This fixes locking and bugs in cpu_down and cpu_up paths of the NUMA slab
allocator. Sonny Rao <sonny@burdell.org> reported problems sometime back on
POWER5 boxes, when the last cpu on the nodes were being offlined. We could
not reproduce the same on x86_64 because the cpumask (node_to_cpumask) was not
being updated on cpu down. Since that issue is now fixed, we can reproduce
Sonny's problems on x86_64 NUMA, and here is the fix.
The problem earlier was on CPU_DOWN, if it was the last cpu on the node to go
down, the array_caches (shared, alien) and the kmem_list3 of the node were
being freed (kfree) with the kmem_list3 lock held. If the l3 or the
array_caches were to come from the same cache being cleared, we hit on
badness.
This patch cleans up the locking in cpu_up and cpu_down path. We cannot
really free l3 on cpu down because, there is no node offlining yet and even
though a cpu is not yet up, node local memory can be allocated for it. So l3s
are usually allocated at keme_cache_create and destroyed at
kmem_cache_destroy. Hence, we don't need cachep->spinlock protection to get
to the cachep->nodelist[nodeid] either.
Patch survived onlining and offlining on a 4 core 2 node Tyan box with a 4
dbench process running all the time.
Signed-off-by: Alok N Kataria <alokk@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Earlier, we had to disable on chip interrupts while taking the
cachep->spinlock because, at cache_grow, on every addition of a slab to a slab
cache, we incremented colour_next which was protected by the cachep->spinlock,
and cache_grow could occur at interrupt context. Since, now we protect the
per-node colour_next with the node's list_lock, we do not need to disable on
chip interrupts while taking the per-cache spinlock, but we just need to
disable interrupts when taking the per-node kmem_list3 list_lock.
Signed-off-by: Alok N Kataria <alokk@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
colour_next is used as an index to add a colouring offset to a new slab in the
cache (colour_off * colour_next). Now with the NUMA aware slab allocator, it
makes sense to colour slabs added on the same node sequentially with
colour_next.
This patch moves the colouring index "colour_next" per-node by placing it on
kmem_list3 rather than kmem_cache.
This also helps simplify locking for CPU up and down paths.
Signed-off-by: Alok N Kataria <alokk@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I just spent some time researching a Bus Error. Turns out that the huge
page fault handler can return VM_FAULT_SIGBUS for various conditions where
no huge page is available.
Add a note explaining the reasoning in the source.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ben points out that:
When writing files out using O_SYNC, jbd's 1 jiffy delay results in a
significant drop in throughput as the disk sits idle. The patch below
results in a 4-5x performance improvement (from 6.5MB/s to ~24-30MB/s on my
IDE test box) when writing out files using O_SYNC.
So optimise the batching code by omitting it entirely if the process which is
doing a sync write is the same as the one which did the most recent sync
write. If that's true, we're unlikely to get any other processes joining the
transaction.
(Has been in -mm for ages - it took me a long time to get on to performance
testing it)
Numbers, on write-cache-disabled IDE:
/usr/bin/time -p synctest -n 10 -uf -t 1 -p 1 dir-name
Unpatched:
40 seconds
Patched:
35 seconds
Batching disabled:
35 seconds
This is the problematic single-process-doing-fsync case. With multiple
fsyncing processes the numbers are AFACIT unaltered by the patch.
Aside: performance testing and instrumentation shows that the transaction
batching almost doesn't help (testing with synctest -n 1 -uf -t 100 -p 10
dir-name on non-writeback-caching IDE). This is because by the time one
process is running a synchronous commit, a bunch of other processes already
have a transaction handle open, so they're all going to batch into the same
transaction anyway.
The batching seems to offer maybe 5-10% speedup with this workload, but I'm
pretty sure it was more important than that when it was first developed 4-odd
years ago...
Cc: "Stephen C. Tweedie" <sct@redhat.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
With CONFIG_REISERFS_FS_XATTR=y, CONFIG_REISERFS_FS_POSIX_ACL=n:
fs/reiserfs/xattr.c: In function `reiserfs_check_acl':
fs/reiserfs/xattr.c:1330: called object is not a function
Cc: Chris Mason <mason@suse.com>
Cc: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
dump_stack() on page allocation failure presently has an irritating habit
of shouting just "====" at everyone: please stop it.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make SELinux depend on SECURITY_NETWORK (which depends on SECURITY), as it
requires the socket hooks for proper operation even in the local case.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It may suck something awful, but it shouldn't taint the kernel.
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The pktcdvd driver uses a compile time macro constant to define the maximum
supported packet length. I changed this from 32 sectors to 128 sectors
because that allows over 100 MB of additional usable space on a 700 MB cdrw,
and increases throughput.
Note that you need a modified cdrwtool program that can format a CDRW disc
with larger packets to benefit from this change.
Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Allocate memory for read-gathering at open time, when it is known just how
much memory is needed. This avoids wasting kernel memory when the real packet
size is smaller than the maximum packet size supported by the driver. This is
always the case when using DVD discs.
Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Unless the help text is outdated, this seems to be logical.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The version information is not useful for a driver that is maintained in
Linus' kernel tree.
Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The pktcdvd driver was using an 8 bit field to store the packet length
obtained from the disc track info. This causes it to overflow packet length
values of 128KB or more. I changed the field to 32 bits to fix this.
The pktcdvd driver defaulted to its maximum allowed packet length when it
detected a 0 in the track info field. I changed this to fail the operation
and refuse to access the media. This seems more sane than attempting to
access it with a value that almost certainly will not work.
Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
migration_cost prints after every CPU hotplug event. Make it print only
once at boot.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The last fix for this function in fact opened up a much more often
triggering race.
It was uncommented tricky code, that was buggy. Add comment, make it less
tricky and fix bug.
Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix http://bugzilla.kernel.org/show_bug.cgi?id=5923
When a scsi command failed, an oops would result.
Back-to-back SMART queries would make the Seagate drives unhappy. The
second SMART query would timeout, and the command would be aborted.
Acked-by: Markus Lidel <Markus.Lidel@shadowconnect.com>
Cc: Kenny Simpson <theonetruekenny@yahoo.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
q->ordcolor must not be flipped on SOFTBARRIER.
Signed-off-by: Tejun Heo <htejun@gmail.com>
Acked-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Previously, if a fs request which was being drained failed and got
requeued, blk_do_ordered() didn't allow it to be reissued, which causes
queue stall. This patch makes blk_do_ordered() use the sequence of each
request to determine whether a request can be issued or not. This fixes
the bug and simplifies code.
Signed-off-by: Tejun Heo <htejun@gmail.com>
Acked-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
percpu_data blindly allocates bootmem memory to store NR_CPUS instances of
cpudata, instead of allocating memory only for possible cpus.
As a preparation for changing that, we need to convert various 0 -> NR_CPUS
loops to use for_each_cpu().
(The above only applies to users of asm-generic/percpu.h. powerpc has gone it
alone and is presently only allocating memory for present CPUs, so it's
currently corrupting memory).
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Jens Axboe <axboe@suse.de>
Cc: Anton Blanchard <anton@samba.org>
Acked-by: William Irwin <wli@holomorphy.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This reverts commit 10f4dc8b27.
Quoth Andi Kleen:
"Kiran decided that it makes the problem worse than it was before.
Fixing it fully requires more work which is too much for 2.6.16. So
please revert that commit for now."
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch contains a printk reorder to remove the current problem of
displaying "PCI-DMA: Disabling IOMMU." and then "PCI-DMA: using GART
IOMMU" 20 lines later in dmesg.
It also constains a printk reorder in swiotlb to state swiotlb
enablement prior to describing the location of the bounce buffers, and a
printk reorder to state gart enablement prior to describing the
aperature.
Also constains a whitespace cleanup in arch/x86_64/kernel/setup.c
Tested (along with patch 2/2) on dual opteron with gart enabled,
iommu=soft, and iommu=off.
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Hack for 2.6.16. In 2.6.17 all code that uses NR_CPUs should
be audited and changed to only touch possible CPUs.
Don't mark the reference per cpu data init data (so it stays
around after boot) and point all impossible CPUs to it. This way
they reference some valid - although shared memory. Usually
this is only initialization like INIT_LIST_HEADs and there
won't be races because these CPUs never run. Still somewhat hackish.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It's bad juju to touch the APIC when it hasn't been enabled.
I also moved ack_bad_irq for x86-64 out of line following i386.
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Some broken BIOS's had processors disabled, but
same apic id as a valid processor. This causes
acpi_processor_start() to think this disabled
cpu is ok, and croak. So we dont record bad
apicid's anymore.
http://bugzilla.kernel.org/show_bug.cgi?id=5930
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Checking of the validity of pointers should be consistently done before
dereferencing the pointer.
Signed-Off-By: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Conditionalize two unwind directives to match other similarly
conditional code.
Signed-Off-By: Jan Beulich <jbeulich@novell.com>
Cc: Jim Houston <jim.houston@ccur.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
On some broken motherboards (at least one NForce3 based AMD64 laptop)
the PIT timer runs at a incorrect frequency. This patch adds a new
option "apicpmtimer" that allows to use the APIC timer and calibrate it
using the PMTimer. It requires the earlier patch that allows to run the
main timer from the APIC.
Specifying apicpmtimer implies apicmaintimer.
The option defaults to off for now.
I tested it on a few systems and the resulting APIC timer frequencies
were usually a bit off, but always <1%, which should be tolerable.
TBD figure out heuristic to enable this automatically on the affected
systems TBD perhaps do it on all NForce3s or using DMI?
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
kprobes cannot deal with the funny calling conventions when it
runs on a different stack when it returns. If someone wants
to instrument context switch they can add a probe to schedule()
instead.
Cc: jkenisto@us.ibm.com, prasanna@in.ibm.com
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Align the start of the per-cpu section to the configured number of bytes in a
cache line. This stops a BUG_ON() from triggering in load_module() when
DEFINE_PER_CPU() is used in a module and the section isn't cacheline-aligned.
Rusty also found this and sent a patch in a while ago
(http://lkml.org/lkml/2004/10/19/17), I don't know what came of that.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[ AK: I redid Kevin's fix to be simpler, but the idea and original
analysis of the problem is from Kevin]
This avoid allocation failures on some SATA systems like Nvidia CK8
when the IOMMU gets fragmented. Modern SATA devices have quite large queues
(128 entries) and the FS with ext2/3 is good enough now that it often
passes whole 128 page sg lists down to the driver. These require
512K of continuous free space in the IOMMU aperture to map when merged.
When the IOMMU is fragmented this could lead to spurious IO errors
due to failing mappings.
Short term fix is to just try to map the SG list again unmerged
page by page - this way fragmentation doesn't matter anymore.
The code for that was already there, but it just wasn't enabled for the
merge case.
According to Kevin at least the Nvidia device doesn't seem to benefit
from merging much anyways, so the only slowdown is from trying
to do an unnecessary merge attempt.
Kevin plans to implement better fragmentation avoidance in the future,
but that wouldn't be 2.6.16 material.
TBD: should add some statistic counters to count how often that really
happens.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I broke this earlier when moving the patch from i386 to x86-64.
Need to return the virtual address here, not the physical address.
This fixes some boot time crashes on x86-64.
Cc: gregkh@suse.de
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- Check if the processor/memory affinity entries are long enough
according to the ACPI 3.0 spec.
- Ignore memory affinity entries that define a zero length region.
All based on BIOS issues found in the field @)
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
attached patch is 2 more cases i found via running the reference_init.pl
script. These were easy to spot just knowing the file names. There is
one another about init/main.c that i cant exactly zero in. (partly
because i dont know how to interpret the data thats spewed out of the tool).
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
> mm/mempolicy.c: In function `huge_zonelist':
> mm/mempolicy.c:1045: error: `HPAGE_SHIFT' undeclared (first use in this function)
> mm/mempolicy.c:1045: error: (Each undeclared identifier is reported only once
> mm/mempolicy.c:1045: error: for each function it appears in.)
> make[1]: *** [mm/mempolicy.o] Error 1
Need to wrap huge_zonelist function with CONFIG_HUGETLBFS.
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It has been enabled by default for some time now and is cheap enough
so it doesn't matter anyways.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently, x86_64 and ia64 arches do not clear the corresponding bits
in the node's cpumask when a cpu goes down or cpu bring up is cancelled.
This is buggy since there are pieces of common code where the cpumask is
checked in the cpu down code path to decide on things (like in the slab
down path). PPC does the right thing, but x86_64 and ia64 don't (This
was the reason Sonny hit upon a slab bug during cpu offline on ppc and
could not reproduce on other arches). This patch fixes it for x86_64.
I won't attempt ia64 as I cannot test it.
Credit for spotting this should go to Alok.
Signed-off-by: Alok N Kataria <alokk@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
They cause quite bad performance regressions on Netburst
This is temporary until we can get new optimized functions
for these CPUs.
This undoes changes that were done in 2.6.15 and in 2.6.16-rc1,
essentially bringing the code back to 2.6.14 level. Only change
is I renamed the X86_FEATURE_K8_C flag to X86_FEATURE_REP_GOOD
and fixed the check for the flag and also fixed some comments.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This avoids BUG_ONs in the low level allocator when an illegal
GFP mask is added.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
At resume time, TSC's value or something similar might be changed a lot
against suspend time. This could make system gets a very big lost ticks.
See http://bugzilla.kernel.org/show_bug.cgi?id=5825
Signed-off-by: Shaohua Li<shaohua.li@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
They all have problems with IRQ 0 routing, so just use the APIC on them.
Can be overwritten with "noapicmaintimer"
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Another piece from the no-idle-tick patch.
This can be enabled with the "apicmaintimer" option.
This is mainly useful when the PIT/HPET interrupt is unreliable.
Note there are some systems that are known to stop the APIC
timer in C3. For those it will never work, but this case
should be automatically detected.
It also only works with PM timer right now. When HPET is used
the way the main timer handler computes the delay doesn't work.
It should be a bit more efficient because there is one less
regular interrupt to process on the boot processor.
Requires earlier bugfix from Venkatesh
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Bug in apic timer removal on C3 patch. We should switch to IPI from APIC timer
only when C3 state is valid.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A kprobe executes IRET early and that could cause NMI recursion
and stack corruption.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
