diff options
| author | Owain Ainsworth <oga@cvs.openbsd.org> | 2009-08-06 15:28:15 +0000 |
|---|---|---|
| committer | Owain Ainsworth <oga@cvs.openbsd.org> | 2009-08-06 15:28:15 +0000 |
| commit | 6e7e8975626fe603529c0e60cefa188eeb5c890e (patch) | |
| tree | 3b3d95f29dc62527a2b78ff502abe9eb379a1445 /sys/kern | |
| parent | f32484ba9d7d14a17687020bdcc98d2e3d42a940 (diff) | |
reintroduce the uvm_tree commit.
Now instead of the global object hashtable, we have a per object tree.
Testing shows no performance difference and a slight code shrink. OTOH when
locking is more fine grained this should be faster due to lock contention on
uvm.hashlock.
ok thib@, art@.
Diffstat (limited to 'sys/kern')
| -rw-r--r-- | sys/kern/vfs_biomem.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/sys/kern/vfs_biomem.c b/sys/kern/vfs_biomem.c index 89ab5f4d114..6b7aaa8d5e2 100644 --- a/sys/kern/vfs_biomem.c +++ b/sys/kern/vfs_biomem.c @@ -1,4 +1,4 @@ -/* $OpenBSD: vfs_biomem.c,v 1.9 2009/06/25 15:49:26 thib Exp $ */ +/* $OpenBSD: vfs_biomem.c,v 1.10 2009/08/06 15:28:14 oga Exp $ */ /* * Copyright (c) 2007 Artur Grabowski <art@openbsd.org> * @@ -64,7 +64,7 @@ buf_mem_init(vsize_t size) buf_object = &buf_object_store; buf_object->pgops = NULL; - TAILQ_INIT(&buf_object->memq); + RB_INIT(&buf_object->memt); buf_object->uo_npages = 0; buf_object->uo_refs = 1; } |