Cody P Schafer [Thu, 23 May 2013 00:37:12 +0000 (10:37 +1000)]
mm/page_alloc: insert memory barriers to allow async update of pcp batch and high
Introduce pageset_update() to perform a safe transision from one set of
pcp->{batch,high} to a new set using memory barriers.
This ensures that batch is always set to a safe value (1) prior to
updating high, and ensure that high is fully updated before setting the
real value of batch. It avoids ->batch ever rising above ->high.
Cody P Schafer [Thu, 23 May 2013 00:37:11 +0000 (10:37 +1000)]
mm/page_alloc: prevent concurrent updaters of pcp ->batch and ->high
Because we are going to rely upon a careful transision between old and new
->high and ->batch values using memory barriers and will remove
stop_machine(), we need to prevent multiple updaters from interweaving
their memory writes.
Add a simple mutex to protect both update loops.
Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com> Cc: Gilad Ben-Yossef <gilad@benyossef.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cody P Schafer [Thu, 23 May 2013 00:37:11 +0000 (10:37 +1000)]
mm/page_alloc: factor out setting of pcp->high and pcp->batch
"Problems" with the current code:
1: there is a lack of synchronization in setting ->high and ->batch in
percpu_pagelist_fraction_sysctl_handler()
2: stop_machine() in zone_pcp_update() is unnecissary.
3: zone_pcp_update() does not consider the case where
percpu_pagelist_fraction is non-zero
To fix:
1: add memory barriers, a safe ->batch value, an update side mutex when
updating ->high and ->batch, and use ACCESS_ONCE() for ->batch users
that expect a stable value.
2: avoid draining pages in zone_pcp_update(), rely upon the memory
barriers added to fix #1
3: factor out quite a few functions, and then call the appropriate one.
Note that it results in a change to the behavior of zone_pcp_update(),
which is used by memory_hotplug. I'm rather certain that I've diserned
(and preserved) the essential behavior (changing ->high and ->batch), and
only eliminated unneeded actions (draining the per cpu pages), but this
may not be the case.
Further note that the draining of pages that previously took place in
zone_pcp_update() occured after repeated draining when attempting to
offline a page, and after the offline has "succeeded". It appears that
the draining was added to zone_pcp_update() to avoid refactoring
setup_pageset() into 2 funtions.
This patch:
Creates pageset_set_batch() for use in setup_pageset().
pageset_set_batch() imitates the functionality of
setup_pagelist_highmark(), but uses the boot time
(percpu_pagelist_fraction == 0) calculations for determining ->high based
on ->batch.
Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com> Cc: Gilad Ben-Yossef <gilad@benyossef.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Shaohua Li [Thu, 23 May 2013 00:37:10 +0000 (10:37 +1000)]
swap: add a simple detector for inappropriate swapin readahead
This is a patch to improve swap readahead algorithm. It's from Hugh and I
slightly changed it.
Hugh's original changelog:
swapin readahead does a blind readahead, whether or not the swapin
is sequential. This may be ok on harddisk, because large reads have
relatively small costs, and if the readahead pages are unneeded they
can be reclaimed easily - though, what if their allocation forced
reclaim of useful pages? But on SSD devices large reads are more
expensive than small ones: if the readahead pages are unneeded,
reading them in caused significant overhead.
This patch adds very simplistic random read detection. Stealing
the PageReadahead technique from Konstantin Khlebnikov's patch,
avoiding the vma/anon_vma sophistications of Shaohua Li's patch,
swapin_nr_pages() simply looks at readahead's current success
rate, and narrows or widens its readahead window accordingly.
There is little science to its heuristic: it's about as stupid
as can be whilst remaining effective.
The table below shows elapsed times (in centiseconds) when running
a single repetitive swapping load across a 1000MB mapping in 900MB
ram with 1GB swap (the harddisk tests had taken painfully too long
when I used mem=500M, but SSD shows similar results for that).
Vanilla is the 3.6-rc7 kernel on which I started; Shaohua denotes
his Sep 3 patch in mmotm and linux-next; HughOld denotes my Oct 1
patch which Shaohua showed to be defective; HughNew this Nov 14
patch, with page_cluster as usual at default of 3 (8-page reads);
HughPC4 this same patch with page_cluster 4 (16-page reads);
HughPC0 with page_cluster 0 (1-page reads: no readahead).
HDD for swapping to harddisk, SSD for swapping to VertexII SSD.
Seq for sequential access to the mapping, cycling five times around;
Rand for the same number of random touches. Anon for a MAP_PRIVATE
anon mapping; Shmem for a MAP_SHARED anon mapping, equivalent to tmpfs.
One weakness of Shaohua's vma/anon_vma approach was that it did
not optimize Shmem: seen below. Konstantin's approach was perhaps
mistuned, 50% slower on Seq: did not compete and is not shown below.
These tests are, of course, two extremes of a very simple case:
under heavier mixed loads I've not yet observed any consistent
improvement or degradation, and wider testing would be welcome.
Shaohua Li:
Test shows Vanilla is slightly better in sequential workload than Hugh's patch.
I observed with Hugh's patch sometimes the readahead size is shrinked too fast
(from 8 to 1 immediately) in sequential workload if there is no hit. And in
such case, continuing doing readahead is good actually.
I don't prepare a sophisticated algorithm for the sequential workload because
so far we can't guarantee sequential accessed pages are swap out sequentially.
So I slightly change Hugh's heuristic - don't shrink readahead size too fast.
Here is my test result (unit second, 3 runs average):
Vanilla Hugh New
Seq 356 370 360
Random 4525 2447 2444
Attached graph is the swapin/swapout throughput I collected with 'vmstat 2'.
The first part is running a random workload (till around 1200 of the x-axis)
and the second part is running a sequential workload. swapin and swapout
throughput are almost identical in steady state in both workloads. These are
expected behavior. while in Vanilla, swapin is much bigger than swapout
especially in random workload (because wrong readahead).
Original patches by: Shaohua Li and Konstantin Khlebnikov.
Signed-off-by: Hugh Dickins <hughd@google.com> Signed-off-by: Shaohua Li <shli@fusionio.com> Cc: Rik van Riel <riel@redhat.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Konstantin Khlebnikov <khlebnikov@openvz.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm/page_alloc.c: fix watermark check in __zone_watermark_ok()
The watermark check consists of two sub-checks. The first one is:
if (free_pages <= min + lowmem_reserve)
return false;
The check assures that there is minimal amount of RAM in the zone. If CMA
is used then the free_pages is reduced by the number of free pages in CMA
prior to the over-mentioned check.
if (!(alloc_flags & ALLOC_CMA))
free_pages -= zone_page_state(z, NR_FREE_CMA_PAGES);
This prevents the zone from being drained from pages available for
non-movable allocations.
The second check prevents the zone from getting too fragmented.
for (o = 0; o < order; o++) {
free_pages -= z->free_area[o].nr_free << o;
min >>= 1;
if (free_pages <= min)
return false;
}
The field z->free_area[o].nr_free is equal to the number of free pages
including free CMA pages. Therefore the CMA pages are subtracted twice.
This may cause a false positive fail of __zone_watermark_ok() if the CMA
area gets strongly fragmented. In such a case there are many 0-order free
pages located in CMA. Those pages are subtracted twice therefore they
will quickly drain free_pages during the check against fragmentation. The
test fails even though there are many free non-cma pages in the zone.
This patch fixes this issue by subtracting CMA pages only for a purpose of
(free_pages <= min + lowmem_reserve) check.
Signed-off-by: Tomasz Stanislawski <t.stanislaws@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Minchan Kim [Thu, 23 May 2013 00:37:09 +0000 (10:37 +1000)]
mm: remove compressed copy from zram in-memory
Swap subsystem does lazy swap slot free with expecting the page would be
swapped out again so we can avoid unnecessary write.
But the problem in in-memory swap(ex, zram) is that it consumes memory
space until vm_swap_full(ie, used half of all of swap device) condition
meet. It could be bad if we use multiple swap device, small in-memory
swap and big storage swap or in-memory swap alone.
This patch makes swap subsystem free swap slot as soon as swap-read is
completed and make the swapcache page dirty so the page should be written
out the swap device to reclaim it. It means we never lose it.
I tested this patch with kernel compile workload.
1. before
compile time : 9882.42
zram max wasted space by fragmentation: 13471881 byte
memory space consumed by zram: 174227456 byte
the number of slot free notify: 206684
2. after
compile time : 9653.90
zram max wasted space by fragmentation: 11805932 byte
memory space consumed by zram: 154001408 byte
the number of slot free notify: 426972
David Rientjes [Thu, 23 May 2013 00:37:08 +0000 (10:37 +1000)]
mm, memcg: don't take task_lock in task_in_mem_cgroup
For processes that have detached their mm's, task_in_mem_cgroup()
unnecessarily takes task_lock() when rcu_read_lock() is all that is
necessary to call mem_cgroup_from_task().
While we're here, switch task_in_mem_cgroup() to return bool.
Signed-off-by: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
mm/thp: use the correct function when updating access flags
We should use pmdp_set_access_flags to update access flags. Archs like
powerpc use extra checks(_PAGE_BUSY) when updating a hugepage PTE. A
set_pmd_at doesn't do those checks. We should use set_pmd_at only when
updating a none hugepage PTE.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Cc: Andrea Arcangeli <aarcange@redhat.com>a Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Pavel Emelyanov [Thu, 23 May 2013 00:37:07 +0000 (10:37 +1000)]
pagemap: prepare to reuse constant bits with page-shift
In order to reuse bits from pagemap entries gracefully, we leave the
entries as is but on pagemap open emit a warning in dmesg, that bits 55-60
are about to change in a couple of releases. Next, if a user issues
soft-dirty clear command via the clear_refs file (it was disabled before
v3.9) we assume that he's aware of the new pagemap format, note that fact
and report the bits in pagemap in the new manner.
The "migration strategy" looks like this then:
1. existing users are not affected -- they don't touch soft-dirty feature, thus
see old bits in pagemap, but are warned and have time to fix themselves
2. those who use soft-dirty know about new pagemap format
3. some time soon we get rid of any signs of page-shift in pagemap as well as
this trick with clear-soft-dirty affecting pagemap format.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com> Cc: Matt Mackall <mpm@selenic.com> Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Cc: Glauber Costa <glommer@parallels.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Pavel Emelyanov [Thu, 23 May 2013 00:37:07 +0000 (10:37 +1000)]
soft-dirty: call mmu notifiers when write-protecting ptes
As noticed by Xiao, since soft-dirty clear command modifies page tables we
have to flush tlbs and call mmu notifiers. While the former is done by
the clear_refs engine itself, the latter is to be done.
One thing to note about this -- in order not to call per-page invalidate
notifier (_all_ address space is about to be changed), the
_invalidate_range_start and _end are used. But for those start and end
are not known exactly. To address this, the same trick as in exit_mmap()
is used -- start is 0 and end is (unsigned long)-1.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com> Cc: Matt Mackall <mpm@selenic.com> Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Cc: Glauber Costa <glommer@parallels.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Pavel Emelyanov [Thu, 23 May 2013 00:37:07 +0000 (10:37 +1000)]
mm: soft-dirty bits for user memory changes tracking
The soft-dirty is a bit on a PTE which helps to track which pages a task
writes to. In order to do this tracking one should
1. Clear soft-dirty bits from PTEs ("echo 4 > /proc/PID/clear_refs)
2. Wait some time.
3. Read soft-dirty bits (55'th in /proc/PID/pagemap2 entries)
To do this tracking, the writable bit is cleared from PTEs when the
soft-dirty bit is. Thus, after this, when the task tries to modify a page
at some virtual address the #PF occurs and the kernel sets the soft-dirty
bit on the respective PTE.
Note, that although all the task's address space is marked as r/o after the
soft-dirty bits clear, the #PF-s that occur after that are processed fast.
This is so, since the pages are still mapped to physical memory, and thus
all the kernel does is finds this fact out and puts back writable, dirty
and soft-dirty bits on the PTE.
Another thing to note, is that when mremap moves PTEs they are marked with
soft-dirty as well, since from the user perspective mremap modifies the
virtual memory at mremap's new address.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com> Cc: Matt Mackall <mpm@selenic.com> Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Cc: Glauber Costa <glommer@parallels.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Pavel Emelyanov [Thu, 23 May 2013 00:37:06 +0000 (10:37 +1000)]
pagemap: introduce pagemap_entry_t without pmshift bits
These bits are always constant (== PAGE_SHIFT) and just occupy space in
the entry. Moreover, in next patch we will need to report one more bit in
the pagemap, but all bits are already busy on it.
That said, describe the pagemap entry that has 6 more free zero bits.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com> Cc: Matt Mackall <mpm@selenic.com> Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Cc: Glauber Costa <glommer@parallels.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This is the implementation of the soft-dirty bit concept that should help
keep track of changes in user memory, which in turn is very-very required
by the checkpoint-restore project (http://criu.org).
To create a dump of an application(s) we save all the information about it
to files, and the biggest part of such dump is the contents of tasks' memory.
However, there are usage scenarios where it's not required to get _all_ the
task memory while creating a dump. For example, when doing periodical dumps,
it's only required to take full memory dump only at the first step and then
take incremental changes of memory. Another example is live migration. We
copy all the memory to the destination node without stopping all tasks, then
stop them, check for what pages has changed, dump it and the rest of the state,
then copy it to the destination node. This decreases freeze time significantly.
That said, some help from kernel to watch how processes modify the contents
of their memory is required.
The proposal is to track changes with the help of new soft-dirty bit this way:
1. First do "echo 4 > /proc/$pid/clear_refs".
At that point kernel clears the soft dirty _and_ the writable bits from all
ptes of process $pid. From now on every write to any page will result in #pf
and the subsequent call to pte_mkdirty/pmd_mkdirty, which in turn will set
the soft dirty flag.
2. Then read the /proc/$pid/pagemap2 and check the soft-dirty bit reported there
(the 55'th one). If set, the respective pte was written to since last call
to clear refs.
The soft-dirty bit is the _PAGE_BIT_HIDDEN one. Although it's used by kmemcheck,
the latter one marks kernel pages with it, while the former bit is put on user
pages so they do not conflict to each other.
This patch:
A new clear-refs type will be added in the next patch, so prepare
code for that.
[akpm@linux-foundation.org: don't assume that sizeof(enum clear_refs_types) == sizeof(int)] Signed-off-by: Pavel Emelyanov <xemul@parallels.com> Cc: Matt Mackall <mpm@selenic.com> Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Cc: Glauber Costa <glommer@parallels.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Sasha Levin [Thu, 23 May 2013 00:37:05 +0000 (10:37 +1000)]
watchdog: trigger all-cpu backtrace when locked up and going to panic
Send an NMI to all CPUs when a lockup is detected and the lockup watchdog
code is configured to panic. This gives us a fairly uptodate snapshot of
all CPUs in the system.
This lets us get stack trace of all CPUs which makes life easier trying to
debug a deadlock, and the NMI doesn't change anything since the next step
is a kernel panic.
Takashi Iwai [Thu, 23 May 2013 00:37:05 +0000 (10:37 +1000)]
vfs: fix invalid ida_remove() call
When the group id of a shared mount is not allocated, the umount still
tries to call mnt_release_group_id(), which eventually hits a kernel
warning at ida_remove() spewing a message like:
ida_remove called for id=0 which is not allocated.
This patch fixes the bug simply checking the group id in the caller.
Signed-off-by: Takashi Iwai <tiwai@suse.de> Reported-by: Cristian Rodríguez <crrodriguez@opensuse.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Josh Hunt [Thu, 23 May 2013 00:37:04 +0000 (10:37 +1000)]
block: restore /proc/partitions to not display non-partitionable removable devices
We found with newer kernels we started seeing the cdrom device showing
up in /proc/partitions, but it was not there before.
Looking into this I found that commit d27769ec ("block: add
GENHD_FL_NO_PART_SCAN") introduces this change in behavior. It's not
clear to me from the commit's changelog if this change was intentional or
not. This comment still remains: /* Don't show non-partitionable
removeable devices or empty devices */ so I've decided to send a patch to
restore the behavior of not printing unpartitionable removable devices.
Signed-off-by: Josh Hunt <johunt@akamai.com> Cc: Tejun Heo <tj@kernel.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Cc: Jens Axboe <axboe@kernel.dk> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
lglock: update lockdep annotations to report recursive local locks
Oleg Nesterov recently noticed that the lockdep annotations in lglock.c
are not sufficient to detect some obvious deadlocks, such as
lg_local_lock(LOCK) + lg_local_lock(LOCK) or spin_lock(X) +
lg_local_lock(Y) vs lg_local_lock(Y) + spin_lock(X).
Both issues are easily fixed by indicating to lockdep that lglock's local
locks are not recursive. We shouldn't use the rwlock acquire/release
functions here, as lglock doesn't share the same semantics. Instead we
can base our lockdep annotations on the lock_acquire_shared (for local
lglock) and lock_acquire_exclusive (for global lglock) helpers.
I am not proposing new lglock specific helpers as I don't see the point of
the existing second level of helpers :)
Signed-off-by: Michel Lespinasse <walken@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@linux.intel.com> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
In lockdep.h, the spinlock/mutex/rwsem/rwlock/lock_map acquire macros have
different definitions based on the value of CONFIG_PROVE_LOCKING. We have
separate ifdefs for each of these definitions, which seems redundant.
Introduce lock_acquire_{exclusive,shared,shared_recursive} helpers which
will have different definitions based on CONFIG_PROVE_LOCKING. Then all
other helper macros can be defined based on the above ones, which reduces
the amount of ifdefined code.
Signed-off-by: Michel Lespinasse <walken@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@linux.intel.com> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Zhang Yanfei [Thu, 23 May 2013 00:37:01 +0000 (10:37 +1000)]
ipvs: change type of netns_ipvs->sysctl_sync_qlen_max
This member of struct netns_ipvs is calculated from nr_free_buffer_pages
so change its type to unsigned long in case of overflow. Also, type of
its related proc var sync_qlen_max and the return type of function
sysctl_sync_qlen_max() should be changed to unsigned long, too.
Besides, the type of ipvs_master_sync_state->sync_queue_len should be
changed to unsigned long accordingly.
Signed-off-by: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Cc: Simon Horman <horms@verge.net.au> Cc: Julian Anastasov <ja@ssi.bg> Cc: David Miller <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Dan Carpenter [Thu, 23 May 2013 00:37:01 +0000 (10:37 +1000)]
configfs: use capped length for ->store_attribute()
The difference between "count" and "len" is that "len" is capped at 4095.
Changing it like this makes it match how sysfs_write_file() is
implemented.
This is a static analysis patch. I haven't found any store_attribute()
functions where this change makes a difference.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Joel Becker <jlbec@evilplan.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Zhao Hongjiang [Thu, 23 May 2013 00:37:00 +0000 (10:37 +1000)]
drivers/infiniband/core/cm.c: convert to using idr_alloc_cyclic()
commit 3e6628c4b347 ("idr: introduce idr_alloc_cyclic()") adds a new
idr_alloc_cyclic routine and converts several of these users to it. This
is just a missed one - add it.
Signed-off-by: Zhao Hongjiang <zhaohongjiang@huawei.com> Cc: Roland Dreier <roland@kernel.org> Cc: Tejun Heo <tj@kernel.org> Cc: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
posix_timers: fix racy timer delta caching on task exit
When a task exits, we perform a caching of the remaining cputime delta
before expiring of its timers.
This is done from the following places:
* When the task is reaped. We iterate through its list of
posix cpu timers and store the remaining timer delta to
the timer struct instead of the absolute value.
(See posix_cpu_timers_exit() / posix_cpu_timers_exit_group() )
* When we call posix_cpu_timer_get() or posix_cpu_timer_schedule().
If the timer's task is considered dying when watched from these
places, the same conversion from absolute to relative expiry time
is performed. Then the given task's reference is released.
(See clear_dead_task() ).
The relevance of this caching is questionable but this is another
and deeper debate.
The big issue here is that these two sources of caching don't mix
up very well together.
More specifically, the caching can easily be done twice, resulting
in a wrong delta as it gets spuriously substracted a second time by
the elapsed clock. This can happen in the following scenario:
1) The task exits and gets reaped: we call posix_cpu_timers_exit()
and the absolute timer expiry values are converted to a relative
delta.
2) timer_gettime() -> posix_cpu_timer_get() is called and relies on
clear_dead_task() because tsk->exit_state == EXIT_DEAD.
The delta gets substracted again by the elapsed clock and we return
a wrong result.
To fix this, just remove the caching done on task reaping time. It
doesn't bring much value on its own. The caching done from
posix_cpu_timer_get/schedule is enough.
And it would also be hard to get it really right: we could make it put and
clear the target task in the timer struct so that readers know if they are
dealing with a relative cached of absolute value. But it would be racy.
The only safe way to do it would be to lock the itimer->it_lock so that we
know nobody reads the cputime expiry value while we modify it and its
target task reference. Doing so would involve some funny workarounds to
avoid circular lock against the sighand lock. There is just no reason to
maintain this.
The user visible effect of this patch can be observed by running the
following code: it creates a subthread that launches a posix cputimer
which expires after 10 seconds. But then the subthread only busy loops for 2
seconds and exits. The parent reaps the subthread and read the timer value.
Its expected value should the be the initial timer's expiration value
minus the cputime elapsed in the subthread. Roughly 10 - 2 = 8 seconds:
/* Arm 10 sec timer */
err = timer_settime(id, 0, &val, NULL);
if (err < 0) {
perror("Can't set timer\n");
return NULL;
}
/* Exit after 2 seconds of execution */
gettimeofday(&start, NULL);
do {
gettimeofday(&end, NULL);
timersub(&end, &start, &diff);
} while (diff.tv_sec < 2);
return NULL;
}
int main(int argc, char **argv)
{
pthread_t pthread;
int err;
err = pthread_create(&pthread, NULL, thread, NULL);
if (err) {
perror("Can't create thread\n");
return -1;
}
pthread_join(pthread, NULL);
/* Just wait a little bit to make sure the child got reaped */
sleep(1);
err = timer_gettime(id, &new);
if (err)
perror("Can't get timer value\n");
printf("%d %ld\n", new.it_value.tv_sec, new.it_value.tv_nsec);
posix-timers: correctly get dying task time sample in posix_cpu_timer_schedule()
In order to re-arm a timer after it fired, we take a sample of the current
process or thread cputime.
If the task is dying though, we don't arm anything but we cache the
remaining timer expiration delta for further reads.
Something similar is performed in posix_cpu_timer_get() but here we forget
to take the process wide cputime sample before caching it.
As a result we are storing random stack content, leading every further
reads of that timer to return junk values.
Fix this by taking the appropriate sample in the case of process wide
timers.
This probably doesn't matter much in practice because, at this stage, the
thread is the last one in the group and we reached exit_notify(). This
implies that we called exit_itimers() and there should be no more timers
to handle for that task.
So this is likely dead code anyway but let's fix the current logic
and the warning that came along:
kernel/posix-cpu-timers.c: In function 'posix_cpu_timer_schedule':
kernel/posix-cpu-timers.c:1127: warning: 'now' may be used uninitialized in this function
Then we can start to think further about cleaning up that code.
Reported-by: Andrew Morton <akpm@linux-foundation.org> Reported-by: Chen Gang <gang.chen@asianux.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Stanislaw Gruszka <sgruszka@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Ingo Molnar <mingo@elte.hu> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Chen Gang <gang.chen@asianux.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Add some initial basic tests on a few posix timers interface such as
setitimer() and timer_settime().
These simply check that expiration happens in a reasonable timeframe after
expected elapsed clock time (user time, user + system time, real time,
...).
This is helpful for finding basic breakages while hacking
on this subsystem.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Stanislaw Gruszka <sgruszka@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The posix cpu timer expiry time is stored in a union of two types: a 64
bits field if we rely on scheduler precise accounting, or a cputime_t if
we rely on jiffies.
This results in quite some duplicate code and special cases to handle the
two types.
Just unify this into a single 64 bits field. cputime_t can always fit
into it.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Stanislaw Gruszka <sgruszka@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Ingo Molnar <mingo@elte.hu> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Ondrej Zary [Thu, 23 May 2013 00:36:58 +0000 (10:36 +1000)]
cyber2000fb: avoid palette corruption at higher clocks
When 1280x1024@75Hz mode is set, console palette is not set properly -
sometimes the background is white, sometimes yellow and text colors are
also messed up. This does not happen at 1280x1024@60Hz and below.
It seems that the HW needs some time before setting the palette - maybe
the PLL needs more time to lock at higher speeds. This patch fixes the
problem but without knowing what register to check for PLL lock(?), the
delay might be excessive.
On Fri, 28 Jan 2011 18:15:37 +0000
Russell King <rmk@arm.linux.org.uk> wrote:
> On Tue, Jan 18, 2011 at 01:14:24PM -0800, Andrew Morton wrote:
> > Russell, I have an (old) note here that this is awaiting an ack from
> > yourself?
>
> Well, I can reproduce this problem on the Netwinders here. I'm not sure
> that we should delay all mode switches by one second - and any attempt
> to reduce this value does result in the palette not being set correctly.
>
> For 1280x1024-75, the dotclock is 135MHz, which gives a PLL values of
> 0x41 and 0x06. That's: M=0x41+1, N=0x06+1, P=0x00 (top 2 bits of 0x06)
> -> Q=1
>
> Fpll = 14.31818MHz * M / N
> Fout = Fpll / Q
>
> The PLL itself is formed by dividing the 14-ish MHz frequency by N and
> phase comparing the output of the VCO, divided by M, and adjusting the
> VCO until the two correlate. As VCOs typically tend to have a limited
> range, it's normal to divide the output frequency to produce a greater
> range - and in this case that's done by Q.
>
> For the 800x600-100 copied from /etc/fb.modes, this has a dotclock of
> 67.5MHz, which is exactly half this rate. The PLL values for this are:
> M=0x41+1, N=0x06+1, P=0x01, giving PLL values of 0x41 and 0x46.
>
> Booting with 800x600-100 does not suffer the problem. So it's not
> related to PLL lock time. There's something else going on.
>
> Another experiment I tried was forcing the PLL values to produce 108MHz
> instead of 135MHz. 108MHz is the dotclock for 1280x1024-60. This too
> doesn't suffer the problem.
>
> I've also tried chosing other delay values. 100ms is too short and
> produces the problem, but 1s works. 1s for a PLL to lock is a hell of
> a time, especially for a PLL operating in the MHz range.
>
> I've tried setting the PLL to a known good freqency, and then switching
> to 135MHz - the problem persists. It's not like 135MHz is reaching the
> limits - it'll go up to 206MHz.
>
> So, I don't think this has anything to do with PLL locking. I think
> there's something else going on which isn't immediately obvious - maybe
> bandwidth starvation preventing us from writing properly to the palette?
> As it's a horrible VGA, where you write the same register multiple times
> I wouldn't be surprised if some writes were going missing.
>
> I'll see if I can play around with it some more this evening, but I've
> spent an awful long time on just this issue already this afternoon...
>
> I think further investigation needs to happen on this patch before it's
> acceptable. Or maybe we should prevent the cyberpro coming up in
Signed-off-by: Ondrej Zary <linux@rainbow-software.org> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Daniel Vetter [Thu, 23 May 2013 00:36:56 +0000 (10:36 +1000)]
drm/fb-helper: don't sleep for screen unblank when an oops is in progress
Otherwise the system will burn even brighter and worse, leave the user
wondering what's going on exactly.
Since we already have a panic handler which will (try) to restore the
entire fbdev console mode, we can just bail out. Inspired by a patch from
Konstantin Khlebnikov. The callchain leading to this, cut&pasted from
Konstantin's original patch:
Note that the entire locking in the fb helper around panic/sysrq and kdbg
is ... non-existant. So we have a decent change of blowing up
everything. But since reworking this ties in with funny concepts like the
fbdev notifier chain or the impressive things which happen around
console_lock while oopsing, I'll leave that as an exercise for braver
souls than me.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Konstantin Khlebnikov <khlebnikov@openvz.org> Cc: Dave Airlie <airlied@gmail.com> Reviewed-by: Rob Clark <robdclark@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Chen Gang [Thu, 23 May 2013 00:36:55 +0000 (10:36 +1000)]
kernel/auditfilter.c: fix leak in audit_add_rule() error path
If both 'tree' and 'watch' are valid we must call audit_put_tree(), just
like the preceding code within audit_add_rule().
Signed-off-by: Chen Gang <gang.chen@asianux.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Eric Paris <eparis@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Both of these look wrong to me. As Steve Grubb pointed out:
"What we need is 1 PATH record that identifies the MQ. The other PATH
records probably should not be there."
Fix it to record the mq root as a parent, and flag it such that it should
be hidden from view when the names are logged, since the root of the mq
filesystem isn't terribly interesting. With this change, we get a single
PATH record that looks more like this:
In order to do this, a new audit_inode_parent_hidden() function is added.
If we do it this way, then we avoid having the existing callers of
audit_inode needing to do any sort of flag conversion if auditing is
inactive.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Reported-by: Jiri Jaburek <jjaburek@redhat.com> Cc: Steve Grubb <sgrubb@redhat.com> Cc: Eric Paris <eparis@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Wen Congyang [Thu, 23 May 2013 00:36:55 +0000 (10:36 +1000)]
x86: make 'mem=' option to work for efi platform
Current mem boot option only can work for non efi environment. If the
user specifies add_efi_memmap, it cannot work for efi environment. In the
efi environment, we call e820_add_region() to add the memory map. So we
can modify __e820_add_region() and the mem boot option can work for efi
environment.
Note: Only E820_RAM is limited, and BOOT_SERVICES_{CODE,DATA} are always
mapped(If its address >= mem_limit, the memory won't be freed in
efi_free_boot_services()).
Signed-off-by: Wen Congyang <wency@cn.fujitsu.com> Cc: Matt Fleming <matt.fleming@intel.com> Cc: Rob Landley <rob@landley.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Yasuaki ISIMATU <isimatu.yasuaki@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Matthew Garrett <mjg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Chen Gang [Thu, 23 May 2013 00:36:53 +0000 (10:36 +1000)]
kernel/audit_tree.c:audit_add_tree_rule(): protect `rule' from kill_rules()
audit_add_tree_rule() must set 'rule->tree = NULL;' firstly, to protect
the rule itself freed in kill_rules().
The reason is when it is killed, the 'rule' itself may have already
released, we should not access it. one example: we add a rule to an
inode, just at the same time the other task is deleting this inode.
The work flow for adding a rule:
audit_receive() -> (need audit_cmd_mutex lock)
audit_receive_skb() ->
audit_receive_msg() ->
audit_receive_filter() ->
audit_add_rule() ->
audit_add_tree_rule() -> (need audit_filter_mutex lock)
...
unlock audit_filter_mutex
get_tree()
...
iterate_mounts() -> (iterate all related inodes)
tag_mount() ->
tag_trunk() ->
create_trunk() -> (assume it is 1st rule)
fsnotify_add_mark() ->
fsnotify_add_inode_mark() -> (add mark to inode->i_fsnotify_marks)
...
get_tree(); (each inode will get one)
...
lock audit_filter_mutex
The work flow for deleting an inode:
__destroy_inode() ->
fsnotify_inode_delete() ->
__fsnotify_inode_delete() ->
fsnotify_clear_marks_by_inode() -> (get mark from inode->i_fsnotify_marks)
fsnotify_destroy_mark() ->
fsnotify_destroy_mark_locked() ->
audit_tree_freeing_mark() ->
evict_chunk() ->
...
tree->goner = 1
...
kill_rules() -> (assume current->audit_context == NULL)
call_rcu() -> (rule->tree != NULL)
audit_free_rule_rcu() ->
audit_free_rule()
...
audit_schedule_prune() -> (assume current->audit_context == NULL)
kthread_run() -> (need audit_cmd_mutex and audit_filter_mutex lock)
prune_one() -> (delete it from prue_list)
put_tree(); (match the original get_tree above)
Signed-off-by: Chen Gang <gang.chen@asianux.com> Cc: Eric Paris <eparis@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Kees Cook [Thu, 23 May 2013 00:36:53 +0000 (10:36 +1000)]
kmsg: honor dmesg_restrict sysctl on /dev/kmsg
The dmesg_restrict sysctl currently covers the syslog method for access
dmesg, however /dev/kmsg isn't covered by the same protections. Most
people haven't noticed because util-linux dmesg(1) defaults to using the
syslog method for access in older versions. With util-linux dmesg(1)
defaults to reading directly from /dev/kmsg.
To fix /dev/kmsg, let's compare the existing interfaces and what they allow:
- /proc/kmsg allows:
- open (SYSLOG_ACTION_OPEN) if CAP_SYSLOG since it uses a destructive
single-reader interface (SYSLOG_ACTION_READ).
- everything, after an open.
- syslog syscall allows:
- anything, if CAP_SYSLOG.
- SYSLOG_ACTION_READ_ALL and SYSLOG_ACTION_SIZE_BUFFER, if dmesg_restrict==0.
- nothing else (EPERM).
The use-cases were:
- dmesg(1) needs to do non-destructive SYSLOG_ACTION_READ_ALLs.
- sysklog(1) needs to open /proc/kmsg, drop privs, and still issue the
destructive SYSLOG_ACTION_READs.
AIUI, dmesg(1) is moving to /dev/kmsg, and systemd-journald doesn't clear
the ring buffer.
Based on the comments in devkmsg_llseek, it sounds like actions besides
reading aren't going to be supported by /dev/kmsg (i.e.
SYSLOG_ACTION_CLEAR), so we have a strict subset of the non-destructive
syslog syscall actions.
To this end, move the check as Josh had done, but also rename the
constants to reflect their new uses (SYSLOG_FROM_CALL becomes
SYSLOG_FROM_READER, and SYSLOG_FROM_FILE becomes SYSLOG_FROM_PROC).
SYSLOG_FROM_READER allows non-destructive actions, and SYSLOG_FROM_PROC
allows destructive actions after a capabilities-constrained
SYSLOG_ACTION_OPEN check.
- /dev/kmsg allows:
- open if CAP_SYSLOG or dmesg_restrict==0
- reading/polling, after open
Robin Holt [Thu, 23 May 2013 00:36:52 +0000 (10:36 +1000)]
reboot: rigrate shutdown/reboot to boot cpu
We recently noticed that reboot of a 1024 cpu machine takes approx 16
minutes of just stopping the cpus. The slowdown was tracked to commit f96972f.
The current implementation does all the work of hot removing the cpus
before halting the system. We are switching to just migrating to the
boot cpu and then continuing with shutdown/reboot.
This also has the effect of not breaking x86's command line parameter for
specifying the reboot cpu. Note, this code was shamelessly copied from
arch/x86/kernel/reboot.c with bits removed pertaining to the reboot_cpu
command line parameter.
Signed-off-by: Robin Holt <holt@sgi.com> Tested-by: Shawn Guo <shawn.guo@linaro.org> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: Russ Anderson <rja@sgi.com> Cc: Robin Holt <holt@sgi.com> Cc: Russell King <linux@arm.linux.org.uk> Cc: Guan Xuetao <gxt@mprc.pku.edu.cn> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Robin Holt <holt@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Srivatsa S. Bhat [Thu, 23 May 2013 00:36:51 +0000 (10:36 +1000)]
CPU hotplug: provide a generic helper to disable/enable CPU hotplug
There are instances in the kernel where we would like to disable CPU
hotplug (from sysfs) during some important operation. Today the freezer
code depends on this and the code to do it was kinda tailor-made for that.
Restructure the code and make it generic enough to be useful for
other usecases too.
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Robin Holt <holt@sgi.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: Russ Anderson <rja@sgi.com> Cc: Robin Holt <holt@sgi.com> Cc: Russell King <linux@arm.linux.org.uk> Cc: Guan Xuetao <gxt@mprc.pku.edu.cn> Cc: Shawn Guo <shawn.guo@linaro.org> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Dave Jones <davej@redhat.com> Cc: Davidlohr Bueso <davidlohr.bueso@hp.com> Cc: Li Zefan <lizefan@huawei.com> Cc: Li Zefan <lizfan@huawei.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
projects / karo-tx-linux.git / log