Home » Mailing list archives » linux-kernel » 2007 » November » 7

Re: [BUG] Linux 2.6.24-rc2 - oom-killer gets invoked

Previous thread: pxeboot broken by commit "x86 setup: handle boot loaders which set up the stack incorrectly" by Yinghai Lu on Tuesday, November 6, 2007 - 8:18 pm. (1 message)

Next thread: [PATCH] fix up perfmon to build on -mm by Greg KH on Tuesday, November 6, 2007 - 8:34 pm. (114 messages)
[view original message]
From: Linus Torvalds <torvalds@...>
To: Linux Kernel Mailing List <linux-kernel@...>
Subject: Linux 2.6.24-rc2
Date: Tuesday, November 6, 2007 - 8:26 pm

Yeah, don't remind me - it's late. 


There was nothng in particular holding this thing up, I just basically

just forgot to cut a -rc2 release last week.


There's not a lot of hugely exciting stuff here. Some arch updates: MIPS,

arm, blackfin, x86, sparc[64], sh, s390..


Also various driver updates: libata, IDE, networking, DVB.. And some more

fallout from the scatter-gather changes.


Some scheduler cleanups, and also fixing the CPU usage statistics that

got scrogged at some point.


Nothing really earthshattering that I can recall or see. A lot of it is

small buglets and compiler warnings (and occasionally broken builds). The

shortlog is still slightly too large for the mailing list limit, so you

need to generate it either from the full log (posted on the normal sites)

or with


	git shortlog v2.6.24-rc1..


but quite frankly, it's no Leo Tolstoy. If you have trouble falling

asleep, you might try to print it out and take it to bed with you: it's

not going to be more than just a couple of pages ("use 2nup and save a

tree"), but I dare you to actually get to the end. Snooze city.


It's full of such *riveting* entries like:


	James Bottomley:

	      x86: voyager: fix bogus conversion to per_cpu for boot_cpu_info

	...

	Jens Axboe:

	      mmc: sg fallout

	...

	Joerg Roedel:

	      x86 gart: rename iommu.h to gart.h

	...

	Mauro Carvalho Chehab:

	      V4L/DVB (6390): Updates missing entries at CARDLIST.em28xx

	...

	Ralf Baechle:

	      [MIPS] Alchemy: Nuke homebrew setup_irq(), it's broken and unnecessary.

	...

	Yu Luming:

	      [IA64] fix typo in per_cpu_offset

	..


and yes, I just picked those out randomly. 


			Linus

-
[view original message]
From: Jon Masters <jonathan@...>
To: Linux Kernel Mailing List <linux-kernel@...>
Subject: Re: Linux 2.6.24-rc2
Date: Friday, November 9, 2007 - 8:07 am

Nobody just happens upon a Voyager quote like that, it takes (at least

some) skill to find, just like actually finding said hardware.


Jon.


-
[view original message]
From: Rafael J. Wysocki <rjw@...>
To: Linux Kernel Mailing List <linux-kernel@...>
Cc: Linus Torvalds <torvalds@...>, Andrew Morton <akpm@...>
Subject: 2.6.24-rc2: Reported regressions from 2.6.23
Date: Wednesday, November 7, 2007 - 9:23 pm

This message contains a list of some regressions from 2.6.23 which have been

reported since 2.6.24-rc1 was released and for which there are no fixes in the

mainline that I know of.  If any of them have been fixed already, please let me

know.


If you know of any other unresolved regressions from 2.6.23, please let me know

either and I'll add them to the list.


Subject		: On 2.6.24-rc1-gc9927c2b BUG: unable to handle kernel paging request at virtual address 3d15b925

Submitter	: Giacomo Catenazzi <cate@cateee.net>

References	: http://lkml.org/lkml/2007/10/24/487

		  http://bugzilla.kernel.org/show_bug.cgi?id=9246

Handled-By	:

Patch		: 


Subject		: Potential regression in -git15: can't resume stopped root shell?

Submitter	: Theodore Tso <tytso@mit.edu>

References	: http://lkml.org/lkml/2007/10/20/114

		  http://bugzilla.kernel.org/show_bug.cgi?id=9247

Handled-By	: Serge Hallyn <serue@us.ibm.com>

Patch		: http://bugzilla.kernel.org/attachment.cgi?id=13361&action=view

		  http://bugzilla.kernel.org/attachment.cgi?id=13375&action=view


Subject		: irq 21: nobody cared 2.6.24-rc1

Submitter	: Bongani Hlope <bonganilinux@mweb.co.za>

References	: http://lkml.org/lkml/2007/10/25/90

		  http://bugzilla.kernel.org/show_bug.cgi?id=9249

Handled-By	:

Patch		: 


Subject		: [BUG] panic after umount (biscted)

Submitter	: Sebastian Siewior <lkml@ml.breakpoint.cc>

References	: http://marc.info/?l=linux-kernel&m=119338387030335&w=2

		  http://bugzilla.kernel.org/show_bug.cgi?id=9250

Handled-By	: Jens Axboe <jens.axboe@oracle.com>

Patch		: http://marc.info/?l=linux-kernel&m=119348520210349&w=2


Subject		: 2.6.24-rc1 sysctl table check failed on PowerMac

Submitter	: Mikael Pettersson <mikpe@it.uu.se>

References	: http://marc.info/?l=linux-kernel&m=119350802331857&w=2

		  http://bugzilla.kernel.org/show_bug.cgi?id=9251

Handled-By	: Alexey Dobriyan <adobriyan@gmail.com>

Patch		: [ message continues ]
" title="http://marc.info/?l=linux-ker...
[ message continues ]
">http://marc.info/?l=linux-ker...
[ message continues ]

[view original message]
From: Matthias Kaehlcke <matthias.kaehlcke@...>
To: Linus Torvalds <torvalds@...>
Cc: Linux Kernel Mailing List <linux-kernel@...>
Subject: Re: Linux 2.6.24-rc2 build
Date: Wednesday, November 7, 2007 - 5:55 am

v2.6.24-rc2 fails to build on my box with the following error:


CC [M]  drivers/kvm/kvm_main.o

drivers/kvm/kvm_main.c: In function
[view original message]
From: Randy Dunlap <randy.dunlap@...>
To: Matthias Kaehlcke <matthias.kaehlcke@...>, <mingo@...>
Cc: Linus Torvalds <torvalds@...>, Linux Kernel Mailing List <linux-kernel@...>
Subject: Re: Linux 2.6.24-rc2 build fails: implicit declaration of function
Date: Wednesday, November 7, 2007 - 12:21 pm

Ingo has a patch that he was going to push for this.


---

~Randy

-
[view original message]
From: Miguel <mboton.lkml@...>
To: Matthias Kaehlcke <matthias.kaehlcke@...>, Linus Torvalds <torvalds@...>, Linux Kernel Mailing List <linux-kernel@...>
Subject: Re: Linux 2.6.24-rc2 build fails: implicit declaration of function
Date: Wednesday, November 7, 2007 - 11:06 am

This patch should fix this error.


Signed-off-by: Miguel Boton <mboton@gmail.com>


Index: linux-2.6.24-rc2/drivers/kvm/kvm_main.c

===================================================================

--- linux-2.6.24-rc2.orig/drivers/kvm/kvm_main.c

+++ linux-2.6.24-rc2/drivers/kvm/kvm_main.c

@@ -217,7 +217,10 @@

 		if (cpu != -1 && cpu != raw_smp_processor_id())

 			cpu_set(cpu, cpus);

 	}

+

+#ifdef CONFIG_SMP

 	smp_call_function_mask(cpus, ack_flush, NULL, 1);

+#endif

 }


 int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)


--

	Miguel Botón

-
[view original message]
From: Kamalesh Babulal <kamalesh@...>
To: Linus Torvalds <torvalds@...>
Cc: Linux Kernel Mailing List <linux-kernel@...>, Andy Whitcroft <apw@...>, Balbir Singh <balbir@...>
Subject: [BUG] Linux 2.6.24-rc2 - oom-killer gets invoked
Date: Wednesday, November 7, 2007 - 3:17 am

Hi,


oom-killer got invoked while running ltp-runall on the 2.6.24-rc2 kernel.


python invoked oom-killer: gfp_mask=0x1201d2, order=0, oomkilladj=0


Call Trace:

 [<ffffffff8026b03b>] oom_kill_process+0x4f/0xf5

 [<ffffffff8026b517>] out_of_memory+0x1bc/0x22d

 [<ffffffff8026dfa2>] __alloc_pages+0x282/0x313

 [<ffffffff804e453c>] __wait_on_bit_lock+0x5b/0x66

 [<ffffffff8026fad8>] __do_page_cache_readahead+0x7c/0x18f

 [<ffffffff8026a723>] filemap_fault+0x15d/0x317

 [<ffffffff8027513e>] __do_fault+0x68/0x3bb

 [<ffffffff80276d44>] handle_mm_fault+0x325/0x694

 [<ffffffff804e739f>] do_page_fault+0x3c5/0x764

 [<ffffffff802118c7>] arch_get_unmapped_area+0x184/0x1f9

 [<ffffffff804e5939>] error_exit+0x0/0x51


Mem-info:

Node 0 DMA per-cpu:

CPU    0: Hot: hi:    0, btch:   1 usd:   0   Cold: hi:    0, btch:   1 usd:   0

CPU    1: Hot: hi:    0, btch:   1 usd:   0   Cold: hi:    0, btch:   1 usd:   0

CPU    2: Hot: hi:    0, btch:   1 usd:   0   Cold: hi:    0, btch:   1 usd:   0

CPU    3: Hot: hi:    0, btch:   1 usd:   0   Cold: hi:    0, btch:   1 usd:   0

Node 0 DMA32 per-cpu:

CPU    0: Hot: hi:  186, btch:  31 usd:  29   Cold: hi:   62, btch:  15 usd:  14

CPU    1: Hot: hi:  186, btch:  31 usd:  95   Cold: hi:   62, btch:  15 usd:  52

CPU    2: Hot: hi:  186, btch:  31 usd:  33   Cold: hi:   62, btch:  15 usd:  51

CPU    3: Hot: hi:  186, btch:  31 usd: 101   Cold: hi:   62, btch:  15 usd:  50

Active:118809 inactive:124570 dirty:0 writeback:1882 unstable:0

 free:2358 slab:2831 mapped:41 pagetables:2058 bounce:0

Node 0 DMA free:3972kB min:28kB low:32kB high:40kB active:996kB inactive:3036kB present:7552kB pages_scanned:0 all_unreclaimable? no

lowmem_reserve[]: 0 992 992 992

Node 0 DMA32 free:177860kB min:4012kB low:5012kB high:6016kB active:408456kB inactive:388372kB present:1015864kB pages_scanned:640 all_unreclaimable? no

lowmem_reserve[]: 0 0 0 0

Node 0 DMA: 57*4kB 39*8kB 25*16kB 10*32kB 2*64kB 3*128kB 1*256kB 0*512kB 1*1...
[ message continues ]
[view original message]
From: Andrew Morton <akpm@...>
To: Kamalesh Babulal <kamalesh@...>
Cc: <torvalds@...>, <linux-kernel@...>, <apw@...>, <balbir@...>, <linux-mm@...>
Subject: Re: [BUG] Linux 2.6.24-rc2 - oom-killer gets invoked
Date: Wednesday, November 7, 2007 - 2:20 pm

This is stupid - there's just no way we should have got within five miles

of out_of_memory() when there's 177MB free in ZONE_DMA32.


I can only think that the zonelists got set up wrongly, or the freelist got


Yeah, I'm sitting on a fix for that - I have a large number of for-2.6.24

fixes which I need to launder through subystem maintainers.


-
[view original message]
From: Balbir Singh <balbir@...>
To: Kamalesh Babulal <kamalesh@...>
Cc: Linus Torvalds <torvalds@...>, Linux Kernel Mailing List <linux-kernel@...>, Andy Whitcroft <apw@...>
Subject: Re: [BUG] Linux 2.6.24-rc2 - oom-killer gets invoked
Date: Wednesday, November 7, 2007 - 3:55 am

You don't see the problem with 2.6.24-rc1 right? Any chance of you


--

	Warm Regards,

	Balbir Singh

	Linux Technology Center

	IBM, ISTL

-
[view original message]
From: Kamalesh Babulal <kamalesh@...>
To: <balbir@...>
Cc: Linus Torvalds <torvalds@...>, Linux Kernel Mailing List <linux-kernel@...>, Andy Whitcroft <apw@...>
Subject: Re: [BUG] Linux 2.6.24-rc2 - oom-killer gets invoked
Date: Wednesday, November 7, 2007 - 8:24 am

Hi Balbir,


I have started doing the git-bisect and will update the results.


--

Thanks & Regards,

Kamalesh Babulal,

Linux Technology Center,

IBM, ISTL.

-
Previous thread: pxeboot broken by commit "x86 setup: handle boot loaders which set up the stack incorrectly" by Yinghai Lu on Tuesday, November 6, 2007 - 8:18 pm. (1 message)

Next thread: [PATCH] fix up perfmon to build on -mm by Greg KH on Tuesday, November 6, 2007 - 8:34 pm. (114 messages)