Gitweb:
http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=e2848a0efedef4dad52d1334d37f8719cd6268fd
Commit: e2848a0efedef4dad52d1334d37f8719cd6268fd
Parent: e31d9eb5c17ae3b80f9e9403f8a5eaf6dba879c9
Author: Nick Piggin <[EMAIL PROTECTED]>
AuthorDate: Mon Feb 4 22:29:10 2008 -0800
Committer: Linus Torvalds <[EMAIL PROTECTED]>
CommitDate: Tue Feb 5 09:44:17 2008 -0800
radix-tree: avoid atomic allocations for preloaded insertions
Most pagecache (and some other) radix tree insertions have the great
opportunity to preallocate a few nodes with relaxed gfp flags. But the
preallocation is squandered when it comes time to allocate a node, we
default to first attempting a GFP_ATOMIC allocation -- that doesn't
normally fail, but it can eat into atomic memory reserves that we don't
need to be using.
Another upshot of this is that it removes the sometimes highly contended
zone->lock from underneath tree_lock. Pagecache insertions are always
performed with a radix tree preload, and after this change, such a
situation will never fall back to kmem_cache_alloc within
radix_tree_node_alloc.
David Miller reports seeing this allocation fail on a highly threaded
sparc64 system:
[527319.459981] dd: page allocation failure. order:0, mode:0x20
[527319.460403] Call Trace:
[527319.460568] [00000000004b71e0] __slab_alloc+0x1b0/0x6a8
[527319.460636] [00000000004b7bbc] kmem_cache_alloc+0x4c/0xa8
[527319.460698] [000000000055309c] radix_tree_node_alloc+0x20/0x90
[527319.460763] [0000000000553238] radix_tree_insert+0x12c/0x260
[527319.460830] [0000000000495cd0] add_to_page_cache+0x38/0xb0
[527319.460893] [00000000004e4794] mpage_readpages+0x6c/0x134
[527319.460955] [000000000049c7fc] __do_page_cache_readahead+0x170/0x280
[527319.461028] [000000000049cc88] ondemand_readahead+0x208/0x214
[527319.461094] [0000000000496018] do_generic_mapping_read+0xe8/0x428
[527319.461152] [0000000000497948] generic_file_aio_read+0x108/0x170
[527319.461217] [00000000004badac] do_sync_read+0x88/0xd0
[527319.461292] [00000000004bb5cc] vfs_read+0x78/0x10c
[527319.461361] [00000000004bb920] sys_read+0x34/0x60
[527319.461424] [0000000000406294] linux_sparc_syscall32+0x3c/0x40
The calltrace is significant: __do_page_cache_readahead allocates a number
of pages with GFP_KERNEL, and hence it should have reclaimed sufficient
memory to satisfy GFP_ATOMIC allocations. However after the list of pages
goes to mpage_readpages, there can be significant intervals (including disk
IO) before all the pages are inserted into the radix-tree. So the reserves
can easily be depleted at that point. The patch is confirmed to fix the
problem.
Signed-off-by: Nick Piggin <[EMAIL PROTECTED]>
Cc: "David S. Miller" <[EMAIL PROTECTED]>
Signed-off-by: Andrew Morton <[EMAIL PROTECTED]>
Signed-off-by: Linus Torvalds <[EMAIL PROTECTED]>
---
lib/radix-tree.c | 15 +++++++++++----
mm/filemap.c | 1 -
mm/rmap.c | 1 -
3 files changed, 11 insertions(+), 6 deletions(-)
diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index 48c250f..65f0e75 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -95,14 +95,17 @@ static inline gfp_t root_gfp_mask(struct radix_tree_root
*root)
static struct radix_tree_node *
radix_tree_node_alloc(struct radix_tree_root *root)
{
- struct radix_tree_node *ret;
+ struct radix_tree_node *ret = NULL;
gfp_t gfp_mask = root_gfp_mask(root);
- ret = kmem_cache_alloc(radix_tree_node_cachep,
- set_migrateflags(gfp_mask, __GFP_RECLAIMABLE));
- if (ret == NULL && !(gfp_mask & __GFP_WAIT)) {
+ if (!(gfp_mask & __GFP_WAIT)) {
struct radix_tree_preload *rtp;
+ /*
+ * Provided the caller has preloaded here, we will always
+ * succeed in getting a node here (and never reach
+ * kmem_cache_alloc)
+ */
rtp = &__get_cpu_var(radix_tree_preloads);
if (rtp->nr) {
ret = rtp->nodes[rtp->nr - 1];
@@ -110,6 +113,10 @@ radix_tree_node_alloc(struct radix_tree_root *root)
rtp->nr--;
}
}
+ if (ret == NULL)
+ ret = kmem_cache_alloc(radix_tree_node_cachep,
+ set_migrateflags(gfp_mask, __GFP_RECLAIMABLE));
+
BUG_ON(radix_tree_is_indirect_ptr(ret));
return ret;
}
diff --git a/mm/filemap.c b/mm/filemap.c
index 76bea88..96920f8 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -65,7 +65,6 @@ generic_file_direct_IO(int rw, struct kiocb *iocb, const
struct iovec *iov,
* ->private_lock (__free_pte->__set_page_dirty_buffers)
* ->swap_lock (exclusive_swap_page, others)
* ->mapping->tree_lock
- * ->zone.lock
*
* ->i_mutex
* ->i_mmap_lock (truncate->unmap_mapping_range)
diff --git a/mm/rmap.c b/mm/rmap.c
index dbc2ca2..0334c8f 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -36,7 +36,6 @@
* mapping->tree_lock (widely used, in set_page_dirty,
* in arch-dependent flush_dcache_mmap_lock,
* within inode_lock in __sync_single_inode)
- * zone->lock (within radix tree node alloc)
*/
#include <linux/mm.h>
-
To unsubscribe from this list: send the line "unsubscribe git-commits-head" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
