David Mosberger wrote:
> It's all very embarassing, but my previous patch was horribly broken:
> it added the NULL terminator to the wrong array... Of course, adding
> it to the correct array uncovered another bug... ;-(
>
> Patch below fixes both problems. Reall, I mean it.
>
> Andrew, you may want to double-check---I didn't look through all of
> slab.c whether there might be problems (there was no other mention of
> ARRAY_SIZE(malloc_sizes) though.
>
> Thanks,
>
> --david
>
My fault, I got a bit overzealous while cleaning that code up. Here's a
better patch that gets rid of ARRAY_SIZE since the null is readded.
-- Brian Gerst
--- linux-2.5.67-bk/mm/slab.c 2003-04-10 08:36:50.000000000 -0400 +++ linux/mm/slab.c 2003-04-10 19:18:08.000000000 -0400 @@ -383,11 +383,12 @@ } malloc_sizes[] = { #define CACHE(x) { .cs_size = (x) }, #include <linux/kmalloc_sizes.h> + { 0, } #undef CACHE }; /* Must match cache_sizes above. Out of line to keep cache footprint low. */ -static struct { +static struct cache_names { char *name; char *name_dma; } cache_names[] = { @@ -596,7 +597,9 @@ */ void __init kmem_cache_sizes_init(void) { - int i; + struct cache_sizes *sizes = malloc_sizes; + struct cache_names *names = cache_names; + /* * Fragmentation resistance on low memory - only use bigger * page orders on machines with more than 32MB of memory. @@ -604,15 +607,14 @@ if (num_physpages > (32 << 20) >> PAGE_SHIFT) slab_break_gfp_order = BREAK_GFP_ORDER_HI; - for (i = 0; i < ARRAY_SIZE(malloc_sizes); i++) { - struct cache_sizes *sizes = malloc_sizes + i; + while (sizes->cs_size) { /* For performance, all the general caches are L1 aligned. * This should be particularly beneficial on SMP boxes, as it * eliminates "false sharing". * Note for systems short on memory removing the alignment will * allow tighter packing of the smaller caches. */ sizes->cs_cachep = kmem_cache_create( - cache_names[i].name, sizes->cs_size, + names->name, sizes->cs_size, 0, SLAB_HWCACHE_ALIGN, NULL, NULL); if (!sizes->cs_cachep) BUG(); @@ -624,10 +626,13 @@ } sizes->cs_dmacachep = kmem_cache_create( - cache_names[i].name_dma, sizes->cs_size, + names->name_dma, sizes->cs_size, 0, SLAB_CACHE_DMA|SLAB_HWCACHE_ALIGN, NULL, NULL); if (!sizes->cs_dmacachep) BUG(); + + sizes++; + names++; } /* * The generic caches are running - time to kick out the
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Apr 15 2003 - 22:00:22 EST