[PATCH 6/6] cmdline: add null checks for invalid input

[PATCH 6/6] cmdline: add null checks for invalid input

[Bruce Richardson]

To harden the public API, add some NULL checks before dereferencing
pointers. Properly written apps should never call these with NULL, but
to increase resilience, we'll add the checks.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 lib/cmdline/cmdline.c        | 3 +++
 lib/cmdline/cmdline_rdline.c | 3 +++
 2 files changed, 6 insertions(+)

diff --git a/lib/cmdline/cmdline.c b/lib/cmdline/cmdline.c
index d1003f0b8e..51fbc36cef 100644
--- a/lib/cmdline/cmdline.c
+++ b/lib/cmdline/cmdline.c
@@ -103,6 +103,9 @@ RTE_EXPORT_SYMBOL(cmdline_get_rdline)
 struct rdline*
 cmdline_get_rdline(struct cmdline *cl)
 {
+	if (!cl)
+		return NULL;
+
 	return &cl->rdl;
 }
 
diff --git a/lib/cmdline/cmdline_rdline.c b/lib/cmdline/cmdline_rdline.c
index 0a5a399b32..15da285c8d 100644
--- a/lib/cmdline/cmdline_rdline.c
+++ b/lib/cmdline/cmdline_rdline.c
@@ -618,6 +618,9 @@ RTE_EXPORT_SYMBOL(rdline_get_history_buffer_size)
 size_t
 rdline_get_history_buffer_size(struct rdline *rdl)
 {
+	if (!rdl)
+		return 0;
+
 	return sizeof(rdl->history_buf);
 }
 
-- 
2.51.0

Re: [PATCH 6/6] cmdline: add null checks for invalid input

[Alex Michael]

pmd_buffer_scatter test suite failed on Ubuntu 24.04 LTS with an Intel XL710 40GbE NIC. Conversely it didn’t fail on Ubuntu 22.04 with an Intel E810 NIC, so the issue lies either with the OS (mbuf size or mempool configuration mismatch / buffer handling logic), the NIC (driver-level bug), the interaction between them or the test suite itself (as it should be unrelated to NULL conditionals).

Re: [PATCH 6/6] cmdline: add null checks for invalid input

[Bruce Richardson]

On Thu, May 07, 2026 at 06:25:25PM -0400, Alex Michael wrote:
> pmd_buffer_scatter test suite failed on Ubuntu 24.04 LTS with an Intel XL710 40GbE NIC. Conversely it didn’t fail on Ubuntu 22.04 with an Intel E810 NIC, so the issue lies either with the OS (mbuf size or mempool configuration mismatch / buffer handling logic), the NIC (driver-level bug), the interaction between them or the test suite itself (as it should be unrelated to NULL conditionals).

Agree. The test failure seem unrelated to this patchset.