From: Cosmin Ratiu <cratiu@nvidia.com>
To: <netdev@vger.kernel.org>, <cratiu@nvidia.com>
Cc: "David S . Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>,
Eric Dumazet <edumazet@google.com>,
Andrew Lunn <andrew+netdev@lunn.ch>,
Paolo Abeni <pabeni@redhat.com>, Joe Damato <jdamato@fastly.com>,
Shuah Khan <shuah@kernel.org>,
Stanislav Fomichev <sdf@fomichev.me>,
Mina Almasry <almasrymina@google.com>,
Saeed Mahameed <saeedm@nvidia.com>,
Tariq Toukan <tariqt@nvidia.com>,
Dragos Tatulea <dtatulea@nvidia.com>,
<linux-kselftest@vger.kernel.org>
Subject: [PATCH net v2] tests/ncdevmem: Fix double-free of queue array
Date: Thu, 8 May 2025 11:44:34 +0300 [thread overview]
Message-ID: <20250508084434.1933069-1-cratiu@nvidia.com> (raw)
netdev_bind_rx takes ownership of the queue array passed as parameter
and frees it, so a queue array buffer cannot be reused across multiple
netdev_bind_rx calls.
This commit fixes that by always passing in a newly created queue array
to all netdev_bind_rx calls in ncdevmem.
Fixes: 85585b4bc8d8 ("selftests: add ncdevmem, netcat for devmem TCP")
Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
---
.../selftests/drivers/net/hw/ncdevmem.c | 55 ++++++++-----------
1 file changed, 22 insertions(+), 33 deletions(-)
diff --git a/tools/testing/selftests/drivers/net/hw/ncdevmem.c b/tools/testing/selftests/drivers/net/hw/ncdevmem.c
index 2bf14ac2b8c6..9d48004ff1a1 100644
--- a/tools/testing/selftests/drivers/net/hw/ncdevmem.c
+++ b/tools/testing/selftests/drivers/net/hw/ncdevmem.c
@@ -431,6 +431,22 @@ static int parse_address(const char *str, int port, struct sockaddr_in6 *sin6)
return 0;
}
+static struct netdev_queue_id *create_queues(void)
+{
+ struct netdev_queue_id *queues;
+ size_t i = 0;
+
+ queues = calloc(num_queues, sizeof(*queues));
+ for (i = 0; i < num_queues; i++) {
+ queues[i]._present.type = 1;
+ queues[i]._present.id = 1;
+ queues[i].type = NETDEV_QUEUE_TYPE_RX;
+ queues[i].id = start_queue + i;
+ }
+
+ return queues;
+}
+
int do_server(struct memory_buffer *mem)
{
char ctrl_data[sizeof(int) * 20000];
@@ -448,7 +464,6 @@ int do_server(struct memory_buffer *mem)
char buffer[256];
int socket_fd;
int client_fd;
- size_t i = 0;
int ret;
ret = parse_address(server_ip, atoi(port), &server_sin);
@@ -471,16 +486,7 @@ int do_server(struct memory_buffer *mem)
sleep(1);
- queues = malloc(sizeof(*queues) * num_queues);
-
- for (i = 0; i < num_queues; i++) {
- queues[i]._present.type = 1;
- queues[i]._present.id = 1;
- queues[i].type = NETDEV_QUEUE_TYPE_RX;
- queues[i].id = start_queue + i;
- }
-
- if (bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys))
+ if (bind_rx_queue(ifindex, mem->fd, create_queues(), num_queues, &ys))
error(1, 0, "Failed to bind\n");
tmp_mem = malloc(mem->size);
@@ -545,7 +551,6 @@ int do_server(struct memory_buffer *mem)
goto cleanup;
}
- i++;
for (cm = CMSG_FIRSTHDR(&msg); cm; cm = CMSG_NXTHDR(&msg, cm)) {
if (cm->cmsg_level != SOL_SOCKET ||
(cm->cmsg_type != SCM_DEVMEM_DMABUF &&
@@ -630,10 +635,8 @@ int do_server(struct memory_buffer *mem)
void run_devmem_tests(void)
{
- struct netdev_queue_id *queues;
struct memory_buffer *mem;
struct ynl_sock *ys;
- size_t i = 0;
mem = provider->alloc(getpagesize() * NUM_PAGES);
@@ -641,38 +644,24 @@ void run_devmem_tests(void)
if (configure_rss())
error(1, 0, "rss error\n");
- queues = calloc(num_queues, sizeof(*queues));
-
if (configure_headersplit(1))
error(1, 0, "Failed to configure header split\n");
- if (!bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys))
+ if (!bind_rx_queue(ifindex, mem->fd,
+ calloc(num_queues, sizeof(struct netdev_queue_id)),
+ num_queues, &ys))
error(1, 0, "Binding empty queues array should have failed\n");
- for (i = 0; i < num_queues; i++) {
- queues[i]._present.type = 1;
- queues[i]._present.id = 1;
- queues[i].type = NETDEV_QUEUE_TYPE_RX;
- queues[i].id = start_queue + i;
- }
-
if (configure_headersplit(0))
error(1, 0, "Failed to configure header split\n");
- if (!bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys))
+ if (!bind_rx_queue(ifindex, mem->fd, create_queues(), num_queues, &ys))
error(1, 0, "Configure dmabuf with header split off should have failed\n");
if (configure_headersplit(1))
error(1, 0, "Failed to configure header split\n");
- for (i = 0; i < num_queues; i++) {
- queues[i]._present.type = 1;
- queues[i]._present.id = 1;
- queues[i].type = NETDEV_QUEUE_TYPE_RX;
- queues[i].id = start_queue + i;
- }
-
- if (bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys))
+ if (bind_rx_queue(ifindex, mem->fd, create_queues(), num_queues, &ys))
error(1, 0, "Failed to bind\n");
/* Deactivating a bound queue should not be legal */
--
2.45.0
next reply other threads:[~2025-05-08 8:45 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-08 8:44 Cosmin Ratiu [this message]
2025-05-08 16:23 ` [PATCH net v2] tests/ncdevmem: Fix double-free of queue array Stanislav Fomichev
2025-05-08 18:31 ` Joe Damato
2025-05-09 7:21 ` Cosmin Ratiu
2025-05-08 20:42 ` Mina Almasry
2025-05-09 7:24 ` Cosmin Ratiu
2025-05-09 23:30 ` patchwork-bot+netdevbpf
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=20250508084434.1933069-1-cratiu@nvidia.com \
--to=cratiu@nvidia.com \
--cc=almasrymina@google.com \
--cc=andrew+netdev@lunn.ch \
--cc=davem@davemloft.net \
--cc=dtatulea@nvidia.com \
--cc=edumazet@google.com \
--cc=jdamato@fastly.com \
--cc=kuba@kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=saeedm@nvidia.com \
--cc=sdf@fomichev.me \
--cc=shuah@kernel.org \
--cc=tariqt@nvidia.com \
/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