page allocation failure message kernel 2.6.31.4 (tty-related)
Subsystems (2)
Hello all,
is this something to worry about, I saw several of those:
Oct 20 11:25:41 box kernel: ssh: page allocation failure. order:1, mode:0x20 Oct 20 11:25:41 box kernel: Pid: 2970, comm: ssh Tainted: G M 2.6.31.4 #1 Oct 20 11:25:41 box kernel: Call Trace: Oct 20 11:25:41 box kernel: [<c105545c>] ? __alloc_pages_nodemask+0x404/0x446 Oct 20 11:25:41 box kernel: [<c106e1d1>] ? cache_alloc_refill+0x249/0x42d Oct 20 11:25:41 box kernel: [<c106e41c>] ? __kmalloc+0x67/0x9e Oct 20 11:25:41 box kernel: [<c1147c5a>] ? tty_buffer_request_room+0xa9/0x112 Oct 20 11:25:42 box kernel: [<c1147deb>] ? tty_insert_flip_string+0x24/0x90 Oct 20 11:25:42 box kernel: [<c114869e>] ? pty_write+0x25/0x3f Oct 20 11:25:42 box kernel: [<c1144b6f>] ? n_tty_write+0x23d/0x2d5 Oct 20 11:25:42 box kernel: [<c1021cc7>] ? default_wake_function+0x0/0x8 Oct 20 11:25:42 box kernel: [<c1142ab7>] ? tty_write+0x144/0x1b3 Oct 20 11:25:42 box kernel: [<c1144932>] ? n_tty_write+0x0/0x2d5 Oct 20 11:25:42 box kernel: [<c1142973>] ? tty_write+0x0/0x1b3 Oct 20 11:25:42 box kernel: [<c107238b>] ? vfs_write+0x84/0x105 Oct 20 11:25:42 box kernel: [<c10724a4>] ? sys_write+0x3c/0x63 Oct 20 11:25:42 box kernel: [<c1002804>] ? sysenter_do_call+0x12/0x22 Oct 20 11:25:44 box kernel: DMA per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 0, btch: 1 usd: 0 Oct 20 11:25:44 box kernel: CPU 1: hi: 0, btch: 1 usd: 0 Oct 20 11:25:44 box kernel: Normal per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 186, btch: 31 usd: 193 Oct 20 11:25:44 box kernel: CPU 1: hi: 186, btch: 31 usd: 46 Oct 20 11:25:44 box kernel: HighMem per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 186, btch: 31 usd: 157 Oct 20 11:25:44 box kernel: CPU 1: hi: 186, btch: 31 usd: 142 Oct 20 11:25:44 box kernel: Active_anon:89762 active_file:96664 inactive_anon:23257 Oct 20 11:25:44 box kernel: inactive_file:119147 unevictable:0 dirty:53 writeback:0 unstable:0 Oct 20 11:25:45 box kernel: free:569624 slab:38452 mapped:14421 pagetables:977 bounce:0 Oct 20 11:25:45 box kernel: DMA free:3488kB min:64kB low:80kB high:96kB active_anon:0kB inactive_anon:0kB active_file:8980kB inactive_file:3024kB unevictable:0kB present:15804kB pages_scanned:0 all_unreclaima Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 865 3698 3698 Oct 20 11:25:45 box kernel: Normal free:5468kB min:3728kB low:4660kB high:5592kB active_anon:0kB inactive_anon:0kB active_file:324840kB inactive_file:330196kB unevictable:0kB present:885944kB pages_scanned:0 Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 0 22669 22669 Oct 20 11:25:45 box kernel: HighMem free:2269540kB min:512kB low:3564kB high:6616kB active_anon:359048kB inactive_anon:93028kB active_file:52836kB inactive_file:143368kB unevictable:0kB present:2901640kB page Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 0 0 0 Oct 20 11:25:45 box kernel: DMA: 172*4kB 6*8kB 0*16kB 0*32kB 3*64kB 2*128kB 3*256kB 3*512kB 0*1024kB 0*2048kB 0*4096kB = 3488kB Oct 20 11:25:45 box kernel: Normal: 1232*4kB 83*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 5592kB Oct 20 11:25:45 box kernel: HighMem: 45*4kB 80*8kB 69*16kB 29*32kB 29*64kB 22*128kB 14*256kB 9*512kB 3*1024kB 3*2048kB 548*4096kB = 2269540kB Oct 20 11:25:45 box kernel: 216788 total pagecache pages Oct 20 11:25:45 box kernel: 0 pages in swap cache Oct 20 11:25:45 box kernel: Swap cache stats: add 0, delete 0, find 0/0 Oct 20 11:25:45 box kernel: Free swap = 2104472kB Oct 20 11:25:45 box kernel: Total swap = 2104472kB
Signature · 2 lines
Regards, Stephan
[ tty related CCs added ]
Quoted 50 lines · Stephan von Krawczynski
Hello all,
is this something to worry about, I saw several of those:
Oct 20 11:25:41 box kernel: ssh: page allocation failure. order:1, mode:0x20 Oct 20 11:25:41 box kernel: Pid: 2970, comm: ssh Tainted: G M 2.6.31.4 #1 Oct 20 11:25:41 box kernel: Call Trace: Oct 20 11:25:41 box kernel: [<c105545c>] ? __alloc_pages_nodemask+0x404/0x446 Oct 20 11:25:41 box kernel: [<c106e1d1>] ? cache_alloc_refill+0x249/0x42d Oct 20 11:25:41 box kernel: [<c106e41c>] ? __kmalloc+0x67/0x9e Oct 20 11:25:41 box kernel: [<c1147c5a>] ? tty_buffer_request_room+0xa9/0x112 Oct 20 11:25:42 box kernel: [<c1147deb>] ? tty_insert_flip_string+0x24/0x90 Oct 20 11:25:42 box kernel: [<c114869e>] ? pty_write+0x25/0x3f Oct 20 11:25:42 box kernel: [<c1144b6f>] ? n_tty_write+0x23d/0x2d5 Oct 20 11:25:42 box kernel: [<c1021cc7>] ? default_wake_function+0x0/0x8 Oct 20 11:25:42 box kernel: [<c1142ab7>] ? tty_write+0x144/0x1b3 Oct 20 11:25:42 box kernel: [<c1144932>] ? n_tty_write+0x0/0x2d5 Oct 20 11:25:42 box kernel: [<c1142973>] ? tty_write+0x0/0x1b3 Oct 20 11:25:42 box kernel: [<c107238b>] ? vfs_write+0x84/0x105 Oct 20 11:25:42 box kernel: [<c10724a4>] ? sys_write+0x3c/0x63 Oct 20 11:25:42 box kernel: [<c1002804>] ? sysenter_do_call+0x12/0x22 Oct 20 11:25:44 box kernel: DMA per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 0, btch: 1 usd: 0 Oct 20 11:25:44 box kernel: CPU 1: hi: 0, btch: 1 usd: 0 Oct 20 11:25:44 box kernel: Normal per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 186, btch: 31 usd: 193 Oct 20 11:25:44 box kernel: CPU 1: hi: 186, btch: 31 usd: 46 Oct 20 11:25:44 box kernel: HighMem per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 186, btch: 31 usd: 157 Oct 20 11:25:44 box kernel: CPU 1: hi: 186, btch: 31 usd: 142 Oct 20 11:25:44 box kernel: Active_anon:89762 active_file:96664 inactive_anon:23257 Oct 20 11:25:44 box kernel: inactive_file:119147 unevictable:0 dirty:53 writeback:0 unstable:0 Oct 20 11:25:45 box kernel: free:569624 slab:38452 mapped:14421 pagetables:977 bounce:0 Oct 20 11:25:45 box kernel: DMA free:3488kB min:64kB low:80kB high:96kB active_anon:0kB inactive_anon:0kB active_file:8980kB inactive_file:3024kB unevictable:0kB present:15804kB pages_scanned:0 all_unreclaima Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 865 3698 3698 Oct 20 11:25:45 box kernel: Normal free:5468kB min:3728kB low:4660kB high:5592kB active_anon:0kB inactive_anon:0kB active_file:324840kB inactive_file:330196kB unevictable:0kB present:885944kB pages_scanned:0 Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 0 22669 22669 Oct 20 11:25:45 box kernel: HighMem free:2269540kB min:512kB low:3564kB high:6616kB active_anon:359048kB inactive_anon:93028kB active_file:52836kB inactive_file:143368kB unevictable:0kB present:2901640kB page Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 0 0 0 Oct 20 11:25:45 box kernel: DMA: 172*4kB 6*8kB 0*16kB 0*32kB 3*64kB 2*128kB 3*256kB 3*512kB 0*1024kB 0*2048kB 0*4096kB = 3488kB Oct 20 11:25:45 box kernel: Normal: 1232*4kB 83*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 5592kB Oct 20 11:25:45 box kernel: HighMem: 45*4kB 80*8kB 69*16kB 29*32kB 29*64kB 22*128kB 14*256kB 9*512kB 3*1024kB 3*2048kB 548*4096kB = 2269540kB Oct 20 11:25:45 box kernel: 216788 total pagecache pages Oct 20 11:25:45 box kernel: 0 pages in swap cache Oct 20 11:25:45 box kernel: Swap cache stats: add 0, delete 0, find 0/0 Oct 20 11:25:45 box kernel: Free swap = 2104472kB Oct 20 11:25:45 box kernel: Total swap = 2104472kB
Signature · 2 lines
Jiri Kosina SUSE Labs, Novell Inc.
Quoted 2 lines · Jiri Kosina
[ tty related CCs added ]
I think this is probably a duplicate of the GFP_ATOMIC page allocator issue at http://bugzilla.kernel.org/show_bug.cgi?id=14141 more than a tty issue.
Have we tried rolling back the refactored page allocator changes in 2.6.31 to see if the logic has changed in other ways unrelated to kswapd? There's subtleties like where rt tasks are given ALLOC_HARDER even when in_interrupt() that are different from previous kernels and could reduce the amount of ALLOC_HIGH and ALLOC_HARDER memory that can be allocated.
Quoted 49 lines · Stephan von Krawczynski
Quoted 48 lines
Hello all,
is this something to worry about, I saw several of those:
Oct 20 11:25:41 box kernel: ssh: page allocation failure. order:1, mode:0x20 Oct 20 11:25:41 box kernel: Pid: 2970, comm: ssh Tainted: G M 2.6.31.4 #1 Oct 20 11:25:41 box kernel: Call Trace: Oct 20 11:25:41 box kernel: [<c105545c>] ? __alloc_pages_nodemask+0x404/0x446 Oct 20 11:25:41 box kernel: [<c106e1d1>] ? cache_alloc_refill+0x249/0x42d Oct 20 11:25:41 box kernel: [<c106e41c>] ? __kmalloc+0x67/0x9e Oct 20 11:25:41 box kernel: [<c1147c5a>] ? tty_buffer_request_room+0xa9/0x112 Oct 20 11:25:42 box kernel: [<c1147deb>] ? tty_insert_flip_string+0x24/0x90 Oct 20 11:25:42 box kernel: [<c114869e>] ? pty_write+0x25/0x3f Oct 20 11:25:42 box kernel: [<c1144b6f>] ? n_tty_write+0x23d/0x2d5 Oct 20 11:25:42 box kernel: [<c1021cc7>] ? default_wake_function+0x0/0x8 Oct 20 11:25:42 box kernel: [<c1142ab7>] ? tty_write+0x144/0x1b3 Oct 20 11:25:42 box kernel: [<c1144932>] ? n_tty_write+0x0/0x2d5 Oct 20 11:25:42 box kernel: [<c1142973>] ? tty_write+0x0/0x1b3 Oct 20 11:25:42 box kernel: [<c107238b>] ? vfs_write+0x84/0x105 Oct 20 11:25:42 box kernel: [<c10724a4>] ? sys_write+0x3c/0x63 Oct 20 11:25:42 box kernel: [<c1002804>] ? sysenter_do_call+0x12/0x22 Oct 20 11:25:44 box kernel: DMA per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 0, btch: 1 usd: 0 Oct 20 11:25:44 box kernel: CPU 1: hi: 0, btch: 1 usd: 0 Oct 20 11:25:44 box kernel: Normal per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 186, btch: 31 usd: 193 Oct 20 11:25:44 box kernel: CPU 1: hi: 186, btch: 31 usd: 46 Oct 20 11:25:44 box kernel: HighMem per-cpu: Oct 20 11:25:44 box kernel: CPU 0: hi: 186, btch: 31 usd: 157 Oct 20 11:25:44 box kernel: CPU 1: hi: 186, btch: 31 usd: 142 Oct 20 11:25:44 box kernel: Active_anon:89762 active_file:96664 inactive_anon:23257 Oct 20 11:25:44 box kernel: inactive_file:119147 unevictable:0 dirty:53 writeback:0 unstable:0 Oct 20 11:25:45 box kernel: free:569624 slab:38452 mapped:14421 pagetables:977 bounce:0 Oct 20 11:25:45 box kernel: DMA free:3488kB min:64kB low:80kB high:96kB active_anon:0kB inactive_anon:0kB active_file:8980kB inactive_file:3024kB unevictable:0kB present:15804kB pages_scanned:0 all_unreclaima Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 865 3698 3698 Oct 20 11:25:45 box kernel: Normal free:5468kB min:3728kB low:4660kB high:5592kB active_anon:0kB inactive_anon:0kB active_file:324840kB inactive_file:330196kB unevictable:0kB present:885944kB pages_scanned:0 Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 0 22669 22669 Oct 20 11:25:45 box kernel: HighMem free:2269540kB min:512kB low:3564kB high:6616kB active_anon:359048kB inactive_anon:93028kB active_file:52836kB inactive_file:143368kB unevictable:0kB present:2901640kB page Oct 20 11:25:45 box kernel: lowmem_reserve[]: 0 0 0 0 Oct 20 11:25:45 box kernel: DMA: 172*4kB 6*8kB 0*16kB 0*32kB 3*64kB 2*128kB 3*256kB 3*512kB 0*1024kB 0*2048kB 0*4096kB = 3488kB Oct 20 11:25:45 box kernel: Normal: 1232*4kB 83*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 5592kB Oct 20 11:25:45 box kernel: HighMem: 45*4kB 80*8kB 69*16kB 29*32kB 29*64kB 22*128kB 14*256kB 9*512kB 3*1024kB 3*2048kB 548*4096kB = 2269540kB Oct 20 11:25:45 box kernel: 216788 total pagecache pages Oct 20 11:25:45 box kernel: 0 pages in swap cache Oct 20 11:25:45 box kernel: Swap cache stats: add 0, delete 0, find 0/0 Oct 20 11:25:45 box kernel: Free swap = 2104472kB Oct 20 11:25:45 box kernel: Total swap = 2104472kB
Quoted 9 lines · David Rientjes
Quoted 2 lines · Jiri Kosina
[ tty related CCs added ]
I think this is probably a duplicate of the
GFP_ATOMICpage allocator issue at http://bugzilla.kernel.org/show_bug.cgi?id=14141 more than a tty issue.
Probably.
Quoted 2 lines
Have we tried rolling back the refactored page allocator changes in 2.6.31 to see if the logic has changed in other ways unrelated to kswapd?
No, but bisects around that area were inconclusive at best and there are dependants that made backing it out problematic.
Quoted 4 lines
There's subtleties like where rt tasks are given
ALLOC_HARDEReven whenin_interrupt()that are different from previous kernels and could reduce the amount ofALLOC_HIGHandALLOC_HARDERmemory that can be allocated.
/me checks again
I think you're right. Correct it with something like?
==== CUT HERE ====
Quoted 1 line
From
bca71e94e10cd93771ec5b17eccb817dd0c85360Mon Sep 17 00:00:00 2001
From: Mel Gorman <[email protected]>
Date: Thu, 22 Oct 2009 11:55:14 +0100
Subject: [PATCH] page allocator: Do not allow interrupts to use ALLOC_HARDER
Commit 341ce06f69abfafa31b9468410a13dbd60e2b237 altered watermark logic slightly by allowing rt_tasks that are handling an interrupt to set ALLOC_HARDER. This patch brings the watermark logic more in line with 2.6.30.
---
mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
mm/page_alloc.c +1 -1
@@ -1769,7 +1769,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)* See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ alloc_flags &= ~ALLOC_CPUSET ; - }else if (unlikely (rt_task (p))) + }else if (unlikely (rt_task (p)) && !in_interrupt ()) alloc_flags |=ALLOC_HARDER ;if (likely (!(gfp_mask & __GFP_NOMEMALLOC))) {
Pekka Enberg
Reviewed-byQuoted 15 lines · Mel Gorman
Quoted 4 lines
There's subtleties like where rt tasks are given
ALLOC_HARDEReven whenin_interrupt()that are different from previous kernels and could reduce the amount ofALLOC_HIGHandALLOC_HARDERmemory that can be allocated./me checks again
I think you're right. Correct it with something like?
==== CUT HERE ====
Frombca71e94e10cd93771ec5b17eccb817dd0c85360Mon Sep 17 00:00:00 2001
From: Mel Gorman <[email protected]>
Date: Thu, 22 Oct 2009 11:55:14 +0100
Subject: [PATCH] page allocator: Do not allow interrupts to useALLOC_HARDER
Quoted 6 lines
Commit
341ce06f69abfafa31b9468410a13dbd60e2b237altered watermark logic slightly by allowingrt_tasksthat are handling an interrupt to setALLOC_HARDER. This patch brings the watermark logic more in line with 2.6.30.Signed-off-by Mel Gorman <[email protected]>
Good catch.
Can we have some of the people that are hitting the OOM problems test this patch in combination with this patch:
http://patchwork.kernel.org/patch/54215/
Mel, can you also resend these page allocator patches to Andrew? I haven't seen him pick the above patch in -mm and we probably want to get fixes into -stable soon because distributors seem to be basing their next releases on 2.6.31...
Quoted 23 lines
---
mm/page_alloc.c | 2 +-1 file changed, 1 insertion(+), 1 deletion(-)mm/page_alloc.c +1
-2@@ -1769,7 +1769,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)* See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ alloc_flags &= ~ALLOC_CPUSET; - } else if (unlikely(rt_task(p))) + } else if (unlikely(rt_task(p)) && !in_interrupt()) alloc_flags |= ALLOC_HARDER; if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) { --To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Quoted 30 lines · Pekka Enberg
Quoted 15 lines · Mel Gorman
Quoted 4 lines
There's subtleties like where rt tasks are given
ALLOC_HARDEReven whenin_interrupt()that are different from previous kernels and could reduce the amount ofALLOC_HIGHandALLOC_HARDERmemory that can be allocated./me checks again
I think you're right. Correct it with something like?
==== CUT HERE ====
Frombca71e94e10cd93771ec5b17eccb817dd0c85360Mon Sep 17 00:00:00 2001
From: Mel Gorman <[email protected]>
Date: Thu, 22 Oct 2009 11:55:14 +0100
Subject: [PATCH] page allocator: Do not allow interrupts to useALLOC_HARDERQuoted 6 lines
Commit
341ce06f69abfafa31b9468410a13dbd60e2b237altered watermark logic slightly by allowingrt_tasksthat are handling an interrupt to setALLOC_HARDER. This patch brings the watermark logic more in line with 2.6.30.Signed-off-by Mel Gorman <[email protected]>Good catch.
Reviewed-by Pekka Enberg <[email protected]>Can we have some of the people that are hitting the OOM problems test this patch in combination with this patch:
Quoted 7 lines
http://patchwork.kernel.org/patch/54215/
Mel, can you also resend these page allocator patches to Andrew? I haven't seen him pick the above patch in -mm and we probably want to get fixes into -stable soon because distributors seem to be basing their next releases on 2.6.31...
What I'm going to do is prepare a set that assembles all the possible patches that fix the variety of errors and get people to retest to make sure we catch what is relevant, what isn't and what bugs (if any) are left. I should have the set ready by the end of the day.
Quoted 24 lines
Quoted 23 lines
---
mm/page_alloc.c | 2 +-1 file changed, 1 insertion(+), 1 deletion(-)mm/page_alloc.c +1
-2@@ -1769,7 +1769,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)* See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ alloc_flags &= ~ALLOC_CPUSET; - } else if (unlikely(rt_task(p))) + } else if (unlikely(rt_task(p)) && !in_interrupt()) alloc_flags |= ALLOC_HARDER; if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) { --To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Signature · 3 lines
Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab
Quoted 5 lines · On Thu, 22 Oct 2009 13:51:20 +0100 Mel Gorman <[email protected]> wrote:
What I'm going to do is prepare a set that assembles all the possible patches that fix the variety of errors and get people to retest to make sure we catch what is relevant, what isn't and what bugs (if any) are left. I should have the set ready by the end of the day.
Feel free to send, I got lots of those messages, so a change should be very visible for me.
Signature · 2 lines
Regards, Stephan
Quoted 11 lines · Stephan von Krawczynski
Quoted 5 lines · On Thu, 22 Oct 2009 13:51:20 +0100 Mel Gorman <[email protected]> wrote:
What I'm going to do is prepare a set that assembles all the possible patches that fix the variety of errors and get people to retest to make sure we catch what is relevant, what isn't and what bugs (if any) are left. I should have the set ready by the end of the day.
Feel free to send, I got lots of those messages, so a change should be very visible for me.
You should have received the patches and test requests in a mail thread starting with "Candidate fix for increased number of GFP_ATOMIC failures V2".
Thanks
Signature · 3 lines
Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab
David Rientjes
Acked-byQuoted 11 lines · Mel Gorman
From
bca71e94e10cd93771ec5b17eccb817dd0c85360Mon Sep 17 00:00:00 2001
From: Mel Gorman <[email protected]>
Date: Thu, 22 Oct 2009 11:55:14 +0100
Subject: [PATCH] page allocator: Do not allow interrupts to useALLOC_HARDERCommit
341ce06f69abfafa31b9468410a13dbd60e2b237altered watermark logic slightly by allowingrt_tasksthat are handling an interrupt to setALLOC_HARDER. This patch brings the watermark logic more in line with 2.6.30.Signed-off-by Mel Gorman <[email protected]>
Rik van Riel
Reviewed-byQuoted 6 lines · Mel Gorman
Commit
341ce06f69abfafa31b9468410a13dbd60e2b237altered watermark logic slightly by allowingrt_tasksthat are handling an interrupt to setALLOC_HARDER. This patch brings the watermark logic more in line with 2.6.30.Signed-off-by Mel Gorman<[email protected]>
Signature · 1 line
All rights reversed.