Skip to contentSkip to conversation

Out of memory error

DISCUSSION Cliffe 2 participants 3 messages Jan 31, 2009 7 messages not in the archive

lore
Share & tools
lore

Cliffe

author
#6 lore raw
In reply to a message not in the archive

Hello,

I get the following error messages from code I am working on. It works fine in userspace.

I suspect it may be due to the kernel's limited stack size as it uses a recursive algorithm. If this is the cause, how can I provide the (lsm) module with more memory? This is for a research project so it is ok if the solution is not quite kosher.

Thank you, any help will be appreciated. Please CC me in on any replies.

Cliffe.

The message:

The following is only an harmless informational message. Unless you get a _continuous_flood_ of these messages it means everything is working fine. Allocations from irqs cannot be perfectly reliable and the kernel is designed to handle that.
bash: page allocation failure. order:0, mode:0x20

[<c015846a>] __alloc_pages+0x2c5/0x2d6
[<c016de72>] cache_alloc_refill+0x2ae/0x4c1
[<c0129640>] irq_exit+0x53/0x6b
[<c016e0fc>] __kmalloc+0x77/0x8f
[<e1060c01>] mem_alloc+0x19/0x5f [fbac_lsm]
[<e1069b14>] copy_to_new_to_send_recursive+0x22/0x91 [fbac_lsm]
[<e1069cf3>] copy_to_new_func_recursive+0xde/0x11d [fbac_lsm]
[<e1069ce3>] copy_to_new_func_recursive+0xce/0x11d [fbac_lsm]
[<e1069ce3>] copy_to_new_func_recursive+0xce/0x11d [fbac_lsm]
[<e1069ce3>] copy_to_new_func_recursive+0xce/0x11d [fbac_lsm]
[<e10633fc>] build_task_tree_found_application+0x82/0x144 [fbac_lsm]
[<e10635d7>] build_task_tree_find_application+0x9d/0x1a5 [fbac_lsm]
[<e106381c>] build_task_tree+0x13d/0x1a1 [fbac_lsm]
[<e1064ca9>] start_task+0x32/0x175 [fbac_lsm]
[<e1060a6a>] fbac_lsm_bprm_set_security+0x124/0x1f1 [fbac_lsm]
[<c017449f>] prepare_binprm+0xa7/0xd0
[<c0173f27>] count+0x14/0x3f
[<c017597f>] do_execve+0xef/0x1e1
[<c01036fe>] sys_execve+0x2f/0x78
[<c0104e22>] sysenter_past_esp+0x6b/0xa9
=======================

Mem-info:
DMA per-cpu:
CPU 0: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0
Normal per-cpu:
CPU 0: Hot: hi: 186, btch: 31 usd: 22 Cold: hi: 62, btch: 15 usd: 12
Active:50484 inactive:64525 dirty:108 writeback:0 unstable:0
free:761 slab:10659 mapped:10607 pagetables:421 bounce:0
DMA free:2000kB min:88kB low:108kB high:132kB active:392kB inactive:8332kB present:16256kB pages_scanned:0 all_unreclaimable? no lowmem_reserve[]: 0 492 492
Normal free:1044kB min:2792kB low:3488kB high:4188kB active:201544kB inactive:249768kB present:503936kB pages_scanned:0 all_unreclaimable? no lowmem_reserve[]: 0 0 0
DMA: 0*4kB 0*8kB 1*16kB 0*32kB 1*64kB 1*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 2000kB
Normal: 1*4kB 0*8kB 1*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 1*1024kB 0*2048kB 0*4096kB = 1044kB
Swap cache: add 0, delete 0, find 0/0, race 0+0
Free swap = 0kB
Total swap = 0kB
Free swap: 0kB
131072 pages of RAM
0 pages of HIGHMEM
2092 reserved pages
92964 pages shared
0 pages swap cached
108 pages dirty
0 pages writeback
10607 pages mapped
10659 pages slab
421 pages pagetables

#7 lore raw
Quoted 4 lines · Cliffe

Hello,

I get the following error messages from code I am working on. It works fine in userspace.

its very very unlikely you'll ever experience malloc() failing in userspace.

Quoted 3 lines

I suspect it may be due to the kernel's limited stack size as it uses a recursive algorithm. If this is the cause, how can I provide the (lsm) module with more memory?

You cannot, you'll have to rewrite your code to be iterative and use an allocated traversal stack.

Quoted 2 lines

This is for a research project so it is ok if
the solution is not quite kosher.

And here you shatter my illusion that science is careful work ;-)

Quoted 1 line

Thank you, any help will be appreciated. Please CC me in on any replies.

Its simply a case of you using kmalloc(GFP_ATOMIC) and that's failing. Furthermore, it appears to me you're not using frame pointers for your kernel builds, please ammend that, it gives far more readable output.

Quoted 65 lines

The following is only an harmless informational message. Unless you get a _continuous_flood_ of these messages it means everything is working fine. Allocations from irqs cannot be perfectly reliable and the kernel is designed to handle that.
bash: page allocation failure. order:0, mode:0x20

 [<c015846a>] __alloc_pages+0x2c5/0x2d6
 [<c016de72>] cache_alloc_refill+0x2ae/0x4c1
 [<c0129640>] irq_exit+0x53/0x6b
 [<c016e0fc>] __kmalloc+0x77/0x8f
 [<e1060c01>] mem_alloc+0x19/0x5f [fbac_lsm]
 [<e1069b14>] copy_to_new_to_send_recursive+0x22/0x91 [fbac_lsm]
 [<e1069cf3>] copy_to_new_func_recursive+0xde/0x11d [fbac_lsm]
 [<e1069ce3>] copy_to_new_func_recursive+0xce/0x11d [fbac_lsm]
 [<e1069ce3>] copy_to_new_func_recursive+0xce/0x11d [fbac_lsm]
 [<e1069ce3>] copy_to_new_func_recursive+0xce/0x11d [fbac_lsm]
 [<e10633fc>] build_task_tree_found_application+0x82/0x144 [fbac_lsm]
 [<e10635d7>] build_task_tree_find_application+0x9d/0x1a5 [fbac_lsm]
 [<e106381c>] build_task_tree+0x13d/0x1a1 [fbac_lsm]
 [<e1064ca9>] start_task+0x32/0x175 [fbac_lsm]
 [<e1060a6a>] fbac_lsm_bprm_set_security+0x124/0x1f1 [fbac_lsm]
 [<c017449f>] prepare_binprm+0xa7/0xd0
 [<c0173f27>] count+0x14/0x3f
 [<c017597f>] do_execve+0xef/0x1e1
 [<c01036fe>] sys_execve+0x2f/0x78
 [<c0104e22>] sysenter_past_esp+0x6b/0xa9
 =======================

Mem-info:
DMA per-cpu:
CPU 0: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1
usd: 0
Normal per-cpu:
CPU 0: Hot: hi: 186, btch: 31 usd: 22 Cold: hi: 62, btch: 15
usd: 12
Active:50484 inactive:64525 dirty:108 writeback:0 unstable:0
free:761 slab:10659 mapped:10607 pagetables:421 bounce:0
DMA free:2000kB min:88kB low:108kB high:132kB active:392kB inactive:8332kB present:16256kB pages_scanned:0 all_unreclaimable? no lowmem_reserve[]: 0 492 492
Normal free:1044kB min:2792kB low:3488kB high:4188kB active:201544kB inactive:249768kB present:503936kB pages_scanned:0 all_unreclaimable? no lowmem_reserve[]: 0 0 0
DMA: 0*4kB 0*8kB 1*16kB 0*32kB 1*64kB 1*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 2000kB
Normal: 1*4kB 0*8kB 1*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 1*1024kB 0*2048kB 0*4096kB = 1044kB
Swap cache: add 0, delete 0, find 0/0, race 0+0
Free swap = 0kB
Total swap = 0kB
Free swap: 0kB
131072 pages of RAM
0 pages of HIGHMEM
2092 reserved pages
92964 pages shared
0 pages swap cached
108 pages dirty
0 pages writeback
10607 pages mapped
10659 pages slab
421 pages pagetables

--
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/

#9 lore raw
In reply to a message not in the archive
Quoted 5 lines · Cliffe
Quoted 2 lines

Its simply a case of you using kmalloc(GFP_ATOMIC) and that's failing.

So it has nothing to do with stack size because we are talking heap? What could be causing the problem?

Right, heap. The thing is, Linux has a greedy memory usage, and tries to keep as much as possible in memory, this means that we're basically in a constant state of low memory.

Regular allocations (GFP_KERNEL) can sleep, and can therefore reclaim memory (write out stuff dirty to disk, swap a little, etc..).

Your allocation (GFP_ATOMIC) isn't allowed to sleep, and can therefore not reclaim memory -- if it really hits rock bottom, not uncommon, it just fails to provide memory.

Any code using GFP_ATOMIC (or for that matter, any kernel allocation, GFP_KERNEL can fail too, just not as easy) must be able to deal with allocation failures.

How to do that is very domain specific.

Quoted 1 line

How limited is the stack?

Depends on the arch and build details, 4k is a reasonable assumption.

Quoted 5 lines
Quoted 3 lines

Furthermore, it appears to me you're not using frame pointers for your kernel builds, please ammend that, it gives far more readable output.

I'll look into that. Is that a kernel compile config option?

CONFIG_FRAME_POINTER=y