aboutsummaryrefslogtreecommitdiffstats
path: root/include/linux
diff options
context:
space:
mode:
authorFengguang Wu <wfg@mail.ustc.edu.cn>2007-07-19 04:48:07 -0400
committerLinus Torvalds <torvalds@woody.linux-foundation.org>2007-07-19 13:04:44 -0400
commitfe3cba17c49471e99d3421e675fc8b3deaaf0b70 (patch)
treedf696c4584c6db2e439f068d2474fcb946ca587d /include/linux
parentd8983910a4045fa21022cfccf76ed13eb40fd7f5 (diff)
mm: share PG_readahead and PG_reclaim
Share the same page flag bit for PG_readahead and PG_reclaim. One is used only on file reads, another is only for emergency writes. One is used mostly for fresh/young pages, another is for old pages. Combinations of possible interactions are: a) clear PG_reclaim => implicit clear of PG_readahead it will delay an asynchronous readahead into a synchronous one it actually does _good_ for readahead: the pages will be reclaimed soon, it's readahead thrashing! in this case, synchronous readahead makes more sense. b) clear PG_readahead => implicit clear of PG_reclaim one(and only one) page will not be reclaimed in time it can be avoided by checking PageWriteback(page) in readahead first c) set PG_reclaim => implicit set of PG_readahead will confuse readahead and make it restart the size rampup process it's a trivial problem, and can mostly be avoided by checking PageWriteback(page) first in readahead d) set PG_readahead => implicit set of PG_reclaim PG_readahead will never be set on already cached pages. PG_reclaim will always be cleared on dirtying a page. so not a problem. In summary, a) we get better behavior b,d) possible interactions can be avoided c) racy condition exists that might affect readahead, but the chance is _really_ low, and the hurt on readahead is trivial. Compound pages also use PG_reclaim, but for now they do not interact with reclaim/readahead code. Signed-off-by: Fengguang Wu <wfg@mail.ustc.edu.cn> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'include/linux')
-rw-r--r--include/linux/page-flags.h4
1 files changed, 3 insertions, 1 deletions
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 709d92fd2877..a454176c3e30 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -83,7 +83,6 @@
83#define PG_private 11 /* If pagecache, has fs-private data */ 83#define PG_private 11 /* If pagecache, has fs-private data */
84 84
85#define PG_writeback 12 /* Page is under writeback */ 85#define PG_writeback 12 /* Page is under writeback */
86#define PG_readahead 13 /* Reminder to do async read-ahead */
87#define PG_compound 14 /* Part of a compound page */ 86#define PG_compound 14 /* Part of a compound page */
88#define PG_swapcache 15 /* Swap page: swp_entry_t in private */ 87#define PG_swapcache 15 /* Swap page: swp_entry_t in private */
89 88
@@ -91,6 +90,9 @@
91#define PG_reclaim 17 /* To be reclaimed asap */ 90#define PG_reclaim 17 /* To be reclaimed asap */
92#define PG_buddy 19 /* Page is free, on buddy lists */ 91#define PG_buddy 19 /* Page is free, on buddy lists */
93 92
93/* PG_readahead is only used for file reads; PG_reclaim is only for writes */
94#define PG_readahead PG_reclaim /* Reminder to do async read-ahead */
95
94/* PG_owner_priv_1 users should have descriptive aliases */ 96/* PG_owner_priv_1 users should have descriptive aliases */
95#define PG_checked PG_owner_priv_1 /* Used by some filesystems */ 97#define PG_checked PG_owner_priv_1 /* Used by some filesystems */
96#define PG_pinned PG_owner_priv_1 /* Xen pinned pagetable */ 98#define PG_pinned PG_owner_priv_1 /* Xen pinned pagetable */