Re: [libvirt] [PATCH 2/5] commandtest: Resolve some coverity resource leaks
On 02/14/13 17:42, John Ferlan wrote:
---
tests/commandtest.c | 17 ++++++++++++-----
1 file changed, 12 insertions(+), 5 deletions(-)
diff --git a/tests/commandtest.c b/tests/commandtest.c
index 93c6333..3bfc358 100644
--- a/tests/commandtest.c
+++ b/tests/commandtest.c
...
@@ -963,16 +964,19 @@ mymain(void)
dup2(fd, 6) < 0 ||
dup2(fd, 7) < 0 ||
dup2(fd, 8) < 0 ||
- (fd > 8 && VIR_CLOSE(fd) < 0))
+ (fd > 8 && VIR_CLOSE(fd) < 0)) {
+ VIR_FORCE_CLOSE(fd);
return EXIT_FAILURE;
+ }
+ sa_assert(fd == -1);
according to man of dup2():
* If oldfd is a valid file descriptor, and newfd has the same value
as oldfd, then dup2() does nothing, and returns newfd.
It is possible that open returns fd < 8, dup2() on that does nothing and
afterwards this assertion won't be true.
/* Prime the debug/verbose settings from the env vars,
* since we're about to reset 'environ' */
ignore_value(virTestGetDebug());
ignore_value(virTestGetVerbose());
- if (virInitialize() < 0)
- return EXIT_FAILURE;
+ /* Make sure to not leak fd's */
+ virinitret = virInitialize();
/* Phase two of killing interfering fds; see above. */
fd = 3;
ACK with the assertion removed or a sufficient explanation provided.
Peter