A recursive lockdep warning occurs if you call
regulator_set_optimum_mode() on a regulator with a supply because
there is no nesting annotation for the rdev->mutex. To avoid this
warning, get the supply's load before locking the regulator's
mutex to avoid grabbing the same class of lock twice.
=============================================
[ INFO: possible recursive locking detected ]
3.4.0 #3257 Tainted: G W
---------------------------------------------
swapper/0/1 is trying to acquire lock:
(&rdev->mutex){+.+.+.}, at: [<c036e9e0>] regulator_get_voltage+0x18/0x38
but task is already holding lock:
(&rdev->mutex){+.+.+.}, at: [<c036ef38>] regulator_set_optimum_mode+0x24/0x224
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&rdev->mutex);
lock(&rdev->mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
3 locks held by swapper/0/1:
#0: (&__lockdep_no_validate__){......}, at: [<c03dbb48>] __driver_attach+0x40/0x8c
#1: (&__lockdep_no_validate__){......}, at: [<c03dbb58>] __driver_attach+0x50/0x8c
#2: (&rdev->mutex){+.+.+.}, at: [<c036ef38>] regulator_set_optimum_mode+0x24/0x224
stack backtrace:
[<c001521c>] (unwind_backtrace+0x0/0x12c) from [<c00cc4d4>] (validate_chain+0x760/0x1080)
[<c00cc4d4>] (validate_chain+0x760/0x1080) from [<c00cd744>] (__lock_acquire+0x950/0xa10)
[<c00cd744>] (__lock_acquire+0x950/0xa10) from [<c00cd990>] (lock_acquire+0x18c/0x1e8)
[<c00cd990>] (lock_acquire+0x18c/0x1e8) from [<c080c248>] (mutex_lock_nested+0x68/0x3c4)
[<c080c248>] (mutex_lock_nested+0x68/0x3c4) from [<c036e9e0>] (regulator_get_voltage+0x18/0x38)
[<c036e9e0>] (regulator_get_voltage+0x18/0x38) from [<c036efb8>] (regulator_set_optimum_mode+0xa4/0x224)
...
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
The below commit introduced a bug in __clk_set_parent()
which could cause it to *skip* the parent validation
which makes sure the parent passed to the api is a valid
one.
commit 7975059db5
Author: Rajendra Nayak <rnayak@ti.com>
Date: Wed Jun 6 14:41:31 2012 +0530
clk: Allow late cache allocation for clk->parents
This was identified by the following compiler warning..
drivers/clk/clk.c: In function '__clk_set_parent':
drivers/clk/clk.c:1083:5: warning: 'i' may be used uninitialized in this function [-Wuninitialized]
.. as reported by Marc Kleine-Budde.
There were various options discussed on how to fix this, one
being initing 'i' to clk->num_parents, but the below approach
was found to be more appropriate as it also makes the 'parent
validation' code simpler to read.
Reported-by: Marc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Mike Turquette <mturquette@linaro.org>
Cc: stable@kernel.org
Just a few driver-specific fixes for ASoC and HD-audio.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
iQIcBAABAgAGBQJP8utuAAoJEGwxgFQ9KSmkG9kQAI3mh986pLpPCnSQq76KOvRJ
1zPk+X+zAAGKJjbPGidZjH1zt5CrYpH40PX/X6/mteCRY6nHdeIL0NQwfThAQvxs
+7xwjf35tlpS9+atvGTlSEWkDB3XgQ+llUixfuN5dZ5QAdA1hDABHlJqZNxuNHa0
NBgQ0UghAWUvSZ3ibf2Tv6d3CH4IRAvyATEqaTz20m6bnv3kUFBfCtuT9BbMg0Wm
Y7dPWylSgr8ERjHpSGIov/hpi9+AtJgxudRS4bOMgUarDqmGLoA2GMlH6wTQ8lHW
HSZywRBJFXlYYhBvKbM+hf48ghN0CeLlRXKmHmOBoEafPTYso+B6+cvloK2g07qo
V+R65jCgXqTu2eaRvSkzn+t1p+K0ZwWqcfniFJ4n7RKi2nx+yKztBDDG0ZdciH+Y
AlKDyRvRoqavSTajrpr9Ii90Q1bm6yVRguQj0AwFpeCPpmvBHtr/uHafwWYuZVF2
7oVAHKt39YsYxJYGiEARRieBUSHkwLDOqzys1z1+0fc+widO91nHh52aVZ+eALOi
BIi7W86eLoQt7ltM817P7jaz0R61Cg/PsRHt8ZbpySkTJELINS8A2Kuqnf/nYchx
o1c1vbcoTjWUnHJBmdbI13WrmOu+tOT8X+YwHdMTgtLt4BVKq/+oJ1hB8Om+QXpe
V8qtG0RII+Hxt7ecQrcC
=iN3h
-----END PGP SIGNATURE-----
Merge tag 'sound-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"Just a few driver-specific fixes for ASoC and HD-audio."
* tag 'sound-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda - Fix no sound from ALC662 after Windows reboot
ASoC: tlv320aic3x: Fix codec pll configure bug
ASoC: wm2200: Add missing BCLK rate
Pull drm fixes from Dave Airlie:
"One regression fix, two radeon fixes (one for an oops), and an i915
fix to unload framebuffers earlier.
We originally were going to leave the i915 fix until -next, but grub2
in some situations causes vesafb/efifb to be loaded now, and this
causes big slowdowns, and I have reports in rawhide I'd like to have
fixed."
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/i915: kick any firmware framebuffers before claiming the gtt
drm: edid: Don't add inferred modes with higher resolution
drm/radeon: fix rare segfault
drm/radeon: fix VM page table setup on SI
When the server doesn't advertise CAP_LARGE_READ_X, then MS-CIFS states
that you must cap the size of the read at the client's MaxBufferSize.
Unfortunately, testing with many older servers shows that they often
can't service a read larger than their own MaxBufferSize.
Since we can't assume what the server will do in this situation, we must
be conservative here for the default. When the server can't do large
reads, then assume that it can't satisfy any read larger than its
MaxBufferSize either.
Luckily almost all modern servers can do large reads, so this won't
affect them. This is really just for older win9x and OS/2 era servers.
Also, note that this patch just governs the default rsize. The admin can
always override this if he so chooses.
Cc: <stable@vger.kernel.org> # 3.2
Reported-by: David H. Durgee <dhdurgee@acm.org>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steven French <sfrench@w500smf.(none)>
You go away for 2 weeks vacation and what do you get when you come back?
Piles of bugs :-)
Some found by inspection, some by testing, some during use in the field,
and some while developing for the next window...
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (GNU/Linux)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=vv4r
-----END PGP SIGNATURE-----
Merge tag 'md-3.5-fixes' of git://neil.brown.name/md
Pull md fixes from NeilBrown:
"md: collection of bug fixes for 3.5
You go away for 2 weeks vacation and what do you get when you come
back? Piles of bugs :-)
Some found by inspection, some by testing, some during use in the
field, and some while developing for the next window..."
* tag 'md-3.5-fixes' of git://neil.brown.name/md:
md: fix up plugging (again).
md: support re-add of recovering devices.
md/raid1: fix bug in read_balance introduced by hot-replace
raid5: delayed stripe fix
md/raid456: When read error cannot be recovered, record bad block
md: make 'name' arg to md_register_thread non-optional.
md/raid10: fix failure when trying to repair a read error.
md/raid5: fix refcount problem when blocked_rdev is set.
md:Add blk_plug in sync_thread.
md/raid5: In ops_run_io, inc nr_pending before calling md_wait_for_blocked_rdev
md/raid5: Do not add data_offset before call to is_badblock
md/raid5: prefer replacing failed devices over want-replacement devices.
md/raid10: Don't try to recovery unmatched (and unused) chunks.
Pull security layer fixes from James Morris.
A documentation update, and a nommu build fix.
* 'for-linus2' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
security: Fix nommu build.
security: document no_new_privs
Veritysetup is now part of cryptsetup package.
Remove on-disk header description (which is not parsed in kernel)
and point users to cryptsetup where it the format is documented.
Mention units for block size paramaters.
Fix target line specification and dmsetup parameters.
Signed-off-by: Milan Broz <mbroz@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
If CONFIG_DM_DEBUG_SPACE_MAPS is enabled and memory is fragmented and a
sufficiently-large metadata device is used in a thin pool then the space
map checker will fail to allocate the memory it requires.
Switch from kmalloc to vmalloc to allow larger virtually contiguous
allocations for the space map checker's internal count arrays.
Reported-by: Vivek Goyal <vgoyal@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
If CONFIG_DM_DEBUG_SPACE_MAPS is enabled and dm_sm_checker_create()
fails, dm_tm_create_internal() would still return success even though it
cleaned up all resources it was supposed to have created. This will
lead to a kernel crash:
general protection fault: 0000 [#1] SMP DEBUG_PAGEALLOC
...
RIP: 0010:[<ffffffff81593659>] [<ffffffff81593659>] dm_bufio_get_block_size+0x9/0x20
Call Trace:
[<ffffffff81599bae>] dm_bm_block_size+0xe/0x10
[<ffffffff8159b8b8>] sm_ll_init+0x78/0xd0
[<ffffffff8159c1a6>] sm_ll_new_disk+0x16/0xa0
[<ffffffff8159c98e>] dm_sm_disk_create+0xfe/0x160
[<ffffffff815abf6e>] dm_pool_metadata_open+0x16e/0x6a0
[<ffffffff815aa010>] pool_ctr+0x3f0/0x900
[<ffffffff8158d565>] dm_table_add_target+0x195/0x450
[<ffffffff815904c4>] table_load+0xe4/0x330
[<ffffffff815917ea>] ctl_ioctl+0x15a/0x2c0
[<ffffffff81591963>] dm_ctl_ioctl+0x13/0x20
[<ffffffff8116a4f8>] do_vfs_ioctl+0x98/0x560
[<ffffffff8116aa51>] sys_ioctl+0x91/0xa0
[<ffffffff81869f52>] system_call_fastpath+0x16/0x1b
Fix the space map checker code to return an appropriate ERR_PTR and have
dm_sm_disk_create() and dm_tm_create_internal() check for it with
IS_ERR.
Reported-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cleanup the shadow table before destroying the transaction manager.
Reference: leak was identified with kmemleak when running
test_discard_random_sectors in the thinp-test-suite.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Userland sometimes sees a corrupt metadata block if metadata is changing
rapidly when a metadata snapshot is reserved for userland, To make the
problem go away, commit before we take the metadata snapshot (which is a
sensible thing to do anyway).
The checksums mean userland spots this corruption immediately so there's
no risk of acting on incorrect data. No corruption exists from the
kernel's point of view, and thin_check passes after pool shutdown.
I believe this is to do with shared blocks at the first level of the
{device, mapping} btree. Prior to the metadata-snap support no sharing
at this level was possible, so this patch is only required after commit
cc8394d86f ("dm thin: provide userspace
access to pool metadata").
Signed-off-by: Joe Thornber <ejt@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
The security + nommu configuration presently blows up with an undefined
reference to BDI_CAP_EXEC_MAP:
security/security.c: In function 'mmap_prot':
security/security.c:687:36: error: dereferencing pointer to incomplete type
security/security.c:688:16: error: 'BDI_CAP_EXEC_MAP' undeclared (first use in this function)
security/security.c:688:16: note: each undeclared identifier is reported only once for each function it appears in
include backing-dev.h directly to fix it up.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: James Morris <james.l.morris@oracle.com>
Especially vesafb likes to map everything as uc- (yikes), and if that
mapping hangs around still while we try to map the gtt as wc the
kernel will downgrade our request to uc-, resulting in abyssal
performance.
Unfortunately we can't do this as early as readon does (i.e. as the
first thing we do when initializing the hw) because our fb/mmio space
region moves around on a per-gen basis. So I've had to move it below
the gtt initialization, but that seems to work, too. The important
thing is that we do this before we set up the gtt wc mapping.
Now an altogether different question is why people compile their
kernels with vesafb enabled, but I guess making things just work isn't
bad per se ...
v2:
- s/radeondrmfb/inteldrmfb/
- fix up error handling
v3: Kill #ifdef X86, this is Intel after all. Noticed by Ben Widawsky.
v4: Jani Nikula complained about the pointless bool primary
initialization.
v5: Don't oops if we can't allocate, noticed by Chris Wilson.
v6: Resolve conflicts with agp rework and fixup whitespace.
This is commit e188719a28 in drm-next.
Backport to 3.5 -fixes queue requested by Dave Airlie - due to grub
using vesa on fedora their initrd seems to load vesafb before loading
the real kms driver. So tons more people actually experience a
dead-slow gpu. Hence also the Cc: stable.
Cc: stable@vger.kernel.org
Reported-and-tested-by: "Kilarski, Bernard R" <bernard.r.kilarski@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
When a monitor EDID doesn't give the preferred bit, driver assumes
that the mode with the higest resolution and rate is the preferred
mode. Meanwhile the recent changes for allowing more modes in the
GFT/CVT ranges give actually more modes, and some modes may be over
the native size. Thus such a mode would be picked up as the preferred
mode although it's no native resolution.
For avoiding such a problem, this patch limits the addition of
inferred modes by checking not to be greater than other modes.
Also, it checks the duplicated mode entry at the same time.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
In gem idle/busy ioctl the radeon object was derefenced after
drm_gem_object_unreference_unlocked which in case the object
have been destroyed lead to use of a possibly free pointer with
possibly wrong data.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The value returned by "mddev_check_plug" is only valid until the
next 'schedule' as that will unplug things. This could happen at any
call to mempool_alloc.
So just calling mddev_check_plug at the start doesn't really make
sense.
So call it just before, or just after, queuing things for the thread.
As the action that happens at unplug is to wake the thread, this makes
lots of sense.
If we cannot add a plug (which requires a small GFP_ATOMIC alloc) we
wake thread immediately.
RAID5 is a bit different. Requests are queued for the thread and the
thread is woken by release_stripe. So we don't need to wake the
thread on failure.
However the thread doesn't perform certain actions when there is any
active plug, so it is important to install a plug before waking the
thread. So for RAID5 we install the plug *before* queuing the request
and waking the thread.
Without this patch it is possible for raid1 or raid10 to queue a
request without then waking the thread, resulting in the array locking
up.
Also change raid10 to only flush_pending_write when there are not
active plugs, just like raid1.
This patch is suitable for 3.0 or later. I plan to submit it to
-stable, but I'll like to let it spend a few weeks in mainline
first to be sure it is completely safe.
Signed-off-by: NeilBrown <neilb@suse.de>
We currently only allow a device to be re-added if it appear to be
in-sync. This is overly restrictive as it may be desirable to re-add
a device that is in the middle of recovery.
So remove the test for "InSync" - the test on rdev->raid_disk is
sufficient to ensure that the re-add will succeed.
Reported-by: Alexander Lyakas <alex.bolshoy@gmail.com>
Tested-by: Alexander Lyakas <alex.bolshoy@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
When we added hot_replace we doubled the number of devices
that could be in a RAID1 array. So we doubled how far read_balance
would search. Unfortunately we didn't double the point at which
it looped back to the beginning - so it effectively loops over
all non-replacement disks twice.
This doesn't cause bad behaviour, but it pointless and means we
never read from replacement devices.
Signed-off-by: NeilBrown <neilb@suse.de>
There isn't locking setting STRIPE_DELAYED and STRIPE_PREREAD_ACTIVE bits, but
the two bits have relationship. A delayed stripe can be moved to hold list only
when preread active stripe count is below IO_THRESHOLD. If a stripe has both
the bits set, such stripe will be in delayed list and preread count not 0,
which will make such stripe never leave delayed list.
Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>
We may not be able to fix a bad block if:
- the array is degraded
- the over-write fails.
In these cases we currently eject the device, but we should
record a bad block if possible.
Signed-off-by: majianpeng <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Having the 'name' arg optional and defaulting to the current
personality name is no necessary and leads to errors, as when
changing the level of an array we can end up using the
name of the old level instead of the new one.
So make it non-optional and always explicitly pass the name
of the level that the array will be.
Reported-by: majianpeng <majianpeng@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
commit 58c54fcca3
md/raid10: handle further errors during fix_read_error better.
in 3.1 added "r10_sync_page_io" which takes an IO size in sectors.
But we were passing the IO size in bytes!!!
This resulting in bio_add_page failing, and empty request being sent
down, and a consequent BUG_ON in scsi_lib.
[fix missing space in error message at same time]
This fix is suitable for 3.1.y and later.
Cc: stable@vger.kernel.org
Reported-by: Christian Balzer <chibi@gol.com>
Signed-off-by: NeilBrown <neilb@suse.de>
The array of names in hugetlbpage.c no longer exists.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Commit 430b01e8f5 ("[POWERPC] Kill
flatdevtree.c") killed the two files including flatdevtree_env.h. It was
apparently just an oversight to not kill that header too. Kill it now.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Warning(arch/powerpc/kernel/pci_of_scan.c:210): Excess function parameter 'node' description in 'of_scan_pci_bridge'
Warning(arch/powerpc/kernel/vio.c:636): No description found for parameter 'desired'
Warning(arch/powerpc/kernel/vio.c:636): Excess function parameter 'new_desired' description in 'vio_cmo_set_dev_desired'
Warning(arch/powerpc/kernel/vio.c:1270): No description found for parameter 'viodrv'
Warning(arch/powerpc/kernel/vio.c:1270): Excess function parameter 'drv' description in '__vio_register_driver'
Warning(arch/powerpc/kernel/vio.c:1289): No description found for parameter 'viodrv'
Warning(arch/powerpc/kernel/vio.c:1289): Excess function parameter 'driver' description in 'vio_unregister_driver'
Signed-off-by: Wanpeng Li <liwp@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
memblock_end_of_DRAM() returns end_address + 1, not end address.
While some code assumes that it returns end address.
Signed-off-by: Bharat Bhushan <bharat.bhushan@freescale.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
I blame Mikey for this. He elevated my slightly dubious testcase:
to benchmark status. And naturally we need to be number 1 at creating
zeros. So lets improve __clear_user some more.
As Paul suggests we can use dcbz for large lengths. This patch gets
the destination cacheline aligned then uses dcbz on whole cachelines.
Before:
10485760000 bytes (10 GB) copied, 0.414744 s, 25.3 GB/s
After:
10485760000 bytes (10 GB) copied, 0.268597 s, 39.0 GB/s
39 GB/s, a new record.
Signed-off-by: Anton Blanchard <anton@samba.org>
Tested-by: Olof Johansson <olof@lixom.net>
Acked-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
At the moment all queues in a multiqueue adapter will serialise
against the IOMMU table lock. This is proving to be a big issue,
especially with 10Gbit ethernet.
This patch creates 4 pools and tries to spread the load across
them. If the table is under 1GB in size we revert back to the
original behaviour of 1 pool and 1 largealloc pool.
We create a hash to map CPUs to pools. Since we prefer interrupts to
be affinitised to primary CPUs, without some form of hashing we are
very likely to end up using the same pool. As an example, POWER7
has 4 way SMT and with 4 pools all primary threads will map to the
same pool.
The largealloc pool is reduced from 1/2 to 1/4 of the space to
partially offset the overhead of breaking the table up into pools.
Some performance numbers were obtained with a Chelsio T3 adapter on
two POWER7 boxes, running a 100 session TCP round robin test.
Performance improved 69% with this patch applied.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
In preparation for IOMMU pools, push the spinlock into
iommu_range_alloc and __iommu_free.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This patch moves tce_free outside of the lock in iommu_free.
Some performance numbers were obtained with a Chelsio T3 adapter on
two POWER7 boxes, running a 100 session TCP round robin test.
Performance improved 25% with this patch applied.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We currently hold the IOMMU spinlock around tce_build and tce_flush.
This causes our spinlock hold times to be much higher than required
and can impact multiqueue adapters.
This patch moves tce_build and tce_flush outside of the lock in
iommu_alloc, and tce_flush outside of the lock in iommu_free.
Some performance numbers were obtained with a Chelsio T3 adapter on
two POWER7 boxes, running a 100 session TCP round robin test.
Performance improved 32% with this patch applied.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
tce_buildmulti_pSeriesLP uses a per cpu page to communicate with the
hypervisor. We currently rely on the IOMMU table spinlock but
subsequent patches will be removing that so disable interrupts
around all accesses of tce_page.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Implement a POWER7 optimised memcpy using VMX and enhanced prefetch
instructions.
This is a copy of the POWER7 optimised copy_to_user/copy_from_user
loop. Detailed implementation and performance details can be found in
commit a66086b819 (powerpc: POWER7 optimised
copy_to_user/copy_from_user using VMX).
I noticed memcpy issues when profiling a RAID6 workload:
.memcpy
.async_memcpy
.async_copy_data
.__raid_run_ops
.handle_stripe
.raid5d
.md_thread
I created a simplified testcase by building a RAID6 array with 4 1GB
ramdisks (booting with brd.rd_size=1048576):
# mdadm -CR -e 1.2 /dev/md0 --level=6 -n4 /dev/ram[0-3]
I then timed how long it took to write to the entire array:
# dd if=/dev/zero of=/dev/md0 bs=1M
Before: 892 MB/s
After: 999 MB/s
A 12% improvement.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Version 2.06 of the POWER ISA introduced enhanced touch instructions,
allowing us to specify a number of attributes including the length of
a stream.
This patch adds a software stream for both loads and stores in the
POWER7 copy_tofrom_user loop. Since the setup is quite complicated
and we have to use an eieio to ensure correct ordering of the "GO"
command we only do this for copies above 4kB.
To quantify any performance improvements we need a working set
bigger than the caches so we operate on a 1GB file:
# dd if=/dev/zero of=/tmp/foo bs=1M count=1024
And we compare how fast we can read the file:
# dd if=/tmp/foo of=/dev/null bs=1M
before: 7.7 GB/s
after: 9.6 GB/s
A 25% improvement.
The worst case for this patch will be a completely L1 cache contained
copy of just over 4kB. We can test this with the copy_to_user
testcase we used to tune copy_tofrom_user originally:
http://ozlabs.org/~anton/junkcode/copy_to_user.c
# time ./copy_to_user2 -l 4224 -i 10000000
before: 6.807 s
after: 6.946 s
A 2% slowdown, which seems reasonable considering our data is unlikely
to be completely L1 contained.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
While creating the PCI root bus through function pci_create_root_bus()
of PCI core, it should have assigned the secondary bus number for the
newly created PCI root bus. Thus we needn't do the explicit assignment
for the secondary bus number again in pcibios_scan_phb().
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The form affinity for NUMA is set to 1 if the firmware supports
OPAL. Otherwise, we have to retrieve that from OF node "/chosen".
For the latter case, OF node "/chosen" reference count was never
decreased.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Implement a POWER7 optimised copy_page using VMX and enhanced
prefetch instructions. We use enhanced prefetch hints to prefetch
both the load and store side. We copy a cacheline at a time and
fall back to regular loads and stores if we are unable to use VMX
(eg we are in an interrupt).
The following microbenchmark was used to assess the impact of
the patch:
http://ozlabs.org/~anton/junkcode/page_fault_file.c
We test MAP_PRIVATE page faults across a 1GB file, 100 times:
# time ./page_fault_file -p -l 1G -i 100
Before: 22.25s
After: 18.89s
17% faster
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Subsequent patches will add more VMX library functions and it makes
sense to keep all the c-code helper functions in the one file.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
mtmsrd is an expensive instruction, we save a few cycles by
doing it once instead of twice.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Version 2.06 of the POWER ISA introduced enhanced touch instructions,
allowing us to specify a number of attributes including the length of
a stream.
This patch adds a software stream for both loads and stores in the
POWER7 copy_tofrom_user loop. Since the setup is quite complicated
and we have to use an eieio to ensure correct ordering of the "GO"
command we only do this for copies above 4kB.
To quantify any performance improvements we need a working set
bigger than the caches so we operate on a 1GB file:
# dd if=/dev/zero of=/tmp/foo bs=1M count=1024
And we compare how fast we can read the file:
# dd if=/tmp/foo of=/dev/null bs=1M
before: 7.7 GB/s
after: 9.6 GB/s
A 25% improvement.
The worst case for this patch will be a completely L1 cache contained
copy of just over 4kB. We can test this with the copy_to_user
testcase we used to tune copy_tofrom_user originally:
http://ozlabs.org/~anton/junkcode/copy_to_user.c
# time ./copy_to_user2 -l 4224 -i 10000000
before: 6.807 s
after: 6.946 s
A 2% slowdown, which seems reasonable considering our data is unlikely
to be completely L1 contained.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
1) call_function.lock used in smp_call_function_many() is just to protect
call_function.queue and &data->refs, cpu_online_mask is outside of the
lock. And it's not necessary to protect cpu_online_mask,
because data->cpumask is pre-calculate and even if a cpu is brougt up
when calling arch_send_call_function_ipi_mask(), it's harmless because
validation test in generic_smp_call_function_interrupt() will take care
of it.
2) For cpu down issue, stop_machine() will guarantee that no concurrent
smp_call_fuction() is processing.
Signed-off-by: Yong Zhang <yong.zhang0@gmail.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
I noticed __clear_user high up in a profile of one of my RAID stress
tests. The testcase was doing a dd from /dev/zero which ends up
calling __clear_user.
__clear_user is basically a loop with a single 4 byte store which
is horribly slow. We can do much better by aligning the desination
and doing 32 bytes of 8 byte stores in a loop.
The following testcase was used to verify the patch:
http://ozlabs.org/~anton/junkcode/stress_clear_user.c
To show the improvement in performance I ran a dd from /dev/zero
to /dev/null on a POWER7 box:
Before:
# dd if=/dev/zero of=/dev/null bs=1M count=10000
10485760000 bytes (10 GB) copied, 3.72379 s, 2.8 GB/s
After:
# time dd if=/dev/zero of=/dev/null bs=1M count=10000
10485760000 bytes (10 GB) copied, 0.728318 s, 14.4 GB/s
Over 5x faster.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
irq_entry, irq_exit, timer_interrupt_entry and timer_interrupt_exit
all do the same thing so use DECLARE_EVENT_CLASS to avoid duplicating
everything 4 times.
This saves quite a lot of space in both instruction text and data:
text data bss dec hex filename
9265 19622 16 28903 70e7 arch/powerpc/kernel/irq.o
6817 19019 16 25852 64fc arch/powerpc/kernel/irq.o
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When looking through some instruction traces I noticed our tracepoint
checks were inline. It turns out we don't have CONFIG_JUMP_LABEL
enabled.
By enabling CONFIG_JUMP_LABEL we replace a load/compare/branch with
a nop at every tracepoint call. For example in do_IRQ:
CONFIG_JUMP_LABEL disabled:
stdx 3,11,9
lwz 0,8(29)
cmpwi 7,0,0
bne- 7,.L124
bl .irq_enter
CONFIG_JUMP_LABEL enabled:
stdx 3,11,9
nop
bl .irq_enter
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The following patch is to remove the pseries_notify_add_cpu() call
and replace it by a hot plug notifier.
This would prevent cpuidle resources being released and allocated each
time cpu comes online on pseries.
The earlier design was causing a lockdep problem
in start_secondary as reported on this thread
-https://lkml.org/lkml/2012/5/17/2
This applies on 3.4-rc7
Signed-off-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
An upcoming release of firmware will add DDW extensions, in particular
an API to "reset" the DMA window to the original configuration (32-bit,
2GB in size). With that API available, we can safely remove the default
window, increasing the resources available to firmware for creation of
larger windows for the slot in question -- if we encounter an error, we
can use the new API to reset the state of the slot.
Further, this same release of firmware will make it a hard requirement
for OSes to release the existing window before any other windows will be
shown as available, to avoid conflicts in addressing between the two
windows.
In anticipation of these changes, always remove the default window
before we do any DDW manipulations.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch_instruction() interface is made to modify kernel text. It is
safer to use that then the probe_kernel_write() when modifying kernel
code.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
For ftrace to use the patch_instruction code, it needs to check for
faults on write. Ftrace updates code all over the kernel, and we need to
know if code is updated or not due to protections that are placed on
some portions of the kernel. If ftrace does not detect a fault, it will
error later on, and it will be much more difficult to find the problem.
By changing patch_instruction() to detect faults, then ftrace will be
able to make use of it too.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Stephen Boyd