NotificationsYou must be signed in to change notification settings
Fork28
Star151

Commit48b2802

committed

When changing select() calls for delays into pg_usleep(), two comments

in s_lock.c were not updated, and still refers to select. Made my grephit the wrong files, so I figured a simple patch was in order.. (otherrefs in the same comment block was changed..)Magnus Hagander

1 parent24614a9 commit48b2802Copy full SHA for 48b2802

File tree

1 file changed

-3

lines changed

src/backend/storage/lmgr
- s_lock.c

1 file changed

-3

lines changed

`‎src/backend/storage/lmgr/s_lock.c`

Lines changed: 3 additions & 3 deletions

Original file line number	Diff line number	Diff line change
`@@ -9,7 +9,7 @@`
`9`	`9`	`*`
`10`	`10`	`*`
`11`	`11`	`* IDENTIFICATION`
`12`		`- * $PostgreSQL: pgsql/src/backend/storage/lmgr/s_lock.c,v 1.26 2004/03/15 16:18:43 momjian Exp $`
	`12`	`+ * $PostgreSQL: pgsql/src/backend/storage/lmgr/s_lock.c,v 1.27 2004/03/23 21:39:46 momjian Exp $`
`13`	`13`	`*`
`14`	`14`	`*-------------------------------------------------------------------------`
`15`	`15`	`*/`
`@@ -45,15 +45,15 @@ void`
`45`	`45`	`s_lock(volatileslock_tlock,constcharfile,intline)`
`46`	`46`	`{`
`47`	`47`	`/*`
`48`		`- * We loop tightly for awhile, then delay usingselect() and try`
	`48`	`+ * We loop tightly for awhile, then delay usingpg_usleep() and try`
`49`	`49`	`* again. Preferably, "awhile" should be a small multiple of the`
`50`	`50`	`* maximum time we expect a spinlock to be held. 100 iterations seems`
`51`	`51`	`* about right. In most multi-CPU scenarios, the spinlock is probably`
`52`	`52`	`* held by a process on another CPU and will be released before we`
`53`	`53`	`* finish 100 iterations. However, on a uniprocessor, the tight loop`
`54`	`54`	`* is just a waste of cycles, so don't iterate thousands of times.`
`55`	`55`	`*`
`56`		`- * Once we do decide to block, we use randomly increasingselect()`
	`56`	`+ * Once we do decide to block, we use randomly increasingpg_usleep()`
`57`	`57`	`* delays. The first delay is 10 msec, then the delay randomly`
`58`	`58`	`* increases to about one second, after which we reset to 10 msec and`
`59`	`59`	`* start again. The idea here is that in the presence of heavy`

0 commit comments

Comments

(0)

Movatterモバイル変換

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commit48b2802

File tree

1 file changed

1 file changed

`‎src/backend/storage/lmgr/s_lock.c`

0 commit comments