Christoph Lameter wrote:
On Wed, 26 Jul 2006, Manfred Spraul wrote:Signed-off-by: Manfred Spraul <[email protected]>Good bye to all those cacheline contentions that helped us find so many race conditions in the past if we switched on SLAB_DEBUG. I thought this was intentional?
Relax, align is nearly never set.- kmalloc uses align==0, except if the architecture requests it (ARCH_KMALLOC_MINALIGN not 0)
- on my i386 system, the following users explicitely use align: * the pmd structure (4096: hardware requirement) * the pgd structure (32 bytes: hardware requirement) * the task structure (16 byte. fxsave)* sigqueue, pid: both request 4 byte alignment (based on __alignof__()). Doesn't affect debugging.
From the other mail:
The kmalloc caches are allocated with HWCACHE_ALIGN+ARCH_KMALLOC_MINALIGN. The logic in kmem_cache_create didn't handle that case correctly. On most architectures, ARCH_KMALLOC_MINALIGN is 0. Thus SLAB_DEBUG redzones everything.Thus the patch is correct, it's a bug in the slab allocator. If HWCACHE_ALIGNis set, then the allocator ignores align or ARCH_SLAB_MINALIGN.But then Heiko does not want to set ARCH_SLAB_MINALIGN at all. This is not the issue we are discussing. In the DEBUG case he wants ARCH_KMALLOC_MINALIGN to be enforced even if ARCH_SLAB_MINALIGN is not set.
On s390, ARCH_KMALLOC_MINALIGN is 8. This disables redzoning. Ok? -- Manfred - 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/
- Follow-Ups:
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Christoph Lameter <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- References:
- Re: [patch] slab: always follow arch requested alignments
- From: Christoph Lameter <[email protected]>
- Re: [patch] slab: always follow arch requested alignments
- From: Heiko Carstens <[email protected]>
- Re: [patch] slab: always follow arch requested alignments
- From: Christoph Lameter <[email protected]>
- Re: [patch] slab: always follow arch requested alignments
- From: Heiko Carstens <[email protected]>
- Re: [patch] slab: always follow arch requested alignments
- From: Christoph Lameter <[email protected]>
- Re: [patch] slab: always follow arch requested alignments
- From: Heiko Carstens <[email protected]>
- [patch 2/2] slab: always consider arch mandated alignment
- From: Heiko Carstens <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Pekka J Enberg <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Heiko Carstens <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Pekka J Enberg <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Heiko Carstens <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Pekka J Enberg <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Manfred Spraul <[email protected]>
- Re: [patch 2/2] slab: always consider arch mandated alignment
- From: Christoph Lameter <[email protected]>
- Re: [patch] slab: always follow arch requested alignments
- Prev by Date: RE: [discuss] Re: [PATCH] Allow all Opteron processors to change pstate at same time
- Next by Date: Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion
- Previous by thread: Re: [patch 2/2] slab: always consider arch mandated alignment
- Next by thread: Re: [patch 2/2] slab: always consider arch mandated alignment
- Index(es):
 |