Revert commit 534b483a86 ("cpumask: don't perform while loop in
cpumask_next_and()").
This was a minor optimization, but it puts a `struct cpumask' on the
stack, which consumes too much stack space.
Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reported-by: Peter Zijlstra <peterz@infradead.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull more MIPS fixes from Ralf Baechle:
"Another round of 4.1 MIPS fixes, one fix to a MIPS-specific #if
condition in lib/mpi, one fix to the MIPS GIC irqchip driver and one
SSB fix.
Details:
- fix handling of clock in chipco SSB driver.
- fix two MIPS-specific #if conditions to correctly work for GCC 5.1.
- fix damage to R6 pgtable bits done by XPA support.
- fix possible crash due to unloading modules that contain statically
defined platform devices.
- fix disabling of the MSA ASE on context switch to also work
correctly when a new thread/process has the CPU for the very first
time.
This is part of linux-next and has been beaten to death on
Imagination's test farm.
While things are not looking too grim this pull request also means the
rate of fixes for 4.1 remains nearly constant so I'd not be unhappy if
you'd delay the release"
* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
MPI: MIPS: Fix compilation error with GCC 5.1
IRQCHIP: mips-gic: Don't nest calls to do_IRQ()
MIPS: MSA: bugfix - disable MSA correctly for new threads/processes.
MIPS: Loongson: Do not register 8250 platform device from module.
MIPS: Cobalt: Do not build MTD platform device registration code as module.
SSB: Fix handling of ssb_pmu_get_alp_clock()
MIPS: pgtable-bits: Fix XPA damage to R6 definitions.
Pull networking fixes from David Miller:
1) Fix stack allocation in s390 BPF JIT, from Michael Holzheu.
2) Disable LRO on openvswitch paths, from Jiri Benc.
3) UDP early demux doesn't handle multicast group membership properly,
fix from Shawn Bohrer.
4) Fix TX queue hang due to incorrect handling of mixed sized fragments
and linearlization in i40e driver, from Anjali Singhai Jain.
5) Cannot use disable_irq() in timer handler of AMD xgbe driver, from
Thomas Lendacky.
6) b2net driver improperly assumes pci_alloc_consistent() gives zero'd
out memory, use dma_zalloc_coherent(). From Sriharsha Basavapatna.
7) Fix use-after-free in MPLS and ipv6, from Robert Shearman.
8) Missing neif_napi_del() calls in cleanup paths of b44 driver, from
Hauke Mehrtens.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
net: replace last open coded skb_orphan_frags with function call
net: bcmgenet: power on MII block for all MII modes
ipv6: Fix protocol resubmission
ipv6: fix possible use after free of dev stats
b44: call netif_napi_del()
bridge: disable softirqs around br_fdb_update to avoid lockup
Revert "bridge: use _bh spinlock variant for br_fdb_update to avoid lockup"
mpls: fix possible use after free of device
be2net: Replace dma/pci_alloc_coherent() calls with dma_zalloc_coherent()
bridge: use _bh spinlock variant for br_fdb_update to avoid lockup
amd-xgbe: Use disable_irq_nosync from within timer function
rhashtable: add missing import <linux/export.h>
i40e: Make sure to be in VEB mode if SRIOV is enabled at probe
i40e: start up in VEPA mode by default
i40e/i40evf: Fix mixed size frags and linearization
ipv4/udp: Verify multicast group is ours in upd_v4_early_demux()
openvswitch: disable LRO
s390/bpf: fix bpf frame pointer setup
s390/bpf: fix stack allocation
rhashtable uses EXPORT_SYMBOL_GPL() without importing linux/export.h
directly it is only imported indirectly through some other includes.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull swiotlb fix from Konrad Rzeszutek Wilk:
"Tiny little fix which just converts an function to be static. Really
tiny"
* 'stable/for-linus-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/swiotlb:
swiotlb: do not export map_single function
The map_single() function is not defined as static, even though it
doesn't seem to be used anywhere else in the kernel. Make it static to
avoid namespace pollution since this is a rather generic symbol.
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
strnlen_user() can return a number in a range 0 to count +
sizeof(unsigned long) - 1. Clarify the comment at the top of the
function so that users don't think the function returns at most count+1.
Signed-off-by: Jan Kara <jack@suse.cz>
[ Also added commentary about preferably not using this function ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If the specified maximum length of the string is a multiple of unsigned
long, we would load one long behind the specified maximum. If that
happens to be in a next page, we can hit a page fault although we were
not expected to.
Fix the off-by-one bug in the test whether we are at the end of the
specified range.
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Changes in this update:
o regression fix for new rename whiteout code
o regression fixes for new superblock generic per-cpu counter code
o fix for incorrect error return sign introduced in 3.17
o metadata corruption fixes that need to go back to -stable kernels
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)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=w5SP
-----END PGP SIGNATURE-----
Merge tag 'xfs-for-linus-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs
Pull xfs fixes from Dave Chinner:
"This is a little larger than I'd like late in the release cycle, but
all the fixes are for regressions introduced in the 4.1-rc1 merge, or
are needed back in -stable kernels fairly quickly as they are
filesystem corruption or userspace visible correctness issues.
Changes in this update:
- regression fix for new rename whiteout code
- regression fixes for new superblock generic per-cpu counter code
- fix for incorrect error return sign introduced in 3.17
- metadata corruption fixes that need to go back to -stable kernels"
* tag 'xfs-for-linus-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs:
xfs: fix broken i_nlink accounting for whiteout tmpfile inode
xfs: xfs_iozero can return positive errno
xfs: xfs_attr_inactive leaves inconsistent attr fork state behind
xfs: extent size hints can round up extents past MAXEXTLEN
xfs: inode and free block counters need to use __percpu_counter_compare
percpu_counter: batch size aware __percpu_counter_compare()
xfs: use percpu_counter_read_positive for mp->m_icount
Two fixes which got lost in my recent distraction. One is a weird
cpumask function which needed to be rewritten, the other is a module
bug which is cc:stable.
Thanks,
Rusty.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=wuP+
-----END PGP SIGNATURE-----
Merge tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull fixes for cpumask and modules from Rusty Russell:
"** NOW WITH TESTING! **
Two fixes which got lost in my recent distraction. One is a weird
cpumask function which needed to be rewritten, the other is a module
bug which is cc:stable"
* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
cpumask_set_cpu_local_first => cpumask_local_spread, lament
module: Call module notifier on failure after complete_formation()
XFS uses non-stanard batch sizes for avoiding frequent global
counter updates on it's allocated inode counters, as they increment
or decrement in batches of 64 inodes. Hence the standard percpu
counter batch of 32 means that the counter is effectively a global
counter. Currently Xfs uses a batch size of 128 so that it doesn't
take the global lock on every single modification.
However, Xfs also needs to compare accurately against zero, which
means we need to use percpu_counter_compare(), and that has a
hard-coded batch size of 32, and hence will spuriously fail to
detect when it is supposed to use precise comparisons and hence
the accounting goes wrong.
Add __percpu_counter_compare() to take a custom batch size so we can
use it sanely in XFS and factor percpu_counter_compare() to use it.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Dave Chinner <david@fromorbit.com>
da91309e0a (cpumask: Utility function to set n'th cpu...) created a
genuinely weird function. I never saw it before, it went through DaveM.
(He only does this to make us other maintainers feel better about our own
mistakes.)
cpumask_set_cpu_local_first's purpose is say "I need to spread things
across N online cpus, choose the ones on this numa node first"; you call
it in a loop.
It can fail. One of the two callers ignores this, the other aborts and
fails the device open.
It can fail in two ways: allocating the off-stack cpumask, or through a
convoluted codepath which AFAICT can only occur if cpu_online_mask
changes. Which shouldn't happen, because if cpu_online_mask can change
while you call this, it could return a now-offline cpu anyway.
It contains a nonsensical test "!cpumask_of_node(numa_node)". This was
drawn to my attention by Geert, who said this causes a warning on Sparc.
It sets a single bit in a cpumask instead of returning a cpu number,
because that's what the callers want.
It could be made more efficient by passing the previous cpu rather than
an index, but that would be more invasive to the callers.
Fixes: da91309e0a
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (then rebased)
Tested-by: Amir Vadai <amirv@mellanox.com>
Acked-by: Amir Vadai <amirv@mellanox.com>
Acked-by: David S. Miller <davem@davemloft.net>
We currently have no limit on the number of elements in a hash table.
This is a problem because some users (tipc) set a ceiling on the
maximum table size and when that is reached the hash table may
degenerate. Others may encounter OOM when growing and if we allow
insertions when that happens the hash table perofrmance may also
suffer.
This patch adds a new paramater insecure_max_entries which becomes
the cap on the table. If unset it defaults to max_size * 2. If
it is also zero it means that there is no cap on the number of
elements in the table. However, the table will grow whenever the
utilisation hits 100% and if that growth fails, you will get ENOMEM
on insertion.
As allowing oversubscription is potentially dangerous, the name
contains the word insecure.
Note that the cap is not a hard limit. This is done for performance
reasons as enforcing a hard limit will result in use of atomic ops
that are heavier than the ones we currently use.
The reasoning is that we're only guarding against a gross over-
subscription of the table, rather than a small breach of the limit.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull RCU fix from Ingo Molnar:
"An RCU Kconfig fix that eliminates an annoying interactive kconfig
question for CONFIG_RCU_TORTURE_TEST_SLOW_INIT"
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
rcu: Control grace-period delays directly from value
The documentation shows a need for gcc > 4.9.2, but it's really >=. The
Kconfig entries don't show require versions so add them. Correct a
latter/later typo too. Also mention that gcc 5 required to catch out of
bounds accesses to global and stack variables.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Andrey Ryabinin <a.ryabinin@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The file lib/find_last_bit.c was no longer used and supposed to be
deleted by commit 8f6f19dd51 ("lib: move find_last_bit to
lib/find_next_bit.c") but that delete didn't happen. This gets rid of
it.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull crypto fixes from Herbert Xu:
"This fixes a build problem with bcm63xx and yet another fix to the
memzero_explicit function to ensure that the memset is not elided"
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
hwrng: bcm63xx - Fix driver compilation
lib: make memzero_explicit more robust against dead store elimination
In commit 0b053c9518 ("lib: memzero_explicit: use barrier instead
of OPTIMIZER_HIDE_VAR"), we made memzero_explicit() more robust in
case LTO would decide to inline memzero_explicit() and eventually
find out it could be elimiated as dead store.
While using barrier() works well for the case of gcc, recent efforts
from LLVMLinux people suggest to use llvm as an alternative to gcc,
and there, Stephan found in a simple stand-alone user space example
that llvm could nevertheless optimize and thus elimitate the memset().
A similar issue has been observed in the referenced llvm bug report,
which is regarded as not-a-bug.
Based on some experiments, icc is a bit special on its own, while it
doesn't seem to eliminate the memset(), it could do so with an own
implementation, and then result in similar findings as with llvm.
The fix in this patch now works for all three compilers (also tested
with more aggressive optimization levels). Arguably, in the current
kernel tree it's more of a theoretical issue, but imho, it's better
to be pedantic about it.
It's clearly visible with gcc/llvm though, with the below code: if we
would have used barrier() only here, llvm would have omitted clearing,
not so with barrier_data() variant:
static inline void memzero_explicit(void *s, size_t count)
{
memset(s, 0, count);
barrier_data(s);
}
int main(void)
{
char buff[20];
memzero_explicit(buff, sizeof(buff));
return 0;
}
$ gcc -O2 test.c
$ gdb a.out
(gdb) disassemble main
Dump of assembler code for function main:
0x0000000000400400 <+0>: lea -0x28(%rsp),%rax
0x0000000000400405 <+5>: movq $0x0,-0x28(%rsp)
0x000000000040040e <+14>: movq $0x0,-0x20(%rsp)
0x0000000000400417 <+23>: movl $0x0,-0x18(%rsp)
0x000000000040041f <+31>: xor %eax,%eax
0x0000000000400421 <+33>: retq
End of assembler dump.
$ clang -O2 test.c
$ gdb a.out
(gdb) disassemble main
Dump of assembler code for function main:
0x00000000004004f0 <+0>: xorps %xmm0,%xmm0
0x00000000004004f3 <+3>: movaps %xmm0,-0x18(%rsp)
0x00000000004004f8 <+8>: movl $0x0,-0x8(%rsp)
0x0000000000400500 <+16>: lea -0x18(%rsp),%rax
0x0000000000400505 <+21>: xor %eax,%eax
0x0000000000400507 <+23>: retq
End of assembler dump.
As gcc, clang, but also icc defines __GNUC__, it's sufficient to define
this in compiler-gcc.h only to be picked up. For a fallback or otherwise
unsupported compiler, we define it as a barrier. Similarly, for ecc which
does not support gcc inline asm.
Reference: https://llvm.org/bugs/show_bug.cgi?id=15495
Reported-by: Stephan Mueller <smueller@chronox.de>
Tested-by: Stephan Mueller <smueller@chronox.de>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Stephan Mueller <smueller@chronox.de>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: mancha security <mancha1@zoho.com>
Cc: Mark Charlebois <charlebm@gmail.com>
Cc: Behan Webster <behanw@converseincode.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Pull networking fixes from David Miller:
1) mlx4 doesn't check fully for supported valid RSS hash function, fix
from Amir Vadai
2) Off by one in ibmveth_change_mtu(), from David Gibson
3) Prevent altera chip from reporting false error interrupts in some
circumstances, from Chee Nouk Phoon
4) Get rid of that stupid endless loop trying to allocate a FIN packet
in TCP, and in the process kill deadlocks. From Eric Dumazet
5) Fix get_rps_cpus() crash due to wrong invalid-cpu value, also from
Eric Dumazet
6) Fix two bugs in async rhashtable resizing, from Thomas Graf
7) Fix topology server listener socket namespace bug in TIPC, from Ying
Xue
8) Add some missing HAS_DMA kconfig dependencies, from Geert
Uytterhoeven
9) bgmac driver intends to force re-polling but does so by returning
the wrong value from it's ->poll() handler. Fix from Rafał Miłecki
10) When the creater of an rhashtable configures a max size for it,
don't bark in the logs and drop insertions when that is exceeded.
Fix from Johannes Berg
11) Recover from out of order packets in ppp mppe properly, from Sylvain
Rochet
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (41 commits)
bnx2x: really disable TPA if 'disable_tpa' option is set
net:treewide: Fix typo in drivers/net
net/mlx4_en: Prevent setting invalid RSS hash function
mdio-mux-gpio: use new gpiod_get_array and gpiod_put_array functions
netfilter; Add some missing default cases to switch statements in nft_reject.
ppp: mppe: discard late packet in stateless mode
ppp: mppe: sanity error path rework
net/bonding: Make DRV macros private
net: rfs: fix crash in get_rps_cpus()
altera tse: add support for fixed-links.
pxa168: fix double deallocation of managed resources
net: fix crash in build_skb()
net: eth: altera: Resolve false errors from MSGDMA to TSE
ehea: Fix memory hook reference counting crashes
net/tg3: Release IRQs on permanent error
net: mdio-gpio: support access that may sleep
inet: fix possible panic in reqsk_queue_unlink()
rhashtable: don't attempt to grow when at max_size
bgmac: fix requests for extra polling calls from NAPI
tcp: avoid looping in tcp_send_fin()
...
Highlights:
- "experimental" code for managing md/raid1 across a cluster using
DLM. Code is not ready for general use and triggers a WARNING if used.
However it is looking good and mostly done and having in mainline
will help co-ordinate development.
- RAID5/6 can now batch multiple (4K wide) stripe_heads so as to
handle a full (chunk wide) stripe as a single unit.
- RAID6 can now perform read-modify-write cycles which should
help performance on larger arrays: 6 or more devices.
- RAID5/6 stripe cache now grows and shrinks dynamically. The value
set is used as a minimum.
- Resync is now allowed to go a little faster than the 'mininum' when
there is competing IO. How much faster depends on the speed of the
devices, so the effective minimum should scale with device speed to
some extent.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=OHG6
-----END PGP SIGNATURE-----
Merge tag 'md/4.1' of git://neil.brown.name/md
Pull md updates from Neil Brown:
"More updates that usual this time. A few have performance impacts
which hould mostly be positive, but RAID5 (in particular) can be very
work-load ensitive... We'll have to wait and see.
Highlights:
- "experimental" code for managing md/raid1 across a cluster using
DLM. Code is not ready for general use and triggers a WARNING if
used. However it is looking good and mostly done and having in
mainline will help co-ordinate development.
- RAID5/6 can now batch multiple (4K wide) stripe_heads so as to
handle a full (chunk wide) stripe as a single unit.
- RAID6 can now perform read-modify-write cycles which should help
performance on larger arrays: 6 or more devices.
- RAID5/6 stripe cache now grows and shrinks dynamically. The value
set is used as a minimum.
- Resync is now allowed to go a little faster than the 'mininum' when
there is competing IO. How much faster depends on the speed of the
devices, so the effective minimum should scale with device speed to
some extent"
* tag 'md/4.1' of git://neil.brown.name/md: (58 commits)
md/raid5: don't do chunk aligned read on degraded array.
md/raid5: allow the stripe_cache to grow and shrink.
md/raid5: change ->inactive_blocked to a bit-flag.
md/raid5: move max_nr_stripes management into grow_one_stripe and drop_one_stripe
md/raid5: pass gfp_t arg to grow_one_stripe()
md/raid5: introduce configuration option rmw_level
md/raid5: activate raid6 rmw feature
md/raid6 algorithms: xor_syndrome() for SSE2
md/raid6 algorithms: xor_syndrome() for generic int
md/raid6 algorithms: improve test program
md/raid6 algorithms: delta syndrome functions
raid5: handle expansion/resync case with stripe batching
raid5: handle io error of batch list
RAID5: batch adjacent full stripe write
raid5: track overwrite disk count
raid5: add a new flag to track if a stripe can be batched
raid5: use flex_array for scribble data
md raid0: access mddev->queue (request queue member) conditionally because it is not set when accessed from dm-raid
md: allow resync to go faster when there is competing IO.
md: remove 'go_faster' option from ->sync_request()
...
The current code currently only stops inserting rehashes into the
chain when no resizes are currently scheduled. As long as resizes
are scheduled and while inserting above the utilization watermark,
more and more rehashes will be scheduled.
This lead to a perfect DoS storm with thousands of rehashes
scheduled which lead to thousands of spinlocks to be taken
sequentially.
Instead, only allow either a series of resizes or a single rehash.
Drop any further rehashes and return -EBUSY.
Fixes: ccd57b1bd3 ("rhashtable: Add immediate rehash during insertion")
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
When rhashtable_insert_rehash() fails with ENOMEM, this indicates that
we can't allocate the necessary memory in the current context but the
limits as set by the user would still allow to grow.
Thus attempt an async resize in the background where we can allocate
using GFP_KERNEL which is more likely to succeed. The insertion itself
will still fail to indicate pressure.
This fixes a bug where the table would never continue growing once the
utilization is above 100%.
Fixes: ccd57b1bd3 ("rhashtable: Add immediate rehash during insertion")
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull sparc fixes from David Miller:
1) ldc_alloc_exp_dring() can be called from softints, so use
GFP_ATOMIC. From Sowmini Varadhan.
2) Some minor warning/build fixups for the new iommu-common code on
certain archs and with certain debug options enabled. Also from
Sowmini Varadhan.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
sparc: Use GFP_ATOMIC in ldc_alloc_exp_dring() as it can be called in softirq context
sparc64: Use M7 PMC write on all chips T4 and onward.
iommu-common: rename iommu_pool_hash to iommu_hash_common
iommu-common: fix x86_64 compiler warnings
The second and (last) optimized XOR syndrome calculation. This version
supports right and left side optimization. All CPUs with architecture
older than Haswell will benefit from it.
It should be noted that SSE2 movntdq kills performance for memory areas
that are read and written simultaneously in chunks smaller than cache
line size. So use movdqa instead for P/Q writes in sse21 and sse22 XOR
functions.
Signed-off-by: Markus Stockhausen <stockhausen@collogia.de>
Signed-off-by: NeilBrown <neilb@suse.de>
Start the algorithms with the very basic one. It is left and right
optimized. That means we can avoid all calculations for unneeded pages
above the right stop offset. For pages below the left start offset we
still need the syndrome multiplication but without reading data pages.
Signed-off-by: Markus Stockhausen <stockhausen@collogia.de>
Signed-off-by: NeilBrown <neilb@suse.de>
It is always helpful to have a test tool in place if we implement
new data critical algorithms. So add some test routines to the raid6
checker that can prove if the new xor_syndrome() works as expected.
Run through all permutations of start/stop pages per algorithm and
simulate a xor_syndrome() assisted rmw run. After each rmw check if
the recovery algorithm still confirms that the stripe is fine.
Signed-off-by: Markus Stockhausen <stockhausen@collogia.de>
Signed-off-by: NeilBrown <neilb@suse.de>
v3: s-o-b comment, explanation of performance and descision for
the start/stop implementation
Implementing rmw functionality for RAID6 requires optimized syndrome
calculation. Up to now we can only generate a complete syndrome. The
target P/Q pages are always overwritten. With this patch we provide
a framework for inplace P/Q modification. In the first place simply
fill those functions with NULL values.
xor_syndrome() has two additional parameters: start & stop. These
will indicate the first and last page that are changing during a
rmw run. That makes it possible to avoid several unneccessary loops
and speed up calculation. The caller needs to implement the following
logic to make the functions work.
1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source
blocks inside P/Q between (and including) start and end.
2) modify any block with start <= block <= stop
3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of
source blocks into P/Q between (and including) start and end.
Pages between start and stop that won't be changed should be filled
with a pointer to the kernel zero page. The reasons for not taking NULL
pages are:
1) Algorithms cross the whole source data line by line. Thus avoid
additional branches.
2) Having a NULL page avoids calculating the XOR P parity but still
need calulation steps for the Q parity. Depending on the algorithm
unrolling that might be only a difference of 2 instructions per loop.
The benchmark numbers of the gen_syndrome() functions are displayed in
the kernel log. Do the same for the xor_syndrome() functions. This
will help to analyze performance problems and give an rough estimate
how well the algorithm works. The choice of the fastest algorithm will
still depend on the gen_syndrome() performance.
With the start/stop page implementation the speed can vary a lot in real
life. E.g. a change of page 0 & page 15 on a stripe will be harder to
compute than the case where page 0 & page 1 are XOR candidates. To be not
to enthusiatic about the expected speeds we will run a worse case test
that simulates a change on the upper half of the stripe. So we do:
1) calculation of P/Q for the upper pages
2) continuation of Q for the lower (empty) pages
Signed-off-by: Markus Stockhausen <stockhausen@collogia.de>
Signed-off-by: NeilBrown <neilb@suse.de>
Here's the big char/misc driver patchset for 4.1-rc1.
Lots of different driver subsystem updates here, nothing major, full
details are in the shortlog below.
All of this has been in linux-next for a while.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
iEYEABECAAYFAlU2IMEACgkQMUfUDdst+yloDQCfbyIRL23WVAn9ckQse/y8gbjB
OT4AoKTJbwndDP9Kb/lrj2tjd9QjNVrC
=xhen
-----END PGP SIGNATURE-----
Merge tag 'char-misc-4.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc driver updates from Greg KH:
"Here's the big char/misc driver patchset for 4.1-rc1.
Lots of different driver subsystem updates here, nothing major, full
details are in the shortlog.
All of this has been in linux-next for a while"
* tag 'char-misc-4.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (133 commits)
mei: trace: remove unused TRACE_SYSTEM_STRING
DTS: ARM: OMAP3-N900: Add lis3lv02d support
Documentation: DT: lis302: update wakeup binding
lis3lv02d: DT: add wakeup unit 2 and wakeup threshold
lis3lv02d: DT: use s32 to support negative values
Drivers: hv: hv_balloon: correctly handle num_pages>INT_MAX case
Drivers: hv: hv_balloon: correctly handle val.freeram<num_pages case
mei: replace check for connection instead of transitioning
mei: use mei_cl_is_connected consistently
mei: fix mei_poll operation
hv_vmbus: Add gradually increased delay for retries in vmbus_post_msg()
Drivers: hv: hv_balloon: survive ballooning request with num_pages=0
Drivers: hv: hv_balloon: eliminate jumps in piecewiese linear floor function
Drivers: hv: hv_balloon: do not online pages in offline blocks
hv: remove the per-channel workqueue
hv: don't schedule new works in vmbus_onoffer()/vmbus_onoffer_rescind()
hv: run non-blocking message handlers in the dispatch tasklet
coresight: moving to new "hwtracing" directory
coresight-tmc: Adding a status interface to sysfs
coresight: remove the unnecessary configuration coresight-default-sink
...
When CONFIG_DEBUG_FORCE_WEAK_PER_CPU is set, the DEFINE_PER_CPU_SECTION
macro will define an extern __pcpu_unique_##name variable that could
conflict with the same definition in powerpc at this time. Avoid that
conflict by renaming iommu_pool_hash in iommu-common.c
Thanks to Guenter Roeck for catching this, and helping to test the fix.
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
functions, prompted by their mis-use in staging.
With these function removed, all cpu functions should only iterate to
nr_cpu_ids, so we finally only allocate that many bits when cpumasks
are allocated offstack.
Thanks,
Rusty.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=VAgA
-----END PGP SIGNATURE-----
Merge tag 'cpumask-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull final removal of deprecated cpus_* cpumask functions from Rusty Russell:
"This is the final removal (after several years!) of the obsolete
cpus_* functions, prompted by their mis-use in staging.
With these function removed, all cpu functions should only iterate to
nr_cpu_ids, so we finally only allocate that many bits when cpumasks
are allocated offstack"
* tag 'cpumask-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux: (25 commits)
cpumask: remove __first_cpu / __next_cpu
cpumask: resurrect CPU_MASK_CPU0
linux/cpumask.h: add typechecking to cpumask_test_cpu
cpumask: only allocate nr_cpumask_bits.
Fix weird uses of num_online_cpus().
cpumask: remove deprecated functions.
mips: fix obsolete cpumask_of_cpu usage.
x86: fix more deprecated cpu function usage.
ia64: remove deprecated cpus_ usage.
powerpc: fix deprecated CPU_MASK_CPU0 usage.
CPU_MASK_ALL/CPU_MASK_NONE: remove from deprecated region.
staging/lustre/o2iblnd: Don't use cpus_weight
staging/lustre/libcfs: replace deprecated cpus_ calls with cpumask_
staging/lustre/ptlrpc: Do not use deprecated cpus_* functions
blackfin: fix up obsolete cpu function usage.
parisc: fix up obsolete cpu function usage.
tile: fix up obsolete cpu function usage.
arm64: fix up obsolete cpu function usage.
mips: fix up obsolete cpu function usage.
x86: fix up obsolete cpu function usage.
...
The test_data_1_le[] array is a const array of const char *. To avoid
dropping any const information, we need to use "const char * const *",
not just "const char **".
I'm not sure why the different test arrays end up having different
const'ness, but let's make the pointer we use to traverse them as const
as possible, since we modify neither the array of pointers _or_ the
pointers we find in the array.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
They were for use by the deprecated first_cpu() and next_cpu() wrappers,
but sparc used them directly.
They're now replaced by cpumask_first / cpumask_next. And __next_cpu_nr
is completely obsolete.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: David S. Miller <davem@davemloft.net>
Fixes warnings due to
- no DMA_ERROR_CODE on PARISC,
- sizeof (unsigned long) == 4 bytes on PARISC.
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Investigation of multithreaded iperf experiments on an ethernet
interface show the iommu->lock as the hottest lock identified by
lockstat, with something of the order of 21M contentions out of
27M acquisitions, and an average wait time of 26 us for the lock.
This is not efficient. A more scalable design is to follow the ppc
model, where the iommu_map_table has multiple pools, each stretching
over a segment of the map, and with a separate lock for each pool.
This model allows for better parallelization of the iommu map search.
This patch adds the iommu range alloc/free function infrastructure.
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
I applied the wrong version of this patch series, V4 instead
of V10, due to a patchwork bundling snafu.
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull RCU fix from Paul E. McKenney:
"This series contains a single change that fixes Kconfig asking pointless
questions."
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull sparc updates from David Miller:
"The PowerPC folks have a really nice scalable IOMMU pool allocator
that we wanted to make use of for sparc. So here we have a series
that abstracts out their code into a common layer that anyone can make
use of.
Sparc is converted, and the PowerPC folks have reviewed and ACK'd this
series and plan to convert PowerPC over as well"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
iommu-common: Fix PARISC compile-time warnings
sparc: Make LDC use common iommu poll management functions
sparc: Make sparc64 use scalable lib/iommu-common.c functions
sparc: Break up monolithic iommu table/lock into finer graularity pools and lock
Fixes warnings due to
- no DMA_ERROR_CODE on PARISC,
- sizeof (unsigned long) == 4 bytes on PARISC.
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Merge third patchbomb from Andrew Morton:
- various misc things
- a couple of lib/ optimisations
- provide DIV_ROUND_CLOSEST_ULL()
- checkpatch updates
- rtc tree
- befs, nilfs2, hfs, hfsplus, fatfs, adfs, affs, bfs
- ptrace fixes
- fork() fixes
- seccomp cleanups
- more mmap_sem hold time reductions from Davidlohr
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (138 commits)
proc: show locks in /proc/pid/fdinfo/X
docs: add missing and new /proc/PID/status file entries, fix typos
drivers/rtc/rtc-at91rm9200.c: make IO endian agnostic
Documentation/spi/spidev_test.c: fix warning
drivers/rtc/rtc-s5m.c: allow usage on device type different than main MFD type
.gitignore: ignore *.tar
MAINTAINERS: add Mediatek SoC mailing list
tomoyo: reduce mmap_sem hold for mm->exe_file
powerpc/oprofile: reduce mmap_sem hold for exe_file
oprofile: reduce mmap_sem hold for mm->exe_file
mips: ip32: add platform data hooks to use DS1685 driver
lib/Kconfig: fix up HAVE_ARCH_BITREVERSE help text
x86: switch to using asm-generic for seccomp.h
sparc: switch to using asm-generic for seccomp.h
powerpc: switch to using asm-generic for seccomp.h
parisc: switch to using asm-generic for seccomp.h
mips: switch to using asm-generic for seccomp.h
microblaze: use asm-generic for seccomp.h
arm: use asm-generic for seccomp.h
seccomp: allow COMPAT sigreturn overrides
...
Cc: Yalin Wang <yalin.wang@sonymobile.com>
Cc: Russell King <linux@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
cpumask_next_and() is looking for cpumask_next() in src1 in a loop and
tests if found cpu is also present in src2. remove that loop, perform
cpumask_and() of src1 and src2 first and use that new mask to find
cpumask_next().
Apart from removing while loop, ./bloat-o-meter on x86_64 shows
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-8 (-8)
function old new delta
cpumask_next_and 62 54 -8
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
bitmap_empty() has its own implementation. But it's clearly as simple as:
find_first_bit(src, nbits) == nbits
The same is true for 'bitmap_full'.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Cc: George Spelvin <linux@horizon.com>
Cc: Alexey Klimov <klimov.linux@gmail.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
I hadn't had enough coffee when I wrote this. Currently, the final
increment of buf depends on the value loaded from the table, and
causes gcc to emit a cmov immediately before the return. It is smarter
to let it depend on r, since the increment can then be computed in
parallel with the final load/store pair. It also shaves 16 bytes of
.text.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
bucket_find_contain() will search the bucket list for a dma_debug_entry.
When the entry isn't found it needs to search other buckets too, since
only the start address of a dma range is hashed (which might be in a
different bucket).
A copy of the dma_debug_entry is used to get the previous hash bucket
but when its list is searched the original dma_debug_entry is to be used
not its modified copy.
This fixes false "device driver tries to sync DMA memory it has not allocated"
warnings.
Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Horia Geanta <horia.geanta@freescale.com>
Cc: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The most expensive part of decimal conversion is the divisions by 10
(albeit done using reciprocal multiplication with appropriately chosen
constants). I decided to see if one could eliminate around half of
these multiplications by emitting two digits at a time, at the cost of a
200 byte lookup table, and it does indeed seem like there is something
to be gained, especially on 64 bits. Microbenchmarking shows
improvements ranging from -50% (for numbers uniformly distributed in [0,
2^64-1]) to -25% (for numbers heavily biased toward the smaller end, a
more realistic distribution).
On a larger scale, perf shows that top, one of the big consumers of /proc
data, uses 0.5-1.0% fewer cpu cycles.
I had to jump through some hoops to get the 32 bit code to compile and run
on my 64 bit machine, so I'm not sure how relevant these numbers are, but
just for comparison the microbenchmark showed improvements between -30%
and -10%.
The bloat-o-meter costs are around 150 bytes (the generated code is a
little smaller, so it's not the full 200 bytes) on both 32 and 64 bit.
I'm aware that extra cache misses won't show up in a microbenchmark as
used above, but on the other hand decimal conversions often happen in bulk
(for example in the case of top).
I have of course tested that the new code generates the same output as the
old, for both the first and last 1e10 numbers in [0,2^64-1] and 4e9
'random' numbers in-between.
Test and verification code on github: https://github.com/Villemoes/dec.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Tested-by: Jeff Epler <jepler@unpythonic.net>
Cc: "Peter Zijlstra (Intel)" <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This file contains implementation for all find_*_bit{,_le}
So giving it more generic name looks reasonable.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Reviewed-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Reviewed-by: George Spelvin <linux@horizon.com>
Cc: Alexey Klimov <klimov.linux@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Mark Salter <msalter@redhat.com>
Cc: AKASHI Takahiro <takahiro.akashi@linaro.org>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Valentin Rothberg <valentinrothberg@gmail.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently all 'find_*_bit' family is located in lib/find_next_bit.c,
except 'find_last_bit', which is in lib/find_last_bit.c. It seems,
there's no major benefit to have it separated.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Reviewed-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Reviewed-by: George Spelvin <linux@horizon.com>
Cc: Alexey Klimov <klimov.linux@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Mark Salter <msalter@redhat.com>
Cc: AKASHI Takahiro <takahiro.akashi@linaro.org>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Valentin Rothberg <valentinrothberg@gmail.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patchset does rework to find_bit function family to achieve better
performance, and decrease size of text. All rework is done in patch 1.
Patches 2 and 3 are about code moving and renaming.
It was boot-tested on x86_64 and MIPS (big-endian) machines.
Performance tests were ran on userspace with code like this:
/* addr[] is filled from /dev/urandom */
start = clock();
while (ret < nbits)
ret = find_next_bit(addr, nbits, ret + 1);
end = clock();
printf("%ld\t", (unsigned long) end - start);
On Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz measurements are: (for
find_next_bit, nbits is 8M, for find_first_bit - 80K)
find_next_bit: find_first_bit:
new current new current
26932 43151 14777 14925
26947 43182 14521 15423
26507 43824 15053 14705
27329 43759 14473 14777
26895 43367 14847 15023
26990 43693 15103 15163
26775 43299 15067 15232
27282 42752 14544 15121
27504 43088 14644 14858
26761 43856 14699 15193
26692 43075 14781 14681
27137 42969 14451 15061
... ...
find_next_bit performance gain is 35-40%;
find_first_bit - no measurable difference.
On ARM machine, there is arch-specific implementation for find_bit.
Thanks a lot to George Spelvin and Rasmus Villemoes for hints and
helpful discussions.
This patch (of 3):
New implementations takes less space in source file (see diffstat) and in
object. For me it's 710 vs 453 bytes of text. It also shows better
performance.
find_last_bit description fixed due to obvious typo.
[akpm@linux-foundation.org: include linux/bitmap.h, per Rasmus]
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Reviewed-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Reviewed-by: George Spelvin <linux@horizon.com>
Cc: Alexey Klimov <klimov.linux@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Mark Salter <msalter@redhat.com>
Cc: AKASHI Takahiro <takahiro.akashi@linaro.org>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Valentin Rothberg <valentinrothberg@gmail.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
