We were taking dcache_lock inside of i_lock, which introduces a dependency
not found elsewhere in the kernel, complicationg the vfs locking
scalability work. Since we don't actually need it here anyway, remove
it.
We only need i_lock to test for the I_COMPLETE flag, so be careful to do
so without dcache_lock held.
Signed-off-by: Sage Weil <sage@newdream.net>
Convert a sequence of kmalloc and memcpy to use kmemdup.
The semantic patch that performs this transformation is:
(http://coccinelle.lip6.fr/)
// <smpl>
@@
expression a,flag,len;
expression arg,e1,e2;
statement S;
@@
a =
- \(kmalloc\|kzalloc\)(len,flag)
+ kmemdup(arg,len,flag)
<... when != a
if (a == NULL || ...) S
...>
- memcpy(a,arg,len+1);
// </smpl>
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Sage Weil <sage@newdream.net>
We should be passing "buf" here insead of "bv". This is tricky because
it's not the same as kmap() and kunmap(). GCC does warn about it if you
compile on i386 with CONFIG_HIGHMEM.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Sage Weil <sage@newdream.net>
ceph_alloc_page_vector() returns ERR_PTR(-ENOMEM) on errors.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Sage Weil <sage@newdream.net>
This only happened when parse_extra_token was not passed
to ceph_parse_option() (hence, only happened in rbd).
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Include "super.h" outside of CONFIG_DEBUG_FS to eliminate a compiler warning:
fs/ceph/debugfs.c:266: warning: 'struct ceph_fs_client' declared inside parameter list
fs/ceph/debugfs.c:266: warning: its scope is only this definition or declaration, which is probably not what you want
fs/ceph/debugfs.c:271: warning: 'struct ceph_fs_client' declared inside parameter list
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
rbd_client_create() doesn't free rbdc, this leads to many leaks.
seg_len in rbd_do_op() is unsigned, so (seg_len < 0) makes no sense.
Also if fixed check fails then seg_name is leaked.
Signed-off-by: Vasiliy Kulikov <segooon@gmail.com>
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Switch from using the BKL explicitly to the new lock_flocks() interface.
Eventually this will turn into a spinlock.
Signed-off-by: Sage Weil <sage@newdream.net>
When the lock_kernel() turns into lock_flocks() and a spinlock, we won't
be able to do allocations with the lock held. Preallocate space without
the lock, and retry if the lock state changes out from underneath us.
Signed-off-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
These facilitate preallocation of pages so that we can encode into the pagelist
in an atomic context.
Signed-off-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
The i_rdcache_gen value only implies we MAY have cached pages; actually
check the mapping to see if it's worth bothering with an invalidate.
Signed-off-by: Sage Weil <sage@newdream.net>
The rados block device (rbd), based on osdblk, creates a block device
that is backed by objects stored in the Ceph distributed object storage
cluster. Each device consists of a single metadata object and data
striped over many data objects.
The rbd driver supports read-only snapshots.
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
This factors out protocol and low-level storage parts of ceph into a
separate libceph module living in net/ceph and include/linux/ceph. This
is mostly a matter of moving files around. However, a few key pieces
of the interface change as well:
- ceph_client becomes ceph_fs_client and ceph_client, where the latter
captures the mon and osd clients, and the fs_client gets the mds client
and file system specific pieces.
- Mount option parsing and debugfs setup is correspondingly broken into
two pieces.
- The mon client gets a generic handler callback for otherwise unknown
messages (mds map, in this case).
- The basic supported/required feature bits can be expanded (and are by
ceph_fs_client).
No functional change, aside from some subtle error handling cases that got
cleaned up in the refactoring process.
Signed-off-by: Sage Weil <sage@newdream.net>
Allow the messenger to send/receive data in a bio. This is added
so that we wouldn't need to copy the data into pages or some other buffer
when doing IO for an rbd block device.
We can now have trailing variable sized data for osd
ops. Also osd ops encoding is more modular.
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
The osd requests creation are being decoupled from the
vino parameter, allowing clients using the osd to use
other arbitrary object names that are not necessarily
vino based. Also, calc_raw_layout now takes a snap id.
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
Implement a pool lookup by name. This will be used by rbd.
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
Currently flush tlb vector allocation is based on below equation:
sender = smp_processor_id() % 8
This isn't optimal, CPUs from different node can have the same vector, this
causes a lot of lock contention. Instead, we can assign the same vectors to
CPUs from the same node, while different node has different vectors. This has
below advantages:
a. if there is lock contention, the lock contention is between CPUs from one
node. This should be much cheaper than the contention between nodes.
b. completely avoid lock contention between nodes. This especially benefits
kswapd, which is the biggest user of tlb flush, since kswapd sets its affinity
to specific node.
In my test, this could reduce > 20% CPU overhead in extreme case.The test
machine has 4 nodes and each node has 16 CPUs. I then bind each node's kswapd
to the first CPU of the node. I run a workload with 4 sequential mmap file
read thread. The files are empty sparse file. This workload will trigger a
lot of page reclaim and tlbflush. The kswapd bind is to easy trigger the
extreme tlb flush lock contention because otherwise kswapd keeps migrating
between CPUs of a node and I can't get stable result. Sure in real workload,
we can't always see so big tlb flush lock contention, but it's possible.
[ hpa: folded in fix from Eric Dumazet to use this_cpu_read() ]
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
LKML-Reference: <1287544023.4571.8.camel@sli10-conroe.sh.intel.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Add a new readmostly percpu section and API. This can be used to
avoid dirtying data lines which are generally not written to, which is
especially important for data which may be accessed by processors
other than the one for which the percpu area belongs to.
[ hpa: moved it *after* the page-aligned section, for obvious
reasons. ]
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
LKML-Reference: <1287544022.4571.7.camel@sli10-conroe.sh.intel.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
* 'upstream' of git://git.linux-mips.org/pub/scm/upstream-linus:
MIPS: O32 compat/N32: Fix to use compat syscall wrappers for AIO syscalls.
MAINTAINERS: Change list for ioc_serial to linux-serial.
SERIAL: ioc3_serial: Return -ENOMEM on memory allocation failure
MIPS: jz4740: Fix Kbuild Platform file.
MIPS: Repair Kbuild make clean breakage.
If the host is slow in reading data or doesn't read data at all,
blocking write calls not only blocked the program that called write()
but the entire guest itself.
To overcome this, let's not block till the host signals it has given
back the virtio ring element we passed it. Instead, send the buffer to
the host and return to userspace. This operation then becomes similar
to how non-blocking writes work, so let's use the existing code for this
path as well.
This code change also ensures blocking write calls do get blocked if
there's not enough room in the virtio ring as well as they don't return
-EAGAIN to userspace.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Acked-by: Hans de Goede <hdegoede@redhat.com>
CC: stable@kernel.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
arch/x86/mm/fault.c: In function 'vmalloc_sync_all':
arch/x86/mm/fault.c:238: warning: assignment makes integer from pointer without a cast
introduced by 617d34d9e5.
Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
LKML-Reference: <20101020103642.GA3135@kryptos.osrc.amd.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
In case of TX only with DMA, the driver assumes that the data
has been transferred once DMA callback in invoked. However,
SPI's shift register may still contain data. Thus, the driver
is supposed to verify that the register is empty and the end of
the SPI transfer has been reached.
Signed-off-by: Ilkka Koskinen <ilkka.koskinen@nokia.com>
Tested-by: Tuomas Katila <ext-tuomas.2.katila@nokia.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
In the TX_ONLY transfer, the SPI controller also receives data
simultaneously and saves them in the rx register. After the TX_ONLY
transfer, the rx register will hold the random data received during
the last tx transaction.
If the direct following transfer is RX_ONLY, this random data has the
possibility to affect this transfer like this:
When the SPI controller is changed from TX_ONLY to RX_ONLY,
the random data makes the rx register full immediately and
triggers a dummy write automatically(in SPI RX_ONLY transfers,
we need a dummy write to trigger the first transaction).
So the first data received in the RX_ONLY transfer will be that
random data instead of something meaningful.
We can avoid this by inserting a Disable/Re-enable toggle of the
channel after the TX_ONLY transfer, since it purges the rx register.
Signed-off-by: Jason Wang <jason77.wang@gmail.com>
Tested-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Since we're now using addruart to establish the debug mapping, we can
remove the io_pg_offst and phys_io members of struct machine_desc.
The various declarations were removed using the following script:
grep -rl MACHINE_START arch/arm | xargs \
sed -i '/MACHINE_START/,/MACHINE_END/ { /\.\(phys_io\|io_pg_offst\)/d }'
[ Initial patch was from Jeremy Kerr, example script from Russell King ]
Signed-off-by: Nicolas Pitre <nicolas.pitre@linaro.org>
Acked-by: Eric Miao <eric.miao at canonical.com>
Since we can get both physical and virtual addresses from the addruart
macro, we can use this to establish the debug mappings.
In the case of CONFIG_DEBUG_ICEDCC, we don't need any mappings, but
may still need to setup r7 correctly.
Incorporating ASM changes from Nicolas Pitre <npitre@fluxnic.net>.
Signed-off-by: Jeremy Kerr <jeremy.kerr@canonical.com>
Tested-by: Kevin Hilman <khilman@deeprootsystems.com>
Rather than checking the MMU status in every instance of addruart, do it
once in kernel/debug.S, and change the existing addruart macros to
return both physical and virtual addresses. The main debug code can then
select the appropriate address to use.
This will also allow us to retreive the address of a uart for the MMU
state that we're not current in.
Updated with fixes for OMAP from Jason Wang <jason77.wang@gmail.com>
and Tony Lindgren <tony@atomide.com>, and fix for versatile express from
Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>.
Signed-off-by: Jeremy Kerr <jeremy.kerr@canonical.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Jason Wang <jason77.wang@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Tested-by: Kevin Hilman <khilman@deeprootsystems.com>
As mentioned in the comment right at the top, the first four fields
are directly accessed by assembly code in head.S. Move nr_irqs so the
comment is true again.
Signed-off-by: Nicolas Pitre <nicolas.pitre@linaro.org>
We want the BIOS to setup the EILVT APIC registers. The offsets
were hardcoded and BIOS settings were overwritten by the OS.
Now, the subsystems for MCE threshold and IBS determine the LVT
offset from the registers the BIOS has setup. If the BIOS setup
is buggy on a family 10h system, a workaround enables IBS. If
the OS determines an invalid register setup, a "[Firmware Bug]:
" error message is reported.
We need this change also for upcomming cpu families.
Signed-off-by: Robert Richter <robert.richter@amd.com>
LKML-Reference: <1286360874-1471-3-git-send-email-robert.richter@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch implements checks for the availability of LVT entries
(APIC500-530) and reserves it if used. The check becomes
necessary since we want to let the BIOS provide the LVT offsets.
The offsets should be determined by the subsystems using it
like those for MCE threshold or IBS. On K8 only offset 0
(APIC500) and MCE interrupts are supported. Beginning with
family 10h at least 4 offsets are available.
Since offsets must be consistent for all cores, we keep track of
the LVT offsets in software and reserve the offset for the same
vector also to be used on other cores. An offset is freed by
setting the entry to APIC_EILVT_MASKED.
If the BIOS is right, there should be no conflicts. Otherwise a
"[Firmware Bug]: ..." error message is generated. However, if
software does not properly determines the offsets, it is not
necessarily a BIOS bug.
Signed-off-by: Robert Richter <robert.richter@amd.com>
LKML-Reference: <1286360874-1471-2-git-send-email-robert.richter@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There may be wakeup sources that aren't associated with any devices
and their statistics information won't be available from sysfs. Also,
for debugging purposes it is convenient to have all of the wakeup
sources statistics available from one place. For these reasons,
introduce new file "wakeup_sources" in debugfs containing those
statistics.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Take mm->page_table_lock while syncing the vmalloc region. This prevents
a race with the Xen pagetable pin/unpin code, which expects that the
page_table_lock is already held. If this race occurs, then Xen can see
an inconsistent page type (a page can either be read/write or a pagetable
page, and pin/unpin converts it between them), which will cause either
the pin or the set_p[gm]d to fail; either will crash the kernel.
vmalloc_sync_all() should be called rarely, so this extra use of
page_table_lock should not interfere with its normal users.
The mm pointer is stashed in the pgd page's index field, as that won't
be otherwise used for pgds.
Reported-by: Ian Campbell <ian.cambell@eu.citrix.com>
Originally-by: Jan Beulich <jbeulich@novell.com>
LKML-Reference: <4CB88A4C.1080305@goop.org>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
[Ralf: Michel's original patch only fixed N32; I replicated the same fix
for O32.]
Signed-off-by: Michel Thebeau <michel.thebeau@windriver.com>
Cc: paul.gortmaker@windriver.com
Cc: bruce.ashfield@windriver.com
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
IOC3 is also being used on SGI MIPS systems but this particular driver is
only being used on IA64 systems so linux-mips made no sense as a list. Pat
also thinks linux-serial@vger.kernel.org is the better list.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Sage Weil
Julia Lawall
Dan Carpenter
Greg Farnum
Yehuda Sadeh
Randy Dunlap
Vasiliy Kulikov
Shaohua Li
Linus Torvalds
Amit Shah
Borislav Petkov
Ilkka Koskinen
Jason Wang
Nicolas Pitre
Jeremy Kerr
Robert Richter
Ingo Molnar
Rafael J. Wysocki
Russell King
Jeremy Fitzhardinge
Michel Thebeau
Ralf Baechle