blkio: Fix another BUG_ON() crash due to cfqq movement across groups
o Once in a while, I was hitting a BUG_ON() in blkio code. empty_time was
assuming that upon slice expiry, group can't be marked empty already (except
forced dispatch).
But this assumption is broken if cfqq can move (group_isolation=0) across
groups after receiving a request.
I think most likely in this case we got a request in a cfqq and accounted
the rq in one group, later while adding the cfqq to tree, we moved the queue
to a different group which was already marked empty and after dispatch from
slice we found group already marked empty and raised alarm.
This patch does not error out if group is already marked empty. This can
introduce some empty_time stat error only in case of group_isolation=0. This
is better than crashing. In case of group_isolation=1 we should still get
same stats as before this patch.
blkio + cfq was crashing even when two sequential readers were put in two
separate cgroups (group_isolation=0).
The reason being that cfqq can migrate across groups based on its being
sync-noidle or not, it can happen that at request insertion time, cfqq
belonged to one cfqg and at request dispatch time, it belonged to root
group. In this case request stats per cgroup can go wrong and it also runs
into BUG_ON().
This patch implements rq stashing away a cfq group pointer and not relying
on cfqq->cfqg pointer alone for rq stat accounting.
This happens because the two symbols aren't exported and hence not available
when blk-cgroup code is built as a module. I've tried to stay consistent with
the use of EXPORT_SYMBOL or EXPORT_SYMBOL_GPL with the other symbols in the
respective files.
Richard Kennedy [Wed, 14 Apr 2010 18:53:37 +0000 (20:53 +0200)]
block: ensure jiffies wrap is handled correctly in blk_rq_timed_out_timer
blk_rq_timed_out_timer() relied on blk_add_timer() never returning a
timer value of zero, but commit 7838c15b8dd18e78a523513749e5b54bda07b0cb
removed the code that bumped this value when it was zero.
Therefore when jiffies is near wrap we could get unlucky & not set the
timeout value correctly.
This patch uses a flag to indicate that the timeout value was set and so
handles jiffies wrap correctly, and it keeps all the logic in one
function so should be easier to maintain in the future.
Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Divyesh Shah [Tue, 13 Apr 2010 17:59:17 +0000 (19:59 +0200)]
block: Update to io-controller stats
Changelog from v1:
o Call blkiocg_update_idle_time_stats() at cfq_rq_enqueued() instead of at
dispatch time.
Changelog from original patchset: (in response to Vivek Goyal's comments)
o group blkiocg_update_blkio_group_dequeue_stats() with other DEBUG functions
o rename blkiocg_update_set_active_queue_stats() to
blkiocg_update_avg_queue_size_stats()
o s/request/io/ in blkiocg_update_request_add_stats() and
blkiocg_update_request_remove_stats()
o Call cfq_del_timer() at request dispatch() instead of
blkiocg_update_idle_time_stats()
io-controller: Add a new interface "weight_device" for IO-Controller
Currently, IO Controller makes use of blkio.weight to assign weight for
all devices. Here a new user interface "blkio.weight_device" is introduced to
assign different weights for different devices. blkio.weight becomes the
default value for devices which are not configured by "blkio.weight_device"
You can use the following format to assigned specific weight for a given
device:
#echo "major:minor weight" > blkio.weight_device
major:minor represents device number.
And you can remove weight for a given device as following:
#echo "major:minor 0" > blkio.weight_device
V1->V2 changes:
- use user interface "weight_device" instead of "policy" suggested by Vivek
- rename some struct suggested by Vivek
- rebase to 2.6-block "for-linus" branch
- remove an useless list_empty check pointed out by Li Zefan
- some trivial typo fix
V2->V3 changes:
- Move policy_*_node() functions up to get rid of forward declarations
- rename related functions by adding prefix "blkio_"
* anonvma:
anonvma: when setting up page->mapping, we need to pick the _oldest_ anonvma
anon_vma: clone the anon_vma chain in the right order
vma_adjust: fix the copying of anon_vma chains
Simplify and comment on anon_vma re-use for anon_vma_prepare()
* master.kernel.org:/home/rmk/linux-2.6-arm: (21 commits)
ARM: Fix ioremap_cached()/ioremap_wc() for SMP platforms
ARM: 6043/1: AT91 slow-clock resume: Don't wait for a disabled PLL to lock
ARM: 6031/1: fix Thumb-2 decompressor
ARM: 6029/1: ep93xx: gpio.c: local functions should be static
ARM: 6028/1: ARM: add MAINTAINERS for U300
ARM: 6024/1: bcmring: fix missing down on semaphore in dma.c
MXC: mach_armadillo5x0: Add USB Host support.
ARM mach-mx3: duplicated include
ARM mach-mx3: duplicated include
imx31: add watchdog device on litekit board.
imx3: Add watchdog platform device support
MXC: mach-mx31_3ds: add support for freescale mc13783 power management device.
MXC: mach-mx31_3ds: Add SPI1 device support.
MXC: mach-mx31_3ds: Add support for on board NAND Flash.
MXC: mach-mx31_3ds: Update variable names over recent mach name modification.
imx31: fix parent clock for rtc
i.MX51: remove NFC AXI static mapping
i.MX51: determine silicon revision dynamically
i.MX51: map TZIC dynamically
i.MX51: Use correct clock for gpt
...
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable
* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: make sure the chunk allocator doesn't create zero length chunks
Btrfs: fix data enospc check overflow
Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6:
quota: Fix possible dq_flags corruption
quota: Hide warnings about writes to the filesystem before quota was turned on
ext3: symlink must be handled via filesystem specific operation
ext2: symlink must be handled via filesystem specific operation
Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus: (36 commits)
MIPS: Calculate proper ebase value for 64-bit kernels
MIPS: Alchemy: DB1200: Remove custom wait implementation
MIPS: Big Sur: Make defconfig more useful.
MIPS: Fix __vmalloc() etc. on MIPS for non-GPL modules
MIPS: Sibyte: Fix M3 TLB exception handler workaround.
MIPS: BCM63xx: Fix build failure in board_bcm963xx.c
MIPS: uasm: Add OR instruction.
MIPS: Sibyte: Apply M3 workaround only on affected chip types and versions.
MIPS: BCM63xx: Initialize gpio_out_low & out_high to current value at boot.
MIPS: BCM63xx: Register SSB SPROM fallback in board's first stage callback
MIPS: BCM63xx: Fix typo in cpu-feature-overrides file.
MIPS: BCM63xx: Add support for second uart.
MIPS: BCM63xx: Fix double gpio registration.
MIPS: BCM63xx: Add DWVS0 board
MIPS: BCM63xx: Add the RTA1025W-16 BCM6348-based board to suppported boards.
MIPS: BCM63xx: Fix BCM6338 and BCM6345 gpio count
MIPS: libgcc.h: Checkpatch cleanup
MIPS: Loongson-2F: Flush the branch target history in BTB and RAS
MIPS: Move signal trampolines off of the stack.
MIPS: Preliminary VDSO
...
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke/nilfs2
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke/nilfs2:
nilfs2: fix typo "numer" -> "number" in alloc.c
nilfs2: Remove an uninitialization warning in nilfs_btree_propagate_v()
nilfs2: fix a wrong type conversion in nilfs_ioctl()
anonvma: when setting up page->mapping, we need to pick the _oldest_ anonvma
Otherwise we might be mapping in a page in a new mapping, but that page
(through the swapcache) would later be mapped into an old mapping too.
The page->mapping must be the case that works for everybody, not just
the mapping that happened to page it in first.
Here's the scenario:
- page gets allocated/mapped by process A. Let's call the anon_vma we
associate the page with 'A' to keep it easy to track.
- Process A forks, creating process B. The anon_vma in B is 'B', and has
a chain that looks like 'B' -> 'A'. Everything is fine.
- Swapping happens. The page (with mapping pointing to 'A') gets swapped
out (perhaps not to disk - it's enough to assume that it's just not
mapped any more, and lives entirely in the swap-cache)
In particular, what happens is that this will now be the "first"
mapping of that page, so page_add_anon_rmap() used to do
if (first)
__page_set_anon_rmap(page, vma, address);
and notice what anon_vma it will use? It will use the anon_vma for
process B!
What happens then? Trivial: process 'A' also pages it in (nothing
happens, it's not the first mapping), and then process 'B' execve's
or exits or unmaps, making anon_vma B go away.
End result: process A has a page that points to anon_vma B, but
anon_vma B does not exist any more. This can go on forever. Forget
about RCU grace periods, forget about locking, forget anything like
that. The bug is simply that page->mapping points to an anon_vma
that was correct at one point, but was _not_ the one that was shared
by all users of that possible mapping.
Changing it to always use the deepest anon_vma in the anonvma chain gets
us to the safest model.
This can be improved in certain cases: if we know the page is private to
just this particular mapping (for example, it's a new page, or it is the
only swapcache entry), we could pick the top (most specific) anon_vma.
But that's a future optimization. Make it _work_ reliably first.
Reviewed-by: Rik van Riel <riel@redhat.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Tested-by: Borislav Petkov <bp@alien8.de> [ "What do you know, I think you fixed it!" ] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
anon_vma: clone the anon_vma chain in the right order
We want to walk the chain in reverse order when cloning it, so that the
order of the result chain will be the same as the order in the source
chain. When we add entries to the chain, they go at the head of the
chain, so we want to add the source head last.
Reviewed-by: Rik van Riel <riel@redhat.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Tested-by: Borislav Petkov <bp@alien8.de> [ "No, it still oopses" ] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When we move the boundaries between two vma's due to things like
mprotect, we need to make sure that the anon_vma of the pages that got
moved from one vma to another gets properly copied around. And that was
not always the case, in this rather hard-to-follow code sequence.
Clarify the code, and fix it so that it copies the anon_vma from the
right source.
Reviewed-by: Rik van Riel <riel@redhat.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Tested-by: Borislav Petkov <bp@alien8.de> [ "Yeah, not so much this one either" ] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Simplify and comment on anon_vma re-use for anon_vma_prepare()
This changes the anon_vma reuse case to require that we only reuse
simple anon_vma's - ie the case when the vma only has a single anon_vma
associated with it.
This means that a reuse of an anon_vma from an adjacent vma will always
guarantee that both vma's are associated not only with the same
anon_vma, they will also have the same anon_vma chain (of just a single
entry in this case).
And since anon_vma re-use was the only case where the same anon_vma
might be associated with different chains of anon_vma's, we now have the
case that every vma that shares the same anon_vma will always also have
the same chain. That makes it much easier to think about merging vma's
that share the same anon_vma's: you can always just drop the other
anon_vma chain in anon_vma_merge() since you know that they are always
identical.
This also splits up the function to validate the anon_vma re-use, and
adds a lot of commentary about the possible races.
dq_flags are modified non-atomically in do_set_dqblk via __set_bit calls and
atomically for example in mark_dquot_dirty or clear_dquot_dirty. Hence a
change done by an atomic operation can be overwritten by a change done by a
non-atomic one. Fix the problem by using atomic bitops even in do_set_dqblk.
Signed-off-by: Andrew Perepechko <andrew.perepechko@sun.com> Signed-off-by: Jan Kara <jack@suse.cz>
Jan Kara [Tue, 6 Apr 2010 16:52:47 +0000 (18:52 +0200)]
quota: Hide warnings about writes to the filesystem before quota was turned on
For a root filesystem write to the filesystem before quota is turned on happens
regularly and there's no way around it because of writes to syslog, /etc/mtab,
and similar. So the warning is rather pointless for ordinary users. It's
still useful during development so we just hide the warning behind
__DQUOT_PARANOIA config option.
David Daney [Tue, 6 Apr 2010 20:29:50 +0000 (13:29 -0700)]
MIPS: Calculate proper ebase value for 64-bit kernels
The ebase is relative to CKSEG0 not CAC_BASE. On a 32-bit kernel they
are the same thing, for a 64-bit kernel they are not.
It happens to kind of work on a 64-bit kernel as they both reference
the same physical memory. However since the CPU uses the CKSEG0 base,
determining if a J instruction will reach always gives the wrong result
unless we use the same number the CPU uses.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
To: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/1093/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
While playing with the out-of-tree MAE driver module, the system would
panic after a while in the db1200 custom wait code after wakeup due to
a clobbered k0 register being used as target address of a store op.
Remove the custom wait implementation and revert back to the Alchemy-
recommended implementation already set as default.
Signed-off-by: Manuel Lauss <manuel.lauss@gmail.com>
To: Linux-MIPS <linux-mips@linux-mips.org>
Patchwork: http://patchwork.linux-mips.org/patch/1092/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
All those functions are EXPORT_SYMBOL() so are meant to be allowed to be
used by non-GPL kernel modules. These calls all take page protection as
an argument which is normally a constant like PAGE_KERNEL.
This commit causes all protection constants like PAGE_KERNEL to not be
constants and instead to contain the GPL-only symbol _page_cachable_default.
This means that all calls to __vmalloc(), vmap(), etc, cause non-GPL
modules to fail to link with the complaint that they are trying to use the
GPL-only symbol _page_cachable_default...
Change EXPORT_SYMBOL_GPL(_page_cachable_default) to EXPORT_SYMBOL() for
non-GPL modules that call __vmalloc(), vmap(), vm_map_ram() etc.
Signed-off-by: Anton Altaparmakov <aia21@cantab.net> Cc: Chris Dearman <chris@mips.com> Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org
Patchwork: http://patchwork.linux-mips.org/patch/1084/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Florian Fainelli [Tue, 23 Mar 2010 09:30:08 +0000 (10:30 +0100)]
MIPS: BCM63xx: Fix build failure in board_bcm963xx.c
Since 2083e8327aeeaf818b0e4522a9d2539835c60423, the SPROM is now registered
in the board_prom_init callback, but it references variables and functions
which are declared below. Move the variables and functions above
board_prom_init.
Ralf Baechle [Mon, 22 Mar 2010 23:02:43 +0000 (00:02 +0100)]
MIPS: Sibyte: Apply M3 workaround only on affected chip types and versions.
Previously it was unconditionally used on all Sibyte family SOCs. The
M3 bug has to be handled in the TLB exception handler which is extremly
performance sensitive, so this modification is expected to deliver around
2-3% performance improvment. This is important as required changes to the
M3 workaround will make it more costly.
Maxime Bizon [Sat, 30 Jan 2010 17:34:55 +0000 (18:34 +0100)]
MIPS: BCM63xx: Add support for second uart.
The BCm63xx SOC has two uarts. Some boards use the second one for
bluetooth. This patch changes platform device registration code to
handle this. Changes to the UART driver were already merged in 6a2c7eabfd09ca7986bf96b8958a87ca041a19d8.
Signed-off-by: Maxime Bizon <mbizon@freebox.fr>
To: linux-mips@linux-mips.org Cc: Maxime Bizon <mbizon@freebox.fr>
Patchwork: http://patchwork.linux-mips.org/patch/900/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Wu Zhangjin [Sat, 13 Mar 2010 04:34:15 +0000 (12:34 +0800)]
MIPS: Loongson-2F: Flush the branch target history in BTB and RAS
As per chapter 15 "Errata: Issue of Out-of-order in loongson"[1] to work
around the Loongson 2F erratum we need to do:
"When switching from user mode to kernel mode, you should flush the
branch target history such as BTB and RAS."
[1] Chinese version: http://www.loongson.cn/uploadfile/file/200808211
[2] English version of chapter 15:
http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source
David Daney [Fri, 19 Feb 2010 00:13:05 +0000 (16:13 -0800)]
MIPS: Move signal trampolines off of the stack.
This is a follow on to the vdso patch.
Since all processes now have signal trampolines permanently mapped, we
can use those instead of putting the trampoline on the stack and
invalidating the corresponding icache across all CPUs. We also get rid
of a bunch of ICACHE_REFILLS_WORKAROUND_WAR code.
[Ralf: GDB 7.1 which has the necessary modifications to allow backtracing
over signal frames will supposedly be released tomorrow. The old signal
frame format obsoleted by this patch exists in two variations, for sane
processors and for those requiring ICACHE_REFILLS_WORKAROUND_WAR. So
there was never a GDB which did support backtracing over signal frames
on all MIPS systems. This convinved me this series should be applied and
pushed upstream as soon as possible.]
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
To: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/974/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
David Daney [Fri, 19 Feb 2010 00:13:04 +0000 (16:13 -0800)]
MIPS: Preliminary VDSO
This is a preliminary patch to add a vdso to all user processes. Still
missing are ELF headers and .eh_frame information. But it is enough to
allow us to move signal trampolines off of the stack. Note that emulation
of branch delay slots in the FPU emulator still requires the stack.
We allocate a single page (the vdso) and write all possible signal
trampolines into it. The stack is moved down by one page and the vdso is
mapped into this space.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
To: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/975/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Andreas Ferber [Tue, 16 Mar 2010 11:35:51 +0000 (12:35 +0100)]
MIPS: Fix SSB PCIcore IO resource management
The SSB PCIcore code reused the IO resource fixup code from the original
2.4.x Broadcom patch for BCM47xx based devices, which was a quick hack
for doing PCI IO resource configuration back then (the boot loader
doesn't configure PCI devices on this platform).
However, this code is no longer necessary since the kernel now can do
PCI resource management fine all by itself, so remove the old code.
When removing the code, it becomes obvious that the mem_offset setting
in the PCIcore driver was wrong, however this was masked by the fixup
code before, except in a few cases involving yenta_socket. For BCM47xx,
the correct offset is 0, and since this is the only device using PCIcore
in host mode, the offset can simply be removed unconditionally.
Signed-off-by: Andreas Ferber <af@chaos-agency.de> Signed-off-by: Michael Buesch <mb@bu3sch.de> Cc: Markus Wigge <markus@cultcom.de> Cc: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/1070/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Florian Fainelli [Wed, 10 Mar 2010 08:51:09 +0000 (09:51 +0100)]
MIPS: make CAC_ADDR and UNCAC_ADDR account for PHYS_OFFSET
On AR7, we already redefine PHYS_OFFSET to match the system specifities, it
is however not sufficient when unsing dma_{map,unmap}_single, specifically
in the ethernet driver, we must also adjust CAC_ADDR and UNCAC_ADDR for DMA
to work correctly. This patch fixes the following issue, seen in cpmac_open:
Robert P. J. Day [Sat, 27 Feb 2010 17:02:51 +0000 (12:02 -0500)]
MIPS: Initialize an atomic_t properly with ATOMIC_INIT(0).
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
To: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/1008/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Ralf Baechle [Thu, 11 Mar 2010 07:48:14 +0000 (08:48 +0100)]
MIPS: Fix elfcore.c build warning
kernel/elfcore.c includes <linux/elf.h> which includes the <asm/elf.h>. In
<asm/elf.h>, struct pt_regs is declared inside the parameter list of the
elf_dump_regs function which causes a kernel build warning.
Fixed by adding a forward declaration of struct pt_regs.
Yang Shi [Wed, 3 Mar 2010 08:43:20 +0000 (16:43 +0800)]
MIPS: Octeon: Remove redundant declaration of octeon_reserve32_memory
octeon_reserve32_memory is defined In Octeon's setup.c, so remove the
redundant extern declaration of this variable.
Signed-off-by: Yang Shi <yang.shi@windriver.com>
To: f.fainelli@gmail.com Cc: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/1022/ Acked-by: David Daney <ddaney@caviumnetworks.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Wu Zhangjin [Tue, 9 Mar 2010 16:27:28 +0000 (00:27 +0800)]
MIPS: Trace: Don't trace irqsoff for the idle process
Like x86 did in arch/x86/kernel/{process_32.c,process_64.c}, also don't
trace irqsoff for idle.
If there's no useful work to be done, we don't care about the irqsoff
duration. If we trace the idle process, the max duration of irqsoff will
be the idle time and make the irqsoff tracer useless.
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
IB/mlx4: Check correct variable for allocation failure
RDMA/nes: Correct cap.max_inline_data assignment in nes_query_qp()
RDMA/cm: Set num_paths when manually assigning path records
IB/cm: Fix device_create() return value check
Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block
* 'for-linus' of git://git.kernel.dk/linux-2.6-block: (34 commits)
cfq-iosched: Fix the incorrect timeslice accounting with forced_dispatch
loop: Update mtime when writing using aops
block: expose the statistics in blkio.time and blkio.sectors for the root cgroup
backing-dev: Handle class_create() failure
Block: Fix block/elevator.c elevator_get() off-by-one error
drbd: lc_element_by_index() never returns NULL
cciss: unlock on error path
cfq-iosched: Do not merge queues of BE and IDLE classes
cfq-iosched: Add additional blktrace log messages in CFQ for easier debugging
i2o: Remove the dangerous kobj_to_i2o_device macro
block: remove 16 bytes of padding from struct request on 64bits
cfq-iosched: fix a kbuild regression
block: make CONFIG_BLK_CGROUP visible
Remove GENHD_FL_DRIVERFS
block: Export max number of segments and max segment size in sysfs
block: Finalize conversion of block limits functions
block: Fix overrun in lcm() and move it to lib
vfs: improve writeback_inodes_wb()
paride: fix off-by-one test
drbd: fix al-to-on-disk-bitmap for 4k logical_block_size
...
Merge branch 'drm-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6
* 'drm-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6: (29 commits)
drm/nouveau: bail out of auxch transaction if we repeatedly recieve defers
drm/nv50: implement gpio set/get routines
drm/nv50: parse/use some more de-magiced parts of gpio table entries
drm/nouveau: store raw gpio table entry in bios gpio structs
drm/nv40: Init some tiling-related PGRAPH state.
drm/nv50: Add NVA3 support in ctxprog/ctxvals generator.
drm/nv50: another dodgy DP hack
drm/nv50: punt hotplug irq handling out to workqueue
drm/nv50: preserve an unknown SOR_MODECTRL value for DP encoders
drm/nv50: Allow using the NVA3 new compute class.
drm/nv50: cleanup properly if PDISPLAY init fails
drm/nouveau: fixup the init failure paths some more
drm/nv50: fix instmem init on IGPs if stolen mem crosses 4GiB mark
drm/nv40: add LVDS table quirk for Dell Latitude D620
drm/nv40: rework lvds table parsing
drm/nouveau: detect vram amount once, and save the value
drm/nouveau: remove some unused members from drm_nouveau_private
drm/nouveau: Make use of TTM busy_placements.
drm/nv50: add more 0x100c80 flushy magic
drm/nv50: fix fbcon when framebuffer above 4GiB mark
...
David Howells [Tue, 6 Apr 2010 21:36:20 +0000 (22:36 +0100)]
radix_tree_tag_get() is not as safe as the docs make out [ver #2]
radix_tree_tag_get() is not safe to use concurrently with radix_tree_tag_set()
or radix_tree_tag_clear(). The problem is that the double tag_get() in
radix_tree_tag_get():
if (!tag_get(node, tag, offset))
saw_unset_tag = 1;
if (height == 1) {
int ret = tag_get(node, tag, offset);
may see the value change due to the action of set/clear. RCU is no protection
against this as no pointers are being changed, no nodes are being replaced
according to a COW protocol - set/clear alter the node directly.
The documentation in linux/radix-tree.h, however, says that
radix_tree_tag_get() is an exception to the rule that "any function modifying
the tree or tags (...) must exclude other modifications, and exclude any
functions reading the tree".
The problem is that the next statement in radix_tree_tag_get() checks that the
tag doesn't vary over time:
To this end, remove the BUG_ON() from radix_tree_tag_get() and note in various
comments that the value of the tag may change whilst the RCU read lock is held,
and thus that the return value of radix_tree_tag_get() may not be relied upon
unless radix_tree_tag_set/clear() and radix_tree_delete() are excluded from
running concurrently with it.
Reported-by: Romain DEGEZ <romain.degez@smartjog.com> Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pekka Enberg [Wed, 7 Apr 2010 16:23:41 +0000 (19:23 +0300)]
slub: Fix kmem_ptr_validate() for non-kernel pointers
As suggested by Linus, fix up kmem_ptr_validate() to handle non-kernel pointers
more graciously. The patch changes kmem_ptr_validate() to use the newly
introduced kern_ptr_validate() helper to check that a pointer is a valid kernel
pointer before we attempt to convert it into a 'struct page'.
Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Matt Mackall <mpm@selenic.com> Cc: Nick Piggin <npiggin@suse.de> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Acked-by: Christoph Lameter <cl@linux-foundation.org> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pekka Enberg [Wed, 7 Apr 2010 16:23:40 +0000 (19:23 +0300)]
slab: Generify kernel pointer validation
As suggested by Linus, introduce a kern_ptr_validate() helper that does some
sanity checks to make sure a pointer is a valid kernel pointer. This is a
preparational step for fixing SLUB kmem_ptr_validate().
Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: David Rientjes <rientjes@google.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Matt Mackall <mpm@selenic.com> Cc: Nick Piggin <npiggin@suse.de> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It changes user-visible sysfs interfaces, and breaks some existing user
space applications which apparently rely on the fact that the output
does not contain the "0x" prefix.
Russell King [Fri, 9 Apr 2010 14:00:11 +0000 (15:00 +0100)]
ARM: Fix ioremap_cached()/ioremap_wc() for SMP platforms
Write combining/cached device mappings are not setting the shared bit,
which could potentially cause problems on SMP systems since the cache
lines won't participate in the cache coherency protocol.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
[S390] nss: add missing .previous statement to asm function
The savesys_ipl_nss asm function is put into the .init.text section
however it is missing a ".previous" section which would restore the
previous section.
Luckily all functions in early.c are init functions so it doesn't
matter currently.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The default size of the vmalloc area is currently 1 GB. The memory resource
controller uses about 10 MB of vmalloc space per gigabyte of memory. That
turns a system with more than ~100 GB memory unbootable with the default
vmalloc size. It costs us nothing to increase the default size to some
more adequate value, e.g. 128 GB.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
commit 6a985c6194017de2c062916ad1cd00dee0302c40
([S390] s390: use change recording override for kernel mapping)
deactivated the change bit recording for the kernel mapping to
improve the performance. This works most of the time, but there
are cases (e.g. kernel runs in home space, futex atomic compare xcmg)
where we modify user memory with the kernel mapping instead of the
user mapping.
Instead of fixing these cases, this patch just deactivates change bit
override to avoid future problems with other kernel code that might
use the kernel mapping for user memory.
CC: stable@kernel.org Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
If a machine check interrupts the io interrupt handler on one of the
instructions between io_return and io_leave the critical section
cleanup code will move the return psw to io_work_loop. By doing that
the switch from the asynchronous interrupt stack to the process stack
is skipped. If e.g. TIF_NEED_RESCHED is set things break because
the scheduler is called with the asynchronous interrupts stack.
Moving the psw back to io_return instead fixes the problem.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Dan Carpenter [Fri, 9 Apr 2010 11:42:59 +0000 (13:42 +0200)]
[S390] sclp_async: potential buffer overflow
"len" hasn't been properly range checked so we shouldn't use it as an
array offset. This can only be written to by root but it would still be
annoying to accidentally write more than 3 characters and corrupt your
memory.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Anders Larsen [Thu, 8 Apr 2010 10:48:16 +0000 (11:48 +0100)]
ARM: 6043/1: AT91 slow-clock resume: Don't wait for a disabled PLL to lock
at91 slow-clock resume: Don't wait for a disabled PLL to lock.
We run into this problem with the PLLB on the at91: ohci-at91 disables
the PLLB when going to suspend. The slowclock code however tries to do
the same: It saves the PLLB register value and when restoring the value
during resume, it waits for the PLLB to lock again. However the PLL will
never lock and the loop would run into its timeout because the slowclock
code just stored and restored an empty register.
This fixes the problem by only restoring PLLA/PLLB when they were enabled
at suspend time.
Cc: Andrew Victor <avictor.za@gmail.com> Signed-off-by: Anders Larsen <al@alarsen.net> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Divyesh Shah [Fri, 9 Apr 2010 07:29:57 +0000 (09:29 +0200)]
cfq-iosched: Fix the incorrect timeslice accounting with forced_dispatch
When CFQ dispatches requests forcefully due to a barrier or changing iosched,
it runs through all cfqq's dispatching requests and then expires each queue.
However, it does not activate a cfqq before flushing its IOs resulting in
using stale values for computing slice_used.
This patch fixes it by calling activate queue before flushing reuqests from
each queue.
This is useful mostly for barrier requests because when the iosched is changing
it really doesnt matter if we have incorrect accounting since we're going to
break down all structures anyway.
We also now expire the current timeslice before moving on with the dispatch
to accurately account slice used for that cfqq.
Divyesh Shah [Fri, 9 Apr 2010 04:15:35 +0000 (21:15 -0700)]
blkio: Add more debug-only per-cgroup stats
1) group_wait_time - This is the amount of time the cgroup had to wait to get a
timeslice for one of its queues from when it became busy, i.e., went from 0
to 1 request queued. This is different from the io_wait_time which is the
cumulative total of the amount of time spent by each IO in that cgroup waiting
in the scheduler queue. This stat is a great way to find out any jobs in the
fleet that are being starved or waiting for longer than what is expected (due
to an IO controller bug or any other issue).
2) empty_time - This is the amount of time a cgroup spends w/o any pending
requests. This stat is useful when a job does not seem to be able to use its
assigned disk share by helping check if that is happening due to an IO
controller bug or because the job is not submitting enough IOs.
3) idle_time - This is the amount of time spent by the IO scheduler idling
for a given cgroup in anticipation of a better request than the exising ones
from other queues/cgroups.
All these stats are recorded using start and stop events. When reading these
stats, we do not add the delta between the current time and the last start time
if we're between the start and stop events. We avoid doing this to make sure
that these numbers are always monotonically increasing when read. Since we're
using sched_clock() which may use the tsc as its source, it may induce some
inconsistency (due to tsc resync across cpus) if we included the current delta.
Divyesh Shah [Fri, 9 Apr 2010 04:15:10 +0000 (21:15 -0700)]
blkio: Add io_queued and avg_queue_size stats
These stats are useful for getting a feel for the queue depth of the cgroup,
i.e., how filled up its queues are at a given instant and over the existence of
the cgroup. This ability is useful when debugging problems in the wild as it
helps understand the application's IO pattern w/o having to read through the
userspace code (coz its tedious or just not available) or w/o the ability
to run blktrace (since you may not have root access and/or not want to disturb
performance).
Divyesh Shah [Fri, 9 Apr 2010 04:14:23 +0000 (21:14 -0700)]
blkio: Add io_merged stat
This includes both the number of bios merged into requests belonging to this
cgroup as well as the number of requests merged together.
In the past, we've observed different merging behavior across upstream kernels,
some by design some actual bugs. This stat helps a lot in debugging such
problems when applications report decreased throughput with a new kernel
version.
This needed adding an extra elevator function to capture bios being merged as I
did not want to pollute elevator code with blkiocg knowledge and hence needed
the accounting invocation to come from CFQ.
Divyesh Shah [Fri, 9 Apr 2010 06:31:19 +0000 (08:31 +0200)]
blkio: Changes to IO controller additional stats patches
that include some minor fixes and addresses all comments.
Changelog: (most based on Vivek Goyal's comments)
o renamed blkiocg_reset_write to blkiocg_reset_stats
o more clarification in the documentation on io_service_time and io_wait_time
o Initialize blkg->stats_lock
o rename io_add_stat to blkio_add_stat and declare it static
o use bool for direction and sync
o derive direction and sync info from existing rq methods
o use 12 for major:minor string length
o define io_service_time better to cover the NCQ case
o add a separate reset_stats interface
o make the indexed stats a 2d array to simplify macro and function pointer code
o blkio.time now exports in jiffies as before
o Added stats description in patch description and
Documentation/cgroup/blkio-controller.txt
o Prefix all stats functions with blkio and make them static as applicable
o replace IO_TYPE_MAX with IO_TYPE_TOTAL
o Moved #define constant to top of blk-cgroup.c
o Pass dev_t around instead of char *
o Add note to documentation file about resetting stats
o use BLK_CGROUP_MODULE in addition to BLK_CGROUP config option in #ifdef
statements
o Avoid struct request specific knowledge in blk-cgroup. blk-cgroup.h now has
rq_direction() and rq_sync() functions which are used by CFQ and when using
io-controller at a higher level, bio_* functions can be added.
Dave Airlie [Fri, 9 Apr 2010 04:27:51 +0000 (14:27 +1000)]
Merge remote branch 'nouveau/for-airlied' of ../drm-nouveau-next into drm-linus
* 'nouveau/for-airlied' of ../drm-nouveau-next: (21 commits)
drm/nouveau: bail out of auxch transaction if we repeatedly recieve defers
drm/nv50: implement gpio set/get routines
drm/nv50: parse/use some more de-magiced parts of gpio table entries
drm/nouveau: store raw gpio table entry in bios gpio structs
drm/nv40: Init some tiling-related PGRAPH state.
drm/nv50: Add NVA3 support in ctxprog/ctxvals generator.
drm/nv50: another dodgy DP hack
drm/nv50: punt hotplug irq handling out to workqueue
drm/nv50: preserve an unknown SOR_MODECTRL value for DP encoders
drm/nv50: Allow using the NVA3 new compute class.
drm/nv50: cleanup properly if PDISPLAY init fails
drm/nouveau: fixup the init failure paths some more
drm/nv50: fix instmem init on IGPs if stolen mem crosses 4GiB mark
drm/nv40: add LVDS table quirk for Dell Latitude D620
drm/nv40: rework lvds table parsing
drm/nouveau: detect vram amount once, and save the value
drm/nouveau: remove some unused members from drm_nouveau_private
drm/nouveau: Make use of TTM busy_placements.
drm/nv50: add more 0x100c80 flushy magic
drm/nv50: fix fbcon when framebuffer above 4GiB mark
...
Francisco Jerez [Tue, 6 Apr 2010 19:11:58 +0000 (21:11 +0200)]
drm/nv40: Init some tiling-related PGRAPH state.
Fixes garbled 3D on an nv46 card.
Reported-by: Francesco Marella <francesco.marella@gmail.com> Signed-off-by: Francisco Jerez <currojerez@riseup.net> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Ben Skeggs [Tue, 30 Mar 2010 06:01:41 +0000 (16:01 +1000)]
drm/nv50: another dodgy DP hack
Allows *some* DP cards to keep working in some corner cases that most
people shouldn't hit. I hit it all the time with development, so this
can stay for now.
Ben Skeggs [Mon, 29 Mar 2010 00:06:09 +0000 (10:06 +1000)]
drm/nv50: preserve an unknown SOR_MODECTRL value for DP encoders
This value interacts with some registers we don't currently know how to
program properly ourselves. The default of 5 that we were using matches
what the VBIOS on early DP cards do, but later ones use 6, which would
cause nouveau to program an incorrect mode on these chips.