From: Olivier Matz <olivier.matz@6wind.com>
To: dev@dpdk.org
Cc: Olivier Matz <olivier.matz@6wind.com>,
Lavanya Govindarajan <lavanyax.govindarajan@intel.com>,
Reshma Pattan <reshma.pattan@intel.com>,
David Marchand <david.marchand@redhat.com>,
stable@dpdk.org
Subject: [dpdk-dev] [PATCH] test/mbuf: fix access to freed memory
Date: Fri, 29 Oct 2021 14:15:44 +0200 [thread overview]
Message-ID: <20211029121544.13298-1-olivier.matz@6wind.com> (raw)
Seen by ASan.
In the external buffer mbuf test, we check that the buffer is freed
by checking that its refcount is 0. This is not a valid condition,
because it accesses to an already freed area.
Fix this by setting a boolean flag in the callback when rte_free()
is actually called, and check this flag instead.
Bugzilla ID: 867
Fixes: 7b295dceea07 ("test/mbuf: add unit test cases")
Cc: stable@dpdk.org
Reported-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
---
app/test/test_mbuf.c | 26 ++++++++++++++++++--------
1 file changed, 18 insertions(+), 8 deletions(-)
diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
index 94d1cdde37..f93bcef8a9 100644
--- a/app/test/test_mbuf.c
+++ b/app/test/test_mbuf.c
@@ -2307,16 +2307,16 @@ test_pktmbuf_read_from_chain(struct rte_mempool *pktmbuf_pool)
/* Define a free call back function to be used for external buffer */
static void
-ext_buf_free_callback_fn(void *addr __rte_unused, void *opaque)
+ext_buf_free_callback_fn(void *addr, void *opaque)
{
- void *ext_buf_addr = opaque;
+ bool *freed = opaque;
- if (ext_buf_addr == NULL) {
+ if (addr == NULL) {
printf("External buffer address is invalid\n");
return;
}
- rte_free(ext_buf_addr);
- ext_buf_addr = NULL;
+ rte_free(addr);
+ *freed = true;
printf("External buffer freed via callback\n");
}
@@ -2340,6 +2340,7 @@ test_pktmbuf_ext_shinfo_init_helper(struct rte_mempool *pktmbuf_pool)
void *ext_buf_addr = NULL;
uint16_t buf_len = EXT_BUF_TEST_DATA_LEN +
sizeof(struct rte_mbuf_ext_shared_info);
+ bool freed = false;
/* alloc a mbuf */
m = rte_pktmbuf_alloc(pktmbuf_pool);
@@ -2355,7 +2356,7 @@ test_pktmbuf_ext_shinfo_init_helper(struct rte_mempool *pktmbuf_pool)
GOTO_FAIL("%s: External buffer allocation failed\n", __func__);
ret_shinfo = rte_pktmbuf_ext_shinfo_init_helper(ext_buf_addr, &buf_len,
- ext_buf_free_callback_fn, ext_buf_addr);
+ ext_buf_free_callback_fn, &freed);
if (ret_shinfo == NULL)
GOTO_FAIL("%s: Shared info initialization failed!\n", __func__);
@@ -2388,26 +2389,35 @@ test_pktmbuf_ext_shinfo_init_helper(struct rte_mempool *pktmbuf_pool)
if (rte_mbuf_ext_refcnt_read(ret_shinfo) != 2)
GOTO_FAIL("%s: Invalid ext_buf ref_cnt\n", __func__);
+ if (freed)
+ GOTO_FAIL("%s: extbuf should not be freed\n", __func__);
/* test to manually update ext_buf_ref_cnt from 2 to 3*/
rte_mbuf_ext_refcnt_update(ret_shinfo, 1);
if (rte_mbuf_ext_refcnt_read(ret_shinfo) != 3)
GOTO_FAIL("%s: Update ext_buf ref_cnt failed\n", __func__);
+ if (freed)
+ GOTO_FAIL("%s: extbuf should not be freed\n", __func__);
/* reset the ext_refcnt before freeing the external buffer */
rte_mbuf_ext_refcnt_set(ret_shinfo, 2);
if (rte_mbuf_ext_refcnt_read(ret_shinfo) != 2)
GOTO_FAIL("%s: set ext_buf ref_cnt failed\n", __func__);
+ if (freed)
+ GOTO_FAIL("%s: extbuf should not be freed\n", __func__);
/* detach the external buffer from mbufs */
rte_pktmbuf_detach_extbuf(m);
/* check if ref cnt is decremented */
if (rte_mbuf_ext_refcnt_read(ret_shinfo) != 1)
GOTO_FAIL("%s: Invalid ext_buf ref_cnt\n", __func__);
+ if (freed)
+ GOTO_FAIL("%s: extbuf should not be freed\n", __func__);
rte_pktmbuf_detach_extbuf(clone);
- if (rte_mbuf_ext_refcnt_read(ret_shinfo) != 0)
- GOTO_FAIL("%s: Invalid ext_buf ref_cnt\n", __func__);
+ if (!freed)
+ GOTO_FAIL("%s: extbuf should be freed\n", __func__);
+ freed = false;
rte_pktmbuf_free(m);
m = NULL;
--
2.30.2
next reply other threads:[~2021-10-29 12:16 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-29 12:15 Olivier Matz [this message]
2021-11-03 15:00 ` David Marchand
2021-11-04 10:09 ` David Marchand
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20211029121544.13298-1-olivier.matz@6wind.com \
--to=olivier.matz@6wind.com \
--cc=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=lavanyax.govindarajan@intel.com \
--cc=reshma.pattan@intel.com \
--cc=stable@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).