aboutsummaryrefslogtreecommitdiffstats
path: root/mm
diff options
context:
space:
mode:
authorFengguang Wu <wfg@mail.ustc.edu.cn>2008-02-05 01:29:36 -0500
committerLinus Torvalds <torvalds@woody.linux-foundation.org>2008-02-05 12:44:19 -0500
commit8bc3be2751b4f74ab90a446da1912fd8204d53f7 (patch)
tree2bc514025a906203244d98de70fb6bd87f3ac9ac /mm
parenta322f8ab66f50b6c0dcdb59abae84fede7a5fded (diff)
writeback: speed up writeback of big dirty files
After making dirty a 100M file, the normal behavior is to start the writeback for all data after 30s delays. But sometimes the following happens instead: - after 30s: ~4M - after 5s: ~4M - after 5s: all remaining 92M Some analyze shows that the internal io dispatch queues goes like this: s_io s_more_io ------------------------- 1) 100M,1K 0 2) 1K 96M 3) 0 96M 1) initial state with a 100M file and a 1K file 2) 4M written, nr_to_write <= 0, so write more 3) 1K written, nr_to_write > 0, no more writes(BUG) nr_to_write > 0 in (3) fools the upper layer to think that data have all been written out. The big dirty file is actually still sitting in s_more_io. We cannot simply splice s_more_io back to s_io as soon as s_io becomes empty, and let the loop in generic_sync_sb_inodes() continue: this may starve newly expired inodes in s_dirty. It is also not an option to draw inodes from both s_more_io and s_dirty, an let the loop go on: this might lead to live locks, and might also starve other superblocks in sync time(well kupdate may still starve some superblocks, that's another bug). We have to return when a full scan of s_io completes. So nr_to_write > 0 does not necessarily mean that "all data are written". This patch introduces a flag writeback_control.more_io to indicate that more io should be done. With it the big dirty file no longer has to wait for the next kupdate invokation 5s later. In sync_sb_inodes() we only set more_io on super_blocks we actually visited. This avoids the interaction between two pdflush deamons. Also in __sync_single_inode() we don't blindly keep requeuing the io if the filesystem cannot progress. Failing to do so may lead to 100% iowait. Tested-by: Mike Snitzer <snitzer@gmail.com> Signed-off-by: Fengguang Wu <wfg@mail.ustc.edu.cn> Cc: Michael Rubin <mrubin@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm')
-rw-r--r--mm/page-writeback.c9
1 files changed, 6 insertions, 3 deletions
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index a4ca162666c5..5e00f1772c20 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -567,6 +567,7 @@ static void background_writeout(unsigned long _min_pages)
567 global_page_state(NR_UNSTABLE_NFS) < background_thresh 567 global_page_state(NR_UNSTABLE_NFS) < background_thresh
568 && min_pages <= 0) 568 && min_pages <= 0)
569 break; 569 break;
570 wbc.more_io = 0;
570 wbc.encountered_congestion = 0; 571 wbc.encountered_congestion = 0;
571 wbc.nr_to_write = MAX_WRITEBACK_PAGES; 572 wbc.nr_to_write = MAX_WRITEBACK_PAGES;
572 wbc.pages_skipped = 0; 573 wbc.pages_skipped = 0;
@@ -574,8 +575,9 @@ static void background_writeout(unsigned long _min_pages)
574 min_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write; 575 min_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write;
575 if (wbc.nr_to_write > 0 || wbc.pages_skipped > 0) { 576 if (wbc.nr_to_write > 0 || wbc.pages_skipped > 0) {
576 /* Wrote less than expected */ 577 /* Wrote less than expected */
577 congestion_wait(WRITE, HZ/10); 578 if (wbc.encountered_congestion || wbc.more_io)
578 if (!wbc.encountered_congestion) 579 congestion_wait(WRITE, HZ/10);
580 else
579 break; 581 break;
580 } 582 }
581 } 583 }
@@ -640,11 +642,12 @@ static void wb_kupdate(unsigned long arg)
640 global_page_state(NR_UNSTABLE_NFS) + 642 global_page_state(NR_UNSTABLE_NFS) +
641 (inodes_stat.nr_inodes - inodes_stat.nr_unused); 643 (inodes_stat.nr_inodes - inodes_stat.nr_unused);
642 while (nr_to_write > 0) { 644 while (nr_to_write > 0) {
645 wbc.more_io = 0;
643 wbc.encountered_congestion = 0; 646 wbc.encountered_congestion = 0;
644 wbc.nr_to_write = MAX_WRITEBACK_PAGES; 647 wbc.nr_to_write = MAX_WRITEBACK_PAGES;
645 writeback_inodes(&wbc); 648 writeback_inodes(&wbc);
646 if (wbc.nr_to_write > 0) { 649 if (wbc.nr_to_write > 0) {
647 if (wbc.encountered_congestion) 650 if (wbc.encountered_congestion || wbc.more_io)
648 congestion_wait(WRITE, HZ/10); 651 congestion_wait(WRITE, HZ/10);
649 else 652 else
650 break; /* All the old data is written */ 653 break; /* All the old data is written */