[Patch] Allocate sparsemem memmap above 4G on X86_64

From: Zou Nan hai
Date: Wed May 16 2007 - 22:39:23 EST


On system with huge amount of physical memory.
VFS cache and memory memmap may eat all available system memory under
4G, then system may fail to allocated swiotlb bounce buffer.

There was a fix in arch/x86_64/mm/numa.c, but that fix does not cover
sparsemem model.
This patch add fix to sparsemem model.

Signed-off-by: Zou Nan hai <nanhai.zou@xxxxxxxxx>
Acked-by: Siddha, Suresh <suresh.b.siddha@xxxxxxxxx>
---
include/asm-x86_64/mmzone.h | 5 +++++
include/linux/bootmem.h | 3 +++
mm/sparse.c | 5 +++++
3 files changed, 13 insertions(+)

diff -Nraup a/include/asm-x86_64/mmzone.h b/include/asm-x86_64/mmzone.h
--- a/include/asm-x86_64/mmzone.h 2007-05-17 09:38:02.000000000 +0800
+++ b/include/asm-x86_64/mmzone.h 2007-05-17 09:54:10.000000000 +0800
@@ -52,5 +52,10 @@ extern int pfn_valid(unsigned long pfn);
#define FAKE_NODE_MIN_HASH_MASK (~(FAKE_NODE_MIN_SIZE - 1uL))
#endif

+#define ARCH_HAS_ALLOC_BOOTMEM_HIGH_NODE 1
+#define alloc_bootmem_high_node(pgdat,size) \
+({__alloc_bootmem_core(pgdat->bdata, size, SMP_CACHE_BYTES, (4UL*1024*1024*1024), 0);})
+
+
#endif
#endif
diff -Nraup a/include/linux/bootmem.h b/include/linux/bootmem.h
--- a/include/linux/bootmem.h 2007-05-17 09:38:02.000000000 +0800
+++ b/include/linux/bootmem.h 2007-05-17 09:37:00.000000000 +0800
@@ -131,5 +131,8 @@ extern void *alloc_large_system_hash(con
#endif
extern int hashdist; /* Distribute hashes across NUMA nodes? */

+#ifndef ARCH_HAS_ALLOC_BOOTMEM_HIGH_NODE
+#define alloc_bootmem_high_node(pgdat, size) ({NULL;})
+#endif

#endif /* _LINUX_BOOTMEM_H */
diff -Nraup a/mm/sparse.c b/mm/sparse.c
--- a/mm/sparse.c 2007-05-17 09:38:03.000000000 +0800
+++ b/mm/sparse.c 2007-05-17 09:54:27.000000000 +0800
@@ -219,6 +219,11 @@ static struct page __init *sparse_early_
if (map)
return map;

+ map = alloc_bootmem_high_node(NODE_DATA(nid),
+ sizeof(struct page) * PAGES_PER_SECTION);
+ if (map)
+ return map;
+
map = alloc_bootmem_node(NODE_DATA(nid),
sizeof(struct page) * PAGES_PER_SECTION);
if (map)




-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/