Under some circumstances the PLLE needs to be retrained, in which case
access to the PMC registers is required. Fix this by passing a pointer
to the PMC registers instead of NULL when registering the PLLE clock.
Signed-off-by: Thierry Reding <thierry.reding@avionic-design.de>
Acked-By: Peter De Schrijver <pdeschrijver@nvidia.com>
Signed-off-by: Mike Turquette <mturquette@linaro.org>
Matthew reported kernels fail the pci_eisa probe and are later successful
with the virtual_eisa_root_init force probe without slot0.
The reason for that is: PNP probing is before pci_eisa_init gets called
as pci_eisa_init is called via pci_driver.
pnp 00:0f has 0xc80 - 0xc84 reserved.
[ 9.700409] pnp 00:0f: [io 0x0c80-0x0c84]
so eisa_probe will fail from pci_eisa_init
==>eisa_root_register
==>eisa_probe path.
as force_probe is not set in pci_eisa_root, it will bail early when
slot0 is not probed and initialized.
Try to use subsys_initcall_sync instead, and will keep following sequence:
pci_subsys_init
pci_eisa_init_early
pnpacpi_init/isapnp_init
After this patch EISA can be initialized properly, and PNP overlapping
resource will not be reserved.
[ 10.104434] system 00:0f: [io 0x0c80-0x0c84] could not be reserved
Reported-by: Matthew Whitehead <mwhitehe@redhat.com>
Tested-by: Matthew Whitehead <mwhitehe@redhat.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org
Matthew found that 3.8.3 is having problems with an old (ancient)
PCI-to-EISA bridge, the Intel 82375. It worked with the 3.2 kernel.
He identified the 82375, but doesn't assign the struct resource *res
pointer inside the struct eisa_root_device, and panics.
pci_eisa_init() was using bus->resource[] directly instead of
pci_bus_resource_n(). The bus->resource[] array is a PCI-internal
implementation detail, and after commit 45ca9e97 (PCI: add helpers for
building PCI bus resource lists) and commit 0efd5aab (PCI: add struct
pci_host_bridge_window with CPU/bus address offset), bus->resource[] is not
used for PCI root buses any more.
The 82375 is a subtractive-decode PCI device, so handle it the same
way we handle PCI-PCI bridges in subtractive-decode mode in
pci_read_bridge_bases().
[bhelgaas: changelog]
Reported-by: Matthew Whitehead <mwhitehe@redhat.com>
Tested-by: Matthew Whitehead <mwhitehe@redhat.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org # v3.3+
Pull arch/tile fix from Chris Metcalf:
"This change allows newer Tilera boot tools to work correctly with
current (and stable) kernels by using the right filename to get the
initramfs from the Tilera hypervisor filesystem."
* 'stable' of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile:
tile: expect new initramfs name from hypervisor file system
Pull networking fixes from David Miller:
1) sadb_msg prepared for IPSEC userspace forgets to initialize the
satype field, fix from Nicolas Dichtel.
2) Fix mac80211 synchronization during station removal, from Johannes
Berg.
3) Fix IPSEC sequence number notifications when they wrap, from Steffen
Klassert.
4) Fix cfg80211 wdev tracing crashes when add_virtual_intf() returns an
error pointer, from Johannes Berg.
5) In mac80211, don't call into the channel context code with the
interface list mutex held. From Johannes Berg.
6) In mac80211, if we don't actually associate, do not restart the STA
timer, otherwise we can crash. From Ben Greear.
7) Missing dma_mapping_error() check in e1000, ixgb, and e1000e. From
Christoph Paasch.
8) Fix sja1000 driver defines to not conflict with SH port, from Marc
Kleine-Budde.
9) Don't call il4965_rs_use_green with a NULL station, from Colin Ian
King.
10) Suspend/Resume in the FEC driver fail because the buffer descriptors
are not initialized at all the moments in which they should. Fix
from Frank Li.
11) cpsw and davinci_emac drivers both use the wrong interface to
restart a stopped TX queue. Use netif_wake_queue not
netif_start_queue, the latter is for initialization/bringup not
active management of the queue. From Mugunthan V N.
12) Fix regression in rate calculations done by
psched_ratecfg_precompute(), missing u64 type promotion. From
Sergey Popovich.
13) Fix length overflow in tg3 VPD parsing, from Kees Cook.
14) AOE driver fails to allocate enough headroom, resulting in crashes.
Fix from Eric Dumazet.
15) RX overflow happens too quickly in sky2 driver because pause packet
thresholds are not programmed correctly. From Mirko Lindner.
16) Bonding driver manages arp_interval and miimon settings incorrectly,
disabling one unintentionally disables both. Fix from Nikolay
Aleksandrov.
17) smsc75xx drivers don't program the RX mac properly for jumbo frames.
Fix from Steve Glendinning.
18) Fix off-by-one in Codel packet scheduler. From Vijay Subramanian.
19) Fix packet corruption in atl1c by disabling MSI support, from Hannes
Frederic Sowa.
20) netdev_rx_handler_unregister() needs a synchronize_net() to fix
crashes in bonding driver unload stress tests. From Eric Dumazet.
21) rxlen field of ks8851 RX packet descriptors not interpreted
correctly (it is 12 bits not 16 bits, so needs to be masked after
shifting the 32-bit value down 16 bits). Fix from Max Nekludov.
22) Fix missed RX/TX enable in sh_eth driver due to mishandling of link
change indications. From Sergei Shtylyov.
23) Fix crashes during spurious ECI interrupts in sh_eth driver, also
from Sergei Shtylyov.
24) dm9000 driver initialization is done wrong for revision B devices
with DSP PHY, from Joseph CHANG.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (53 commits)
DM9000B: driver initialization upgrade
sh_eth: make 'link' field of 'struct sh_eth_private' *int*
sh_eth: workaround for spurious ECI interrupt
sh_eth: fix handling of no LINK signal
ks8851: Fix interpretation of rxlen field.
net: add a synchronize_net() in netdev_rx_handler_unregister()
MAINTAINERS: Update netxen_nic maintainers list
atl1e: drop pci-msi support because of packet corruption
net: fq_codel: Fix off-by-one error
net: calxedaxgmac: Wake-on-LAN fixes
net: calxedaxgmac: fix rx ring handling when OOM
net: core: Remove redundant call to 'nf_reset' in 'dev_forward_skb'
smsc75xx: fix jumbo frame support
net: fix the use of this_cpu_ptr
bonding: fix disabling of arp_interval and miimon
ipv6: don't accept node local multicast traffic from the wire
sky2: Threshold for Pause Packet is set wrong
sky2: Receive Overflows not counted
aoe: reserve enough headroom on skbs
line up comment for ndo_bridge_getlink
...
Some SPI slave devices require asserted chip select signal across
multiple transfer segments of an SPI message. Currently the driver
always de-asserts the internal SS signal for every single transfer
segment of the message and ignores the 'cs_change' flag of the
transfer description. Disable the internal chip select (SS) only
if this is needed and indicated by the 'cs_change' flag.
Without this change, each partial transfer of a surrounding
multi-part SPI transaction might erroneously change the SS
signal, which might prevent slaves from answering the request
that was sent in a previous transfer segment because the
transaction could be considered aborted (SS was de-asserted
before reading the response).
Reported-by: Gerhard Sittig <gerhard.sittig@ifm.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org
spi_pump_messages() calls into a controller driver with
unprepare_transfer_hardware() which is documented as "This may sleep".
As in the prepare_transfer_hardware() call below, we should release the
queue_lock spinlock before making the call.
Rework the logic a bit to hold queue_lock to protect the 'busy' flag,
then release it to call unprepare_transfer_hardware().
Signed-off-by: Bryan Freed <bfreed@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>
Signed-off-by: Doug Anderson <dianders@chromium.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
The status of the interrupt is available in the status register,
so reading the clear pending register and writing back the same
value will not actually clear the pending interrupts. This patch
modifies the interrupt handler to read the status register and
clear the corresponding pending bit in the clear pending register.
Modified the hwInit function to clear all the pending interrupts.
Signed-off-by: Girish K S <ks.giri@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org
Fix bug for DM9000 revision B which contain a DSP PHY
DM9000B use DSP PHY instead previouse DM9000 revisions' analog PHY,
So need extra change in initialization, For
explicity PHY Reset and PHY init parameter, and
first DM9000_NCR reset need NCR_MAC_LBK bit by dm9000_probe().
Following DM9000_NCR reset cause by dm9000_open() clear the
NCR_MAC_LBK bit.
Without this fix, Power-up FIFO pointers error happen around 2%
rate among Davicom's customers' boards. With this fix, All above
cases can be solved.
Signed-off-by: Joseph CHANG <josright123@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The 'link' field of 'struct sh_eth_private' has type 'enum phy_state' while the
'link' field of 'struct phy_device' is merely *int* (having values 0 and 1) and
the former field gets assigned from the latter. Make the field match, getting
rid of incorrectly used PHY_DOWN value in assignments/comparisons.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
At least on Renesas R8A7778, EESR.ECI interrupt seems to fire regardless of its
mask in EESIPR register. I can 100% reproduce it with the following scenario:
target is booted with 'ip=on' option, and so IP-Config opens SoC Ether device
but doesn't get a proper reply and then succeeds with on-board SMC chip; then
I login and try to bring up the SoC Ether device with 'ifconfig', and I get
an ECI interrupt once request_irq() is called by sh_eth_open() (while interrupt
mask in EESIPR register is all 0), if that interrupt is accompanied by a pending
EESR.FRC (frame receive completion) interrupt, I get kernel oops in sh_eth_rx()
because sh_eth_ring_init() hasn't been called yet!
The solution I worked out is the following: in sh_eth_interrupt(), mask the
interrupt status from EESR register with the interrupt mask from EESIPR register
in order not to handle the disabled interrupts -- but forcing EESIPR.M_ECI bit
in this mask set because we always need to fully handle EESR.ECI interrupt in
sh_eth_error() in order to quench it (as it doesn't get cleared by just writing
1 to the this bit as all the other interrupts).
While at it, remove unneeded initializer for 'intr_status' variable and give it
*unsigned long* type, matching the type of sh_eth_read()'s result; fix comment.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Max Filippov <max.filippov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The code handling the absent LINK signal (or the absent PSR register -- which
reflects the state of this signal) is quite naive and has probably never really
worked. It's probably enough to say that this code is executed only on the LINK
change interrupt (sic!) but even if we actually have the signal and choose to
ignore it (it might be connected to PHY's link/activity LED output as on the
Renesas BOCK-W board), sh_eth_adjust_link() on which this code relies to update
'mdp->link' gets executed later than the LINK change interrupt where it is
checked, and so RX/TX never get enabled via ECMR register.
So, ignore the LINK changed interrupt iff LINK signal is absent (or just chosen
not to be used) or PSR register is absent, and enable/disable RX/TX directly in
sh_eth_adjust_link() in this case.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull slave-dmaengine fixes from Vinod Koul:
"Two fixes for slave-dmaengine.
The first one is for making slave_id value correct for dw_dmac and
the other one fixes the endieness in DT parsing"
* 'fixes' of git://git.infradead.org/users/vkoul/slave-dma:
dw_dmac: adjust slave_id accordingly to request line base
dmaengine: dw_dma: fix endianess for DT xlate function
Pull media fixes from Mauro Carvalho Chehab:
"For a some fixes for Kernel 3.9:
- subsystem build fix when VIDEO_DEV=y, VIDEO_V4L2=m and I2C=m
- compilation fix for arm multiarch preventing IR_RX51 to be selected
- regression fix at bttv crop logic
- s5p-mfc/m5mols/exynos: a few fixes for cameras on exynos hardware"
* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
[media] [REGRESSION] bt8xx: Fix too large height in cropcap
[media] fix compilation with both V4L2 and I2C as 'm'
[media] m5mols: Fix bug in stream on handler
[media] s5p-fimc: Do not attempt to disable not enabled media pipeline
[media] s5p-mfc: Fix encoder control 15 issue
[media] s5p-mfc: Fix frame skip bug
[media] s5p-fimc: send valid m2m ctx to fimc_m2m_job_finish
[media] exynos-gsc: send valid m2m ctx to gsc_m2m_job_finish
[media] fimc-lite: Fix the variable type to avoid possible crash
[media] fimc-lite: Initialize 'step' field in fimc_lite_ctrl structure
[media] ir: IR_RX51 only works on OMAP2
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=u2RH
-----END PGP SIGNATURE-----
Merge tag 'for-linus-20130331' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
"Alright, this time from 10K up in the air.
Collection of fixes that have been queued up since the merge window
opened, hence postponed until later in the cycle. The pull request
contains:
- A bunch of fixes for the xen blk front/back driver.
- A round of fixes for the new IBM RamSan driver, fixing various
nasty issues.
- Fixes for multiple drives from Wei Yongjun, bad handling of return
values and wrong pointer math.
- A fix for loop properly killing partitions when being detached."
* tag 'for-linus-20130331' of git://git.kernel.dk/linux-block: (25 commits)
mg_disk: fix error return code in mg_probe()
rsxx: remove unused variable
rsxx: enable error return of rsxx_eeh_save_issued_dmas()
block: removes dynamic allocation on stack
Block: blk-flush: Fixed indent code style
cciss: fix invalid use of sizeof in cciss_find_cfgtables()
loop: cleanup partitions when detaching loop device
loop: fix error return code in loop_add()
mtip32xx: fix error return code in mtip_pci_probe()
xen-blkfront: remove frame list from blk_shadow
xen-blkfront: pre-allocate pages for requests
xen-blkback: don't store dev_bus_addr
xen-blkfront: switch from llist to list
xen-blkback: fix foreach_grant_safe to handle empty lists
xen-blkfront: replace kmalloc and then memcpy with kmemdup
xen-blkback: fix dispatch_rw_block_io() error path
rsxx: fix missing unlock on error return in rsxx_eeh_remap_dmas()
Adding in EEH support to the IBM FlashSystem 70/80 device driver
block: IBM RamSan 70/80 error message bug fix.
block: IBM RamSan 70/80 branding changes.
...
This reverts commit 6aa9707099.
Commit 6aa9707099 ("lockdep: check that no locks held at freeze time")
causes problems with NFS root filesystems. The failures were noticed on
OMAP2 and 3 boards during kernel init:
[ BUG: swapper/0/1 still has locks held! ]
3.9.0-rc3-00344-ga937536 #1 Not tainted
-------------------------------------
1 lock held by swapper/0/1:
#0: (&type->s_umount_key#13/1){+.+.+.}, at: [<c011e84c>] sget+0x248/0x574
stack backtrace:
rpc_wait_bit_killable
__wait_on_bit
out_of_line_wait_on_bit
__rpc_execute
rpc_run_task
rpc_call_sync
nfs_proc_get_root
nfs_get_root
nfs_fs_mount_common
nfs_try_mount
nfs_fs_mount
mount_fs
vfs_kern_mount
do_mount
sys_mount
do_mount_root
mount_root
prepare_namespace
kernel_init_freeable
kernel_init
Although the rootfs mounts, the system is unstable. Here's a transcript
from a PM test:
http://www.pwsan.com/omap/testlogs/test_v3.9-rc3/20130317194234/pm/37xxevm/37xxevm_log.txt
Here's what the test log should look like:
http://www.pwsan.com/omap/testlogs/test_v3.8/20130218214403/pm/37xxevm/37xxevm_log.txt
Mailing list discussion is here:
http://lkml.org/lkml/2013/3/4/221
Deal with this for v3.9 by reverting the problem commit, until folks can
figure out the right long-term course of action.
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Mandeep Singh Baines <msb@chromium.org>
Cc: Jeff Layton <jlayton@redhat.com>
Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: <maciej.rutecki@gmail.com>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Ben Chan <benchan@chromium.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull SCSI target fixes from Nicholas Bellinger:
"This includes the bug-fix for a >= v3.8-rc1 regression specific to
iscsi-target persistent reservation conflict handling (CC'ed to
stable), and a tcm_vhost patch to drop VIRTIO_RING_F_EVENT_IDX usage
so that in-flight qemu vhost-scsi-pci device code can detect the
proper vhost feature bits.
Also, there are two more tcm_vhost patches still being discussed by
MST and Asias for v3.9 that will be required for the in-flight qemu
vhost-scsi-pci device patch to function properly, and that should
(hopefully) be the last target fixes for this round."
* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
target: Fix RESERVATION_CONFLICT status regression for iscsi-target special case
tcm_vhost: Avoid VIRTIO_RING_F_EVENT_IDX feature bit
Since commit 0536bdf33f (ARM: move iotable mappings within the vmalloc
region), the Cavium CNS3xxx cannot boot anymore.
This is caused by the pre-defined iotable mappings is not in the vmalloc
region. This patch move the iotable mappings into the vmalloc region, and
merge the MPCore private memory region (containing the SCU, the GIC and
the TWD) as a single region.
Signed-off-by: Mac Lin <mkl0301@gmail.com>
Signed-off-by: Anton Vorontsov <anton@enomsg.org>
Cc: stable@vger.kernel.org [v3.3+]
When multiple ovq operations are being performed (lots of open/close
operations on virtio_console fds), the __send_control_msg() function can
get confused without locking.
A simple recipe to cause badness is:
* create a QEMU VM with two virtio-serial ports
* in the guest, do
while true;do echo abc >/dev/vport0p1;done
while true;do echo edf >/dev/vport0p2;done
In one run, this caused a panic in __send_control_msg(). In another, I
got
virtio_console virtio0: control-o:id 0 is not a head!
This also results repeated messages similar to these on the host:
qemu-kvm: virtio-serial-bus: Unexpected port id 478762112 for device virtio-serial-bus.0
qemu-kvm: virtio-serial-bus: Unexpected port id 478762368 for device virtio-serial-bus.0
Reported-by: FuXiangChun <xfu@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Reviewed-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Reviewed-by: Asias He <asias@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org
The cvq_lock was taken for the c_ivq. Rename the lock to make that
obvious.
We'll also add a lock around the c_ovq in the next commit, so there's no
ambiguity.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Reviewed-by: Asias He <asias@redhat.com>
Reviewed-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org
On some hardware configurations we have got the request line with the offset.
The patch introduces convert_slave_id() helper for that cases. The request line
base is came from the driver data provided by the platform_device_id table.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
As reported by Wu Fengguang's build robot tracking sparse warnings, the
dma_spec arguments in the dw_dma_xlate are already byte swapped on
little-endian platforms and must not get swapped again. This code is
currently not used anywhere, but will be used in Linux 3.10 when the
ARM SPEAr platform starts using the generic DMA DT binding.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
The Adam Belay's e-mail address in MAINTAINERS under PNP SUPPORT
is not valid any more and I started to maintain that code in the
meantime as a matter of fact, so list myself as a maintainer of it
along with Bjorn and remove the Adam's entry from it.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
According to the Datasheet (page 52):
15-12 Reserved
11-0 RXBC Receive Byte Count
This field indicates the present received frame byte size.
The code has a bug:
rxh = ks8851_rdreg32(ks, KS_RXFHSR);
rxstat = rxh & 0xffff;
rxlen = rxh >> 16; // BUG!!! 0xFFF mask should be applied
Signed-off-by: Max Nekludov <Max.Nekludov@us.elster.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 35d48903e9 (bonding: fix rx_handler locking) added a race
in bonding driver, reported by Steven Rostedt who did a very good
diagnosis :
<quoting Steven>
I'm currently debugging a crash in an old 3.0-rt kernel that one of our
customers is seeing. The bug happens with a stress test that loads and
unloads the bonding module in a loop (I don't know all the details as
I'm not the one that is directly interacting with the customer). But the
bug looks to be something that may still be present and possibly present
in mainline too. It will just be much harder to trigger it in mainline.
In -rt, interrupts are threads, and can schedule in and out just like
any other thread. Note, mainline now supports interrupt threads so this
may be easily reproducible in mainline as well. I don't have the ability
to tell the customer to try mainline or other kernels, so my hands are
somewhat tied to what I can do.
But according to a core dump, I tracked down that the eth irq thread
crashed in bond_handle_frame() here:
slave = bond_slave_get_rcu(skb->dev);
bond = slave->bond; <--- BUG
the slave returned was NULL and accessing slave->bond caused a NULL
pointer dereference.
Looking at the code that unregisters the handler:
void netdev_rx_handler_unregister(struct net_device *dev)
{
ASSERT_RTNL();
RCU_INIT_POINTER(dev->rx_handler, NULL);
RCU_INIT_POINTER(dev->rx_handler_data, NULL);
}
Which is basically:
dev->rx_handler = NULL;
dev->rx_handler_data = NULL;
And looking at __netif_receive_skb() we have:
rx_handler = rcu_dereference(skb->dev->rx_handler);
if (rx_handler) {
if (pt_prev) {
ret = deliver_skb(skb, pt_prev, orig_dev);
pt_prev = NULL;
}
switch (rx_handler(&skb)) {
My question to all of you is, what stops this interrupt from happening
while the bonding module is unloading? What happens if the interrupt
triggers and we have this:
CPU0 CPU1
---- ----
rx_handler = skb->dev->rx_handler
netdev_rx_handler_unregister() {
dev->rx_handler = NULL;
dev->rx_handler_data = NULL;
rx_handler()
bond_handle_frame() {
slave = skb->dev->rx_handler;
bond = slave->bond; <-- NULL pointer dereference!!!
What protection am I missing in the bond release handler that would
prevent the above from happening?
</quoting Steven>
We can fix bug this in two ways. First is adding a test in
bond_handle_frame() and others to check if rx_handler_data is NULL.
A second way is adding a synchronize_net() in
netdev_rx_handler_unregister() to make sure that a rcu protected reader
has the guarantee to see a non NULL rx_handler_data.
The second way is better as it avoids an extra test in fast path.
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jiri Pirko <jpirko@redhat.com>
Cc: Paul E. McKenney <paulmck@us.ibm.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o Add myself to netxen_nic maintainers list
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Usage of pci-msi results in corrupted dma packet transfers to the host.
Reported-by: rebelyouth <rebelyouth.hacklab@gmail.com>
Cc: Huang, Xiong <xiong@qca.qualcomm.com>
Tested-by: Christian Sünkenberg <christian.suenkenberg@student.kit.edu>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, we hold a max of sch->limit -1 number of packets instead of
sch->limit packets. Fix this off-by-one error.
Signed-off-by: Vijay Subramanian <subramanian.vijay@gmail.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
WOL is broken because the magic packet status bit is getting set rather
than the enable bit. The PMT interrupt is not getting serviced because
the PMT interrupt is also enabled on the global interrupt, but not
cleared by the global interrupt and the global interrupt is higher
priority. This fixes both of these issues to get WOL working.
There's still a problem with receive after resume, but at least now we
can wake-up.
Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If skb allocation for the rx ring fails repeatedly, we can reach a point
were the ring is empty. In this condition, the driver is out of sync with
the h/w. While this has always been possible, the removal of the skb
recycling seems to have made triggering this problem easier.
Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
'nf_reset' is called just prior calling 'netif_rx'.
No need to call it twice.
Reported-by: Igor Michailov <rgohita@gmail.com>
Signed-off-by: Shmulik Ladkani <shmulik.ladkani@gmail.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch enables RX of jumbo frames for LAN7500.
Previously the driver would transmit jumbo frames succesfully but
would drop received jumbo frames (incrementing the interface errors
count).
With this patch applied the device can succesfully receive jumbo
frames up to MTU 9000 (9014 bytes on the wire including ethernet
header).
Signed-off-by: Steve Glendinning <steve.glendinning@shawell.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
flush_tasklet is not percpu var, and percpu is percpu var, and
this_cpu_ptr(&info->cache->percpu->flush_tasklet)
is not equal to
&this_cpu_ptr(info->cache->percpu)->flush_tasklet
1f743b076(use this_cpu_ptr per-cpu helper) introduced this bug.
Signed-off-by: Li RongQing <roy.qing.li@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current Tilera boot infrastructure now provides the initramfs
to Linux as a Tilera-hypervisor file named "initramfs", rather than
"initramfs.cpio.gz", as before. (This makes it reasonable to use
other compression techniques than gzip on the file without having to
worry about the name causing confusion.) Adapt to use the new name,
but also fall back to checking for the old name.
Cc'ing to stable so that older kernels will remain compatible with
newer Tilera boot infrastructure.
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
Cc: stable@vger.kernel.org
Currently if either arp_interval or miimon is disabled, they both get
disabled, and upon disabling they get executed once more which is not
the proper behaviour. Also when doing a no-op and disabling an already
disabled one, the other again gets disabled.
Also fix the error messages with the proper valid ranges, and a small
typo fix in the up delay error message (outputting "down delay", instead
of "up delay").
Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Erik Hugne's errata proposal (Errata ID: 3480) to RFC4291 has been
verified: http://www.rfc-editor.org/errata_search.php?eid=3480
We have to check for pkt_type and loopback flag because either the
packets are allowed to travel over the loopback interface (in which case
pkt_type is PACKET_HOST and IFF_LOOPBACK flag is set) or they travel
over a non-loopback interface back to us (in which case PACKET_TYPE is
PACKET_LOOPBACK and IFF_LOOPBACK flag is not set).
Cc: Erik Hugne <erik.hugne@ericsson.com>
Cc: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The sky2 driver sets the Rx Upper Threshold for Pause Packet generation to a
wrong value which leads to only 2kB of RAM remaining space. This can lead to
Rx overflow errors even with activated flow-control.
Fix: We should increase the value to 8192/8
Signed-off-by: Mirko Lindner <mlindner@marvell.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The sky2 driver doesn't count the Receive Overflows because the MAC
interrupt for this event is not set in the MAC's interrupt mask.
The MAC's interrupt mask is set only for Transmit FIFO Underruns.
Fix: The correct setting should be (GM_IS_TX_FF_UR | GM_IS_RX_FF_OR)
Otherwise the Receive Overflow event will not generate any interrupt.
The Receive Overflow interrupt is handled correctly
Signed-off-by: Mirko Lindner <mlindner@marvell.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull ceph fix from Sage Weil:
"This fixes a regression introduced during the last merge window when
mapping non-existent images."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
rbd: don't zero-fill non-image object requests
A result of ENOENT from a read request for an object that's part of
an rbd image indicates that there is a hole in that portion of the
image. Similarly, a short read for such an object indicates that
the remainder of the read should be interpreted a full read with
zeros filling out the end of the request.
This behavior is not correct for objects that are not backing rbd
image data. Currently rbd_img_obj_request_callback() assumes it
should be done for all objects.
Change rbd_img_obj_request_callback() so it only does this zeroing
for image objects. Encapsulate that special handling in its own
function. Add an assertion that the image object request is a bio
request, since we assume that (and we currently don't support any
other types).
This resolves a problem identified here:
http://tracker.ceph.com/issues/4559
The regression was introduced by bf0d5f503d.
Reported-by: Dan van der Ster <dan@vanderster.com>
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-off-by: Sage Weil <sage@inktank.com>
Pull btrfs fixes from Chris Mason:
"We've had a busy two weeks of bug fixing. The biggest patches in here
are some long standing early-enospc problems (Josef) and a very old
race where compression and mmap combine forces to lose writes (me).
I'm fairly sure the mmap bug goes all the way back to the introduction
of the compression code, which is proof that fsx doesn't trigger every
possible mmap corner after all.
I'm sure you'll notice one of these is from this morning, it's a small
and isolated use-after-free fix in our scrub error reporting. I
double checked it here."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: don't drop path when printing out tree errors in scrub
Btrfs: fix wrong return value of btrfs_lookup_csum()
Btrfs: fix wrong reservation of csums
Btrfs: fix double free in the btrfs_qgroup_account_ref()
Btrfs: limit the global reserve to 512mb
Btrfs: hold the ordered operations mutex when waiting on ordered extents
Btrfs: fix space accounting for unlink and rename
Btrfs: fix space leak when we fail to reserve metadata space
Btrfs: fix EIO from btrfs send in is_extent_unchanged for punched holes
Btrfs: fix race between mmap writes and compression
Btrfs: fix memory leak in btrfs_create_tree()
Btrfs: fix locking on ROOT_REPLACE operations in tree mod log
Btrfs: fix missing qgroup reservation before fallocating
Btrfs: handle a bogus chunk tree nicely
Btrfs: update to use fs_state bit
Commit 3e7fc708eb ("ia64 idle: delete pm_idle") in 3.9-rc1 didn't
finish the job, leaving an un-initialized reference to (*idle)().
[ Haven't seen a crash from this - but seems like we are just being
lucky that "idle" is zero so it does get initialized before we jump to
randomland - Len ]
Reported-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull arc architecture fixes from Vineet Gupta:
"This includes fix for a serious bug in DMA mapping API, make
allyesconfig wreckage, removal of bogus email-list placeholder in
MAINTAINERS, a typo in ptrace helper code and last remaining changes
for syscall ABI v3 which we are finally starting to transition-to
internally.
The request is late than I intended to - but I was held up with
debugging a timer link list corruption, for which a proposed fix to
generic timer code was sent out to lkml/tglx earlier today."
* 'for-curr' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
ARC: Fix the typo in event identifier flags used by ptrace
arc: fix dma_address assignment during dma_map_sg()
ARC: Remove SET_PERSONALITY (tracks cross-arch change)
ARC: ABIv3: fork/vfork wrappers not needed in "no-legacy-syscall" ABI
ARC: ABIv3: Print the correct ABI ver
ARC: make allyesconfig build breakages
ARC: MAINTAINERS update for ARC
After commit 21d8a15a (lookup_one_len: don't accept . and ..) reiserfs
started failing to delete xattrs from inode. This was due to a buggy
test for '.' and '..' in fill_with_dentries() which resulted in passing
'.' and '..' entries to lookup_one_len() in some cases. That returned
error and so we failed to iterate over all xattrs of and inode.
Fix the test in fill_with_dentries() along the lines of the one in
lookup_one_len().
Reported-by: Pawel Zawora <pzawora@gmail.com>
CC: stable@vger.kernel.org
Signed-off-by: Jan Kara <jack@suse.cz>
John W. Linville
Thierry Reding
Yinghai Lu
Linus Torvalds
Anatolij Gustschin
Bryan Freed
Girish K S
Joseph CHANG
Sergei Shtylyov
Samuel Ortiz
Mark Brown
Paul Walmsley
Mac Lin
Amit Shah
Andy Shevchenko
Arnd Bergmann
Rafael J. Wysocki
Max.Nekludov@us.elster.com
Eric Dumazet
Manish Chopra
Hannes Frederic Sowa
Vijay Subramanian
Rob Herring
Shmulik Ladkani
Steve Glendinning
Li RongQing
Chris Metcalf
nikolay@redhat.com
Mirko Lindner
Alex Elder
Len Brown
Jan Kara