[BUG] 2.6.25-rc3-mm1 kernel panic while bootup on powerpc ()

[Mel Gorman]

On (04/03/08 21:18), Pekka Enberg didst pronounce:
> Andrew Morton wrote:
> >> [c000000009edf5f0] [c0000000000b56e4] .__alloc_pages_internal+0xf8/0x470
> >> [c000000009edf6e0] [c0000000000e0458] .kmem_getpages+0x8c/0x194
> >> [c000000009edf770] [c0000000000e1050] .fallback_alloc+0x194/0x254
> >> [c000000009edf820] [c0000000000e14b0] .kmem_cache_alloc+0xd8/0x144
> >> [c000000009edf8c0] [c0000000001fe0f8] .radix_tree_preload+0x50/0xd4
> >> [c000000009edf960] [c0000000000ad048] .add_to_page_cache+0x38/0x12c
> >> [c000000009edfa00] [c0000000000ad158] .add_to_page_cache_lru+0x1c/0x4c
> >> [c000000009edfa90] [c0000000000add58] .find_or_create_page+0x60/0xa8
> >> [c000000009edfb30] [c00000000011e478] .__getblk+0x140/0x310
> >> [c000000009edfc00] [c0000000001b78c4] 
> >.journal_get_descriptor_buffer+0x44/0xd8
> >> [c000000009edfca0] [c0000000001b236c] 
> >.journal_commit_transaction+0x948/0x1590
> >> [c000000009edfe00] [c0000000001b585c] .kjournald+0xf4/0x2ac
> >> [c000000009edff00] [c00000000007ff4c] .kthread+0x84/0xd0
> >> [c000000009edff90] [c000000000028900] .kernel_thread+0x4c/0x68
> >> Instruction dump:
> >> 7dc57378 48009575 60000000 2fa30000 419e0490 56c902d8 3c000018 7dd907b4 
> >> 7ad2c7e2 7f890000 7c000026 5400fffe <0b000000> e93e8128 3b000000 
> >80090000 
> >/* Convert GFP flags to their corresponding migrate type */
> >static inline int allocflags_to_migratetype(gfp_t gfp_flags)
> >{
> >        WARN_ON((gfp_flags & GFP_MOVABLE_MASK) == GFP_MOVABLE_MASK);
> >
> >Mel, Pekka: would you have some head-scratching time for this one please?
> 
> What we have is __getblk() -> __getblk_slow() -> grow_buffers() -> 
> grow_dev_page() doing find_or_create_page() with __GFP_MOVABLE set. That 
> path then eventually does radix_tree_preload -> kmem_cache_alloc() to a 
> cache that has SLAB_RECLAIM_ACCOUNT set which implies __GFP_RECLAIMABLE 
> (for both SLAB and SLUB). So we oops there.
> 
> I suspect the WARN_ON() is bogus although I really don't know that part 
> of the code all too well. Mel?
> 

The warn-on is valid. A situation should not exist that allows both flags to
be set. I suspect  if remove-set_migrateflags.patch was reverted from -mm
the warning would not trigger. Christoph, would it be reasonable to always
clear __GFP_MOVABLE when __GFP_RECLAIMABLE is set for SLAB_RECLAIM_ACCOUNT.

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab