[PATCH v2 02/10] app/test: fix undefined behaviour in red autotest

[PATCH v2 02/10] app/test: fix undefined behaviour in red autotest

[Bruce Richardson]

The shift of a negative number (or very large positive) is undefined
behaviour which causes errors when run with UBSan. Fix this by making
the behaviour explicit for the edge case of n being zero in the
calculation.

Fixes: de3cfa2c9823 ("sched: initial import")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 lib/sched/rte_red.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/lib/sched/rte_red.h b/lib/sched/rte_red.h
index e62abb9295..3b90cc46a9 100644
--- a/lib/sched/rte_red.h
+++ b/lib/sched/rte_red.h
@@ -172,8 +172,15 @@ __rte_red_calc_qempty_factor(uint8_t wq_log2, uint16_t m)
 	f = (n >> 6) & 0xf;
 	n >>= 10;
 
-	if (n < RTE_RED_SCALING)
+	if (n < RTE_RED_SCALING) {
+		/* When n == 0, no rounding or shifting needed.
+		 * For n > 0, add 2^(n-1) for rounding before right shift.
+		 * This avoids UB from (1 << -1) when n == 0.
+		 */
+		if (n == 0)
+			return (uint16_t) rte_red_pow2_frac_inv[f];
 		return (uint16_t) ((rte_red_pow2_frac_inv[f] + (1 << (n - 1))) >> n);
+	}
 
 	return 0;
 }
-- 
2.51.0

[PATCH v2 04/10] app/test: fix timer loop hang on secondary process failure

[Bruce Richardson]

When the secondary process run from timer_secontary_autotest fails, the
timer loop is never stopped so the whole process hangs until timeout.
Fix this by setting the stop flag before checking for success or failure
of the secondary process.

Fixes: 50247fe03fe0 ("test/timer: exercise new APIs in secondary process")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 app/test/test_timer_secondary.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/app/test/test_timer_secondary.c b/app/test/test_timer_secondary.c
index ee2675a821..8580880a56 100644
--- a/app/test/test_timer_secondary.c
+++ b/app/test/test_timer_secondary.c
@@ -160,11 +160,12 @@ test_timer_secondary(void)
 		TEST_ASSERT_SUCCESS(ret, "Failed to launch timer manage loop");
 
 		ret = timer_secondary_spawn_wait(*sec_lcorep);
-		TEST_ASSERT_SUCCESS(ret, "Secondary process execution failed");
+		/* must set exit flag even on error case, so check ret later */
 
 		rte_delay_ms(500);
-
 		test_info->exit_flag = 1;
+
+		TEST_ASSERT_SUCCESS(ret, "Secondary process execution failed");
 		rte_eal_wait_lcore(*mgr_lcorep);
 
 #ifdef RTE_LIBRTE_TIMER_DEBUG
-- 
2.51.0

[PATCH v2 10/10] buildtools/get-test-suites: suppress empty output

[Bruce Richardson]

If there are no tests which are not assigned to a test suite, omit the
line printing the non_suite_tests. This avoid having error messages from
meson about test "" not being in any test suite.

Fixes: 25065ef1f6c0 ("test: emit warning for orphaned tests")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 buildtools/get-test-suites.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/buildtools/get-test-suites.py b/buildtools/get-test-suites.py
index 73f1d9c9e1..36ed4bedc2 100644
--- a/buildtools/get-test-suites.py
+++ b/buildtools/get-test-suites.py
@@ -38,4 +38,5 @@ def get_fast_test_params(test_name, ln):
 
 for suite in test_suites.keys():
     print(f"{suite}={','.join(test_suites[suite])}")
-print(f"non_suite_tests={','.join(non_suite_tests)}")
+if non_suite_tests:
+    print(f"non_suite_tests={','.join(non_suite_tests)}")
-- 
2.51.0

RE: [PATCH v2 02/10] app/test: fix undefined behaviour in red autotest

[Marat Khalili]

> -----Original Message-----
> From: Bruce Richardson <bruce.richardson@intel.com>
> Sent: Thursday 4 December 2025 18:21
> To: dev@dpdk.org
> Cc: Bruce Richardson <bruce.richardson@intel.com>; stable@dpdk.org
> Subject: [PATCH v2 02/10] app/test: fix undefined behaviour in red autotest
> 
> The shift of a negative number (or very large positive) is undefined
> behaviour which causes errors when run with UBSan. Fix this by making
> the behaviour explicit for the edge case of n being zero in the
> calculation.
> 
> Fixes: de3cfa2c9823 ("sched: initial import")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
> ---
>  lib/sched/rte_red.h | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/lib/sched/rte_red.h b/lib/sched/rte_red.h
> index e62abb9295..3b90cc46a9 100644
> --- a/lib/sched/rte_red.h
> +++ b/lib/sched/rte_red.h
> @@ -172,8 +172,15 @@ __rte_red_calc_qempty_factor(uint8_t wq_log2, uint16_t m)
>  	f = (n >> 6) & 0xf;
>  	n >>= 10;
> 
> -	if (n < RTE_RED_SCALING)
> +	if (n < RTE_RED_SCALING) {
> +		/* When n == 0, no rounding or shifting needed.
> +		 * For n > 0, add 2^(n-1) for rounding before right shift.
> +		 * This avoids UB from (1 << -1) when n == 0.
> +		 */
> +		if (n == 0)
> +			return (uint16_t) rte_red_pow2_frac_inv[f];
>  		return (uint16_t) ((rte_red_pow2_frac_inv[f] + (1 << (n - 1))) >> n);
> +	}
> 
>  	return 0;
>  }
> --
> 2.51.0

Acked-by: Marat Khalili <marat.khalili@huawei.com>

RE: [PATCH v2 10/10] buildtools/get-test-suites: suppress empty output

[Marat Khalili]

> -----Original Message-----
> From: Bruce Richardson <bruce.richardson@intel.com>
> Sent: Thursday 4 December 2025 18:21
> To: dev@dpdk.org
> Cc: Bruce Richardson <bruce.richardson@intel.com>; stable@dpdk.org
> Subject: [PATCH v2 10/10] buildtools/get-test-suites: suppress empty output
> 
> If there are no tests which are not assigned to a test suite, omit the
> line printing the non_suite_tests. This avoid having error messages from
> meson about test "" not being in any test suite.
> 
> Fixes: 25065ef1f6c0 ("test: emit warning for orphaned tests")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
> ---
>  buildtools/get-test-suites.py | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/buildtools/get-test-suites.py b/buildtools/get-test-suites.py
> index 73f1d9c9e1..36ed4bedc2 100644
> --- a/buildtools/get-test-suites.py
> +++ b/buildtools/get-test-suites.py
> @@ -38,4 +38,5 @@ def get_fast_test_params(test_name, ln):
> 
>  for suite in test_suites.keys():
>      print(f"{suite}={','.join(test_suites[suite])}")
> -print(f"non_suite_tests={','.join(non_suite_tests)}")
> +if non_suite_tests:
> +    print(f"non_suite_tests={','.join(non_suite_tests)}")
> --
> 2.51.0

Acked-by: Marat Khalili <marat.khalili@huawei.com>

[PATCH v3 02/11] app/test: fix undefined behaviour in red autotest

[Bruce Richardson]

The shift of a negative number (or very large positive) is undefined
behaviour which causes errors when run with UBSan. Fix this by making
the behaviour explicit for the edge case of n being zero in the
calculation.

Fixes: de3cfa2c9823 ("sched: initial import")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Marat Khalili <marat.khalili@huawei.com>
---
 lib/sched/rte_red.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/lib/sched/rte_red.h b/lib/sched/rte_red.h
index e62abb9295..3b90cc46a9 100644
--- a/lib/sched/rte_red.h
+++ b/lib/sched/rte_red.h
@@ -172,8 +172,15 @@ __rte_red_calc_qempty_factor(uint8_t wq_log2, uint16_t m)
 	f = (n >> 6) & 0xf;
 	n >>= 10;
 
-	if (n < RTE_RED_SCALING)
+	if (n < RTE_RED_SCALING) {
+		/* When n == 0, no rounding or shifting needed.
+		 * For n > 0, add 2^(n-1) for rounding before right shift.
+		 * This avoids UB from (1 << -1) when n == 0.
+		 */
+		if (n == 0)
+			return (uint16_t) rte_red_pow2_frac_inv[f];
 		return (uint16_t) ((rte_red_pow2_frac_inv[f] + (1 << (n - 1))) >> n);
+	}
 
 	return 0;
 }
-- 
2.51.0

[PATCH v3 04/11] app/test: fix timer loop hang on secondary process failure

[Bruce Richardson]

When the secondary process run from timer_secontary_autotest fails, the
timer loop is never stopped so the whole process hangs until timeout.
Fix this by setting the stop flag before checking for success or failure
of the secondary process.

Fixes: 50247fe03fe0 ("test/timer: exercise new APIs in secondary process")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 app/test/test_timer_secondary.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/app/test/test_timer_secondary.c b/app/test/test_timer_secondary.c
index ee2675a821..8580880a56 100644
--- a/app/test/test_timer_secondary.c
+++ b/app/test/test_timer_secondary.c
@@ -160,11 +160,12 @@ test_timer_secondary(void)
 		TEST_ASSERT_SUCCESS(ret, "Failed to launch timer manage loop");
 
 		ret = timer_secondary_spawn_wait(*sec_lcorep);
-		TEST_ASSERT_SUCCESS(ret, "Secondary process execution failed");
+		/* must set exit flag even on error case, so check ret later */
 
 		rte_delay_ms(500);
-
 		test_info->exit_flag = 1;
+
+		TEST_ASSERT_SUCCESS(ret, "Secondary process execution failed");
 		rte_eal_wait_lcore(*mgr_lcorep);
 
 #ifdef RTE_LIBRTE_TIMER_DEBUG
-- 
2.51.0

[PATCH v3 10/11] buildtools/get-test-suites: suppress empty output

[Bruce Richardson]

If there are no tests which are not assigned to a test suite, omit the
line printing the non_suite_tests. This avoid having error messages from
meson about test "" not being in any test suite.

Fixes: 25065ef1f6c0 ("test: emit warning for orphaned tests")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Marat Khalili <marat.khalili@huawei.com>
---
 buildtools/get-test-suites.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/buildtools/get-test-suites.py b/buildtools/get-test-suites.py
index 73f1d9c9e1..36ed4bedc2 100644
--- a/buildtools/get-test-suites.py
+++ b/buildtools/get-test-suites.py
@@ -38,4 +38,5 @@ def get_fast_test_params(test_name, ln):
 
 for suite in test_suites.keys():
     print(f"{suite}={','.join(test_suites[suite])}")
-print(f"non_suite_tests={','.join(non_suite_tests)}")
+if non_suite_tests:
+    print(f"non_suite_tests={','.join(non_suite_tests)}")
-- 
2.51.0

[PATCH v4 02/11] app/test: fix undefined behaviour in red autotest

[Bruce Richardson]

The shift of a negative number (or very large positive) is undefined
behaviour which causes errors when run with UBSan. Fix this by making
the behaviour explicit for the edge case of n being zero in the
calculation.

Fixes: de3cfa2c9823 ("sched: initial import")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Marat Khalili <marat.khalili@huawei.com>
---
 lib/sched/rte_red.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/lib/sched/rte_red.h b/lib/sched/rte_red.h
index e62abb9295..3b90cc46a9 100644
--- a/lib/sched/rte_red.h
+++ b/lib/sched/rte_red.h
@@ -172,8 +172,15 @@ __rte_red_calc_qempty_factor(uint8_t wq_log2, uint16_t m)
 	f = (n >> 6) & 0xf;
 	n >>= 10;
 
-	if (n < RTE_RED_SCALING)
+	if (n < RTE_RED_SCALING) {
+		/* When n == 0, no rounding or shifting needed.
+		 * For n > 0, add 2^(n-1) for rounding before right shift.
+		 * This avoids UB from (1 << -1) when n == 0.
+		 */
+		if (n == 0)
+			return (uint16_t) rte_red_pow2_frac_inv[f];
 		return (uint16_t) ((rte_red_pow2_frac_inv[f] + (1 << (n - 1))) >> n);
+	}
 
 	return 0;
 }
-- 
2.51.0

[PATCH v4 04/11] app/test: fix timer loop hang on secondary process failure

[Bruce Richardson]

When the secondary process run from timer_secontary_autotest fails, the
timer loop is never stopped so the whole process hangs until timeout.
Fix this by setting the stop flag before checking for success or failure
of the secondary process.

Fixes: 50247fe03fe0 ("test/timer: exercise new APIs in secondary process")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
 app/test/test_timer_secondary.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/app/test/test_timer_secondary.c b/app/test/test_timer_secondary.c
index 864f175d6a..fb2cc43895 100644
--- a/app/test/test_timer_secondary.c
+++ b/app/test/test_timer_secondary.c
@@ -160,11 +160,12 @@ test_timer_secondary(void)
 		TEST_ASSERT_SUCCESS(ret, "Failed to launch timer manage loop");
 
 		ret = timer_secondary_spawn_wait(*sec_lcorep);
-		TEST_ASSERT_SUCCESS(ret, "Secondary process execution failed");
+		/* must set exit flag even on error case, so check ret later */
 
 		rte_delay_ms(500);
-
 		test_info->exit_flag = 1;
+
+		TEST_ASSERT_SUCCESS(ret, "Secondary process execution failed");
 		rte_eal_wait_lcore(*mgr_lcorep);
 
 #ifdef RTE_LIBRTE_TIMER_DEBUG
-- 
2.51.0

[PATCH v4 10/11] buildtools/get-test-suites: suppress empty output

[Bruce Richardson]

If there are no tests which are not assigned to a test suite, omit the
line printing the non_suite_tests. This avoid having error messages from
meson about test "" not being in any test suite.

Fixes: 25065ef1f6c0 ("test: emit warning for orphaned tests")
Cc: stable@dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Marat Khalili <marat.khalili@huawei.com>
---
 buildtools/get-test-suites.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/buildtools/get-test-suites.py b/buildtools/get-test-suites.py
index 73f1d9c9e1..36ed4bedc2 100644
--- a/buildtools/get-test-suites.py
+++ b/buildtools/get-test-suites.py
@@ -38,4 +38,5 @@ def get_fast_test_params(test_name, ln):
 
 for suite in test_suites.keys():
     print(f"{suite}={','.join(test_suites[suite])}")
-print(f"non_suite_tests={','.join(non_suite_tests)}")
+if non_suite_tests:
+    print(f"non_suite_tests={','.join(non_suite_tests)}")
-- 
2.51.0