Clarify the description a bit, and document the Linux data integrity guarantees. Signed-off-by: Christoph Hellwig <hch@xxxxxx> diff --git a/man2/sync.2 b/man2/sync.2 index 5e222cb..28a42c4 100644 --- a/man2/sync.2 +++ b/man2/sync.2 @@ -35,7 +35,7 @@ .\" .TH SYNC 2 2015-08-08 "Linux" "Linux Programmer's Manual" .SH NAME -sync, syncfs \- commit buffer cache to disk +sync, syncfs \- commit filesystem caches to disk .SH SYNOPSIS .B #include <unistd.h> .sp @@ -64,8 +64,8 @@ _GNU_SOURCE .ad .SH DESCRIPTION .BR sync () -causes all buffered modifications to file metadata and data to be -written to the underlying filesystems. +causes all pending modifications to file system metadata and cached file +data to be written to the underlying filesystems. .BR syncfs () is like @@ -108,16 +108,21 @@ In glibc 2.2.1 and earlier, it was "int sync(void)", and .BR sync () always returned 0. -.SH BUGS + According to the standard specification (e.g., POSIX.1-2001), .BR sync () schedules the writes, but may return before the actual -writing is done. -However, since version 1.3.20 Linux does actually wait. -(This still does not guarantee data integrity: modern disks have -large caches.) +writing is done. However Linux waits for I/O completions, +and thus +.BR sync () +or +.BR syncfs () +provide the same guarantees as fsync called on every file in +the system or filesystem respectively. +.SH BUGS +Before version 1.3.20 Linux did not wait for I/O to complete +before returning. .SH SEE ALSO -.BR bdflush (2), .BR fdatasync (2), .BR fsync (2), .BR sync (1) -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
