || IS_RUNNING_ON_FAKE_PROM())
return 0;
/*
diff --git a/arch/ia64/sn/kernel/iomv.c b/arch/ia64/sn/kernel/iomv.c
index fec6d8b8237b..7ce3cdad627b 100644
--- a/arch/ia64/sn/kernel/iomv.c
+++ b/arch/ia64/sn/kernel/iomv.c
@@ -9,12 +9,16 @@
#include
#include
#include
+#include
#include
#include
#include
#include
#include
+#define IS_LEGACY_VGA_IOPORT(p) \
+ (((p) >= 0x3b0 && (p) <= 0x3bb) || ((p) >= 0x3c0 && (p) <= 0x3df))
+
/**
* sn_io_addr - convert an in/out port to an i/o address
* @port: port to convert
@@ -26,6 +30,8 @@
void *sn_io_addr(unsigned long port)
{
if (!IS_RUNNING_ON_SIMULATOR()) {
+ if (IS_LEGACY_VGA_IOPORT(port))
+ port += vga_console_iobase;
/* On sn2, legacy I/O ports don't point at anything */
if (port < (64 * 1024))
return NULL;
diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index 44bfc7f318cb..22e10d282c7f 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -36,6 +36,7 @@
#include
#include
#include
+#include
#include
#include
#include
@@ -95,6 +96,7 @@ u8 sn_coherency_id;
EXPORT_SYMBOL(sn_coherency_id);
u8 sn_region_size;
EXPORT_SYMBOL(sn_region_size);
+int sn_prom_type; /* 0=hardware, 1=medusa/realprom, 2=medusa/fakeprom */
short physical_node_map[MAX_PHYSNODE_ID];
@@ -273,14 +275,17 @@ void __init sn_setup(char **cmdline_p)
ia64_sn_plat_set_error_handling_features();
- /*
- * If the generic code has enabled vga console support - lets
- * get rid of it again. This is a kludge for the fact that ACPI
- * currtently has no way of informing us if legacy VGA is available
- * or not.
- */
#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
- if (conswitchp == &vga_con) {
+ /*
+ * If there was a primary vga adapter identified through the
+ * EFI PCDP table, make it the preferred console. Otherwise
+ * zero out conswitchp.
+ */
+
+ if (vga_console_membase) {
+ /* usable vga ... make tty0 the preferred default console */
+ add_preferred_console("tty", 0, NULL);
+ } else {
printk(KERN_DEBUG "SGI: Disabling VGA console\n");
#ifdef CONFIG_DUMMY_CONSOLE
conswitchp = &dummy_con;
@@ -350,7 +355,7 @@ void __init sn_setup(char **cmdline_p)
ia64_mark_idle = &snidle;
- /*
+ /*
* For the bootcpu, we do this here. All other cpus will make the
* call as part of cpu_init in slave cpu initialization.
*/
@@ -397,7 +402,7 @@ static void __init sn_init_pdas(char **cmdline_p)
nodepdaindr[cnode] =
alloc_bootmem_node(NODE_DATA(cnode), sizeof(nodepda_t));
memset(nodepdaindr[cnode], 0, sizeof(nodepda_t));
- memset(nodepdaindr[cnode]->phys_cpuid, -1,
+ memset(nodepdaindr[cnode]->phys_cpuid, -1,
sizeof(nodepdaindr[cnode]->phys_cpuid));
}
@@ -427,7 +432,7 @@ static void __init sn_init_pdas(char **cmdline_p)
}
/*
- * Initialize the per node hubdev. This includes IO Nodes and
+ * Initialize the per node hubdev. This includes IO Nodes and
* headless/memless nodes.
*/
for (cnode = 0; cnode < numionodes; cnode++) {
@@ -455,6 +460,14 @@ void __init sn_cpu_init(void)
int i;
static int wars_have_been_checked;
+ if (smp_processor_id() == 0 && IS_MEDUSA()) {
+ if (ia64_sn_is_fake_prom())
+ sn_prom_type = 2;
+ else
+ sn_prom_type = 1;
+ printk("Running on medusa with %s PROM\n", (sn_prom_type == 1) ? "real" : "fake");
+ }
+
memset(pda, 0, sizeof(pda));
if (ia64_sn_get_sn_info(0, &sn_hub_info->shub2, &sn_hub_info->nasid_bitmask, &sn_hub_info->nasid_shift,
&sn_system_size, &sn_sharing_domain_size, &sn_partition_id,
@@ -520,7 +533,7 @@ void __init sn_cpu_init(void)
*/
{
u64 pio1[] = {SH1_PIO_WRITE_STATUS_0, 0, SH1_PIO_WRITE_STATUS_1, 0};
- u64 pio2[] = {SH2_PIO_WRITE_STATUS_0, SH2_PIO_WRITE_STATUS_1,
+ u64 pio2[] = {SH2_PIO_WRITE_STATUS_0, SH2_PIO_WRITE_STATUS_1,
SH2_PIO_WRITE_STATUS_2, SH2_PIO_WRITE_STATUS_3};
u64 *pio;
pio = is_shub1() ? pio1 : pio2;
@@ -552,6 +565,10 @@ static void __init scan_for_ionodes(void)
int nasid = 0;
lboard_t *brd;
+ /* fakeprom does not support klgraph */
+ if (IS_RUNNING_ON_FAKE_PROM())
+ return;
+
/* Setup ionodes with memory */
for (nasid = 0; nasid < MAX_PHYSNODE_ID; nasid += 2) {
char *klgraph_header;
@@ -563,8 +580,6 @@ static void __init scan_for_ionodes(void)
cnodeid = -1;
klgraph_header = __va(ia64_sn_get_klconfig_addr(nasid));
if (!klgraph_header) {
- if (IS_RUNNING_ON_SIMULATOR())
- continue;
BUG(); /* All nodes must have klconfig tables! */
}
cnodeid = nasid_to_cnodeid(nasid);
@@ -630,8 +645,8 @@ int
nasid_slice_to_cpuid(int nasid, int slice)
{
long cpu;
-
- for (cpu=0; cpu < NR_CPUS; cpu++)
+
+ for (cpu=0; cpu < NR_CPUS; cpu++)
if (cpuid_to_nasid(cpu) == nasid &&
cpuid_to_slice(cpu) == slice)
return cpu;
diff --git a/arch/ia64/sn/kernel/sn2/ptc_deadlock.S b/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
index 7947312801ec..96cb71d15682 100644
--- a/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
+++ b/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
@@ -6,6 +6,7 @@
* Copyright (C) 2000-2004 Silicon Graphics, Inc. All rights reserved.
*/
+#include
#include
#define DEADLOCKBIT SH_PIO_WRITE_STATUS_WRITE_DEADLOCK_SHFT
diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c
index a087b274847e..8716f4d5314b 100644
--- a/arch/ia64/sn/kernel/tiocx.c
+++ b/arch/ia64/sn/kernel/tiocx.c
@@ -204,8 +204,8 @@ cx_device_register(nasid_t nasid, int part_num, int mfg_num,
cx_dev->dev.parent = NULL;
cx_dev->dev.bus = &tiocx_bus_type;
cx_dev->dev.release = tiocx_bus_release;
- snprintf(cx_dev->dev.bus_id, BUS_ID_SIZE, "%d.0x%x",
- cx_dev->cx_id.nasid, cx_dev->cx_id.part_num);
+ snprintf(cx_dev->dev.bus_id, BUS_ID_SIZE, "%d",
+ cx_dev->cx_id.nasid);
device_register(&cx_dev->dev);
get_device(&cx_dev->dev);
@@ -236,7 +236,6 @@ int cx_device_unregister(struct cx_dev *cx_dev)
*/
static int cx_device_reload(struct cx_dev *cx_dev)
{
- device_remove_file(&cx_dev->dev, &dev_attr_cxdev_control);
cx_device_unregister(cx_dev);
return cx_device_register(cx_dev->cx_id.nasid, cx_dev->cx_id.part_num,
cx_dev->cx_id.mfg_num, cx_dev->hubdev);
@@ -383,6 +382,7 @@ static int is_fpga_brick(int nasid)
switch (tiocx_btchar_get(nasid)) {
case L1_BRICKTYPE_SA:
case L1_BRICKTYPE_ATHENA:
+ case L1_BRICKTYPE_DAYTONA:
return 1;
}
return 0;
@@ -409,7 +409,7 @@ static int tiocx_reload(struct cx_dev *cx_dev)
uint64_t cx_id;
cx_id =
- *(volatile int32_t *)(TIO_SWIN_BASE(nasid, TIOCX_CORELET) +
+ *(volatile uint64_t *)(TIO_SWIN_BASE(nasid, TIOCX_CORELET) +
WIDGET_ID);
part_num = XWIDGET_PART_NUM(cx_id);
mfg_num = XWIDGET_MFG_NUM(cx_id);
@@ -458,6 +458,10 @@ static ssize_t store_cxdev_control(struct device *dev, struct device_attribute *
switch (n) {
case 1:
+ tio_corelet_reset(cx_dev->cx_id.nasid, TIOCX_CORELET);
+ tiocx_reload(cx_dev);
+ break;
+ case 2:
tiocx_reload(cx_dev);
break;
case 3:
@@ -537,7 +541,7 @@ static void __exit tiocx_exit(void)
bus_unregister(&tiocx_bus_type);
}
-module_init(tiocx_init);
+subsys_initcall(tiocx_init);
module_exit(tiocx_exit);
/************************************************************************
diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c
index 8dae9eb45456..05aa8c2fe9bb 100644
--- a/arch/ia64/sn/pci/tioca_provider.c
+++ b/arch/ia64/sn/pci/tioca_provider.c
@@ -336,7 +336,7 @@ tioca_dma_d48(struct pci_dev *pdev, uint64_t paddr)
if (!ct_addr)
return 0;
- bus_addr = (dma_addr_t) (ct_addr & 0xffffffffffff);
+ bus_addr = (dma_addr_t) (ct_addr & 0xffffffffffffUL);
node_upper = ct_addr >> 48;
if (node_upper > 64) {
@@ -464,7 +464,7 @@ map_return:
* For mappings created using the direct modes (64 or 48) there are no
* resources to release.
*/
-void
+static void
tioca_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir)
{
int i, entry;
@@ -514,7 +514,7 @@ tioca_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir)
* The mapping mode used is based on the devices dma_mask. As a last resort
* use the GART mapped mode.
*/
-uint64_t
+static uint64_t
tioca_dma_map(struct pci_dev *pdev, uint64_t paddr, size_t byte_count)
{
uint64_t mapaddr;
@@ -580,7 +580,7 @@ tioca_error_intr_handler(int irq, void *arg, struct pt_regs *pt)
* On successful setup, returns the kernel version of tioca_common back to
* the caller.
*/
-void *
+static void *
tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft)
{
struct tioca_common *tioca_common;
diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
index 508026ae5842..65ee15396ffd 100644
--- a/arch/mips/kernel/signal.c
+++ b/arch/mips/kernel/signal.c
@@ -457,7 +457,7 @@ static int do_signal(sigset_t *oldset, struct pt_regs *regs)
if (!user_mode(regs))
return 1;
- if (try_to_freeze(0))
+ if (try_to_freeze())
goto no_signal;
if (!oldset)
diff --git a/arch/parisc/configs/712_defconfig b/arch/parisc/configs/712_defconfig
index 872085dea8a8..6efaa9293eef 100644
--- a/arch/parisc/configs/712_defconfig
+++ b/arch/parisc/configs/712_defconfig
@@ -506,7 +506,7 @@ CONFIG_HW_CONSOLE=y
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=8
+CONFIG_SERIAL_8250_NR_UARTS=17
CONFIG_SERIAL_8250_EXTENDED=y
CONFIG_SERIAL_8250_MANY_PORTS=y
CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/configs/a500_defconfig b/arch/parisc/configs/a500_defconfig
index d28ebfa1070d..30fc03ed0cfb 100644
--- a/arch/parisc/configs/a500_defconfig
+++ b/arch/parisc/configs/a500_defconfig
@@ -662,7 +662,7 @@ CONFIG_HW_CONSOLE=y
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
CONFIG_SERIAL_8250_CS=m
-CONFIG_SERIAL_8250_NR_UARTS=8
+CONFIG_SERIAL_8250_NR_UARTS=17
CONFIG_SERIAL_8250_EXTENDED=y
CONFIG_SERIAL_8250_MANY_PORTS=y
CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/configs/b180_defconfig b/arch/parisc/configs/b180_defconfig
index 1700d7aec686..46c9511f3229 100644
--- a/arch/parisc/configs/b180_defconfig
+++ b/arch/parisc/configs/b180_defconfig
@@ -514,7 +514,7 @@ CONFIG_HW_CONSOLE=y
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_NR_UARTS=13
CONFIG_SERIAL_8250_EXTENDED=y
CONFIG_SERIAL_8250_MANY_PORTS=y
CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/configs/c3000_defconfig b/arch/parisc/configs/c3000_defconfig
index b27980161c31..67aca6ccc9b0 100644
--- a/arch/parisc/configs/c3000_defconfig
+++ b/arch/parisc/configs/c3000_defconfig
@@ -661,7 +661,7 @@ CONFIG_HW_CONSOLE=y
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_NR_UARTS=13
CONFIG_SERIAL_8250_EXTENDED=y
CONFIG_SERIAL_8250_MANY_PORTS=y
CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/parisc/defconfig b/arch/parisc/defconfig
index ebd6301aa599..fdae21c503d7 100644
--- a/arch/parisc/defconfig
+++ b/arch/parisc/defconfig
@@ -517,7 +517,7 @@ CONFIG_HW_CONSOLE=y
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_NR_UARTS=13
CONFIG_SERIAL_8250_EXTENDED=y
CONFIG_SERIAL_8250_MANY_PORTS=y
CONFIG_SERIAL_8250_SHARE_IRQ=y
diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S
index b6a63a49a232..191a8def3bdb 100644
--- a/arch/ppc/kernel/misc.S
+++ b/arch/ppc/kernel/misc.S
@@ -1449,3 +1449,5 @@ _GLOBAL(sys_call_table)
.long sys_request_key /* 270 */
.long sys_keyctl
.long sys_waitid
+ .long sys_ioprio_set
+ .long sys_ioprio_get
diff --git a/arch/ppc/kernel/pci.c b/arch/ppc/kernel/pci.c
index 6d7b92d72458..70cfb6ffd877 100644
--- a/arch/ppc/kernel/pci.c
+++ b/arch/ppc/kernel/pci.c
@@ -1495,7 +1495,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
*offset += hose->pci_mem_offset;
res_bit = IORESOURCE_MEM;
} else {
- io_offset = (unsigned long)hose->io_base_virt;
+ io_offset = hose->io_base_virt - ___IO_BASE;
*offset += io_offset;
res_bit = IORESOURCE_IO;
}
@@ -1522,7 +1522,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
/* found it! construct the final physical address */
if (mmap_state == pci_mmap_io)
- *offset += hose->io_base_phys - _IO_BASE;
+ *offset += hose->io_base_phys - io_offset;
return rp;
}
@@ -1739,6 +1739,23 @@ long sys_pciconfig_iobase(long which, unsigned long bus, unsigned long devfn)
return result;
}
+void pci_resource_to_user(const struct pci_dev *dev, int bar,
+ const struct resource *rsrc,
+ u64 *start, u64 *end)
+{
+ struct pci_controller *hose = pci_bus_to_hose(dev->bus->number);
+ unsigned long offset = 0;
+
+ if (hose == NULL)
+ return;
+
+ if (rsrc->flags & IORESOURCE_IO)
+ offset = ___IO_BASE - hose->io_base_virt + hose->io_base_phys;
+
+ *start = rsrc->start + offset;
+ *end = rsrc->end + offset;
+}
+
void __init
pci_init_resource(struct resource *res, unsigned long start, unsigned long end,
int flags, char *name)
diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c
index 334ef4150d92..6164a2b34733 100644
--- a/arch/ppc/mm/init.c
+++ b/arch/ppc/mm/init.c
@@ -606,9 +606,19 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address,
struct page *page = pfn_to_page(pfn);
if (!PageReserved(page)
&& !test_bit(PG_arch_1, &page->flags)) {
- if (vma->vm_mm == current->active_mm)
+ if (vma->vm_mm == current->active_mm) {
+#ifdef CONFIG_8xx
+ /* On 8xx, cache control instructions (particularly
+ * "dcbst" from flush_dcache_icache) fault as write
+ * operation if there is an unpopulated TLB entry
+ * for the address in question. To workaround that,
+ * we invalidate the TLB here, thus avoiding dcbst
+ * misbehaviour.
+ */
+ _tlbie(address);
+#endif
__flush_dcache_icache((void *) address);
- else
+ } else
flush_dcache_icache_page(page);
set_bit(PG_arch_1, &page->flags);
}
diff --git a/arch/ppc/platforms/pmac_sleep.S b/arch/ppc/platforms/pmac_sleep.S
index f459ade1bd63..016a74649155 100644
--- a/arch/ppc/platforms/pmac_sleep.S
+++ b/arch/ppc/platforms/pmac_sleep.S
@@ -46,7 +46,7 @@
.section .text
.align 5
-#if defined(CONFIG_PMAC_PBOOK) || defined(CONFIG_CPU_FREQ_PMAC)
+#if defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ_PMAC)
/* This gets called by via-pmu.c late during the sleep process.
* The PMU was already send the sleep command and will shut us down
@@ -382,7 +382,7 @@ turn_on_mmu:
isync
rfi
-#endif /* defined(CONFIG_PMAC_PBOOK) || defined(CONFIG_CPU_FREQ) */
+#endif /* defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ) */
.section .data
.balign L1_CACHE_LINE_SIZE
diff --git a/arch/ppc/platforms/pmac_time.c b/arch/ppc/platforms/pmac_time.c
index de60ccc7db9f..778ce4fec368 100644
--- a/arch/ppc/platforms/pmac_time.c
+++ b/arch/ppc/platforms/pmac_time.c
@@ -206,7 +206,7 @@ via_calibrate_decr(void)
return 1;
}
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
/*
* Reset the time after a sleep.
*/
@@ -238,7 +238,7 @@ time_sleep_notify(struct pmu_sleep_notifier *self, int when)
static struct pmu_sleep_notifier time_sleep_notifier __pmacdata = {
time_sleep_notify, SLEEP_LEVEL_MISC,
};
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
/*
* Query the OF and get the decr frequency.
@@ -251,9 +251,9 @@ pmac_calibrate_decr(void)
struct device_node *cpu;
unsigned int freq, *fp;
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
pmu_register_sleep_notifier(&time_sleep_notifier);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
/* We assume MacRISC2 machines have correct device-tree
* calibration. That's better since the VIA itself seems
diff --git a/arch/ppc/platforms/sandpoint.c b/arch/ppc/platforms/sandpoint.c
index 70e58f43f2b8..8b149c2fc54f 100644
--- a/arch/ppc/platforms/sandpoint.c
+++ b/arch/ppc/platforms/sandpoint.c
@@ -324,6 +324,7 @@ sandpoint_setup_arch(void)
pdata[1].irq = 0;
pdata[1].mapbase = 0;
}
+ }
printk(KERN_INFO "Motorola SPS Sandpoint Test Platform\n");
printk(KERN_INFO "Port by MontaVista Software, Inc. (source@mvista.com)\n");
diff --git a/arch/ppc/syslib/open_pic.c b/arch/ppc/syslib/open_pic.c
index b45d8268bf93..ad39b86ca92c 100644
--- a/arch/ppc/syslib/open_pic.c
+++ b/arch/ppc/syslib/open_pic.c
@@ -370,8 +370,9 @@ void __init openpic_init(int offset)
/* Initialize IPI interrupts */
if ( ppc_md.progress ) ppc_md.progress("openpic: ipi",0x3bb);
for (i = 0; i < OPENPIC_NUM_IPI; i++) {
- /* Disabled, Priority 10..13 */
- openpic_initipi(i, 10+i, OPENPIC_VEC_IPI+i+offset);
+ /* Disabled, increased priorities 10..13 */
+ openpic_initipi(i, OPENPIC_PRIORITY_IPI_BASE+i,
+ OPENPIC_VEC_IPI+i+offset);
/* IPIs are per-CPU */
irq_desc[OPENPIC_VEC_IPI+i+offset].status |= IRQ_PER_CPU;
irq_desc[OPENPIC_VEC_IPI+i+offset].handler = &open_pic_ipi;
@@ -399,8 +400,9 @@ void __init openpic_init(int offset)
if (sense & IRQ_SENSE_MASK)
irq_desc[i+offset].status = IRQ_LEVEL;
- /* Enabled, Priority 8 */
- openpic_initirq(i, 8, i+offset, (sense & IRQ_POLARITY_MASK),
+ /* Enabled, Default priority */
+ openpic_initirq(i, OPENPIC_PRIORITY_DEFAULT, i+offset,
+ (sense & IRQ_POLARITY_MASK),
(sense & IRQ_SENSE_MASK));
/* Processor 0 */
openpic_mapirq(i, CPU_MASK_CPU0, CPU_MASK_NONE);
@@ -655,6 +657,18 @@ static void __init openpic_maptimer(u_int timer, cpumask_t cpumask)
cpus_addr(phys)[0]);
}
+/*
+ * Change the priority of an interrupt
+ */
+void __init
+openpic_set_irq_priority(u_int irq, u_int pri)
+{
+ check_arg_irq(irq);
+ openpic_safe_writefield(&ISR[irq - open_pic_irq_offset]->Vector_Priority,
+ OPENPIC_PRIORITY_MASK,
+ pri << OPENPIC_PRIORITY_SHIFT);
+}
+
/*
* Initalize the interrupt source which will generate an NMI.
* This raises the interrupt's priority from 8 to 9.
@@ -665,9 +679,7 @@ void __init
openpic_init_nmi_irq(u_int irq)
{
check_arg_irq(irq);
- openpic_safe_writefield(&ISR[irq - open_pic_irq_offset]->Vector_Priority,
- OPENPIC_PRIORITY_MASK,
- 9 << OPENPIC_PRIORITY_SHIFT);
+ openpic_set_irq_priority(irq, OPENPIC_PRIORITY_NMI);
}
/*
diff --git a/arch/ppc64/kernel/irq.c b/arch/ppc64/kernel/irq.c
index 3defc8c33adf..ffe300611f00 100644
--- a/arch/ppc64/kernel/irq.c
+++ b/arch/ppc64/kernel/irq.c
@@ -245,7 +245,7 @@ void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq)
spin_lock(&desc->lock);
if (!noirqdebug)
- note_interrupt(irq, desc, action_ret);
+ note_interrupt(irq, desc, action_ret, regs);
if (likely(!(desc->status & IRQ_PENDING)))
break;
desc->status &= ~IRQ_PENDING;
diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c
index 782ce3efa2c1..1d2ff6d6b0b3 100644
--- a/arch/ppc64/kernel/kprobes.c
+++ b/arch/ppc64/kernel/kprobes.c
@@ -36,6 +36,8 @@
#include
#include
+static DECLARE_MUTEX(kprobe_mutex);
+
static struct kprobe *current_kprobe;
static unsigned long kprobe_status, kprobe_saved_msr;
static struct kprobe *kprobe_prev;
@@ -54,6 +56,15 @@ int arch_prepare_kprobe(struct kprobe *p)
printk("Cannot register a kprobe on rfid or mtmsrd\n");
ret = -EINVAL;
}
+
+ /* insn must be on a special executable page on ppc64 */
+ if (!ret) {
+ up(&kprobe_mutex);
+ p->ainsn.insn = get_insn_slot();
+ down(&kprobe_mutex);
+ if (!p->ainsn.insn)
+ ret = -ENOMEM;
+ }
return ret;
}
@@ -79,16 +90,22 @@ void arch_disarm_kprobe(struct kprobe *p)
void arch_remove_kprobe(struct kprobe *p)
{
+ up(&kprobe_mutex);
+ free_insn_slot(p->ainsn.insn);
+ down(&kprobe_mutex);
}
static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
{
+ kprobe_opcode_t insn = *p->ainsn.insn;
+
regs->msr |= MSR_SE;
- /*single step inline if it a breakpoint instruction*/
- if (p->opcode == BREAKPOINT_INSTRUCTION)
+
+ /* single step inline if it is a trap variant */
+ if (IS_TW(insn) || IS_TD(insn) || IS_TWI(insn) || IS_TDI(insn))
regs->nip = (unsigned long)p->addr;
else
- regs->nip = (unsigned long)&p->ainsn.insn;
+ regs->nip = (unsigned long)p->ainsn.insn;
}
static inline void save_previous_kprobe(void)
@@ -105,6 +122,23 @@ static inline void restore_previous_kprobe(void)
kprobe_saved_msr = kprobe_saved_msr_prev;
}
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+ struct kretprobe_instance *ri;
+
+ if ((ri = get_free_rp_inst(rp)) != NULL) {
+ ri->rp = rp;
+ ri->task = current;
+ ri->ret_addr = (kprobe_opcode_t *)regs->link;
+
+ /* Replace the return addr with trampoline addr */
+ regs->link = (unsigned long)kretprobe_trampoline;
+ add_rp_inst(ri);
+ } else {
+ rp->nmissed++;
+ }
+}
+
static inline int kprobe_handler(struct pt_regs *regs)
{
struct kprobe *p;
@@ -194,6 +228,78 @@ no_kprobe:
return ret;
}
+/*
+ * Function return probe trampoline:
+ * - init_kprobes() establishes a probepoint here
+ * - When the probed function returns, this probe
+ * causes the handlers to fire
+ */
+void kretprobe_trampoline_holder(void)
+{
+ asm volatile(".global kretprobe_trampoline\n"
+ "kretprobe_trampoline:\n"
+ "nop\n");
+}
+
+/*
+ * Called when the probe at kretprobe trampoline is hit
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+ struct kretprobe_instance *ri = NULL;
+ struct hlist_head *head;
+ struct hlist_node *node, *tmp;
+ unsigned long orig_ret_address = 0;
+ unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
+
+ head = kretprobe_inst_table_head(current);
+
+ /*
+ * It is possible to have multiple instances associated with a given
+ * task either because an multiple functions in the call path
+ * have a return probe installed on them, and/or more then one return
+ * return probe was registered for a target function.
+ *
+ * We can handle this because:
+ * - instances are always inserted at the head of the list
+ * - when multiple return probes are registered for the same
+ * function, the first instance's ret_addr will point to the
+ * real return address, and all the rest will point to
+ * kretprobe_trampoline
+ */
+ hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+ if (ri->task != current)
+ /* another task is sharing our hash bucket */
+ continue;
+
+ if (ri->rp && ri->rp->handler)
+ ri->rp->handler(ri, regs);
+
+ orig_ret_address = (unsigned long)ri->ret_addr;
+ recycle_rp_inst(ri);
+
+ if (orig_ret_address != trampoline_address)
+ /*
+ * This is the real return address. Any other
+ * instances associated with this task are for
+ * other calls deeper on the call stack
+ */
+ break;
+ }
+
+ BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+ regs->nip = orig_ret_address;
+
+ unlock_kprobes();
+
+ /*
+ * By returning a non-zero value, we are telling
+ * kprobe_handler() that we have handled unlocking
+ * and re-enabling preemption.
+ */
+ return 1;
+}
+
/*
* Called after single-stepping. p->addr is the address of the
* instruction whose first byte has been replaced by the "breakpoint"
@@ -205,9 +311,10 @@ no_kprobe:
static void resume_execution(struct kprobe *p, struct pt_regs *regs)
{
int ret;
+ unsigned int insn = *p->ainsn.insn;
regs->nip = (unsigned long)p->addr;
- ret = emulate_step(regs, p->ainsn.insn[0]);
+ ret = emulate_step(regs, insn);
if (ret == 0)
regs->nip = (unsigned long)p->addr + 4;
}
@@ -331,3 +438,13 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
memcpy(regs, &jprobe_saved_regs, sizeof(struct pt_regs));
return 1;
}
+
+static struct kprobe trampoline_p = {
+ .addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+ .pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+ return register_kprobe(&trampoline_p);
+}
diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c
index 580676f87d23..ae6f579d3fa0 100644
--- a/arch/ppc64/kernel/pci.c
+++ b/arch/ppc64/kernel/pci.c
@@ -351,7 +351,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
*offset += hose->pci_mem_offset;
res_bit = IORESOURCE_MEM;
} else {
- io_offset = (unsigned long)hose->io_base_virt;
+ io_offset = (unsigned long)hose->io_base_virt - pci_io_base;
*offset += io_offset;
res_bit = IORESOURCE_IO;
}
@@ -378,7 +378,7 @@ static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
/* found it! construct the final physical address */
if (mmap_state == pci_mmap_io)
- *offset += hose->io_base_phys - io_offset;
+ *offset += hose->io_base_phys - io_offset;
return rp;
}
@@ -944,4 +944,22 @@ int pci_read_irq_line(struct pci_dev *pci_dev)
}
EXPORT_SYMBOL(pci_read_irq_line);
+void pci_resource_to_user(const struct pci_dev *dev, int bar,
+ const struct resource *rsrc,
+ u64 *start, u64 *end)
+{
+ struct pci_controller *hose = pci_bus_to_host(dev->bus);
+ unsigned long offset = 0;
+
+ if (hose == NULL)
+ return;
+
+ if (rsrc->flags & IORESOURCE_IO)
+ offset = pci_io_base - (unsigned long)hose->io_base_virt +
+ hose->io_base_phys;
+
+ *start = rsrc->start + offset;
+ *end = rsrc->end + offset;
+}
+
#endif /* CONFIG_PPC_MULTIPLATFORM */
diff --git a/arch/ppc64/kernel/ppc_ksyms.c b/arch/ppc64/kernel/ppc_ksyms.c
index b230a63fe4c8..705742f4eec6 100644
--- a/arch/ppc64/kernel/ppc_ksyms.c
+++ b/arch/ppc64/kernel/ppc_ksyms.c
@@ -75,6 +75,7 @@ EXPORT_SYMBOL(giveup_fpu);
EXPORT_SYMBOL(giveup_altivec);
#endif
EXPORT_SYMBOL(__flush_icache_range);
+EXPORT_SYMBOL(flush_dcache_range);
#ifdef CONFIG_SMP
#ifdef CONFIG_PPC_ISERIES
diff --git a/arch/ppc64/kernel/process.c b/arch/ppc64/kernel/process.c
index aba89554d89d..f7cae05e40fb 100644
--- a/arch/ppc64/kernel/process.c
+++ b/arch/ppc64/kernel/process.c
@@ -36,6 +36,7 @@
#include
#include
#include
+#include
#include
#include
@@ -307,6 +308,8 @@ void show_regs(struct pt_regs * regs)
void exit_thread(void)
{
+ kprobe_flush_task(current);
+
#ifndef CONFIG_SMP
if (last_task_used_math == current)
last_task_used_math = NULL;
@@ -321,6 +324,7 @@ void flush_thread(void)
{
struct thread_info *t = current_thread_info();
+ kprobe_flush_task(current);
if (t->flags & _TIF_ABI_PENDING)
t->flags ^= (_TIF_ABI_PENDING | _TIF_32BIT);
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index 2348a75e050d..2a532db9138a 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -91,6 +91,7 @@ unsigned long tb_to_xs;
unsigned tb_to_us;
unsigned long processor_freq;
DEFINE_SPINLOCK(rtc_lock);
+EXPORT_SYMBOL_GPL(rtc_lock);
unsigned long tb_to_ns_scale;
unsigned long tb_to_ns_shift;
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 262e13d086fe..7a117ef473c5 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -270,66 +270,10 @@ endmenu
source "drivers/Kconfig"
-config PRINTER
- tristate "Parallel printer support"
- depends on PARPORT
- ---help---
- If you intend to attach a printer to the parallel port of your Linux
- box (as opposed to using a serial printer; if the connector at the
- printer has 9 or 25 holes ["female"], then it's serial), say Y.
- Also read the Printing-HOWTO, available from
- .
-
- It is possible to share one parallel port among several devices
- (e.g. printer and ZIP drive) and it is safe to compile the
- corresponding drivers into the kernel. If you want to compile this
- driver as a module however, choose M here and read
- . The module will be called lp.
-
- If you have several parallel ports, you can specify which ports to
- use with the "lp" kernel command line option. (Try "man bootparam"
- or see the documentation of your boot loader (silo) about how to pass
- options to the kernel at boot time.) The syntax of the "lp" command
- line option can be found in .
-
- If you have more than 8 printers, you need to increase the LP_NO
- macro in lp.c and the PARPORT_MAX macro in parport.h.
-
-source "mm/Kconfig"
-
-endmenu
-
-source "drivers/base/Kconfig"
-
-source "drivers/video/Kconfig"
-
-source "drivers/mtd/Kconfig"
-
-source "drivers/serial/Kconfig"
-
if !SUN4
source "drivers/sbus/char/Kconfig"
endif
-source "drivers/block/Kconfig"
-
-# Don't frighten a common SBus user
-if PCI
-
-source "drivers/ide/Kconfig"
-
-endif
-
-source "drivers/isdn/Kconfig"
-
-source "drivers/scsi/Kconfig"
-
-source "drivers/fc4/Kconfig"
-
-source "drivers/md/Kconfig"
-
-source "net/Kconfig"
-
# This one must be before the filesystem configs. -DaveM
menu "Unix98 PTY support"
diff --git a/arch/sparc64/kernel/auxio.c b/arch/sparc64/kernel/auxio.c
index a0716ccc2f4a..8852c20c8d99 100644
--- a/arch/sparc64/kernel/auxio.c
+++ b/arch/sparc64/kernel/auxio.c
@@ -16,7 +16,7 @@
#include
#include
-/* This cannot be static, as it is referenced in entry.S */
+/* This cannot be static, as it is referenced in irq.c */
void __iomem *auxio_register = NULL;
enum auxio_type {
diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S
index a47f2d0b1a29..eee516a71c14 100644
--- a/arch/sparc64/kernel/entry.S
+++ b/arch/sparc64/kernel/entry.S
@@ -271,8 +271,9 @@ cplus_fptrap_insn_1:
fmuld %f0, %f2, %f26
faddd %f0, %f2, %f28
fmuld %f0, %f2, %f30
+ membar #Sync
b,pt %xcc, fpdis_exit
- membar #Sync
+ nop
2: andcc %g5, FPRS_DU, %g0
bne,pt %icc, 3f
fzero %f32
@@ -301,8 +302,9 @@ cplus_fptrap_insn_2:
fmuld %f32, %f34, %f58
faddd %f32, %f34, %f60
fmuld %f32, %f34, %f62
+ membar #Sync
ba,pt %xcc, fpdis_exit
- membar #Sync
+ nop
3: mov SECONDARY_CONTEXT, %g3
add %g6, TI_FPREGS, %g1
ldxa [%g3] ASI_DMMU, %g5
@@ -699,116 +701,6 @@ utrap_ill:
ba,pt %xcc, rtrap
clr %l6
-#ifdef CONFIG_BLK_DEV_FD
- .globl floppy_hardint
-floppy_hardint:
- wr %g0, (1 << 11), %clear_softint
- sethi %hi(doing_pdma), %g1
- ld [%g1 + %lo(doing_pdma)], %g2
- brz,pn %g2, floppy_dosoftint
- sethi %hi(fdc_status), %g3
- ldx [%g3 + %lo(fdc_status)], %g3
- sethi %hi(pdma_vaddr), %g5
- ldx [%g5 + %lo(pdma_vaddr)], %g4
- sethi %hi(pdma_size), %g5
- ldx [%g5 + %lo(pdma_size)], %g5
-
-next_byte:
- lduba [%g3] ASI_PHYS_BYPASS_EC_E, %g7
- andcc %g7, 0x80, %g0
- be,pn %icc, floppy_fifo_emptied
- andcc %g7, 0x20, %g0
- be,pn %icc, floppy_overrun
- andcc %g7, 0x40, %g0
- be,pn %icc, floppy_write
- sub %g5, 1, %g5
-
- inc %g3
- lduba [%g3] ASI_PHYS_BYPASS_EC_E, %g7
- dec %g3
- orcc %g0, %g5, %g0
- stb %g7, [%g4]
- bne,pn %xcc, next_byte
- add %g4, 1, %g4
-
- b,pt %xcc, floppy_tdone
- nop
-
-floppy_write:
- ldub [%g4], %g7
- orcc %g0, %g5, %g0
- inc %g3
- stba %g7, [%g3] ASI_PHYS_BYPASS_EC_E
- dec %g3
- bne,pn %xcc, next_byte
- add %g4, 1, %g4
-
-floppy_tdone:
- sethi %hi(pdma_vaddr), %g1
- stx %g4, [%g1 + %lo(pdma_vaddr)]
- sethi %hi(pdma_size), %g1
- stx %g5, [%g1 + %lo(pdma_size)]
- sethi %hi(auxio_register), %g1
- ldx [%g1 + %lo(auxio_register)], %g7
- lduba [%g7] ASI_PHYS_BYPASS_EC_E, %g5
- or %g5, AUXIO_AUX1_FTCNT, %g5
-/* andn %g5, AUXIO_AUX1_MASK, %g5 */
- stba %g5, [%g7] ASI_PHYS_BYPASS_EC_E
- andn %g5, AUXIO_AUX1_FTCNT, %g5
-/* andn %g5, AUXIO_AUX1_MASK, %g5 */
-
- nop; nop; nop; nop; nop; nop;
- nop; nop; nop; nop; nop; nop;
-
- stba %g5, [%g7] ASI_PHYS_BYPASS_EC_E
- sethi %hi(doing_pdma), %g1
- b,pt %xcc, floppy_dosoftint
- st %g0, [%g1 + %lo(doing_pdma)]
-
-floppy_fifo_emptied:
- sethi %hi(pdma_vaddr), %g1
- stx %g4, [%g1 + %lo(pdma_vaddr)]
- sethi %hi(pdma_size), %g1
- stx %g5, [%g1 + %lo(pdma_size)]
- sethi %hi(irq_action), %g1
- or %g1, %lo(irq_action), %g1
- ldx [%g1 + (11 << 3)], %g3 ! irqaction[floppy_irq]
- ldx [%g3 + 0x08], %g4 ! action->flags>>48==ino
- sethi %hi(ivector_table), %g3
- srlx %g4, 48, %g4
- or %g3, %lo(ivector_table), %g3
- sllx %g4, 5, %g4
- ldx [%g3 + %g4], %g4 ! &ivector_table[ino]
- ldx [%g4 + 0x10], %g4 ! bucket->iclr
- stwa %g0, [%g4] ASI_PHYS_BYPASS_EC_E ! ICLR_IDLE
- membar #Sync ! probably not needed...
- retry
-
-floppy_overrun:
- sethi %hi(pdma_vaddr), %g1
- stx %g4, [%g1 + %lo(pdma_vaddr)]
- sethi %hi(pdma_size), %g1
- stx %g5, [%g1 + %lo(pdma_size)]
- sethi %hi(doing_pdma), %g1
- st %g0, [%g1 + %lo(doing_pdma)]
-
-floppy_dosoftint:
- rdpr %pil, %g2
- wrpr %g0, 15, %pil
- sethi %hi(109f), %g7
- b,pt %xcc, etrap_irq
-109: or %g7, %lo(109b), %g7
-
- mov 11, %o0
- mov 0, %o1
- call sparc_floppy_irq
- add %sp, PTREGS_OFF, %o2
-
- b,pt %xcc, rtrap_irq
- nop
-
-#endif /* CONFIG_BLK_DEV_FD */
-
/* XXX Here is stuff we still need to write... -DaveM XXX */
.globl netbsd_syscall
netbsd_syscall:
diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index 4dcb8af94090..424712577307 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -37,6 +37,7 @@
#include
#include
#include
+#include
#ifdef CONFIG_SMP
static void distribute_irqs(void);
@@ -834,138 +835,66 @@ void handler_irq(int irq, struct pt_regs *regs)
}
#ifdef CONFIG_BLK_DEV_FD
-extern void floppy_interrupt(int irq, void *dev_cookie, struct pt_regs *regs);
+extern irqreturn_t floppy_interrupt(int, void *, struct pt_regs *);;
-void sparc_floppy_irq(int irq, void *dev_cookie, struct pt_regs *regs)
+/* XXX No easy way to include asm/floppy.h XXX */
+extern unsigned char *pdma_vaddr;
+extern unsigned long pdma_size;
+extern volatile int doing_pdma;
+extern unsigned long fdc_status;
+
+irqreturn_t sparc_floppy_irq(int irq, void *dev_cookie, struct pt_regs *regs)
{
- struct irqaction *action = *(irq + irq_action);
- struct ino_bucket *bucket;
- int cpu = smp_processor_id();
+ if (likely(doing_pdma)) {
+ void __iomem *stat = (void __iomem *) fdc_status;
+ unsigned char *vaddr = pdma_vaddr;
+ unsigned long size = pdma_size;
+ u8 val;
- irq_enter();
- kstat_this_cpu.irqs[irq]++;
+ while (size) {
+ val = readb(stat);
+ if (unlikely(!(val & 0x80))) {
+ pdma_vaddr = vaddr;
+ pdma_size = size;
+ return IRQ_HANDLED;
+ }
+ if (unlikely(!(val & 0x20))) {
+ pdma_vaddr = vaddr;
+ pdma_size = size;
+ doing_pdma = 0;
+ goto main_interrupt;
+ }
+ if (val & 0x40) {
+ /* read */
+ *vaddr++ = readb(stat + 1);
+ } else {
+ unsigned char data = *vaddr++;
- *(irq_work(cpu, irq)) = 0;
- bucket = get_ino_in_irqaction(action) + ivector_table;
+ /* write */
+ writeb(data, stat + 1);
+ }
+ size--;
+ }
- bucket->flags |= IBF_INPROGRESS;
+ pdma_vaddr = vaddr;
+ pdma_size = size;
- floppy_interrupt(irq, dev_cookie, regs);
- upa_writel(ICLR_IDLE, bucket->iclr);
+ /* Send Terminal Count pulse to floppy controller. */
+ val = readb(auxio_register);
+ val |= AUXIO_AUX1_FTCNT;
+ writeb(val, auxio_register);
+ val &= AUXIO_AUX1_FTCNT;
+ writeb(val, auxio_register);
- bucket->flags &= ~IBF_INPROGRESS;
+ doing_pdma = 0;
+ }
- irq_exit();
+main_interrupt:
+ return floppy_interrupt(irq, dev_cookie, regs);
}
+EXPORT_SYMBOL(sparc_floppy_irq);
#endif
-/* The following assumes that the branch lies before the place we
- * are branching to. This is the case for a trap vector...
- * You have been warned.
- */
-#define SPARC_BRANCH(dest_addr, inst_addr) \
- (0x10800000 | ((((dest_addr)-(inst_addr))>>2)&0x3fffff))
-
-#define SPARC_NOP (0x01000000)
-
-static void install_fast_irq(unsigned int cpu_irq,
- irqreturn_t (*handler)(int, void *, struct pt_regs *))
-{
- extern unsigned long sparc64_ttable_tl0;
- unsigned long ttent = (unsigned long) &sparc64_ttable_tl0;
- unsigned int *insns;
-
- ttent += 0x820;
- ttent += (cpu_irq - 1) << 5;
- insns = (unsigned int *) ttent;
- insns[0] = SPARC_BRANCH(((unsigned long) handler),
- ((unsigned long)&insns[0]));
- insns[1] = SPARC_NOP;
- __asm__ __volatile__("membar #StoreStore; flush %0" : : "r" (ttent));
-}
-
-int request_fast_irq(unsigned int irq,
- irqreturn_t (*handler)(int, void *, struct pt_regs *),
- unsigned long irqflags, const char *name, void *dev_id)
-{
- struct irqaction *action;
- struct ino_bucket *bucket = __bucket(irq);
- unsigned long flags;
-
- /* No pil0 dummy buckets allowed here. */
- if (bucket < &ivector_table[0] ||
- bucket >= &ivector_table[NUM_IVECS]) {
- unsigned int *caller;
-
- __asm__ __volatile__("mov %%i7, %0" : "=r" (caller));
- printk(KERN_CRIT "request_fast_irq: Old style IRQ registry attempt "
- "from %p, irq %08x.\n", caller, irq);
- return -EINVAL;
- }
-
- if (!handler)
- return -EINVAL;
-
- if ((bucket->pil == 0) || (bucket->pil == 14)) {
- printk("request_fast_irq: Trying to register shared IRQ 0 or 14.\n");
- return -EBUSY;
- }
-
- spin_lock_irqsave(&irq_action_lock, flags);
-
- action = *(bucket->pil + irq_action);
- if (action) {
- if (action->flags & SA_SHIRQ)
- panic("Trying to register fast irq when already shared.\n");
- if (irqflags & SA_SHIRQ)
- panic("Trying to register fast irq as shared.\n");
- printk("request_fast_irq: Trying to register yet already owned.\n");
- spin_unlock_irqrestore(&irq_action_lock, flags);
- return -EBUSY;
- }
-
- /*
- * We do not check for SA_SAMPLE_RANDOM in this path. Neither do we
- * support smp intr affinity in this path.
- */
- if (irqflags & SA_STATIC_ALLOC) {
- if (static_irq_count < MAX_STATIC_ALLOC)
- action = &static_irqaction[static_irq_count++];
- else
- printk("Request for IRQ%d (%s) SA_STATIC_ALLOC failed "
- "using kmalloc\n", bucket->pil, name);
- }
- if (action == NULL)
- action = (struct irqaction *)kmalloc(sizeof(struct irqaction),
- GFP_ATOMIC);
- if (!action) {
- spin_unlock_irqrestore(&irq_action_lock, flags);
- return -ENOMEM;
- }
- install_fast_irq(bucket->pil, handler);
-
- bucket->irq_info = action;
- bucket->flags |= IBF_ACTIVE;
-
- action->handler = handler;
- action->flags = irqflags;
- action->dev_id = NULL;
- action->name = name;
- action->next = NULL;
- put_ino_in_irqaction(action, irq);
- put_smpaff_in_irqaction(action, CPU_MASK_NONE);
-
- *(bucket->pil + irq_action) = action;
- enable_irq(irq);
-
- spin_unlock_irqrestore(&irq_action_lock, flags);
-
-#ifdef CONFIG_SMP
- distribute_irqs();
-#endif
- return 0;
-}
-
/* We really don't need these at all on the Sparc. We only have
* stubs here because they are exported to modules.
*/
diff --git a/arch/sparc64/kernel/semaphore.c b/arch/sparc64/kernel/semaphore.c
index 63496c43fe17..a809e63f03ef 100644
--- a/arch/sparc64/kernel/semaphore.c
+++ b/arch/sparc64/kernel/semaphore.c
@@ -32,8 +32,9 @@ static __inline__ int __sem_update_count(struct semaphore *sem, int incr)
" add %1, %4, %1\n"
" cas [%3], %0, %1\n"
" cmp %0, %1\n"
+" membar #StoreLoad | #StoreStore\n"
" bne,pn %%icc, 1b\n"
-" membar #StoreLoad | #StoreStore\n"
+" nop\n"
: "=&r" (old_count), "=&r" (tmp), "=m" (sem->count)
: "r" (&sem->count), "r" (incr), "m" (sem->count)
: "cc");
@@ -71,8 +72,9 @@ void up(struct semaphore *sem)
" cmp %%g1, %%g7\n"
" bne,pn %%icc, 1b\n"
" addcc %%g7, 1, %%g0\n"
+" membar #StoreLoad | #StoreStore\n"
" ble,pn %%icc, 3f\n"
-" membar #StoreLoad | #StoreStore\n"
+" nop\n"
"2:\n"
" .subsection 2\n"
"3: mov %0, %%g1\n"
@@ -128,8 +130,9 @@ void __sched down(struct semaphore *sem)
" cmp %%g1, %%g7\n"
" bne,pn %%icc, 1b\n"
" cmp %%g7, 1\n"
+" membar #StoreLoad | #StoreStore\n"
" bl,pn %%icc, 3f\n"
-" membar #StoreLoad | #StoreStore\n"
+" nop\n"
"2:\n"
" .subsection 2\n"
"3: mov %0, %%g1\n"
@@ -233,8 +236,9 @@ int __sched down_interruptible(struct semaphore *sem)
" cmp %%g1, %%g7\n"
" bne,pn %%icc, 1b\n"
" cmp %%g7, 1\n"
+" membar #StoreLoad | #StoreStore\n"
" bl,pn %%icc, 3f\n"
-" membar #StoreLoad | #StoreStore\n"
+" nop\n"
"2:\n"
" .subsection 2\n"
"3: mov %2, %%g1\n"
diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c
index e78cc53594fa..56cd96f4a5cd 100644
--- a/arch/sparc64/kernel/sparc64_ksyms.c
+++ b/arch/sparc64/kernel/sparc64_ksyms.c
@@ -227,7 +227,6 @@ EXPORT_SYMBOL(__flush_dcache_range);
EXPORT_SYMBOL(mostek_lock);
EXPORT_SYMBOL(mstk48t02_regs);
-EXPORT_SYMBOL(request_fast_irq);
#ifdef CONFIG_SUN_AUXIO
EXPORT_SYMBOL(auxio_set_led);
EXPORT_SYMBOL(auxio_set_lte);
diff --git a/arch/sparc64/kernel/trampoline.S b/arch/sparc64/kernel/trampoline.S
index 2c8f9344b4ee..3a145fc39cf2 100644
--- a/arch/sparc64/kernel/trampoline.S
+++ b/arch/sparc64/kernel/trampoline.S
@@ -98,8 +98,9 @@ startup_continue:
sethi %hi(prom_entry_lock), %g2
1: ldstub [%g2 + %lo(prom_entry_lock)], %g1
+ membar #StoreLoad | #StoreStore
brnz,pn %g1, 1b
- membar #StoreLoad | #StoreStore
+ nop
sethi %hi(p1275buf), %g2
or %g2, %lo(p1275buf), %g2
diff --git a/arch/sparc64/lib/U1memcpy.S b/arch/sparc64/lib/U1memcpy.S
index da9b520c7189..bafd2fc07acb 100644
--- a/arch/sparc64/lib/U1memcpy.S
+++ b/arch/sparc64/lib/U1memcpy.S
@@ -87,14 +87,17 @@
#define LOOP_CHUNK3(src, dest, len, branch_dest) \
MAIN_LOOP_CHUNK(src, dest, f32, f48, len, branch_dest)
+#define DO_SYNC membar #Sync;
#define STORE_SYNC(dest, fsrc) \
EX_ST(STORE_BLK(%fsrc, %dest)); \
- add %dest, 0x40, %dest;
+ add %dest, 0x40, %dest; \
+ DO_SYNC
#define STORE_JUMP(dest, fsrc, target) \
EX_ST(STORE_BLK(%fsrc, %dest)); \
add %dest, 0x40, %dest; \
- ba,pt %xcc, target;
+ ba,pt %xcc, target; \
+ nop;
#define FINISH_VISCHUNK(dest, f0, f1, left) \
subcc %left, 8, %left;\
@@ -239,17 +242,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f0, %f2, %f48
1: FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
- STORE_JUMP(o0, f48, 40f) membar #Sync
+ STORE_JUMP(o0, f48, 40f)
2: FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
- STORE_JUMP(o0, f48, 48f) membar #Sync
+ STORE_JUMP(o0, f48, 48f)
3: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
- STORE_JUMP(o0, f48, 56f) membar #Sync
+ STORE_JUMP(o0, f48, 56f)
1: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -260,17 +263,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f2, %f4, %f48
1: FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
- STORE_JUMP(o0, f48, 41f) membar #Sync
+ STORE_JUMP(o0, f48, 41f)
2: FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
- STORE_JUMP(o0, f48, 49f) membar #Sync
+ STORE_JUMP(o0, f48, 49f)
3: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
- STORE_JUMP(o0, f48, 57f) membar #Sync
+ STORE_JUMP(o0, f48, 57f)
1: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -281,17 +284,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f4, %f6, %f48
1: FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
- STORE_JUMP(o0, f48, 42f) membar #Sync
+ STORE_JUMP(o0, f48, 42f)
2: FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
- STORE_JUMP(o0, f48, 50f) membar #Sync
+ STORE_JUMP(o0, f48, 50f)
3: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
- STORE_JUMP(o0, f48, 58f) membar #Sync
+ STORE_JUMP(o0, f48, 58f)
1: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -302,17 +305,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f6, %f8, %f48
1: FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
- STORE_JUMP(o0, f48, 43f) membar #Sync
+ STORE_JUMP(o0, f48, 43f)
2: FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
- STORE_JUMP(o0, f48, 51f) membar #Sync
+ STORE_JUMP(o0, f48, 51f)
3: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
- STORE_JUMP(o0, f48, 59f) membar #Sync
+ STORE_JUMP(o0, f48, 59f)
1: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -323,17 +326,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f8, %f10, %f48
1: FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
- STORE_JUMP(o0, f48, 44f) membar #Sync
+ STORE_JUMP(o0, f48, 44f)
2: FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
- STORE_JUMP(o0, f48, 52f) membar #Sync
+ STORE_JUMP(o0, f48, 52f)
3: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
- STORE_JUMP(o0, f48, 60f) membar #Sync
+ STORE_JUMP(o0, f48, 60f)
1: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -344,17 +347,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f10, %f12, %f48
1: FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
- STORE_JUMP(o0, f48, 45f) membar #Sync
+ STORE_JUMP(o0, f48, 45f)
2: FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
- STORE_JUMP(o0, f48, 53f) membar #Sync
+ STORE_JUMP(o0, f48, 53f)
3: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
- STORE_JUMP(o0, f48, 61f) membar #Sync
+ STORE_JUMP(o0, f48, 61f)
1: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -365,17 +368,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f12, %f14, %f48
1: FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
- STORE_JUMP(o0, f48, 46f) membar #Sync
+ STORE_JUMP(o0, f48, 46f)
2: FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
- STORE_JUMP(o0, f48, 54f) membar #Sync
+ STORE_JUMP(o0, f48, 54f)
3: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
- STORE_JUMP(o0, f48, 62f) membar #Sync
+ STORE_JUMP(o0, f48, 62f)
1: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
@@ -386,17 +389,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %xcc, 1b+4
faligndata %f14, %f16, %f48
1: FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
- STORE_JUMP(o0, f48, 47f) membar #Sync
+ STORE_JUMP(o0, f48, 47f)
2: FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
- STORE_JUMP(o0, f48, 55f) membar #Sync
+ STORE_JUMP(o0, f48, 55f)
3: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
- STORE_SYNC(o0, f48) membar #Sync
+ STORE_SYNC(o0, f48)
FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
- STORE_JUMP(o0, f48, 63f) membar #Sync
+ STORE_JUMP(o0, f48, 63f)
40: FINISH_VISCHUNK(o0, f0, f2, g3)
41: FINISH_VISCHUNK(o0, f2, f4, g3)
diff --git a/arch/sparc64/lib/VISsave.S b/arch/sparc64/lib/VISsave.S
index 65e328d600a8..4e18989bd602 100644
--- a/arch/sparc64/lib/VISsave.S
+++ b/arch/sparc64/lib/VISsave.S
@@ -72,7 +72,11 @@ vis1: ldub [%g6 + TI_FPSAVED], %g3
stda %f48, [%g3 + %g1] ASI_BLK_P
5: membar #Sync
- jmpl %g7 + %g0, %g0
+ ba,pt %xcc, 80f
+ nop
+
+ .align 32
+80: jmpl %g7 + %g0, %g0
nop
6: ldub [%g3 + TI_FPSAVED], %o5
@@ -87,8 +91,11 @@ vis1: ldub [%g6 + TI_FPSAVED], %g3
stda %f32, [%g2 + %g1] ASI_BLK_P
stda %f48, [%g3 + %g1] ASI_BLK_P
membar #Sync
- jmpl %g7 + %g0, %g0
+ ba,pt %xcc, 80f
+ nop
+ .align 32
+80: jmpl %g7 + %g0, %g0
nop
.align 32
@@ -126,6 +133,10 @@ VISenterhalf:
stda %f0, [%g2 + %g1] ASI_BLK_P
stda %f16, [%g3 + %g1] ASI_BLK_P
membar #Sync
+ ba,pt %xcc, 4f
+ nop
+
+ .align 32
4: and %o5, FPRS_DU, %o5
jmpl %g7 + %g0, %g0
wr %o5, FPRS_FEF, %fprs
diff --git a/arch/sparc64/lib/atomic.S b/arch/sparc64/lib/atomic.S
index e528b8d1a3e6..faf87c31598b 100644
--- a/arch/sparc64/lib/atomic.S
+++ b/arch/sparc64/lib/atomic.S
@@ -7,18 +7,6 @@
#include
#include
- /* On SMP we need to use memory barriers to ensure
- * correct memory operation ordering, nop these out
- * for uniprocessor.
- */
-#ifdef CONFIG_SMP
-#define ATOMIC_PRE_BARRIER membar #StoreLoad | #LoadLoad
-#define ATOMIC_POST_BARRIER membar #StoreLoad | #StoreStore
-#else
-#define ATOMIC_PRE_BARRIER nop
-#define ATOMIC_POST_BARRIER nop
-#endif
-
.text
/* Two versions of the atomic routines, one that
@@ -52,6 +40,24 @@ atomic_sub: /* %o0 = decrement, %o1 = atomic_ptr */
nop
.size atomic_sub, .-atomic_sub
+ /* On SMP we need to use memory barriers to ensure
+ * correct memory operation ordering, nop these out
+ * for uniprocessor.
+ */
+#ifdef CONFIG_SMP
+
+#define ATOMIC_PRE_BARRIER membar #StoreLoad | #LoadLoad;
+#define ATOMIC_POST_BARRIER \
+ ba,pt %xcc, 80b; \
+ membar #StoreLoad | #StoreStore
+
+80: retl
+ nop
+#else
+#define ATOMIC_PRE_BARRIER
+#define ATOMIC_POST_BARRIER
+#endif
+
.globl atomic_add_ret
.type atomic_add_ret,#function
atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
@@ -62,9 +68,10 @@ atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
cmp %g1, %g7
bne,pn %icc, 1b
add %g7, %o0, %g7
+ sra %g7, 0, %o0
ATOMIC_POST_BARRIER
retl
- sra %g7, 0, %o0
+ nop
.size atomic_add_ret, .-atomic_add_ret
.globl atomic_sub_ret
@@ -77,9 +84,10 @@ atomic_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */
cmp %g1, %g7
bne,pn %icc, 1b
sub %g7, %o0, %g7
+ sra %g7, 0, %o0
ATOMIC_POST_BARRIER
retl
- sra %g7, 0, %o0
+ nop
.size atomic_sub_ret, .-atomic_sub_ret
.globl atomic64_add
@@ -118,9 +126,10 @@ atomic64_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
cmp %g1, %g7
bne,pn %xcc, 1b
add %g7, %o0, %g7
+ mov %g7, %o0
ATOMIC_POST_BARRIER
retl
- mov %g7, %o0
+ nop
.size atomic64_add_ret, .-atomic64_add_ret
.globl atomic64_sub_ret
@@ -133,7 +142,8 @@ atomic64_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */
cmp %g1, %g7
bne,pn %xcc, 1b
sub %g7, %o0, %g7
+ mov %g7, %o0
ATOMIC_POST_BARRIER
retl
- mov %g7, %o0
+ nop
.size atomic64_sub_ret, .-atomic64_sub_ret
diff --git a/arch/sparc64/lib/bitops.S b/arch/sparc64/lib/bitops.S
index 886dcd2b376a..31afbfe6c1e8 100644
--- a/arch/sparc64/lib/bitops.S
+++ b/arch/sparc64/lib/bitops.S
@@ -7,19 +7,25 @@
#include
#include
+ .text
+
/* On SMP we need to use memory barriers to ensure
* correct memory operation ordering, nop these out
* for uniprocessor.
*/
+
#ifdef CONFIG_SMP
#define BITOP_PRE_BARRIER membar #StoreLoad | #LoadLoad
-#define BITOP_POST_BARRIER membar #StoreLoad | #StoreStore
-#else
-#define BITOP_PRE_BARRIER nop
-#define BITOP_POST_BARRIER nop
-#endif
+#define BITOP_POST_BARRIER \
+ ba,pt %xcc, 80b; \
+ membar #StoreLoad | #StoreStore
- .text
+80: retl
+ nop
+#else
+#define BITOP_PRE_BARRIER
+#define BITOP_POST_BARRIER
+#endif
.globl test_and_set_bit
.type test_and_set_bit,#function
@@ -37,10 +43,11 @@ test_and_set_bit: /* %o0=nr, %o1=addr */
cmp %g7, %g1
bne,pn %xcc, 1b
and %g7, %o2, %g2
- BITOP_POST_BARRIER
clr %o0
+ movrne %g2, 1, %o0
+ BITOP_POST_BARRIER
retl
- movrne %g2, 1, %o0
+ nop
.size test_and_set_bit, .-test_and_set_bit
.globl test_and_clear_bit
@@ -59,10 +66,11 @@ test_and_clear_bit: /* %o0=nr, %o1=addr */
cmp %g7, %g1
bne,pn %xcc, 1b
and %g7, %o2, %g2
- BITOP_POST_BARRIER
clr %o0
+ movrne %g2, 1, %o0
+ BITOP_POST_BARRIER
retl
- movrne %g2, 1, %o0
+ nop
.size test_and_clear_bit, .-test_and_clear_bit
.globl test_and_change_bit
@@ -81,10 +89,11 @@ test_and_change_bit: /* %o0=nr, %o1=addr */
cmp %g7, %g1
bne,pn %xcc, 1b
and %g7, %o2, %g2
- BITOP_POST_BARRIER
clr %o0
+ movrne %g2, 1, %o0
+ BITOP_POST_BARRIER
retl
- movrne %g2, 1, %o0
+ nop
.size test_and_change_bit, .-test_and_change_bit
.globl set_bit
diff --git a/arch/sparc64/lib/debuglocks.c b/arch/sparc64/lib/debuglocks.c
index c421e0c65325..f03344cf784e 100644
--- a/arch/sparc64/lib/debuglocks.c
+++ b/arch/sparc64/lib/debuglocks.c
@@ -252,8 +252,9 @@ wlock_again:
" andn %%g1, %%g3, %%g7\n"
" casx [%0], %%g1, %%g7\n"
" cmp %%g1, %%g7\n"
+" membar #StoreLoad | #StoreStore\n"
" bne,pn %%xcc, 1b\n"
-" membar #StoreLoad | #StoreStore"
+" nop"
: /* no outputs */
: "r" (&(rw->lock))
: "g3", "g1", "g7", "cc", "memory");
@@ -351,8 +352,9 @@ int _do_write_trylock (rwlock_t *rw, char *str)
" andn %%g1, %%g3, %%g7\n"
" casx [%0], %%g1, %%g7\n"
" cmp %%g1, %%g7\n"
+" membar #StoreLoad | #StoreStore\n"
" bne,pn %%xcc, 1b\n"
-" membar #StoreLoad | #StoreStore"
+" nop"
: /* no outputs */
: "r" (&(rw->lock))
: "g3", "g1", "g7", "cc", "memory");
diff --git a/arch/sparc64/lib/dec_and_lock.S b/arch/sparc64/lib/dec_and_lock.S
index 7e6fdaebedba..8ee288dd0afc 100644
--- a/arch/sparc64/lib/dec_and_lock.S
+++ b/arch/sparc64/lib/dec_and_lock.S
@@ -48,8 +48,9 @@ start_to_zero:
#endif
to_zero:
ldstub [%o1], %g3
+ membar #StoreLoad | #StoreStore
brnz,pn %g3, spin_on_lock
- membar #StoreLoad | #StoreStore
+ nop
loop2: cas [%o0], %g2, %g7 /* ASSERT(g7 == 0) */
cmp %g2, %g7
@@ -71,8 +72,9 @@ loop2: cas [%o0], %g2, %g7 /* ASSERT(g7 == 0) */
nop
spin_on_lock:
ldub [%o1], %g3
+ membar #LoadLoad
brnz,pt %g3, spin_on_lock
- membar #LoadLoad
+ nop
ba,pt %xcc, to_zero
nop
nop
diff --git a/arch/sparc64/lib/rwsem.S b/arch/sparc64/lib/rwsem.S
index 174ff7b9164c..75f0e6b951d6 100644
--- a/arch/sparc64/lib/rwsem.S
+++ b/arch/sparc64/lib/rwsem.S
@@ -17,8 +17,9 @@ __down_read:
bne,pn %icc, 1b
add %g7, 1, %g7
cmp %g7, 0
+ membar #StoreLoad | #StoreStore
bl,pn %icc, 3f
- membar #StoreLoad | #StoreStore
+ nop
2:
retl
nop
@@ -57,8 +58,9 @@ __down_write:
cmp %g3, %g7
bne,pn %icc, 1b
cmp %g7, 0
+ membar #StoreLoad | #StoreStore
bne,pn %icc, 3f
- membar #StoreLoad | #StoreStore
+ nop
2: retl
nop
3:
@@ -97,8 +99,9 @@ __up_read:
cmp %g1, %g7
bne,pn %icc, 1b
cmp %g7, 0
+ membar #StoreLoad | #StoreStore
bl,pn %icc, 3f
- membar #StoreLoad | #StoreStore
+ nop
2: retl
nop
3: sethi %hi(RWSEM_ACTIVE_MASK), %g1
@@ -126,8 +129,9 @@ __up_write:
bne,pn %icc, 1b
sub %g7, %g1, %g7
cmp %g7, 0
+ membar #StoreLoad | #StoreStore
bl,pn %icc, 3f
- membar #StoreLoad | #StoreStore
+ nop
2:
retl
nop
@@ -151,8 +155,9 @@ __downgrade_write:
bne,pn %icc, 1b
sub %g7, %g1, %g7
cmp %g7, 0
+ membar #StoreLoad | #StoreStore
bl,pn %icc, 3f
- membar #StoreLoad | #StoreStore
+ nop
2:
retl
nop
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index 9c5222075da9..8fc413cb6acd 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -136,8 +136,9 @@ static __inline__ void set_dcache_dirty(struct page *page, int this_cpu)
"or %%g1, %0, %%g1\n\t"
"casx [%2], %%g7, %%g1\n\t"
"cmp %%g7, %%g1\n\t"
+ "membar #StoreLoad | #StoreStore\n\t"
"bne,pn %%xcc, 1b\n\t"
- " membar #StoreLoad | #StoreStore"
+ " nop"
: /* no outputs */
: "r" (mask), "r" (non_cpu_bits), "r" (&page->flags)
: "g1", "g7");
@@ -157,8 +158,9 @@ static __inline__ void clear_dcache_dirty_cpu(struct page *page, unsigned long c
" andn %%g7, %1, %%g1\n\t"
"casx [%2], %%g7, %%g1\n\t"
"cmp %%g7, %%g1\n\t"
+ "membar #StoreLoad | #StoreStore\n\t"
"bne,pn %%xcc, 1b\n\t"
- " membar #StoreLoad | #StoreStore\n"
+ " nop\n"
"2:"
: /* no outputs */
: "r" (cpu), "r" (mask), "r" (&page->flags),
diff --git a/arch/sparc64/mm/ultra.S b/arch/sparc64/mm/ultra.S
index 7a0934321010..7a2431d3abc7 100644
--- a/arch/sparc64/mm/ultra.S
+++ b/arch/sparc64/mm/ultra.S
@@ -266,8 +266,9 @@ __cheetah_flush_tlb_pending: /* 22 insns */
andn %o3, 1, %o3
stxa %g0, [%o3] ASI_IMMU_DEMAP
2: stxa %g0, [%o3] ASI_DMMU_DEMAP
+ membar #Sync
brnz,pt %o1, 1b
- membar #Sync
+ nop
stxa %g2, [%o4] ASI_DMMU
flush %g6
wrpr %g0, 0, %tl
diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 4e680f87a75f..acd2a778ebe6 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -38,7 +38,7 @@
#include
#include
#include
-#include
+
#include
#include
#include
@@ -51,8 +51,6 @@ static struct kprobe *kprobe_prev;
static unsigned long kprobe_status_prev, kprobe_old_rflags_prev, kprobe_saved_rflags_prev;
static struct pt_regs jprobe_saved_regs;
static long *jprobe_saved_rsp;
-static kprobe_opcode_t *get_insn_slot(void);
-static void free_insn_slot(kprobe_opcode_t *slot);
void jprobe_return_end(void);
/* copy of the kernel stack at the probe fire time */
@@ -274,48 +272,23 @@ static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
regs->rip = (unsigned long)p->ainsn.insn;
}
-struct task_struct *arch_get_kprobe_task(void *ptr)
-{
- return ((struct thread_info *) (((unsigned long) ptr) &
- (~(THREAD_SIZE -1))))->task;
-}
-
void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
{
unsigned long *sara = (unsigned long *)regs->rsp;
- struct kretprobe_instance *ri;
- static void *orig_ret_addr;
+ struct kretprobe_instance *ri;
- /*
- * Save the return address when the return probe hits
- * the first time, and use it to populate the (krprobe
- * instance)->ret_addr for subsequent return probes at
- * the same addrress since stack address would have
- * the kretprobe_trampoline by then.
- */
- if (((void*) *sara) != kretprobe_trampoline)
- orig_ret_addr = (void*) *sara;
+ if ((ri = get_free_rp_inst(rp)) != NULL) {
+ ri->rp = rp;
+ ri->task = current;
+ ri->ret_addr = (kprobe_opcode_t *) *sara;
- if ((ri = get_free_rp_inst(rp)) != NULL) {
- ri->rp = rp;
- ri->stack_addr = sara;
- ri->ret_addr = orig_ret_addr;
- add_rp_inst(ri);
/* Replace the return addr with trampoline addr */
*sara = (unsigned long) &kretprobe_trampoline;
- } else {
- rp->nmissed++;
- }
-}
-void arch_kprobe_flush_task(struct task_struct *tk)
-{
- struct kretprobe_instance *ri;
- while ((ri = get_rp_inst_tsk(tk)) != NULL) {
- *((unsigned long *)(ri->stack_addr)) =
- (unsigned long) ri->ret_addr;
- recycle_rp_inst(ri);
- }
+ add_rp_inst(ri);
+ } else {
+ rp->nmissed++;
+ }
}
/*
@@ -428,36 +401,59 @@ no_kprobe:
*/
int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
{
- struct task_struct *tsk;
- struct kretprobe_instance *ri;
- struct hlist_head *head;
- struct hlist_node *node;
- unsigned long *sara = (unsigned long *)regs->rsp - 1;
+ struct kretprobe_instance *ri = NULL;
+ struct hlist_head *head;
+ struct hlist_node *node, *tmp;
+ unsigned long orig_ret_address = 0;
+ unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
- tsk = arch_get_kprobe_task(sara);
- head = kretprobe_inst_table_head(tsk);
+ head = kretprobe_inst_table_head(current);
- hlist_for_each_entry(ri, node, head, hlist) {
- if (ri->stack_addr == sara && ri->rp) {
- if (ri->rp->handler)
- ri->rp->handler(ri, regs);
- }
- }
- return 0;
-}
+ /*
+ * It is possible to have multiple instances associated with a given
+ * task either because an multiple functions in the call path
+ * have a return probe installed on them, and/or more then one return
+ * return probe was registered for a target function.
+ *
+ * We can handle this because:
+ * - instances are always inserted at the head of the list
+ * - when multiple return probes are registered for the same
+ * function, the first instance's ret_addr will point to the
+ * real return address, and all the rest will point to
+ * kretprobe_trampoline
+ */
+ hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+ if (ri->task != current)
+ /* another task is sharing our hash bucket */
+ continue;
-void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
- unsigned long flags)
-{
- struct kretprobe_instance *ri;
- /* RA already popped */
- unsigned long *sara = ((unsigned long *)regs->rsp) - 1;
+ if (ri->rp && ri->rp->handler)
+ ri->rp->handler(ri, regs);
- while ((ri = get_rp_inst(sara))) {
- regs->rip = (unsigned long)ri->ret_addr;
+ orig_ret_address = (unsigned long)ri->ret_addr;
recycle_rp_inst(ri);
+
+ if (orig_ret_address != trampoline_address)
+ /*
+ * This is the real return address. Any other
+ * instances associated with this task are for
+ * other calls deeper on the call stack
+ */
+ break;
}
- regs->eflags &= ~TF_MASK;
+
+ BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+ regs->rip = orig_ret_address;
+
+ unlock_kprobes();
+ preempt_enable_no_resched();
+
+ /*
+ * By returning a non-zero value, we are telling
+ * kprobe_handler() that we have handled unlocking
+ * and re-enabling preemption.
+ */
+ return 1;
}
/*
@@ -550,8 +546,7 @@ int post_kprobe_handler(struct pt_regs *regs)
current_kprobe->post_handler(current_kprobe, regs, 0);
}
- if (current_kprobe->post_handler != trampoline_post_handler)
- resume_execution(current_kprobe, regs);
+ resume_execution(current_kprobe, regs);
regs->eflags |= kprobe_saved_rflags;
/* Restore the original saved kprobes variables and continue. */
@@ -682,111 +677,12 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
return 0;
}
-/*
- * kprobe->ainsn.insn points to the copy of the instruction to be single-stepped.
- * By default on x86_64, pages we get from kmalloc or vmalloc are not
- * executable. Single-stepping an instruction on such a page yields an
- * oops. So instead of storing the instruction copies in their respective
- * kprobe objects, we allocate a page, map it executable, and store all the
- * instruction copies there. (We can allocate additional pages if somebody
- * inserts a huge number of probes.) Each page can hold up to INSNS_PER_PAGE
- * instruction slots, each of which is MAX_INSN_SIZE*sizeof(kprobe_opcode_t)
- * bytes.
- */
-#define INSNS_PER_PAGE (PAGE_SIZE/(MAX_INSN_SIZE*sizeof(kprobe_opcode_t)))
-struct kprobe_insn_page {
- struct hlist_node hlist;
- kprobe_opcode_t *insns; /* page of instruction slots */
- char slot_used[INSNS_PER_PAGE];
- int nused;
+static struct kprobe trampoline_p = {
+ .addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+ .pre_handler = trampoline_probe_handler
};
-static struct hlist_head kprobe_insn_pages;
-
-/**
- * get_insn_slot() - Find a slot on an executable page for an instruction.
- * We allocate an executable page if there's no room on existing ones.
- */
-static kprobe_opcode_t *get_insn_slot(void)
+int __init arch_init(void)
{
- struct kprobe_insn_page *kip;
- struct hlist_node *pos;
-
- hlist_for_each(pos, &kprobe_insn_pages) {
- kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
- if (kip->nused < INSNS_PER_PAGE) {
- int i;
- for (i = 0; i < INSNS_PER_PAGE; i++) {
- if (!kip->slot_used[i]) {
- kip->slot_used[i] = 1;
- kip->nused++;
- return kip->insns + (i*MAX_INSN_SIZE);
- }
- }
- /* Surprise! No unused slots. Fix kip->nused. */
- kip->nused = INSNS_PER_PAGE;
- }
- }
-
- /* All out of space. Need to allocate a new page. Use slot 0.*/
- kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL);
- if (!kip) {
- return NULL;
- }
-
- /*
- * For the %rip-relative displacement fixups to be doable, we
- * need our instruction copy to be within +/- 2GB of any data it
- * might access via %rip. That is, within 2GB of where the
- * kernel image and loaded module images reside. So we allocate
- * a page in the module loading area.
- */
- kip->insns = module_alloc(PAGE_SIZE);
- if (!kip->insns) {
- kfree(kip);
- return NULL;
- }
- INIT_HLIST_NODE(&kip->hlist);
- hlist_add_head(&kip->hlist, &kprobe_insn_pages);
- memset(kip->slot_used, 0, INSNS_PER_PAGE);
- kip->slot_used[0] = 1;
- kip->nused = 1;
- return kip->insns;
-}
-
-/**
- * free_insn_slot() - Free instruction slot obtained from get_insn_slot().
- */
-static void free_insn_slot(kprobe_opcode_t *slot)
-{
- struct kprobe_insn_page *kip;
- struct hlist_node *pos;
-
- hlist_for_each(pos, &kprobe_insn_pages) {
- kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
- if (kip->insns <= slot
- && slot < kip->insns+(INSNS_PER_PAGE*MAX_INSN_SIZE)) {
- int i = (slot - kip->insns) / MAX_INSN_SIZE;
- kip->slot_used[i] = 0;
- kip->nused--;
- if (kip->nused == 0) {
- /*
- * Page is no longer in use. Free it unless
- * it's the last one. We keep the last one
- * so as not to have to set it up again the
- * next time somebody inserts a probe.
- */
- hlist_del(&kip->hlist);
- if (hlist_empty(&kprobe_insn_pages)) {
- INIT_HLIST_NODE(&kip->hlist);
- hlist_add_head(&kip->hlist,
- &kprobe_insn_pages);
- } else {
- module_free(NULL, kip->insns);
- kfree(kip);
- }
- }
- return;
- }
- }
+ return register_kprobe(&trampoline_p);
}
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 1d91271796e5..7577f9d7a75d 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -481,6 +481,33 @@ out:
return err;
}
+/*
+ * This function selects if the context switch from prev to next
+ * has to tweak the TSC disable bit in the cr4.
+ */
+static inline void disable_tsc(struct task_struct *prev_p,
+ struct task_struct *next_p)
+{
+ struct thread_info *prev, *next;
+
+ /*
+ * gcc should eliminate the ->thread_info dereference if
+ * has_secure_computing returns 0 at compile time (SECCOMP=n).
+ */
+ prev = prev_p->thread_info;
+ next = next_p->thread_info;
+
+ if (has_secure_computing(prev) || has_secure_computing(next)) {
+ /* slow path here */
+ if (has_secure_computing(prev) &&
+ !has_secure_computing(next)) {
+ write_cr4(read_cr4() & ~X86_CR4_TSD);
+ } else if (!has_secure_computing(prev) &&
+ has_secure_computing(next))
+ write_cr4(read_cr4() | X86_CR4_TSD);
+ }
+}
+
/*
* This special macro can be used to load a debugging register
*/
@@ -599,6 +626,8 @@ struct task_struct *__switch_to(struct task_struct *prev_p, struct task_struct *
}
}
+ disable_tsc(prev_p, next_p);
+
return prev_p;
}
diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index b693c232fd07..657e88aa0902 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -7,25 +7,50 @@
#include
#include
+#include
#include "pci.h"
#define MMCONFIG_APER_SIZE (256*1024*1024)
-/* The physical address of the MMCONFIG aperture. Set from ACPI tables. */
-u32 pci_mmcfg_base_addr;
-
/* Static virtual mapping of the MMCONFIG aperture */
-char *pci_mmcfg_virt;
+struct mmcfg_virt {
+ struct acpi_table_mcfg_config *cfg;
+ char *virt;
+};
+static struct mmcfg_virt *pci_mmcfg_virt;
-static inline char *pci_dev_base(unsigned int bus, unsigned int devfn)
+static char *get_virt(unsigned int seg, int bus)
{
- return pci_mmcfg_virt + ((bus << 20) | (devfn << 12));
+ int cfg_num = -1;
+ struct acpi_table_mcfg_config *cfg;
+
+ while (1) {
+ ++cfg_num;
+ if (cfg_num >= pci_mmcfg_config_num) {
+ /* something bad is going on, no cfg table is found. */
+ /* so we fall back to the old way we used to do this */
+ /* and just rely on the first entry to be correct. */
+ return pci_mmcfg_virt[0].virt;
+ }
+ cfg = pci_mmcfg_virt[cfg_num].cfg;
+ if (cfg->pci_segment_group_number != seg)
+ continue;
+ if ((cfg->start_bus_number <= bus) &&
+ (cfg->end_bus_number >= bus))
+ return pci_mmcfg_virt[cfg_num].virt;
+ }
+}
+
+static inline char *pci_dev_base(unsigned int seg, unsigned int bus, unsigned int devfn)
+{
+
+ return get_virt(seg, bus) + ((bus << 20) | (devfn << 12));
}
static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
- char *addr = pci_dev_base(bus, devfn);
+ char *addr = pci_dev_base(seg, bus, devfn);
if (unlikely(!value || (bus > 255) || (devfn > 255) || (reg > 4095)))
return -EINVAL;
@@ -48,7 +73,7 @@ static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
static int pci_mmcfg_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
- char *addr = pci_dev_base(bus,devfn);
+ char *addr = pci_dev_base(seg, bus, devfn);
if (unlikely((bus > 255) || (devfn > 255) || (reg > 4095)))
return -EINVAL;
@@ -75,9 +100,15 @@ static struct pci_raw_ops pci_mmcfg = {
static int __init pci_mmcfg_init(void)
{
+ int i;
+
if ((pci_probe & PCI_PROBE_MMCONF) == 0)
return 0;
- if (!pci_mmcfg_base_addr)
+
+ acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg);
+ if ((pci_mmcfg_config_num == 0) ||
+ (pci_mmcfg_config == NULL) ||
+ (pci_mmcfg_config[0].base_address == 0))
return 0;
/* Kludge for now. Don't use mmconfig on AMD systems because
@@ -88,13 +119,22 @@ static int __init pci_mmcfg_init(void)
return 0;
/* RED-PEN i386 doesn't do _nocache right now */
- pci_mmcfg_virt = ioremap_nocache(pci_mmcfg_base_addr, MMCONFIG_APER_SIZE);
- if (!pci_mmcfg_virt) {
- printk("PCI: Cannot map mmconfig aperture\n");
+ pci_mmcfg_virt = kmalloc(sizeof(*pci_mmcfg_virt) * pci_mmcfg_config_num, GFP_KERNEL);
+ if (pci_mmcfg_virt == NULL) {
+ printk("PCI: Can not allocate memory for mmconfig structures\n");
return 0;
- }
+ }
+ for (i = 0; i < pci_mmcfg_config_num; ++i) {
+ pci_mmcfg_virt[i].cfg = &pci_mmcfg_config[i];
+ pci_mmcfg_virt[i].virt = ioremap_nocache(pci_mmcfg_config[i].base_address, MMCONFIG_APER_SIZE);
+ if (!pci_mmcfg_virt[i].virt) {
+ printk("PCI: Cannot map mmconfig aperture for segment %d\n",
+ pci_mmcfg_config[i].pci_segment_group_number);
+ return 0;
+ }
+ printk(KERN_INFO "PCI: Using MMCONFIG at %x\n", pci_mmcfg_config[i].base_address);
+ }
- printk(KERN_INFO "PCI: Using MMCONFIG at %x\n", pci_mmcfg_base_addr);
raw_pci_ops = &pci_mmcfg;
pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
diff --git a/drivers/acpi/container.c b/drivers/acpi/container.c
index 5a0adbf8bc04..97013ddfa202 100644
--- a/drivers/acpi/container.c
+++ b/drivers/acpi/container.c
@@ -153,7 +153,7 @@ container_device_add(struct acpi_device **device, acpi_handle handle)
return_VALUE(-ENODEV);
}
- result = acpi_bus_scan(*device);
+ result = acpi_bus_start(*device);
return_VALUE(result);
}
diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index 5d19b39e9e2b..5148f3c10b5c 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -61,15 +61,14 @@ acpi_pci_data_handler (
/**
- * acpi_os_get_pci_id
+ * acpi_get_pci_id
* ------------------
* This function is used by the ACPI Interpreter (a.k.a. Core Subsystem)
* to resolve PCI information for ACPI-PCI devices defined in the namespace.
* This typically occurs when resolving PCI operation region information.
*/
-#ifdef ACPI_FUTURE_USAGE
acpi_status
-acpi_os_get_pci_id (
+acpi_get_pci_id (
acpi_handle handle,
struct acpi_pci_id *id)
{
@@ -78,7 +77,7 @@ acpi_os_get_pci_id (
struct acpi_device *device = NULL;
struct acpi_pci_data *data = NULL;
- ACPI_FUNCTION_TRACE("acpi_os_get_pci_id");
+ ACPI_FUNCTION_TRACE("acpi_get_pci_id");
if (!id)
return_ACPI_STATUS(AE_BAD_PARAMETER);
@@ -92,7 +91,7 @@ acpi_os_get_pci_id (
}
status = acpi_get_data(handle, acpi_pci_data_handler, (void**) &data);
- if (ACPI_FAILURE(status) || !data || !data->dev) {
+ if (ACPI_FAILURE(status) || !data) {
ACPI_DEBUG_PRINT((ACPI_DB_ERROR,
"Invalid ACPI-PCI context for device %s\n",
acpi_device_bid(device)));
@@ -115,7 +114,7 @@ acpi_os_get_pci_id (
return_ACPI_STATUS(AE_OK);
}
-#endif /* ACPI_FUTURE_USAGE */
+EXPORT_SYMBOL(acpi_get_pci_id);
int
@@ -129,6 +128,8 @@ acpi_pci_bind (
char *pathname = NULL;
struct acpi_buffer buffer = {0, NULL};
acpi_handle handle = NULL;
+ struct pci_dev *dev;
+ struct pci_bus *bus;
ACPI_FUNCTION_TRACE("acpi_pci_bind");
@@ -193,8 +194,20 @@ acpi_pci_bind (
* Locate matching device in PCI namespace. If it doesn't exist
* this typically means that the device isn't currently inserted
* (e.g. docking station, port replicator, etc.).
+ * We cannot simply search the global pci device list, since
+ * PCI devices are added to the global pci list when the root
+ * bridge start ops are run, which may not have happened yet.
*/
- data->dev = pci_find_slot(data->id.bus, PCI_DEVFN(data->id.device, data->id.function));
+ bus = pci_find_bus(data->id.segment, data->id.bus);
+ if (bus) {
+ list_for_each_entry(dev, &bus->devices, bus_list) {
+ if (dev->devfn == PCI_DEVFN(data->id.device,
+ data->id.function)) {
+ data->dev = dev;
+ break;
+ }
+ }
+ }
if (!data->dev) {
ACPI_DEBUG_PRINT((ACPI_DB_INFO,
"Device %02x:%02x:%02x.%02x not present in PCI namespace\n",
diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 7e6b8e3b2ed4..5d2f77fcd50c 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -46,6 +46,7 @@ ACPI_MODULE_NAME ("pci_root")
static int acpi_pci_root_add (struct acpi_device *device);
static int acpi_pci_root_remove (struct acpi_device *device, int type);
+static int acpi_pci_root_start (struct acpi_device *device);
static struct acpi_driver acpi_pci_root_driver = {
.name = ACPI_PCI_ROOT_DRIVER_NAME,
@@ -54,6 +55,7 @@ static struct acpi_driver acpi_pci_root_driver = {
.ops = {
.add = acpi_pci_root_add,
.remove = acpi_pci_root_remove,
+ .start = acpi_pci_root_start,
},
};
@@ -169,6 +171,7 @@ acpi_pci_root_add (
if (!root)
return_VALUE(-ENOMEM);
memset(root, 0, sizeof(struct acpi_pci_root));
+ INIT_LIST_HEAD(&root->node);
root->handle = device->handle;
strcpy(acpi_device_name(device), ACPI_PCI_ROOT_DEVICE_NAME);
@@ -298,12 +301,31 @@ acpi_pci_root_add (
root->id.bus);
end:
- if (result)
+ if (result) {
+ if (!list_empty(&root->node))
+ list_del(&root->node);
kfree(root);
+ }
return_VALUE(result);
}
+static int
+acpi_pci_root_start (
+ struct acpi_device *device)
+{
+ struct acpi_pci_root *root;
+
+ ACPI_FUNCTION_TRACE("acpi_pci_root_start");
+
+ list_for_each_entry(root, &acpi_pci_roots, node) {
+ if (root->handle == device->handle) {
+ pci_bus_add_devices(root->bus);
+ return_VALUE(0);
+ }
+ }
+ return_VALUE(-ENODEV);
+}
static int
acpi_pci_root_remove (
diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index f4778747e889..76156ac91bd3 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -723,7 +723,7 @@ int acpi_processor_device_add(
return_VALUE(-ENODEV);
}
- acpi_bus_scan(*device);
+ acpi_bus_start(*device);
pr = acpi_driver_data(*device);
if (!pr)
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index e85885593280..337d49b5564b 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -553,20 +553,29 @@ acpi_bus_driver_init (
* upon possible configuration and currently allocated resources.
*/
+ ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Driver successfully bound to device\n"));
+ return_VALUE(0);
+}
+
+int
+acpi_start_single_object (
+ struct acpi_device *device)
+{
+ int result = 0;
+ struct acpi_driver *driver;
+
+ ACPI_FUNCTION_TRACE("acpi_start_single_object");
+
+ if (!(driver = device->driver))
+ return_VALUE(0);
+
if (driver->ops.start) {
result = driver->ops.start(device);
if (result && driver->ops.remove)
driver->ops.remove(device, ACPI_BUS_REMOVAL_NORMAL);
- return_VALUE(result);
}
- ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Driver successfully bound to device\n"));
-
- if (driver->ops.scan) {
- driver->ops.scan(device);
- }
-
- return_VALUE(0);
+ return_VALUE(result);
}
static int acpi_driver_attach(struct acpi_driver * drv)
@@ -586,6 +595,7 @@ static int acpi_driver_attach(struct acpi_driver * drv)
if (!acpi_bus_match(dev, drv)) {
if (!acpi_bus_driver_init(dev, drv)) {
+ acpi_start_single_object(dev);
atomic_inc(&drv->references);
count++;
ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Found driver [%s] for device [%s]\n",
@@ -1009,8 +1019,8 @@ acpi_bus_remove (
}
-int
-acpi_bus_add (
+static int
+acpi_add_single_object (
struct acpi_device **child,
struct acpi_device *parent,
acpi_handle handle,
@@ -1019,7 +1029,7 @@ acpi_bus_add (
int result = 0;
struct acpi_device *device = NULL;
- ACPI_FUNCTION_TRACE("acpi_bus_add");
+ ACPI_FUNCTION_TRACE("acpi_add_single_object");
if (!child)
return_VALUE(-EINVAL);
@@ -1140,7 +1150,7 @@ acpi_bus_add (
*
* TBD: Assumes LDM provides driver hot-plug capability.
*/
- acpi_bus_find_driver(device);
+ result = acpi_bus_find_driver(device);
end:
if (!result)
@@ -1153,10 +1163,10 @@ end:
return_VALUE(result);
}
-EXPORT_SYMBOL(acpi_bus_add);
-int acpi_bus_scan (struct acpi_device *start)
+static int acpi_bus_scan (struct acpi_device *start,
+ struct acpi_bus_ops *ops)
{
acpi_status status = AE_OK;
struct acpi_device *parent = NULL;
@@ -1229,9 +1239,20 @@ int acpi_bus_scan (struct acpi_device *start)
continue;
}
- status = acpi_bus_add(&child, parent, chandle, type);
- if (ACPI_FAILURE(status))
- continue;
+ if (ops->acpi_op_add)
+ status = acpi_add_single_object(&child, parent,
+ chandle, type);
+ else
+ status = acpi_bus_get_device(chandle, &child);
+
+ if (ACPI_FAILURE(status))
+ continue;
+
+ if (ops->acpi_op_start) {
+ status = acpi_start_single_object(child);
+ if (ACPI_FAILURE(status))
+ continue;
+ }
/*
* If the device is present, enabled, and functioning then
@@ -1257,8 +1278,50 @@ int acpi_bus_scan (struct acpi_device *start)
return_VALUE(0);
}
-EXPORT_SYMBOL(acpi_bus_scan);
+int
+acpi_bus_add (
+ struct acpi_device **child,
+ struct acpi_device *parent,
+ acpi_handle handle,
+ int type)
+{
+ int result;
+ struct acpi_bus_ops ops;
+
+ ACPI_FUNCTION_TRACE("acpi_bus_add");
+
+ result = acpi_add_single_object(child, parent, handle, type);
+ if (!result) {
+ memset(&ops, 0, sizeof(ops));
+ ops.acpi_op_add = 1;
+ result = acpi_bus_scan(*child, &ops);
+ }
+ return_VALUE(result);
+}
+EXPORT_SYMBOL(acpi_bus_add);
+
+int
+acpi_bus_start (
+ struct acpi_device *device)
+{
+ int result;
+ struct acpi_bus_ops ops;
+
+ ACPI_FUNCTION_TRACE("acpi_bus_start");
+
+ if (!device)
+ return_VALUE(-EINVAL);
+
+ result = acpi_start_single_object(device);
+ if (!result) {
+ memset(&ops, 0, sizeof(ops));
+ ops.acpi_op_start = 1;
+ result = acpi_bus_scan(device, &ops);
+ }
+ return_VALUE(result);
+}
+EXPORT_SYMBOL(acpi_bus_start);
static int
acpi_bus_trim(struct acpi_device *start,
@@ -1331,13 +1394,19 @@ acpi_bus_scan_fixed (
/*
* Enumerate all fixed-feature devices.
*/
- if (acpi_fadt.pwr_button == 0)
- result = acpi_bus_add(&device, acpi_root,
+ if (acpi_fadt.pwr_button == 0) {
+ result = acpi_add_single_object(&device, acpi_root,
NULL, ACPI_BUS_TYPE_POWER_BUTTON);
+ if (!result)
+ result = acpi_start_single_object(device);
+ }
- if (acpi_fadt.sleep_button == 0)
- result = acpi_bus_add(&device, acpi_root,
+ if (acpi_fadt.sleep_button == 0) {
+ result = acpi_add_single_object(&device, acpi_root,
NULL, ACPI_BUS_TYPE_SLEEP_BUTTON);
+ if (!result)
+ result = acpi_start_single_object(device);
+ }
return_VALUE(result);
}
@@ -1346,6 +1415,7 @@ acpi_bus_scan_fixed (
static int __init acpi_scan_init(void)
{
int result;
+ struct acpi_bus_ops ops;
ACPI_FUNCTION_TRACE("acpi_scan_init");
@@ -1357,17 +1427,23 @@ static int __init acpi_scan_init(void)
/*
* Create the root device in the bus's device tree
*/
- result = acpi_bus_add(&acpi_root, NULL, ACPI_ROOT_OBJECT,
+ result = acpi_add_single_object(&acpi_root, NULL, ACPI_ROOT_OBJECT,
ACPI_BUS_TYPE_SYSTEM);
if (result)
goto Done;
+ result = acpi_start_single_object(acpi_root);
+
/*
* Enumerate devices in the ACPI namespace.
*/
result = acpi_bus_scan_fixed(acpi_root);
- if (!result)
- result = acpi_bus_scan(acpi_root);
+ if (!result) {
+ memset(&ops, 0, sizeof(ops));
+ ops.acpi_op_add = 1;
+ ops.acpi_op_start = 1;
+ result = acpi_bus_scan(acpi_root, &ops);
+ }
if (result)
acpi_device_unregister(acpi_root, ACPI_BUS_REMOVAL_NORMAL);
diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c
index 97fe13f7f07c..652281402c92 100644
--- a/drivers/base/firmware_class.c
+++ b/drivers/base/firmware_class.c
@@ -74,6 +74,8 @@ static ssize_t
firmware_timeout_store(struct class *class, const char *buf, size_t count)
{
loading_timeout = simple_strtol(buf, NULL, 10);
+ if (loading_timeout < 0)
+ loading_timeout = 0;
return count;
}
@@ -138,6 +140,10 @@ firmware_loading_store(struct class_device *class_dev,
switch (loading) {
case 1:
down(&fw_lock);
+ if (!fw_priv->fw) {
+ up(&fw_lock);
+ break;
+ }
vfree(fw_priv->fw->data);
fw_priv->fw->data = NULL;
fw_priv->fw->size = 0;
@@ -178,7 +184,7 @@ firmware_data_read(struct kobject *kobj,
down(&fw_lock);
fw = fw_priv->fw;
- if (test_bit(FW_STATUS_DONE, &fw_priv->status)) {
+ if (!fw || test_bit(FW_STATUS_DONE, &fw_priv->status)) {
ret_count = -ENODEV;
goto out;
}
@@ -238,9 +244,10 @@ firmware_data_write(struct kobject *kobj,
if (!capable(CAP_SYS_RAWIO))
return -EPERM;
+
down(&fw_lock);
fw = fw_priv->fw;
- if (test_bit(FW_STATUS_DONE, &fw_priv->status)) {
+ if (!fw || test_bit(FW_STATUS_DONE, &fw_priv->status)) {
retval = -ENODEV;
goto out;
}
@@ -418,7 +425,7 @@ request_firmware(const struct firmware **firmware_p, const char *name,
fw_priv = class_get_devdata(class_dev);
- if (loading_timeout) {
+ if (loading_timeout > 0) {
fw_priv->timeout.expires = jiffies + loading_timeout * HZ;
add_timer(&fw_priv->timeout);
}
diff --git a/drivers/block/as-iosched.c b/drivers/block/as-iosched.c
index 3410b4d294b9..91aeb678135d 100644
--- a/drivers/block/as-iosched.c
+++ b/drivers/block/as-iosched.c
@@ -1806,7 +1806,8 @@ static void as_put_request(request_queue_t *q, struct request *rq)
rq->elevator_private = NULL;
}
-static int as_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+static int as_set_request(request_queue_t *q, struct request *rq,
+ struct bio *bio, int gfp_mask)
{
struct as_data *ad = q->elevator->elevator_data;
struct as_rq *arq = mempool_alloc(ad->arq_pool, gfp_mask);
@@ -1827,7 +1828,7 @@ static int as_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
return 1;
}
-static int as_may_queue(request_queue_t *q, int rw)
+static int as_may_queue(request_queue_t *q, int rw, struct bio *bio)
{
int ret = ELV_MQUEUE_MAY;
struct as_data *ad = q->elevator->elevator_data;
diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index abde27027c06..3e9fb6e4a52a 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -1,6 +1,6 @@
/*
* Disk Array driver for HP SA 5xxx and 6xxx Controllers
- * Copyright 2000, 2002 Hewlett-Packard Development Company, L.P.
+ * Copyright 2000, 2005 Hewlett-Packard Development Company, L.P.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -54,7 +54,7 @@
MODULE_AUTHOR("Hewlett-Packard Company");
MODULE_DESCRIPTION("Driver for HP Controller SA5xxx SA6xxx version 2.6.6");
MODULE_SUPPORTED_DEVICE("HP SA5i SA5i+ SA532 SA5300 SA5312 SA641 SA642 SA6400"
- " SA6i P600 P800 E400");
+ " SA6i P600 P800 E400 E300");
MODULE_LICENSE("GPL");
#include "cciss_cmd.h"
@@ -85,8 +85,10 @@ static const struct pci_device_id cciss_pci_device_id[] = {
0x103C, 0x3225, 0, 0, 0},
{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSB,
0x103c, 0x3223, 0, 0, 0},
- { PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSB,
+ { PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC,
0x103c, 0x3231, 0, 0, 0},
+ { PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC,
+ 0x103c, 0x3233, 0, 0, 0},
{0,}
};
MODULE_DEVICE_TABLE(pci, cciss_pci_device_id);
@@ -110,6 +112,7 @@ static struct board_type products[] = {
{ 0x3225103C, "Smart Array P600", &SA5_access},
{ 0x3223103C, "Smart Array P800", &SA5_access},
{ 0x3231103C, "Smart Array E400", &SA5_access},
+ { 0x3233103C, "Smart Array E300", &SA5_access},
};
/* How long to wait (in millesconds) for board to go into simple mode */
@@ -635,6 +638,7 @@ static int cciss_ioctl(struct inode *inode, struct file *filep,
cciss_pci_info_struct pciinfo;
if (!arg) return -EINVAL;
+ pciinfo.domain = pci_domain_nr(host->pdev->bus);
pciinfo.bus = host->pdev->bus->number;
pciinfo.dev_fn = host->pdev->devfn;
pciinfo.board_id = host->board_id;
@@ -782,18 +786,10 @@ static int cciss_ioctl(struct inode *inode, struct file *filep,
case CCISS_GETLUNINFO: {
LogvolInfo_struct luninfo;
- int i;
luninfo.LunID = drv->LunID;
luninfo.num_opens = drv->usage_count;
luninfo.num_parts = 0;
- /* count partitions 1 to 15 with sizes > 0 */
- for (i = 0; i < MAX_PART - 1; i++) {
- if (!disk->part[i])
- continue;
- if (disk->part[i]->nr_sects != 0)
- luninfo.num_parts++;
- }
if (copy_to_user(argp, &luninfo,
sizeof(LogvolInfo_struct)))
return -EFAULT;
diff --git a/drivers/block/cfq-iosched.c b/drivers/block/cfq-iosched.c
index 3ac47dde64da..de5746e38af9 100644
--- a/drivers/block/cfq-iosched.c
+++ b/drivers/block/cfq-iosched.c
@@ -21,22 +21,34 @@
#include
#include
#include
-
-static unsigned long max_elapsed_crq;
-static unsigned long max_elapsed_dispatch;
+#include
+#include
/*
* tunables
*/
static int cfq_quantum = 4; /* max queue in one round of service */
static int cfq_queued = 8; /* minimum rq allocate limit per-queue*/
-static int cfq_service = HZ; /* period over which service is avg */
-static int cfq_fifo_expire_r = HZ / 2; /* fifo timeout for sync requests */
-static int cfq_fifo_expire_w = 5 * HZ; /* fifo timeout for async requests */
-static int cfq_fifo_rate = HZ / 8; /* fifo expiry rate */
+static int cfq_fifo_expire[2] = { HZ / 4, HZ / 8 };
static int cfq_back_max = 16 * 1024; /* maximum backwards seek, in KiB */
static int cfq_back_penalty = 2; /* penalty of a backwards seek */
+static int cfq_slice_sync = HZ / 10;
+static int cfq_slice_async = HZ / 25;
+static int cfq_slice_async_rq = 2;
+static int cfq_slice_idle = HZ / 100;
+
+#define CFQ_IDLE_GRACE (HZ / 10)
+#define CFQ_SLICE_SCALE (5)
+
+#define CFQ_KEY_ASYNC (0)
+#define CFQ_KEY_ANY (0xffff)
+
+/*
+ * disable queueing at the driver/hardware level
+ */
+static int cfq_max_depth = 1;
+
/*
* for the hash of cfqq inside the cfqd
*/
@@ -55,6 +67,7 @@ static int cfq_back_penalty = 2; /* penalty of a backwards seek */
#define list_entry_hash(ptr) hlist_entry((ptr), struct cfq_rq, hash)
#define list_entry_cfqq(ptr) list_entry((ptr), struct cfq_queue, cfq_list)
+#define list_entry_fifo(ptr) list_entry((ptr), struct request, queuelist)
#define RQ_DATA(rq) (rq)->elevator_private
@@ -75,78 +88,110 @@ static int cfq_back_penalty = 2; /* penalty of a backwards seek */
#define rb_entry_crq(node) rb_entry((node), struct cfq_rq, rb_node)
#define rq_rb_key(rq) (rq)->sector
-/*
- * threshold for switching off non-tag accounting
- */
-#define CFQ_MAX_TAG (4)
-
-/*
- * sort key types and names
- */
-enum {
- CFQ_KEY_PGID,
- CFQ_KEY_TGID,
- CFQ_KEY_UID,
- CFQ_KEY_GID,
- CFQ_KEY_LAST,
-};
-
-static char *cfq_key_types[] = { "pgid", "tgid", "uid", "gid", NULL };
-
static kmem_cache_t *crq_pool;
static kmem_cache_t *cfq_pool;
static kmem_cache_t *cfq_ioc_pool;
+#define CFQ_PRIO_LISTS IOPRIO_BE_NR
+#define cfq_class_idle(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_IDLE)
+#define cfq_class_be(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_BE)
+#define cfq_class_rt(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_RT)
+
+#define ASYNC (0)
+#define SYNC (1)
+
+#define cfq_cfqq_dispatched(cfqq) \
+ ((cfqq)->on_dispatch[ASYNC] + (cfqq)->on_dispatch[SYNC])
+
+#define cfq_cfqq_class_sync(cfqq) ((cfqq)->key != CFQ_KEY_ASYNC)
+
+#define cfq_cfqq_sync(cfqq) \
+ (cfq_cfqq_class_sync(cfqq) || (cfqq)->on_dispatch[SYNC])
+
+/*
+ * Per block device queue structure
+ */
struct cfq_data {
- struct list_head rr_list;
+ atomic_t ref;
+ request_queue_t *queue;
+
+ /*
+ * rr list of queues with requests and the count of them
+ */
+ struct list_head rr_list[CFQ_PRIO_LISTS];
+ struct list_head busy_rr;
+ struct list_head cur_rr;
+ struct list_head idle_rr;
+ unsigned int busy_queues;
+
+ /*
+ * non-ordered list of empty cfqq's
+ */
struct list_head empty_list;
+ /*
+ * cfqq lookup hash
+ */
struct hlist_head *cfq_hash;
- struct hlist_head *crq_hash;
- /* queues on rr_list (ie they have pending requests */
- unsigned int busy_queues;
+ /*
+ * global crq hash for all queues
+ */
+ struct hlist_head *crq_hash;
unsigned int max_queued;
- atomic_t ref;
-
- int key_type;
-
mempool_t *crq_pool;
- request_queue_t *queue;
+ int rq_in_driver;
+
+ /*
+ * schedule slice state info
+ */
+ /*
+ * idle window management
+ */
+ struct timer_list idle_slice_timer;
+ struct work_struct unplug_work;
+
+ struct cfq_queue *active_queue;
+ struct cfq_io_context *active_cic;
+ int cur_prio, cur_end_prio;
+ unsigned int dispatch_slice;
+
+ struct timer_list idle_class_timer;
sector_t last_sector;
+ unsigned long last_end_request;
- int rq_in_driver;
+ unsigned int rq_starved;
/*
* tunables, see top of file
*/
unsigned int cfq_quantum;
unsigned int cfq_queued;
- unsigned int cfq_fifo_expire_r;
- unsigned int cfq_fifo_expire_w;
- unsigned int cfq_fifo_batch_expire;
+ unsigned int cfq_fifo_expire[2];
unsigned int cfq_back_penalty;
unsigned int cfq_back_max;
- unsigned int find_best_crq;
-
- unsigned int cfq_tagged;
+ unsigned int cfq_slice[2];
+ unsigned int cfq_slice_async_rq;
+ unsigned int cfq_slice_idle;
+ unsigned int cfq_max_depth;
};
+/*
+ * Per process-grouping structure
+ */
struct cfq_queue {
/* reference count */
atomic_t ref;
/* parent cfq_data */
struct cfq_data *cfqd;
- /* hash of mergeable requests */
+ /* cfqq lookup hash */
struct hlist_node cfq_hash;
/* hash key */
- unsigned long key;
- /* whether queue is on rr (or empty) list */
- int on_rr;
+ unsigned int key;
/* on either rr or empty list of cfqd */
struct list_head cfq_list;
/* sorted list of pending requests */
@@ -158,21 +203,22 @@ struct cfq_queue {
/* currently allocated requests */
int allocated[2];
/* fifo list of requests in sort_list */
- struct list_head fifo[2];
- /* last time fifo expired */
- unsigned long last_fifo_expire;
+ struct list_head fifo;
- int key_type;
+ unsigned long slice_start;
+ unsigned long slice_end;
+ unsigned long slice_left;
+ unsigned long service_last;
- unsigned long service_start;
- unsigned long service_used;
+ /* number of requests that are on the dispatch list */
+ int on_dispatch[2];
- unsigned int max_rate;
+ /* io prio of this group */
+ unsigned short ioprio, org_ioprio;
+ unsigned short ioprio_class, org_ioprio_class;
- /* number of requests that have been handed to the driver */
- int in_flight;
- /* number of currently allocated requests */
- int alloc_limit[2];
+ /* various state flags, see below */
+ unsigned int flags;
};
struct cfq_rq {
@@ -184,42 +230,78 @@ struct cfq_rq {
struct cfq_queue *cfq_queue;
struct cfq_io_context *io_context;
- unsigned long service_start;
- unsigned long queue_start;
-
- unsigned int in_flight : 1;
- unsigned int accounted : 1;
- unsigned int is_sync : 1;
- unsigned int is_write : 1;
+ unsigned int crq_flags;
};
-static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *, unsigned long);
+enum cfqq_state_flags {
+ CFQ_CFQQ_FLAG_on_rr = 0,
+ CFQ_CFQQ_FLAG_wait_request,
+ CFQ_CFQQ_FLAG_must_alloc,
+ CFQ_CFQQ_FLAG_must_alloc_slice,
+ CFQ_CFQQ_FLAG_must_dispatch,
+ CFQ_CFQQ_FLAG_fifo_expire,
+ CFQ_CFQQ_FLAG_idle_window,
+ CFQ_CFQQ_FLAG_prio_changed,
+ CFQ_CFQQ_FLAG_expired,
+};
+
+#define CFQ_CFQQ_FNS(name) \
+static inline void cfq_mark_cfqq_##name(struct cfq_queue *cfqq) \
+{ \
+ cfqq->flags |= (1 << CFQ_CFQQ_FLAG_##name); \
+} \
+static inline void cfq_clear_cfqq_##name(struct cfq_queue *cfqq) \
+{ \
+ cfqq->flags &= ~(1 << CFQ_CFQQ_FLAG_##name); \
+} \
+static inline int cfq_cfqq_##name(const struct cfq_queue *cfqq) \
+{ \
+ return (cfqq->flags & (1 << CFQ_CFQQ_FLAG_##name)) != 0; \
+}
+
+CFQ_CFQQ_FNS(on_rr);
+CFQ_CFQQ_FNS(wait_request);
+CFQ_CFQQ_FNS(must_alloc);
+CFQ_CFQQ_FNS(must_alloc_slice);
+CFQ_CFQQ_FNS(must_dispatch);
+CFQ_CFQQ_FNS(fifo_expire);
+CFQ_CFQQ_FNS(idle_window);
+CFQ_CFQQ_FNS(prio_changed);
+CFQ_CFQQ_FNS(expired);
+#undef CFQ_CFQQ_FNS
+
+enum cfq_rq_state_flags {
+ CFQ_CRQ_FLAG_in_flight = 0,
+ CFQ_CRQ_FLAG_in_driver,
+ CFQ_CRQ_FLAG_is_sync,
+ CFQ_CRQ_FLAG_requeued,
+};
+
+#define CFQ_CRQ_FNS(name) \
+static inline void cfq_mark_crq_##name(struct cfq_rq *crq) \
+{ \
+ crq->crq_flags |= (1 << CFQ_CRQ_FLAG_##name); \
+} \
+static inline void cfq_clear_crq_##name(struct cfq_rq *crq) \
+{ \
+ crq->crq_flags &= ~(1 << CFQ_CRQ_FLAG_##name); \
+} \
+static inline int cfq_crq_##name(const struct cfq_rq *crq) \
+{ \
+ return (crq->crq_flags & (1 << CFQ_CRQ_FLAG_##name)) != 0; \
+}
+
+CFQ_CRQ_FNS(in_flight);
+CFQ_CRQ_FNS(in_driver);
+CFQ_CRQ_FNS(is_sync);
+CFQ_CRQ_FNS(requeued);
+#undef CFQ_CRQ_FNS
+
+static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *, unsigned int, unsigned short);
static void cfq_dispatch_sort(request_queue_t *, struct cfq_rq *);
-static void cfq_update_next_crq(struct cfq_rq *);
static void cfq_put_cfqd(struct cfq_data *cfqd);
-/*
- * what the fairness is based on (ie how processes are grouped and
- * differentiated)
- */
-static inline unsigned long
-cfq_hash_key(struct cfq_data *cfqd, struct task_struct *tsk)
-{
- /*
- * optimize this so that ->key_type is the offset into the struct
- */
- switch (cfqd->key_type) {
- case CFQ_KEY_PGID:
- return process_group(tsk);
- default:
- case CFQ_KEY_TGID:
- return tsk->tgid;
- case CFQ_KEY_UID:
- return tsk->uid;
- case CFQ_KEY_GID:
- return tsk->gid;
- }
-}
+#define process_sync(tsk) ((tsk)->flags & PF_SYNCWRITE)
/*
* lots of deadline iosched dupes, can be abstracted later...
@@ -235,16 +317,12 @@ static void cfq_remove_merge_hints(request_queue_t *q, struct cfq_rq *crq)
if (q->last_merge == crq->request)
q->last_merge = NULL;
-
- cfq_update_next_crq(crq);
}
static inline void cfq_add_crq_hash(struct cfq_data *cfqd, struct cfq_rq *crq)
{
const int hash_idx = CFQ_MHASH_FN(rq_hash_key(crq->request));
- BUG_ON(!hlist_unhashed(&crq->hash));
-
hlist_add_head(&crq->hash, &cfqd->crq_hash[hash_idx]);
}
@@ -257,8 +335,6 @@ static struct request *cfq_find_rq_hash(struct cfq_data *cfqd, sector_t offset)
struct cfq_rq *crq = list_entry_hash(entry);
struct request *__rq = crq->request;
- BUG_ON(hlist_unhashed(&crq->hash));
-
if (!rq_mergeable(__rq)) {
cfq_del_crq_hash(crq);
continue;
@@ -271,6 +347,28 @@ static struct request *cfq_find_rq_hash(struct cfq_data *cfqd, sector_t offset)
return NULL;
}
+static inline int cfq_pending_requests(struct cfq_data *cfqd)
+{
+ return !list_empty(&cfqd->queue->queue_head) || cfqd->busy_queues;
+}
+
+/*
+ * scheduler run of queue, if there are requests pending and no one in the
+ * driver that will restart queueing
+ */
+static inline void cfq_schedule_dispatch(struct cfq_data *cfqd)
+{
+ if (!cfqd->rq_in_driver && cfq_pending_requests(cfqd))
+ kblockd_schedule_work(&cfqd->unplug_work);
+}
+
+static int cfq_queue_empty(request_queue_t *q)
+{
+ struct cfq_data *cfqd = q->elevator->elevator_data;
+
+ return !cfq_pending_requests(cfqd);
+}
+
/*
* Lifted from AS - choose which of crq1 and crq2 that is best served now.
* We choose the request that is closest to the head right now. Distance
@@ -287,36 +385,16 @@ cfq_choose_req(struct cfq_data *cfqd, struct cfq_rq *crq1, struct cfq_rq *crq2)
return crq2;
if (crq2 == NULL)
return crq1;
+ if (cfq_crq_requeued(crq1))
+ return crq1;
+ if (cfq_crq_requeued(crq2))
+ return crq2;
s1 = crq1->request->sector;
s2 = crq2->request->sector;
last = cfqd->last_sector;
-#if 0
- if (!list_empty(&cfqd->queue->queue_head)) {
- struct list_head *entry = &cfqd->queue->queue_head;
- unsigned long distance = ~0UL;
- struct request *rq;
-
- while ((entry = entry->prev) != &cfqd->queue->queue_head) {
- rq = list_entry_rq(entry);
-
- if (blk_barrier_rq(rq))
- break;
-
- if (distance < abs(s1 - rq->sector + rq->nr_sectors)) {
- distance = abs(s1 - rq->sector +rq->nr_sectors);
- last = rq->sector + rq->nr_sectors;
- }
- if (distance < abs(s2 - rq->sector + rq->nr_sectors)) {
- distance = abs(s2 - rq->sector +rq->nr_sectors);
- last = rq->sector + rq->nr_sectors;
- }
- }
- }
-#endif
-
/*
* by definition, 1KiB is 2 sectors
*/
@@ -377,11 +455,14 @@ cfq_find_next_crq(struct cfq_data *cfqd, struct cfq_queue *cfqq,
struct cfq_rq *crq_next = NULL, *crq_prev = NULL;
struct rb_node *rbnext, *rbprev;
- if (!ON_RB(&last->rb_node))
- return NULL;
-
- if ((rbnext = rb_next(&last->rb_node)) == NULL)
+ rbnext = NULL;
+ if (ON_RB(&last->rb_node))
+ rbnext = rb_next(&last->rb_node);
+ if (!rbnext) {
rbnext = rb_first(&cfqq->sort_list);
+ if (rbnext == &last->rb_node)
+ rbnext = NULL;
+ }
rbprev = rb_prev(&last->rb_node);
@@ -401,67 +482,53 @@ static void cfq_update_next_crq(struct cfq_rq *crq)
cfqq->next_crq = cfq_find_next_crq(cfqq->cfqd, cfqq, crq);
}
-static int cfq_check_sort_rr_list(struct cfq_queue *cfqq)
+static void cfq_resort_rr_list(struct cfq_queue *cfqq, int preempted)
{
- struct list_head *head = &cfqq->cfqd->rr_list;
- struct list_head *next, *prev;
+ struct cfq_data *cfqd = cfqq->cfqd;
+ struct list_head *list, *entry;
- /*
- * list might still be ordered
- */
- next = cfqq->cfq_list.next;
- if (next != head) {
- struct cfq_queue *cnext = list_entry_cfqq(next);
-
- if (cfqq->service_used > cnext->service_used)
- return 1;
- }
-
- prev = cfqq->cfq_list.prev;
- if (prev != head) {
- struct cfq_queue *cprev = list_entry_cfqq(prev);
-
- if (cfqq->service_used < cprev->service_used)
- return 1;
- }
-
- return 0;
-}
-
-static void cfq_sort_rr_list(struct cfq_queue *cfqq, int new_queue)
-{
- struct list_head *entry = &cfqq->cfqd->rr_list;
-
- if (!cfqq->on_rr)
- return;
- if (!new_queue && !cfq_check_sort_rr_list(cfqq))
- return;
+ BUG_ON(!cfq_cfqq_on_rr(cfqq));
list_del(&cfqq->cfq_list);
+ if (cfq_class_rt(cfqq))
+ list = &cfqd->cur_rr;
+ else if (cfq_class_idle(cfqq))
+ list = &cfqd->idle_rr;
+ else {
+ /*
+ * if cfqq has requests in flight, don't allow it to be
+ * found in cfq_set_active_queue before it has finished them.
+ * this is done to increase fairness between a process that
+ * has lots of io pending vs one that only generates one
+ * sporadically or synchronously
+ */
+ if (cfq_cfqq_dispatched(cfqq))
+ list = &cfqd->busy_rr;
+ else
+ list = &cfqd->rr_list[cfqq->ioprio];
+ }
+
/*
- * sort by our mean service_used, sub-sort by in-flight requests
+ * if queue was preempted, just add to front to be fair. busy_rr
+ * isn't sorted.
*/
- while ((entry = entry->prev) != &cfqq->cfqd->rr_list) {
+ if (preempted || list == &cfqd->busy_rr) {
+ list_add(&cfqq->cfq_list, list);
+ return;
+ }
+
+ /*
+ * sort by when queue was last serviced
+ */
+ entry = list;
+ while ((entry = entry->prev) != list) {
struct cfq_queue *__cfqq = list_entry_cfqq(entry);
- if (cfqq->service_used > __cfqq->service_used)
+ if (!__cfqq->service_last)
+ break;
+ if (time_before(__cfqq->service_last, cfqq->service_last))
break;
- else if (cfqq->service_used == __cfqq->service_used) {
- struct list_head *prv;
-
- while ((prv = entry->prev) != &cfqq->cfqd->rr_list) {
- __cfqq = list_entry_cfqq(prv);
-
- WARN_ON(__cfqq->service_used > cfqq->service_used);
- if (cfqq->service_used != __cfqq->service_used)
- break;
- if (cfqq->in_flight > __cfqq->in_flight)
- break;
-
- entry = prv;
- }
- }
}
list_add(&cfqq->cfq_list, entry);
@@ -469,28 +536,24 @@ static void cfq_sort_rr_list(struct cfq_queue *cfqq, int new_queue)
/*
* add to busy list of queues for service, trying to be fair in ordering
- * the pending list according to requests serviced
+ * the pending list according to last request service
*/
static inline void
-cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+cfq_add_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq, int requeue)
{
- /*
- * it's currently on the empty list
- */
- cfqq->on_rr = 1;
+ BUG_ON(cfq_cfqq_on_rr(cfqq));
+ cfq_mark_cfqq_on_rr(cfqq);
cfqd->busy_queues++;
- if (time_after(jiffies, cfqq->service_start + cfq_service))
- cfqq->service_used >>= 3;
-
- cfq_sort_rr_list(cfqq, 1);
+ cfq_resort_rr_list(cfqq, requeue);
}
static inline void
cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq)
{
+ BUG_ON(!cfq_cfqq_on_rr(cfqq));
+ cfq_clear_cfqq_on_rr(cfqq);
list_move(&cfqq->cfq_list, &cfqd->empty_list);
- cfqq->on_rr = 0;
BUG_ON(!cfqd->busy_queues);
cfqd->busy_queues--;
@@ -505,16 +568,17 @@ static inline void cfq_del_crq_rb(struct cfq_rq *crq)
if (ON_RB(&crq->rb_node)) {
struct cfq_data *cfqd = cfqq->cfqd;
+ const int sync = cfq_crq_is_sync(crq);
- BUG_ON(!cfqq->queued[crq->is_sync]);
+ BUG_ON(!cfqq->queued[sync]);
+ cfqq->queued[sync]--;
cfq_update_next_crq(crq);
- cfqq->queued[crq->is_sync]--;
rb_erase(&crq->rb_node, &cfqq->sort_list);
RB_CLEAR_COLOR(&crq->rb_node);
- if (RB_EMPTY(&cfqq->sort_list) && cfqq->on_rr)
+ if (cfq_cfqq_on_rr(cfqq) && RB_EMPTY(&cfqq->sort_list))
cfq_del_cfqq_rr(cfqd, cfqq);
}
}
@@ -550,7 +614,7 @@ static void cfq_add_crq_rb(struct cfq_rq *crq)
struct cfq_rq *__alias;
crq->rb_key = rq_rb_key(rq);
- cfqq->queued[crq->is_sync]++;
+ cfqq->queued[cfq_crq_is_sync(crq)]++;
/*
* looks a little odd, but the first insert might return an alias.
@@ -561,8 +625,8 @@ static void cfq_add_crq_rb(struct cfq_rq *crq)
rb_insert_color(&crq->rb_node, &cfqq->sort_list);
- if (!cfqq->on_rr)
- cfq_add_cfqq_rr(cfqd, cfqq);
+ if (!cfq_cfqq_on_rr(cfqq))
+ cfq_add_cfqq_rr(cfqd, cfqq, cfq_crq_requeued(crq));
/*
* check if this request is a better next-serve candidate
@@ -575,17 +639,16 @@ cfq_reposition_crq_rb(struct cfq_queue *cfqq, struct cfq_rq *crq)
{
if (ON_RB(&crq->rb_node)) {
rb_erase(&crq->rb_node, &cfqq->sort_list);
- cfqq->queued[crq->is_sync]--;
+ cfqq->queued[cfq_crq_is_sync(crq)]--;
}
cfq_add_crq_rb(crq);
}
-static struct request *
-cfq_find_rq_rb(struct cfq_data *cfqd, sector_t sector)
+static struct request *cfq_find_rq_rb(struct cfq_data *cfqd, sector_t sector)
+
{
- const unsigned long key = cfq_hash_key(cfqd, current);
- struct cfq_queue *cfqq = cfq_find_cfq_hash(cfqd, key);
+ struct cfq_queue *cfqq = cfq_find_cfq_hash(cfqd, current->pid, CFQ_KEY_ANY);
struct rb_node *n;
if (!cfqq)
@@ -609,20 +672,25 @@ out:
static void cfq_deactivate_request(request_queue_t *q, struct request *rq)
{
+ struct cfq_data *cfqd = q->elevator->elevator_data;
struct cfq_rq *crq = RQ_DATA(rq);
if (crq) {
struct cfq_queue *cfqq = crq->cfq_queue;
- if (cfqq->cfqd->cfq_tagged) {
- cfqq->service_used--;
- cfq_sort_rr_list(cfqq, 0);
+ if (cfq_crq_in_driver(crq)) {
+ cfq_clear_crq_in_driver(crq);
+ WARN_ON(!cfqd->rq_in_driver);
+ cfqd->rq_in_driver--;
}
+ if (cfq_crq_in_flight(crq)) {
+ const int sync = cfq_crq_is_sync(crq);
- if (crq->accounted) {
- crq->accounted = 0;
- cfqq->cfqd->rq_in_driver--;
+ cfq_clear_crq_in_flight(crq);
+ WARN_ON(!cfqq->on_dispatch[sync]);
+ cfqq->on_dispatch[sync]--;
}
+ cfq_mark_crq_requeued(crq);
}
}
@@ -640,11 +708,10 @@ static void cfq_remove_request(request_queue_t *q, struct request *rq)
struct cfq_rq *crq = RQ_DATA(rq);
if (crq) {
- cfq_remove_merge_hints(q, crq);
list_del_init(&rq->queuelist);
+ cfq_del_crq_rb(crq);
+ cfq_remove_merge_hints(q, crq);
- if (crq->cfq_queue)
- cfq_del_crq_rb(crq);
}
}
@@ -662,21 +729,15 @@ cfq_merge(request_queue_t *q, struct request **req, struct bio *bio)
}
__rq = cfq_find_rq_hash(cfqd, bio->bi_sector);
- if (__rq) {
- BUG_ON(__rq->sector + __rq->nr_sectors != bio->bi_sector);
-
- if (elv_rq_merge_ok(__rq, bio)) {
- ret = ELEVATOR_BACK_MERGE;
- goto out;
- }
+ if (__rq && elv_rq_merge_ok(__rq, bio)) {
+ ret = ELEVATOR_BACK_MERGE;
+ goto out;
}
__rq = cfq_find_rq_rb(cfqd, bio->bi_sector + bio_sectors(bio));
- if (__rq) {
- if (elv_rq_merge_ok(__rq, bio)) {
- ret = ELEVATOR_FRONT_MERGE;
- goto out;
- }
+ if (__rq && elv_rq_merge_ok(__rq, bio)) {
+ ret = ELEVATOR_FRONT_MERGE;
+ goto out;
}
return ELEVATOR_NO_MERGE;
@@ -709,20 +770,220 @@ static void
cfq_merged_requests(request_queue_t *q, struct request *rq,
struct request *next)
{
- struct cfq_rq *crq = RQ_DATA(rq);
- struct cfq_rq *cnext = RQ_DATA(next);
-
cfq_merged_request(q, rq);
- if (!list_empty(&rq->queuelist) && !list_empty(&next->queuelist)) {
- if (time_before(cnext->queue_start, crq->queue_start)) {
- list_move(&rq->queuelist, &next->queuelist);
- crq->queue_start = cnext->queue_start;
- }
+ /*
+ * reposition in fifo if next is older than rq
+ */
+ if (!list_empty(&rq->queuelist) && !list_empty(&next->queuelist) &&
+ time_before(next->start_time, rq->start_time))
+ list_move(&rq->queuelist, &next->queuelist);
+
+ cfq_remove_request(q, next);
+}
+
+static inline void
+__cfq_set_active_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+ if (cfqq) {
+ /*
+ * stop potential idle class queues waiting service
+ */
+ del_timer(&cfqd->idle_class_timer);
+
+ cfqq->slice_start = jiffies;
+ cfqq->slice_end = 0;
+ cfqq->slice_left = 0;
+ cfq_clear_cfqq_must_alloc_slice(cfqq);
+ cfq_clear_cfqq_fifo_expire(cfqq);
+ cfq_clear_cfqq_expired(cfqq);
}
- cfq_update_next_crq(cnext);
- cfq_remove_request(q, next);
+ cfqd->active_queue = cfqq;
+}
+
+/*
+ * 0
+ * 0,1
+ * 0,1,2
+ * 0,1,2,3
+ * 0,1,2,3,4
+ * 0,1,2,3,4,5
+ * 0,1,2,3,4,5,6
+ * 0,1,2,3,4,5,6,7
+ */
+static int cfq_get_next_prio_level(struct cfq_data *cfqd)
+{
+ int prio, wrap;
+
+ prio = -1;
+ wrap = 0;
+ do {
+ int p;
+
+ for (p = cfqd->cur_prio; p <= cfqd->cur_end_prio; p++) {
+ if (!list_empty(&cfqd->rr_list[p])) {
+ prio = p;
+ break;
+ }
+ }
+
+ if (prio != -1)
+ break;
+ cfqd->cur_prio = 0;
+ if (++cfqd->cur_end_prio == CFQ_PRIO_LISTS) {
+ cfqd->cur_end_prio = 0;
+ if (wrap)
+ break;
+ wrap = 1;
+ }
+ } while (1);
+
+ if (unlikely(prio == -1))
+ return -1;
+
+ BUG_ON(prio >= CFQ_PRIO_LISTS);
+
+ list_splice_init(&cfqd->rr_list[prio], &cfqd->cur_rr);
+
+ cfqd->cur_prio = prio + 1;
+ if (cfqd->cur_prio > cfqd->cur_end_prio) {
+ cfqd->cur_end_prio = cfqd->cur_prio;
+ cfqd->cur_prio = 0;
+ }
+ if (cfqd->cur_end_prio == CFQ_PRIO_LISTS) {
+ cfqd->cur_prio = 0;
+ cfqd->cur_end_prio = 0;
+ }
+
+ return prio;
+}
+
+static struct cfq_queue *cfq_set_active_queue(struct cfq_data *cfqd)
+{
+ struct cfq_queue *cfqq;
+
+ /*
+ * if current queue is expired but not done with its requests yet,
+ * wait for that to happen
+ */
+ if ((cfqq = cfqd->active_queue) != NULL) {
+ if (cfq_cfqq_expired(cfqq) && cfq_cfqq_dispatched(cfqq))
+ return NULL;
+ }
+
+ /*
+ * if current list is non-empty, grab first entry. if it is empty,
+ * get next prio level and grab first entry then if any are spliced
+ */
+ if (!list_empty(&cfqd->cur_rr) || cfq_get_next_prio_level(cfqd) != -1)
+ cfqq = list_entry_cfqq(cfqd->cur_rr.next);
+
+ /*
+ * if we have idle queues and no rt or be queues had pending
+ * requests, either allow immediate service if the grace period
+ * has passed or arm the idle grace timer
+ */
+ if (!cfqq && !list_empty(&cfqd->idle_rr)) {
+ unsigned long end = cfqd->last_end_request + CFQ_IDLE_GRACE;
+
+ if (time_after_eq(jiffies, end))
+ cfqq = list_entry_cfqq(cfqd->idle_rr.next);
+ else
+ mod_timer(&cfqd->idle_class_timer, end);
+ }
+
+ __cfq_set_active_queue(cfqd, cfqq);
+ return cfqq;
+}
+
+/*
+ * current cfqq expired its slice (or was too idle), select new one
+ */
+static void
+__cfq_slice_expired(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+ int preempted)
+{
+ unsigned long now = jiffies;
+
+ if (cfq_cfqq_wait_request(cfqq))
+ del_timer(&cfqd->idle_slice_timer);
+
+ if (!preempted && !cfq_cfqq_dispatched(cfqq))
+ cfqq->service_last = now;
+
+ cfq_clear_cfqq_must_dispatch(cfqq);
+ cfq_clear_cfqq_wait_request(cfqq);
+
+ /*
+ * store what was left of this slice, if the queue idled out
+ * or was preempted
+ */
+ if (time_after(now, cfqq->slice_end))
+ cfqq->slice_left = now - cfqq->slice_end;
+ else
+ cfqq->slice_left = 0;
+
+ if (cfq_cfqq_on_rr(cfqq))
+ cfq_resort_rr_list(cfqq, preempted);
+
+ if (cfqq == cfqd->active_queue)
+ cfqd->active_queue = NULL;
+
+ if (cfqd->active_cic) {
+ put_io_context(cfqd->active_cic->ioc);
+ cfqd->active_cic = NULL;
+ }
+
+ cfqd->dispatch_slice = 0;
+}
+
+static inline void cfq_slice_expired(struct cfq_data *cfqd, int preempted)
+{
+ struct cfq_queue *cfqq = cfqd->active_queue;
+
+ if (cfqq) {
+ /*
+ * use deferred expiry, if there are requests in progress as
+ * not to disturb the slice of the next queue
+ */
+ if (cfq_cfqq_dispatched(cfqq))
+ cfq_mark_cfqq_expired(cfqq);
+ else
+ __cfq_slice_expired(cfqd, cfqq, preempted);
+ }
+}
+
+static int cfq_arm_slice_timer(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+
+{
+ WARN_ON(!RB_EMPTY(&cfqq->sort_list));
+ WARN_ON(cfqq != cfqd->active_queue);
+
+ /*
+ * idle is disabled, either manually or by past process history
+ */
+ if (!cfqd->cfq_slice_idle)
+ return 0;
+ if (!cfq_cfqq_idle_window(cfqq))
+ return 0;
+ /*
+ * task has exited, don't wait
+ */
+ if (cfqd->active_cic && !cfqd->active_cic->ioc->task)
+ return 0;
+
+ cfq_mark_cfqq_must_dispatch(cfqq);
+ cfq_mark_cfqq_wait_request(cfqq);
+
+ if (!timer_pending(&cfqd->idle_slice_timer)) {
+ unsigned long slice_left = min(cfqq->slice_end - 1, (unsigned long) cfqd->cfq_slice_idle);
+
+ cfqd->idle_slice_timer.expires = jiffies + slice_left;
+ add_timer(&cfqd->idle_slice_timer);
+ }
+
+ return 1;
}
/*
@@ -738,31 +999,40 @@ static void cfq_dispatch_sort(request_queue_t *q, struct cfq_rq *crq)
struct request *__rq;
sector_t last;
- cfq_del_crq_rb(crq);
- cfq_remove_merge_hints(q, crq);
list_del(&crq->request->queuelist);
last = cfqd->last_sector;
- while ((entry = entry->prev) != head) {
- __rq = list_entry_rq(entry);
+ list_for_each_entry_reverse(__rq, head, queuelist) {
+ struct cfq_rq *__crq = RQ_DATA(__rq);
- if (blk_barrier_rq(crq->request))
+ if (blk_barrier_rq(__rq))
break;
- if (!blk_fs_request(crq->request))
+ if (!blk_fs_request(__rq))
+ break;
+ if (cfq_crq_requeued(__crq))
break;
- if (crq->request->sector > __rq->sector)
+ if (__rq->sector <= crq->request->sector)
break;
if (__rq->sector > last && crq->request->sector < last) {
- last = crq->request->sector;
+ last = crq->request->sector + crq->request->nr_sectors;
break;
}
+ entry = &__rq->queuelist;
}
cfqd->last_sector = last;
- crq->in_flight = 1;
- cfqq->in_flight++;
- list_add(&crq->request->queuelist, entry);
+
+ cfqq->next_crq = cfq_find_next_crq(cfqd, cfqq, crq);
+
+ cfq_del_crq_rb(crq);
+ cfq_remove_merge_hints(q, crq);
+
+ cfq_mark_crq_in_flight(crq);
+ cfq_clear_crq_requeued(crq);
+
+ cfqq->on_dispatch[cfq_crq_is_sync(crq)]++;
+ list_add_tail(&crq->request->queuelist, entry);
}
/*
@@ -771,173 +1041,225 @@ static void cfq_dispatch_sort(request_queue_t *q, struct cfq_rq *crq)
static inline struct cfq_rq *cfq_check_fifo(struct cfq_queue *cfqq)
{
struct cfq_data *cfqd = cfqq->cfqd;
- const int reads = !list_empty(&cfqq->fifo[0]);
- const int writes = !list_empty(&cfqq->fifo[1]);
- unsigned long now = jiffies;
+ struct request *rq;
struct cfq_rq *crq;
- if (time_before(now, cfqq->last_fifo_expire + cfqd->cfq_fifo_batch_expire))
+ if (cfq_cfqq_fifo_expire(cfqq))
return NULL;
- crq = RQ_DATA(list_entry(cfqq->fifo[0].next, struct request, queuelist));
- if (reads && time_after(now, crq->queue_start + cfqd->cfq_fifo_expire_r)) {
- cfqq->last_fifo_expire = now;
- return crq;
- }
+ if (!list_empty(&cfqq->fifo)) {
+ int fifo = cfq_cfqq_class_sync(cfqq);
- crq = RQ_DATA(list_entry(cfqq->fifo[1].next, struct request, queuelist));
- if (writes && time_after(now, crq->queue_start + cfqd->cfq_fifo_expire_w)) {
- cfqq->last_fifo_expire = now;
- return crq;
+ crq = RQ_DATA(list_entry_fifo(cfqq->fifo.next));
+ rq = crq->request;
+ if (time_after(jiffies, rq->start_time + cfqd->cfq_fifo_expire[fifo])) {
+ cfq_mark_cfqq_fifo_expire(cfqq);
+ return crq;
+ }
}
return NULL;
}
/*
- * dispatch a single request from given queue
+ * Scale schedule slice based on io priority. Use the sync time slice only
+ * if a queue is marked sync and has sync io queued. A sync queue with async
+ * io only, should not get full sync slice length.
*/
-static inline void
-cfq_dispatch_request(request_queue_t *q, struct cfq_data *cfqd,
- struct cfq_queue *cfqq)
+static inline int
+cfq_prio_to_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
{
- struct cfq_rq *crq;
+ const int base_slice = cfqd->cfq_slice[cfq_cfqq_sync(cfqq)];
- /*
- * follow expired path, else get first next available
- */
- if ((crq = cfq_check_fifo(cfqq)) == NULL) {
- if (cfqd->find_best_crq)
- crq = cfqq->next_crq;
- else
- crq = rb_entry_crq(rb_first(&cfqq->sort_list));
- }
+ WARN_ON(cfqq->ioprio >= IOPRIO_BE_NR);
- cfqd->last_sector = crq->request->sector + crq->request->nr_sectors;
-
- /*
- * finally, insert request into driver list
- */
- cfq_dispatch_sort(q, crq);
+ return base_slice + (base_slice/CFQ_SLICE_SCALE * (4 - cfqq->ioprio));
}
-static int cfq_dispatch_requests(request_queue_t *q, int max_dispatch)
+static inline void
+cfq_set_prio_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+ cfqq->slice_end = cfq_prio_to_slice(cfqd, cfqq) + jiffies;
+}
+
+static inline int
+cfq_prio_to_maxrq(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+ const int base_rq = cfqd->cfq_slice_async_rq;
+
+ WARN_ON(cfqq->ioprio >= IOPRIO_BE_NR);
+
+ return 2 * (base_rq + base_rq * (CFQ_PRIO_LISTS - 1 - cfqq->ioprio));
+}
+
+/*
+ * get next queue for service
+ */
+static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd, int force)
+{
+ unsigned long now = jiffies;
+ struct cfq_queue *cfqq;
+
+ cfqq = cfqd->active_queue;
+ if (!cfqq)
+ goto new_queue;
+
+ if (cfq_cfqq_expired(cfqq))
+ goto new_queue;
+
+ /*
+ * slice has expired
+ */
+ if (!cfq_cfqq_must_dispatch(cfqq) && time_after(now, cfqq->slice_end))
+ goto expire;
+
+ /*
+ * if queue has requests, dispatch one. if not, check if
+ * enough slice is left to wait for one
+ */
+ if (!RB_EMPTY(&cfqq->sort_list))
+ goto keep_queue;
+ else if (!force && cfq_cfqq_class_sync(cfqq) &&
+ time_before(now, cfqq->slice_end)) {
+ if (cfq_arm_slice_timer(cfqd, cfqq))
+ return NULL;
+ }
+
+expire:
+ cfq_slice_expired(cfqd, 0);
+new_queue:
+ cfqq = cfq_set_active_queue(cfqd);
+keep_queue:
+ return cfqq;
+}
+
+static int
+__cfq_dispatch_requests(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+ int max_dispatch)
+{
+ int dispatched = 0;
+
+ BUG_ON(RB_EMPTY(&cfqq->sort_list));
+
+ do {
+ struct cfq_rq *crq;
+
+ /*
+ * follow expired path, else get first next available
+ */
+ if ((crq = cfq_check_fifo(cfqq)) == NULL)
+ crq = cfqq->next_crq;
+
+ /*
+ * finally, insert request into driver dispatch list
+ */
+ cfq_dispatch_sort(cfqd->queue, crq);
+
+ cfqd->dispatch_slice++;
+ dispatched++;
+
+ if (!cfqd->active_cic) {
+ atomic_inc(&crq->io_context->ioc->refcount);
+ cfqd->active_cic = crq->io_context;
+ }
+
+ if (RB_EMPTY(&cfqq->sort_list))
+ break;
+
+ } while (dispatched < max_dispatch);
+
+ /*
+ * if slice end isn't set yet, set it. if at least one request was
+ * sync, use the sync time slice value
+ */
+ if (!cfqq->slice_end)
+ cfq_set_prio_slice(cfqd, cfqq);
+
+ /*
+ * expire an async queue immediately if it has used up its slice. idle
+ * queue always expire after 1 dispatch round.
+ */
+ if ((!cfq_cfqq_sync(cfqq) &&
+ cfqd->dispatch_slice >= cfq_prio_to_maxrq(cfqd, cfqq)) ||
+ cfq_class_idle(cfqq))
+ cfq_slice_expired(cfqd, 0);
+
+ return dispatched;
+}
+
+static int
+cfq_dispatch_requests(request_queue_t *q, int max_dispatch, int force)
{
struct cfq_data *cfqd = q->elevator->elevator_data;
struct cfq_queue *cfqq;
- struct list_head *entry, *tmp;
- int queued, busy_queues, first_round;
- if (list_empty(&cfqd->rr_list))
+ if (!cfqd->busy_queues)
return 0;
- queued = 0;
- first_round = 1;
-restart:
- busy_queues = 0;
- list_for_each_safe(entry, tmp, &cfqd->rr_list) {
- cfqq = list_entry_cfqq(entry);
+ cfqq = cfq_select_queue(cfqd, force);
+ if (cfqq) {
+ cfq_clear_cfqq_must_dispatch(cfqq);
+ cfq_clear_cfqq_wait_request(cfqq);
+ del_timer(&cfqd->idle_slice_timer);
- BUG_ON(RB_EMPTY(&cfqq->sort_list));
+ if (cfq_class_idle(cfqq))
+ max_dispatch = 1;
- /*
- * first round of queueing, only select from queues that
- * don't already have io in-flight
- */
- if (first_round && cfqq->in_flight)
- continue;
-
- cfq_dispatch_request(q, cfqd, cfqq);
-
- if (!RB_EMPTY(&cfqq->sort_list))
- busy_queues++;
-
- queued++;
+ return __cfq_dispatch_requests(cfqd, cfqq, max_dispatch);
}
- if ((queued < max_dispatch) && (busy_queues || first_round)) {
- first_round = 0;
- goto restart;
- }
-
- return queued;
+ return 0;
}
static inline void cfq_account_dispatch(struct cfq_rq *crq)
{
struct cfq_queue *cfqq = crq->cfq_queue;
struct cfq_data *cfqd = cfqq->cfqd;
- unsigned long now, elapsed;
- if (!blk_fs_request(crq->request))
+ if (unlikely(!blk_fs_request(crq->request)))
return;
/*
* accounted bit is necessary since some drivers will call
* elv_next_request() many times for the same request (eg ide)
*/
- if (crq->accounted)
+ if (cfq_crq_in_driver(crq))
return;
- now = jiffies;
- if (cfqq->service_start == ~0UL)
- cfqq->service_start = now;
-
- /*
- * on drives with tagged command queueing, command turn-around time
- * doesn't necessarily reflect the time spent processing this very
- * command inside the drive. so do the accounting differently there,
- * by just sorting on the number of requests
- */
- if (cfqd->cfq_tagged) {
- if (time_after(now, cfqq->service_start + cfq_service)) {
- cfqq->service_start = now;
- cfqq->service_used /= 10;
- }
-
- cfqq->service_used++;
- cfq_sort_rr_list(cfqq, 0);
- }
-
- elapsed = now - crq->queue_start;
- if (elapsed > max_elapsed_dispatch)
- max_elapsed_dispatch = elapsed;
-
- crq->accounted = 1;
- crq->service_start = now;
-
- if (++cfqd->rq_in_driver >= CFQ_MAX_TAG && !cfqd->cfq_tagged) {
- cfqq->cfqd->cfq_tagged = 1;
- printk("cfq: depth %d reached, tagging now on\n", CFQ_MAX_TAG);
- }
+ cfq_mark_crq_in_driver(crq);
+ cfqd->rq_in_driver++;
}
static inline void
cfq_account_completion(struct cfq_queue *cfqq, struct cfq_rq *crq)
{
struct cfq_data *cfqd = cfqq->cfqd;
+ unsigned long now;
- if (!crq->accounted)
+ if (!cfq_crq_in_driver(crq))
return;
+ now = jiffies;
+
WARN_ON(!cfqd->rq_in_driver);
cfqd->rq_in_driver--;
- if (!cfqd->cfq_tagged) {
- unsigned long now = jiffies;
- unsigned long duration = now - crq->service_start;
+ if (!cfq_class_idle(cfqq))
+ cfqd->last_end_request = now;
- if (time_after(now, cfqq->service_start + cfq_service)) {
- cfqq->service_start = now;
- cfqq->service_used >>= 3;
+ if (!cfq_cfqq_dispatched(cfqq)) {
+ if (cfq_cfqq_on_rr(cfqq)) {
+ cfqq->service_last = now;
+ cfq_resort_rr_list(cfqq, 0);
+ }
+ if (cfq_cfqq_expired(cfqq)) {
+ __cfq_slice_expired(cfqd, cfqq, 0);
+ cfq_schedule_dispatch(cfqd);
}
-
- cfqq->service_used += duration;
- cfq_sort_rr_list(cfqq, 0);
-
- if (duration > max_elapsed_crq)
- max_elapsed_crq = duration;
}
+
+ if (cfq_crq_is_sync(crq))
+ crq->io_context->last_end_request = now;
}
static struct request *cfq_next_request(request_queue_t *q)
@@ -950,7 +1272,18 @@ static struct request *cfq_next_request(request_queue_t *q)
dispatch:
rq = list_entry_rq(q->queue_head.next);
- if ((crq = RQ_DATA(rq)) != NULL) {
+ crq = RQ_DATA(rq);
+ if (crq) {
+ struct cfq_queue *cfqq = crq->cfq_queue;
+
+ /*
+ * if idle window is disabled, allow queue buildup
+ */
+ if (!cfq_crq_in_driver(crq) &&
+ !cfq_cfqq_idle_window(cfqq) &&
+ cfqd->rq_in_driver >= cfqd->cfq_max_depth)
+ return NULL;
+
cfq_remove_merge_hints(q, crq);
cfq_account_dispatch(crq);
}
@@ -958,7 +1291,7 @@ dispatch:
return rq;
}
- if (cfq_dispatch_requests(q, cfqd->cfq_quantum))
+ if (cfq_dispatch_requests(q, cfqd->cfq_quantum, 0))
goto dispatch;
return NULL;
@@ -972,13 +1305,21 @@ dispatch:
*/
static void cfq_put_queue(struct cfq_queue *cfqq)
{
- BUG_ON(!atomic_read(&cfqq->ref));
+ struct cfq_data *cfqd = cfqq->cfqd;
+
+ BUG_ON(atomic_read(&cfqq->ref) <= 0);
if (!atomic_dec_and_test(&cfqq->ref))
return;
BUG_ON(rb_first(&cfqq->sort_list));
- BUG_ON(cfqq->on_rr);
+ BUG_ON(cfqq->allocated[READ] + cfqq->allocated[WRITE]);
+ BUG_ON(cfq_cfqq_on_rr(cfqq));
+
+ if (unlikely(cfqd->active_queue == cfqq)) {
+ __cfq_slice_expired(cfqd, cfqq, 0);
+ cfq_schedule_dispatch(cfqd);
+ }
cfq_put_cfqd(cfqq->cfqd);
@@ -991,15 +1332,17 @@ static void cfq_put_queue(struct cfq_queue *cfqq)
}
static inline struct cfq_queue *
-__cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned long key, const int hashval)
+__cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key, unsigned int prio,
+ const int hashval)
{
struct hlist_head *hash_list = &cfqd->cfq_hash[hashval];
struct hlist_node *entry, *next;
hlist_for_each_safe(entry, next, hash_list) {
struct cfq_queue *__cfqq = list_entry_qhash(entry);
+ const unsigned short __p = IOPRIO_PRIO_VALUE(__cfqq->ioprio_class, __cfqq->ioprio);
- if (__cfqq->key == key)
+ if (__cfqq->key == key && (__p == prio || prio == CFQ_KEY_ANY))
return __cfqq;
}
@@ -1007,210 +1350,187 @@ __cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned long key, const int hashval)
}
static struct cfq_queue *
-cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned long key)
+cfq_find_cfq_hash(struct cfq_data *cfqd, unsigned int key, unsigned short prio)
{
- return __cfq_find_cfq_hash(cfqd, key, hash_long(key, CFQ_QHASH_SHIFT));
-}
-
-static inline void
-cfq_rehash_cfqq(struct cfq_data *cfqd, struct cfq_queue **cfqq,
- struct cfq_io_context *cic)
-{
- unsigned long hashkey = cfq_hash_key(cfqd, current);
- unsigned long hashval = hash_long(hashkey, CFQ_QHASH_SHIFT);
- struct cfq_queue *__cfqq;
- unsigned long flags;
-
- spin_lock_irqsave(cfqd->queue->queue_lock, flags);
-
- hlist_del(&(*cfqq)->cfq_hash);
-
- __cfqq = __cfq_find_cfq_hash(cfqd, hashkey, hashval);
- if (!__cfqq || __cfqq == *cfqq) {
- __cfqq = *cfqq;
- hlist_add_head(&__cfqq->cfq_hash, &cfqd->cfq_hash[hashval]);
- __cfqq->key_type = cfqd->key_type;
- } else {
- atomic_inc(&__cfqq->ref);
- cic->cfqq = __cfqq;
- cfq_put_queue(*cfqq);
- *cfqq = __cfqq;
- }
-
- cic->cfqq = __cfqq;
- spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
+ return __cfq_find_cfq_hash(cfqd, key, prio, hash_long(key, CFQ_QHASH_SHIFT));
}
static void cfq_free_io_context(struct cfq_io_context *cic)
{
+ struct cfq_io_context *__cic;
+ struct list_head *entry, *next;
+
+ list_for_each_safe(entry, next, &cic->list) {
+ __cic = list_entry(entry, struct cfq_io_context, list);
+ kmem_cache_free(cfq_ioc_pool, __cic);
+ }
+
kmem_cache_free(cfq_ioc_pool, cic);
}
/*
- * locking hierarchy is: io_context lock -> queue locks
+ * Called with interrupts disabled
+ */
+static void cfq_exit_single_io_context(struct cfq_io_context *cic)
+{
+ struct cfq_data *cfqd = cic->cfqq->cfqd;
+ request_queue_t *q = cfqd->queue;
+
+ WARN_ON(!irqs_disabled());
+
+ spin_lock(q->queue_lock);
+
+ if (unlikely(cic->cfqq == cfqd->active_queue)) {
+ __cfq_slice_expired(cfqd, cic->cfqq, 0);
+ cfq_schedule_dispatch(cfqd);
+ }
+
+ cfq_put_queue(cic->cfqq);
+ cic->cfqq = NULL;
+ spin_unlock(q->queue_lock);
+}
+
+/*
+ * Another task may update the task cic list, if it is doing a queue lookup
+ * on its behalf. cfq_cic_lock excludes such concurrent updates
*/
static void cfq_exit_io_context(struct cfq_io_context *cic)
{
- struct cfq_queue *cfqq = cic->cfqq;
- struct list_head *entry = &cic->list;
- request_queue_t *q;
+ struct cfq_io_context *__cic;
+ struct list_head *entry;
unsigned long flags;
+ local_irq_save(flags);
+
/*
* put the reference this task is holding to the various queues
*/
- spin_lock_irqsave(&cic->ioc->lock, flags);
- while ((entry = cic->list.next) != &cic->list) {
- struct cfq_io_context *__cic;
-
+ list_for_each(entry, &cic->list) {
__cic = list_entry(entry, struct cfq_io_context, list);
- list_del(entry);
-
- q = __cic->cfqq->cfqd->queue;
- spin_lock(q->queue_lock);
- cfq_put_queue(__cic->cfqq);
- spin_unlock(q->queue_lock);
+ cfq_exit_single_io_context(__cic);
}
- q = cfqq->cfqd->queue;
- spin_lock(q->queue_lock);
- cfq_put_queue(cfqq);
- spin_unlock(q->queue_lock);
-
- cic->cfqq = NULL;
- spin_unlock_irqrestore(&cic->ioc->lock, flags);
+ cfq_exit_single_io_context(cic);
+ local_irq_restore(flags);
}
-static struct cfq_io_context *cfq_alloc_io_context(int gfp_flags)
+static struct cfq_io_context *
+cfq_alloc_io_context(struct cfq_data *cfqd, int gfp_mask)
{
- struct cfq_io_context *cic = kmem_cache_alloc(cfq_ioc_pool, gfp_flags);
+ struct cfq_io_context *cic = kmem_cache_alloc(cfq_ioc_pool, gfp_mask);
if (cic) {
- cic->dtor = cfq_free_io_context;
- cic->exit = cfq_exit_io_context;
INIT_LIST_HEAD(&cic->list);
cic->cfqq = NULL;
+ cic->key = NULL;
+ cic->last_end_request = jiffies;
+ cic->ttime_total = 0;
+ cic->ttime_samples = 0;
+ cic->ttime_mean = 0;
+ cic->dtor = cfq_free_io_context;
+ cic->exit = cfq_exit_io_context;
}
return cic;
}
+static void cfq_init_prio_data(struct cfq_queue *cfqq)
+{
+ struct task_struct *tsk = current;
+ int ioprio_class;
+
+ if (!cfq_cfqq_prio_changed(cfqq))
+ return;
+
+ ioprio_class = IOPRIO_PRIO_CLASS(tsk->ioprio);
+ switch (ioprio_class) {
+ default:
+ printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class);
+ case IOPRIO_CLASS_NONE:
+ /*
+ * no prio set, place us in the middle of the BE classes
+ */
+ cfqq->ioprio = task_nice_ioprio(tsk);
+ cfqq->ioprio_class = IOPRIO_CLASS_BE;
+ break;
+ case IOPRIO_CLASS_RT:
+ cfqq->ioprio = task_ioprio(tsk);
+ cfqq->ioprio_class = IOPRIO_CLASS_RT;
+ break;
+ case IOPRIO_CLASS_BE:
+ cfqq->ioprio = task_ioprio(tsk);
+ cfqq->ioprio_class = IOPRIO_CLASS_BE;
+ break;
+ case IOPRIO_CLASS_IDLE:
+ cfqq->ioprio_class = IOPRIO_CLASS_IDLE;
+ cfqq->ioprio = 7;
+ cfq_clear_cfqq_idle_window(cfqq);
+ break;
+ }
+
+ /*
+ * keep track of original prio settings in case we have to temporarily
+ * elevate the priority of this queue
+ */
+ cfqq->org_ioprio = cfqq->ioprio;
+ cfqq->org_ioprio_class = cfqq->ioprio_class;
+
+ if (cfq_cfqq_on_rr(cfqq))
+ cfq_resort_rr_list(cfqq, 0);
+
+ cfq_clear_cfqq_prio_changed(cfqq);
+}
+
+static inline void changed_ioprio(struct cfq_queue *cfqq)
+{
+ if (cfqq) {
+ struct cfq_data *cfqd = cfqq->cfqd;
+
+ spin_lock(cfqd->queue->queue_lock);
+ cfq_mark_cfqq_prio_changed(cfqq);
+ cfq_init_prio_data(cfqq);
+ spin_unlock(cfqd->queue->queue_lock);
+ }
+}
+
/*
- * Setup general io context and cfq io context. There can be several cfq
- * io contexts per general io context, if this process is doing io to more
- * than one device managed by cfq. Note that caller is holding a reference to
- * cfqq, so we don't need to worry about it disappearing
+ * callback from sys_ioprio_set, irqs are disabled
*/
-static struct cfq_io_context *
-cfq_get_io_context(struct cfq_queue **cfqq, int gfp_flags)
+static int cfq_ioc_set_ioprio(struct io_context *ioc, unsigned int ioprio)
{
- struct cfq_data *cfqd = (*cfqq)->cfqd;
- struct cfq_queue *__cfqq = *cfqq;
- struct cfq_io_context *cic;
- struct io_context *ioc;
+ struct cfq_io_context *cic = ioc->cic;
- might_sleep_if(gfp_flags & __GFP_WAIT);
+ changed_ioprio(cic->cfqq);
- ioc = get_io_context(gfp_flags);
- if (!ioc)
- return NULL;
+ list_for_each_entry(cic, &cic->list, list)
+ changed_ioprio(cic->cfqq);
- if ((cic = ioc->cic) == NULL) {
- cic = cfq_alloc_io_context(gfp_flags);
-
- if (cic == NULL)
- goto err;
-
- ioc->cic = cic;
- cic->ioc = ioc;
- cic->cfqq = __cfqq;
- atomic_inc(&__cfqq->ref);
- } else {
- struct cfq_io_context *__cic;
- unsigned long flags;
-
- /*
- * since the first cic on the list is actually the head
- * itself, need to check this here or we'll duplicate an
- * cic per ioc for no reason
- */
- if (cic->cfqq == __cfqq)
- goto out;
-
- /*
- * cic exists, check if we already are there. linear search
- * should be ok here, the list will usually not be more than
- * 1 or a few entries long
- */
- spin_lock_irqsave(&ioc->lock, flags);
- list_for_each_entry(__cic, &cic->list, list) {
- /*
- * this process is already holding a reference to
- * this queue, so no need to get one more
- */
- if (__cic->cfqq == __cfqq) {
- cic = __cic;
- spin_unlock_irqrestore(&ioc->lock, flags);
- goto out;
- }
- }
- spin_unlock_irqrestore(&ioc->lock, flags);
-
- /*
- * nope, process doesn't have a cic assoicated with this
- * cfqq yet. get a new one and add to list
- */
- __cic = cfq_alloc_io_context(gfp_flags);
- if (__cic == NULL)
- goto err;
-
- __cic->ioc = ioc;
- __cic->cfqq = __cfqq;
- atomic_inc(&__cfqq->ref);
- spin_lock_irqsave(&ioc->lock, flags);
- list_add(&__cic->list, &cic->list);
- spin_unlock_irqrestore(&ioc->lock, flags);
-
- cic = __cic;
- *cfqq = __cfqq;
- }
-
-out:
- /*
- * if key_type has been changed on the fly, we lazily rehash
- * each queue at lookup time
- */
- if ((*cfqq)->key_type != cfqd->key_type)
- cfq_rehash_cfqq(cfqd, cfqq, cic);
-
- return cic;
-err:
- put_io_context(ioc);
- return NULL;
+ return 0;
}
static struct cfq_queue *
-__cfq_get_queue(struct cfq_data *cfqd, unsigned long key, int gfp_mask)
+cfq_get_queue(struct cfq_data *cfqd, unsigned int key, unsigned short ioprio,
+ int gfp_mask)
{
const int hashval = hash_long(key, CFQ_QHASH_SHIFT);
struct cfq_queue *cfqq, *new_cfqq = NULL;
retry:
- cfqq = __cfq_find_cfq_hash(cfqd, key, hashval);
+ cfqq = __cfq_find_cfq_hash(cfqd, key, ioprio, hashval);
if (!cfqq) {
if (new_cfqq) {
cfqq = new_cfqq;
new_cfqq = NULL;
- } else {
+ } else if (gfp_mask & __GFP_WAIT) {
spin_unlock_irq(cfqd->queue->queue_lock);
new_cfqq = kmem_cache_alloc(cfq_pool, gfp_mask);
spin_lock_irq(cfqd->queue->queue_lock);
-
- if (!new_cfqq && !(gfp_mask & __GFP_WAIT))
- goto out;
-
goto retry;
+ } else {
+ cfqq = kmem_cache_alloc(cfq_pool, gfp_mask);
+ if (!cfqq)
+ goto out;
}
memset(cfqq, 0, sizeof(*cfqq));
@@ -1218,16 +1538,21 @@ retry:
INIT_HLIST_NODE(&cfqq->cfq_hash);
INIT_LIST_HEAD(&cfqq->cfq_list);
RB_CLEAR_ROOT(&cfqq->sort_list);
- INIT_LIST_HEAD(&cfqq->fifo[0]);
- INIT_LIST_HEAD(&cfqq->fifo[1]);
+ INIT_LIST_HEAD(&cfqq->fifo);
cfqq->key = key;
hlist_add_head(&cfqq->cfq_hash, &cfqd->cfq_hash[hashval]);
atomic_set(&cfqq->ref, 0);
cfqq->cfqd = cfqd;
atomic_inc(&cfqd->ref);
- cfqq->key_type = cfqd->key_type;
- cfqq->service_start = ~0UL;
+ cfqq->service_last = 0;
+ /*
+ * set ->slice_left to allow preemption for a new process
+ */
+ cfqq->slice_left = 2 * cfqd->cfq_slice_idle;
+ cfq_mark_cfqq_idle_window(cfqq);
+ cfq_mark_cfqq_prio_changed(cfqq);
+ cfq_init_prio_data(cfqq);
}
if (new_cfqq)
@@ -1239,55 +1564,297 @@ out:
return cfqq;
}
-static void cfq_enqueue(struct cfq_data *cfqd, struct cfq_rq *crq)
+/*
+ * Setup general io context and cfq io context. There can be several cfq
+ * io contexts per general io context, if this process is doing io to more
+ * than one device managed by cfq. Note that caller is holding a reference to
+ * cfqq, so we don't need to worry about it disappearing
+ */
+static struct cfq_io_context *
+cfq_get_io_context(struct cfq_data *cfqd, pid_t pid, int gfp_mask)
{
- crq->is_sync = 0;
- if (rq_data_dir(crq->request) == READ || current->flags & PF_SYNCWRITE)
- crq->is_sync = 1;
+ struct io_context *ioc = NULL;
+ struct cfq_io_context *cic;
+
+ might_sleep_if(gfp_mask & __GFP_WAIT);
+
+ ioc = get_io_context(gfp_mask);
+ if (!ioc)
+ return NULL;
+
+ if ((cic = ioc->cic) == NULL) {
+ cic = cfq_alloc_io_context(cfqd, gfp_mask);
+
+ if (cic == NULL)
+ goto err;
+
+ /*
+ * manually increment generic io_context usage count, it
+ * cannot go away since we are already holding one ref to it
+ */
+ ioc->cic = cic;
+ ioc->set_ioprio = cfq_ioc_set_ioprio;
+ cic->ioc = ioc;
+ cic->key = cfqd;
+ atomic_inc(&cfqd->ref);
+ } else {
+ struct cfq_io_context *__cic;
+
+ /*
+ * the first cic on the list is actually the head itself
+ */
+ if (cic->key == cfqd)
+ goto out;
+
+ /*
+ * cic exists, check if we already are there. linear search
+ * should be ok here, the list will usually not be more than
+ * 1 or a few entries long
+ */
+ list_for_each_entry(__cic, &cic->list, list) {
+ /*
+ * this process is already holding a reference to
+ * this queue, so no need to get one more
+ */
+ if (__cic->key == cfqd) {
+ cic = __cic;
+ goto out;
+ }
+ }
+
+ /*
+ * nope, process doesn't have a cic assoicated with this
+ * cfqq yet. get a new one and add to list
+ */
+ __cic = cfq_alloc_io_context(cfqd, gfp_mask);
+ if (__cic == NULL)
+ goto err;
+
+ __cic->ioc = ioc;
+ __cic->key = cfqd;
+ atomic_inc(&cfqd->ref);
+ list_add(&__cic->list, &cic->list);
+ cic = __cic;
+ }
+
+out:
+ return cic;
+err:
+ put_io_context(ioc);
+ return NULL;
+}
+
+static void
+cfq_update_io_thinktime(struct cfq_data *cfqd, struct cfq_io_context *cic)
+{
+ unsigned long elapsed, ttime;
+
+ /*
+ * if this context already has stuff queued, thinktime is from
+ * last queue not last end
+ */
+#if 0
+ if (time_after(cic->last_end_request, cic->last_queue))
+ elapsed = jiffies - cic->last_end_request;
+ else
+ elapsed = jiffies - cic->last_queue;
+#else
+ elapsed = jiffies - cic->last_end_request;
+#endif
+
+ ttime = min(elapsed, 2UL * cfqd->cfq_slice_idle);
+
+ cic->ttime_samples = (7*cic->ttime_samples + 256) / 8;
+ cic->ttime_total = (7*cic->ttime_total + 256*ttime) / 8;
+ cic->ttime_mean = (cic->ttime_total + 128) / cic->ttime_samples;
+}
+
+#define sample_valid(samples) ((samples) > 80)
+
+/*
+ * Disable idle window if the process thinks too long or seeks so much that
+ * it doesn't matter
+ */
+static void
+cfq_update_idle_window(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+ struct cfq_io_context *cic)
+{
+ int enable_idle = cfq_cfqq_idle_window(cfqq);
+
+ if (!cic->ioc->task || !cfqd->cfq_slice_idle)
+ enable_idle = 0;
+ else if (sample_valid(cic->ttime_samples)) {
+ if (cic->ttime_mean > cfqd->cfq_slice_idle)
+ enable_idle = 0;
+ else
+ enable_idle = 1;
+ }
+
+ if (enable_idle)
+ cfq_mark_cfqq_idle_window(cfqq);
+ else
+ cfq_clear_cfqq_idle_window(cfqq);
+}
+
+
+/*
+ * Check if new_cfqq should preempt the currently active queue. Return 0 for
+ * no or if we aren't sure, a 1 will cause a preempt.
+ */
+static int
+cfq_should_preempt(struct cfq_data *cfqd, struct cfq_queue *new_cfqq,
+ struct cfq_rq *crq)
+{
+ struct cfq_queue *cfqq = cfqd->active_queue;
+
+ if (cfq_class_idle(new_cfqq))
+ return 0;
+
+ if (!cfqq)
+ return 1;
+
+ if (cfq_class_idle(cfqq))
+ return 1;
+ if (!cfq_cfqq_wait_request(new_cfqq))
+ return 0;
+ /*
+ * if it doesn't have slice left, forget it
+ */
+ if (new_cfqq->slice_left < cfqd->cfq_slice_idle)
+ return 0;
+ if (cfq_crq_is_sync(crq) && !cfq_cfqq_sync(cfqq))
+ return 1;
+
+ return 0;
+}
+
+/*
+ * cfqq preempts the active queue. if we allowed preempt with no slice left,
+ * let it have half of its nominal slice.
+ */
+static void cfq_preempt_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+ struct cfq_queue *__cfqq, *next;
+
+ list_for_each_entry_safe(__cfqq, next, &cfqd->cur_rr, cfq_list)
+ cfq_resort_rr_list(__cfqq, 1);
+
+ if (!cfqq->slice_left)
+ cfqq->slice_left = cfq_prio_to_slice(cfqd, cfqq) / 2;
+
+ cfqq->slice_end = cfqq->slice_left + jiffies;
+ __cfq_slice_expired(cfqd, cfqq, 1);
+ __cfq_set_active_queue(cfqd, cfqq);
+}
+
+/*
+ * should really be a ll_rw_blk.c helper
+ */
+static void cfq_start_queueing(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+ request_queue_t *q = cfqd->queue;
+
+ if (!blk_queue_plugged(q))
+ q->request_fn(q);
+ else
+ __generic_unplug_device(q);
+}
+
+/*
+ * Called when a new fs request (crq) is added (to cfqq). Check if there's
+ * something we should do about it
+ */
+static void
+cfq_crq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+ struct cfq_rq *crq)
+{
+ const int sync = cfq_crq_is_sync(crq);
+
+ cfqq->next_crq = cfq_choose_req(cfqd, cfqq->next_crq, crq);
+
+ if (sync) {
+ struct cfq_io_context *cic = crq->io_context;
+
+ cfq_update_io_thinktime(cfqd, cic);
+ cfq_update_idle_window(cfqd, cfqq, cic);
+
+ cic->last_queue = jiffies;
+ }
+
+ if (cfqq == cfqd->active_queue) {
+ /*
+ * if we are waiting for a request for this queue, let it rip
+ * immediately and flag that we must not expire this queue
+ * just now
+ */
+ if (cfq_cfqq_wait_request(cfqq)) {
+ cfq_mark_cfqq_must_dispatch(cfqq);
+ del_timer(&cfqd->idle_slice_timer);
+ cfq_start_queueing(cfqd, cfqq);
+ }
+ } else if (cfq_should_preempt(cfqd, cfqq, crq)) {
+ /*
+ * not the active queue - expire current slice if it is
+ * idle and has expired it's mean thinktime or this new queue
+ * has some old slice time left and is of higher priority
+ */
+ cfq_preempt_queue(cfqd, cfqq);
+ cfq_mark_cfqq_must_dispatch(cfqq);
+ cfq_start_queueing(cfqd, cfqq);
+ }
+}
+
+static void cfq_enqueue(struct cfq_data *cfqd, struct request *rq)
+{
+ struct cfq_rq *crq = RQ_DATA(rq);
+ struct cfq_queue *cfqq = crq->cfq_queue;
+
+ cfq_init_prio_data(cfqq);
cfq_add_crq_rb(crq);
- crq->queue_start = jiffies;
- list_add_tail(&crq->request->queuelist, &crq->cfq_queue->fifo[crq->is_sync]);
+ list_add_tail(&rq->queuelist, &cfqq->fifo);
+
+ if (rq_mergeable(rq)) {
+ cfq_add_crq_hash(cfqd, crq);
+
+ if (!cfqd->queue->last_merge)
+ cfqd->queue->last_merge = rq;
+ }
+
+ cfq_crq_enqueued(cfqd, cfqq, crq);
}
static void
cfq_insert_request(request_queue_t *q, struct request *rq, int where)
{
struct cfq_data *cfqd = q->elevator->elevator_data;
- struct cfq_rq *crq = RQ_DATA(rq);
switch (where) {
case ELEVATOR_INSERT_BACK:
- while (cfq_dispatch_requests(q, cfqd->cfq_quantum))
+ while (cfq_dispatch_requests(q, INT_MAX, 1))
;
list_add_tail(&rq->queuelist, &q->queue_head);
+ /*
+ * If we were idling with pending requests on
+ * inactive cfqqs, force dispatching will
+ * remove the idle timer and the queue won't
+ * be kicked by __make_request() afterward.
+ * Kick it here.
+ */
+ cfq_schedule_dispatch(cfqd);
break;
case ELEVATOR_INSERT_FRONT:
list_add(&rq->queuelist, &q->queue_head);
break;
case ELEVATOR_INSERT_SORT:
BUG_ON(!blk_fs_request(rq));
- cfq_enqueue(cfqd, crq);
+ cfq_enqueue(cfqd, rq);
break;
default:
printk("%s: bad insert point %d\n", __FUNCTION__,where);
return;
}
-
- if (rq_mergeable(rq)) {
- cfq_add_crq_hash(cfqd, crq);
-
- if (!q->last_merge)
- q->last_merge = rq;
- }
-}
-
-static int cfq_queue_empty(request_queue_t *q)
-{
- struct cfq_data *cfqd = q->elevator->elevator_data;
-
- return list_empty(&q->queue_head) && list_empty(&cfqd->rr_list);
}
static void cfq_completed_request(request_queue_t *q, struct request *rq)
@@ -1300,9 +1867,11 @@ static void cfq_completed_request(request_queue_t *q, struct request *rq)
cfqq = crq->cfq_queue;
- if (crq->in_flight) {
- WARN_ON(!cfqq->in_flight);
- cfqq->in_flight--;
+ if (cfq_crq_in_flight(crq)) {
+ const int sync = cfq_crq_is_sync(crq);
+
+ WARN_ON(!cfqq->on_dispatch[sync]);
+ cfqq->on_dispatch[sync]--;
}
cfq_account_completion(cfqq, crq);
@@ -1332,51 +1901,136 @@ cfq_latter_request(request_queue_t *q, struct request *rq)
return NULL;
}
-static int cfq_may_queue(request_queue_t *q, int rw)
+/*
+ * we temporarily boost lower priority queues if they are holding fs exclusive
+ * resources. they are boosted to normal prio (CLASS_BE/4)
+ */
+static void cfq_prio_boost(struct cfq_queue *cfqq)
{
- struct cfq_data *cfqd = q->elevator->elevator_data;
- struct cfq_queue *cfqq;
- int ret = ELV_MQUEUE_MAY;
+ const int ioprio_class = cfqq->ioprio_class;
+ const int ioprio = cfqq->ioprio;
- if (current->flags & PF_MEMALLOC)
- return ELV_MQUEUE_MAY;
-
- cfqq = cfq_find_cfq_hash(cfqd, cfq_hash_key(cfqd, current));
- if (cfqq) {
- int limit = cfqd->max_queued;
-
- if (cfqq->allocated[rw] < cfqd->cfq_queued)
- return ELV_MQUEUE_MUST;
-
- if (cfqd->busy_queues)
- limit = q->nr_requests / cfqd->busy_queues;
-
- if (limit < cfqd->cfq_queued)
- limit = cfqd->cfq_queued;
- else if (limit > cfqd->max_queued)
- limit = cfqd->max_queued;
-
- if (cfqq->allocated[rw] >= limit) {
- if (limit > cfqq->alloc_limit[rw])
- cfqq->alloc_limit[rw] = limit;
-
- ret = ELV_MQUEUE_NO;
- }
+ if (has_fs_excl()) {
+ /*
+ * boost idle prio on transactions that would lock out other
+ * users of the filesystem
+ */
+ if (cfq_class_idle(cfqq))
+ cfqq->ioprio_class = IOPRIO_CLASS_BE;
+ if (cfqq->ioprio > IOPRIO_NORM)
+ cfqq->ioprio = IOPRIO_NORM;
+ } else {
+ /*
+ * check if we need to unboost the queue
+ */
+ if (cfqq->ioprio_class != cfqq->org_ioprio_class)
+ cfqq->ioprio_class = cfqq->org_ioprio_class;
+ if (cfqq->ioprio != cfqq->org_ioprio)
+ cfqq->ioprio = cfqq->org_ioprio;
}
- return ret;
+ /*
+ * refile between round-robin lists if we moved the priority class
+ */
+ if ((ioprio_class != cfqq->ioprio_class || ioprio != cfqq->ioprio) &&
+ cfq_cfqq_on_rr(cfqq))
+ cfq_resort_rr_list(cfqq, 0);
+}
+
+static inline pid_t cfq_queue_pid(struct task_struct *task, int rw)
+{
+ if (rw == READ || process_sync(task))
+ return task->pid;
+
+ return CFQ_KEY_ASYNC;
+}
+
+static inline int
+__cfq_may_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+ struct task_struct *task, int rw)
+{
+#if 1
+ if ((cfq_cfqq_wait_request(cfqq) || cfq_cfqq_must_alloc(cfqq)) &&
+ !cfq_cfqq_must_alloc_slice(cfqq)) {
+ cfq_mark_cfqq_must_alloc_slice(cfqq);
+ return ELV_MQUEUE_MUST;
+ }
+
+ return ELV_MQUEUE_MAY;
+#else
+ if (!cfqq || task->flags & PF_MEMALLOC)
+ return ELV_MQUEUE_MAY;
+ if (!cfqq->allocated[rw] || cfq_cfqq_must_alloc(cfqq)) {
+ if (cfq_cfqq_wait_request(cfqq))
+ return ELV_MQUEUE_MUST;
+
+ /*
+ * only allow 1 ELV_MQUEUE_MUST per slice, otherwise we
+ * can quickly flood the queue with writes from a single task
+ */
+ if (rw == READ || !cfq_cfqq_must_alloc_slice(cfqq)) {
+ cfq_mark_cfqq_must_alloc_slice(cfqq);
+ return ELV_MQUEUE_MUST;
+ }
+
+ return ELV_MQUEUE_MAY;
+ }
+ if (cfq_class_idle(cfqq))
+ return ELV_MQUEUE_NO;
+ if (cfqq->allocated[rw] >= cfqd->max_queued) {
+ struct io_context *ioc = get_io_context(GFP_ATOMIC);
+ int ret = ELV_MQUEUE_NO;
+
+ if (ioc && ioc->nr_batch_requests)
+ ret = ELV_MQUEUE_MAY;
+
+ put_io_context(ioc);
+ return ret;
+ }
+
+ return ELV_MQUEUE_MAY;
+#endif
+}
+
+static int cfq_may_queue(request_queue_t *q, int rw, struct bio *bio)
+{
+ struct cfq_data *cfqd = q->elevator->elevator_data;
+ struct task_struct *tsk = current;
+ struct cfq_queue *cfqq;
+
+ /*
+ * don't force setup of a queue from here, as a call to may_queue
+ * does not necessarily imply that a request actually will be queued.
+ * so just lookup a possibly existing queue, or return 'may queue'
+ * if that fails
+ */
+ cfqq = cfq_find_cfq_hash(cfqd, cfq_queue_pid(tsk, rw), tsk->ioprio);
+ if (cfqq) {
+ cfq_init_prio_data(cfqq);
+ cfq_prio_boost(cfqq);
+
+ return __cfq_may_queue(cfqd, cfqq, tsk, rw);
+ }
+
+ return ELV_MQUEUE_MAY;
}
static void cfq_check_waiters(request_queue_t *q, struct cfq_queue *cfqq)
{
+ struct cfq_data *cfqd = q->elevator->elevator_data;
struct request_list *rl = &q->rq;
- const int write = waitqueue_active(&rl->wait[WRITE]);
- const int read = waitqueue_active(&rl->wait[READ]);
- if (read && cfqq->allocated[READ] < cfqq->alloc_limit[READ])
- wake_up(&rl->wait[READ]);
- if (write && cfqq->allocated[WRITE] < cfqq->alloc_limit[WRITE])
- wake_up(&rl->wait[WRITE]);
+ if (cfqq->allocated[READ] <= cfqd->max_queued || cfqd->rq_starved) {
+ smp_mb();
+ if (waitqueue_active(&rl->wait[READ]))
+ wake_up(&rl->wait[READ]);
+ }
+
+ if (cfqq->allocated[WRITE] <= cfqd->max_queued || cfqd->rq_starved) {
+ smp_mb();
+ if (waitqueue_active(&rl->wait[WRITE]))
+ wake_up(&rl->wait[WRITE]);
+ }
}
/*
@@ -1389,69 +2043,61 @@ static void cfq_put_request(request_queue_t *q, struct request *rq)
if (crq) {
struct cfq_queue *cfqq = crq->cfq_queue;
+ const int rw = rq_data_dir(rq);
- BUG_ON(q->last_merge == rq);
- BUG_ON(!hlist_unhashed(&crq->hash));
+ BUG_ON(!cfqq->allocated[rw]);
+ cfqq->allocated[rw]--;
- if (crq->io_context)
- put_io_context(crq->io_context->ioc);
-
- BUG_ON(!cfqq->allocated[crq->is_write]);
- cfqq->allocated[crq->is_write]--;
+ put_io_context(crq->io_context->ioc);
mempool_free(crq, cfqd->crq_pool);
rq->elevator_private = NULL;
- smp_mb();
cfq_check_waiters(q, cfqq);
cfq_put_queue(cfqq);
}
}
/*
- * Allocate cfq data structures associated with this request. A queue and
+ * Allocate cfq data structures associated with this request.
*/
-static int cfq_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+static int
+cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+ int gfp_mask)
{
struct cfq_data *cfqd = q->elevator->elevator_data;
+ struct task_struct *tsk = current;
struct cfq_io_context *cic;
const int rw = rq_data_dir(rq);
- struct cfq_queue *cfqq, *saved_cfqq;
+ pid_t key = cfq_queue_pid(tsk, rw);
+ struct cfq_queue *cfqq;
struct cfq_rq *crq;
unsigned long flags;
might_sleep_if(gfp_mask & __GFP_WAIT);
+ cic = cfq_get_io_context(cfqd, key, gfp_mask);
+
spin_lock_irqsave(q->queue_lock, flags);
- cfqq = __cfq_get_queue(cfqd, cfq_hash_key(cfqd, current), gfp_mask);
- if (!cfqq)
- goto out_lock;
+ if (!cic)
+ goto queue_fail;
-repeat:
- if (cfqq->allocated[rw] >= cfqd->max_queued)
- goto out_lock;
+ if (!cic->cfqq) {
+ cfqq = cfq_get_queue(cfqd, key, tsk->ioprio, gfp_mask);
+ if (!cfqq)
+ goto queue_fail;
+
+ cic->cfqq = cfqq;
+ } else
+ cfqq = cic->cfqq;
cfqq->allocated[rw]++;
+ cfq_clear_cfqq_must_alloc(cfqq);
+ cfqd->rq_starved = 0;
+ atomic_inc(&cfqq->ref);
spin_unlock_irqrestore(q->queue_lock, flags);
- /*
- * if hashing type has changed, the cfq_queue might change here.
- */
- saved_cfqq = cfqq;
- cic = cfq_get_io_context(&cfqq, gfp_mask);
- if (!cic)
- goto err;
-
- /*
- * repeat allocation checks on queue change
- */
- if (unlikely(saved_cfqq != cfqq)) {
- spin_lock_irqsave(q->queue_lock, flags);
- saved_cfqq->allocated[rw]--;
- goto repeat;
- }
-
crq = mempool_alloc(cfqd->crq_pool, gfp_mask);
if (crq) {
RB_CLEAR(&crq->rb_node);
@@ -1460,24 +2106,141 @@ repeat:
INIT_HLIST_NODE(&crq->hash);
crq->cfq_queue = cfqq;
crq->io_context = cic;
- crq->service_start = crq->queue_start = 0;
- crq->in_flight = crq->accounted = crq->is_sync = 0;
- crq->is_write = rw;
+ cfq_clear_crq_in_flight(crq);
+ cfq_clear_crq_in_driver(crq);
+ cfq_clear_crq_requeued(crq);
+
+ if (rw == READ || process_sync(tsk))
+ cfq_mark_crq_is_sync(crq);
+ else
+ cfq_clear_crq_is_sync(crq);
+
rq->elevator_private = crq;
- cfqq->alloc_limit[rw] = 0;
return 0;
}
- put_io_context(cic->ioc);
-err:
spin_lock_irqsave(q->queue_lock, flags);
cfqq->allocated[rw]--;
+ if (!(cfqq->allocated[0] + cfqq->allocated[1]))
+ cfq_mark_cfqq_must_alloc(cfqq);
cfq_put_queue(cfqq);
-out_lock:
+queue_fail:
+ if (cic)
+ put_io_context(cic->ioc);
+ /*
+ * mark us rq allocation starved. we need to kickstart the process
+ * ourselves if there are no pending requests that can do it for us.
+ * that would be an extremely rare OOM situation
+ */
+ cfqd->rq_starved = 1;
+ cfq_schedule_dispatch(cfqd);
spin_unlock_irqrestore(q->queue_lock, flags);
return 1;
}
+static void cfq_kick_queue(void *data)
+{
+ request_queue_t *q = data;
+ struct cfq_data *cfqd = q->elevator->elevator_data;
+ unsigned long flags;
+
+ spin_lock_irqsave(q->queue_lock, flags);
+
+ if (cfqd->rq_starved) {
+ struct request_list *rl = &q->rq;
+
+ /*
+ * we aren't guaranteed to get a request after this, but we
+ * have to be opportunistic
+ */
+ smp_mb();
+ if (waitqueue_active(&rl->wait[READ]))
+ wake_up(&rl->wait[READ]);
+ if (waitqueue_active(&rl->wait[WRITE]))
+ wake_up(&rl->wait[WRITE]);
+ }
+
+ blk_remove_plug(q);
+ q->request_fn(q);
+ spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
+/*
+ * Timer running if the active_queue is currently idling inside its time slice
+ */
+static void cfq_idle_slice_timer(unsigned long data)
+{
+ struct cfq_data *cfqd = (struct cfq_data *) data;
+ struct cfq_queue *cfqq;
+ unsigned long flags;
+
+ spin_lock_irqsave(cfqd->queue->queue_lock, flags);
+
+ if ((cfqq = cfqd->active_queue) != NULL) {
+ unsigned long now = jiffies;
+
+ /*
+ * expired
+ */
+ if (time_after(now, cfqq->slice_end))
+ goto expire;
+
+ /*
+ * only expire and reinvoke request handler, if there are
+ * other queues with pending requests
+ */
+ if (!cfq_pending_requests(cfqd)) {
+ cfqd->idle_slice_timer.expires = min(now + cfqd->cfq_slice_idle, cfqq->slice_end);
+ add_timer(&cfqd->idle_slice_timer);
+ goto out_cont;
+ }
+
+ /*
+ * not expired and it has a request pending, let it dispatch
+ */
+ if (!RB_EMPTY(&cfqq->sort_list)) {
+ cfq_mark_cfqq_must_dispatch(cfqq);
+ goto out_kick;
+ }
+ }
+expire:
+ cfq_slice_expired(cfqd, 0);
+out_kick:
+ cfq_schedule_dispatch(cfqd);
+out_cont:
+ spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
+}
+
+/*
+ * Timer running if an idle class queue is waiting for service
+ */
+static void cfq_idle_class_timer(unsigned long data)
+{
+ struct cfq_data *cfqd = (struct cfq_data *) data;
+ unsigned long flags, end;
+
+ spin_lock_irqsave(cfqd->queue->queue_lock, flags);
+
+ /*
+ * race with a non-idle queue, reset timer
+ */
+ end = cfqd->last_end_request + CFQ_IDLE_GRACE;
+ if (!time_after_eq(jiffies, end)) {
+ cfqd->idle_class_timer.expires = end;
+ add_timer(&cfqd->idle_class_timer);
+ } else
+ cfq_schedule_dispatch(cfqd);
+
+ spin_unlock_irqrestore(cfqd->queue->queue_lock, flags);
+}
+
+static void cfq_shutdown_timer_wq(struct cfq_data *cfqd)
+{
+ del_timer_sync(&cfqd->idle_slice_timer);
+ del_timer_sync(&cfqd->idle_class_timer);
+ blk_sync_queue(cfqd->queue);
+}
+
static void cfq_put_cfqd(struct cfq_data *cfqd)
{
request_queue_t *q = cfqd->queue;
@@ -1487,6 +2250,9 @@ static void cfq_put_cfqd(struct cfq_data *cfqd)
blk_put_queue(q);
+ cfq_shutdown_timer_wq(cfqd);
+ q->elevator->elevator_data = NULL;
+
mempool_destroy(cfqd->crq_pool);
kfree(cfqd->crq_hash);
kfree(cfqd->cfq_hash);
@@ -1495,7 +2261,10 @@ static void cfq_put_cfqd(struct cfq_data *cfqd)
static void cfq_exit_queue(elevator_t *e)
{
- cfq_put_cfqd(e->elevator_data);
+ struct cfq_data *cfqd = e->elevator_data;
+
+ cfq_shutdown_timer_wq(cfqd);
+ cfq_put_cfqd(cfqd);
}
static int cfq_init_queue(request_queue_t *q, elevator_t *e)
@@ -1508,7 +2277,13 @@ static int cfq_init_queue(request_queue_t *q, elevator_t *e)
return -ENOMEM;
memset(cfqd, 0, sizeof(*cfqd));
- INIT_LIST_HEAD(&cfqd->rr_list);
+
+ for (i = 0; i < CFQ_PRIO_LISTS; i++)
+ INIT_LIST_HEAD(&cfqd->rr_list[i]);
+
+ INIT_LIST_HEAD(&cfqd->busy_rr);
+ INIT_LIST_HEAD(&cfqd->cur_rr);
+ INIT_LIST_HEAD(&cfqd->idle_rr);
INIT_LIST_HEAD(&cfqd->empty_list);
cfqd->crq_hash = kmalloc(sizeof(struct hlist_head) * CFQ_MHASH_ENTRIES, GFP_KERNEL);
@@ -1533,24 +2308,32 @@ static int cfq_init_queue(request_queue_t *q, elevator_t *e)
cfqd->queue = q;
atomic_inc(&q->refcnt);
- /*
- * just set it to some high value, we want anyone to be able to queue
- * some requests. fairness is handled differently
- */
- q->nr_requests = 1024;
- cfqd->max_queued = q->nr_requests / 16;
+ cfqd->max_queued = q->nr_requests / 4;
q->nr_batching = cfq_queued;
- cfqd->key_type = CFQ_KEY_TGID;
- cfqd->find_best_crq = 1;
+
+ init_timer(&cfqd->idle_slice_timer);
+ cfqd->idle_slice_timer.function = cfq_idle_slice_timer;
+ cfqd->idle_slice_timer.data = (unsigned long) cfqd;
+
+ init_timer(&cfqd->idle_class_timer);
+ cfqd->idle_class_timer.function = cfq_idle_class_timer;
+ cfqd->idle_class_timer.data = (unsigned long) cfqd;
+
+ INIT_WORK(&cfqd->unplug_work, cfq_kick_queue, q);
+
atomic_set(&cfqd->ref, 1);
cfqd->cfq_queued = cfq_queued;
cfqd->cfq_quantum = cfq_quantum;
- cfqd->cfq_fifo_expire_r = cfq_fifo_expire_r;
- cfqd->cfq_fifo_expire_w = cfq_fifo_expire_w;
- cfqd->cfq_fifo_batch_expire = cfq_fifo_rate;
+ cfqd->cfq_fifo_expire[0] = cfq_fifo_expire[0];
+ cfqd->cfq_fifo_expire[1] = cfq_fifo_expire[1];
cfqd->cfq_back_max = cfq_back_max;
cfqd->cfq_back_penalty = cfq_back_penalty;
+ cfqd->cfq_slice[0] = cfq_slice_async;
+ cfqd->cfq_slice[1] = cfq_slice_sync;
+ cfqd->cfq_slice_async_rq = cfq_slice_async_rq;
+ cfqd->cfq_slice_idle = cfq_slice_idle;
+ cfqd->cfq_max_depth = cfq_max_depth;
return 0;
out_crqpool:
@@ -1595,7 +2378,6 @@ fail:
return -ENOMEM;
}
-
/*
* sysfs parts below -->
*/
@@ -1620,45 +2402,6 @@ cfq_var_store(unsigned int *var, const char *page, size_t count)
return count;
}
-static ssize_t
-cfq_clear_elapsed(struct cfq_data *cfqd, const char *page, size_t count)
-{
- max_elapsed_dispatch = max_elapsed_crq = 0;
- return count;
-}
-
-static ssize_t
-cfq_set_key_type(struct cfq_data *cfqd, const char *page, size_t count)
-{
- spin_lock_irq(cfqd->queue->queue_lock);
- if (!strncmp(page, "pgid", 4))
- cfqd->key_type = CFQ_KEY_PGID;
- else if (!strncmp(page, "tgid", 4))
- cfqd->key_type = CFQ_KEY_TGID;
- else if (!strncmp(page, "uid", 3))
- cfqd->key_type = CFQ_KEY_UID;
- else if (!strncmp(page, "gid", 3))
- cfqd->key_type = CFQ_KEY_GID;
- spin_unlock_irq(cfqd->queue->queue_lock);
- return count;
-}
-
-static ssize_t
-cfq_read_key_type(struct cfq_data *cfqd, char *page)
-{
- ssize_t len = 0;
- int i;
-
- for (i = CFQ_KEY_PGID; i < CFQ_KEY_LAST; i++) {
- if (cfqd->key_type == i)
- len += sprintf(page+len, "[%s] ", cfq_key_types[i]);
- else
- len += sprintf(page+len, "%s ", cfq_key_types[i]);
- }
- len += sprintf(page+len, "\n");
- return len;
-}
-
#define SHOW_FUNCTION(__FUNC, __VAR, __CONV) \
static ssize_t __FUNC(struct cfq_data *cfqd, char *page) \
{ \
@@ -1669,12 +2412,15 @@ static ssize_t __FUNC(struct cfq_data *cfqd, char *page) \
}
SHOW_FUNCTION(cfq_quantum_show, cfqd->cfq_quantum, 0);
SHOW_FUNCTION(cfq_queued_show, cfqd->cfq_queued, 0);
-SHOW_FUNCTION(cfq_fifo_expire_r_show, cfqd->cfq_fifo_expire_r, 1);
-SHOW_FUNCTION(cfq_fifo_expire_w_show, cfqd->cfq_fifo_expire_w, 1);
-SHOW_FUNCTION(cfq_fifo_batch_expire_show, cfqd->cfq_fifo_batch_expire, 1);
-SHOW_FUNCTION(cfq_find_best_show, cfqd->find_best_crq, 0);
+SHOW_FUNCTION(cfq_fifo_expire_sync_show, cfqd->cfq_fifo_expire[1], 1);
+SHOW_FUNCTION(cfq_fifo_expire_async_show, cfqd->cfq_fifo_expire[0], 1);
SHOW_FUNCTION(cfq_back_max_show, cfqd->cfq_back_max, 0);
SHOW_FUNCTION(cfq_back_penalty_show, cfqd->cfq_back_penalty, 0);
+SHOW_FUNCTION(cfq_slice_idle_show, cfqd->cfq_slice_idle, 1);
+SHOW_FUNCTION(cfq_slice_sync_show, cfqd->cfq_slice[1], 1);
+SHOW_FUNCTION(cfq_slice_async_show, cfqd->cfq_slice[0], 1);
+SHOW_FUNCTION(cfq_slice_async_rq_show, cfqd->cfq_slice_async_rq, 0);
+SHOW_FUNCTION(cfq_max_depth_show, cfqd->cfq_max_depth, 0);
#undef SHOW_FUNCTION
#define STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, __CONV) \
@@ -1694,12 +2440,15 @@ static ssize_t __FUNC(struct cfq_data *cfqd, const char *page, size_t count) \
}
STORE_FUNCTION(cfq_quantum_store, &cfqd->cfq_quantum, 1, UINT_MAX, 0);
STORE_FUNCTION(cfq_queued_store, &cfqd->cfq_queued, 1, UINT_MAX, 0);
-STORE_FUNCTION(cfq_fifo_expire_r_store, &cfqd->cfq_fifo_expire_r, 1, UINT_MAX, 1);
-STORE_FUNCTION(cfq_fifo_expire_w_store, &cfqd->cfq_fifo_expire_w, 1, UINT_MAX, 1);
-STORE_FUNCTION(cfq_fifo_batch_expire_store, &cfqd->cfq_fifo_batch_expire, 0, UINT_MAX, 1);
-STORE_FUNCTION(cfq_find_best_store, &cfqd->find_best_crq, 0, 1, 0);
+STORE_FUNCTION(cfq_fifo_expire_sync_store, &cfqd->cfq_fifo_expire[1], 1, UINT_MAX, 1);
+STORE_FUNCTION(cfq_fifo_expire_async_store, &cfqd->cfq_fifo_expire[0], 1, UINT_MAX, 1);
STORE_FUNCTION(cfq_back_max_store, &cfqd->cfq_back_max, 0, UINT_MAX, 0);
STORE_FUNCTION(cfq_back_penalty_store, &cfqd->cfq_back_penalty, 1, UINT_MAX, 0);
+STORE_FUNCTION(cfq_slice_idle_store, &cfqd->cfq_slice_idle, 0, UINT_MAX, 1);
+STORE_FUNCTION(cfq_slice_sync_store, &cfqd->cfq_slice[1], 1, UINT_MAX, 1);
+STORE_FUNCTION(cfq_slice_async_store, &cfqd->cfq_slice[0], 1, UINT_MAX, 1);
+STORE_FUNCTION(cfq_slice_async_rq_store, &cfqd->cfq_slice_async_rq, 1, UINT_MAX, 0);
+STORE_FUNCTION(cfq_max_depth_store, &cfqd->cfq_max_depth, 1, UINT_MAX, 0);
#undef STORE_FUNCTION
static struct cfq_fs_entry cfq_quantum_entry = {
@@ -1712,25 +2461,15 @@ static struct cfq_fs_entry cfq_queued_entry = {
.show = cfq_queued_show,
.store = cfq_queued_store,
};
-static struct cfq_fs_entry cfq_fifo_expire_r_entry = {
+static struct cfq_fs_entry cfq_fifo_expire_sync_entry = {
.attr = {.name = "fifo_expire_sync", .mode = S_IRUGO | S_IWUSR },
- .show = cfq_fifo_expire_r_show,
- .store = cfq_fifo_expire_r_store,
+ .show = cfq_fifo_expire_sync_show,
+ .store = cfq_fifo_expire_sync_store,
};
-static struct cfq_fs_entry cfq_fifo_expire_w_entry = {
+static struct cfq_fs_entry cfq_fifo_expire_async_entry = {
.attr = {.name = "fifo_expire_async", .mode = S_IRUGO | S_IWUSR },
- .show = cfq_fifo_expire_w_show,
- .store = cfq_fifo_expire_w_store,
-};
-static struct cfq_fs_entry cfq_fifo_batch_expire_entry = {
- .attr = {.name = "fifo_batch_expire", .mode = S_IRUGO | S_IWUSR },
- .show = cfq_fifo_batch_expire_show,
- .store = cfq_fifo_batch_expire_store,
-};
-static struct cfq_fs_entry cfq_find_best_entry = {
- .attr = {.name = "find_best_crq", .mode = S_IRUGO | S_IWUSR },
- .show = cfq_find_best_show,
- .store = cfq_find_best_store,
+ .show = cfq_fifo_expire_async_show,
+ .store = cfq_fifo_expire_async_store,
};
static struct cfq_fs_entry cfq_back_max_entry = {
.attr = {.name = "back_seek_max", .mode = S_IRUGO | S_IWUSR },
@@ -1742,27 +2481,44 @@ static struct cfq_fs_entry cfq_back_penalty_entry = {
.show = cfq_back_penalty_show,
.store = cfq_back_penalty_store,
};
-static struct cfq_fs_entry cfq_clear_elapsed_entry = {
- .attr = {.name = "clear_elapsed", .mode = S_IWUSR },
- .store = cfq_clear_elapsed,
+static struct cfq_fs_entry cfq_slice_sync_entry = {
+ .attr = {.name = "slice_sync", .mode = S_IRUGO | S_IWUSR },
+ .show = cfq_slice_sync_show,
+ .store = cfq_slice_sync_store,
};
-static struct cfq_fs_entry cfq_key_type_entry = {
- .attr = {.name = "key_type", .mode = S_IRUGO | S_IWUSR },
- .show = cfq_read_key_type,
- .store = cfq_set_key_type,
+static struct cfq_fs_entry cfq_slice_async_entry = {
+ .attr = {.name = "slice_async", .mode = S_IRUGO | S_IWUSR },
+ .show = cfq_slice_async_show,
+ .store = cfq_slice_async_store,
+};
+static struct cfq_fs_entry cfq_slice_async_rq_entry = {
+ .attr = {.name = "slice_async_rq", .mode = S_IRUGO | S_IWUSR },
+ .show = cfq_slice_async_rq_show,
+ .store = cfq_slice_async_rq_store,
+};
+static struct cfq_fs_entry cfq_slice_idle_entry = {
+ .attr = {.name = "slice_idle", .mode = S_IRUGO | S_IWUSR },
+ .show = cfq_slice_idle_show,
+ .store = cfq_slice_idle_store,
+};
+static struct cfq_fs_entry cfq_max_depth_entry = {
+ .attr = {.name = "max_depth", .mode = S_IRUGO | S_IWUSR },
+ .show = cfq_max_depth_show,
+ .store = cfq_max_depth_store,
};
static struct attribute *default_attrs[] = {
&cfq_quantum_entry.attr,
&cfq_queued_entry.attr,
- &cfq_fifo_expire_r_entry.attr,
- &cfq_fifo_expire_w_entry.attr,
- &cfq_fifo_batch_expire_entry.attr,
- &cfq_key_type_entry.attr,
- &cfq_find_best_entry.attr,
+ &cfq_fifo_expire_sync_entry.attr,
+ &cfq_fifo_expire_async_entry.attr,
&cfq_back_max_entry.attr,
&cfq_back_penalty_entry.attr,
- &cfq_clear_elapsed_entry.attr,
+ &cfq_slice_sync_entry.attr,
+ &cfq_slice_async_entry.attr,
+ &cfq_slice_async_rq_entry.attr,
+ &cfq_slice_idle_entry.attr,
+ &cfq_max_depth_entry.attr,
NULL,
};
@@ -1832,21 +2588,46 @@ static int __init cfq_init(void)
{
int ret;
+ /*
+ * could be 0 on HZ < 1000 setups
+ */
+ if (!cfq_slice_async)
+ cfq_slice_async = 1;
+ if (!cfq_slice_idle)
+ cfq_slice_idle = 1;
+
if (cfq_slab_setup())
return -ENOMEM;
ret = elv_register(&iosched_cfq);
- if (!ret) {
- __module_get(THIS_MODULE);
- return 0;
- }
+ if (ret)
+ cfq_slab_kill();
- cfq_slab_kill();
return ret;
}
static void __exit cfq_exit(void)
{
+ struct task_struct *g, *p;
+ unsigned long flags;
+
+ read_lock_irqsave(&tasklist_lock, flags);
+
+ /*
+ * iterate each process in the system, removing our io_context
+ */
+ do_each_thread(g, p) {
+ struct io_context *ioc = p->io_context;
+
+ if (ioc && ioc->cic) {
+ ioc->cic->exit(ioc->cic);
+ cfq_free_io_context(ioc->cic);
+ ioc->cic = NULL;
+ }
+ } while_each_thread(g, p);
+
+ read_unlock_irqrestore(&tasklist_lock, flags);
+
cfq_slab_kill();
elv_unregister(&iosched_cfq);
}
diff --git a/drivers/block/deadline-iosched.c b/drivers/block/deadline-iosched.c
index 4bc2fea73273..ff5201e02153 100644
--- a/drivers/block/deadline-iosched.c
+++ b/drivers/block/deadline-iosched.c
@@ -760,7 +760,8 @@ static void deadline_put_request(request_queue_t *q, struct request *rq)
}
static int
-deadline_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+deadline_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+ int gfp_mask)
{
struct deadline_data *dd = q->elevator->elevator_data;
struct deadline_rq *drq;
diff --git a/drivers/block/elevator.c b/drivers/block/elevator.c
index f831f08f839c..98f0126a2deb 100644
--- a/drivers/block/elevator.c
+++ b/drivers/block/elevator.c
@@ -486,12 +486,13 @@ struct request *elv_former_request(request_queue_t *q, struct request *rq)
return NULL;
}
-int elv_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+int elv_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+ int gfp_mask)
{
elevator_t *e = q->elevator;
if (e->ops->elevator_set_req_fn)
- return e->ops->elevator_set_req_fn(q, rq, gfp_mask);
+ return e->ops->elevator_set_req_fn(q, rq, bio, gfp_mask);
rq->elevator_private = NULL;
return 0;
@@ -505,12 +506,12 @@ void elv_put_request(request_queue_t *q, struct request *rq)
e->ops->elevator_put_req_fn(q, rq);
}
-int elv_may_queue(request_queue_t *q, int rw)
+int elv_may_queue(request_queue_t *q, int rw, struct bio *bio)
{
elevator_t *e = q->elevator;
if (e->ops->elevator_may_queue_fn)
- return e->ops->elevator_may_queue_fn(q, rw);
+ return e->ops->elevator_may_queue_fn(q, rw, bio);
return ELV_MQUEUE_MAY;
}
diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c
index 60e64091de1b..692a5fced76e 100644
--- a/drivers/block/ll_rw_blk.c
+++ b/drivers/block/ll_rw_blk.c
@@ -276,6 +276,7 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
rq->errors = 0;
rq->rq_status = RQ_ACTIVE;
rq->bio = rq->biotail = NULL;
+ rq->ioprio = 0;
rq->buffer = NULL;
rq->ref_count = 1;
rq->q = q;
@@ -1442,11 +1443,7 @@ void __generic_unplug_device(request_queue_t *q)
if (!blk_remove_plug(q))
return;
- /*
- * was plugged, fire request_fn if queue has stuff to do
- */
- if (elv_next_request(q))
- q->request_fn(q);
+ q->request_fn(q);
}
EXPORT_SYMBOL(__generic_unplug_device);
@@ -1776,8 +1773,8 @@ static inline void blk_free_request(request_queue_t *q, struct request *rq)
mempool_free(rq, q->rq.rq_pool);
}
-static inline struct request *blk_alloc_request(request_queue_t *q, int rw,
- int gfp_mask)
+static inline struct request *
+blk_alloc_request(request_queue_t *q, int rw, struct bio *bio, int gfp_mask)
{
struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask);
@@ -1790,7 +1787,7 @@ static inline struct request *blk_alloc_request(request_queue_t *q, int rw,
*/
rq->flags = rw;
- if (!elv_set_request(q, rq, gfp_mask))
+ if (!elv_set_request(q, rq, bio, gfp_mask))
return rq;
mempool_free(rq, q->rq.rq_pool);
@@ -1870,18 +1867,20 @@ static void freed_request(request_queue_t *q, int rw)
#define blkdev_free_rq(list) list_entry((list)->next, struct request, queuelist)
/*
- * Get a free request, queue_lock must not be held
+ * Get a free request, queue_lock must be held.
+ * Returns NULL on failure, with queue_lock held.
+ * Returns !NULL on success, with queue_lock *not held*.
*/
-static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
+static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
+ int gfp_mask)
{
struct request *rq = NULL;
struct request_list *rl = &q->rq;
- struct io_context *ioc = get_io_context(gfp_mask);
+ struct io_context *ioc = current_io_context(GFP_ATOMIC);
if (unlikely(test_bit(QUEUE_FLAG_DRAIN, &q->queue_flags)))
goto out;
- spin_lock_irq(q->queue_lock);
if (rl->count[rw]+1 >= q->nr_requests) {
/*
* The queue will fill after this allocation, so set it as
@@ -1895,7 +1894,7 @@ static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
}
}
- switch (elv_may_queue(q, rw)) {
+ switch (elv_may_queue(q, rw, bio)) {
case ELV_MQUEUE_NO:
goto rq_starved;
case ELV_MQUEUE_MAY:
@@ -1909,18 +1908,25 @@ static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
* The queue is full and the allocating process is not a
* "batcher", and not exempted by the IO scheduler
*/
- spin_unlock_irq(q->queue_lock);
goto out;
}
get_rq:
+ /*
+ * Only allow batching queuers to allocate up to 50% over the defined
+ * limit of requests, otherwise we could have thousands of requests
+ * allocated with any setting of ->nr_requests
+ */
+ if (rl->count[rw] >= (3 * q->nr_requests / 2))
+ goto out;
+
rl->count[rw]++;
rl->starved[rw] = 0;
if (rl->count[rw] >= queue_congestion_on_threshold(q))
set_queue_congested(q, rw);
spin_unlock_irq(q->queue_lock);
- rq = blk_alloc_request(q, rw, gfp_mask);
+ rq = blk_alloc_request(q, rw, bio, gfp_mask);
if (!rq) {
/*
* Allocation failed presumably due to memory. Undo anything
@@ -1943,7 +1949,6 @@ rq_starved:
if (unlikely(rl->count[rw] == 0))
rl->starved[rw] = 1;
- spin_unlock_irq(q->queue_lock);
goto out;
}
@@ -1953,31 +1958,35 @@ rq_starved:
rq_init(q, rq);
rq->rl = rl;
out:
- put_io_context(ioc);
return rq;
}
/*
* No available requests for this queue, unplug the device and wait for some
* requests to become available.
+ *
+ * Called with q->queue_lock held, and returns with it unlocked.
*/
-static struct request *get_request_wait(request_queue_t *q, int rw)
+static struct request *get_request_wait(request_queue_t *q, int rw,
+ struct bio *bio)
{
- DEFINE_WAIT(wait);
struct request *rq;
- do {
+ rq = get_request(q, rw, bio, GFP_NOIO);
+ while (!rq) {
+ DEFINE_WAIT(wait);
struct request_list *rl = &q->rq;
prepare_to_wait_exclusive(&rl->wait[rw], &wait,
TASK_UNINTERRUPTIBLE);
- rq = get_request(q, rw, GFP_NOIO);
+ rq = get_request(q, rw, bio, GFP_NOIO);
if (!rq) {
struct io_context *ioc;
- generic_unplug_device(q);
+ __generic_unplug_device(q);
+ spin_unlock_irq(q->queue_lock);
io_schedule();
/*
@@ -1986,12 +1995,13 @@ static struct request *get_request_wait(request_queue_t *q, int rw)
* up to a big batch of them for a small period time.
* See ioc_batching, ioc_set_batching
*/
- ioc = get_io_context(GFP_NOIO);
+ ioc = current_io_context(GFP_NOIO);
ioc_set_batching(q, ioc);
- put_io_context(ioc);
+
+ spin_lock_irq(q->queue_lock);
}
finish_wait(&rl->wait[rw], &wait);
- } while (!rq);
+ }
return rq;
}
@@ -2002,14 +2012,18 @@ struct request *blk_get_request(request_queue_t *q, int rw, int gfp_mask)
BUG_ON(rw != READ && rw != WRITE);
- if (gfp_mask & __GFP_WAIT)
- rq = get_request_wait(q, rw);
- else
- rq = get_request(q, rw, gfp_mask);
+ spin_lock_irq(q->queue_lock);
+ if (gfp_mask & __GFP_WAIT) {
+ rq = get_request_wait(q, rw, NULL);
+ } else {
+ rq = get_request(q, rw, NULL, gfp_mask);
+ if (!rq)
+ spin_unlock_irq(q->queue_lock);
+ }
+ /* q->queue_lock is unlocked at this point */
return rq;
}
-
EXPORT_SYMBOL(blk_get_request);
/**
@@ -2333,7 +2347,6 @@ static void __blk_put_request(request_queue_t *q, struct request *req)
return;
req->rq_status = RQ_INACTIVE;
- req->q = NULL;
req->rl = NULL;
/*
@@ -2462,6 +2475,8 @@ static int attempt_merge(request_queue_t *q, struct request *req,
req->rq_disk->in_flight--;
}
+ req->ioprio = ioprio_best(req->ioprio, next->ioprio);
+
__blk_put_request(q, next);
return 1;
}
@@ -2512,13 +2527,15 @@ EXPORT_SYMBOL(blk_attempt_remerge);
static int __make_request(request_queue_t *q, struct bio *bio)
{
- struct request *req, *freereq = NULL;
+ struct request *req;
int el_ret, rw, nr_sectors, cur_nr_sectors, barrier, err, sync;
+ unsigned short prio;
sector_t sector;
sector = bio->bi_sector;
nr_sectors = bio_sectors(bio);
cur_nr_sectors = bio_cur_sectors(bio);
+ prio = bio_prio(bio);
rw = bio_data_dir(bio);
sync = bio_sync(bio);
@@ -2538,14 +2555,9 @@ static int __make_request(request_queue_t *q, struct bio *bio)
goto end_io;
}
-again:
spin_lock_irq(q->queue_lock);
- if (elv_queue_empty(q)) {
- blk_plug_device(q);
- goto get_rq;
- }
- if (barrier)
+ if (unlikely(barrier) || elv_queue_empty(q))
goto get_rq;
el_ret = elv_merge(q, &req, bio);
@@ -2559,6 +2571,7 @@ again:
req->biotail->bi_next = bio;
req->biotail = bio;
req->nr_sectors = req->hard_nr_sectors += nr_sectors;
+ req->ioprio = ioprio_best(req->ioprio, prio);
drive_stat_acct(req, nr_sectors, 0);
if (!attempt_back_merge(q, req))
elv_merged_request(q, req);
@@ -2583,45 +2596,30 @@ again:
req->hard_cur_sectors = cur_nr_sectors;
req->sector = req->hard_sector = sector;
req->nr_sectors = req->hard_nr_sectors += nr_sectors;
+ req->ioprio = ioprio_best(req->ioprio, prio);
drive_stat_acct(req, nr_sectors, 0);
if (!attempt_front_merge(q, req))
elv_merged_request(q, req);
goto out;
- /*
- * elevator says don't/can't merge. get new request
- */
- case ELEVATOR_NO_MERGE:
- break;
-
+ /* ELV_NO_MERGE: elevator says don't/can't merge. */
default:
- printk("elevator returned crap (%d)\n", el_ret);
- BUG();
+ ;
}
+get_rq:
+ /*
+ * Grab a free request. This is might sleep but can not fail.
+ * Returns with the queue unlocked.
+ */
+ req = get_request_wait(q, rw, bio);
+
/*
- * Grab a free request from the freelist - if that is empty, check
- * if we are doing read ahead and abort instead of blocking for
- * a free slot.
+ * After dropping the lock and possibly sleeping here, our request
+ * may now be mergeable after it had proven unmergeable (above).
+ * We don't worry about that case for efficiency. It won't happen
+ * often, and the elevators are able to handle it.
*/
-get_rq:
- if (freereq) {
- req = freereq;
- freereq = NULL;
- } else {
- spin_unlock_irq(q->queue_lock);
- if ((freereq = get_request(q, rw, GFP_ATOMIC)) == NULL) {
- /*
- * READA bit set
- */
- err = -EWOULDBLOCK;
- if (bio_rw_ahead(bio))
- goto end_io;
-
- freereq = get_request_wait(q, rw);
- }
- goto again;
- }
req->flags |= REQ_CMD;
@@ -2646,13 +2644,15 @@ get_rq:
req->buffer = bio_data(bio); /* see ->buffer comment above */
req->waiting = NULL;
req->bio = req->biotail = bio;
+ req->ioprio = prio;
req->rq_disk = bio->bi_bdev->bd_disk;
req->start_time = jiffies;
+ spin_lock_irq(q->queue_lock);
+ if (elv_queue_empty(q))
+ blk_plug_device(q);
add_request(q, req);
out:
- if (freereq)
- __blk_put_request(q, freereq);
if (sync)
__generic_unplug_device(q);
@@ -2674,7 +2674,7 @@ static inline void blk_partition_remap(struct bio *bio)
if (bdev != bdev->bd_contains) {
struct hd_struct *p = bdev->bd_part;
- switch (bio->bi_rw) {
+ switch (bio_data_dir(bio)) {
case READ:
p->read_sectors += bio_sectors(bio);
p->reads++;
@@ -2693,6 +2693,7 @@ void blk_finish_queue_drain(request_queue_t *q)
{
struct request_list *rl = &q->rq;
struct request *rq;
+ int requeued = 0;
spin_lock_irq(q->queue_lock);
clear_bit(QUEUE_FLAG_DRAIN, &q->queue_flags);
@@ -2701,9 +2702,13 @@ void blk_finish_queue_drain(request_queue_t *q)
rq = list_entry_rq(q->drain_list.next);
list_del_init(&rq->queuelist);
- __elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 1);
+ elv_requeue_request(q, rq);
+ requeued++;
}
+ if (requeued)
+ q->request_fn(q);
+
spin_unlock_irq(q->queue_lock);
wake_up(&rl->wait[0]);
@@ -2900,7 +2905,7 @@ void submit_bio(int rw, struct bio *bio)
BIO_BUG_ON(!bio->bi_size);
BIO_BUG_ON(!bio->bi_io_vec);
- bio->bi_rw = rw;
+ bio->bi_rw |= rw;
if (rw & WRITE)
mod_page_state(pgpgout, count);
else
@@ -3257,8 +3262,11 @@ void exit_io_context(void)
struct io_context *ioc;
local_irq_save(flags);
+ task_lock(current);
ioc = current->io_context;
current->io_context = NULL;
+ ioc->task = NULL;
+ task_unlock(current);
local_irq_restore(flags);
if (ioc->aic && ioc->aic->exit)
@@ -3271,55 +3279,51 @@ void exit_io_context(void)
/*
* If the current task has no IO context then create one and initialise it.
- * If it does have a context, take a ref on it.
+ * Otherwise, return its existing IO context.
*
- * This is always called in the context of the task which submitted the I/O.
- * But weird things happen, so we disable local interrupts to ensure exclusive
- * access to *current.
+ * This returned IO context doesn't have a specifically elevated refcount,
+ * but since the current task itself holds a reference, the context can be
+ * used in general code, so long as it stays within `current` context.
*/
-struct io_context *get_io_context(int gfp_flags)
+struct io_context *current_io_context(int gfp_flags)
{
struct task_struct *tsk = current;
- unsigned long flags;
struct io_context *ret;
- local_irq_save(flags);
ret = tsk->io_context;
- if (ret)
- goto out;
-
- local_irq_restore(flags);
+ if (likely(ret))
+ return ret;
ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
if (ret) {
atomic_set(&ret->refcount, 1);
- ret->pid = tsk->pid;
+ ret->task = current;
+ ret->set_ioprio = NULL;
ret->last_waited = jiffies; /* doesn't matter... */
ret->nr_batch_requests = 0; /* because this is 0 */
ret->aic = NULL;
ret->cic = NULL;
- spin_lock_init(&ret->lock);
-
- local_irq_save(flags);
-
- /*
- * very unlikely, someone raced with us in setting up the task
- * io context. free new context and just grab a reference.
- */
- if (!tsk->io_context)
- tsk->io_context = ret;
- else {
- kmem_cache_free(iocontext_cachep, ret);
- ret = tsk->io_context;
- }
-
-out:
- atomic_inc(&ret->refcount);
- local_irq_restore(flags);
+ tsk->io_context = ret;
}
return ret;
}
+EXPORT_SYMBOL(current_io_context);
+
+/*
+ * If the current task has no IO context then create one and initialise it.
+ * If it does have a context, take a ref on it.
+ *
+ * This is always called in the context of the task which submitted the I/O.
+ */
+struct io_context *get_io_context(int gfp_flags)
+{
+ struct io_context *ret;
+ ret = current_io_context(gfp_flags);
+ if (likely(ret))
+ atomic_inc(&ret->refcount);
+ return ret;
+}
EXPORT_SYMBOL(get_io_context);
void copy_io_context(struct io_context **pdst, struct io_context **psrc)
diff --git a/drivers/block/swim3.c b/drivers/block/swim3.c
index 5b09cf154ac7..e5f7494c00ee 100644
--- a/drivers/block/swim3.c
+++ b/drivers/block/swim3.c
@@ -253,7 +253,7 @@ static int floppy_revalidate(struct gendisk *disk);
static int swim3_add_device(struct device_node *swims);
int swim3_init(void);
-#ifndef CONFIG_PMAC_PBOOK
+#ifndef CONFIG_PMAC_MEDIABAY
#define check_media_bay(which, what) 1
#endif
@@ -297,9 +297,11 @@ static void do_fd_request(request_queue_t * q)
int i;
for(i=0;imedia_bay && check_media_bay(fs->media_bay, MB_FD))
return -ENXIO;
+#endif
switch (cmd) {
case FDEJECT:
@@ -881,8 +885,10 @@ static int floppy_open(struct inode *inode, struct file *filp)
int n, err = 0;
if (fs->ref_count == 0) {
+#ifdef CONFIG_PMAC_MEDIABAY
if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
return -ENXIO;
+#endif
out_8(&sw->setup, S_IBM_DRIVE | S_FCLK_DIV2);
out_8(&sw->control_bic, 0xff);
out_8(&sw->mode, 0x95);
@@ -967,8 +973,10 @@ static int floppy_revalidate(struct gendisk *disk)
struct swim3 __iomem *sw;
int ret, n;
+#ifdef CONFIG_PMAC_MEDIABAY
if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
return -ENXIO;
+#endif
sw = fs->swim3;
grab_drive(fs, revalidating, 0);
diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c
index 5ed3a6379452..9db0a9e3e59c 100644
--- a/drivers/block/sx8.c
+++ b/drivers/block/sx8.c
@@ -26,6 +26,7 @@
#include
#include
#include
+#include
#include
#include
#include
@@ -1582,9 +1583,9 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
goto err_out;
#if IF_64BIT_DMA_IS_POSSIBLE /* grrrr... */
- rc = pci_set_dma_mask(pdev, 0xffffffffffffffffULL);
+ rc = pci_set_dma_mask(pdev, DMA_64BIT_MASK);
if (!rc) {
- rc = pci_set_consistent_dma_mask(pdev, 0xffffffffffffffffULL);
+ rc = pci_set_consistent_dma_mask(pdev, DMA_64BIT_MASK);
if (rc) {
printk(KERN_ERR DRV_NAME "(%s): consistent DMA mask failure\n",
pci_name(pdev));
@@ -1593,7 +1594,7 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
pci_dac = 1;
} else {
#endif
- rc = pci_set_dma_mask(pdev, 0xffffffffULL);
+ rc = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
if (rc) {
printk(KERN_ERR DRV_NAME "(%s): DMA mask failure\n",
pci_name(pdev));
diff --git a/drivers/bluetooth/bluecard_cs.c b/drivers/bluetooth/bluecard_cs.c
index e481cc411b5d..5ef9adb9fe73 100644
--- a/drivers/bluetooth/bluecard_cs.c
+++ b/drivers/bluetooth/bluecard_cs.c
@@ -1089,6 +1089,14 @@ static int bluecard_event(event_t event, int priority, event_callback_args_t *ar
return 0;
}
+static struct pcmcia_device_id bluecard_ids[] = {
+ PCMCIA_DEVICE_PROD_ID12("BlueCard", "LSE041", 0xbaf16fbf, 0x657cc15e),
+ PCMCIA_DEVICE_PROD_ID12("BTCFCARD", "LSE139", 0xe3987764, 0x2524b59c),
+ PCMCIA_DEVICE_PROD_ID12("WSS", "LSE039", 0x0a0736ec, 0x24e6dfab),
+ PCMCIA_DEVICE_NULL
+};
+MODULE_DEVICE_TABLE(pcmcia, bluecard_ids);
+
static struct pcmcia_driver bluecard_driver = {
.owner = THIS_MODULE,
.drv = {
@@ -1096,6 +1104,7 @@ static struct pcmcia_driver bluecard_driver = {
},
.attach = bluecard_attach,
.detach = bluecard_detach,
+ .id_table = bluecard_ids,
};
static int __init init_bluecard_cs(void)
diff --git a/drivers/bluetooth/bt3c_cs.c b/drivers/bluetooth/bt3c_cs.c
index f71e5c76963d..9013cd759afb 100644
--- a/drivers/bluetooth/bt3c_cs.c
+++ b/drivers/bluetooth/bt3c_cs.c
@@ -935,6 +935,12 @@ static int bt3c_event(event_t event, int priority, event_callback_args_t *args)
return 0;
}
+static struct pcmcia_device_id bt3c_ids[] = {
+ PCMCIA_DEVICE_PROD_ID13("3COM", "Bluetooth PC Card", 0xefce0a31, 0xd4ce9b02),
+ PCMCIA_DEVICE_NULL
+};
+MODULE_DEVICE_TABLE(pcmcia, bt3c_ids);
+
static struct pcmcia_driver bt3c_driver = {
.owner = THIS_MODULE,
.drv = {
@@ -942,6 +948,7 @@ static struct pcmcia_driver bt3c_driver = {
},
.attach = bt3c_attach,
.detach = bt3c_detach,
+ .id_table = bt3c_ids,
};
static int __init init_bt3c_cs(void)
diff --git a/drivers/bluetooth/btuart_cs.c b/drivers/bluetooth/btuart_cs.c
index ad8d972444a5..c479484a1f7f 100644
--- a/drivers/bluetooth/btuart_cs.c
+++ b/drivers/bluetooth/btuart_cs.c
@@ -855,6 +855,12 @@ static int btuart_event(event_t event, int priority, event_callback_args_t *args
return 0;
}
+static struct pcmcia_device_id btuart_ids[] = {
+ /* don't use this driver. Use serial_cs + hci_uart instead */
+ PCMCIA_DEVICE_NULL
+};
+MODULE_DEVICE_TABLE(pcmcia, btuart_ids);
+
static struct pcmcia_driver btuart_driver = {
.owner = THIS_MODULE,
.drv = {
@@ -862,6 +868,7 @@ static struct pcmcia_driver btuart_driver = {
},
.attach = btuart_attach,
.detach = btuart_detach,
+ .id_table = btuart_ids,
};
static int __init init_btuart_cs(void)
diff --git a/drivers/bluetooth/dtl1_cs.c b/drivers/bluetooth/dtl1_cs.c
index fe954e5d9a1d..bb12f7daeb91 100644
--- a/drivers/bluetooth/dtl1_cs.c
+++ b/drivers/bluetooth/dtl1_cs.c
@@ -807,6 +807,13 @@ static int dtl1_event(event_t event, int priority, event_callback_args_t *args)
return 0;
}
+static struct pcmcia_device_id dtl1_ids[] = {
+ PCMCIA_DEVICE_PROD_ID12("Nokia Mobile Phones", "DTL-1", 0xe1bfdd64, 0xe168480d),
+ PCMCIA_DEVICE_PROD_ID12("Socket", "CF", 0xb38bcc2e, 0x44ebf863),
+ PCMCIA_DEVICE_NULL
+};
+MODULE_DEVICE_TABLE(pcmcia, dtl1_ids);
+
static struct pcmcia_driver dtl1_driver = {
.owner = THIS_MODULE,
.drv = {
@@ -814,6 +821,7 @@ static struct pcmcia_driver dtl1_driver = {
},
.attach = dtl1_attach,
.detach = dtl1_detach,
+ .id_table = dtl1_ids,
};
static int __init init_dtl1_cs(void)
diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c
index 1407945a5892..59f589d733f9 100644
--- a/drivers/char/agp/amd64-agp.c
+++ b/drivers/char/agp/amd64-agp.c
@@ -686,6 +686,15 @@ static struct pci_device_id agp_amd64_pci_table[] = {
.subvendor = PCI_ANY_ID,
.subdevice = PCI_ANY_ID,
},
+ /* SIS 760 */
+ {
+ .class = (PCI_CLASS_BRIDGE_HOST << 8),
+ .class_mask = ~0,
+ .vendor = PCI_VENDOR_ID_SI,
+ .device = PCI_DEVICE_ID_SI_760,
+ .subvendor = PCI_ANY_ID,
+ .subdevice = PCI_ANY_ID,
+ },
{ }
};
diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c
index 1813d0d198f1..e16c13fe698d 100644
--- a/drivers/char/ipmi/ipmi_msghandler.c
+++ b/drivers/char/ipmi/ipmi_msghandler.c
@@ -1088,8 +1088,8 @@ static inline int i_ipmi_request(ipmi_user_t user,
long seqid;
int broadcast = 0;
- if (addr->channel > IPMI_NUM_CHANNELS) {
- spin_lock_irqsave(&intf->counter_lock, flags);
+ if (addr->channel >= IPMI_MAX_CHANNELS) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
intf->sent_invalid_commands++;
spin_unlock_irqrestore(&intf->counter_lock, flags);
rv = -EINVAL;
diff --git a/drivers/char/misc.c b/drivers/char/misc.c
index 31cf84d69026..931efd58f87a 100644
--- a/drivers/char/misc.c
+++ b/drivers/char/misc.c
@@ -308,9 +308,6 @@ static int __init misc_init(void)
#endif
#ifdef CONFIG_BVME6000
rtc_DP8570A_init();
-#endif
-#ifdef CONFIG_PMAC_PBOOK
- pmu_device_init();
#endif
if (register_chrdev(MISC_MAJOR,"misc",&misc_fops)) {
printk("unable to get major %d for misc devices\n",
diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c
index 7c24fbe831f8..95f7046ff059 100644
--- a/drivers/char/moxa.c
+++ b/drivers/char/moxa.c
@@ -451,7 +451,7 @@ static int __init moxa_init(void)
int n = (sizeof(moxa_pcibrds) / sizeof(moxa_pcibrds[0])) - 1;
i = 0;
while (i < n) {
- while ((p = pci_find_device(moxa_pcibrds[i].vendor, moxa_pcibrds[i].device, p))!=NULL)
+ while ((p = pci_get_device(moxa_pcibrds[i].vendor, moxa_pcibrds[i].device, p))!=NULL)
{
if (pci_enable_device(p))
continue;
diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 1c8d866a49dc..8f36b1758eb6 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -581,7 +581,7 @@ static dev_link_t *mgslpc_attach(void)
/* Interrupt setup */
link->irq.Attributes = IRQ_TYPE_EXCLUSIVE;
- link->irq.IRQInfo1 = IRQ_INFO2_VALID | IRQ_LEVEL_ID;
+ link->irq.IRQInfo1 = IRQ_LEVEL_ID;
link->irq.Handler = NULL;
link->conf.Attributes = 0;
@@ -3081,6 +3081,12 @@ void mgslpc_remove_device(MGSLPC_INFO *remove_info)
}
}
+static struct pcmcia_device_id mgslpc_ids[] = {
+ PCMCIA_DEVICE_MANF_CARD(0x02c5, 0x0050),
+ PCMCIA_DEVICE_NULL
+};
+MODULE_DEVICE_TABLE(pcmcia, mgslpc_ids);
+
static struct pcmcia_driver mgslpc_driver = {
.owner = THIS_MODULE,
.drv = {
@@ -3088,6 +3094,7 @@ static struct pcmcia_driver mgslpc_driver = {
},
.attach = mgslpc_attach,
.detach = mgslpc_detach,
+ .id_table = mgslpc_ids,
};
static struct tty_operations mgslpc_ops = {
diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c
index 7db3370f4972..d7d484024e2b 100644
--- a/drivers/char/rio/rio_linux.c
+++ b/drivers/char/rio/rio_linux.c
@@ -1095,7 +1095,7 @@ static int __init rio_init(void)
#ifdef CONFIG_PCI
/* First look for the JET devices: */
- while ((pdev = pci_find_device (PCI_VENDOR_ID_SPECIALIX,
+ while ((pdev = pci_get_device (PCI_VENDOR_ID_SPECIALIX,
PCI_DEVICE_ID_SPECIALIX_SX_XIO_IO8,
pdev))) {
if (pci_enable_device(pdev)) continue;
@@ -1169,7 +1169,7 @@ static int __init rio_init(void)
*/
/* Then look for the older RIO/PCI devices: */
- while ((pdev = pci_find_device (PCI_VENDOR_ID_SPECIALIX,
+ while ((pdev = pci_get_device (PCI_VENDOR_ID_SPECIALIX,
PCI_DEVICE_ID_SPECIALIX_RIO,
pdev))) {
if (pci_enable_device(pdev)) continue;
diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c
index ff4f09804865..d8f9e94ae475 100644
--- a/drivers/char/rtc.c
+++ b/drivers/char/rtc.c
@@ -78,6 +78,7 @@
#include
#include
#include
+#include
#include
#include
@@ -894,7 +895,6 @@ static int __init rtc_init(void)
struct proc_dir_entry *ent;
#if defined(__alpha__) || defined(__mips__)
unsigned int year, ctrl;
- unsigned long uip_watchdog;
char *guess = NULL;
#endif
#ifdef __sparc__
@@ -1000,12 +1000,8 @@ no_irq:
/* Each operating system on an Alpha uses its own epoch.
Let's try to guess which one we are using now. */
- uip_watchdog = jiffies;
if (rtc_is_updating() != 0)
- while (jiffies - uip_watchdog < 2*HZ/100) {
- barrier();
- cpu_relax();
- }
+ msleep(20);
spin_lock_irq(&rtc_lock);
year = CMOS_READ(RTC_YEAR);
@@ -1213,7 +1209,6 @@ static int rtc_proc_open(struct inode *inode, struct file *file)
void rtc_get_rtc_time(struct rtc_time *rtc_tm)
{
- unsigned long uip_watchdog = jiffies;
unsigned char ctrl;
#ifdef CONFIG_MACH_DECSTATION
unsigned int real_year;
@@ -1221,7 +1216,7 @@ void rtc_get_rtc_time(struct rtc_time *rtc_tm)
/*
* read RTC once any update in progress is done. The update
- * can take just over 2ms. We wait 10 to 20ms. There is no need to
+ * can take just over 2ms. We wait 20ms. There is no need to
* to poll-wait (up to 1s - eeccch) for the falling edge of RTC_UIP.
* If you need to know *exactly* when a second has started, enable
* periodic update complete interrupts, (via ioctl) and then
@@ -1230,10 +1225,7 @@ void rtc_get_rtc_time(struct rtc_time *rtc_tm)
*/
if (rtc_is_updating() != 0)
- while (jiffies - uip_watchdog < 2*HZ/100) {
- barrier();
- cpu_relax();
- }
+ msleep(20);
/*
* Only the values that we read from the RTC are set. We leave
diff --git a/drivers/char/tipar.c b/drivers/char/tipar.c
index 659335d80ee7..ec78d2f161f7 100644
--- a/drivers/char/tipar.c
+++ b/drivers/char/tipar.c
@@ -396,7 +396,7 @@ static struct file_operations tipar_fops = {
static int __init
tipar_setup(char *str)
{
- int ints[2];
+ int ints[3];
str = get_options(str, ARRAY_SIZE(ints), ints);
diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c
index 58597993954f..f19cf9d7792d 100644
--- a/drivers/char/tty_ioctl.c
+++ b/drivers/char/tty_ioctl.c
@@ -476,11 +476,11 @@ int n_tty_ioctl(struct tty_struct * tty, struct file * file,
ld = tty_ldisc_ref(tty);
switch (arg) {
case TCIFLUSH:
- if (ld->flush_buffer)
+ if (ld && ld->flush_buffer)
ld->flush_buffer(tty);
break;
case TCIOFLUSH:
- if (ld->flush_buffer)
+ if (ld && ld->flush_buffer)
ld->flush_buffer(tty);
/* fall through */
case TCOFLUSH:
diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c
index 8971484b956b..1d44f69e1fda 100644
--- a/drivers/char/vt_ioctl.c
+++ b/drivers/char/vt_ioctl.c
@@ -25,6 +25,7 @@
#include
#include
#include
+#include
#include
#include
@@ -386,7 +387,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file,
if (!perm)
return -EPERM;
if (arg)
- arg = 1193182 / arg;
+ arg = CLOCK_TICK_RATE / arg;
kd_mksound(arg, 0);
return 0;
@@ -403,7 +404,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file,
ticks = HZ * ((arg >> 16) & 0xffff) / 1000;
count = ticks ? (arg & 0xffff) : 0;
if (count)
- count = 1193182 / count;
+ count = CLOCK_TICK_RATE / count;
kd_mksound(count, ticks);
return 0;
}
diff --git a/drivers/char/watchdog/ixp2000_wdt.c b/drivers/char/watchdog/ixp2000_wdt.c
index 4e98c215e5b1..4b039516cc86 100644
--- a/drivers/char/watchdog/ixp2000_wdt.c
+++ b/drivers/char/watchdog/ixp2000_wdt.c
@@ -162,7 +162,7 @@ ixp2000_wdt_release(struct inode *inode, struct file *file)
if (test_bit(WDT_OK_TO_CLOSE, &wdt_status)) {
wdt_disable();
} else {
- printk(KERN_CRIT "WATCHDOG: Device closed unexpectdly - "
+ printk(KERN_CRIT "WATCHDOG: Device closed unexpectedly - "
"timer will not stop\n");
}
diff --git a/drivers/char/watchdog/ixp4xx_wdt.c b/drivers/char/watchdog/ixp4xx_wdt.c
index 82396e06c8a8..83df369113a4 100644
--- a/drivers/char/watchdog/ixp4xx_wdt.c
+++ b/drivers/char/watchdog/ixp4xx_wdt.c
@@ -156,7 +156,7 @@ ixp4xx_wdt_release(struct inode *inode, struct file *file)
if (test_bit(WDT_OK_TO_CLOSE, &wdt_status)) {
wdt_disable();
} else {
- printk(KERN_CRIT "WATCHDOG: Device closed unexpectdly - "
+ printk(KERN_CRIT "WATCHDOG: Device closed unexpectedly - "
"timer will not stop\n");
}
diff --git a/drivers/firmware/pcdp.c b/drivers/firmware/pcdp.c
index 839b44a7e08b..53c95c0bbf46 100644
--- a/drivers/firmware/pcdp.c
+++ b/drivers/firmware/pcdp.c
@@ -16,6 +16,7 @@
#include
#include
#include
+#include
#include "pcdp.h"
static int __init
@@ -40,10 +41,27 @@ setup_serial_console(struct pcdp_uart *uart)
}
static int __init
-setup_vga_console(struct pcdp_vga *vga)
+setup_vga_console(struct pcdp_device *dev)
{
#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
- if (efi_mem_type(0xA0000) == EFI_CONVENTIONAL_MEMORY) {
+ u8 *if_ptr;
+
+ if_ptr = ((u8 *)dev + sizeof(struct pcdp_device));
+ if (if_ptr[0] == PCDP_IF_PCI) {
+ struct pcdp_if_pci if_pci;
+
+ /* struct copy since ifptr might not be correctly aligned */
+
+ memcpy(&if_pci, if_ptr, sizeof(if_pci));
+
+ if (if_pci.trans & PCDP_PCI_TRANS_IOPORT)
+ vga_console_iobase = if_pci.ioport_tra;
+
+ if (if_pci.trans & PCDP_PCI_TRANS_MMIO)
+ vga_console_membase = if_pci.mmio_tra;
+ }
+
+ if (efi_mem_type(vga_console_membase + 0xA0000) == EFI_CONVENTIONAL_MEMORY) {
printk(KERN_ERR "PCDP: VGA selected, but frame buffer is not MMIO!\n");
return -ENODEV;
}
@@ -95,7 +113,7 @@ efi_setup_pcdp_console(char *cmdline)
dev = (struct pcdp_device *) ((u8 *) dev + dev->length)) {
if (dev->flags & PCDP_PRIMARY_CONSOLE) {
if (dev->type == PCDP_CONSOLE_VGA) {
- return setup_vga_console((struct pcdp_vga *) dev);
+ return setup_vga_console(dev);
}
}
}
diff --git a/drivers/firmware/pcdp.h b/drivers/firmware/pcdp.h
index 1dc7c88b7b4d..e72cc47de33b 100644
--- a/drivers/firmware/pcdp.h
+++ b/drivers/firmware/pcdp.h
@@ -52,11 +52,34 @@ struct pcdp_uart {
u32 clock_rate;
u8 pci_prog_intfc;
u8 flags;
-};
+} __attribute__((packed));
+
+#define PCDP_IF_PCI 1
+
+/* pcdp_if_pci.trans */
+#define PCDP_PCI_TRANS_IOPORT 0x02
+#define PCDP_PCI_TRANS_MMIO 0x01
+
+struct pcdp_if_pci {
+ u8 interconnect;
+ u8 reserved;
+ u16 length;
+ u8 segment;
+ u8 bus;
+ u8 dev;
+ u8 fun;
+ u16 dev_id;
+ u16 vendor_id;
+ u32 acpi_interrupt;
+ u64 mmio_tra;
+ u64 ioport_tra;
+ u8 flags;
+ u8 trans;
+} __attribute__((packed));
struct pcdp_vga {
u8 count; /* address space descriptors */
-};
+} __attribute__((packed));
/* pcdp_device.flags */
#define PCDP_PRIMARY_CONSOLE 1
@@ -66,7 +89,9 @@ struct pcdp_device {
u8 flags;
u16 length;
u16 efi_index;
-};
+ /* next data is pcdp_if_pci or pcdp_if_acpi (not yet supported) */
+ /* next data is device specific type (currently only pcdp_vga) */
+} __attribute__((packed));
struct pcdp {
u8 signature[4];
@@ -81,4 +106,4 @@ struct pcdp {
u32 num_uarts;
struct pcdp_uart uart[0]; /* actual size is num_uarts */
/* remainder of table is pcdp_device structures */
-};
+} __attribute__((packed));
diff --git a/drivers/i2c/chips/atxp1.c b/drivers/i2c/chips/atxp1.c
index 5c6597aa2c7f..0bcf82b4c07b 100644
--- a/drivers/i2c/chips/atxp1.c
+++ b/drivers/i2c/chips/atxp1.c
@@ -144,7 +144,7 @@ static ssize_t atxp1_storevcore(struct device *dev, struct device_attribute *att
if (vid == cvid)
return count;
- dev_info(dev, "Setting VCore to %d mV (0x%02x)\n", vcore, vid);
+ dev_dbg(dev, "Setting VCore to %d mV (0x%02x)\n", vcore, vid);
/* Write every 25 mV step to increase stability */
if (cvid > vid) {
diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 0273f124a4f7..5f33df47aa74 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -606,6 +606,12 @@ config BLK_DEV_IT8172
; picture of the
board at .
+config BLK_DEV_IT821X
+ tristate "IT821X IDE support"
+ help
+ This driver adds support for the ITE 8211 IDE controller and the
+ IT 8212 IDE RAID controller in both RAID and pass-through mode.
+
config BLK_DEV_NS87415
tristate "NS87415 chipset support"
help
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index d6f934886b04..f9c1acb4ed6a 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -119,6 +119,10 @@ static int lba_capacity_is_ok (struct hd_driveid *id)
{
unsigned long lba_sects, chs_sects, head, tail;
+ /* No non-LBA info .. so valid! */
+ if (id->cyls == 0)
+ return 1;
+
/*
* The ATA spec tells large drives to return
* C/H/S = 16383/16/63 independent of their size.
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 2d2eefb610dd..1e1531334c25 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -132,7 +132,6 @@ static const struct drive_list_entry drive_blacklist [] = {
{ "SAMSUNG CD-ROM SC-148C", "ALL" },
{ "SAMSUNG CD-ROM SC", "ALL" },
{ "SanDisk SDP3B-64" , "ALL" },
- { "SAMSUNG CD-ROM SN-124", "ALL" },
{ "ATAPI CD-ROM DRIVE 40X MAXIMUM", "ALL" },
{ "_NEC DV5800A", "ALL" },
{ NULL , NULL }
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 53024942a7eb..b443b04a4c5a 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -1181,7 +1181,8 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
pre_reset(drive);
SELECT_DRIVE(drive);
udelay (20);
- hwif->OUTB(WIN_SRST, IDE_COMMAND_REG);
+ hwif->OUTBSYNC(drive, WIN_SRST, IDE_COMMAND_REG);
+ ndelay(400);
hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
hwgroup->polling = 1;
__ide_set_handler(drive, &atapi_reset_pollfunc, HZ/20, NULL);
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index e20327e54b1a..978d27d6452d 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -457,6 +457,40 @@ int ide_event(event_t event, int priority,
return 0;
} /* ide_event */
+static struct pcmcia_device_id ide_ids[] = {
+ PCMCIA_DEVICE_FUNC_ID(4),
+ PCMCIA_DEVICE_MANF_CARD(0x0032, 0x0704),
+ PCMCIA_DEVICE_MANF_CARD(0x00a4, 0x002d),
+ PCMCIA_DEVICE_MANF_CARD(0x2080, 0x0001),
+ PCMCIA_DEVICE_MANF_CARD(0x0045, 0x0401),
+ PCMCIA_DEVICE_PROD_ID123("Caravelle", "PSC-IDE ", "PSC000", 0x8c36137c, 0xd0693ab8, 0x2768a9f0),
+ PCMCIA_DEVICE_PROD_ID123("CDROM", "IDE", "MCD-601p", 0x1b9179ca, 0xede88951, 0x0d902f74),
+ PCMCIA_DEVICE_PROD_ID123("PCMCIA", "IDE CARD", "F1", 0x281f1c5d, 0x1907960c, 0xf7fde8b9),
+ PCMCIA_DEVICE_PROD_ID12("ARGOSY", "CD-ROM", 0x78f308dc, 0x66536591),
+ PCMCIA_DEVICE_PROD_ID12("ARGOSY", "PnPIDE", 0x78f308dc, 0x0c694728),
+ PCMCIA_DEVICE_PROD_ID12("CNF CD-M", "CD-ROM", 0x7d93b852, 0x66536591),
+ PCMCIA_DEVICE_PROD_ID12("Creative Technology Ltd.", "PCMCIA CD-ROM Interface Card", 0xff8c8a45, 0xfe8020c4),
+ PCMCIA_DEVICE_PROD_ID12("Digital Equipment Corporation.", "Digital Mobile Media CD-ROM", 0x17692a66, 0xef1dcbde),
+ PCMCIA_DEVICE_PROD_ID12("EXP", "CD", 0x6f58c983, 0xaae5994f),
+ PCMCIA_DEVICE_PROD_ID12("EXP ", "CD-ROM", 0x0a5c52fd, 0x66536591),
+ PCMCIA_DEVICE_PROD_ID12("EXP ", "PnPIDE", 0x0a5c52fd, 0x0c694728),
+ PCMCIA_DEVICE_PROD_ID12("FREECOM", "PCCARD-IDE", 0x5714cbf7, 0x48e0ab8e),
+ PCMCIA_DEVICE_PROD_ID12("IBM", "IBM17JSSFP20", 0xb569a6e5, 0xf2508753),
+ PCMCIA_DEVICE_PROD_ID12("IO DATA", "CBIDE2 ", 0x547e66dc, 0x8671043b),
+ PCMCIA_DEVICE_PROD_ID12("IO DATA", "PCIDE", 0x547e66dc, 0x5c5ab149),
+ PCMCIA_DEVICE_PROD_ID12("IO DATA", "PCIDEII", 0x547e66dc, 0xb3662674),
+ PCMCIA_DEVICE_PROD_ID12("LOOKMEET", "CBIDE2 ", 0xe37be2b5, 0x8671043b),
+ PCMCIA_DEVICE_PROD_ID12(" ", "NinjaATA-", 0x3b6e20c8, 0xebe0bd79),
+ PCMCIA_DEVICE_PROD_ID12("PCMCIA", "CD-ROM", 0x281f1c5d, 0x66536591),
+ PCMCIA_DEVICE_PROD_ID12("PCMCIA", "PnPIDE", 0x281f1c5d, 0x0c694728),
+ PCMCIA_DEVICE_PROD_ID12("SHUTTLE TECHNOLOGY LTD.", "PCCARD-IDE/ATAPI Adapter", 0x4a3f0ba0, 0x322560e1),
+ PCMCIA_DEVICE_PROD_ID12("TOSHIBA", "MK2001MPL", 0xb4585a1a, 0x3489e003),
+ PCMCIA_DEVICE_PROD_ID12("WIT", "IDE16", 0x244e5994, 0x3e232852),
+ PCMCIA_DEVICE_PROD_ID1("STI Flash", 0xe4a13209),
+ PCMCIA_DEVICE_NULL,
+};
+MODULE_DEVICE_TABLE(pcmcia, ide_ids);
+
static struct pcmcia_driver ide_cs_driver = {
.owner = THIS_MODULE,
.drv = {
@@ -464,6 +498,7 @@ static struct pcmcia_driver ide_cs_driver = {
},
.attach = ide_attach,
.detach = ide_detach,
+ .id_table = ide_ids,
};
static int __init init_ide_cs(void)
diff --git a/drivers/ide/pci/Makefile b/drivers/ide/pci/Makefile
index 55e6e553e497..af46226c1796 100644
--- a/drivers/ide/pci/Makefile
+++ b/drivers/ide/pci/Makefile
@@ -12,6 +12,7 @@ obj-$(CONFIG_BLK_DEV_HPT34X) += hpt34x.o
obj-$(CONFIG_BLK_DEV_HPT366) += hpt366.o
#obj-$(CONFIG_BLK_DEV_HPT37X) += hpt37x.o
obj-$(CONFIG_BLK_DEV_IT8172) += it8172.o
+obj-$(CONFIG_BLK_DEV_IT821X) += it821x.o
obj-$(CONFIG_BLK_DEV_NS87415) += ns87415.o
obj-$(CONFIG_BLK_DEV_OPTI621) += opti621.o
obj-$(CONFIG_BLK_DEV_PDC202XX_OLD) += pdc202xx_old.o
diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c
index 4565cc311ff3..da46577380f3 100644
--- a/drivers/ide/pci/generic.c
+++ b/drivers/ide/pci/generic.c
@@ -39,6 +39,17 @@
#include
+static int ide_generic_all; /* Set to claim all devices */
+
+static int __init ide_generic_all_on(char *unused)
+{
+ ide_generic_all = 1;
+ printk(KERN_INFO "IDE generic will claim all unknown PCI IDE storage controllers.\n");
+ return 1;
+}
+
+__setup("all-generic-ide", ide_generic_all_on);
+
static void __devinit init_hwif_generic (ide_hwif_t *hwif)
{
switch(hwif->pci_dev->device) {
@@ -78,79 +89,85 @@ static void __devinit init_hwif_generic (ide_hwif_t *hwif)
static ide_pci_device_t generic_chipsets[] __devinitdata = {
{ /* 0 */
+ .name = "Unknown",
+ .init_hwif = init_hwif_generic,
+ .channels = 2,
+ .autodma = AUTODMA,
+ .bootable = ON_BOARD,
+ },{ /* 1 */
.name = "NS87410",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = AUTODMA,
.enablebits = {{0x43,0x08,0x08}, {0x47,0x08,0x08}},
.bootable = ON_BOARD,
- },{ /* 1 */
+ },{ /* 2 */
.name = "SAMURAI",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = AUTODMA,
.bootable = ON_BOARD,
- },{ /* 2 */
+ },{ /* 3 */
.name = "HT6565",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = AUTODMA,
.bootable = ON_BOARD,
- },{ /* 3 */
+ },{ /* 4 */
.name = "UM8673F",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = NODMA,
.bootable = ON_BOARD,
- },{ /* 4 */
+ },{ /* 5 */
.name = "UM8886A",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = NODMA,
.bootable = ON_BOARD,
- },{ /* 5 */
+ },{ /* 6 */
.name = "UM8886BF",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = NODMA,
.bootable = ON_BOARD,
- },{ /* 6 */
+ },{ /* 7 */
.name = "HINT_IDE",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = AUTODMA,
.bootable = ON_BOARD,
- },{ /* 7 */
+ },{ /* 8 */
.name = "VIA_IDE",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = NOAUTODMA,
.bootable = ON_BOARD,
- },{ /* 8 */
+ },{ /* 9 */
.name = "OPTI621V",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = NOAUTODMA,
.bootable = ON_BOARD,
- },{ /* 9 */
+ },{ /* 10 */
.name = "VIA8237SATA",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = AUTODMA,
.bootable = OFF_BOARD,
- },{ /* 10 */
+ },{ /* 11 */
.name = "Piccolo0102",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = NOAUTODMA,
.bootable = ON_BOARD,
- },{ /* 11 */
+ },{ /* 12 */
.name = "Piccolo0103",
.init_hwif = init_hwif_generic,
.channels = 2,
.autodma = NOAUTODMA,
.bootable = ON_BOARD,
- },{ /* 12 */
+ },{ /* 13 */
.name = "Piccolo0105",
.init_hwif = init_hwif_generic,
.channels = 2,
@@ -174,6 +191,10 @@ static int __devinit generic_init_one(struct pci_dev *dev, const struct pci_devi
u16 command;
int ret = -ENODEV;
+ /* Don't use the generic entry unless instructed to do so */
+ if (id->driver_data == 0 && ide_generic_all == 0)
+ goto out;
+
if (dev->vendor == PCI_VENDOR_ID_UMC &&
dev->device == PCI_DEVICE_ID_UMC_UM8886A &&
(!(PCI_FUNC(dev->devfn) & 1)))
@@ -195,21 +216,23 @@ out:
}
static struct pci_device_id generic_pci_tbl[] = {
- { PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_87410, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
- { PCI_VENDOR_ID_PCTECH, PCI_DEVICE_ID_PCTECH_SAMURAI_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1},
- { PCI_VENDOR_ID_HOLTEK, PCI_DEVICE_ID_HOLTEK_6565, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2},
- { PCI_VENDOR_ID_UMC, PCI_DEVICE_ID_UMC_UM8673F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 3},
- { PCI_VENDOR_ID_UMC, PCI_DEVICE_ID_UMC_UM8886A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 4},
- { PCI_VENDOR_ID_UMC, PCI_DEVICE_ID_UMC_UM8886BF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 5},
- { PCI_VENDOR_ID_HINT, PCI_DEVICE_ID_HINT_VXPROII_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 6},
- { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C561, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 7},
- { PCI_VENDOR_ID_OPTI, PCI_DEVICE_ID_OPTI_82C558, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8},
+ { PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_87410, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1},
+ { PCI_VENDOR_ID_PCTECH, PCI_DEVICE_ID_PCTECH_SAMURAI_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2},
+ { PCI_VENDOR_ID_HOLTEK, PCI_DEVICE_ID_HOLTEK_6565, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 3},
+ { PCI_VENDOR_ID_UMC, PCI_DEVICE_ID_UMC_UM8673F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 4},
+ { PCI_VENDOR_ID_UMC, PCI_DEVICE_ID_UMC_UM8886A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 5},
+ { PCI_VENDOR_ID_UMC, PCI_DEVICE_ID_UMC_UM8886BF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 6},
+ { PCI_VENDOR_ID_HINT, PCI_DEVICE_ID_HINT_VXPROII_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 7},
+ { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C561, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8},
+ { PCI_VENDOR_ID_OPTI, PCI_DEVICE_ID_OPTI_82C558, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 9},
#ifdef CONFIG_BLK_DEV_IDE_SATA
- { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237_SATA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 9},
+ { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237_SATA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 10},
#endif
- { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 10},
- { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 11},
- { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 12},
+ { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 11},
+ { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 12},
+ { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 13},
+ /* Must come last. If you add entries adjust this table appropriately and the init_one code */
+ { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xFFFFFF00UL, 0},
{ 0, },
};
MODULE_DEVICE_TABLE(pci, generic_pci_tbl);
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index c8ee0b8c0292..7b64db10d1b0 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -10,6 +10,11 @@
* donation of an ABit BP6 mainboard, processor, and memory acellerated
* development and support.
*
+ *
+ * Highpoint have their own driver (source except for the raid part)
+ * available from http://www.highpoint-tech.com/hpt3xx-opensource-v131.tgz
+ * This may be useful to anyone wanting to work on the mainstream hpt IDE.
+ *
* Note that final HPT370 support was done by force extraction of GPL.
*
* - add function for getting/setting power status of drive
@@ -446,44 +451,29 @@ static struct chipset_bus_clock_list_entry sixty_six_base_hpt374[] = {
#define F_LOW_PCI_50 0x2d
#define F_LOW_PCI_66 0x42
-/* FIXME: compare with driver's code before removing */
-#if 0
- if (hpt_minimum_revision(dev, 3)) {
- u8 cbl;
- cbl = inb(iobase + 0x7b);
- outb(cbl | 1, iobase + 0x7b);
- outb(cbl & ~1, iobase + 0x7b);
- cbl = inb(iobase + 0x7a);
- p += sprintf(p, "Cable: ATA-%d"
- " ATA-%d\n",
- (cbl & 0x02) ? 33 : 66,
- (cbl & 0x01) ? 33 : 66);
- p += sprintf(p, "\n");
- }
- {
- u8 c2, c3;
- /* older revs don't have these registers mapped
- * into io space */
- pci_read_config_byte(dev, 0x43, &c0);
- pci_read_config_byte(dev, 0x47, &c1);
- pci_read_config_byte(dev, 0x4b, &c2);
- pci_read_config_byte(dev, 0x4f, &c3);
+/*
+ * Hold all the highpoint quirks and revision information in one
+ * place.
+ */
- p += sprintf(p, "Mode: %s %s"
- " %s %s\n",
- (c0 & 0x10) ? "UDMA" : (c0 & 0x20) ? "DMA " :
- (c0 & 0x80) ? "PIO " : "off ",
- (c1 & 0x10) ? "UDMA" : (c1 & 0x20) ? "DMA " :
- (c1 & 0x80) ? "PIO " : "off ",
- (c2 & 0x10) ? "UDMA" : (c2 & 0x20) ? "DMA " :
- (c2 & 0x80) ? "PIO " : "off ",
- (c3 & 0x10) ? "UDMA" : (c3 & 0x20) ? "DMA " :
- (c3 & 0x80) ? "PIO " : "off ");
- }
- }
-#endif
+struct hpt_info
+{
+ u8 max_mode; /* Speeds allowed */
+ int revision; /* Chipset revision */
+ int flags; /* Chipset properties */
+#define PLL_MODE 1
+#define IS_372N 2
+ /* Speed table */
+ struct chipset_bus_clock_list_entry *speed;
+};
-static u32 hpt_revision (struct pci_dev *dev)
+/*
+ * This wants fixing so that we do everything not by classrev
+ * (which breaks on the newest chips) but by creating an
+ * enumeration of chip variants and using that
+ */
+
+static __devinit u32 hpt_revision (struct pci_dev *dev)
{
u32 class_rev;
pci_read_config_dword(dev, PCI_CLASS_REVISION, &class_rev);
@@ -507,37 +497,33 @@ static u32 hpt_revision (struct pci_dev *dev)
return class_rev;
}
-static u32 hpt_minimum_revision (struct pci_dev *dev, int revision)
-{
- unsigned int class_rev = hpt_revision(dev);
- revision--;
- return ((int) (class_rev > revision) ? 1 : 0);
-}
-
static int check_in_drive_lists(ide_drive_t *drive, const char **list);
static u8 hpt3xx_ratemask (ide_drive_t *drive)
{
- struct pci_dev *dev = HWIF(drive)->pci_dev;
+ ide_hwif_t *hwif = drive->hwif;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
u8 mode = 0;
- if (hpt_minimum_revision(dev, 8)) { /* HPT374 */
+ /* FIXME: TODO - move this to set info->mode once at boot */
+
+ if (info->revision >= 8) { /* HPT374 */
mode = (HPT374_ALLOW_ATA133_6) ? 4 : 3;
- } else if (hpt_minimum_revision(dev, 7)) { /* HPT371 */
+ } else if (info->revision >= 7) { /* HPT371 */
mode = (HPT371_ALLOW_ATA133_6) ? 4 : 3;
- } else if (hpt_minimum_revision(dev, 6)) { /* HPT302 */
+ } else if (info->revision >= 6) { /* HPT302 */
mode = (HPT302_ALLOW_ATA133_6) ? 4 : 3;
- } else if (hpt_minimum_revision(dev, 5)) { /* HPT372 */
+ } else if (info->revision >= 5) { /* HPT372 */
mode = (HPT372_ALLOW_ATA133_6) ? 4 : 3;
- } else if (hpt_minimum_revision(dev, 4)) { /* HPT370A */
+ } else if (info->revision >= 4) { /* HPT370A */
mode = (HPT370_ALLOW_ATA100_5) ? 3 : 2;
- } else if (hpt_minimum_revision(dev, 3)) { /* HPT370 */
+ } else if (info->revision >= 3) { /* HPT370 */
mode = (HPT370_ALLOW_ATA100_5) ? 3 : 2;
mode = (check_in_drive_lists(drive, bad_ata33)) ? 0 : mode;
} else { /* HPT366 and HPT368 */
mode = (check_in_drive_lists(drive, bad_ata33)) ? 0 : 2;
}
- if (!eighty_ninty_three(drive) && (mode))
+ if (!eighty_ninty_three(drive) && mode)
mode = min(mode, (u8)1);
return mode;
}
@@ -549,7 +535,8 @@ static u8 hpt3xx_ratemask (ide_drive_t *drive)
static u8 hpt3xx_ratefilter (ide_drive_t *drive, u8 speed)
{
- struct pci_dev *dev = HWIF(drive)->pci_dev;
+ ide_hwif_t *hwif = drive->hwif;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
u8 mode = hpt3xx_ratemask(drive);
if (drive->media != ide_disk)
@@ -561,7 +548,7 @@ static u8 hpt3xx_ratefilter (ide_drive_t *drive, u8 speed)
break;
case 0x03:
speed = min(speed, (u8)XFER_UDMA_5);
- if (hpt_minimum_revision(dev, 5))
+ if (info->revision >= 5)
break;
if (check_in_drive_lists(drive, bad_ata100_5))
speed = min(speed, (u8)XFER_UDMA_4);
@@ -571,7 +558,7 @@ static u8 hpt3xx_ratefilter (ide_drive_t *drive, u8 speed)
/*
* CHECK ME, Does this need to be set to 5 ??
*/
- if (hpt_minimum_revision(dev, 3))
+ if (info->revision >= 3)
break;
if ((check_in_drive_lists(drive, bad_ata66_4)) ||
(!(HPT366_ALLOW_ATA66_4)))
@@ -585,7 +572,7 @@ static u8 hpt3xx_ratefilter (ide_drive_t *drive, u8 speed)
/*
* CHECK ME, Does this need to be set to 5 ??
*/
- if (hpt_minimum_revision(dev, 3))
+ if (info->revision >= 3)
break;
if (check_in_drive_lists(drive, bad_ata33))
speed = min(speed, (u8)XFER_MW_DMA_2);
@@ -624,11 +611,12 @@ static unsigned int pci_bus_clock_list (u8 speed, struct chipset_bus_clock_list_
static int hpt36x_tune_chipset(ide_drive_t *drive, u8 xferspeed)
{
- struct pci_dev *dev = HWIF(drive)->pci_dev;
+ ide_hwif_t *hwif = drive->hwif;
+ struct pci_dev *dev = hwif->pci_dev;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
u8 speed = hpt3xx_ratefilter(drive, xferspeed);
-// u8 speed = ide_rate_filter(hpt3xx_ratemask(drive), xferspeed);
u8 regtime = (drive->select.b.unit & 0x01) ? 0x44 : 0x40;
- u8 regfast = (HWIF(drive)->channel) ? 0x55 : 0x51;
+ u8 regfast = (hwif->channel) ? 0x55 : 0x51;
u8 drive_fast = 0;
u32 reg1 = 0, reg2 = 0;
@@ -636,16 +624,11 @@ static int hpt36x_tune_chipset(ide_drive_t *drive, u8 xferspeed)
* Disable the "fast interrupt" prediction.
*/
pci_read_config_byte(dev, regfast, &drive_fast);
-#if 0
- if (drive_fast & 0x02)
- pci_write_config_byte(dev, regfast, drive_fast & ~0x20);
-#else
if (drive_fast & 0x80)
pci_write_config_byte(dev, regfast, drive_fast & ~0x80);
-#endif
- reg2 = pci_bus_clock_list(speed,
- (struct chipset_bus_clock_list_entry *) pci_get_drvdata(dev));
+ reg2 = pci_bus_clock_list(speed, info->speed);
+
/*
* Disable on-chip PIO FIFO/buffer
* (to avoid problems handling I/O errors later)
@@ -665,10 +648,11 @@ static int hpt36x_tune_chipset(ide_drive_t *drive, u8 xferspeed)
static int hpt370_tune_chipset(ide_drive_t *drive, u8 xferspeed)
{
- struct pci_dev *dev = HWIF(drive)->pci_dev;
+ ide_hwif_t *hwif = drive->hwif;
+ struct pci_dev *dev = hwif->pci_dev;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
u8 speed = hpt3xx_ratefilter(drive, xferspeed);
-// u8 speed = ide_rate_filter(hpt3xx_ratemask(drive), xferspeed);
- u8 regfast = (HWIF(drive)->channel) ? 0x55 : 0x51;
+ u8 regfast = (drive->hwif->channel) ? 0x55 : 0x51;
u8 drive_pci = 0x40 + (drive->dn * 4);
u8 new_fast = 0, drive_fast = 0;
u32 list_conf = 0, drive_conf = 0;
@@ -693,17 +677,13 @@ static int hpt370_tune_chipset(ide_drive_t *drive, u8 xferspeed)
if (new_fast != drive_fast)
pci_write_config_byte(dev, regfast, new_fast);
- list_conf = pci_bus_clock_list(speed,
- (struct chipset_bus_clock_list_entry *)
- pci_get_drvdata(dev));
+ list_conf = pci_bus_clock_list(speed, info->speed);
pci_read_config_dword(dev, drive_pci, &drive_conf);
list_conf = (list_conf & ~conf_mask) | (drive_conf & conf_mask);
- if (speed < XFER_MW_DMA_0) {
+ if (speed < XFER_MW_DMA_0)
list_conf &= ~0x80000000; /* Disable on-chip PIO FIFO/buffer */
- }
-
pci_write_config_dword(dev, drive_pci, list_conf);
return ide_config_drive_speed(drive, speed);
@@ -711,10 +691,11 @@ static int hpt370_tune_chipset(ide_drive_t *drive, u8 xferspeed)
static int hpt372_tune_chipset(ide_drive_t *drive, u8 xferspeed)
{
- struct pci_dev *dev = HWIF(drive)->pci_dev;
+ ide_hwif_t *hwif = drive->hwif;
+ struct pci_dev *dev = hwif->pci_dev;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
u8 speed = hpt3xx_ratefilter(drive, xferspeed);
-// u8 speed = ide_rate_filter(hpt3xx_ratemask(drive), xferspeed);
- u8 regfast = (HWIF(drive)->channel) ? 0x55 : 0x51;
+ u8 regfast = (drive->hwif->channel) ? 0x55 : 0x51;
u8 drive_fast = 0, drive_pci = 0x40 + (drive->dn * 4);
u32 list_conf = 0, drive_conf = 0;
u32 conf_mask = (speed >= XFER_MW_DMA_0) ? 0xc0000000 : 0x30070000;
@@ -726,10 +707,8 @@ static int hpt372_tune_chipset(ide_drive_t *drive, u8 xferspeed)
pci_read_config_byte(dev, regfast, &drive_fast);
drive_fast &= ~0x07;
pci_write_config_byte(dev, regfast, drive_fast);
-
- list_conf = pci_bus_clock_list(speed,
- (struct chipset_bus_clock_list_entry *)
- pci_get_drvdata(dev));
+
+ list_conf = pci_bus_clock_list(speed, info->speed);
pci_read_config_dword(dev, drive_pci, &drive_conf);
list_conf = (list_conf & ~conf_mask) | (drive_conf & conf_mask);
if (speed < XFER_MW_DMA_0)
@@ -741,19 +720,14 @@ static int hpt372_tune_chipset(ide_drive_t *drive, u8 xferspeed)
static int hpt3xx_tune_chipset (ide_drive_t *drive, u8 speed)
{
- struct pci_dev *dev = HWIF(drive)->pci_dev;
+ ide_hwif_t *hwif = drive->hwif;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
- if (hpt_minimum_revision(dev, 8))
+ if (info->revision >= 8)
return hpt372_tune_chipset(drive, speed); /* not a typo */
-#if 0
- else if (hpt_minimum_revision(dev, 7))
- hpt371_tune_chipset(drive, speed);
- else if (hpt_minimum_revision(dev, 6))
- hpt302_tune_chipset(drive, speed);
-#endif
- else if (hpt_minimum_revision(dev, 5))
+ else if (info->revision >= 5)
return hpt372_tune_chipset(drive, speed);
- else if (hpt_minimum_revision(dev, 3))
+ else if (info->revision >= 3)
return hpt370_tune_chipset(drive, speed);
else /* hpt368: hpt_minimum_revision(dev, 2) */
return hpt36x_tune_chipset(drive, speed);
@@ -779,8 +753,14 @@ static void hpt3xx_tune_drive (ide_drive_t *drive, u8 pio)
static int config_chipset_for_dma (ide_drive_t *drive)
{
u8 speed = ide_dma_speed(drive, hpt3xx_ratemask(drive));
+ ide_hwif_t *hwif = drive->hwif;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
- if (!(speed))
+ if (!speed)
+ return 0;
+
+ /* If we don't have any timings we can't do a lot */
+ if (info->speed == NULL)
return 0;
(void) hpt3xx_tune_chipset(drive, speed);
@@ -794,7 +774,7 @@ static int hpt3xx_quirkproc (ide_drive_t *drive)
static void hpt3xx_intrproc (ide_drive_t *drive)
{
- ide_hwif_t *hwif = HWIF(drive);
+ ide_hwif_t *hwif = drive->hwif;
if (drive->quirk_list)
return;
@@ -804,24 +784,26 @@ static void hpt3xx_intrproc (ide_drive_t *drive)
static void hpt3xx_maskproc (ide_drive_t *drive, int mask)
{
- struct pci_dev *dev = HWIF(drive)->pci_dev;
+ ide_hwif_t *hwif = drive->hwif;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
+ struct pci_dev *dev = hwif->pci_dev;
if (drive->quirk_list) {
- if (hpt_minimum_revision(dev,3)) {
+ if (info->revision >= 3) {
u8 reg5a = 0;
pci_read_config_byte(dev, 0x5a, ®5a);
if (((reg5a & 0x10) >> 4) != mask)
pci_write_config_byte(dev, 0x5a, mask ? (reg5a | 0x10) : (reg5a & ~0x10));
} else {
if (mask) {
- disable_irq(HWIF(drive)->irq);
+ disable_irq(hwif->irq);
} else {
- enable_irq(HWIF(drive)->irq);
+ enable_irq(hwif->irq);
}
}
} else {
if (IDE_CONTROL_REG)
- HWIF(drive)->OUTB(mask ? (drive->ctl | 2) :
+ hwif->OUTB(mask ? (drive->ctl | 2) :
(drive->ctl & ~2),
IDE_CONTROL_REG);
}
@@ -829,12 +811,12 @@ static void hpt3xx_maskproc (ide_drive_t *drive, int mask)
static int hpt366_config_drive_xfer_rate (ide_drive_t *drive)
{
- ide_hwif_t *hwif = HWIF(drive);
+ ide_hwif_t *hwif = drive->hwif;
struct hd_driveid *id = drive->id;
drive->init_speed = 0;
- if (id && (id->capability & 1) && drive->autodma) {
+ if ((id->capability & 1) && drive->autodma) {
if (ide_use_dma(drive)) {
if (config_chipset_for_dma(drive))
@@ -868,15 +850,6 @@ static int hpt366_ide_dma_lostirq (ide_drive_t *drive)
drive->name, __FUNCTION__, reg50h, reg52h, reg5ah);
if (reg5ah & 0x10)
pci_write_config_byte(dev, 0x5a, reg5ah & ~0x10);
-#if 0
- /* how about we flush and reset, mmmkay? */
- pci_write_config_byte(dev, 0x51, 0x1F);
- /* fall through to a reset */
- case dma_start:
- case ide_dma_end:
- /* reset the chips state over and over.. */
- pci_write_config_byte(dev, 0x51, 0x13);
-#endif
return __ide_dma_lostirq(drive);
}
@@ -919,7 +892,7 @@ static void hpt370_lostirq_timeout (ide_drive_t *drive)
u8 dma_stat = 0, dma_cmd = 0;
pci_read_config_byte(HWIF(drive)->pci_dev, reginfo, &bfifo);
- printk("%s: %d bytes in FIFO\n", drive->name, bfifo);
+ printk(KERN_DEBUG "%s: %d bytes in FIFO\n", drive->name, bfifo);
hpt370_clear_engine(drive);
/* get dma command mode */
dma_cmd = hwif->INB(hwif->dma_command);
@@ -1047,15 +1020,6 @@ static void hpt372n_rw_disk(ide_drive_t *drive, struct request *rq)
static void hpt3xx_reset (ide_drive_t *drive)
{
-#if 0
- unsigned long high_16 = pci_resource_start(HWIF(drive)->pci_dev, 4);
- u8 reset = (HWIF(drive)->channel) ? 0x80 : 0x40;
- u8 reg59h = 0;
-
- pci_read_config_byte(HWIF(drive)->pci_dev, 0x59, ®59h);
- pci_write_config_byte(HWIF(drive)->pci_dev, 0x59, reg59h|reset);
- pci_write_config_byte(HWIF(drive)->pci_dev, 0x59, reg59h);
-#endif
}
static int hpt3xx_tristate (ide_drive_t * drive, int state)
@@ -1065,8 +1029,6 @@ static int hpt3xx_tristate (ide_drive_t * drive, int state)
u8 reg59h = 0, reset = (hwif->channel) ? 0x80 : 0x40;
u8 regXXh = 0, state_reg= (hwif->channel) ? 0x57 : 0x53;
-// hwif->bus_state = state;
-
pci_read_config_byte(dev, 0x59, ®59h);
pci_read_config_byte(dev, state_reg, ®XXh);
@@ -1093,7 +1055,7 @@ static int hpt3xx_tristate (ide_drive_t * drive, int state)
#define TRISTATE_BIT 0x8000
static int hpt370_busproc(ide_drive_t * drive, int state)
{
- ide_hwif_t *hwif = HWIF(drive);
+ ide_hwif_t *hwif = drive->hwif;
struct pci_dev *dev = hwif->pci_dev;
u8 tristate = 0, resetmask = 0, bus_reg = 0;
u16 tri_reg;
@@ -1148,33 +1110,44 @@ static int hpt370_busproc(ide_drive_t * drive, int state)
return 0;
}
-static int __devinit init_hpt37x(struct pci_dev *dev)
+static void __devinit hpt366_clocking(ide_hwif_t *hwif)
{
+ u32 reg1 = 0;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
+
+ pci_read_config_dword(hwif->pci_dev, 0x40, ®1);
+
+ /* detect bus speed by looking at control reg timing: */
+ switch((reg1 >> 8) & 7) {
+ case 5:
+ info->speed = forty_base_hpt366;
+ break;
+ case 9:
+ info->speed = twenty_five_base_hpt366;
+ break;
+ case 7:
+ default:
+ info->speed = thirty_three_base_hpt366;
+ break;
+ }
+}
+
+static void __devinit hpt37x_clocking(ide_hwif_t *hwif)
+{
+ struct hpt_info *info = ide_get_hwifdata(hwif);
+ struct pci_dev *dev = hwif->pci_dev;
int adjust, i;
u16 freq;
u32 pll;
u8 reg5bh;
- u8 reg5ah = 0;
- unsigned long dmabase = pci_resource_start(dev, 4);
- u8 did, rid;
- int is_372n = 0;
- pci_read_config_byte(dev, 0x5a, ®5ah);
- /* interrupt force enable */
- pci_write_config_byte(dev, 0x5a, (reg5ah & ~0x10));
-
- if(dmabase)
- {
- did = inb(dmabase + 0x22);
- rid = inb(dmabase + 0x28);
-
- if((did == 4 && rid == 6) || (did == 5 && rid > 1))
- is_372n = 1;
- }
-
/*
* default to pci clock. make sure MA15/16 are set to output
- * to prevent drives having problems with 40-pin cables.
+ * to prevent drives having problems with 40-pin cables. Needed
+ * for some drives such as IBM-DTLA which will not enter ready
+ * state on reset when PDIAG is a input.
+ *
+ * ToDo: should we set 0x21 when using PLL mode ?
*/
pci_write_config_byte(dev, 0x5b, 0x23);
@@ -1197,9 +1170,7 @@ static int __devinit init_hpt37x(struct pci_dev *dev)
* Currently we always set up the PLL for the 372N
*/
- pci_set_drvdata(dev, NULL);
-
- if(is_372n)
+ if(info->flags & IS_372N)
{
printk(KERN_INFO "hpt: HPT372N detected, using 372N timing.\n");
if(freq < 0x55)
@@ -1227,39 +1198,38 @@ static int __devinit init_hpt37x(struct pci_dev *dev)
pll = F_LOW_PCI_66;
if (pll == F_LOW_PCI_33) {
- if (hpt_minimum_revision(dev,8))
- pci_set_drvdata(dev, (void *) thirty_three_base_hpt374);
- else if (hpt_minimum_revision(dev,5))
- pci_set_drvdata(dev, (void *) thirty_three_base_hpt372);
- else if (hpt_minimum_revision(dev,4))
- pci_set_drvdata(dev, (void *) thirty_three_base_hpt370a);
+ if (info->revision >= 8)
+ info->speed = thirty_three_base_hpt374;
+ else if (info->revision >= 5)
+ info->speed = thirty_three_base_hpt372;
+ else if (info->revision >= 4)
+ info->speed = thirty_three_base_hpt370a;
else
- pci_set_drvdata(dev, (void *) thirty_three_base_hpt370);
- printk("HPT37X: using 33MHz PCI clock\n");
+ info->speed = thirty_three_base_hpt370;
+ printk(KERN_DEBUG "HPT37X: using 33MHz PCI clock\n");
} else if (pll == F_LOW_PCI_40) {
/* Unsupported */
} else if (pll == F_LOW_PCI_50) {
- if (hpt_minimum_revision(dev,8))
- pci_set_drvdata(dev, (void *) fifty_base_hpt370a);
- else if (hpt_minimum_revision(dev,5))
- pci_set_drvdata(dev, (void *) fifty_base_hpt372);
- else if (hpt_minimum_revision(dev,4))
- pci_set_drvdata(dev, (void *) fifty_base_hpt370a);
+ if (info->revision >= 8)
+ info->speed = fifty_base_hpt370a;
+ else if (info->revision >= 5)
+ info->speed = fifty_base_hpt372;
+ else if (info->revision >= 4)
+ info->speed = fifty_base_hpt370a;
else
- pci_set_drvdata(dev, (void *) fifty_base_hpt370a);
- printk("HPT37X: using 50MHz PCI clock\n");
+ info->speed = fifty_base_hpt370a;
+ printk(KERN_DEBUG "HPT37X: using 50MHz PCI clock\n");
} else {
- if (hpt_minimum_revision(dev,8))
- {
+ if (info->revision >= 8) {
printk(KERN_ERR "HPT37x: 66MHz timings are not supported.\n");
}
- else if (hpt_minimum_revision(dev,5))
- pci_set_drvdata(dev, (void *) sixty_six_base_hpt372);
- else if (hpt_minimum_revision(dev,4))
- pci_set_drvdata(dev, (void *) sixty_six_base_hpt370a);
+ else if (info->revision >= 5)
+ info->speed = sixty_six_base_hpt372;
+ else if (info->revision >= 4)
+ info->speed = sixty_six_base_hpt370a;
else
- pci_set_drvdata(dev, (void *) sixty_six_base_hpt370);
- printk("HPT37X: using 66MHz PCI clock\n");
+ info->speed = sixty_six_base_hpt370;
+ printk(KERN_DEBUG "HPT37X: using 66MHz PCI clock\n");
}
}
@@ -1269,11 +1239,19 @@ static int __devinit init_hpt37x(struct pci_dev *dev)
* result in slow reads when using a 33MHz PCI clock. we also
* don't like to use the PLL because it will cause glitches
* on PRST/SRST when the HPT state engine gets reset.
+ *
+ * ToDo: Use 66MHz PLL when ATA133 devices are present on a
+ * 372 device so we can get ATA133 support
*/
- if (pci_get_drvdata(dev))
+ if (info->speed)
goto init_hpt37X_done;
+
+ info->flags |= PLL_MODE;
/*
+ * FIXME: make this work correctly, esp with 372N as per
+ * reference driver code.
+ *
* adjust PLL based upon PCI clock, enable it, and wait for
* stabilization.
*/
@@ -1298,14 +1276,14 @@ static int __devinit init_hpt37x(struct pci_dev *dev)
pci_write_config_dword(dev, 0x5c,
pll & ~0x100);
pci_write_config_byte(dev, 0x5b, 0x21);
- if (hpt_minimum_revision(dev,8))
- pci_set_drvdata(dev, (void *) fifty_base_hpt370a);
- else if (hpt_minimum_revision(dev,5))
- pci_set_drvdata(dev, (void *) fifty_base_hpt372);
- else if (hpt_minimum_revision(dev,4))
- pci_set_drvdata(dev, (void *) fifty_base_hpt370a);
+ if (info->revision >= 8)
+ info->speed = fifty_base_hpt370a;
+ else if (info->revision >= 5)
+ info->speed = fifty_base_hpt372;
+ else if (info->revision >= 4)
+ info->speed = fifty_base_hpt370a;
else
- pci_set_drvdata(dev, (void *) fifty_base_hpt370a);
+ info->speed = fifty_base_hpt370a;
printk("HPT37X: using 50MHz internal PLL\n");
goto init_hpt37X_done;
}
@@ -1318,10 +1296,22 @@ pll_recal:
}
init_hpt37X_done:
+ if (!info->speed)
+ printk(KERN_ERR "HPT37X%s: unknown bus timing [%d %d].\n",
+ (info->flags & IS_372N)?"N":"", pll, freq);
/* reset state engine */
pci_write_config_byte(dev, 0x50, 0x37);
pci_write_config_byte(dev, 0x54, 0x37);
udelay(100);
+}
+
+static int __devinit init_hpt37x(struct pci_dev *dev)
+{
+ u8 reg5ah;
+
+ pci_read_config_byte(dev, 0x5a, ®5ah);
+ /* interrupt force enable */
+ pci_write_config_byte(dev, 0x5a, (reg5ah & ~0x10));
return 0;
}
@@ -1338,59 +1328,27 @@ static int __devinit init_hpt366(struct pci_dev *dev)
pci_write_config_byte(dev, 0x51, drive_fast & ~0x80);
pci_read_config_dword(dev, 0x40, ®1);
- /* detect bus speed by looking at control reg timing: */
- switch((reg1 >> 8) & 7) {
- case 5:
- pci_set_drvdata(dev, (void *) forty_base_hpt366);
- break;
- case 9:
- pci_set_drvdata(dev, (void *) twenty_five_base_hpt366);
- break;
- case 7:
- default:
- pci_set_drvdata(dev, (void *) thirty_three_base_hpt366);
- break;
- }
-
- if (!pci_get_drvdata(dev))
- {
- printk(KERN_ERR "hpt366: unknown bus timing.\n");
- pci_set_drvdata(dev, NULL);
- }
return 0;
}
static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const char *name)
{
int ret = 0;
- u8 test = 0;
-
+ /* FIXME: Not portable */
if (dev->resource[PCI_ROM_RESOURCE].start)
pci_write_config_byte(dev, PCI_ROM_ADDRESS,
dev->resource[PCI_ROM_RESOURCE].start | PCI_ROM_ADDRESS_ENABLE);
- pci_read_config_byte(dev, PCI_CACHE_LINE_SIZE, &test);
- if (test != (L1_CACHE_BYTES / 4))
- pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE,
- (L1_CACHE_BYTES / 4));
+ pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, (L1_CACHE_BYTES / 4));
+ pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x78);
+ pci_write_config_byte(dev, PCI_MIN_GNT, 0x08);
+ pci_write_config_byte(dev, PCI_MAX_LAT, 0x08);
- pci_read_config_byte(dev, PCI_LATENCY_TIMER, &test);
- if (test != 0x78)
- pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x78);
-
- pci_read_config_byte(dev, PCI_MIN_GNT, &test);
- if (test != 0x08)
- pci_write_config_byte(dev, PCI_MIN_GNT, 0x08);
-
- pci_read_config_byte(dev, PCI_MAX_LAT, &test);
- if (test != 0x08)
- pci_write_config_byte(dev, PCI_MAX_LAT, 0x08);
-
- if (hpt_minimum_revision(dev, 3)) {
+ if (hpt_revision(dev) >= 3)
ret = init_hpt37x(dev);
- } else {
- ret =init_hpt366(dev);
- }
+ else
+ ret = init_hpt366(dev);
+
if (ret)
return ret;
@@ -1400,27 +1358,16 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
{
struct pci_dev *dev = hwif->pci_dev;
+ struct hpt_info *info = ide_get_hwifdata(hwif);
u8 ata66 = 0, regmask = (hwif->channel) ? 0x01 : 0x02;
- u8 did, rid;
- unsigned long dmabase = hwif->dma_base;
- int is_372n = 0;
- if(dmabase)
- {
- did = inb(dmabase + 0x22);
- rid = inb(dmabase + 0x28);
-
- if((did == 4 && rid == 6) || (did == 5 && rid > 1))
- is_372n = 1;
- }
-
hwif->tuneproc = &hpt3xx_tune_drive;
hwif->speedproc = &hpt3xx_tune_chipset;
hwif->quirkproc = &hpt3xx_quirkproc;
hwif->intrproc = &hpt3xx_intrproc;
hwif->maskproc = &hpt3xx_maskproc;
- if(is_372n)
+ if(info->flags & IS_372N)
hwif->rw_disk = &hpt372n_rw_disk;
/*
@@ -1428,7 +1375,7 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
* address lines to access an external eeprom. To read valid
* cable detect state the pins must be enabled as inputs.
*/
- if (hpt_minimum_revision(dev, 8) && PCI_FUNC(dev->devfn) & 1) {
+ if (info->revision >= 8 && (PCI_FUNC(dev->devfn) & 1)) {
/*
* HPT374 PCI function 1
* - set bit 15 of reg 0x52 to enable TCBLID as input
@@ -1443,7 +1390,7 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
pci_read_config_byte(dev, 0x5a, &ata66);
pci_write_config_word(dev, 0x52, mcr3);
pci_write_config_word(dev, 0x56, mcr6);
- } else if (hpt_minimum_revision(dev, 3)) {
+ } else if (info->revision >= 3) {
/*
* HPT370/372 and 374 pcifn 0
* - clear bit 0 of 0x5b to enable P/SCBLID as inputs
@@ -1470,7 +1417,7 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
hwif->serialized = hwif->mate->serialized = 1;
#endif
- if (hpt_minimum_revision(dev,3)) {
+ if (info->revision >= 3) {
u8 reg5ah = 0;
pci_write_config_byte(dev, 0x5a, reg5ah & ~0x10);
/*
@@ -1480,8 +1427,7 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
*/
hwif->resetproc = &hpt3xx_reset;
hwif->busproc = &hpt370_busproc;
-// hwif->drives[0].autotune = hwif->drives[1].autotune = 1;
- } else if (hpt_minimum_revision(dev,2)) {
+ } else if (info->revision >= 2) {
hwif->resetproc = &hpt3xx_reset;
hwif->busproc = &hpt3xx_tristate;
} else {
@@ -1502,18 +1448,18 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
hwif->udma_four = ((ata66 & regmask) ? 0 : 1);
hwif->ide_dma_check = &hpt366_config_drive_xfer_rate;
- if (hpt_minimum_revision(dev,8)) {
+ if (info->revision >= 8) {
hwif->ide_dma_test_irq = &hpt374_ide_dma_test_irq;
hwif->ide_dma_end = &hpt374_ide_dma_end;
- } else if (hpt_minimum_revision(dev,5)) {
+ } else if (info->revision >= 5) {
hwif->ide_dma_test_irq = &hpt374_ide_dma_test_irq;
hwif->ide_dma_end = &hpt374_ide_dma_end;
- } else if (hpt_minimum_revision(dev,3)) {
+ } else if (info->revision >= 3) {
hwif->dma_start = &hpt370_ide_dma_start;
hwif->ide_dma_end = &hpt370_ide_dma_end;
hwif->ide_dma_timeout = &hpt370_ide_dma_timeout;
hwif->ide_dma_lostirq = &hpt370_ide_dma_lostirq;
- } else if (hpt_minimum_revision(dev,2))
+ } else if (info->revision >= 2)
hwif->ide_dma_lostirq = &hpt366_ide_dma_lostirq;
else
hwif->ide_dma_lostirq = &hpt366_ide_dma_lostirq;
@@ -1526,6 +1472,7 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
static void __devinit init_dma_hpt366(ide_hwif_t *hwif, unsigned long dmabase)
{
+ struct hpt_info *info = ide_get_hwifdata(hwif);
u8 masterdma = 0, slavedma = 0;
u8 dma_new = 0, dma_old = 0;
u8 primary = hwif->channel ? 0x4b : 0x43;
@@ -1535,8 +1482,7 @@ static void __devinit init_dma_hpt366(ide_hwif_t *hwif, unsigned long dmabase)
if (!dmabase)
return;
- if(pci_get_drvdata(hwif->pci_dev) == NULL)
- {
+ if(info->speed == NULL) {
printk(KERN_WARNING "hpt: no known IDE timings, disabling DMA.\n");
return;
}
@@ -1559,6 +1505,40 @@ static void __devinit init_dma_hpt366(ide_hwif_t *hwif, unsigned long dmabase)
ide_setup_dma(hwif, dmabase, 8);
}
+/*
+ * We "borrow" this hook in order to set the data structures
+ * up early enough before dma or init_hwif calls are made.
+ */
+
+static void __devinit init_iops_hpt366(ide_hwif_t *hwif)
+{
+ struct hpt_info *info = kmalloc(sizeof(struct hpt_info), GFP_KERNEL);
+ unsigned long dmabase = pci_resource_start(hwif->pci_dev, 4);
+ u8 did, rid;
+
+ if(info == NULL) {
+ printk(KERN_WARNING "hpt366: out of memory.\n");
+ return;
+ }
+ memset(info, 0, sizeof(struct hpt_info));
+ ide_set_hwifdata(hwif, info);
+
+ if(dmabase) {
+ did = inb(dmabase + 0x22);
+ rid = inb(dmabase + 0x28);
+
+ if((did == 4 && rid == 6) || (did == 5 && rid > 1))
+ info->flags |= IS_372N;
+ }
+
+ info->revision = hpt_revision(hwif->pci_dev);
+
+ if (info->revision >= 3)
+ hpt37x_clocking(hwif);
+ else
+ hpt366_clocking(hwif);
+}
+
static int __devinit init_setup_hpt374(struct pci_dev *dev, ide_pci_device_t *d)
{
struct pci_dev *findev = NULL;
@@ -1646,6 +1626,7 @@ static ide_pci_device_t hpt366_chipsets[] __devinitdata = {
.name = "HPT366",
.init_setup = init_setup_hpt366,
.init_chipset = init_chipset_hpt366,
+ .init_iops = init_iops_hpt366,
.init_hwif = init_hwif_hpt366,
.init_dma = init_dma_hpt366,
.channels = 2,
@@ -1656,6 +1637,7 @@ static ide_pci_device_t hpt366_chipsets[] __devinitdata = {
.name = "HPT372A",
.init_setup = init_setup_hpt37x,
.init_chipset = init_chipset_hpt366,
+ .init_iops = init_iops_hpt366,
.init_hwif = init_hwif_hpt366,
.init_dma = init_dma_hpt366,
.channels = 2,
@@ -1665,6 +1647,7 @@ static ide_pci_device_t hpt366_chipsets[] __devinitdata = {
.name = "HPT302",
.init_setup = init_setup_hpt37x,
.init_chipset = init_chipset_hpt366,
+ .init_iops = init_iops_hpt366,
.init_hwif = init_hwif_hpt366,
.init_dma = init_dma_hpt366,
.channels = 2,
@@ -1674,6 +1657,7 @@ static ide_pci_device_t hpt366_chipsets[] __devinitdata = {
.name = "HPT371",
.init_setup = init_setup_hpt37x,
.init_chipset = init_chipset_hpt366,
+ .init_iops = init_iops_hpt366,
.init_hwif = init_hwif_hpt366,
.init_dma = init_dma_hpt366,
.channels = 2,
@@ -1683,6 +1667,7 @@ static ide_pci_device_t hpt366_chipsets[] __devinitdata = {
.name = "HPT374",
.init_setup = init_setup_hpt374,
.init_chipset = init_chipset_hpt366,
+ .init_iops = init_iops_hpt366,
.init_hwif = init_hwif_hpt366,
.init_dma = init_dma_hpt366,
.channels = 2, /* 4 */
@@ -1692,6 +1677,7 @@ static ide_pci_device_t hpt366_chipsets[] __devinitdata = {
.name = "HPT372N",
.init_setup = init_setup_hpt37x,
.init_chipset = init_chipset_hpt366,
+ .init_iops = init_iops_hpt366,
.init_hwif = init_hwif_hpt366,
.init_dma = init_dma_hpt366,
.channels = 2, /* 4 */
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
new file mode 100644
index 000000000000..e440036e651f
--- /dev/null
+++ b/drivers/ide/pci/it821x.c
@@ -0,0 +1,812 @@
+
+/*
+ * linux/drivers/ide/pci/it821x.c Version 0.09 December 2004
+ *
+ * Copyright (C) 2004 Red Hat
+ *
+ * May be copied or modified under the terms of the GNU General Public License
+ * Based in part on the ITE vendor provided SCSI driver.
+ *
+ * Documentation available from
+ * http://www.ite.com.tw/pc/IT8212F_V04.pdf
+ * Some other documents are NDA.
+ *
+ * The ITE8212 isn't exactly a standard IDE controller. It has two
+ * modes. In pass through mode then it is an IDE controller. In its smart
+ * mode its actually quite a capable hardware raid controller disguised
+ * as an IDE controller. Smart mode only understands DMA read/write and
+ * identify, none of the fancier commands apply. The IT8211 is identical
+ * in other respects but lacks the raid mode.
+ *
+ * Errata:
+ * o Rev 0x10 also requires master/slave hold the same DMA timings and
+ * cannot do ATAPI MWDMA.
+ * o The identify data for raid volumes lacks CHS info (technically ok)
+ * but also fails to set the LBA28 and other bits. We fix these in
+ * the IDE probe quirk code.
+ * o If you write LBA48 sized I/O's (ie > 256 sector) in smart mode
+ * raid then the controller firmware dies
+ * o Smart mode without RAID doesn't clear all the necessary identify
+ * bits to reduce the command set to the one used
+ *
+ * This has a few impacts on the driver
+ * - In pass through mode we do all the work you would expect
+ * - In smart mode the clocking set up is done by the controller generally
+ * but we must watch the other limits and filter.
+ * - There are a few extra vendor commands that actually talk to the
+ * controller but only work PIO with no IRQ.
+ *
+ * Vendor areas of the identify block in smart mode are used for the
+ * timing and policy set up. Each HDD in raid mode also has a serial
+ * block on the disk. The hardware extra commands are get/set chip status,
+ * rebuild, get rebuild status.
+ *
+ * In Linux the driver supports pass through mode as if the device was
+ * just another IDE controller. If the smart mode is running then
+ * volumes are managed by the controller firmware and each IDE "disk"
+ * is a raid volume. Even more cute - the controller can do automated
+ * hotplug and rebuild.
+ *
+ * The pass through controller itself is a little demented. It has a
+ * flaw that it has a single set of PIO/MWDMA timings per channel so
+ * non UDMA devices restrict each others performance. It also has a
+ * single clock source per channel so mixed UDMA100/133 performance
+ * isn't perfect and we have to pick a clock. Thankfully none of this
+ * matters in smart mode. ATAPI DMA is not currently supported.
+ *
+ * It seems the smart mode is a win for RAID1/RAID10 but otherwise not.
+ *
+ * TODO
+ * - ATAPI UDMA is ok but not MWDMA it seems
+ * - RAID configuration ioctls
+ * - Move to libata once it grows up
+ */
+
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+
+#include
+
+struct it821x_dev
+{
+ unsigned int smart:1, /* Are we in smart raid mode */
+ timing10:1; /* Rev 0x10 */
+ u8 clock_mode; /* 0, ATA_50 or ATA_66 */
+ u8 want[2][2]; /* Mode/Pri log for master slave */
+ /* We need these for switching the clock when DMA goes on/off
+ The high byte is the 66Mhz timing */
+ u16 pio[2]; /* Cached PIO values */
+ u16 mwdma[2]; /* Cached MWDMA values */
+ u16 udma[2]; /* Cached UDMA values (per drive) */
+};
+
+#define ATA_66 0
+#define ATA_50 1
+#define ATA_ANY 2
+
+#define UDMA_OFF 0
+#define MWDMA_OFF 0
+
+/*
+ * We allow users to force the card into non raid mode without
+ * flashing the alternative BIOS. This is also neccessary right now
+ * for embedded platforms that cannot run a PC BIOS but are using this
+ * device.
+ */
+
+static int it8212_noraid;
+
+/**
+ * it821x_program - program the PIO/MWDMA registers
+ * @drive: drive to tune
+ *
+ * Program the PIO/MWDMA timing for this channel according to the
+ * current clock.
+ */
+
+static void it821x_program(ide_drive_t *drive, u16 timing)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ int channel = hwif->channel;
+ u8 conf;
+
+ /* Program PIO/MWDMA timing bits */
+ if(itdev->clock_mode == ATA_66)
+ conf = timing >> 8;
+ else
+ conf = timing & 0xFF;
+ pci_write_config_byte(hwif->pci_dev, 0x54 + 4 * channel, conf);
+}
+
+/**
+ * it821x_program_udma - program the UDMA registers
+ * @drive: drive to tune
+ *
+ * Program the UDMA timing for this drive according to the
+ * current clock.
+ */
+
+static void it821x_program_udma(ide_drive_t *drive, u16 timing)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ int channel = hwif->channel;
+ int unit = drive->select.b.unit;
+ u8 conf;
+
+ /* Program UDMA timing bits */
+ if(itdev->clock_mode == ATA_66)
+ conf = timing >> 8;
+ else
+ conf = timing & 0xFF;
+ if(itdev->timing10 == 0)
+ pci_write_config_byte(hwif->pci_dev, 0x56 + 4 * channel + unit, conf);
+ else {
+ pci_write_config_byte(hwif->pci_dev, 0x56 + 4 * channel, conf);
+ pci_write_config_byte(hwif->pci_dev, 0x56 + 4 * channel + 1, conf);
+ }
+}
+
+
+/**
+ * it821x_clock_strategy
+ * @hwif: hardware interface
+ *
+ * Select between the 50 and 66Mhz base clocks to get the best
+ * results for this interface.
+ */
+
+static void it821x_clock_strategy(ide_drive_t *drive)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+
+ u8 unit = drive->select.b.unit;
+ ide_drive_t *pair = &hwif->drives[1-unit];
+
+ int clock, altclock;
+ u8 v;
+ int sel = 0;
+
+ if(itdev->want[0][0] > itdev->want[1][0]) {
+ clock = itdev->want[0][1];
+ altclock = itdev->want[1][1];
+ } else {
+ clock = itdev->want[1][1];
+ altclock = itdev->want[0][1];
+ }
+
+ /* Master doesn't care does the slave ? */
+ if(clock == ATA_ANY)
+ clock = altclock;
+
+ /* Nobody cares - keep the same clock */
+ if(clock == ATA_ANY)
+ return;
+ /* No change */
+ if(clock == itdev->clock_mode)
+ return;
+
+ /* Load this into the controller ? */
+ if(clock == ATA_66)
+ itdev->clock_mode = ATA_66;
+ else {
+ itdev->clock_mode = ATA_50;
+ sel = 1;
+ }
+ pci_read_config_byte(hwif->pci_dev, 0x50, &v);
+ v &= ~(1 << (1 + hwif->channel));
+ v |= sel << (1 + hwif->channel);
+ pci_write_config_byte(hwif->pci_dev, 0x50, v);
+
+ /*
+ * Reprogram the UDMA/PIO of the pair drive for the switch
+ * MWDMA will be dealt with by the dma switcher
+ */
+ if(pair && itdev->udma[1-unit] != UDMA_OFF) {
+ it821x_program_udma(pair, itdev->udma[1-unit]);
+ it821x_program(pair, itdev->pio[1-unit]);
+ }
+ /*
+ * Reprogram the UDMA/PIO of our drive for the switch.
+ * MWDMA will be dealt with by the dma switcher
+ */
+ if(itdev->udma[unit] != UDMA_OFF) {
+ it821x_program_udma(drive, itdev->udma[unit]);
+ it821x_program(drive, itdev->pio[unit]);
+ }
+}
+
+/**
+ * it821x_ratemask - Compute available modes
+ * @drive: IDE drive
+ *
+ * Compute the available speeds for the devices on the interface. This
+ * is all modes to ATA133 clipped by drive cable setup.
+ */
+
+static u8 it821x_ratemask (ide_drive_t *drive)
+{
+ u8 mode = 4;
+ if (!eighty_ninty_three(drive))
+ mode = min(mode, (u8)1);
+ return mode;
+}
+
+/**
+ * it821x_tuneproc - tune a drive
+ * @drive: drive to tune
+ * @mode_wanted: the target operating mode
+ *
+ * Load the timing settings for this device mode into the
+ * controller. By the time we are called the mode has been
+ * modified as neccessary to handle the absence of seperate
+ * master/slave timers for MWDMA/PIO.
+ *
+ * This code is only used in pass through mode.
+ */
+
+static void it821x_tuneproc (ide_drive_t *drive, byte mode_wanted)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ int unit = drive->select.b.unit;
+
+ /* Spec says 89 ref driver uses 88 */
+ static u16 pio[] = { 0xAA88, 0xA382, 0xA181, 0x3332, 0x3121 };
+ static u8 pio_want[] = { ATA_66, ATA_66, ATA_66, ATA_66, ATA_ANY };
+
+ if(itdev->smart)
+ return;
+
+ /* We prefer 66Mhz clock for PIO 0-3, don't care for PIO4 */
+ itdev->want[unit][1] = pio_want[mode_wanted];
+ itdev->want[unit][0] = 1; /* PIO is lowest priority */
+ itdev->pio[unit] = pio[mode_wanted];
+ it821x_clock_strategy(drive);
+ it821x_program(drive, itdev->pio[unit]);
+}
+
+/**
+ * it821x_tune_mwdma - tune a channel for MWDMA
+ * @drive: drive to set up
+ * @mode_wanted: the target operating mode
+ *
+ * Load the timing settings for this device mode into the
+ * controller when doing MWDMA in pass through mode. The caller
+ * must manage the whole lack of per device MWDMA/PIO timings and
+ * the shared MWDMA/PIO timing register.
+ */
+
+static void it821x_tune_mwdma (ide_drive_t *drive, byte mode_wanted)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = (void *)ide_get_hwifdata(hwif);
+ int unit = drive->select.b.unit;
+ int channel = hwif->channel;
+ u8 conf;
+
+ static u16 dma[] = { 0x8866, 0x3222, 0x3121 };
+ static u8 mwdma_want[] = { ATA_ANY, ATA_66, ATA_ANY };
+
+ itdev->want[unit][1] = mwdma_want[mode_wanted];
+ itdev->want[unit][0] = 2; /* MWDMA is low priority */
+ itdev->mwdma[unit] = dma[mode_wanted];
+ itdev->udma[unit] = UDMA_OFF;
+
+ /* UDMA bits off - Revision 0x10 do them in pairs */
+ pci_read_config_byte(hwif->pci_dev, 0x50, &conf);
+ if(itdev->timing10)
+ conf |= channel ? 0x60: 0x18;
+ else
+ conf |= 1 << (3 + 2 * channel + unit);
+ pci_write_config_byte(hwif->pci_dev, 0x50, conf);
+
+ it821x_clock_strategy(drive);
+ /* FIXME: do we need to program this ? */
+ /* it821x_program(drive, itdev->mwdma[unit]); */
+}
+
+/**
+ * it821x_tune_udma - tune a channel for UDMA
+ * @drive: drive to set up
+ * @mode_wanted: the target operating mode
+ *
+ * Load the timing settings for this device mode into the
+ * controller when doing UDMA modes in pass through.
+ */
+
+static void it821x_tune_udma (ide_drive_t *drive, byte mode_wanted)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ int unit = drive->select.b.unit;
+ int channel = hwif->channel;
+ u8 conf;
+
+ static u16 udma[] = { 0x4433, 0x4231, 0x3121, 0x2121, 0x1111, 0x2211, 0x1111 };
+ static u8 udma_want[] = { ATA_ANY, ATA_50, ATA_ANY, ATA_66, ATA_66, ATA_50, ATA_66 };
+
+ itdev->want[unit][1] = udma_want[mode_wanted];
+ itdev->want[unit][0] = 3; /* UDMA is high priority */
+ itdev->mwdma[unit] = MWDMA_OFF;
+ itdev->udma[unit] = udma[mode_wanted];
+ if(mode_wanted >= 5)
+ itdev->udma[unit] |= 0x8080; /* UDMA 5/6 select on */
+
+ /* UDMA on. Again revision 0x10 must do the pair */
+ pci_read_config_byte(hwif->pci_dev, 0x50, &conf);
+ if(itdev->timing10)
+ conf &= channel ? 0x9F: 0xE7;
+ else
+ conf &= ~ (1 << (3 + 2 * channel + unit));
+ pci_write_config_byte(hwif->pci_dev, 0x50, conf);
+
+ it821x_clock_strategy(drive);
+ it821x_program_udma(drive, itdev->udma[unit]);
+
+}
+
+/**
+ * config_it821x_chipset_for_pio - set drive timings
+ * @drive: drive to tune
+ * @speed we want
+ *
+ * Compute the best pio mode we can for a given device. We must
+ * pick a speed that does not cause problems with the other device
+ * on the cable.
+ */
+
+static void config_it821x_chipset_for_pio (ide_drive_t *drive, byte set_speed)
+{
+ u8 unit = drive->select.b.unit;
+ ide_hwif_t *hwif = drive->hwif;
+ ide_drive_t *pair = &hwif->drives[1-unit];
+ u8 speed = 0, set_pio = ide_get_best_pio_mode(drive, 255, 5, NULL);
+ u8 pair_pio;
+
+ /* We have to deal with this mess in pairs */
+ if(pair != NULL) {
+ pair_pio = ide_get_best_pio_mode(pair, 255, 5, NULL);
+ /* Trim PIO to the slowest of the master/slave */
+ if(pair_pio < set_pio)
+ set_pio = pair_pio;
+ }
+ it821x_tuneproc(drive, set_pio);
+ speed = XFER_PIO_0 + set_pio;
+ /* XXX - We trim to the lowest of the pair so the other drive
+ will always be fine at this point until we do hotplug passthru */
+
+ if (set_speed)
+ (void) ide_config_drive_speed(drive, speed);
+}
+
+/**
+ * it821x_dma_read - DMA hook
+ * @drive: drive for DMA
+ *
+ * The IT821x has a single timing register for MWDMA and for PIO
+ * operations. As we flip back and forth we have to reload the
+ * clock. In addition the rev 0x10 device only works if the same
+ * timing value is loaded into the master and slave UDMA clock
+ * so we must also reload that.
+ *
+ * FIXME: we could figure out in advance if we need to do reloads
+ */
+
+static void it821x_dma_start(ide_drive_t *drive)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ int unit = drive->select.b.unit;
+ if(itdev->mwdma[unit] != MWDMA_OFF)
+ it821x_program(drive, itdev->mwdma[unit]);
+ else if(itdev->udma[unit] != UDMA_OFF && itdev->timing10)
+ it821x_program_udma(drive, itdev->udma[unit]);
+ ide_dma_start(drive);
+}
+
+/**
+ * it821x_dma_write - DMA hook
+ * @drive: drive for DMA stop
+ *
+ * The IT821x has a single timing register for MWDMA and for PIO
+ * operations. As we flip back and forth we have to reload the
+ * clock.
+ */
+
+static int it821x_dma_end(ide_drive_t *drive)
+{
+ ide_hwif_t *hwif = drive->hwif;
+ int unit = drive->select.b.unit;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ int ret = __ide_dma_end(drive);
+ if(itdev->mwdma[unit] != MWDMA_OFF)
+ it821x_program(drive, itdev->pio[unit]);
+ return ret;
+}
+
+
+/**
+ * it821x_tune_chipset - set controller timings
+ * @drive: Drive to set up
+ * @xferspeed: speed we want to achieve
+ *
+ * Tune the ITE chipset for the desired mode. If we can't achieve
+ * the desired mode then tune for a lower one, but ultimately
+ * make the thing work.
+ */
+
+static int it821x_tune_chipset (ide_drive_t *drive, byte xferspeed)
+{
+
+ ide_hwif_t *hwif = drive->hwif;
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ u8 speed = ide_rate_filter(it821x_ratemask(drive), xferspeed);
+
+ if(!itdev->smart) {
+ switch(speed) {
+ case XFER_PIO_4:
+ case XFER_PIO_3:
+ case XFER_PIO_2:
+ case XFER_PIO_1:
+ case XFER_PIO_0:
+ it821x_tuneproc(drive, (speed - XFER_PIO_0));
+ break;
+ /* MWDMA tuning is really hard because our MWDMA and PIO
+ timings are kept in the same place. We can switch in the
+ host dma on/off callbacks */
+ case XFER_MW_DMA_2:
+ case XFER_MW_DMA_1:
+ case XFER_MW_DMA_0:
+ it821x_tune_mwdma(drive, (speed - XFER_MW_DMA_0));
+ break;
+ case XFER_UDMA_6:
+ case XFER_UDMA_5:
+ case XFER_UDMA_4:
+ case XFER_UDMA_3:
+ case XFER_UDMA_2:
+ case XFER_UDMA_1:
+ case XFER_UDMA_0:
+ it821x_tune_udma(drive, (speed - XFER_UDMA_0));
+ break;
+ default:
+ return 1;
+ }
+ }
+ /*
+ * In smart mode the clocking is done by the host controller
+ * snooping the mode we picked. The rest of it is not our problem
+ */
+ return ide_config_drive_speed(drive, speed);
+}
+
+/**
+ * config_chipset_for_dma - configure for DMA
+ * @drive: drive to configure
+ *
+ * Called by the IDE layer when it wants the timings set up.
+ */
+
+static int config_chipset_for_dma (ide_drive_t *drive)
+{
+ u8 speed = ide_dma_speed(drive, it821x_ratemask(drive));
+
+ config_it821x_chipset_for_pio(drive, !speed);
+ it821x_tune_chipset(drive, speed);
+ return ide_dma_enable(drive);
+}
+
+/**
+ * it821x_configure_drive_for_dma - set up for DMA transfers
+ * @drive: drive we are going to set up
+ *
+ * Set up the drive for DMA, tune the controller and drive as
+ * required. If the drive isn't suitable for DMA or we hit
+ * other problems then we will drop down to PIO and set up
+ * PIO appropriately
+ */
+
+static int it821x_config_drive_for_dma (ide_drive_t *drive)
+{
+ ide_hwif_t *hwif = drive->hwif;
+
+ if (ide_use_dma(drive)) {
+ if (config_chipset_for_dma(drive))
+ return hwif->ide_dma_on(drive);
+ }
+ config_it821x_chipset_for_pio(drive, 1);
+ return hwif->ide_dma_off_quietly(drive);
+}
+
+/**
+ * ata66_it821x - check for 80 pin cable
+ * @hwif: interface to check
+ *
+ * Check for the presence of an ATA66 capable cable on the
+ * interface. Problematic as it seems some cards don't have
+ * the needed logic onboard.
+ */
+
+static unsigned int __devinit ata66_it821x(ide_hwif_t *hwif)
+{
+ /* The reference driver also only does disk side */
+ return 1;
+}
+
+/**
+ * it821x_fixup - post init callback
+ * @hwif: interface
+ *
+ * This callback is run after the drives have been probed but
+ * before anything gets attached. It allows drivers to do any
+ * final tuning that is needed, or fixups to work around bugs.
+ */
+
+static void __devinit it821x_fixups(ide_hwif_t *hwif)
+{
+ struct it821x_dev *itdev = ide_get_hwifdata(hwif);
+ int i;
+
+ if(!itdev->smart) {
+ /*
+ * If we are in pass through mode then not much
+ * needs to be done, but we do bother to clear the
+ * IRQ mask as we may well be in PIO (eg rev 0x10)
+ * for now and we know unmasking is safe on this chipset.
+ */
+ for (i = 0; i < 2; i++) {
+ ide_drive_t *drive = &hwif->drives[i];
+ if(drive->present)
+ drive->unmask = 1;
+ }
+ return;
+ }
+ /*
+ * Perform fixups on smart mode. We need to "lose" some
+ * capabilities the firmware lacks but does not filter, and
+ * also patch up some capability bits that it forgets to set
+ * in RAID mode.
+ */
+
+ for(i = 0; i < 2; i++) {
+ ide_drive_t *drive = &hwif->drives[i];
+ struct hd_driveid *id;
+ u16 *idbits;
+
+ if(!drive->present)
+ continue;
+ id = drive->id;
+ idbits = (u16 *)drive->id;
+
+ /* Check for RAID v native */
+ if(strstr(id->model, "Integrated Technology Express")) {
+ /* In raid mode the ident block is slightly buggy
+ We need to set the bits so that the IDE layer knows
+ LBA28. LBA48 and DMA ar valid */
+ id->capability |= 3; /* LBA28, DMA */
+ id->command_set_2 |= 0x0400; /* LBA48 valid */
+ id->cfs_enable_2 |= 0x0400; /* LBA48 on */
+ /* Reporting logic */
+ printk(KERN_INFO "%s: IT8212 %sRAID %d volume",
+ drive->name,
+ idbits[147] ? "Bootable ":"",
+ idbits[129]);
+ if(idbits[129] != 1)
+ printk("(%dK stripe)", idbits[146]);
+ printk(".\n");
+ /* Now the core code will have wrongly decided no DMA
+ so we need to fix this */
+ hwif->ide_dma_off_quietly(drive);
+#ifdef CONFIG_IDEDMA_ONLYDISK
+ if (drive->media == ide_disk)
+#endif
+ hwif->ide_dma_check(drive);
+ } else {
+ /* Non RAID volume. Fixups to stop the core code
+ doing unsupported things */
+ id->field_valid &= 1;
+ id->queue_depth = 0;
+ id->command_set_1 = 0;
+ id->command_set_2 &= 0xC400;
+ id->cfsse &= 0xC000;
+ id->cfs_enable_1 = 0;
+ id->cfs_enable_2 &= 0xC400;
+ id->csf_default &= 0xC000;
+ id->word127 = 0;
+ id->dlf = 0;
+ id->csfo = 0;
+ id->cfa_power = 0;
+ printk(KERN_INFO "%s: Performing identify fixups.\n",
+ drive->name);
+ }
+ }
+
+}
+
+/**
+ * init_hwif_it821x - set up hwif structs
+ * @hwif: interface to set up
+ *
+ * We do the basic set up of the interface structure. The IT8212
+ * requires several custom handlers so we override the default
+ * ide DMA handlers appropriately
+ */
+
+static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
+{
+ struct it821x_dev *idev = kmalloc(sizeof(struct it821x_dev), GFP_KERNEL);
+ u8 conf;
+
+ if(idev == NULL) {
+ printk(KERN_ERR "it821x: out of memory, falling back to legacy behaviour.\n");
+ goto fallback;
+ }
+ memset(idev, 0, sizeof(struct it821x_dev));
+ ide_set_hwifdata(hwif, idev);
+
+ pci_read_config_byte(hwif->pci_dev, 0x50, &conf);
+ if(conf & 1) {
+ idev->smart = 1;
+ hwif->atapi_dma = 0;
+ /* Long I/O's although allowed in LBA48 space cause the
+ onboard firmware to enter the twighlight zone */
+ hwif->rqsize = 256;
+ }
+
+ /* Pull the current clocks from 0x50 also */
+ if (conf & (1 << (1 + hwif->channel)))
+ idev->clock_mode = ATA_50;
+ else
+ idev->clock_mode = ATA_66;
+
+ idev->want[0][1] = ATA_ANY;
+ idev->want[1][1] = ATA_ANY;
+
+ /*
+ * Not in the docs but according to the reference driver
+ * this is neccessary.
+ */
+
+ pci_read_config_byte(hwif->pci_dev, 0x08, &conf);
+ if(conf == 0x10) {
+ idev->timing10 = 1;
+ hwif->atapi_dma = 0;
+ if(!idev->smart)
+ printk(KERN_WARNING "it821x: Revision 0x10, workarounds activated.\n");
+ }
+
+ hwif->speedproc = &it821x_tune_chipset;
+ hwif->tuneproc = &it821x_tuneproc;
+
+ /* MWDMA/PIO clock switching for pass through mode */
+ if(!idev->smart) {
+ hwif->dma_start = &it821x_dma_start;
+ hwif->ide_dma_end = &it821x_dma_end;
+ }
+
+ hwif->drives[0].autotune = 1;
+ hwif->drives[1].autotune = 1;
+
+ if (!hwif->dma_base)
+ goto fallback;
+
+ hwif->ultra_mask = 0x7f;
+ hwif->mwdma_mask = 0x07;
+ hwif->swdma_mask = 0x07;
+
+ hwif->ide_dma_check = &it821x_config_drive_for_dma;
+ if (!(hwif->udma_four))
+ hwif->udma_four = ata66_it821x(hwif);
+
+ /*
+ * The BIOS often doesn't set up DMA on this controller
+ * so we always do it.
+ */
+
+ hwif->autodma = 1;
+ hwif->drives[0].autodma = hwif->autodma;
+ hwif->drives[1].autodma = hwif->autodma;
+ return;
+fallback:
+ hwif->autodma = 0;
+ return;
+}
+
+static void __devinit it8212_disable_raid(struct pci_dev *dev)
+{
+ /* Reset local CPU, and set BIOS not ready */
+ pci_write_config_byte(dev, 0x5E, 0x01);
+
+ /* Set to bypass mode, and reset PCI bus */
+ pci_write_config_byte(dev, 0x50, 0x00);
+ pci_write_config_word(dev, PCI_COMMAND,
+ PCI_COMMAND_PARITY | PCI_COMMAND_IO |
+ PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER);
+ pci_write_config_word(dev, 0x40, 0xA0F3);
+
+ pci_write_config_dword(dev,0x4C, 0x02040204);
+ pci_write_config_byte(dev, 0x42, 0x36);
+ pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0);
+}
+
+static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev, const char *name)
+{
+ u8 conf;
+ static char *mode[2] = { "pass through", "smart" };
+
+ /* Force the card into bypass mode if so requested */
+ if (it8212_noraid) {
+ printk(KERN_INFO "it8212: forcing bypass mode.\n");
+ it8212_disable_raid(dev);
+ }
+ pci_read_config_byte(dev, 0x50, &conf);
+ printk(KERN_INFO "it821x: controller in %s mode.\n", mode[conf & 1]);
+ return 0;
+}
+
+
+#define DECLARE_ITE_DEV(name_str) \
+ { \
+ .name = name_str, \
+ .init_chipset = init_chipset_it821x, \
+ .init_hwif = init_hwif_it821x, \
+ .channels = 2, \
+ .autodma = AUTODMA, \
+ .bootable = ON_BOARD, \
+ .fixup = it821x_fixups \
+ }
+
+static ide_pci_device_t it821x_chipsets[] __devinitdata = {
+ /* 0 */ DECLARE_ITE_DEV("IT8212"),
+};
+
+/**
+ * it821x_init_one - pci layer discovery entry
+ * @dev: PCI device
+ * @id: ident table entry
+ *
+ * Called by the PCI code when it finds an ITE821x controller.
+ * We then use the IDE PCI generic helper to do most of the work.
+ */
+
+static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+ ide_setup_pci_device(dev, &it821x_chipsets[id->driver_data]);
+ return 0;
+}
+
+static struct pci_device_id it821x_pci_tbl[] = {
+ { PCI_VENDOR_ID_ITE, PCI_DEVICE_ID_ITE_8211, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+ { PCI_VENDOR_ID_ITE, PCI_DEVICE_ID_ITE_8212, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+ { 0, },
+};
+
+MODULE_DEVICE_TABLE(pci, it821x_pci_tbl);
+
+static struct pci_driver driver = {
+ .name = "ITE821x IDE",
+ .id_table = it821x_pci_tbl,
+ .probe = it821x_init_one,
+};
+
+static int __init it821x_ide_init(void)
+{
+ return ide_pci_register_driver(&driver);
+}
+
+module_init(it821x_ide_init);
+
+module_param_named(noraid, it8212_noraid, int, S_IRUGO);
+MODULE_PARM_DESC(it8212_noraid, "Force card into bypass mode");
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("PCI driver module for the ITE 821x");
+MODULE_LICENSE("GPL");
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index 82a1103b2413..c6f5fa4b4ca6 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -442,7 +442,7 @@ static unsigned int __devinit init_chipset_svwks (struct pci_dev *dev, const cha
return (dev->irq) ? dev->irq : 0;
}
-static unsigned int __init ata66_svwks_svwks (ide_hwif_t *hwif)
+static unsigned int __devinit ata66_svwks_svwks (ide_hwif_t *hwif)
{
return 1;
}
@@ -454,7 +454,7 @@ static unsigned int __init ata66_svwks_svwks (ide_hwif_t *hwif)
* Bit 14 clear = primary IDE channel does not have 80-pin cable.
* Bit 14 set = primary IDE channel has 80-pin cable.
*/
-static unsigned int __init ata66_svwks_dell (ide_hwif_t *hwif)
+static unsigned int __devinit ata66_svwks_dell (ide_hwif_t *hwif)
{
struct pci_dev *dev = hwif->pci_dev;
if (dev->subsystem_vendor == PCI_VENDOR_ID_DELL &&
@@ -472,7 +472,7 @@ static unsigned int __init ata66_svwks_dell (ide_hwif_t *hwif)
*
* WARNING: this only works on Alpine hardware!
*/
-static unsigned int __init ata66_svwks_cobalt (ide_hwif_t *hwif)
+static unsigned int __devinit ata66_svwks_cobalt (ide_hwif_t *hwif)
{
struct pci_dev *dev = hwif->pci_dev;
if (dev->subsystem_vendor == PCI_VENDOR_ID_SUN &&
@@ -483,7 +483,7 @@ static unsigned int __init ata66_svwks_cobalt (ide_hwif_t *hwif)
return 0;
}
-static unsigned int __init ata66_svwks (ide_hwif_t *hwif)
+static unsigned int __devinit ata66_svwks (ide_hwif_t *hwif)
{
struct pci_dev *dev = hwif->pci_dev;
@@ -573,7 +573,7 @@ static int __devinit init_setup_svwks (struct pci_dev *dev, ide_pci_device_t *d)
return ide_setup_pci_device(dev, d);
}
-static int __init init_setup_csb6 (struct pci_dev *dev, ide_pci_device_t *d)
+static int __devinit init_setup_csb6 (struct pci_dev *dev, ide_pci_device_t *d)
{
if (!(PCI_FUNC(dev->devfn) & 1)) {
d->bootable = NEVER_BOARD;
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 569f16767442..818380b5fd27 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1324,9 +1324,9 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
/* XXX FIXME: Media bay stuff need re-organizing */
if (np->parent && np->parent->name
&& strcasecmp(np->parent->name, "media-bay") == 0) {
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PMAC_MEDIABAY
media_bay_set_ide_infos(np->parent, pmif->regbase, pmif->irq, hwif->index);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PMAC_MEDIABAY */
pmif->mediabay = 1;
if (!bidp)
pmif->aapl_bus_id = 1;
@@ -1382,10 +1382,10 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
hwif->index, model_name[pmif->kind], pmif->aapl_bus_id,
pmif->mediabay ? " (mediabay)" : "", hwif->irq);
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PMAC_MEDIABAY
if (pmif->mediabay && check_media_bay_by_base(pmif->regbase, MB_CD) == 0)
hwif->noprobe = 0;
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PMAC_MEDIABAY */
hwif->sg_max_nents = MAX_DCMDS;
diff --git a/drivers/ieee1394/ohci1394.c b/drivers/ieee1394/ohci1394.c
index 36e25ac823dc..b3d3d22fde64 100644
--- a/drivers/ieee1394/ohci1394.c
+++ b/drivers/ieee1394/ohci1394.c
@@ -3538,8 +3538,8 @@ static void ohci1394_pci_remove(struct pci_dev *pdev)
static int ohci1394_pci_resume (struct pci_dev *pdev)
{
-#ifdef CONFIG_PMAC_PBOOK
- {
+#ifdef CONFIG_PPC_PMAC
+ if (_machine == _MACH_Pmac) {
struct device_node *of_node;
/* Re-enable 1394 */
@@ -3547,7 +3547,7 @@ static int ohci1394_pci_resume (struct pci_dev *pdev)
if (of_node)
pmac_call_feature (PMAC_FTR_1394_ENABLE, of_node, 0, 1);
}
-#endif
+#endif /* CONFIG_PPC_PMAC */
pci_enable_device(pdev);
@@ -3557,8 +3557,8 @@ static int ohci1394_pci_resume (struct pci_dev *pdev)
static int ohci1394_pci_suspend (struct pci_dev *pdev, pm_message_t state)
{
-#ifdef CONFIG_PMAC_PBOOK
- {
+#ifdef CONFIG_PPC_PMAC
+ if (_machine == _MACH_Pmac) {
struct device_node *of_node;
/* Disable 1394 */
diff --git a/drivers/infiniband/core/packer.c b/drivers/infiniband/core/packer.c
index 5f15feffeae2..eb5ff54c10d7 100644
--- a/drivers/infiniband/core/packer.c
+++ b/drivers/infiniband/core/packer.c
@@ -96,7 +96,7 @@ void ib_pack(const struct ib_field *desc,
else
val = 0;
- mask = cpu_to_be64(((1ull << desc[i].size_bits) - 1) << shift);
+ mask = cpu_to_be64((~0ull >> (64 - desc[i].size_bits)) << shift);
addr = (__be64 *) ((__be32 *) buf + desc[i].offset_words);
*addr = (*addr & ~mask) | (cpu_to_be64(val) & mask);
} else {
@@ -176,7 +176,7 @@ void ib_unpack(const struct ib_field *desc,
__be64 *addr;
shift = 64 - desc[i].offset_bits - desc[i].size_bits;
- mask = ((1ull << desc[i].size_bits) - 1) << shift;
+ mask = (~0ull >> (64 - desc[i].size_bits)) << shift;
addr = (__be64 *) buf + desc[i].offset_words;
val = (be64_to_cpup(addr) & mask) >> shift;
value_write(desc[i].struct_offset_bytes,
diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index 276e1a53010d..5a08e81fa827 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -507,7 +507,13 @@ retry:
spin_unlock_irqrestore(&idr_lock, flags);
}
- return ret;
+ /*
+ * It's not safe to dereference query any more, because the
+ * send may already have completed and freed the query in
+ * another context. So use wr.wr_id, which has a copy of the
+ * query's id.
+ */
+ return ret ? ret : wr.wr_id;
}
static void ib_sa_path_rec_callback(struct ib_sa_query *sa_query,
@@ -598,14 +604,15 @@ int ib_sa_path_rec_get(struct ib_device *device, u8 port_num,
rec, query->sa_query.mad->data);
*sa_query = &query->sa_query;
+
ret = send_mad(&query->sa_query, timeout_ms);
- if (ret) {
+ if (ret < 0) {
*sa_query = NULL;
kfree(query->sa_query.mad);
kfree(query);
}
- return ret ? ret : query->sa_query.id;
+ return ret;
}
EXPORT_SYMBOL(ib_sa_path_rec_get);
@@ -674,14 +681,15 @@ int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num,
rec, query->sa_query.mad->data);
*sa_query = &query->sa_query;
+
ret = send_mad(&query->sa_query, timeout_ms);
- if (ret) {
+ if (ret < 0) {
*sa_query = NULL;
kfree(query->sa_query.mad);
kfree(query);
}
- return ret ? ret : query->sa_query.id;
+ return ret;
}
EXPORT_SYMBOL(ib_sa_mcmember_rec_query);
diff --git a/drivers/infiniband/hw/mthca/mthca_av.c b/drivers/infiniband/hw/mthca/mthca_av.c
index 085baf393ca4..d58dcbe66488 100644
--- a/drivers/infiniband/hw/mthca/mthca_av.c
+++ b/drivers/infiniband/hw/mthca/mthca_av.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2004 Topspin Communications. All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c
index cd9ed958d92f..1557a522d831 100644
--- a/drivers/infiniband/hw/mthca/mthca_cmd.c
+++ b/drivers/infiniband/hw/mthca/mthca_cmd.c
@@ -431,6 +431,36 @@ static int mthca_cmd_imm(struct mthca_dev *dev,
timeout, status);
}
+int mthca_cmd_init(struct mthca_dev *dev)
+{
+ sema_init(&dev->cmd.hcr_sem, 1);
+ sema_init(&dev->cmd.poll_sem, 1);
+ dev->cmd.use_events = 0;
+
+ dev->hcr = ioremap(pci_resource_start(dev->pdev, 0) + MTHCA_HCR_BASE,
+ MTHCA_HCR_SIZE);
+ if (!dev->hcr) {
+ mthca_err(dev, "Couldn't map command register.");
+ return -ENOMEM;
+ }
+
+ dev->cmd.pool = pci_pool_create("mthca_cmd", dev->pdev,
+ MTHCA_MAILBOX_SIZE,
+ MTHCA_MAILBOX_SIZE, 0);
+ if (!dev->cmd.pool) {
+ iounmap(dev->hcr);
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+void mthca_cmd_cleanup(struct mthca_dev *dev)
+{
+ pci_pool_destroy(dev->cmd.pool);
+ iounmap(dev->hcr);
+}
+
/*
* Switch to using events to issue FW commands (should be called after
* event queue to command events has been initialized).
@@ -489,6 +519,33 @@ void mthca_cmd_use_polling(struct mthca_dev *dev)
up(&dev->cmd.poll_sem);
}
+struct mthca_mailbox *mthca_alloc_mailbox(struct mthca_dev *dev,
+ unsigned int gfp_mask)
+{
+ struct mthca_mailbox *mailbox;
+
+ mailbox = kmalloc(sizeof *mailbox, gfp_mask);
+ if (!mailbox)
+ return ERR_PTR(-ENOMEM);
+
+ mailbox->buf = pci_pool_alloc(dev->cmd.pool, gfp_mask, &mailbox->dma);
+ if (!mailbox->buf) {
+ kfree(mailbox);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ return mailbox;
+}
+
+void mthca_free_mailbox(struct mthca_dev *dev, struct mthca_mailbox *mailbox)
+{
+ if (!mailbox)
+ return;
+
+ pci_pool_free(dev->cmd.pool, mailbox->buf, mailbox->dma);
+ kfree(mailbox);
+}
+
int mthca_SYS_EN(struct mthca_dev *dev, u8 *status)
{
u64 out;
@@ -513,20 +570,20 @@ int mthca_SYS_DIS(struct mthca_dev *dev, u8 *status)
static int mthca_map_cmd(struct mthca_dev *dev, u16 op, struct mthca_icm *icm,
u64 virt, u8 *status)
{
- u32 *inbox;
- dma_addr_t indma;
+ struct mthca_mailbox *mailbox;
struct mthca_icm_iter iter;
+ __be64 *pages;
int lg;
int nent = 0;
int i;
int err = 0;
int ts = 0, tc = 0;
- inbox = pci_alloc_consistent(dev->pdev, PAGE_SIZE, &indma);
- if (!inbox)
- return -ENOMEM;
-
- memset(inbox, 0, PAGE_SIZE);
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ memset(mailbox->buf, 0, MTHCA_MAILBOX_SIZE);
+ pages = mailbox->buf;
for (mthca_icm_first(icm, &iter);
!mthca_icm_last(&iter);
@@ -546,19 +603,17 @@ static int mthca_map_cmd(struct mthca_dev *dev, u16 op, struct mthca_icm *icm,
}
for (i = 0; i < mthca_icm_size(&iter) / (1 << lg); ++i, ++nent) {
if (virt != -1) {
- *((__be64 *) (inbox + nent * 4)) =
- cpu_to_be64(virt);
+ pages[nent * 2] = cpu_to_be64(virt);
virt += 1 << lg;
}
- *((__be64 *) (inbox + nent * 4 + 2)) =
- cpu_to_be64((mthca_icm_addr(&iter) +
- (i << lg)) | (lg - 12));
+ pages[nent * 2 + 1] = cpu_to_be64((mthca_icm_addr(&iter) +
+ (i << lg)) | (lg - 12));
ts += 1 << (lg - 10);
++tc;
- if (nent == PAGE_SIZE / 16) {
- err = mthca_cmd(dev, indma, nent, 0, op,
+ if (nent == MTHCA_MAILBOX_SIZE / 16) {
+ err = mthca_cmd(dev, mailbox->dma, nent, 0, op,
CMD_TIME_CLASS_B, status);
if (err || *status)
goto out;
@@ -568,7 +623,7 @@ static int mthca_map_cmd(struct mthca_dev *dev, u16 op, struct mthca_icm *icm,
}
if (nent)
- err = mthca_cmd(dev, indma, nent, 0, op,
+ err = mthca_cmd(dev, mailbox->dma, nent, 0, op,
CMD_TIME_CLASS_B, status);
switch (op) {
@@ -585,7 +640,7 @@ static int mthca_map_cmd(struct mthca_dev *dev, u16 op, struct mthca_icm *icm,
}
out:
- pci_free_consistent(dev->pdev, PAGE_SIZE, inbox, indma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
@@ -606,8 +661,8 @@ int mthca_RUN_FW(struct mthca_dev *dev, u8 *status)
int mthca_QUERY_FW(struct mthca_dev *dev, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u32 *outbox;
- dma_addr_t outdma;
int err = 0;
u8 lg;
@@ -625,12 +680,12 @@ int mthca_QUERY_FW(struct mthca_dev *dev, u8 *status)
#define QUERY_FW_EQ_ARM_BASE_OFFSET 0x40
#define QUERY_FW_EQ_SET_CI_BASE_OFFSET 0x48
- outbox = pci_alloc_consistent(dev->pdev, QUERY_FW_OUT_SIZE, &outdma);
- if (!outbox) {
- return -ENOMEM;
- }
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ outbox = mailbox->buf;
- err = mthca_cmd_box(dev, 0, outdma, 0, 0, CMD_QUERY_FW,
+ err = mthca_cmd_box(dev, 0, mailbox->dma, 0, 0, CMD_QUERY_FW,
CMD_TIME_CLASS_A, status);
if (err)
@@ -681,15 +736,15 @@ int mthca_QUERY_FW(struct mthca_dev *dev, u8 *status)
}
out:
- pci_free_consistent(dev->pdev, QUERY_FW_OUT_SIZE, outbox, outdma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
int mthca_ENABLE_LAM(struct mthca_dev *dev, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u8 info;
u32 *outbox;
- dma_addr_t outdma;
int err = 0;
#define ENABLE_LAM_OUT_SIZE 0x100
@@ -700,11 +755,12 @@ int mthca_ENABLE_LAM(struct mthca_dev *dev, u8 *status)
#define ENABLE_LAM_INFO_HIDDEN_FLAG (1 << 4)
#define ENABLE_LAM_INFO_ECC_MASK 0x3
- outbox = pci_alloc_consistent(dev->pdev, ENABLE_LAM_OUT_SIZE, &outdma);
- if (!outbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ outbox = mailbox->buf;
- err = mthca_cmd_box(dev, 0, outdma, 0, 0, CMD_ENABLE_LAM,
+ err = mthca_cmd_box(dev, 0, mailbox->dma, 0, 0, CMD_ENABLE_LAM,
CMD_TIME_CLASS_C, status);
if (err)
@@ -733,7 +789,7 @@ int mthca_ENABLE_LAM(struct mthca_dev *dev, u8 *status)
(unsigned long long) dev->ddr_end);
out:
- pci_free_consistent(dev->pdev, ENABLE_LAM_OUT_SIZE, outbox, outdma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
@@ -744,9 +800,9 @@ int mthca_DISABLE_LAM(struct mthca_dev *dev, u8 *status)
int mthca_QUERY_DDR(struct mthca_dev *dev, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u8 info;
u32 *outbox;
- dma_addr_t outdma;
int err = 0;
#define QUERY_DDR_OUT_SIZE 0x100
@@ -757,11 +813,12 @@ int mthca_QUERY_DDR(struct mthca_dev *dev, u8 *status)
#define QUERY_DDR_INFO_HIDDEN_FLAG (1 << 4)
#define QUERY_DDR_INFO_ECC_MASK 0x3
- outbox = pci_alloc_consistent(dev->pdev, QUERY_DDR_OUT_SIZE, &outdma);
- if (!outbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ outbox = mailbox->buf;
- err = mthca_cmd_box(dev, 0, outdma, 0, 0, CMD_QUERY_DDR,
+ err = mthca_cmd_box(dev, 0, mailbox->dma, 0, 0, CMD_QUERY_DDR,
CMD_TIME_CLASS_A, status);
if (err)
@@ -787,15 +844,15 @@ int mthca_QUERY_DDR(struct mthca_dev *dev, u8 *status)
(unsigned long long) dev->ddr_end);
out:
- pci_free_consistent(dev->pdev, QUERY_DDR_OUT_SIZE, outbox, outdma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
int mthca_QUERY_DEV_LIM(struct mthca_dev *dev,
struct mthca_dev_lim *dev_lim, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u32 *outbox;
- dma_addr_t outdma;
u8 field;
u16 size;
int err;
@@ -860,11 +917,12 @@ int mthca_QUERY_DEV_LIM(struct mthca_dev *dev,
#define QUERY_DEV_LIM_LAMR_OFFSET 0x9f
#define QUERY_DEV_LIM_MAX_ICM_SZ_OFFSET 0xa0
- outbox = pci_alloc_consistent(dev->pdev, QUERY_DEV_LIM_OUT_SIZE, &outdma);
- if (!outbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ outbox = mailbox->buf;
- err = mthca_cmd_box(dev, 0, outdma, 0, 0, CMD_QUERY_DEV_LIM,
+ err = mthca_cmd_box(dev, 0, mailbox->dma, 0, 0, CMD_QUERY_DEV_LIM,
CMD_TIME_CLASS_A, status);
if (err)
@@ -1020,15 +1078,15 @@ int mthca_QUERY_DEV_LIM(struct mthca_dev *dev,
}
out:
- pci_free_consistent(dev->pdev, QUERY_DEV_LIM_OUT_SIZE, outbox, outdma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
int mthca_QUERY_ADAPTER(struct mthca_dev *dev,
struct mthca_adapter *adapter, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u32 *outbox;
- dma_addr_t outdma;
int err;
#define QUERY_ADAPTER_OUT_SIZE 0x100
@@ -1037,23 +1095,24 @@ int mthca_QUERY_ADAPTER(struct mthca_dev *dev,
#define QUERY_ADAPTER_REVISION_ID_OFFSET 0x08
#define QUERY_ADAPTER_INTA_PIN_OFFSET 0x10
- outbox = pci_alloc_consistent(dev->pdev, QUERY_ADAPTER_OUT_SIZE, &outdma);
- if (!outbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ outbox = mailbox->buf;
- err = mthca_cmd_box(dev, 0, outdma, 0, 0, CMD_QUERY_ADAPTER,
+ err = mthca_cmd_box(dev, 0, mailbox->dma, 0, 0, CMD_QUERY_ADAPTER,
CMD_TIME_CLASS_A, status);
if (err)
goto out;
- MTHCA_GET(adapter->vendor_id, outbox, QUERY_ADAPTER_VENDOR_ID_OFFSET);
- MTHCA_GET(adapter->device_id, outbox, QUERY_ADAPTER_DEVICE_ID_OFFSET);
+ MTHCA_GET(adapter->vendor_id, outbox, QUERY_ADAPTER_VENDOR_ID_OFFSET);
+ MTHCA_GET(adapter->device_id, outbox, QUERY_ADAPTER_DEVICE_ID_OFFSET);
MTHCA_GET(adapter->revision_id, outbox, QUERY_ADAPTER_REVISION_ID_OFFSET);
- MTHCA_GET(adapter->inta_pin, outbox, QUERY_ADAPTER_INTA_PIN_OFFSET);
+ MTHCA_GET(adapter->inta_pin, outbox, QUERY_ADAPTER_INTA_PIN_OFFSET);
out:
- pci_free_consistent(dev->pdev, QUERY_DEV_LIM_OUT_SIZE, outbox, outdma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
@@ -1061,8 +1120,8 @@ int mthca_INIT_HCA(struct mthca_dev *dev,
struct mthca_init_hca_param *param,
u8 *status)
{
+ struct mthca_mailbox *mailbox;
u32 *inbox;
- dma_addr_t indma;
int err;
#define INIT_HCA_IN_SIZE 0x200
@@ -1102,9 +1161,10 @@ int mthca_INIT_HCA(struct mthca_dev *dev,
#define INIT_HCA_UAR_SCATCH_BASE_OFFSET (INIT_HCA_UAR_OFFSET + 0x10)
#define INIT_HCA_UAR_CTX_BASE_OFFSET (INIT_HCA_UAR_OFFSET + 0x18)
- inbox = pci_alloc_consistent(dev->pdev, INIT_HCA_IN_SIZE, &indma);
- if (!inbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ inbox = mailbox->buf;
memset(inbox, 0, INIT_HCA_IN_SIZE);
@@ -1167,10 +1227,9 @@ int mthca_INIT_HCA(struct mthca_dev *dev,
MTHCA_PUT(inbox, param->uarc_base, INIT_HCA_UAR_CTX_BASE_OFFSET);
}
- err = mthca_cmd(dev, indma, 0, 0, CMD_INIT_HCA,
- HZ, status);
+ err = mthca_cmd(dev, mailbox->dma, 0, 0, CMD_INIT_HCA, HZ, status);
- pci_free_consistent(dev->pdev, INIT_HCA_IN_SIZE, inbox, indma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
@@ -1178,8 +1237,8 @@ int mthca_INIT_IB(struct mthca_dev *dev,
struct mthca_init_ib_param *param,
int port, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u32 *inbox;
- dma_addr_t indma;
int err;
u32 flags;
@@ -1199,9 +1258,10 @@ int mthca_INIT_IB(struct mthca_dev *dev,
#define INIT_IB_NODE_GUID_OFFSET 0x18
#define INIT_IB_SI_GUID_OFFSET 0x20
- inbox = pci_alloc_consistent(dev->pdev, INIT_IB_IN_SIZE, &indma);
- if (!inbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ inbox = mailbox->buf;
memset(inbox, 0, INIT_IB_IN_SIZE);
@@ -1221,10 +1281,10 @@ int mthca_INIT_IB(struct mthca_dev *dev,
MTHCA_PUT(inbox, param->node_guid, INIT_IB_NODE_GUID_OFFSET);
MTHCA_PUT(inbox, param->si_guid, INIT_IB_SI_GUID_OFFSET);
- err = mthca_cmd(dev, indma, port, 0, CMD_INIT_IB,
+ err = mthca_cmd(dev, mailbox->dma, port, 0, CMD_INIT_IB,
CMD_TIME_CLASS_A, status);
- pci_free_consistent(dev->pdev, INIT_HCA_IN_SIZE, inbox, indma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
@@ -1241,8 +1301,8 @@ int mthca_CLOSE_HCA(struct mthca_dev *dev, int panic, u8 *status)
int mthca_SET_IB(struct mthca_dev *dev, struct mthca_set_ib_param *param,
int port, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u32 *inbox;
- dma_addr_t indma;
int err;
u32 flags = 0;
@@ -1253,9 +1313,10 @@ int mthca_SET_IB(struct mthca_dev *dev, struct mthca_set_ib_param *param,
#define SET_IB_CAP_MASK_OFFSET 0x04
#define SET_IB_SI_GUID_OFFSET 0x08
- inbox = pci_alloc_consistent(dev->pdev, SET_IB_IN_SIZE, &indma);
- if (!inbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ inbox = mailbox->buf;
memset(inbox, 0, SET_IB_IN_SIZE);
@@ -1266,10 +1327,10 @@ int mthca_SET_IB(struct mthca_dev *dev, struct mthca_set_ib_param *param,
MTHCA_PUT(inbox, param->cap_mask, SET_IB_CAP_MASK_OFFSET);
MTHCA_PUT(inbox, param->si_guid, SET_IB_SI_GUID_OFFSET);
- err = mthca_cmd(dev, indma, port, 0, CMD_SET_IB,
+ err = mthca_cmd(dev, mailbox->dma, port, 0, CMD_SET_IB,
CMD_TIME_CLASS_B, status);
- pci_free_consistent(dev->pdev, INIT_HCA_IN_SIZE, inbox, indma);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
@@ -1280,20 +1341,22 @@ int mthca_MAP_ICM(struct mthca_dev *dev, struct mthca_icm *icm, u64 virt, u8 *st
int mthca_MAP_ICM_page(struct mthca_dev *dev, u64 dma_addr, u64 virt, u8 *status)
{
+ struct mthca_mailbox *mailbox;
u64 *inbox;
- dma_addr_t indma;
int err;
- inbox = pci_alloc_consistent(dev->pdev, 16, &indma);
- if (!inbox)
- return -ENOMEM;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ inbox = mailbox->buf;
inbox[0] = cpu_to_be64(virt);
inbox[1] = cpu_to_be64(dma_addr);
- err = mthca_cmd(dev, indma, 1, 0, CMD_MAP_ICM, CMD_TIME_CLASS_B, status);
+ err = mthca_cmd(dev, mailbox->dma, 1, 0, CMD_MAP_ICM,
+ CMD_TIME_CLASS_B, status);
- pci_free_consistent(dev->pdev, 16, inbox, indma);
+ mthca_free_mailbox(dev, mailbox);
if (!err)
mthca_dbg(dev, "Mapped page at %llx to %llx for ICM.\n",
@@ -1338,69 +1401,26 @@ int mthca_SET_ICM_SIZE(struct mthca_dev *dev, u64 icm_size, u64 *aux_pages,
return 0;
}
-int mthca_SW2HW_MPT(struct mthca_dev *dev, void *mpt_entry,
+int mthca_SW2HW_MPT(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int mpt_index, u8 *status)
{
- dma_addr_t indma;
- int err;
-
- indma = pci_map_single(dev->pdev, mpt_entry,
- MTHCA_MPT_ENTRY_SIZE,
- PCI_DMA_TODEVICE);
- if (pci_dma_mapping_error(indma))
- return -ENOMEM;
-
- err = mthca_cmd(dev, indma, mpt_index, 0, CMD_SW2HW_MPT,
- CMD_TIME_CLASS_B, status);
-
- pci_unmap_single(dev->pdev, indma,
- MTHCA_MPT_ENTRY_SIZE, PCI_DMA_TODEVICE);
- return err;
+ return mthca_cmd(dev, mailbox->dma, mpt_index, 0, CMD_SW2HW_MPT,
+ CMD_TIME_CLASS_B, status);
}
-int mthca_HW2SW_MPT(struct mthca_dev *dev, void *mpt_entry,
+int mthca_HW2SW_MPT(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int mpt_index, u8 *status)
{
- dma_addr_t outdma = 0;
- int err;
-
- if (mpt_entry) {
- outdma = pci_map_single(dev->pdev, mpt_entry,
- MTHCA_MPT_ENTRY_SIZE,
- PCI_DMA_FROMDEVICE);
- if (pci_dma_mapping_error(outdma))
- return -ENOMEM;
- }
-
- err = mthca_cmd_box(dev, 0, outdma, mpt_index, !mpt_entry,
- CMD_HW2SW_MPT,
- CMD_TIME_CLASS_B, status);
-
- if (mpt_entry)
- pci_unmap_single(dev->pdev, outdma,
- MTHCA_MPT_ENTRY_SIZE,
- PCI_DMA_FROMDEVICE);
- return err;
+ return mthca_cmd_box(dev, 0, mailbox ? mailbox->dma : 0, mpt_index,
+ !mailbox, CMD_HW2SW_MPT,
+ CMD_TIME_CLASS_B, status);
}
-int mthca_WRITE_MTT(struct mthca_dev *dev, u64 *mtt_entry,
+int mthca_WRITE_MTT(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int num_mtt, u8 *status)
{
- dma_addr_t indma;
- int err;
-
- indma = pci_map_single(dev->pdev, mtt_entry,
- (num_mtt + 2) * 8,
- PCI_DMA_TODEVICE);
- if (pci_dma_mapping_error(indma))
- return -ENOMEM;
-
- err = mthca_cmd(dev, indma, num_mtt, 0, CMD_WRITE_MTT,
- CMD_TIME_CLASS_B, status);
-
- pci_unmap_single(dev->pdev, indma,
- (num_mtt + 2) * 8, PCI_DMA_TODEVICE);
- return err;
+ return mthca_cmd(dev, mailbox->dma, num_mtt, 0, CMD_WRITE_MTT,
+ CMD_TIME_CLASS_B, status);
}
int mthca_SYNC_TPT(struct mthca_dev *dev, u8 *status)
@@ -1418,92 +1438,38 @@ int mthca_MAP_EQ(struct mthca_dev *dev, u64 event_mask, int unmap,
0, CMD_MAP_EQ, CMD_TIME_CLASS_B, status);
}
-int mthca_SW2HW_EQ(struct mthca_dev *dev, void *eq_context,
+int mthca_SW2HW_EQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int eq_num, u8 *status)
{
- dma_addr_t indma;
- int err;
-
- indma = pci_map_single(dev->pdev, eq_context,
- MTHCA_EQ_CONTEXT_SIZE,
- PCI_DMA_TODEVICE);
- if (pci_dma_mapping_error(indma))
- return -ENOMEM;
-
- err = mthca_cmd(dev, indma, eq_num, 0, CMD_SW2HW_EQ,
- CMD_TIME_CLASS_A, status);
-
- pci_unmap_single(dev->pdev, indma,
- MTHCA_EQ_CONTEXT_SIZE, PCI_DMA_TODEVICE);
- return err;
+ return mthca_cmd(dev, mailbox->dma, eq_num, 0, CMD_SW2HW_EQ,
+ CMD_TIME_CLASS_A, status);
}
-int mthca_HW2SW_EQ(struct mthca_dev *dev, void *eq_context,
+int mthca_HW2SW_EQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int eq_num, u8 *status)
{
- dma_addr_t outdma = 0;
- int err;
-
- outdma = pci_map_single(dev->pdev, eq_context,
- MTHCA_EQ_CONTEXT_SIZE,
- PCI_DMA_FROMDEVICE);
- if (pci_dma_mapping_error(outdma))
- return -ENOMEM;
-
- err = mthca_cmd_box(dev, 0, outdma, eq_num, 0,
- CMD_HW2SW_EQ,
- CMD_TIME_CLASS_A, status);
-
- pci_unmap_single(dev->pdev, outdma,
- MTHCA_EQ_CONTEXT_SIZE,
- PCI_DMA_FROMDEVICE);
- return err;
+ return mthca_cmd_box(dev, 0, mailbox->dma, eq_num, 0,
+ CMD_HW2SW_EQ,
+ CMD_TIME_CLASS_A, status);
}
-int mthca_SW2HW_CQ(struct mthca_dev *dev, void *cq_context,
+int mthca_SW2HW_CQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int cq_num, u8 *status)
{
- dma_addr_t indma;
- int err;
-
- indma = pci_map_single(dev->pdev, cq_context,
- MTHCA_CQ_CONTEXT_SIZE,
- PCI_DMA_TODEVICE);
- if (pci_dma_mapping_error(indma))
- return -ENOMEM;
-
- err = mthca_cmd(dev, indma, cq_num, 0, CMD_SW2HW_CQ,
+ return mthca_cmd(dev, mailbox->dma, cq_num, 0, CMD_SW2HW_CQ,
CMD_TIME_CLASS_A, status);
-
- pci_unmap_single(dev->pdev, indma,
- MTHCA_CQ_CONTEXT_SIZE, PCI_DMA_TODEVICE);
- return err;
}
-int mthca_HW2SW_CQ(struct mthca_dev *dev, void *cq_context,
+int mthca_HW2SW_CQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int cq_num, u8 *status)
{
- dma_addr_t outdma = 0;
- int err;
-
- outdma = pci_map_single(dev->pdev, cq_context,
- MTHCA_CQ_CONTEXT_SIZE,
- PCI_DMA_FROMDEVICE);
- if (pci_dma_mapping_error(outdma))
- return -ENOMEM;
-
- err = mthca_cmd_box(dev, 0, outdma, cq_num, 0,
- CMD_HW2SW_CQ,
- CMD_TIME_CLASS_A, status);
-
- pci_unmap_single(dev->pdev, outdma,
- MTHCA_CQ_CONTEXT_SIZE,
- PCI_DMA_FROMDEVICE);
- return err;
+ return mthca_cmd_box(dev, 0, mailbox->dma, cq_num, 0,
+ CMD_HW2SW_CQ,
+ CMD_TIME_CLASS_A, status);
}
int mthca_MODIFY_QP(struct mthca_dev *dev, int trans, u32 num,
- int is_ee, void *qp_context, u32 optmask,
+ int is_ee, struct mthca_mailbox *mailbox, u32 optmask,
u8 *status)
{
static const u16 op[] = {
@@ -1520,36 +1486,34 @@ int mthca_MODIFY_QP(struct mthca_dev *dev, int trans, u32 num,
[MTHCA_TRANS_ANY2RST] = CMD_ERR2RST_QPEE
};
u8 op_mod = 0;
-
- dma_addr_t indma;
+ int my_mailbox = 0;
int err;
if (trans < 0 || trans >= ARRAY_SIZE(op))
return -EINVAL;
if (trans == MTHCA_TRANS_ANY2RST) {
- indma = 0;
op_mod = 3; /* don't write outbox, any->reset */
/* For debugging */
- qp_context = pci_alloc_consistent(dev->pdev, MTHCA_QP_CONTEXT_SIZE,
- &indma);
- op_mod = 2; /* write outbox, any->reset */
+ if (!mailbox) {
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (!IS_ERR(mailbox)) {
+ my_mailbox = 1;
+ op_mod = 2; /* write outbox, any->reset */
+ } else
+ mailbox = NULL;
+ }
} else {
- indma = pci_map_single(dev->pdev, qp_context,
- MTHCA_QP_CONTEXT_SIZE,
- PCI_DMA_TODEVICE);
- if (pci_dma_mapping_error(indma))
- return -ENOMEM;
-
if (0) {
int i;
mthca_dbg(dev, "Dumping QP context:\n");
- printk(" opt param mask: %08x\n", be32_to_cpup(qp_context));
+ printk(" opt param mask: %08x\n", be32_to_cpup(mailbox->buf));
for (i = 0; i < 0x100 / 4; ++i) {
if (i % 8 == 0)
printk(" [%02x] ", i * 4);
- printk(" %08x", be32_to_cpu(((u32 *) qp_context)[i + 2]));
+ printk(" %08x",
+ be32_to_cpu(((u32 *) mailbox->buf)[i + 2]));
if ((i + 1) % 8 == 0)
printk("\n");
}
@@ -1557,55 +1521,39 @@ int mthca_MODIFY_QP(struct mthca_dev *dev, int trans, u32 num,
}
if (trans == MTHCA_TRANS_ANY2RST) {
- err = mthca_cmd_box(dev, 0, indma, (!!is_ee << 24) | num,
- op_mod, op[trans], CMD_TIME_CLASS_C, status);
+ err = mthca_cmd_box(dev, 0, mailbox ? mailbox->dma : 0,
+ (!!is_ee << 24) | num, op_mod,
+ op[trans], CMD_TIME_CLASS_C, status);
- if (0) {
+ if (0 && mailbox) {
int i;
mthca_dbg(dev, "Dumping QP context:\n");
- printk(" %08x\n", be32_to_cpup(qp_context));
+ printk(" %08x\n", be32_to_cpup(mailbox->buf));
for (i = 0; i < 0x100 / 4; ++i) {
if (i % 8 == 0)
printk("[%02x] ", i * 4);
- printk(" %08x", be32_to_cpu(((u32 *) qp_context)[i + 2]));
+ printk(" %08x",
+ be32_to_cpu(((u32 *) mailbox->buf)[i + 2]));
if ((i + 1) % 8 == 0)
printk("\n");
}
}
} else
- err = mthca_cmd(dev, indma, (!!is_ee << 24) | num,
+ err = mthca_cmd(dev, mailbox->dma, (!!is_ee << 24) | num,
op_mod, op[trans], CMD_TIME_CLASS_C, status);
- if (trans != MTHCA_TRANS_ANY2RST)
- pci_unmap_single(dev->pdev, indma,
- MTHCA_QP_CONTEXT_SIZE, PCI_DMA_TODEVICE);
- else
- pci_free_consistent(dev->pdev, MTHCA_QP_CONTEXT_SIZE,
- qp_context, indma);
+ if (my_mailbox)
+ mthca_free_mailbox(dev, mailbox);
+
return err;
}
int mthca_QUERY_QP(struct mthca_dev *dev, u32 num, int is_ee,
- void *qp_context, u8 *status)
+ struct mthca_mailbox *mailbox, u8 *status)
{
- dma_addr_t outdma = 0;
- int err;
-
- outdma = pci_map_single(dev->pdev, qp_context,
- MTHCA_QP_CONTEXT_SIZE,
- PCI_DMA_FROMDEVICE);
- if (pci_dma_mapping_error(outdma))
- return -ENOMEM;
-
- err = mthca_cmd_box(dev, 0, outdma, (!!is_ee << 24) | num, 0,
- CMD_QUERY_QPEE,
- CMD_TIME_CLASS_A, status);
-
- pci_unmap_single(dev->pdev, outdma,
- MTHCA_QP_CONTEXT_SIZE,
- PCI_DMA_FROMDEVICE);
- return err;
+ return mthca_cmd_box(dev, 0, mailbox->dma, (!!is_ee << 24) | num, 0,
+ CMD_QUERY_QPEE, CMD_TIME_CLASS_A, status);
}
int mthca_CONF_SPECIAL_QP(struct mthca_dev *dev, int type, u32 qpn,
@@ -1635,11 +1583,11 @@ int mthca_CONF_SPECIAL_QP(struct mthca_dev *dev, int type, u32 qpn,
}
int mthca_MAD_IFC(struct mthca_dev *dev, int ignore_mkey, int ignore_bkey,
- int port, struct ib_wc* in_wc, struct ib_grh* in_grh,
+ int port, struct ib_wc *in_wc, struct ib_grh *in_grh,
void *in_mad, void *response_mad, u8 *status)
{
- void *box;
- dma_addr_t dma;
+ struct mthca_mailbox *inmailbox, *outmailbox;
+ void *inbox;
int err;
u32 in_modifier = port;
u8 op_modifier = 0;
@@ -1653,11 +1601,18 @@ int mthca_MAD_IFC(struct mthca_dev *dev, int ignore_mkey, int ignore_bkey,
#define MAD_IFC_PKEY_OFFSET 0x10e
#define MAD_IFC_GRH_OFFSET 0x140
- box = pci_alloc_consistent(dev->pdev, MAD_IFC_BOX_SIZE, &dma);
- if (!box)
- return -ENOMEM;
+ inmailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(inmailbox))
+ return PTR_ERR(inmailbox);
+ inbox = inmailbox->buf;
- memcpy(box, in_mad, 256);
+ outmailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(outmailbox)) {
+ mthca_free_mailbox(dev, inmailbox);
+ return PTR_ERR(outmailbox);
+ }
+
+ memcpy(inbox, in_mad, 256);
/*
* Key check traps can't be generated unless we have in_wc to
@@ -1671,97 +1626,65 @@ int mthca_MAD_IFC(struct mthca_dev *dev, int ignore_mkey, int ignore_bkey,
if (in_wc) {
u8 val;
- memset(box + 256, 0, 256);
+ memset(inbox + 256, 0, 256);
- MTHCA_PUT(box, in_wc->qp_num, MAD_IFC_MY_QPN_OFFSET);
- MTHCA_PUT(box, in_wc->src_qp, MAD_IFC_RQPN_OFFSET);
+ MTHCA_PUT(inbox, in_wc->qp_num, MAD_IFC_MY_QPN_OFFSET);
+ MTHCA_PUT(inbox, in_wc->src_qp, MAD_IFC_RQPN_OFFSET);
val = in_wc->sl << 4;
- MTHCA_PUT(box, val, MAD_IFC_SL_OFFSET);
+ MTHCA_PUT(inbox, val, MAD_IFC_SL_OFFSET);
val = in_wc->dlid_path_bits |
(in_wc->wc_flags & IB_WC_GRH ? 0x80 : 0);
- MTHCA_PUT(box, val, MAD_IFC_GRH_OFFSET);
+ MTHCA_PUT(inbox, val, MAD_IFC_GRH_OFFSET);
- MTHCA_PUT(box, in_wc->slid, MAD_IFC_RLID_OFFSET);
- MTHCA_PUT(box, in_wc->pkey_index, MAD_IFC_PKEY_OFFSET);
+ MTHCA_PUT(inbox, in_wc->slid, MAD_IFC_RLID_OFFSET);
+ MTHCA_PUT(inbox, in_wc->pkey_index, MAD_IFC_PKEY_OFFSET);
if (in_grh)
- memcpy((u8 *) box + MAD_IFC_GRH_OFFSET, in_grh, 40);
+ memcpy(inbox + MAD_IFC_GRH_OFFSET, in_grh, 40);
op_modifier |= 0x10;
in_modifier |= in_wc->slid << 16;
}
- err = mthca_cmd_box(dev, dma, dma + 512, in_modifier, op_modifier,
+ err = mthca_cmd_box(dev, inmailbox->dma, outmailbox->dma,
+ in_modifier, op_modifier,
CMD_MAD_IFC, CMD_TIME_CLASS_C, status);
if (!err && !*status)
- memcpy(response_mad, box + 512, 256);
+ memcpy(response_mad, outmailbox->buf, 256);
- pci_free_consistent(dev->pdev, MAD_IFC_BOX_SIZE, box, dma);
+ mthca_free_mailbox(dev, inmailbox);
+ mthca_free_mailbox(dev, outmailbox);
return err;
}
-int mthca_READ_MGM(struct mthca_dev *dev, int index, void *mgm,
- u8 *status)
+int mthca_READ_MGM(struct mthca_dev *dev, int index,
+ struct mthca_mailbox *mailbox, u8 *status)
{
- dma_addr_t outdma = 0;
- int err;
-
- outdma = pci_map_single(dev->pdev, mgm,
- MTHCA_MGM_ENTRY_SIZE,
- PCI_DMA_FROMDEVICE);
- if (pci_dma_mapping_error(outdma))
- return -ENOMEM;
-
- err = mthca_cmd_box(dev, 0, outdma, index, 0,
- CMD_READ_MGM,
- CMD_TIME_CLASS_A, status);
-
- pci_unmap_single(dev->pdev, outdma,
- MTHCA_MGM_ENTRY_SIZE,
- PCI_DMA_FROMDEVICE);
- return err;
+ return mthca_cmd_box(dev, 0, mailbox->dma, index, 0,
+ CMD_READ_MGM, CMD_TIME_CLASS_A, status);
}
-int mthca_WRITE_MGM(struct mthca_dev *dev, int index, void *mgm,
- u8 *status)
+int mthca_WRITE_MGM(struct mthca_dev *dev, int index,
+ struct mthca_mailbox *mailbox, u8 *status)
{
- dma_addr_t indma;
- int err;
-
- indma = pci_map_single(dev->pdev, mgm,
- MTHCA_MGM_ENTRY_SIZE,
- PCI_DMA_TODEVICE);
- if (pci_dma_mapping_error(indma))
- return -ENOMEM;
-
- err = mthca_cmd(dev, indma, index, 0, CMD_WRITE_MGM,
- CMD_TIME_CLASS_A, status);
-
- pci_unmap_single(dev->pdev, indma,
- MTHCA_MGM_ENTRY_SIZE, PCI_DMA_TODEVICE);
- return err;
+ return mthca_cmd(dev, mailbox->dma, index, 0, CMD_WRITE_MGM,
+ CMD_TIME_CLASS_A, status);
}
-int mthca_MGID_HASH(struct mthca_dev *dev, void *gid, u16 *hash,
- u8 *status)
+int mthca_MGID_HASH(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
+ u16 *hash, u8 *status)
{
- dma_addr_t indma;
u64 imm;
int err;
- indma = pci_map_single(dev->pdev, gid, 16, PCI_DMA_TODEVICE);
- if (pci_dma_mapping_error(indma))
- return -ENOMEM;
-
- err = mthca_cmd_imm(dev, indma, &imm, 0, 0, CMD_MGID_HASH,
+ err = mthca_cmd_imm(dev, mailbox->dma, &imm, 0, 0, CMD_MGID_HASH,
CMD_TIME_CLASS_A, status);
- *hash = imm;
- pci_unmap_single(dev->pdev, indma, 16, PCI_DMA_TODEVICE);
+ *hash = imm;
return err;
}
diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.h b/drivers/infiniband/hw/mthca/mthca_cmd.h
index adf039b3c540..ed517f175dd6 100644
--- a/drivers/infiniband/hw/mthca/mthca_cmd.h
+++ b/drivers/infiniband/hw/mthca/mthca_cmd.h
@@ -37,8 +37,7 @@
#include
-#define MTHCA_CMD_MAILBOX_ALIGN 16UL
-#define MTHCA_CMD_MAILBOX_EXTRA (MTHCA_CMD_MAILBOX_ALIGN - 1)
+#define MTHCA_MAILBOX_SIZE 4096
enum {
/* command completed successfully: */
@@ -112,6 +111,11 @@ enum {
DEV_LIM_FLAG_UD_MULTI = 1 << 21,
};
+struct mthca_mailbox {
+ dma_addr_t dma;
+ void *buf;
+};
+
struct mthca_dev_lim {
int max_srq_sz;
int max_qp_sz;
@@ -235,11 +239,17 @@ struct mthca_set_ib_param {
u32 cap_mask;
};
+int mthca_cmd_init(struct mthca_dev *dev);
+void mthca_cmd_cleanup(struct mthca_dev *dev);
int mthca_cmd_use_events(struct mthca_dev *dev);
void mthca_cmd_use_polling(struct mthca_dev *dev);
void mthca_cmd_event(struct mthca_dev *dev, u16 token,
u8 status, u64 out_param);
+struct mthca_mailbox *mthca_alloc_mailbox(struct mthca_dev *dev,
+ unsigned int gfp_mask);
+void mthca_free_mailbox(struct mthca_dev *dev, struct mthca_mailbox *mailbox);
+
int mthca_SYS_EN(struct mthca_dev *dev, u8 *status);
int mthca_SYS_DIS(struct mthca_dev *dev, u8 *status);
int mthca_MAP_FA(struct mthca_dev *dev, struct mthca_icm *icm, u8 *status);
@@ -270,41 +280,39 @@ int mthca_MAP_ICM_AUX(struct mthca_dev *dev, struct mthca_icm *icm, u8 *status);
int mthca_UNMAP_ICM_AUX(struct mthca_dev *dev, u8 *status);
int mthca_SET_ICM_SIZE(struct mthca_dev *dev, u64 icm_size, u64 *aux_pages,
u8 *status);
-int mthca_SW2HW_MPT(struct mthca_dev *dev, void *mpt_entry,
+int mthca_SW2HW_MPT(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int mpt_index, u8 *status);
-int mthca_HW2SW_MPT(struct mthca_dev *dev, void *mpt_entry,
+int mthca_HW2SW_MPT(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int mpt_index, u8 *status);
-int mthca_WRITE_MTT(struct mthca_dev *dev, u64 *mtt_entry,
+int mthca_WRITE_MTT(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int num_mtt, u8 *status);
int mthca_SYNC_TPT(struct mthca_dev *dev, u8 *status);
int mthca_MAP_EQ(struct mthca_dev *dev, u64 event_mask, int unmap,
int eq_num, u8 *status);
-int mthca_SW2HW_EQ(struct mthca_dev *dev, void *eq_context,
+int mthca_SW2HW_EQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int eq_num, u8 *status);
-int mthca_HW2SW_EQ(struct mthca_dev *dev, void *eq_context,
+int mthca_HW2SW_EQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int eq_num, u8 *status);
-int mthca_SW2HW_CQ(struct mthca_dev *dev, void *cq_context,
+int mthca_SW2HW_CQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int cq_num, u8 *status);
-int mthca_HW2SW_CQ(struct mthca_dev *dev, void *cq_context,
+int mthca_HW2SW_CQ(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
int cq_num, u8 *status);
int mthca_MODIFY_QP(struct mthca_dev *dev, int trans, u32 num,
- int is_ee, void *qp_context, u32 optmask,
+ int is_ee, struct mthca_mailbox *mailbox, u32 optmask,
u8 *status);
int mthca_QUERY_QP(struct mthca_dev *dev, u32 num, int is_ee,
- void *qp_context, u8 *status);
+ struct mthca_mailbox *mailbox, u8 *status);
int mthca_CONF_SPECIAL_QP(struct mthca_dev *dev, int type, u32 qpn,
u8 *status);
int mthca_MAD_IFC(struct mthca_dev *dev, int ignore_mkey, int ignore_bkey,
- int port, struct ib_wc* in_wc, struct ib_grh* in_grh,
+ int port, struct ib_wc *in_wc, struct ib_grh *in_grh,
void *in_mad, void *response_mad, u8 *status);
-int mthca_READ_MGM(struct mthca_dev *dev, int index, void *mgm,
- u8 *status);
-int mthca_WRITE_MGM(struct mthca_dev *dev, int index, void *mgm,
- u8 *status);
-int mthca_MGID_HASH(struct mthca_dev *dev, void *gid, u16 *hash,
- u8 *status);
+int mthca_READ_MGM(struct mthca_dev *dev, int index,
+ struct mthca_mailbox *mailbox, u8 *status);
+int mthca_WRITE_MGM(struct mthca_dev *dev, int index,
+ struct mthca_mailbox *mailbox, u8 *status);
+int mthca_MGID_HASH(struct mthca_dev *dev, struct mthca_mailbox *mailbox,
+ u16 *hash, u8 *status);
int mthca_NOP(struct mthca_dev *dev, u8 *status);
-#define MAILBOX_ALIGN(x) ((void *) ALIGN((unsigned long) (x), MTHCA_CMD_MAILBOX_ALIGN))
-
#endif /* MTHCA_CMD_H */
diff --git a/drivers/infiniband/hw/mthca/mthca_cq.c b/drivers/infiniband/hw/mthca/mthca_cq.c
index 2bf347b84c31..766e9031ec45 100644
--- a/drivers/infiniband/hw/mthca/mthca_cq.c
+++ b/drivers/infiniband/hw/mthca/mthca_cq.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2004, 2005 Topspin Communications. All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
@@ -171,6 +172,17 @@ static inline void set_cqe_hw(struct mthca_cqe *cqe)
cqe->owner = MTHCA_CQ_ENTRY_OWNER_HW;
}
+static void dump_cqe(struct mthca_dev *dev, void *cqe_ptr)
+{
+ __be32 *cqe = cqe_ptr;
+
+ (void) cqe; /* avoid warning if mthca_dbg compiled away... */
+ mthca_dbg(dev, "CQE contents %08x %08x %08x %08x %08x %08x %08x %08x\n",
+ be32_to_cpu(cqe[0]), be32_to_cpu(cqe[1]), be32_to_cpu(cqe[2]),
+ be32_to_cpu(cqe[3]), be32_to_cpu(cqe[4]), be32_to_cpu(cqe[5]),
+ be32_to_cpu(cqe[6]), be32_to_cpu(cqe[7]));
+}
+
/*
* incr is ignored in native Arbel (mem-free) mode, so cq->cons_index
* should be correct before calling update_cons_index().
@@ -280,16 +292,12 @@ static int handle_error_cqe(struct mthca_dev *dev, struct mthca_cq *cq,
int dbd;
u32 new_wqe;
- if (1 && cqe->syndrome != SYNDROME_WR_FLUSH_ERR) {
- int j;
-
- mthca_dbg(dev, "%x/%d: error CQE -> QPN %06x, WQE @ %08x\n",
- cq->cqn, cq->cons_index, be32_to_cpu(cqe->my_qpn),
- be32_to_cpu(cqe->wqe));
-
- for (j = 0; j < 8; ++j)
- printk(KERN_DEBUG " [%2x] %08x\n",
- j * 4, be32_to_cpu(((u32 *) cqe)[j]));
+ if (cqe->syndrome == SYNDROME_LOCAL_QP_OP_ERR) {
+ mthca_dbg(dev, "local QP operation err "
+ "(QPN %06x, WQE @ %08x, CQN %06x, index %d)\n",
+ be32_to_cpu(cqe->my_qpn), be32_to_cpu(cqe->wqe),
+ cq->cqn, cq->cons_index);
+ dump_cqe(dev, cqe);
}
/*
@@ -377,15 +385,6 @@ static int handle_error_cqe(struct mthca_dev *dev, struct mthca_cq *cq,
return 0;
}
-static void dump_cqe(struct mthca_cqe *cqe)
-{
- int j;
-
- for (j = 0; j < 8; ++j)
- printk(KERN_DEBUG " [%2x] %08x\n",
- j * 4, be32_to_cpu(((u32 *) cqe)[j]));
-}
-
static inline int mthca_poll_one(struct mthca_dev *dev,
struct mthca_cq *cq,
struct mthca_qp **cur_qp,
@@ -414,8 +413,7 @@ static inline int mthca_poll_one(struct mthca_dev *dev,
mthca_dbg(dev, "%x/%d: CQE -> QPN %06x, WQE @ %08x\n",
cq->cqn, cq->cons_index, be32_to_cpu(cqe->my_qpn),
be32_to_cpu(cqe->wqe));
-
- dump_cqe(cqe);
+ dump_cqe(dev, cqe);
}
is_error = (cqe->opcode & MTHCA_ERROR_CQE_OPCODE_MASK) ==
@@ -638,19 +636,19 @@ static void mthca_free_cq_buf(struct mthca_dev *dev, struct mthca_cq *cq)
int size;
if (cq->is_direct)
- pci_free_consistent(dev->pdev,
- (cq->ibcq.cqe + 1) * MTHCA_CQ_ENTRY_SIZE,
- cq->queue.direct.buf,
- pci_unmap_addr(&cq->queue.direct,
- mapping));
+ dma_free_coherent(&dev->pdev->dev,
+ (cq->ibcq.cqe + 1) * MTHCA_CQ_ENTRY_SIZE,
+ cq->queue.direct.buf,
+ pci_unmap_addr(&cq->queue.direct,
+ mapping));
else {
size = (cq->ibcq.cqe + 1) * MTHCA_CQ_ENTRY_SIZE;
for (i = 0; i < (size + PAGE_SIZE - 1) / PAGE_SIZE; ++i)
if (cq->queue.page_list[i].buf)
- pci_free_consistent(dev->pdev, PAGE_SIZE,
- cq->queue.page_list[i].buf,
- pci_unmap_addr(&cq->queue.page_list[i],
- mapping));
+ dma_free_coherent(&dev->pdev->dev, PAGE_SIZE,
+ cq->queue.page_list[i].buf,
+ pci_unmap_addr(&cq->queue.page_list[i],
+ mapping));
kfree(cq->queue.page_list);
}
@@ -670,8 +668,8 @@ static int mthca_alloc_cq_buf(struct mthca_dev *dev, int size,
npages = 1;
shift = get_order(size) + PAGE_SHIFT;
- cq->queue.direct.buf = pci_alloc_consistent(dev->pdev,
- size, &t);
+ cq->queue.direct.buf = dma_alloc_coherent(&dev->pdev->dev,
+ size, &t, GFP_KERNEL);
if (!cq->queue.direct.buf)
return -ENOMEM;
@@ -709,7 +707,8 @@ static int mthca_alloc_cq_buf(struct mthca_dev *dev, int size,
for (i = 0; i < npages; ++i) {
cq->queue.page_list[i].buf =
- pci_alloc_consistent(dev->pdev, PAGE_SIZE, &t);
+ dma_alloc_coherent(&dev->pdev->dev, PAGE_SIZE,
+ &t, GFP_KERNEL);
if (!cq->queue.page_list[i].buf)
goto err_free;
@@ -746,7 +745,7 @@ int mthca_init_cq(struct mthca_dev *dev, int nent,
struct mthca_cq *cq)
{
int size = nent * MTHCA_CQ_ENTRY_SIZE;
- void *mailbox = NULL;
+ struct mthca_mailbox *mailbox;
struct mthca_cq_context *cq_context;
int err = -ENOMEM;
u8 status;
@@ -780,12 +779,11 @@ int mthca_init_cq(struct mthca_dev *dev, int nent,
goto err_out_ci;
}
- mailbox = kmalloc(sizeof (struct mthca_cq_context) + MTHCA_CMD_MAILBOX_EXTRA,
- GFP_KERNEL);
- if (!mailbox)
- goto err_out_mailbox;
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ goto err_out_arm;
- cq_context = MAILBOX_ALIGN(mailbox);
+ cq_context = mailbox->buf;
err = mthca_alloc_cq_buf(dev, size, cq);
if (err)
@@ -816,7 +814,7 @@ int mthca_init_cq(struct mthca_dev *dev, int nent,
cq_context->state_db = cpu_to_be32(cq->arm_db_index);
}
- err = mthca_SW2HW_CQ(dev, cq_context, cq->cqn, &status);
+ err = mthca_SW2HW_CQ(dev, mailbox, cq->cqn, &status);
if (err) {
mthca_warn(dev, "SW2HW_CQ failed (%d)\n", err);
goto err_out_free_mr;
@@ -840,7 +838,7 @@ int mthca_init_cq(struct mthca_dev *dev, int nent,
cq->cons_index = 0;
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
return 0;
@@ -849,8 +847,9 @@ err_out_free_mr:
mthca_free_cq_buf(dev, cq);
err_out_mailbox:
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
+err_out_arm:
if (mthca_is_memfree(dev))
mthca_free_db(dev, MTHCA_DB_TYPE_CQ_ARM, cq->arm_db_index);
@@ -870,28 +869,26 @@ err_out:
void mthca_free_cq(struct mthca_dev *dev,
struct mthca_cq *cq)
{
- void *mailbox;
+ struct mthca_mailbox *mailbox;
int err;
u8 status;
might_sleep();
- mailbox = kmalloc(sizeof (struct mthca_cq_context) + MTHCA_CMD_MAILBOX_EXTRA,
- GFP_KERNEL);
- if (!mailbox) {
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox)) {
mthca_warn(dev, "No memory for mailbox to free CQ.\n");
return;
}
- err = mthca_HW2SW_CQ(dev, MAILBOX_ALIGN(mailbox), cq->cqn, &status);
+ err = mthca_HW2SW_CQ(dev, mailbox, cq->cqn, &status);
if (err)
mthca_warn(dev, "HW2SW_CQ failed (%d)\n", err);
else if (status)
- mthca_warn(dev, "HW2SW_CQ returned status 0x%02x\n",
- status);
+ mthca_warn(dev, "HW2SW_CQ returned status 0x%02x\n", status);
if (0) {
- u32 *ctx = MAILBOX_ALIGN(mailbox);
+ u32 *ctx = mailbox->buf;
int j;
printk(KERN_ERR "context for CQN %x (cons index %x, next sw %d)\n",
@@ -919,11 +916,11 @@ void mthca_free_cq(struct mthca_dev *dev,
if (mthca_is_memfree(dev)) {
mthca_free_db(dev, MTHCA_DB_TYPE_CQ_ARM, cq->arm_db_index);
mthca_free_db(dev, MTHCA_DB_TYPE_CQ_SET_CI, cq->set_ci_db_index);
- mthca_table_put(dev, dev->cq_table.table, cq->cqn);
}
+ mthca_table_put(dev, dev->cq_table.table, cq->cqn);
mthca_free(&dev->cq_table.alloc, cq->cqn);
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
}
int __devinit mthca_init_cq_table(struct mthca_dev *dev)
diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h
index e3d79e267dc9..4127f09dc5ec 100644
--- a/drivers/infiniband/hw/mthca/mthca_dev.h
+++ b/drivers/infiniband/hw/mthca/mthca_dev.h
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2004, 2005 Topspin Communications. All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
@@ -46,8 +47,8 @@
#define DRV_NAME "ib_mthca"
#define PFX DRV_NAME ": "
-#define DRV_VERSION "0.06-pre"
-#define DRV_RELDATE "November 8, 2004"
+#define DRV_VERSION "0.06"
+#define DRV_RELDATE "June 23, 2005"
enum {
MTHCA_FLAG_DDR_HIDDEN = 1 << 1,
@@ -98,6 +99,7 @@ enum {
};
struct mthca_cmd {
+ struct pci_pool *pool;
int use_events;
struct semaphore hcr_sem;
struct semaphore poll_sem;
@@ -379,6 +381,12 @@ void mthca_uar_free(struct mthca_dev *dev, struct mthca_uar *uar);
int mthca_pd_alloc(struct mthca_dev *dev, struct mthca_pd *pd);
void mthca_pd_free(struct mthca_dev *dev, struct mthca_pd *pd);
+struct mthca_mtt *mthca_alloc_mtt(struct mthca_dev *dev, int size);
+void mthca_free_mtt(struct mthca_dev *dev, struct mthca_mtt *mtt);
+int mthca_write_mtt(struct mthca_dev *dev, struct mthca_mtt *mtt,
+ int start_index, u64 *buffer_list, int list_len);
+int mthca_mr_alloc(struct mthca_dev *dev, u32 pd, int buffer_size_shift,
+ u64 iova, u64 total_size, u32 access, struct mthca_mr *mr);
int mthca_mr_alloc_notrans(struct mthca_dev *dev, u32 pd,
u32 access, struct mthca_mr *mr);
int mthca_mr_alloc_phys(struct mthca_dev *dev, u32 pd,
diff --git a/drivers/infiniband/hw/mthca/mthca_doorbell.h b/drivers/infiniband/hw/mthca/mthca_doorbell.h
index 821039a49049..535fad7710fb 100644
--- a/drivers/infiniband/hw/mthca/mthca_doorbell.h
+++ b/drivers/infiniband/hw/mthca/mthca_doorbell.h
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2004 Topspin Communications. All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mthca/mthca_eq.c b/drivers/infiniband/hw/mthca/mthca_eq.c
index f46d615d396f..cbcf2b4722e4 100644
--- a/drivers/infiniband/hw/mthca/mthca_eq.c
+++ b/drivers/infiniband/hw/mthca/mthca_eq.c
@@ -469,7 +469,7 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev,
PAGE_SIZE;
u64 *dma_list = NULL;
dma_addr_t t;
- void *mailbox = NULL;
+ struct mthca_mailbox *mailbox;
struct mthca_eq_context *eq_context;
int err = -ENOMEM;
int i;
@@ -494,17 +494,16 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev,
if (!dma_list)
goto err_out_free;
- mailbox = kmalloc(sizeof *eq_context + MTHCA_CMD_MAILBOX_EXTRA,
- GFP_KERNEL);
- if (!mailbox)
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
goto err_out_free;
- eq_context = MAILBOX_ALIGN(mailbox);
+ eq_context = mailbox->buf;
for (i = 0; i < npages; ++i) {
- eq->page_list[i].buf = pci_alloc_consistent(dev->pdev,
- PAGE_SIZE, &t);
+ eq->page_list[i].buf = dma_alloc_coherent(&dev->pdev->dev,
+ PAGE_SIZE, &t, GFP_KERNEL);
if (!eq->page_list[i].buf)
- goto err_out_free;
+ goto err_out_free_pages;
dma_list[i] = t;
pci_unmap_addr_set(&eq->page_list[i], mapping, t);
@@ -517,7 +516,7 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev,
eq->eqn = mthca_alloc(&dev->eq_table.alloc);
if (eq->eqn == -1)
- goto err_out_free;
+ goto err_out_free_pages;
err = mthca_mr_alloc_phys(dev, dev->driver_pd.pd_num,
dma_list, PAGE_SHIFT, npages,
@@ -548,7 +547,7 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev,
eq_context->intr = intr;
eq_context->lkey = cpu_to_be32(eq->mr.ibmr.lkey);
- err = mthca_SW2HW_EQ(dev, eq_context, eq->eqn, &status);
+ err = mthca_SW2HW_EQ(dev, mailbox, eq->eqn, &status);
if (err) {
mthca_warn(dev, "SW2HW_EQ failed (%d)\n", err);
goto err_out_free_mr;
@@ -561,7 +560,7 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev,
}
kfree(dma_list);
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
eq->eqn_mask = swab32(1 << eq->eqn);
eq->cons_index = 0;
@@ -579,17 +578,19 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev,
err_out_free_eq:
mthca_free(&dev->eq_table.alloc, eq->eqn);
- err_out_free:
+ err_out_free_pages:
for (i = 0; i < npages; ++i)
if (eq->page_list[i].buf)
- pci_free_consistent(dev->pdev, PAGE_SIZE,
- eq->page_list[i].buf,
- pci_unmap_addr(&eq->page_list[i],
- mapping));
+ dma_free_coherent(&dev->pdev->dev, PAGE_SIZE,
+ eq->page_list[i].buf,
+ pci_unmap_addr(&eq->page_list[i],
+ mapping));
+ mthca_free_mailbox(dev, mailbox);
+
+ err_out_free:
kfree(eq->page_list);
kfree(dma_list);
- kfree(mailbox);
err_out:
return err;
@@ -598,25 +599,22 @@ static int __devinit mthca_create_eq(struct mthca_dev *dev,
static void mthca_free_eq(struct mthca_dev *dev,
struct mthca_eq *eq)
{
- void *mailbox = NULL;
+ struct mthca_mailbox *mailbox;
int err;
u8 status;
int npages = (eq->nent * MTHCA_EQ_ENTRY_SIZE + PAGE_SIZE - 1) /
PAGE_SIZE;
int i;
- mailbox = kmalloc(sizeof (struct mthca_eq_context) + MTHCA_CMD_MAILBOX_EXTRA,
- GFP_KERNEL);
- if (!mailbox)
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
return;
- err = mthca_HW2SW_EQ(dev, MAILBOX_ALIGN(mailbox),
- eq->eqn, &status);
+ err = mthca_HW2SW_EQ(dev, mailbox, eq->eqn, &status);
if (err)
mthca_warn(dev, "HW2SW_EQ failed (%d)\n", err);
if (status)
- mthca_warn(dev, "HW2SW_EQ returned status 0x%02x\n",
- status);
+ mthca_warn(dev, "HW2SW_EQ returned status 0x%02x\n", status);
dev->eq_table.arm_mask &= ~eq->eqn_mask;
@@ -625,7 +623,7 @@ static void mthca_free_eq(struct mthca_dev *dev,
for (i = 0; i < sizeof (struct mthca_eq_context) / 4; ++i) {
if (i % 4 == 0)
printk("[%02x] ", i * 4);
- printk(" %08x", be32_to_cpup(MAILBOX_ALIGN(mailbox) + i * 4));
+ printk(" %08x", be32_to_cpup(mailbox->buf + i * 4));
if ((i + 1) % 4 == 0)
printk("\n");
}
@@ -638,7 +636,7 @@ static void mthca_free_eq(struct mthca_dev *dev,
pci_unmap_addr(&eq->page_list[i], mapping));
kfree(eq->page_list);
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
}
static void mthca_free_irqs(struct mthca_dev *dev)
@@ -709,8 +707,7 @@ static int __devinit mthca_map_eq_regs(struct mthca_dev *dev)
if (mthca_map_reg(dev, ((pci_resource_len(dev->pdev, 0) - 1) &
dev->fw.arbel.eq_arm_base) + 4, 4,
&dev->eq_regs.arbel.eq_arm)) {
- mthca_err(dev, "Couldn't map interrupt clear register, "
- "aborting.\n");
+ mthca_err(dev, "Couldn't map EQ arm register, aborting.\n");
mthca_unmap_reg(dev, (pci_resource_len(dev->pdev, 0) - 1) &
dev->fw.arbel.clr_int_base, MTHCA_CLR_INT_SIZE,
dev->clr_base);
@@ -721,8 +718,7 @@ static int __devinit mthca_map_eq_regs(struct mthca_dev *dev)
dev->fw.arbel.eq_set_ci_base,
MTHCA_EQ_SET_CI_SIZE,
&dev->eq_regs.arbel.eq_set_ci_base)) {
- mthca_err(dev, "Couldn't map interrupt clear register, "
- "aborting.\n");
+ mthca_err(dev, "Couldn't map EQ CI register, aborting.\n");
mthca_unmap_reg(dev, ((pci_resource_len(dev->pdev, 0) - 1) &
dev->fw.arbel.eq_arm_base) + 4, 4,
dev->eq_regs.arbel.eq_arm);
diff --git a/drivers/infiniband/hw/mthca/mthca_main.c b/drivers/infiniband/hw/mthca/mthca_main.c
index d40590356df8..09519b604c08 100644
--- a/drivers/infiniband/hw/mthca/mthca_main.c
+++ b/drivers/infiniband/hw/mthca/mthca_main.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2004, 2005 Topspin Communications. All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
@@ -69,7 +70,7 @@ MODULE_PARM_DESC(msi, "attempt to use MSI if nonzero");
#endif /* CONFIG_PCI_MSI */
static const char mthca_version[] __devinitdata =
- "ib_mthca: Mellanox InfiniBand HCA driver v"
+ DRV_NAME ": Mellanox InfiniBand HCA driver v"
DRV_VERSION " (" DRV_RELDATE ")\n";
static struct mthca_profile default_profile = {
@@ -927,13 +928,13 @@ static int __devinit mthca_init_one(struct pci_dev *pdev,
*/
if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM) ||
pci_resource_len(pdev, 0) != 1 << 20) {
- dev_err(&pdev->dev, "Missing DCS, aborting.");
+ dev_err(&pdev->dev, "Missing DCS, aborting.\n");
err = -ENODEV;
goto err_disable_pdev;
}
if (!(pci_resource_flags(pdev, 2) & IORESOURCE_MEM) ||
pci_resource_len(pdev, 2) != 1 << 23) {
- dev_err(&pdev->dev, "Missing UAR, aborting.");
+ dev_err(&pdev->dev, "Missing UAR, aborting.\n");
err = -ENODEV;
goto err_disable_pdev;
}
@@ -1004,25 +1005,18 @@ static int __devinit mthca_init_one(struct pci_dev *pdev,
!pci_enable_msi(pdev))
mdev->mthca_flags |= MTHCA_FLAG_MSI;
- sema_init(&mdev->cmd.hcr_sem, 1);
- sema_init(&mdev->cmd.poll_sem, 1);
- mdev->cmd.use_events = 0;
-
- mdev->hcr = ioremap(pci_resource_start(pdev, 0) + MTHCA_HCR_BASE, MTHCA_HCR_SIZE);
- if (!mdev->hcr) {
- mthca_err(mdev, "Couldn't map command register, "
- "aborting.\n");
- err = -ENOMEM;
+ if (mthca_cmd_init(mdev)) {
+ mthca_err(mdev, "Failed to init command interface, aborting.\n");
goto err_free_dev;
}
err = mthca_tune_pci(mdev);
if (err)
- goto err_iounmap;
+ goto err_cmd;
err = mthca_init_hca(mdev);
if (err)
- goto err_iounmap;
+ goto err_cmd;
if (mdev->fw_ver < mthca_hca_table[id->driver_data].latest_fw) {
mthca_warn(mdev, "HCA FW version %x.%x.%x is old (%x.%x.%x is current).\n",
@@ -1070,8 +1064,8 @@ err_cleanup:
err_close:
mthca_close_hca(mdev);
-err_iounmap:
- iounmap(mdev->hcr);
+err_cmd:
+ mthca_cmd_cleanup(mdev);
err_free_dev:
if (mdev->mthca_flags & MTHCA_FLAG_MSI_X)
@@ -1118,10 +1112,8 @@ static void __devexit mthca_remove_one(struct pci_dev *pdev)
iounmap(mdev->kar);
mthca_uar_free(mdev, &mdev->driver_uar);
mthca_cleanup_uar_table(mdev);
-
mthca_close_hca(mdev);
-
- iounmap(mdev->hcr);
+ mthca_cmd_cleanup(mdev);
if (mdev->mthca_flags & MTHCA_FLAG_MSI_X)
pci_disable_msix(pdev);
@@ -1163,7 +1155,7 @@ static struct pci_device_id mthca_pci_table[] = {
MODULE_DEVICE_TABLE(pci, mthca_pci_table);
static struct pci_driver mthca_driver = {
- .name = "ib_mthca",
+ .name = DRV_NAME,
.id_table = mthca_pci_table,
.probe = mthca_init_one,
.remove = __devexit_p(mthca_remove_one)
diff --git a/drivers/infiniband/hw/mthca/mthca_mcg.c b/drivers/infiniband/hw/mthca/mthca_mcg.c
index 70a6553a588e..5be7d949dbf6 100644
--- a/drivers/infiniband/hw/mthca/mthca_mcg.c
+++ b/drivers/infiniband/hw/mthca/mthca_mcg.c
@@ -66,22 +66,23 @@ static const u8 zero_gid[16]; /* automatically initialized to 0 */
* entry in hash chain and *mgm holds end of hash chain.
*/
static int find_mgm(struct mthca_dev *dev,
- u8 *gid, struct mthca_mgm *mgm,
+ u8 *gid, struct mthca_mailbox *mgm_mailbox,
u16 *hash, int *prev, int *index)
{
- void *mailbox;
+ struct mthca_mailbox *mailbox;
+ struct mthca_mgm *mgm = mgm_mailbox->buf;
u8 *mgid;
int err;
u8 status;
- mailbox = kmalloc(16 + MTHCA_CMD_MAILBOX_EXTRA, GFP_KERNEL);
- if (!mailbox)
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
return -ENOMEM;
- mgid = MAILBOX_ALIGN(mailbox);
+ mgid = mailbox->buf;
memcpy(mgid, gid, 16);
- err = mthca_MGID_HASH(dev, mgid, hash, &status);
+ err = mthca_MGID_HASH(dev, mailbox, hash, &status);
if (err)
goto out;
if (status) {
@@ -103,7 +104,7 @@ static int find_mgm(struct mthca_dev *dev,
*prev = -1;
do {
- err = mthca_READ_MGM(dev, *index, mgm, &status);
+ err = mthca_READ_MGM(dev, *index, mgm_mailbox, &status);
if (err)
goto out;
if (status) {
@@ -129,14 +130,14 @@ static int find_mgm(struct mthca_dev *dev,
*index = -1;
out:
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
{
struct mthca_dev *dev = to_mdev(ibqp->device);
- void *mailbox;
+ struct mthca_mailbox *mailbox;
struct mthca_mgm *mgm;
u16 hash;
int index, prev;
@@ -145,15 +146,15 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
int err;
u8 status;
- mailbox = kmalloc(sizeof *mgm + MTHCA_CMD_MAILBOX_EXTRA, GFP_KERNEL);
- if (!mailbox)
- return -ENOMEM;
- mgm = MAILBOX_ALIGN(mailbox);
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ mgm = mailbox->buf;
if (down_interruptible(&dev->mcg_table.sem))
return -EINTR;
- err = find_mgm(dev, gid->raw, mgm, &hash, &prev, &index);
+ err = find_mgm(dev, gid->raw, mailbox, &hash, &prev, &index);
if (err)
goto out;
@@ -170,7 +171,7 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
goto out;
}
- err = mthca_READ_MGM(dev, index, mgm, &status);
+ err = mthca_READ_MGM(dev, index, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -195,7 +196,7 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
goto out;
}
- err = mthca_WRITE_MGM(dev, index, mgm, &status);
+ err = mthca_WRITE_MGM(dev, index, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -206,7 +207,7 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
if (!link)
goto out;
- err = mthca_READ_MGM(dev, prev, mgm, &status);
+ err = mthca_READ_MGM(dev, prev, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -217,7 +218,7 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
mgm->next_gid_index = cpu_to_be32(index << 5);
- err = mthca_WRITE_MGM(dev, prev, mgm, &status);
+ err = mthca_WRITE_MGM(dev, prev, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -227,14 +228,14 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
out:
up(&dev->mcg_table.sem);
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
{
struct mthca_dev *dev = to_mdev(ibqp->device);
- void *mailbox;
+ struct mthca_mailbox *mailbox;
struct mthca_mgm *mgm;
u16 hash;
int prev, index;
@@ -242,15 +243,15 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
int err;
u8 status;
- mailbox = kmalloc(sizeof *mgm + MTHCA_CMD_MAILBOX_EXTRA, GFP_KERNEL);
- if (!mailbox)
- return -ENOMEM;
- mgm = MAILBOX_ALIGN(mailbox);
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ mgm = mailbox->buf;
if (down_interruptible(&dev->mcg_table.sem))
return -EINTR;
- err = find_mgm(dev, gid->raw, mgm, &hash, &prev, &index);
+ err = find_mgm(dev, gid->raw, mailbox, &hash, &prev, &index);
if (err)
goto out;
@@ -285,7 +286,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
mgm->qp[loc] = mgm->qp[i - 1];
mgm->qp[i - 1] = 0;
- err = mthca_WRITE_MGM(dev, index, mgm, &status);
+ err = mthca_WRITE_MGM(dev, index, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -304,7 +305,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
if (be32_to_cpu(mgm->next_gid_index) >> 5) {
err = mthca_READ_MGM(dev,
be32_to_cpu(mgm->next_gid_index) >> 5,
- mgm, &status);
+ mailbox, &status);
if (err)
goto out;
if (status) {
@@ -316,7 +317,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
} else
memset(mgm->gid, 0, 16);
- err = mthca_WRITE_MGM(dev, index, mgm, &status);
+ err = mthca_WRITE_MGM(dev, index, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -327,7 +328,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
} else {
/* Remove entry from AMGM */
index = be32_to_cpu(mgm->next_gid_index) >> 5;
- err = mthca_READ_MGM(dev, prev, mgm, &status);
+ err = mthca_READ_MGM(dev, prev, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -338,7 +339,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
mgm->next_gid_index = cpu_to_be32(index << 5);
- err = mthca_WRITE_MGM(dev, prev, mgm, &status);
+ err = mthca_WRITE_MGM(dev, prev, mailbox, &status);
if (err)
goto out;
if (status) {
@@ -350,7 +351,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
out:
up(&dev->mcg_table.sem);
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
return err;
}
diff --git a/drivers/infiniband/hw/mthca/mthca_memfree.c b/drivers/infiniband/hw/mthca/mthca_memfree.c
index 637b30e35592..6d3b05dd9e3f 100644
--- a/drivers/infiniband/hw/mthca/mthca_memfree.c
+++ b/drivers/infiniband/hw/mthca/mthca_memfree.c
@@ -179,9 +179,14 @@ out:
void mthca_table_put(struct mthca_dev *dev, struct mthca_icm_table *table, int obj)
{
- int i = (obj & (table->num_obj - 1)) * table->obj_size / MTHCA_TABLE_CHUNK_SIZE;
+ int i;
u8 status;
+ if (!mthca_is_memfree(dev))
+ return;
+
+ i = (obj & (table->num_obj - 1)) * table->obj_size / MTHCA_TABLE_CHUNK_SIZE;
+
down(&table->mutex);
if (--table->icm[i]->refcount == 0) {
@@ -256,6 +261,9 @@ void mthca_table_put_range(struct mthca_dev *dev, struct mthca_icm_table *table,
{
int i;
+ if (!mthca_is_memfree(dev))
+ return;
+
for (i = start; i <= end; i += MTHCA_TABLE_CHUNK_SIZE / table->obj_size)
mthca_table_put(dev, table, i);
}
diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c
index 8960fc2306be..cbe50feaf680 100644
--- a/drivers/infiniband/hw/mthca/mthca_mr.c
+++ b/drivers/infiniband/hw/mthca/mthca_mr.c
@@ -40,6 +40,12 @@
#include "mthca_cmd.h"
#include "mthca_memfree.h"
+struct mthca_mtt {
+ struct mthca_buddy *buddy;
+ int order;
+ u32 first_seg;
+};
+
/*
* Must be packed because mtt_seg is 64 bits but only aligned to 32 bits.
*/
@@ -173,8 +179,8 @@ static void __devexit mthca_buddy_cleanup(struct mthca_buddy *buddy)
kfree(buddy->bits);
}
-static u32 mthca_alloc_mtt(struct mthca_dev *dev, int order,
- struct mthca_buddy *buddy)
+static u32 mthca_alloc_mtt_range(struct mthca_dev *dev, int order,
+ struct mthca_buddy *buddy)
{
u32 seg = mthca_buddy_alloc(buddy, order);
@@ -191,14 +197,102 @@ static u32 mthca_alloc_mtt(struct mthca_dev *dev, int order,
return seg;
}
-static void mthca_free_mtt(struct mthca_dev *dev, u32 seg, int order,
- struct mthca_buddy* buddy)
+static struct mthca_mtt *__mthca_alloc_mtt(struct mthca_dev *dev, int size,
+ struct mthca_buddy *buddy)
{
- mthca_buddy_free(buddy, seg, order);
+ struct mthca_mtt *mtt;
+ int i;
- if (mthca_is_memfree(dev))
- mthca_table_put_range(dev, dev->mr_table.mtt_table, seg,
- seg + (1 << order) - 1);
+ if (size <= 0)
+ return ERR_PTR(-EINVAL);
+
+ mtt = kmalloc(sizeof *mtt, GFP_KERNEL);
+ if (!mtt)
+ return ERR_PTR(-ENOMEM);
+
+ mtt->buddy = buddy;
+ mtt->order = 0;
+ for (i = MTHCA_MTT_SEG_SIZE / 8; i < size; i <<= 1)
+ ++mtt->order;
+
+ mtt->first_seg = mthca_alloc_mtt_range(dev, mtt->order, buddy);
+ if (mtt->first_seg == -1) {
+ kfree(mtt);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ return mtt;
+}
+
+struct mthca_mtt *mthca_alloc_mtt(struct mthca_dev *dev, int size)
+{
+ return __mthca_alloc_mtt(dev, size, &dev->mr_table.mtt_buddy);
+}
+
+void mthca_free_mtt(struct mthca_dev *dev, struct mthca_mtt *mtt)
+{
+ if (!mtt)
+ return;
+
+ mthca_buddy_free(mtt->buddy, mtt->first_seg, mtt->order);
+
+ mthca_table_put_range(dev, dev->mr_table.mtt_table,
+ mtt->first_seg,
+ mtt->first_seg + (1 << mtt->order) - 1);
+
+ kfree(mtt);
+}
+
+int mthca_write_mtt(struct mthca_dev *dev, struct mthca_mtt *mtt,
+ int start_index, u64 *buffer_list, int list_len)
+{
+ struct mthca_mailbox *mailbox;
+ u64 *mtt_entry;
+ int err = 0;
+ u8 status;
+ int i;
+
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ mtt_entry = mailbox->buf;
+
+ while (list_len > 0) {
+ mtt_entry[0] = cpu_to_be64(dev->mr_table.mtt_base +
+ mtt->first_seg * MTHCA_MTT_SEG_SIZE +
+ start_index * 8);
+ mtt_entry[1] = 0;
+ for (i = 0; i < list_len && i < MTHCA_MAILBOX_SIZE / 8 - 2; ++i)
+ mtt_entry[i + 2] = cpu_to_be64(buffer_list[i] |
+ MTHCA_MTT_FLAG_PRESENT);
+
+ /*
+ * If we have an odd number of entries to write, add
+ * one more dummy entry for firmware efficiency.
+ */
+ if (i & 1)
+ mtt_entry[i + 2] = 0;
+
+ err = mthca_WRITE_MTT(dev, mailbox, (i + 1) & ~1, &status);
+ if (err) {
+ mthca_warn(dev, "WRITE_MTT failed (%d)\n", err);
+ goto out;
+ }
+ if (status) {
+ mthca_warn(dev, "WRITE_MTT returned status 0x%02x\n",
+ status);
+ err = -EINVAL;
+ goto out;
+ }
+
+ list_len -= i;
+ start_index += i;
+ buffer_list += i;
+ }
+
+out:
+ mthca_free_mailbox(dev, mailbox);
+ return err;
}
static inline u32 tavor_hw_index_to_key(u32 ind)
@@ -237,91 +331,18 @@ static inline u32 key_to_hw_index(struct mthca_dev *dev, u32 key)
return tavor_key_to_hw_index(key);
}
-int mthca_mr_alloc_notrans(struct mthca_dev *dev, u32 pd,
- u32 access, struct mthca_mr *mr)
+int mthca_mr_alloc(struct mthca_dev *dev, u32 pd, int buffer_size_shift,
+ u64 iova, u64 total_size, u32 access, struct mthca_mr *mr)
{
- void *mailbox = NULL;
+ struct mthca_mailbox *mailbox;
struct mthca_mpt_entry *mpt_entry;
u32 key;
+ int i;
int err;
u8 status;
might_sleep();
- mr->order = -1;
- key = mthca_alloc(&dev->mr_table.mpt_alloc);
- if (key == -1)
- return -ENOMEM;
- mr->ibmr.rkey = mr->ibmr.lkey = hw_index_to_key(dev, key);
-
- if (mthca_is_memfree(dev)) {
- err = mthca_table_get(dev, dev->mr_table.mpt_table, key);
- if (err)
- goto err_out_mpt_free;
- }
-
- mailbox = kmalloc(sizeof *mpt_entry + MTHCA_CMD_MAILBOX_EXTRA,
- GFP_KERNEL);
- if (!mailbox) {
- err = -ENOMEM;
- goto err_out_table;
- }
- mpt_entry = MAILBOX_ALIGN(mailbox);
-
- mpt_entry->flags = cpu_to_be32(MTHCA_MPT_FLAG_SW_OWNS |
- MTHCA_MPT_FLAG_MIO |
- MTHCA_MPT_FLAG_PHYSICAL |
- MTHCA_MPT_FLAG_REGION |
- access);
- mpt_entry->page_size = 0;
- mpt_entry->key = cpu_to_be32(key);
- mpt_entry->pd = cpu_to_be32(pd);
- mpt_entry->start = 0;
- mpt_entry->length = ~0ULL;
-
- memset(&mpt_entry->lkey, 0,
- sizeof *mpt_entry - offsetof(struct mthca_mpt_entry, lkey));
-
- err = mthca_SW2HW_MPT(dev, mpt_entry,
- key & (dev->limits.num_mpts - 1),
- &status);
- if (err) {
- mthca_warn(dev, "SW2HW_MPT failed (%d)\n", err);
- goto err_out_table;
- } else if (status) {
- mthca_warn(dev, "SW2HW_MPT returned status 0x%02x\n",
- status);
- err = -EINVAL;
- goto err_out_table;
- }
-
- kfree(mailbox);
- return err;
-
-err_out_table:
- if (mthca_is_memfree(dev))
- mthca_table_put(dev, dev->mr_table.mpt_table, key);
-
-err_out_mpt_free:
- mthca_free(&dev->mr_table.mpt_alloc, key);
- kfree(mailbox);
- return err;
-}
-
-int mthca_mr_alloc_phys(struct mthca_dev *dev, u32 pd,
- u64 *buffer_list, int buffer_size_shift,
- int list_len, u64 iova, u64 total_size,
- u32 access, struct mthca_mr *mr)
-{
- void *mailbox;
- u64 *mtt_entry;
- struct mthca_mpt_entry *mpt_entry;
- u32 key;
- int err = -ENOMEM;
- u8 status;
- int i;
-
- might_sleep();
WARN_ON(buffer_size_shift >= 32);
key = mthca_alloc(&dev->mr_table.mpt_alloc);
@@ -335,75 +356,33 @@ int mthca_mr_alloc_phys(struct mthca_dev *dev, u32 pd,
goto err_out_mpt_free;
}
- for (i = MTHCA_MTT_SEG_SIZE / 8, mr->order = 0;
- i < list_len;
- i <<= 1, ++mr->order)
- ; /* nothing */
-
- mr->first_seg = mthca_alloc_mtt(dev, mr->order,
- &dev->mr_table.mtt_buddy);
- if (mr->first_seg == -1)
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox)) {
+ err = PTR_ERR(mailbox);
goto err_out_table;
-
- /*
- * If list_len is odd, we add one more dummy entry for
- * firmware efficiency.
- */
- mailbox = kmalloc(max(sizeof *mpt_entry,
- (size_t) 8 * (list_len + (list_len & 1) + 2)) +
- MTHCA_CMD_MAILBOX_EXTRA,
- GFP_KERNEL);
- if (!mailbox)
- goto err_out_free_mtt;
-
- mtt_entry = MAILBOX_ALIGN(mailbox);
-
- mtt_entry[0] = cpu_to_be64(dev->mr_table.mtt_base +
- mr->first_seg * MTHCA_MTT_SEG_SIZE);
- mtt_entry[1] = 0;
- for (i = 0; i < list_len; ++i)
- mtt_entry[i + 2] = cpu_to_be64(buffer_list[i] |
- MTHCA_MTT_FLAG_PRESENT);
- if (list_len & 1) {
- mtt_entry[i + 2] = 0;
- ++list_len;
}
-
- if (0) {
- mthca_dbg(dev, "Dumping MPT entry\n");
- for (i = 0; i < list_len + 2; ++i)
- printk(KERN_ERR "[%2d] %016llx\n",
- i, (unsigned long long) be64_to_cpu(mtt_entry[i]));
- }
-
- err = mthca_WRITE_MTT(dev, mtt_entry, list_len, &status);
- if (err) {
- mthca_warn(dev, "WRITE_MTT failed (%d)\n", err);
- goto err_out_mailbox_free;
- }
- if (status) {
- mthca_warn(dev, "WRITE_MTT returned status 0x%02x\n",
- status);
- err = -EINVAL;
- goto err_out_mailbox_free;
- }
-
- mpt_entry = MAILBOX_ALIGN(mailbox);
+ mpt_entry = mailbox->buf;
mpt_entry->flags = cpu_to_be32(MTHCA_MPT_FLAG_SW_OWNS |
MTHCA_MPT_FLAG_MIO |
MTHCA_MPT_FLAG_REGION |
access);
+ if (!mr->mtt)
+ mpt_entry->flags |= cpu_to_be32(MTHCA_MPT_FLAG_PHYSICAL);
mpt_entry->page_size = cpu_to_be32(buffer_size_shift - 12);
mpt_entry->key = cpu_to_be32(key);
mpt_entry->pd = cpu_to_be32(pd);
mpt_entry->start = cpu_to_be64(iova);
mpt_entry->length = cpu_to_be64(total_size);
+
memset(&mpt_entry->lkey, 0,
sizeof *mpt_entry - offsetof(struct mthca_mpt_entry, lkey));
- mpt_entry->mtt_seg = cpu_to_be64(dev->mr_table.mtt_base +
- mr->first_seg * MTHCA_MTT_SEG_SIZE);
+
+ if (mr->mtt)
+ mpt_entry->mtt_seg =
+ cpu_to_be64(dev->mr_table.mtt_base +
+ mr->mtt->first_seg * MTHCA_MTT_SEG_SIZE);
if (0) {
mthca_dbg(dev, "Dumping MPT entry %08x:\n", mr->ibmr.lkey);
@@ -416,45 +395,70 @@ int mthca_mr_alloc_phys(struct mthca_dev *dev, u32 pd,
}
}
- err = mthca_SW2HW_MPT(dev, mpt_entry,
+ err = mthca_SW2HW_MPT(dev, mailbox,
key & (dev->limits.num_mpts - 1),
&status);
- if (err)
+ if (err) {
mthca_warn(dev, "SW2HW_MPT failed (%d)\n", err);
- else if (status) {
+ goto err_out_mailbox;
+ } else if (status) {
mthca_warn(dev, "SW2HW_MPT returned status 0x%02x\n",
status);
err = -EINVAL;
+ goto err_out_mailbox;
}
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
return err;
-err_out_mailbox_free:
- kfree(mailbox);
-
-err_out_free_mtt:
- mthca_free_mtt(dev, mr->first_seg, mr->order, &dev->mr_table.mtt_buddy);
+err_out_mailbox:
+ mthca_free_mailbox(dev, mailbox);
err_out_table:
- if (mthca_is_memfree(dev))
- mthca_table_put(dev, dev->mr_table.mpt_table, key);
+ mthca_table_put(dev, dev->mr_table.mpt_table, key);
err_out_mpt_free:
mthca_free(&dev->mr_table.mpt_alloc, key);
return err;
}
-/* Free mr or fmr */
-static void mthca_free_region(struct mthca_dev *dev, u32 lkey, int order,
- u32 first_seg, struct mthca_buddy *buddy)
+int mthca_mr_alloc_notrans(struct mthca_dev *dev, u32 pd,
+ u32 access, struct mthca_mr *mr)
{
- if (order >= 0)
- mthca_free_mtt(dev, first_seg, order, buddy);
+ mr->mtt = NULL;
+ return mthca_mr_alloc(dev, pd, 12, 0, ~0ULL, access, mr);
+}
- if (mthca_is_memfree(dev))
- mthca_table_put(dev, dev->mr_table.mpt_table,
- arbel_key_to_hw_index(lkey));
+int mthca_mr_alloc_phys(struct mthca_dev *dev, u32 pd,
+ u64 *buffer_list, int buffer_size_shift,
+ int list_len, u64 iova, u64 total_size,
+ u32 access, struct mthca_mr *mr)
+{
+ int err;
+
+ mr->mtt = mthca_alloc_mtt(dev, list_len);
+ if (IS_ERR(mr->mtt))
+ return PTR_ERR(mr->mtt);
+
+ err = mthca_write_mtt(dev, mr->mtt, 0, buffer_list, list_len);
+ if (err) {
+ mthca_free_mtt(dev, mr->mtt);
+ return err;
+ }
+
+ err = mthca_mr_alloc(dev, pd, buffer_size_shift, iova,
+ total_size, access, mr);
+ if (err)
+ mthca_free_mtt(dev, mr->mtt);
+
+ return err;
+}
+
+/* Free mr or fmr */
+static void mthca_free_region(struct mthca_dev *dev, u32 lkey)
+{
+ mthca_table_put(dev, dev->mr_table.mpt_table,
+ arbel_key_to_hw_index(lkey));
mthca_free(&dev->mr_table.mpt_alloc, key_to_hw_index(dev, lkey));
}
@@ -476,15 +480,15 @@ void mthca_free_mr(struct mthca_dev *dev, struct mthca_mr *mr)
mthca_warn(dev, "HW2SW_MPT returned status 0x%02x\n",
status);
- mthca_free_region(dev, mr->ibmr.lkey, mr->order, mr->first_seg,
- &dev->mr_table.mtt_buddy);
+ mthca_free_region(dev, mr->ibmr.lkey);
+ mthca_free_mtt(dev, mr->mtt);
}
int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
u32 access, struct mthca_fmr *mr)
{
struct mthca_mpt_entry *mpt_entry;
- void *mailbox;
+ struct mthca_mailbox *mailbox;
u64 mtt_seg;
u32 key, idx;
u8 status;
@@ -522,31 +526,24 @@ int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
mr->mem.tavor.mpt = dev->mr_table.tavor_fmr.mpt_base +
sizeof *(mr->mem.tavor.mpt) * idx;
- for (i = MTHCA_MTT_SEG_SIZE / 8, mr->order = 0;
- i < list_len;
- i <<= 1, ++mr->order)
- ; /* nothing */
-
- mr->first_seg = mthca_alloc_mtt(dev, mr->order,
- dev->mr_table.fmr_mtt_buddy);
- if (mr->first_seg == -1)
+ mr->mtt = __mthca_alloc_mtt(dev, list_len, dev->mr_table.fmr_mtt_buddy);
+ if (IS_ERR(mr->mtt))
goto err_out_table;
- mtt_seg = mr->first_seg * MTHCA_MTT_SEG_SIZE;
+ mtt_seg = mr->mtt->first_seg * MTHCA_MTT_SEG_SIZE;
if (mthca_is_memfree(dev)) {
mr->mem.arbel.mtts = mthca_table_find(dev->mr_table.mtt_table,
- mr->first_seg);
+ mr->mtt->first_seg);
BUG_ON(!mr->mem.arbel.mtts);
} else
mr->mem.tavor.mtts = dev->mr_table.tavor_fmr.mtt_base + mtt_seg;
- mailbox = kmalloc(sizeof *mpt_entry + MTHCA_CMD_MAILBOX_EXTRA,
- GFP_KERNEL);
- if (!mailbox)
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
goto err_out_free_mtt;
- mpt_entry = MAILBOX_ALIGN(mailbox);
+ mpt_entry = mailbox->buf;
mpt_entry->flags = cpu_to_be32(MTHCA_MPT_FLAG_SW_OWNS |
MTHCA_MPT_FLAG_MIO |
@@ -571,7 +568,7 @@ int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
}
}
- err = mthca_SW2HW_MPT(dev, mpt_entry,
+ err = mthca_SW2HW_MPT(dev, mailbox,
key & (dev->limits.num_mpts - 1),
&status);
if (err) {
@@ -585,19 +582,17 @@ int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
goto err_out_mailbox_free;
}
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
return 0;
err_out_mailbox_free:
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
err_out_free_mtt:
- mthca_free_mtt(dev, mr->first_seg, mr->order,
- dev->mr_table.fmr_mtt_buddy);
+ mthca_free_mtt(dev, mr->mtt);
err_out_table:
- if (mthca_is_memfree(dev))
- mthca_table_put(dev, dev->mr_table.mpt_table, key);
+ mthca_table_put(dev, dev->mr_table.mpt_table, key);
err_out_mpt_free:
mthca_free(&dev->mr_table.mpt_alloc, mr->ibmr.lkey);
@@ -609,8 +604,9 @@ int mthca_free_fmr(struct mthca_dev *dev, struct mthca_fmr *fmr)
if (fmr->maps)
return -EBUSY;
- mthca_free_region(dev, fmr->ibmr.lkey, fmr->order, fmr->first_seg,
- dev->mr_table.fmr_mtt_buddy);
+ mthca_free_region(dev, fmr->ibmr.lkey);
+ mthca_free_mtt(dev, fmr->mtt);
+
return 0;
}
@@ -826,7 +822,8 @@ int __devinit mthca_init_mr_table(struct mthca_dev *dev)
if (dev->limits.reserved_mtts) {
i = fls(dev->limits.reserved_mtts - 1);
- if (mthca_alloc_mtt(dev, i, dev->mr_table.fmr_mtt_buddy) == -1) {
+ if (mthca_alloc_mtt_range(dev, i,
+ dev->mr_table.fmr_mtt_buddy) == -1) {
mthca_warn(dev, "MTT table of order %d is too small.\n",
dev->mr_table.fmr_mtt_buddy->max_order);
err = -ENOMEM;
diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c
index 159f4e6c312d..0b5adfd91597 100644
--- a/drivers/infiniband/hw/mthca/mthca_provider.c
+++ b/drivers/infiniband/hw/mthca/mthca_provider.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2004, 2005 Topspin Communications. All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
@@ -52,7 +53,7 @@ static int mthca_query_device(struct ib_device *ibdev,
if (!in_mad || !out_mad)
goto out;
- memset(props, 0, sizeof props);
+ memset(props, 0, sizeof *props);
props->fw_ver = mdev->fw_ver;
@@ -558,6 +559,7 @@ static struct ib_mr *mthca_reg_phys_mr(struct ib_pd *pd,
convert_access(acc), mr);
if (err) {
+ kfree(page_list);
kfree(mr);
return ERR_PTR(err);
}
diff --git a/drivers/infiniband/hw/mthca/mthca_provider.h b/drivers/infiniband/hw/mthca/mthca_provider.h
index 619710f95a87..4d976cccb1a8 100644
--- a/drivers/infiniband/hw/mthca/mthca_provider.h
+++ b/drivers/infiniband/hw/mthca/mthca_provider.h
@@ -54,18 +54,18 @@ struct mthca_uar {
int index;
};
+struct mthca_mtt;
+
struct mthca_mr {
- struct ib_mr ibmr;
- int order;
- u32 first_seg;
+ struct ib_mr ibmr;
+ struct mthca_mtt *mtt;
};
struct mthca_fmr {
- struct ib_fmr ibmr;
+ struct ib_fmr ibmr;
struct ib_fmr_attr attr;
- int order;
- u32 first_seg;
- int maps;
+ struct mthca_mtt *mtt;
+ int maps;
union {
struct {
struct mthca_mpt_entry __iomem *mpt;
diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c
index ca73bab11a02..163a8ef4186f 100644
--- a/drivers/infiniband/hw/mthca/mthca_qp.c
+++ b/drivers/infiniband/hw/mthca/mthca_qp.c
@@ -357,6 +357,9 @@ static const struct {
[UD] = (IB_QP_PKEY_INDEX |
IB_QP_PORT |
IB_QP_QKEY),
+ [UC] = (IB_QP_PKEY_INDEX |
+ IB_QP_PORT |
+ IB_QP_ACCESS_FLAGS),
[RC] = (IB_QP_PKEY_INDEX |
IB_QP_PORT |
IB_QP_ACCESS_FLAGS),
@@ -378,6 +381,9 @@ static const struct {
[UD] = (IB_QP_PKEY_INDEX |
IB_QP_PORT |
IB_QP_QKEY),
+ [UC] = (IB_QP_PKEY_INDEX |
+ IB_QP_PORT |
+ IB_QP_ACCESS_FLAGS),
[RC] = (IB_QP_PKEY_INDEX |
IB_QP_PORT |
IB_QP_ACCESS_FLAGS),
@@ -388,6 +394,11 @@ static const struct {
[IB_QPS_RTR] = {
.trans = MTHCA_TRANS_INIT2RTR,
.req_param = {
+ [UC] = (IB_QP_AV |
+ IB_QP_PATH_MTU |
+ IB_QP_DEST_QPN |
+ IB_QP_RQ_PSN |
+ IB_QP_MAX_DEST_RD_ATOMIC),
[RC] = (IB_QP_AV |
IB_QP_PATH_MTU |
IB_QP_DEST_QPN |
@@ -398,6 +409,9 @@ static const struct {
.opt_param = {
[UD] = (IB_QP_PKEY_INDEX |
IB_QP_QKEY),
+ [UC] = (IB_QP_ALT_PATH |
+ IB_QP_ACCESS_FLAGS |
+ IB_QP_PKEY_INDEX),
[RC] = (IB_QP_ALT_PATH |
IB_QP_ACCESS_FLAGS |
IB_QP_PKEY_INDEX),
@@ -413,6 +427,8 @@ static const struct {
.trans = MTHCA_TRANS_RTR2RTS,
.req_param = {
[UD] = IB_QP_SQ_PSN,
+ [UC] = (IB_QP_SQ_PSN |
+ IB_QP_MAX_QP_RD_ATOMIC),
[RC] = (IB_QP_TIMEOUT |
IB_QP_RETRY_CNT |
IB_QP_RNR_RETRY |
@@ -423,6 +439,11 @@ static const struct {
.opt_param = {
[UD] = (IB_QP_CUR_STATE |
IB_QP_QKEY),
+ [UC] = (IB_QP_CUR_STATE |
+ IB_QP_ALT_PATH |
+ IB_QP_ACCESS_FLAGS |
+ IB_QP_PKEY_INDEX |
+ IB_QP_PATH_MIG_STATE),
[RC] = (IB_QP_CUR_STATE |
IB_QP_ALT_PATH |
IB_QP_ACCESS_FLAGS |
@@ -442,6 +463,9 @@ static const struct {
.opt_param = {
[UD] = (IB_QP_CUR_STATE |
IB_QP_QKEY),
+ [UC] = (IB_QP_ACCESS_FLAGS |
+ IB_QP_ALT_PATH |
+ IB_QP_PATH_MIG_STATE),
[RC] = (IB_QP_ACCESS_FLAGS |
IB_QP_ALT_PATH |
IB_QP_PATH_MIG_STATE |
@@ -462,6 +486,10 @@ static const struct {
.opt_param = {
[UD] = (IB_QP_CUR_STATE |
IB_QP_QKEY),
+ [UC] = (IB_QP_CUR_STATE |
+ IB_QP_ALT_PATH |
+ IB_QP_ACCESS_FLAGS |
+ IB_QP_PATH_MIG_STATE),
[RC] = (IB_QP_CUR_STATE |
IB_QP_ALT_PATH |
IB_QP_ACCESS_FLAGS |
@@ -476,6 +504,14 @@ static const struct {
.opt_param = {
[UD] = (IB_QP_PKEY_INDEX |
IB_QP_QKEY),
+ [UC] = (IB_QP_AV |
+ IB_QP_MAX_QP_RD_ATOMIC |
+ IB_QP_MAX_DEST_RD_ATOMIC |
+ IB_QP_CUR_STATE |
+ IB_QP_ALT_PATH |
+ IB_QP_ACCESS_FLAGS |
+ IB_QP_PKEY_INDEX |
+ IB_QP_PATH_MIG_STATE),
[RC] = (IB_QP_AV |
IB_QP_TIMEOUT |
IB_QP_RETRY_CNT |
@@ -501,6 +537,7 @@ static const struct {
.opt_param = {
[UD] = (IB_QP_CUR_STATE |
IB_QP_QKEY),
+ [UC] = (IB_QP_CUR_STATE),
[RC] = (IB_QP_CUR_STATE |
IB_QP_MIN_RNR_TIMER),
[MLX] = (IB_QP_CUR_STATE |
@@ -552,7 +589,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
struct mthca_dev *dev = to_mdev(ibqp->device);
struct mthca_qp *qp = to_mqp(ibqp);
enum ib_qp_state cur_state, new_state;
- void *mailbox = NULL;
+ struct mthca_mailbox *mailbox;
struct mthca_qp_param *qp_param;
struct mthca_qp_context *qp_context;
u32 req_param, opt_param;
@@ -609,10 +646,10 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
return -EINVAL;
}
- mailbox = kmalloc(sizeof (*qp_param) + MTHCA_CMD_MAILBOX_EXTRA, GFP_KERNEL);
- if (!mailbox)
- return -ENOMEM;
- qp_param = MAILBOX_ALIGN(mailbox);
+ mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ qp_param = mailbox->buf;
qp_context = &qp_param->context;
memset(qp_param, 0, sizeof *qp_param);
@@ -683,7 +720,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
if (attr_mask & IB_QP_AV) {
qp_context->pri_path.g_mylmc = attr->ah_attr.src_path_bits & 0x7f;
qp_context->pri_path.rlid = cpu_to_be16(attr->ah_attr.dlid);
- qp_context->pri_path.static_rate = (!!attr->ah_attr.static_rate) << 3;
+ qp_context->pri_path.static_rate = !!attr->ah_attr.static_rate;
if (attr->ah_attr.ah_flags & IB_AH_GRH) {
qp_context->pri_path.g_mylmc |= 1 << 7;
qp_context->pri_path.mgid_index = attr->ah_attr.grh.sgid_index;
@@ -724,9 +761,9 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
qp_param->opt_param_mask |= cpu_to_be32(MTHCA_QP_OPTPAR_RETRY_COUNT);
}
- if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC) {
- qp_context->params1 |= cpu_to_be32(min(attr->max_dest_rd_atomic ?
- ffs(attr->max_dest_rd_atomic) - 1 : 0,
+ if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC) {
+ qp_context->params1 |= cpu_to_be32(min(attr->max_rd_atomic ?
+ ffs(attr->max_rd_atomic) - 1 : 0,
7) << 21);
qp_param->opt_param_mask |= cpu_to_be32(MTHCA_QP_OPTPAR_SRA_MAX);
}
@@ -764,10 +801,10 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
qp->atomic_rd_en = attr->qp_access_flags;
}
- if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC) {
+ if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC) {
u8 rra_max;
- if (qp->resp_depth && !attr->max_rd_atomic) {
+ if (qp->resp_depth && !attr->max_dest_rd_atomic) {
/*
* Lowering our responder resources to zero.
* Turn off RDMA/atomics as responder.
@@ -778,7 +815,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
MTHCA_QP_OPTPAR_RAE);
}
- if (!qp->resp_depth && attr->max_rd_atomic) {
+ if (!qp->resp_depth && attr->max_dest_rd_atomic) {
/*
* Increasing our responder resources from
* zero. Turn on RDMA/atomics as appropriate.
@@ -799,7 +836,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
}
for (rra_max = 0;
- 1 << rra_max < attr->max_rd_atomic &&
+ 1 << rra_max < attr->max_dest_rd_atomic &&
rra_max < dev->qp_table.rdb_shift;
++rra_max)
; /* nothing */
@@ -807,7 +844,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
qp_context->params2 |= cpu_to_be32(rra_max << 21);
qp_param->opt_param_mask |= cpu_to_be32(MTHCA_QP_OPTPAR_RRA_MAX);
- qp->resp_depth = attr->max_rd_atomic;
+ qp->resp_depth = attr->max_dest_rd_atomic;
}
qp_context->params2 |= cpu_to_be32(MTHCA_QP_BIT_RSC);
@@ -835,7 +872,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
}
err = mthca_MODIFY_QP(dev, state_table[cur_state][new_state].trans,
- qp->qpn, 0, qp_param, 0, &status);
+ qp->qpn, 0, mailbox, 0, &status);
if (status) {
mthca_warn(dev, "modify QP %d returned status %02x.\n",
state_table[cur_state][new_state].trans, status);
@@ -845,7 +882,7 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
if (!err)
qp->state = new_state;
- kfree(mailbox);
+ mthca_free_mailbox(dev, mailbox);
if (is_sqp(dev, qp))
store_attrs(to_msqp(qp), attr, attr_mask);
@@ -934,7 +971,8 @@ static int mthca_alloc_wqe_buf(struct mthca_dev *dev,
mthca_dbg(dev, "Creating direct QP of size %d (shift %d)\n",
size, shift);
- qp->queue.direct.buf = pci_alloc_consistent(dev->pdev, size, &t);
+ qp->queue.direct.buf = dma_alloc_coherent(&dev->pdev->dev, size,
+ &t, GFP_KERNEL);
if (!qp->queue.direct.buf)
goto err_out;
@@ -973,7 +1011,8 @@ static int mthca_alloc_wqe_buf(struct mthca_dev *dev,
for (i = 0; i < npages; ++i) {
qp->queue.page_list[i].buf =
- pci_alloc_consistent(dev->pdev, PAGE_SIZE, &t);
+ dma_alloc_coherent(&dev->pdev->dev, PAGE_SIZE,
+ &t, GFP_KERNEL);
if (!qp->queue.page_list[i].buf)
goto err_out_free;
@@ -996,16 +1035,15 @@ static int mthca_alloc_wqe_buf(struct mthca_dev *dev,
err_out_free:
if (qp->is_direct) {
- pci_free_consistent(dev->pdev, size,
- qp->queue.direct.buf,
- pci_unmap_addr(&qp->queue.direct, mapping));
+ dma_free_coherent(&dev->pdev->dev, size, qp->queue.direct.buf,
+ pci_unmap_addr(&qp->queue.direct, mapping));
} else
for (i = 0; i < npages; ++i) {
if (qp->queue.page_list[i].buf)
- pci_free_consistent(dev->pdev, PAGE_SIZE,
- qp->queue.page_list[i].buf,
- pci_unmap_addr(&qp->queue.page_list[i],
- mapping));
+ dma_free_coherent(&dev->pdev->dev, PAGE_SIZE,
+ qp->queue.page_list[i].buf,
+ pci_unmap_addr(&qp->queue.page_list[i],
+ mapping));
}
@@ -1073,11 +1111,12 @@ static void mthca_free_memfree(struct mthca_dev *dev,
if (mthca_is_memfree(dev)) {
mthca_free_db(dev, MTHCA_DB_TYPE_SQ, qp->sq.db_index);
mthca_free_db(dev, MTHCA_DB_TYPE_RQ, qp->rq.db_index);
- mthca_table_put(dev, dev->qp_table.rdb_table,
- qp->qpn << dev->qp_table.rdb_shift);
- mthca_table_put(dev, dev->qp_table.eqp_table, qp->qpn);
- mthca_table_put(dev, dev->qp_table.qp_table, qp->qpn);
}
+
+ mthca_table_put(dev, dev->qp_table.rdb_table,
+ qp->qpn << dev->qp_table.rdb_shift);
+ mthca_table_put(dev, dev->qp_table.eqp_table, qp->qpn);
+ mthca_table_put(dev, dev->qp_table.qp_table, qp->qpn);
}
static void mthca_wq_init(struct mthca_wq* wq)
@@ -1529,6 +1568,26 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
break;
+ case UC:
+ switch (wr->opcode) {
+ case IB_WR_RDMA_WRITE:
+ case IB_WR_RDMA_WRITE_WITH_IMM:
+ ((struct mthca_raddr_seg *) wqe)->raddr =
+ cpu_to_be64(wr->wr.rdma.remote_addr);
+ ((struct mthca_raddr_seg *) wqe)->rkey =
+ cpu_to_be32(wr->wr.rdma.rkey);
+ ((struct mthca_raddr_seg *) wqe)->reserved = 0;
+ wqe += sizeof (struct mthca_raddr_seg);
+ size += sizeof (struct mthca_raddr_seg) / 16;
+ break;
+
+ default:
+ /* No extra segments required for sends */
+ break;
+ }
+
+ break;
+
case UD:
((struct mthca_tavor_ud_seg *) wqe)->lkey =
cpu_to_be32(to_mah(wr->wr.ud.ah)->key);
@@ -1814,9 +1873,29 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
sizeof (struct mthca_atomic_seg);
break;
+ case IB_WR_RDMA_READ:
+ case IB_WR_RDMA_WRITE:
+ case IB_WR_RDMA_WRITE_WITH_IMM:
+ ((struct mthca_raddr_seg *) wqe)->raddr =
+ cpu_to_be64(wr->wr.rdma.remote_addr);
+ ((struct mthca_raddr_seg *) wqe)->rkey =
+ cpu_to_be32(wr->wr.rdma.rkey);
+ ((struct mthca_raddr_seg *) wqe)->reserved = 0;
+ wqe += sizeof (struct mthca_raddr_seg);
+ size += sizeof (struct mthca_raddr_seg) / 16;
+ break;
+
+ default:
+ /* No extra segments required for sends */
+ break;
+ }
+
+ break;
+
+ case UC:
+ switch (wr->opcode) {
case IB_WR_RDMA_WRITE:
case IB_WR_RDMA_WRITE_WITH_IMM:
- case IB_WR_RDMA_READ:
((struct mthca_raddr_seg *) wqe)->raddr =
cpu_to_be64(wr->wr.rdma.remote_addr);
((struct mthca_raddr_seg *) wqe)->rkey =
diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
index 556264b43425..374f404e81da 100644
--- a/drivers/input/evdev.c
+++ b/drivers/input/evdev.c
@@ -21,6 +21,7 @@
#include
#include
#include
+#include
struct evdev {
int exist;
@@ -145,6 +146,41 @@ static int evdev_open(struct inode * inode, struct file * file)
return 0;
}
+#ifdef CONFIG_COMPAT
+struct input_event_compat {
+ struct compat_timeval time;
+ __u16 type;
+ __u16 code;
+ __s32 value;
+};
+
+#ifdef CONFIG_X86_64
+# define COMPAT_TEST test_thread_flag(TIF_IA32)
+#elif defined(CONFIG_IA64)
+# define COMPAT_TEST IS_IA32_PROCESS(ia64_task_regs(current))
+#elif defined(CONFIG_ARCH_S390)
+# define COMPAT_TEST test_thread_flag(TIF_31BIT)
+#else
+# define COMPAT_TEST test_thread_flag(TIF_32BIT)
+#endif
+
+static ssize_t evdev_write_compat(struct file * file, const char __user * buffer, size_t count, loff_t *ppos)
+{
+ struct evdev_list *list = file->private_data;
+ struct input_event_compat event;
+ int retval = 0;
+
+ while (retval < count) {
+ if (copy_from_user(&event, buffer + retval, sizeof(struct input_event_compat)))
+ return -EFAULT;
+ input_event(list->evdev->handle.dev, event.type, event.code, event.value);
+ retval += sizeof(struct input_event_compat);
+ }
+
+ return retval;
+}
+#endif
+
static ssize_t evdev_write(struct file * file, const char __user * buffer, size_t count, loff_t *ppos)
{
struct evdev_list *list = file->private_data;
@@ -153,6 +189,11 @@ static ssize_t evdev_write(struct file * file, const char __user * buffer, size_
if (!list->evdev->exist) return -ENODEV;
+#ifdef CONFIG_COMPAT
+ if (COMPAT_TEST)
+ return evdev_write_compat(file, buffer, count, ppos);
+#endif
+
while (retval < count) {
if (copy_from_user(&event, buffer + retval, sizeof(struct input_event)))
@@ -164,11 +205,56 @@ static ssize_t evdev_write(struct file * file, const char __user * buffer, size_
return retval;
}
+#ifdef CONFIG_COMPAT
+static ssize_t evdev_read_compat(struct file * file, char __user * buffer, size_t count, loff_t *ppos)
+{
+ struct evdev_list *list = file->private_data;
+ int retval;
+
+ if (count < sizeof(struct input_event_compat))
+ return -EINVAL;
+
+ if (list->head == list->tail && list->evdev->exist && (file->f_flags & O_NONBLOCK))
+ return -EAGAIN;
+
+ retval = wait_event_interruptible(list->evdev->wait,
+ list->head != list->tail || (!list->evdev->exist));
+
+ if (retval)
+ return retval;
+
+ if (!list->evdev->exist)
+ return -ENODEV;
+
+ while (list->head != list->tail && retval + sizeof(struct input_event_compat) <= count) {
+ struct input_event *event = (struct input_event *) list->buffer + list->tail;
+ struct input_event_compat event_compat;
+ event_compat.time.tv_sec = event->time.tv_sec;
+ event_compat.time.tv_usec = event->time.tv_usec;
+ event_compat.type = event->type;
+ event_compat.code = event->code;
+ event_compat.value = event->value;
+
+ if (copy_to_user(buffer + retval, &event_compat,
+ sizeof(struct input_event_compat))) return -EFAULT;
+ list->tail = (list->tail + 1) & (EVDEV_BUFFER_SIZE - 1);
+ retval += sizeof(struct input_event_compat);
+ }
+
+ return retval;
+}
+#endif
+
static ssize_t evdev_read(struct file * file, char __user * buffer, size_t count, loff_t *ppos)
{
struct evdev_list *list = file->private_data;
int retval;
+#ifdef CONFIG_COMPAT
+ if (COMPAT_TEST)
+ return evdev_read_compat(file, buffer, count, ppos);
+#endif
+
if (count < sizeof(struct input_event))
return -EINVAL;
@@ -186,7 +272,7 @@ static ssize_t evdev_read(struct file * file, char __user * buffer, size_t count
while (list->head != list->tail && retval + sizeof(struct input_event) <= count) {
if (copy_to_user(buffer + retval, list->buffer + list->tail,
- sizeof(struct input_event))) return -EFAULT;
+ sizeof(struct input_event))) return -EFAULT;
list->tail = (list->tail + 1) & (EVDEV_BUFFER_SIZE - 1);
retval += sizeof(struct input_event);
}
@@ -203,7 +289,7 @@ static unsigned int evdev_poll(struct file *file, poll_table *wait)
(list->evdev->exist ? 0 : (POLLHUP | POLLERR));
}
-static int evdev_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg)
+static long evdev_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
struct evdev_list *list = file->private_data;
struct evdev *evdev = list->evdev;
@@ -285,110 +371,268 @@ static int evdev_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
default:
- if (_IOC_TYPE(cmd) != 'E' || _IOC_DIR(cmd) != _IOC_READ)
+ if (_IOC_TYPE(cmd) != 'E')
return -EINVAL;
- if ((_IOC_NR(cmd) & ~EV_MAX) == _IOC_NR(EVIOCGBIT(0,0))) {
+ if (_IOC_DIR(cmd) == _IOC_READ) {
- long *bits;
- int len;
+ if ((_IOC_NR(cmd) & ~EV_MAX) == _IOC_NR(EVIOCGBIT(0,0))) {
- switch (_IOC_NR(cmd) & EV_MAX) {
- case 0: bits = dev->evbit; len = EV_MAX; break;
- case EV_KEY: bits = dev->keybit; len = KEY_MAX; break;
- case EV_REL: bits = dev->relbit; len = REL_MAX; break;
- case EV_ABS: bits = dev->absbit; len = ABS_MAX; break;
- case EV_MSC: bits = dev->mscbit; len = MSC_MAX; break;
- case EV_LED: bits = dev->ledbit; len = LED_MAX; break;
- case EV_SND: bits = dev->sndbit; len = SND_MAX; break;
- case EV_FF: bits = dev->ffbit; len = FF_MAX; break;
- default: return -EINVAL;
+ long *bits;
+ int len;
+
+ switch (_IOC_NR(cmd) & EV_MAX) {
+ case 0: bits = dev->evbit; len = EV_MAX; break;
+ case EV_KEY: bits = dev->keybit; len = KEY_MAX; break;
+ case EV_REL: bits = dev->relbit; len = REL_MAX; break;
+ case EV_ABS: bits = dev->absbit; len = ABS_MAX; break;
+ case EV_MSC: bits = dev->mscbit; len = MSC_MAX; break;
+ case EV_LED: bits = dev->ledbit; len = LED_MAX; break;
+ case EV_SND: bits = dev->sndbit; len = SND_MAX; break;
+ case EV_FF: bits = dev->ffbit; len = FF_MAX; break;
+ default: return -EINVAL;
+ }
+ len = NBITS(len) * sizeof(long);
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, bits, len) ? -EFAULT : len;
}
- len = NBITS(len) * sizeof(long);
- if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
- return copy_to_user(p, bits, len) ? -EFAULT : len;
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGKEY(0))) {
+ int len;
+ len = NBITS(KEY_MAX) * sizeof(long);
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->key, len) ? -EFAULT : len;
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGLED(0))) {
+ int len;
+ len = NBITS(LED_MAX) * sizeof(long);
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->led, len) ? -EFAULT : len;
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGSND(0))) {
+ int len;
+ len = NBITS(SND_MAX) * sizeof(long);
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->snd, len) ? -EFAULT : len;
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGNAME(0))) {
+ int len;
+ if (!dev->name) return -ENOENT;
+ len = strlen(dev->name) + 1;
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->name, len) ? -EFAULT : len;
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGPHYS(0))) {
+ int len;
+ if (!dev->phys) return -ENOENT;
+ len = strlen(dev->phys) + 1;
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->phys, len) ? -EFAULT : len;
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGUNIQ(0))) {
+ int len;
+ if (!dev->uniq) return -ENOENT;
+ len = strlen(dev->uniq) + 1;
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->uniq, len) ? -EFAULT : len;
+ }
+
+ if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCGABS(0))) {
+
+ int t = _IOC_NR(cmd) & ABS_MAX;
+
+ abs.value = dev->abs[t];
+ abs.minimum = dev->absmin[t];
+ abs.maximum = dev->absmax[t];
+ abs.fuzz = dev->absfuzz[t];
+ abs.flat = dev->absflat[t];
+
+ if (copy_to_user(p, &abs, sizeof(struct input_absinfo)))
+ return -EFAULT;
+
+ return 0;
+ }
+
}
- if (_IOC_NR(cmd) == _IOC_NR(EVIOCGKEY(0))) {
- int len;
- len = NBITS(KEY_MAX) * sizeof(long);
- if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
- return copy_to_user(p, dev->key, len) ? -EFAULT : len;
- }
+ if (_IOC_DIR(cmd) == _IOC_WRITE) {
- if (_IOC_NR(cmd) == _IOC_NR(EVIOCGLED(0))) {
- int len;
- len = NBITS(LED_MAX) * sizeof(long);
- if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
- return copy_to_user(p, dev->led, len) ? -EFAULT : len;
- }
+ if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCSABS(0))) {
- if (_IOC_NR(cmd) == _IOC_NR(EVIOCGSND(0))) {
- int len;
- len = NBITS(SND_MAX) * sizeof(long);
- if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
- return copy_to_user(p, dev->snd, len) ? -EFAULT : len;
- }
+ int t = _IOC_NR(cmd) & ABS_MAX;
- if (_IOC_NR(cmd) == _IOC_NR(EVIOCGNAME(0))) {
- int len;
- if (!dev->name) return -ENOENT;
- len = strlen(dev->name) + 1;
- if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
- return copy_to_user(p, dev->name, len) ? -EFAULT : len;
- }
+ if (copy_from_user(&abs, p, sizeof(struct input_absinfo)))
+ return -EFAULT;
- if (_IOC_NR(cmd) == _IOC_NR(EVIOCGPHYS(0))) {
- int len;
- if (!dev->phys) return -ENOENT;
- len = strlen(dev->phys) + 1;
- if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
- return copy_to_user(p, dev->phys, len) ? -EFAULT : len;
- }
+ dev->abs[t] = abs.value;
+ dev->absmin[t] = abs.minimum;
+ dev->absmax[t] = abs.maximum;
+ dev->absfuzz[t] = abs.fuzz;
+ dev->absflat[t] = abs.flat;
- if (_IOC_NR(cmd) == _IOC_NR(EVIOCGUNIQ(0))) {
- int len;
- if (!dev->uniq) return -ENOENT;
- len = strlen(dev->uniq) + 1;
- if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
- return copy_to_user(p, dev->uniq, len) ? -EFAULT : len;
- }
-
- if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCGABS(0))) {
-
- int t = _IOC_NR(cmd) & ABS_MAX;
-
- abs.value = dev->abs[t];
- abs.minimum = dev->absmin[t];
- abs.maximum = dev->absmax[t];
- abs.fuzz = dev->absfuzz[t];
- abs.flat = dev->absflat[t];
-
- if (copy_to_user(p, &abs, sizeof(struct input_absinfo)))
- return -EFAULT;
-
- return 0;
- }
-
- if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCSABS(0))) {
-
- int t = _IOC_NR(cmd) & ABS_MAX;
-
- if (copy_from_user(&abs, p, sizeof(struct input_absinfo)))
- return -EFAULT;
-
- dev->abs[t] = abs.value;
- dev->absmin[t] = abs.minimum;
- dev->absmax[t] = abs.maximum;
- dev->absfuzz[t] = abs.fuzz;
- dev->absflat[t] = abs.flat;
-
- return 0;
+ return 0;
+ }
}
}
return -EINVAL;
}
+#ifdef CONFIG_COMPAT
+
+#define BITS_PER_LONG_COMPAT (sizeof(compat_long_t) * 8)
+#define NBITS_COMPAT(x) ((((x)-1)/BITS_PER_LONG_COMPAT)+1)
+#define OFF_COMPAT(x) ((x)%BITS_PER_LONG_COMPAT)
+#define BIT_COMPAT(x) (1UL<> OFF_COMPAT(bit)) & 1)
+
+#ifdef __BIG_ENDIAN
+#define bit_to_user(bit, max) \
+do { \
+ int i; \
+ int len = NBITS_COMPAT((max)) * sizeof(compat_long_t); \
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd); \
+ for (i = 0; i < len / sizeof(compat_long_t); i++) \
+ if (copy_to_user((compat_long_t*) p + i, \
+ (compat_long_t*) (bit) + i + 1 - ((i % 2) << 1), \
+ sizeof(compat_long_t))) \
+ return -EFAULT; \
+ return len; \
+} while (0)
+#else
+#define bit_to_user(bit, max) \
+do { \
+ int len = NBITS_COMPAT((max)) * sizeof(compat_long_t); \
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd); \
+ return copy_to_user(p, (bit), len) ? -EFAULT : len; \
+} while (0)
+#endif
+
+static long evdev_ioctl_compat(struct file *file, unsigned int cmd, unsigned long arg)
+{
+ struct evdev_list *list = file->private_data;
+ struct evdev *evdev = list->evdev;
+ struct input_dev *dev = evdev->handle.dev;
+ struct input_absinfo abs;
+ void __user *p = compat_ptr(arg);
+
+ if (!evdev->exist) return -ENODEV;
+
+ switch (cmd) {
+
+ case EVIOCGVERSION:
+ case EVIOCGID:
+ case EVIOCGKEYCODE:
+ case EVIOCSKEYCODE:
+ case EVIOCSFF:
+ case EVIOCRMFF:
+ case EVIOCGEFFECTS:
+ case EVIOCGRAB:
+ return evdev_ioctl(file, cmd, (unsigned long) p);
+
+ default:
+
+ if (_IOC_TYPE(cmd) != 'E')
+ return -EINVAL;
+
+ if (_IOC_DIR(cmd) == _IOC_READ) {
+
+ if ((_IOC_NR(cmd) & ~EV_MAX) == _IOC_NR(EVIOCGBIT(0,0))) {
+ long *bits;
+ int max;
+
+ switch (_IOC_NR(cmd) & EV_MAX) {
+ case 0: bits = dev->evbit; max = EV_MAX; break;
+ case EV_KEY: bits = dev->keybit; max = KEY_MAX; break;
+ case EV_REL: bits = dev->relbit; max = REL_MAX; break;
+ case EV_ABS: bits = dev->absbit; max = ABS_MAX; break;
+ case EV_MSC: bits = dev->mscbit; max = MSC_MAX; break;
+ case EV_LED: bits = dev->ledbit; max = LED_MAX; break;
+ case EV_SND: bits = dev->sndbit; max = SND_MAX; break;
+ case EV_FF: bits = dev->ffbit; max = FF_MAX; break;
+ default: return -EINVAL;
+ }
+ bit_to_user(bits, max);
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGKEY(0)))
+ bit_to_user(dev->key, KEY_MAX);
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGLED(0)))
+ bit_to_user(dev->led, LED_MAX);
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGSND(0)))
+ bit_to_user(dev->snd, SND_MAX);
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGNAME(0))) {
+ int len;
+ if (!dev->name) return -ENOENT;
+ len = strlen(dev->name) + 1;
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->name, len) ? -EFAULT : len;
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGPHYS(0))) {
+ int len;
+ if (!dev->phys) return -ENOENT;
+ len = strlen(dev->phys) + 1;
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->phys, len) ? -EFAULT : len;
+ }
+
+ if (_IOC_NR(cmd) == _IOC_NR(EVIOCGUNIQ(0))) {
+ int len;
+ if (!dev->uniq) return -ENOENT;
+ len = strlen(dev->uniq) + 1;
+ if (len > _IOC_SIZE(cmd)) len = _IOC_SIZE(cmd);
+ return copy_to_user(p, dev->uniq, len) ? -EFAULT : len;
+ }
+
+ if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCGABS(0))) {
+
+ int t = _IOC_NR(cmd) & ABS_MAX;
+
+ abs.value = dev->abs[t];
+ abs.minimum = dev->absmin[t];
+ abs.maximum = dev->absmax[t];
+ abs.fuzz = dev->absfuzz[t];
+ abs.flat = dev->absflat[t];
+
+ if (copy_to_user(p, &abs, sizeof(struct input_absinfo)))
+ return -EFAULT;
+
+ return 0;
+ }
+ }
+
+ if (_IOC_DIR(cmd) == _IOC_WRITE) {
+
+ if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCSABS(0))) {
+
+ int t = _IOC_NR(cmd) & ABS_MAX;
+
+ if (copy_from_user(&abs, p, sizeof(struct input_absinfo)))
+ return -EFAULT;
+
+ dev->abs[t] = abs.value;
+ dev->absmin[t] = abs.minimum;
+ dev->absmax[t] = abs.maximum;
+ dev->absfuzz[t] = abs.fuzz;
+ dev->absflat[t] = abs.flat;
+
+ return 0;
+ }
+ }
+ }
+ return -EINVAL;
+}
+#endif
+
static struct file_operations evdev_fops = {
.owner = THIS_MODULE,
.read = evdev_read,
@@ -396,7 +640,10 @@ static struct file_operations evdev_fops = {
.poll = evdev_poll,
.open = evdev_open,
.release = evdev_release,
- .ioctl = evdev_ioctl,
+ .unlocked_ioctl = evdev_ioctl,
+#ifdef CONFIG_COMPAT
+ .compat_ioctl = evdev_ioctl_compat,
+#endif
.fasync = evdev_fasync,
.flush = evdev_flush
};
diff --git a/drivers/input/gameport/Kconfig b/drivers/input/gameport/Kconfig
index 1d93f5092904..7524bd7d8b8f 100644
--- a/drivers/input/gameport/Kconfig
+++ b/drivers/input/gameport/Kconfig
@@ -49,22 +49,8 @@ config GAMEPORT_EMU10K1
To compile this driver as a module, choose M here: the
module will be called emu10k1-gp.
-config GAMEPORT_VORTEX
- tristate "Aureal Vortex, Vortex 2 gameport support"
- depends on PCI
- help
- Say Y here if you have an Aureal Vortex 1 or 2 card and want
- to use its gameport.
-
- To compile this driver as a module, choose M here: the
- module will be called vortex.
-
config GAMEPORT_FM801
tristate "ForteMedia FM801 gameport support"
depends on PCI
-config GAMEPORT_CS461X
- tristate "Crystal SoundFusion gameport support"
- depends on PCI
-
endif
diff --git a/drivers/input/gameport/Makefile b/drivers/input/gameport/Makefile
index 5367b4267adf..b6f6097bd8c4 100644
--- a/drivers/input/gameport/Makefile
+++ b/drivers/input/gameport/Makefile
@@ -5,9 +5,7 @@
# Each configuration option enables a list of files.
obj-$(CONFIG_GAMEPORT) += gameport.o
-obj-$(CONFIG_GAMEPORT_CS461X) += cs461x.o
obj-$(CONFIG_GAMEPORT_EMU10K1) += emu10k1-gp.o
obj-$(CONFIG_GAMEPORT_FM801) += fm801-gp.o
obj-$(CONFIG_GAMEPORT_L4) += lightning.o
obj-$(CONFIG_GAMEPORT_NS558) += ns558.o
-obj-$(CONFIG_GAMEPORT_VORTEX) += vortex.o
diff --git a/drivers/input/gameport/cs461x.c b/drivers/input/gameport/cs461x.c
deleted file mode 100644
index d4013ff98623..000000000000
--- a/drivers/input/gameport/cs461x.c
+++ /dev/null
@@ -1,322 +0,0 @@
-/*
- The all defines and part of code (such as cs461x_*) are
- contributed from ALSA 0.5.8 sources.
- See http://www.alsa-project.org/ for sources
-
- Tested on Linux 686 2.4.0-test9, ALSA 0.5.8a and CS4610
-*/
-
-#include
-
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-
-MODULE_AUTHOR("Victor Krapivin");
-MODULE_LICENSE("GPL");
-
-/*
- These options are experimental
-
-#define CS461X_FULL_MAP
-*/
-
-
-#ifndef PCI_VENDOR_ID_CIRRUS
-#define PCI_VENDOR_ID_CIRRUS 0x1013
-#endif
-#ifndef PCI_DEVICE_ID_CIRRUS_4610
-#define PCI_DEVICE_ID_CIRRUS_4610 0x6001
-#endif
-#ifndef PCI_DEVICE_ID_CIRRUS_4612
-#define PCI_DEVICE_ID_CIRRUS_4612 0x6003
-#endif
-#ifndef PCI_DEVICE_ID_CIRRUS_4615
-#define PCI_DEVICE_ID_CIRRUS_4615 0x6004
-#endif
-
-/* Registers */
-
-#define BA0_JSPT 0x00000480
-#define BA0_JSCTL 0x00000484
-#define BA0_JSC1 0x00000488
-#define BA0_JSC2 0x0000048C
-#define BA0_JSIO 0x000004A0
-
-/* Bits for JSPT */
-
-#define JSPT_CAX 0x00000001
-#define JSPT_CAY 0x00000002
-#define JSPT_CBX 0x00000004
-#define JSPT_CBY 0x00000008
-#define JSPT_BA1 0x00000010
-#define JSPT_BA2 0x00000020
-#define JSPT_BB1 0x00000040
-#define JSPT_BB2 0x00000080
-
-/* Bits for JSCTL */
-
-#define JSCTL_SP_MASK 0x00000003
-#define JSCTL_SP_SLOW 0x00000000
-#define JSCTL_SP_MEDIUM_SLOW 0x00000001
-#define JSCTL_SP_MEDIUM_FAST 0x00000002
-#define JSCTL_SP_FAST 0x00000003
-#define JSCTL_ARE 0x00000004
-
-/* Data register pairs masks */
-
-#define JSC1_Y1V_MASK 0x0000FFFF
-#define JSC1_X1V_MASK 0xFFFF0000
-#define JSC1_Y1V_SHIFT 0
-#define JSC1_X1V_SHIFT 16
-#define JSC2_Y2V_MASK 0x0000FFFF
-#define JSC2_X2V_MASK 0xFFFF0000
-#define JSC2_Y2V_SHIFT 0
-#define JSC2_X2V_SHIFT 16
-
-/* JS GPIO */
-
-#define JSIO_DAX 0x00000001
-#define JSIO_DAY 0x00000002
-#define JSIO_DBX 0x00000004
-#define JSIO_DBY 0x00000008
-#define JSIO_AXOE 0x00000010
-#define JSIO_AYOE 0x00000020
-#define JSIO_BXOE 0x00000040
-#define JSIO_BYOE 0x00000080
-
-/*
- The card initialization code is obfuscated; the module cs461x
- need to be loaded after ALSA modules initialized and something
- played on the CS 4610 chip (see sources for details of CS4610
- initialization code from ALSA)
-*/
-
-/* Card specific definitions */
-
-#define CS461X_BA0_SIZE 0x2000
-#define CS461X_BA1_DATA0_SIZE 0x3000
-#define CS461X_BA1_DATA1_SIZE 0x3800
-#define CS461X_BA1_PRG_SIZE 0x7000
-#define CS461X_BA1_REG_SIZE 0x0100
-
-#define BA1_SP_DMEM0 0x00000000
-#define BA1_SP_DMEM1 0x00010000
-#define BA1_SP_PMEM 0x00020000
-#define BA1_SP_REG 0x00030000
-
-#define BA1_DWORD_SIZE (13 * 1024 + 512)
-#define BA1_MEMORY_COUNT 3
-
-/*
- Only one CS461x card is still suppoted; the code requires
- redesign to avoid this limitatuion.
-*/
-
-static unsigned long ba0_addr;
-static unsigned int __iomem *ba0;
-
-#ifdef CS461X_FULL_MAP
-static unsigned long ba1_addr;
-static union ba1_t {
- struct {
- unsigned int __iomem *data0;
- unsigned int __iomem *data1;
- unsigned int __iomem *pmem;
- unsigned int __iomem *reg;
- } name;
- unsigned int __iomem *idx[4];
-} ba1;
-
-static void cs461x_poke(unsigned long reg, unsigned int val)
-{
- writel(val, &ba1.idx[(reg >> 16) & 3][(reg >> 2) & 0x3fff]);
-}
-
-static unsigned int cs461x_peek(unsigned long reg)
-{
- return readl(&ba1.idx[(reg >> 16) & 3][(reg >> 2) & 0x3fff]);
-}
-
-#endif
-
-static void cs461x_pokeBA0(unsigned long reg, unsigned int val)
-{
- writel(val, &ba0[reg >> 2]);
-}
-
-static unsigned int cs461x_peekBA0(unsigned long reg)
-{
- return readl(&ba0[reg >> 2]);
-}
-
-static int cs461x_free(struct pci_dev *pdev)
-{
- struct gameport *port = pci_get_drvdata(pdev);
-
- if (port)
- gameport_unregister_port(port);
-
- if (ba0) iounmap(ba0);
-#ifdef CS461X_FULL_MAP
- if (ba1.name.data0) iounmap(ba1.name.data0);
- if (ba1.name.data1) iounmap(ba1.name.data1);
- if (ba1.name.pmem) iounmap(ba1.name.pmem);
- if (ba1.name.reg) iounmap(ba1.name.reg);
-#endif
- return 0;
-}
-
-static void cs461x_gameport_trigger(struct gameport *gameport)
-{
- cs461x_pokeBA0(BA0_JSPT, 0xFF); //outb(gameport->io, 0xFF);
-}
-
-static unsigned char cs461x_gameport_read(struct gameport *gameport)
-{
- return cs461x_peekBA0(BA0_JSPT); //inb(gameport->io);
-}
-
-static int cs461x_gameport_cooked_read(struct gameport *gameport, int *axes, int *buttons)
-{
- unsigned js1, js2, jst;
-
- js1 = cs461x_peekBA0(BA0_JSC1);
- js2 = cs461x_peekBA0(BA0_JSC2);
- jst = cs461x_peekBA0(BA0_JSPT);
-
- *buttons = (~jst >> 4) & 0x0F;
-
- axes[0] = ((js1 & JSC1_Y1V_MASK) >> JSC1_Y1V_SHIFT) & 0xFFFF;
- axes[1] = ((js1 & JSC1_X1V_MASK) >> JSC1_X1V_SHIFT) & 0xFFFF;
- axes[2] = ((js2 & JSC2_Y2V_MASK) >> JSC2_Y2V_SHIFT) & 0xFFFF;
- axes[3] = ((js2 & JSC2_X2V_MASK) >> JSC2_X2V_SHIFT) & 0xFFFF;
-
- for(jst=0;jst<4;++jst)
- if(axes[jst]==0xFFFF) axes[jst] = -1;
- return 0;
-}
-
-static int cs461x_gameport_open(struct gameport *gameport, int mode)
-{
- switch (mode) {
- case GAMEPORT_MODE_COOKED:
- case GAMEPORT_MODE_RAW:
- return 0;
- default:
- return -1;
- }
- return 0;
-}
-
-static struct pci_device_id cs461x_pci_tbl[] = {
- { PCI_VENDOR_ID_CIRRUS, 0x6001, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* Cirrus CS4610 */
- { PCI_VENDOR_ID_CIRRUS, 0x6003, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* Cirrus CS4612 */
- { PCI_VENDOR_ID_CIRRUS, 0x6005, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* Cirrus CS4615 */
- { 0, }
-};
-MODULE_DEVICE_TABLE(pci, cs461x_pci_tbl);
-
-static int __devinit cs461x_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
-{
- int rc;
- struct gameport* port;
-
- rc = pci_enable_device(pdev);
- if (rc) {
- printk(KERN_ERR "cs461x: Cannot enable PCI gameport (bus %d, devfn %d) error=%d\n",
- pdev->bus->number, pdev->devfn, rc);
- return rc;
- }
-
- ba0_addr = pci_resource_start(pdev, 0);
-#ifdef CS461X_FULL_MAP
- ba1_addr = pci_resource_start(pdev, 1);
-#endif
- if (ba0_addr == 0 || ba0_addr == ~0
-#ifdef CS461X_FULL_MAP
- || ba1_addr == 0 || ba1_addr == ~0
-#endif
- ) {
- printk(KERN_ERR "cs461x: wrong address - ba0 = 0x%lx\n", ba0_addr);
-#ifdef CS461X_FULL_MAP
- printk(KERN_ERR "cs461x: wrong address - ba1 = 0x%lx\n", ba1_addr);
-#endif
- cs461x_free(pdev);
- return -ENOMEM;
- }
-
- ba0 = ioremap(ba0_addr, CS461X_BA0_SIZE);
-#ifdef CS461X_FULL_MAP
- ba1.name.data0 = ioremap(ba1_addr + BA1_SP_DMEM0, CS461X_BA1_DATA0_SIZE);
- ba1.name.data1 = ioremap(ba1_addr + BA1_SP_DMEM1, CS461X_BA1_DATA1_SIZE);
- ba1.name.pmem = ioremap(ba1_addr + BA1_SP_PMEM, CS461X_BA1_PRG_SIZE);
- ba1.name.reg = ioremap(ba1_addr + BA1_SP_REG, CS461X_BA1_REG_SIZE);
-
- if (ba0 == NULL || ba1.name.data0 == NULL ||
- ba1.name.data1 == NULL || ba1.name.pmem == NULL ||
- ba1.name.reg == NULL) {
- cs461x_free(pdev);
- return -ENOMEM;
- }
-#else
- if (ba0 == NULL) {
- cs461x_free(pdev);
- return -ENOMEM;
- }
-#endif
-
- if (!(port = gameport_allocate_port())) {
- printk(KERN_ERR "cs461x: Memory allocation failed\n");
- cs461x_free(pdev);
- return -ENOMEM;
- }
-
- pci_set_drvdata(pdev, port);
-
- port->open = cs461x_gameport_open;
- port->trigger = cs461x_gameport_trigger;
- port->read = cs461x_gameport_read;
- port->cooked_read = cs461x_gameport_cooked_read;
-
- gameport_set_name(port, "CS416x");
- gameport_set_phys(port, "pci%s/gameport0", pci_name(pdev));
- port->dev.parent = &pdev->dev;
-
- cs461x_pokeBA0(BA0_JSIO, 0xFF); // ?
- cs461x_pokeBA0(BA0_JSCTL, JSCTL_SP_MEDIUM_SLOW);
-
- gameport_register_port(port);
-
- return 0;
-}
-
-static void __devexit cs461x_pci_remove(struct pci_dev *pdev)
-{
- cs461x_free(pdev);
-}
-
-static struct pci_driver cs461x_pci_driver = {
- .name = "CS461x_gameport",
- .id_table = cs461x_pci_tbl,
- .probe = cs461x_pci_probe,
- .remove = __devexit_p(cs461x_pci_remove),
-};
-
-static int __init cs461x_init(void)
-{
- return pci_register_driver(&cs461x_pci_driver);
-}
-
-static void __exit cs461x_exit(void)
-{
- pci_unregister_driver(&cs461x_pci_driver);
-}
-
-module_init(cs461x_init);
-module_exit(cs461x_exit);
-
diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c
index c77a82e46055..3e72c9b1461e 100644
--- a/drivers/input/gameport/gameport.c
+++ b/drivers/input/gameport/gameport.c
@@ -17,11 +17,10 @@
#include
#include
#include
-#include
#include
-#include
#include
#include
+#include
/*#include */
@@ -238,8 +237,7 @@ struct gameport_event {
static DEFINE_SPINLOCK(gameport_event_lock); /* protects gameport_event_list */
static LIST_HEAD(gameport_event_list);
static DECLARE_WAIT_QUEUE_HEAD(gameport_wait);
-static DECLARE_COMPLETION(gameport_exited);
-static int gameport_pid;
+static struct task_struct *gameport_task;
static void gameport_queue_event(void *object, struct module *owner,
enum gameport_event_type event_type)
@@ -250,12 +248,12 @@ static void gameport_queue_event(void *object, struct module *owner,
spin_lock_irqsave(&gameport_event_lock, flags);
/*
- * Scan event list for the other events for the same gameport port,
+ * Scan event list for the other events for the same gameport port,
* starting with the most recent one. If event is the same we
* do not need add new one. If event is of different type we
* need to add this event and should not look further because
* we need to preseve sequence of distinct events.
- */
+ */
list_for_each_entry_reverse(event, &gameport_event_list, node) {
if (event->object == object) {
if (event->type == event_type)
@@ -432,20 +430,15 @@ static struct gameport *gameport_get_pending_child(struct gameport *parent)
static int gameport_thread(void *nothing)
{
- lock_kernel();
- daemonize("kgameportd");
- allow_signal(SIGTERM);
-
do {
gameport_handle_events();
- wait_event_interruptible(gameport_wait, !list_empty(&gameport_event_list));
+ wait_event_interruptible(gameport_wait,
+ kthread_should_stop() || !list_empty(&gameport_event_list));
try_to_freeze();
- } while (!signal_pending(current));
+ } while (!kthread_should_stop());
printk(KERN_DEBUG "gameport: kgameportd exiting\n");
-
- unlock_kernel();
- complete_and_exit(&gameport_exited, 0);
+ return 0;
}
@@ -773,9 +766,10 @@ void gameport_close(struct gameport *gameport)
static int __init gameport_init(void)
{
- if (!(gameport_pid = kernel_thread(gameport_thread, NULL, CLONE_KERNEL))) {
+ gameport_task = kthread_run(gameport_thread, NULL, "kgameportd");
+ if (IS_ERR(gameport_task)) {
printk(KERN_ERR "gameport: Failed to start kgameportd\n");
- return -1;
+ return PTR_ERR(gameport_task);
}
gameport_bus.dev_attrs = gameport_device_attrs;
@@ -789,8 +783,7 @@ static int __init gameport_init(void)
static void __exit gameport_exit(void)
{
bus_unregister(&gameport_bus);
- kill_proc(gameport_pid, SIGTERM, 1);
- wait_for_completion(&gameport_exited);
+ kthread_stop(gameport_task);
}
module_init(gameport_init);
diff --git a/drivers/input/gameport/ns558.c b/drivers/input/gameport/ns558.c
index 7c5c6318eeb9..1ab5f2dc8a2a 100644
--- a/drivers/input/gameport/ns558.c
+++ b/drivers/input/gameport/ns558.c
@@ -258,18 +258,18 @@ static int __init ns558_init(void)
{
int i = 0;
+ if (pnp_register_driver(&ns558_pnp_driver) >= 0)
+ pnp_registered = 1;
+
/*
- * Probe ISA ports first so that PnP gets to choose free port addresses
- * not occupied by the ISA ports.
+ * Probe ISA ports after PnP, so that PnP ports that are already
+ * enabled get detected as PnP. This may be suboptimal in multi-device
+ * configurations, but saves hassle with simple setups.
*/
while (ns558_isa_portlist[i])
ns558_isa_probe(ns558_isa_portlist[i++]);
- if (pnp_register_driver(&ns558_pnp_driver) >= 0)
- pnp_registered = 1;
-
-
return (list_empty(&ns558_list) && !pnp_registered) ? -ENODEV : 0;
}
diff --git a/drivers/input/gameport/vortex.c b/drivers/input/gameport/vortex.c
deleted file mode 100644
index 36b0309c8bf6..000000000000
--- a/drivers/input/gameport/vortex.c
+++ /dev/null
@@ -1,186 +0,0 @@
-/*
- * $Id: vortex.c,v 1.5 2002/07/01 15:39:30 vojtech Exp $
- *
- * Copyright (c) 2000-2001 Vojtech Pavlik
- *
- * Based on the work of:
- * Raymond Ingles
- */
-
-/*
- * Trident 4DWave and Aureal Vortex gameport driver for Linux
- */
-
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- *
- * Should you need to contact me, the author, you can do so either by
- * e-mail - mail your message to , or by paper mail:
- * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic
- */
-
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-
-MODULE_AUTHOR("Vojtech Pavlik ");
-MODULE_DESCRIPTION("Aureal Vortex and Vortex2 gameport driver");
-MODULE_LICENSE("GPL");
-
-#define VORTEX_GCR 0x0c /* Gameport control register */
-#define VORTEX_LEG 0x08 /* Legacy port location */
-#define VORTEX_AXD 0x10 /* Axes start */
-#define VORTEX_DATA_WAIT 20 /* 20 ms */
-
-struct vortex {
- struct gameport *gameport;
- struct pci_dev *dev;
- unsigned char __iomem *base;
- unsigned char __iomem *io;
-};
-
-static unsigned char vortex_read(struct gameport *gameport)
-{
- struct vortex *vortex = gameport->port_data;
- return readb(vortex->io + VORTEX_LEG);
-}
-
-static void vortex_trigger(struct gameport *gameport)
-{
- struct vortex *vortex = gameport->port_data;
- writeb(0xff, vortex->io + VORTEX_LEG);
-}
-
-static int vortex_cooked_read(struct gameport *gameport, int *axes, int *buttons)
-{
- struct vortex *vortex = gameport->port_data;
- int i;
-
- *buttons = (~readb(vortex->base + VORTEX_LEG) >> 4) & 0xf;
-
- for (i = 0; i < 4; i++) {
- axes[i] = readw(vortex->io + VORTEX_AXD + i * sizeof(u32));
- if (axes[i] == 0x1fff) axes[i] = -1;
- }
-
- return 0;
-}
-
-static int vortex_open(struct gameport *gameport, int mode)
-{
- struct vortex *vortex = gameport->port_data;
-
- switch (mode) {
- case GAMEPORT_MODE_COOKED:
- writeb(0x40, vortex->io + VORTEX_GCR);
- msleep(VORTEX_DATA_WAIT);
- return 0;
- case GAMEPORT_MODE_RAW:
- writeb(0x00, vortex->io + VORTEX_GCR);
- return 0;
- default:
- return -1;
- }
-
- return 0;
-}
-
-static int __devinit vortex_probe(struct pci_dev *dev, const struct pci_device_id *id)
-{
- struct vortex *vortex;
- struct gameport *port;
- int i;
-
- vortex = kcalloc(1, sizeof(struct vortex), GFP_KERNEL);
- port = gameport_allocate_port();
- if (!vortex || !port) {
- printk(KERN_ERR "vortex: Memory allocation failed.\n");
- kfree(vortex);
- gameport_free_port(port);
- return -ENOMEM;
- }
-
- for (i = 0; i < 6; i++)
- if (~pci_resource_flags(dev, i) & IORESOURCE_IO)
- break;
-
- pci_enable_device(dev);
-
- vortex->dev = dev;
- vortex->gameport = port;
- vortex->base = ioremap(pci_resource_start(vortex->dev, i),
- pci_resource_len(vortex->dev, i));
- vortex->io = vortex->base + id->driver_data;
-
- pci_set_drvdata(dev, vortex);
-
- port->port_data = vortex;
- port->fuzz = 64;
-
- gameport_set_name(port, "AU88x0");
- gameport_set_phys(port, "pci%s/gameport0", pci_name(dev));
- port->dev.parent = &dev->dev;
- port->read = vortex_read;
- port->trigger = vortex_trigger;
- port->cooked_read = vortex_cooked_read;
- port->open = vortex_open;
-
- gameport_register_port(port);
-
- return 0;
-}
-
-static void __devexit vortex_remove(struct pci_dev *dev)
-{
- struct vortex *vortex = pci_get_drvdata(dev);
-
- gameport_unregister_port(vortex->gameport);
- iounmap(vortex->base);
- kfree(vortex);
-}
-
-static struct pci_device_id vortex_id_table[] = {
- { 0x12eb, 0x0001, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0x11000 },
- { 0x12eb, 0x0002, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0x28800 },
- { 0 }
-};
-
-static struct pci_driver vortex_driver = {
- .name = "vortex_gameport",
- .id_table = vortex_id_table,
- .probe = vortex_probe,
- .remove = __devexit_p(vortex_remove),
-};
-
-static int __init vortex_init(void)
-{
- return pci_register_driver(&vortex_driver);
-}
-
-static void __exit vortex_exit(void)
-{
- pci_unregister_driver(&vortex_driver);
-}
-
-module_init(vortex_init);
-module_exit(vortex_exit);
diff --git a/drivers/input/input.c b/drivers/input/input.c
index 83c77c990dda..7c4b4d37b3e6 100644
--- a/drivers/input/input.c
+++ b/drivers/input/input.c
@@ -219,10 +219,24 @@ void input_release_device(struct input_handle *handle)
int input_open_device(struct input_handle *handle)
{
+ struct input_dev *dev = handle->dev;
+ int err;
+
+ err = down_interruptible(&dev->sem);
+ if (err)
+ return err;
+
handle->open++;
- if (handle->dev->open)
- return handle->dev->open(handle->dev);
- return 0;
+
+ if (!dev->users++ && dev->open)
+ err = dev->open(dev);
+
+ if (err)
+ handle->open--;
+
+ up(&dev->sem);
+
+ return err;
}
int input_flush_device(struct input_handle* handle, struct file* file)
@@ -235,10 +249,17 @@ int input_flush_device(struct input_handle* handle, struct file* file)
void input_close_device(struct input_handle *handle)
{
+ struct input_dev *dev = handle->dev;
+
input_release_device(handle);
- if (handle->dev->close)
- handle->dev->close(handle->dev);
+
+ down(&dev->sem);
+
+ if (!--dev->users && dev->close)
+ dev->close(dev);
handle->open--;
+
+ up(&dev->sem);
}
static void input_link_handle(struct input_handle *handle)
@@ -415,6 +436,8 @@ void input_register_device(struct input_dev *dev)
set_bit(EV_SYN, dev->evbit);
+ init_MUTEX(&dev->sem);
+
/*
* If delay and period are pre-set by the driver, then autorepeating
* is handled by the driver itself and we don't do it in input.c.
@@ -674,6 +697,8 @@ static int input_handlers_read(char *buf, char **start, off_t pos, int count, in
return (count > cnt) ? cnt : count;
}
+static struct file_operations input_fileops;
+
static int __init input_proc_init(void)
{
struct proc_dir_entry *entry;
@@ -688,6 +713,8 @@ static int __init input_proc_init(void)
return -ENOMEM;
}
entry->owner = THIS_MODULE;
+ input_fileops = *entry->proc_fops;
+ entry->proc_fops = &input_fileops;
entry->proc_fops->poll = input_devices_poll;
entry = create_proc_read_entry("handlers", 0, proc_bus_input_dir, input_handlers_read, NULL);
if (entry == NULL) {
diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c
index 39775fc380c7..ff8e1bbd0e13 100644
--- a/drivers/input/joydev.c
+++ b/drivers/input/joydev.c
@@ -285,48 +285,33 @@ static unsigned int joydev_poll(struct file *file, poll_table *wait)
(POLLIN | POLLRDNORM) : 0) | (list->joydev->exist ? 0 : (POLLHUP | POLLERR));
}
-static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg)
+static int joydev_ioctl_common(struct joydev *joydev, unsigned int cmd, void __user *argp)
{
- struct joydev_list *list = file->private_data;
- struct joydev *joydev = list->joydev;
struct input_dev *dev = joydev->handle.dev;
- void __user *argp = (void __user *)arg;
int i, j;
- if (!joydev->exist) return -ENODEV;
-
switch (cmd) {
case JS_SET_CAL:
return copy_from_user(&joydev->glue.JS_CORR, argp,
- sizeof(struct JS_DATA_TYPE)) ? -EFAULT : 0;
+ sizeof(joydev->glue.JS_CORR)) ? -EFAULT : 0;
case JS_GET_CAL:
return copy_to_user(argp, &joydev->glue.JS_CORR,
- sizeof(struct JS_DATA_TYPE)) ? -EFAULT : 0;
+ sizeof(joydev->glue.JS_CORR)) ? -EFAULT : 0;
case JS_SET_TIMEOUT:
- return get_user(joydev->glue.JS_TIMEOUT, (int __user *) arg);
+ return get_user(joydev->glue.JS_TIMEOUT, (s32 __user *) argp);
case JS_GET_TIMEOUT:
- return put_user(joydev->glue.JS_TIMEOUT, (int __user *) arg);
- case JS_SET_TIMELIMIT:
- return get_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
- case JS_GET_TIMELIMIT:
- return put_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
- case JS_SET_ALL:
- return copy_from_user(&joydev->glue, argp,
- sizeof(struct JS_DATA_SAVE_TYPE)) ? -EFAULT : 0;
- case JS_GET_ALL:
- return copy_to_user(argp, &joydev->glue,
- sizeof(struct JS_DATA_SAVE_TYPE)) ? -EFAULT : 0;
+ return put_user(joydev->glue.JS_TIMEOUT, (s32 __user *) argp);
case JSIOCGVERSION:
- return put_user(JS_VERSION, (__u32 __user *) arg);
+ return put_user(JS_VERSION, (__u32 __user *) argp);
case JSIOCGAXES:
- return put_user(joydev->nabs, (__u8 __user *) arg);
+ return put_user(joydev->nabs, (__u8 __user *) argp);
case JSIOCGBUTTONS:
- return put_user(joydev->nkey, (__u8 __user *) arg);
+ return put_user(joydev->nkey, (__u8 __user *) argp);
case JSIOCSCORR:
if (copy_from_user(joydev->corr, argp,
- sizeof(struct js_corr) * joydev->nabs))
+ sizeof(joydev->corr[0]) * joydev->nabs))
return -EFAULT;
for (i = 0; i < joydev->nabs; i++) {
j = joydev->abspam[i];
@@ -335,7 +320,7 @@ static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd
return 0;
case JSIOCGCORR:
return copy_to_user(argp, joydev->corr,
- sizeof(struct js_corr) * joydev->nabs) ? -EFAULT : 0;
+ sizeof(joydev->corr[0]) * joydev->nabs) ? -EFAULT : 0;
case JSIOCSAXMAP:
if (copy_from_user(joydev->abspam, argp, sizeof(__u8) * (ABS_MAX + 1)))
return -EFAULT;
@@ -371,6 +356,84 @@ static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd
return -EINVAL;
}
+#ifdef CONFIG_COMPAT
+static long joydev_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+ struct joydev_list *list = file->private_data;
+ struct joydev *joydev = list->joydev;
+ void __user *argp = (void __user *)arg;
+ s32 tmp32;
+ struct JS_DATA_SAVE_TYPE_32 ds32;
+ int err;
+
+ if (!joydev->exist) return -ENODEV;
+ switch(cmd) {
+ case JS_SET_TIMELIMIT:
+ err = get_user(tmp32, (s32 __user *) arg);
+ if (err == 0)
+ joydev->glue.JS_TIMELIMIT = tmp32;
+ break;
+ case JS_GET_TIMELIMIT:
+ tmp32 = joydev->glue.JS_TIMELIMIT;
+ err = put_user(tmp32, (s32 __user *) arg);
+ break;
+
+ case JS_SET_ALL:
+ err = copy_from_user(&ds32, argp,
+ sizeof(ds32)) ? -EFAULT : 0;
+ if (err == 0) {
+ joydev->glue.JS_TIMEOUT = ds32.JS_TIMEOUT;
+ joydev->glue.BUSY = ds32.BUSY;
+ joydev->glue.JS_EXPIRETIME = ds32.JS_EXPIRETIME;
+ joydev->glue.JS_TIMELIMIT = ds32.JS_TIMELIMIT;
+ joydev->glue.JS_SAVE = ds32.JS_SAVE;
+ joydev->glue.JS_CORR = ds32.JS_CORR;
+ }
+ break;
+
+ case JS_GET_ALL:
+ ds32.JS_TIMEOUT = joydev->glue.JS_TIMEOUT;
+ ds32.BUSY = joydev->glue.BUSY;
+ ds32.JS_EXPIRETIME = joydev->glue.JS_EXPIRETIME;
+ ds32.JS_TIMELIMIT = joydev->glue.JS_TIMELIMIT;
+ ds32.JS_SAVE = joydev->glue.JS_SAVE;
+ ds32.JS_CORR = joydev->glue.JS_CORR;
+
+ err = copy_to_user(argp, &ds32,
+ sizeof(ds32)) ? -EFAULT : 0;
+ break;
+
+ default:
+ err = joydev_ioctl_common(joydev, cmd, argp);
+ }
+ return err;
+}
+#endif /* CONFIG_COMPAT */
+
+static int joydev_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg)
+{
+ struct joydev_list *list = file->private_data;
+ struct joydev *joydev = list->joydev;
+ void __user *argp = (void __user *)arg;
+
+ if (!joydev->exist) return -ENODEV;
+
+ switch(cmd) {
+ case JS_SET_TIMELIMIT:
+ return get_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
+ case JS_GET_TIMELIMIT:
+ return put_user(joydev->glue.JS_TIMELIMIT, (long __user *) arg);
+ case JS_SET_ALL:
+ return copy_from_user(&joydev->glue, argp,
+ sizeof(joydev->glue)) ? -EFAULT : 0;
+ case JS_GET_ALL:
+ return copy_to_user(argp, &joydev->glue,
+ sizeof(joydev->glue)) ? -EFAULT : 0;
+ default:
+ return joydev_ioctl_common(joydev, cmd, argp);
+ }
+}
+
static struct file_operations joydev_fops = {
.owner = THIS_MODULE,
.read = joydev_read,
@@ -379,6 +442,9 @@ static struct file_operations joydev_fops = {
.open = joydev_open,
.release = joydev_release,
.ioctl = joydev_ioctl,
+#ifdef CONFIG_COMPAT
+ .compat_ioctl = joydev_compat_ioctl,
+#endif
.fasync = joydev_fasync,
};
diff --git a/drivers/input/joystick/a3d.c b/drivers/input/joystick/a3d.c
index ad39fe4bf35f..bf34f75b9467 100644
--- a/drivers/input/joystick/a3d.c
+++ b/drivers/input/joystick/a3d.c
@@ -185,7 +185,7 @@ static void a3d_poll(struct gameport *gameport)
a3d->reads++;
if (a3d_read_packet(a3d->gameport, a3d->length, data) != a3d->length ||
data[0] != a3d->mode || a3d_csum(data, a3d->length))
- a3d->bads++;
+ a3d->bads++;
else
a3d_read(a3d, data);
}
diff --git a/drivers/input/joystick/adi.c b/drivers/input/joystick/adi.c
index 83f6dafc1716..265962956c63 100644
--- a/drivers/input/joystick/adi.c
+++ b/drivers/input/joystick/adi.c
@@ -82,7 +82,7 @@ static char adi_cm2_abs[] = { ABS_X, ABS_Y, ABS_Z, ABS_RX, ABS_RY, ABS_RZ };
static char adi_wmf_abs[] = { ABS_WHEEL, ABS_GAS, ABS_BRAKE, ABS_HAT0X, ABS_HAT0Y, ABS_HAT1X, ABS_HAT1Y, ABS_HAT2X, ABS_HAT2Y };
static short adi_wmgpe_key[] = { BTN_A, BTN_B, BTN_C, BTN_X, BTN_Y, BTN_Z, BTN_TL, BTN_TR, BTN_START, BTN_MODE, BTN_SELECT };
-static short adi_wmi_key[] = { BTN_TRIGGER, BTN_TOP, BTN_THUMB, BTN_TOP2, BTN_BASE, BTN_BASE2, BTN_BASE3, BTN_BASE4, BTN_EXTRA };
+static short adi_wmi_key[] = { BTN_TRIGGER, BTN_TOP, BTN_THUMB, BTN_TOP2, BTN_BASE, BTN_BASE2, BTN_BASE3, BTN_BASE4, BTN_EXTRA };
static short adi_wmed3d_key[] = { BTN_TRIGGER, BTN_THUMB, BTN_THUMB2, BTN_TOP, BTN_TOP2, BTN_BASE, BTN_BASE2 };
static short adi_cm2_key[] = { BTN_1, BTN_2, BTN_3, BTN_4, BTN_5, BTN_6, BTN_7, BTN_8 };
@@ -183,7 +183,7 @@ static void adi_move_bits(struct adi_port *port, int length)
int i;
struct adi *adi = port->adi;
- adi[0].idx = adi[1].idx = 0;
+ adi[0].idx = adi[1].idx = 0;
if (adi[0].ret <= 0 || adi[1].ret <= 0) return;
if (adi[0].data[0] & 0x20 || ~adi[1].data[0] & 0x20) return;
diff --git a/drivers/input/joystick/amijoy.c b/drivers/input/joystick/amijoy.c
index cf36ca9b92f3..033456bb9fe0 100644
--- a/drivers/input/joystick/amijoy.c
+++ b/drivers/input/joystick/amijoy.c
@@ -51,7 +51,8 @@ MODULE_PARM_DESC(map, "Map of attached joysticks in form of , (default is
__obsolete_setup("amijoy=");
-static int amijoy_used[2] = { 0, 0 };
+static int amijoy_used;
+static DECLARE_MUTEX(amijoy_sem);
static struct input_dev amijoy_dev[2];
static char *amijoy_phys[2] = { "amijoy/input0", "amijoy/input1" };
@@ -84,26 +85,30 @@ static irqreturn_t amijoy_interrupt(int irq, void *dummy, struct pt_regs *fp)
static int amijoy_open(struct input_dev *dev)
{
- int *used = dev->private;
+ int err;
- if ((*used)++)
- return 0;
+ err = down_interruptible(&amijoy_sem);
+ if (err)
+ return err;
- if (request_irq(IRQ_AMIGA_VERTB, amijoy_interrupt, 0, "amijoy", amijoy_interrupt)) {
- (*used)--;
+ if (!amijoy_used && request_irq(IRQ_AMIGA_VERTB, amijoy_interrupt, 0, "amijoy", amijoy_interrupt)) {
printk(KERN_ERR "amijoy.c: Can't allocate irq %d\n", IRQ_AMIGA_VERTB);
- return -EBUSY;
+ err = -EBUSY;
+ goto out;
}
- return 0;
+ amijoy_used++;
+out:
+ up(&amijoy_sem);
+ return err;
}
static void amijoy_close(struct input_dev *dev)
{
- int *used = dev->private;
-
- if (!--(*used))
+ down(&amijoysem);
+ if (!--amijoy_used)
free_irq(IRQ_AMIGA_VERTB, amijoy_interrupt);
+ up(&amijoy_sem);
}
static int __init amijoy_init(void)
@@ -138,8 +143,6 @@ static int __init amijoy_init(void)
amijoy_dev[i].id.product = 0x0003;
amijoy_dev[i].id.version = 0x0100;
- amijoy_dev[i].private = amijoy_used + i;
-
input_register_device(amijoy_dev + i);
printk(KERN_INFO "input: %s at joy%ddat\n", amijoy_name, i);
}
diff --git a/drivers/input/joystick/db9.c b/drivers/input/joystick/db9.c
index cfdd3acf06a1..fbd3eed07f90 100644
--- a/drivers/input/joystick/db9.c
+++ b/drivers/input/joystick/db9.c
@@ -87,7 +87,7 @@ __obsolete_setup("db9_3=");
#define DB9_NORMAL 0x0a
#define DB9_NOSELECT 0x08
-#define DB9_MAX_DEVICES 2
+#define DB9_MAX_DEVICES 2
#define DB9_GENESIS6_DELAY 14
#define DB9_REFRESH_TIME HZ/100
@@ -98,6 +98,7 @@ struct db9 {
struct pardevice *pd;
int mode;
int used;
+ struct semaphore sem;
char phys[2][32];
};
@@ -503,6 +504,11 @@ static int db9_open(struct input_dev *dev)
{
struct db9 *db9 = dev->private;
struct parport *port = db9->pd->port;
+ int err;
+
+ err = down_interruptible(&db9->sem);
+ if (err)
+ return err;
if (!db9->used++) {
parport_claim(db9->pd);
@@ -514,6 +520,7 @@ static int db9_open(struct input_dev *dev)
mod_timer(&db9->timer, jiffies + DB9_REFRESH_TIME);
}
+ up(&db9->sem);
return 0;
}
@@ -522,12 +529,14 @@ static void db9_close(struct input_dev *dev)
struct db9 *db9 = dev->private;
struct parport *port = db9->pd->port;
+ down(&db9->sem);
if (!--db9->used) {
- del_timer(&db9->timer);
+ del_timer_sync(&db9->timer);
parport_write_control(port, 0x00);
parport_data_forward(port);
parport_release(db9->pd);
}
+ up(&db9->sem);
}
static struct db9 __init *db9_probe(int *config, int nargs)
@@ -563,12 +572,12 @@ static struct db9 __init *db9_probe(int *config, int nargs)
}
}
- if (!(db9 = kmalloc(sizeof(struct db9), GFP_KERNEL))) {
+ if (!(db9 = kcalloc(1, sizeof(struct db9), GFP_KERNEL))) {
parport_put_port(pp);
return NULL;
}
- memset(db9, 0, sizeof(struct db9));
+ init_MUTEX(&db9->sem);
db9->mode = config[1];
init_timer(&db9->timer);
db9->timer.data = (long) db9;
diff --git a/drivers/input/joystick/gamecon.c b/drivers/input/joystick/gamecon.c
index 8732f52bdd08..95bbdd302aad 100644
--- a/drivers/input/joystick/gamecon.c
+++ b/drivers/input/joystick/gamecon.c
@@ -1,12 +1,12 @@
/*
* NES, SNES, N64, MultiSystem, PSX gamepad driver for Linux
*
- * Copyright (c) 1999-2004 Vojtech Pavlik
- * Copyright (c) 2004 Peter Nelson
+ * Copyright (c) 1999-2004 Vojtech Pavlik