Re: Slab BUG with DEBUG_* options

[Pekka Enberg <penberg@iki.fi>]

On 11/30/2013 01:42 PM, Meelis Roos wrote:
> I am debugging a reboot problem on Sun Ultra 5 (sparc64) with 512M RAM
> and turned on DEBUG_PAGEALLOC DEBUG_SLAB and DEBUG_SLAB_LEAK (and most
> other debug options) and got the following BUG and hang on startup. This
> happened originally with 3.11-rc2-00058 where my bisection of
> another problem lead, but I retested 3.12 to have the same BUG in the
> same place.
>
> kernel BUG at mm/slab.c:2391!
>                \|/ ____ \|/
>                "@'/ .. \`@"
>                /_| \__/ |_\
>                   \__U_/
> swapper(0): Kernel bad sw trap 5 [#1]
> CPU: 0 PID: 0 Comm: swapper Not tainted 3.11.0-rc2-00058-g20bafb3-dirty #127
> task: 00000000008ac468 ti: 000000000089c000 task.ti: 000000000089c000
> TSTATE: 0000004480e01606 TPC: 00000000004f57d4 TNPC: 00000000004f57d8 Y: 00000000    Not tainted
> TPC: <__kmem_cache_create+0x374/0x480>
> g0: 00000000000000f8 g1: 00000000008bb400 g2: 000000000002780b g3: 00000000008b5120
> g4: 00000000008ac468 g5: 0000000000000000 g6: 000000000089c000 g7: 0000000000000000
> o0: 0000000000845f08 o1: 0000000000000957 o2: ffffffffffffffe0 o3: 0000000000000000
> o4: 0000000000002004 o5: 0000000000000000 sp: 000000000089f301 ret_pc: 00000000004f57cc
> RPC: <__kmem_cache_create+0x36c/0x480>
> l0: fffff8001e812040 l1: fffff8001e819f80 l2: fffff8001e819fb8 l3: fffff8001e819fd8
> l4: 0000000000000001 l5: fffff8001e819fc8 l6: 0000000000845f08 l7: fffff8001e8300a0
> i0: fffff8001e831fa0 i1: 0000000080002800 i2: 0000000080000000 i3: 0000000000000034
> i4: 0000000000000000 i5: 0000000000002000 i6: 000000000089f3b1 i7: 0000000000907464
> I7: <create_boot_cache+0x4c/0x84>
> Call Trace:
>   [0000000000907464] create_boot_cache+0x4c/0x84
>   [00000000009074d0] create_kmalloc_cache+0x34/0x60
>   [0000000000907540] create_kmalloc_caches+0x44/0x168
>   [0000000000908dfc] kmem_cache_init+0x1d0/0x1e0
>   [00000000008fc658] start_kernel+0x18c/0x370
>   [0000000000761df4] tlb_fixup_done+0x88/0x94
>   [0000000000000000]           (null)
> Disabling lock debugging due to kernel taint
> Caller[0000000000907464]: create_boot_cache+0x4c/0x84
> Caller[00000000009074d0]: create_kmalloc_cache+0x34/0x60
> Caller[0000000000907540]: create_kmalloc_caches+0x44/0x168
> Caller[0000000000908dfc]: kmem_cache_init+0x1d0/0x1e0
> Caller[00000000008fc658]: start_kernel+0x18c/0x370
> Caller[0000000000761df4]: tlb_fixup_done+0x88/0x94
> Caller[0000000000000000]:           (null)
> Instruction DUMP: 92102957  7ffccb35  90122308 <91d02005> 90100018  4009b371  920f20d0  ba922000  02480006
> Kernel panic - not syncing: Attempted to kill the idle task!
> Press Stop-A (L1-A) to return to the boot prom
>
> The line shows that __kmem_cache_create gets a NULL from kmalloc_slab().
>
> I instrumented the code and found the following:
>
> __kmem_cache_create: starting, size=248, flags=8192
> __kmem_cache_create: now flags=76800
> __kmem_cache_create: aligned size to 248 because of redzoning
> __kmem_cache_create: pagealloc debug, setting size to 8192
> __kmem_cache_create: aligned size to 8192
> __kmem_cache_create: num=1, slab_size=64
> __kmem_cache_create: starting, size=96, flags=8192
> __kmem_cache_create: now flags=76800
> __kmem_cache_create: aligned size to 96 because of redzoning
> __kmem_cache_create: pagealloc debug, setting size to 8192
> __kmem_cache_create: aligned size to 8192
> __kmem_cache_create: num=1, slab_size=64
> __kmem_cache_create: starting, size=192, flags=8192
> __kmem_cache_create: now flags=76800
> __kmem_cache_create: aligned size to 192 because of redzoning
> __kmem_cache_create: pagealloc debug, setting size to 8192
> __kmem_cache_create: aligned size to 8192
> __kmem_cache_create: num=1, slab_size=64
> __kmem_cache_create: starting, size=32, flags=8192
> __kmem_cache_create: now flags=76800
> __kmem_cache_create: aligned size to 32 because of redzoning
> __kmem_cache_create: aligned size to 32
> __kmem_cache_create: num=226, slab_size=960
> __kmem_cache_create: starting, size=64, flags=8192
> __kmem_cache_create: now flags=76800
> __kmem_cache_create: aligned size to 64 because of redzoning
> __kmem_cache_create: pagealloc debug, setting size to 8192
> __kmem_cache_create: turning on CFLGS_OFF_SLAB, size=8192
> __kmem_cache_create: aligned size to 8192
> __kmem_cache_create: num=1, slab_size=64
> __kmem_cache_create: CFLGS_OFF_SLAB, size=8192, slab_size=52
> __kmem_cache_create: CFLGS_OFF_SLAB, allocating slab 52
>
> With slab size 64, it turns on CFLGS_OFF_SLAB and off slab allocation
> with this size fails. I do not know slab internals so I can not tell if
> this just happens because of the debug paths, or is it a real problem
> without the debug options too.
>

There was a rather large change to mm/slab.c that changed it to use 
'struct page' like SLUB. Perhaps slab debugging was broken in the 
process. Joonsoo, does the problem Meelis describes ring a bell?

                         Pekka

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>