Normally, tailq entry should have a valid fd by the time we attempt
to map the segment. However, in case it doesn't, we're leaking fd,
so fix it.
Coverity issue: 272570
Fixes: 2a04139f66b4 ("eal: add single file segments option")
Cc: anatoly.bura...@intel.com
Signed-off-by: Anatoly Burakov <anatoly.bura...@intel.com>
---
Notes:
v4:
- Unconditionally close fd on remove
- Clarify what happens if file is not removed
lib/librte_eal/linuxapp/eal/eal_memalloc.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_memalloc.c
b/lib/librte_eal/linuxapp/eal/eal_memalloc.c
index 5ea6dd3..d366f0a 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memalloc.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memalloc.c
@@ -522,13 +522,14 @@ alloc_seg(struct rte_memseg *ms, void *addr, int
socket_id,
resize_hugefile(fd, map_offset, alloc_sz, false);
if (is_zero_length(fd)) {
struct msl_entry *te = get_msl_entry_by_idx(list_idx);
- if (te != NULL && te->fd >= 0) {
- close(te->fd);
+ /* te->fd is equivalent to fd */
+ if (te != NULL && te->fd >= 0)
te->fd = -1;
- }
/* ignore errors, can't make it any worse */
unlink(path);
+ close(fd);
}
+ /* if we're not removing the file, fd stays in the tailq */
} else {
close(fd);
unlink(path);
--
2.7.4
