[gentoo-commits] linux-patches r1536 - genpatches-2.6/trunk/2.6.28

[Mike Pagano (mpagano)]

Author: mpagano
Date: 2009-03-26 16:42:19 +0000 (Thu, 26 Mar 2009)
New Revision: 1536

Added:
   genpatches-2.6/trunk/2.6.28/1920_write-sync-regression-fix.patch
Modified:
   genpatches-2.6/trunk/2.6.28/0000_README
Log:
Fix for performance regression when WRITE_SYNC writes and writes submitted without sync flag caused excession idling in IO scheduler

Modified: genpatches-2.6/trunk/2.6.28/0000_README
===================================================================
--- genpatches-2.6/trunk/2.6.28/0000_README	2009-03-24 23:58:01 UTC (rev 1535)
+++ genpatches-2.6/trunk/2.6.28/0000_README	2009-03-26 16:42:19 UTC (rev 1536)
@@ -99,6 +99,10 @@
 From:   Aneesh Kumar K.V <aneesh.kumar@××××××××××××××.com>
 Desc:   ext4: Fix discard of inode prealloc space with delayed allocation
 
+Patch:  1920_write-sync-regression-fix.patch
+From:   http://bugs.gentoo.org/show_bug.cgi?id=263870
+Desc:   Fix for performance regression when WRITE_SYNC writes and writes submitted without sync flag caused excession idling in IO scheduler
+
 Patch	2300_ppc-legacymem-anon-memory.patch
 From:	https://bugs.gentoo.org/253149
 Desc:	Fix launching of X.org on some PPC platforms

Added: genpatches-2.6/trunk/2.6.28/1920_write-sync-regression-fix.patch
===================================================================
--- genpatches-2.6/trunk/2.6.28/1920_write-sync-regression-fix.patch	                        (rev 0)
+++ genpatches-2.6/trunk/2.6.28/1920_write-sync-regression-fix.patch	2009-03-26 16:42:19 UTC (rev 1536)
@@ -0,0 +1,68 @@
+From: Jens Axboe <jens.axboe@××××××.com>
+Date: Tue, 17 Feb 2009 12:59:08 +0000 (+0100)
+Subject: block: revert part of 18ce3751ccd488c78d3827e9f6bf54e6322676fb
+X-Git-Tag: v2.6.29-rc6~40^2~2
+X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=78f707bfc723552e8309b7c38a8d0cc51012e813
+
+block: revert part of 18ce3751ccd488c78d3827e9f6bf54e6322676fb
+
+The above commit added WRITE_SYNC and switched various places to using
+that for committing writes that will be waited upon immediately after
+submission. However, this causes a performance regression with AS and CFQ
+for ext3 at least, since sync_dirty_buffer() will submit some writes with
+WRITE_SYNC while ext3 has sumitted others dependent writes without the sync
+flag set. This causes excessive anticipation/idling in the IO scheduler
+because sync and async writes get interleaved, causing a big performance
+regression for the below test case (which is meant to simulate sqlite
+like behaviour).
+
+---- test case ----
+
+int main(int argc, char **argv)
+{
+
+	int fdes, i;
+	FILE *fp;
+	struct timeval start;
+	struct timeval end;
+	struct timeval res;
+
+	gettimeofday(&start, NULL);
+	for (i=0; i<ROWS; i++) {
+		fp = fopen("test_file", "a");
+		fprintf(fp, "Some Text Data\n");
+		fdes = fileno(fp);
+		fsync(fdes);
+		fclose(fp);
+	}
+	gettimeofday(&end, NULL);
+
+	timersub(&end, &start, &res);
+	fprintf(stdout, "time to write %d lines is %ld(msec)\n", ROWS,
+			(res.tv_sec*1000000 + res.tv_usec)/1000);
+
+	return 0;
+}
+
+-------------------
+
+Thanks to Sean.White@××××.com for tracking down this performance
+regression and providing a test case.
+
+Signed-off-by: Jens Axboe <jens.axboe@××××××.com>
+---
+
+diff --git a/fs/buffer.c b/fs/buffer.c
+index 665d446..62b57e3 100644
+--- a/fs/buffer.c
++++ b/fs/buffer.c
+@@ -3108,7 +3108,7 @@ int sync_dirty_buffer(struct buffer_head *bh)
+ 	if (test_clear_buffer_dirty(bh)) {
+ 		get_bh(bh);
+ 		bh->b_end_io = end_buffer_write_sync;
+-		ret = submit_bh(WRITE_SYNC, bh);
++		ret = submit_bh(WRITE, bh);
+ 		wait_on_buffer(bh);
+ 		if (buffer_eopnotsupp(bh)) {
+ 			clear_buffer_eopnotsupp(bh);
+