[PATCH] xfs_fsr: fix test for short write

Eric Sandeen sandeen at sandeen.net
Thu Sep 17 23:08:13 CDT 2009


Clang found this one too, as a dead nested assignment.

The point is to see if we write all the bytes; but the initial
assignment is backwards.  It's safe in the end, because if they
weren't already equal anyway, we'd have exited prior to this, but
it's worth fixing up for clarity.

Signed-of-by: Eric Sandeen <sandeen at sandeen.net>
---

clang record:
http://sandeen.net/clang/xfsprogs/2009-09-09-1/report-xOZEHP.html#EndPath

diff --git a/fsr/xfs_fsr.c b/fsr/xfs_fsr.c
index 9f2bf24..253a55d 100644
--- a/fsr/xfs_fsr.c
+++ b/fsr/xfs_fsr.c
@@ -1190,7 +1190,7 @@ packfile(char *fname, char *tname, int fd,
 			}
 			if (nfrags) {
 				/* Do a matching write to the tmp file */
-				wc = wc_b4;
+				wc_b4 = wc;
 				if (((wc = write(ffd, fbuf, wc)) != wc_b4)) {
 					fsrprintf(_("bad write of %d bytes "	
 						"to %s: %s\n"),




More information about the xfs mailing list