[FFmpeg-devel] [PATCH] tests/checkasm/svq1enc: Use proper range for input

[FFmpeg-devel] [PATCH] tests/checkasm/svq1enc: Use proper range for input

[Andreas Rheinhardt]

ssd_int8_vs_int16 is only called from encode_block()
in svq1enc.c; it calls it in stages: At stage 0,
the int16_t array contains the difference of two
uint16_t. At each of the following stages, the
int16_t array is filled by subtracting an int8_t from
the current stage's int16_t array. The maximum stage
is five, so the int16_t are in the range
(-255 + 5 * 127)..(255 + 5 * 128).

This commit modifies the checkasm test to only use
values from this range, fixing (undefined) integer overflow
in the test.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
---
 tests/checkasm/svq1enc.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/tests/checkasm/svq1enc.c b/tests/checkasm/svq1enc.c
index 1a6f531141..f9abdcbff8 100644
--- a/tests/checkasm/svq1enc.c
+++ b/tests/checkasm/svq1enc.c
@@ -18,7 +18,6 @@
  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
  */
 
-#include "libavutil/mem.h"
 #include "libavutil/mem_internal.h"
 
 #include "libavcodec/svq1encdsp.h"
@@ -26,11 +25,13 @@
 #include "checkasm.h"
 
 #define BUF_SIZE 1024
+#define MIN_VAL (-255 - 5 * 127)
+#define MAX_VAL ( 255 + 5 * 128)
 
 #define randomize(buf, len) \
     do { \
         for (int i = 0; i < len; i++) \
-            buf[i] = ((rnd() % 65281) - 32641); \
+            buf[i] = ((rnd() % (MAX_VAL - MIN_VAL + 1)) + MIN_VAL); \
     } while (0)
 
 static void test_ssd_int8_vs_int16(SVQ1EncDSPContext *s) {
-- 
2.40.1

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH] tests/checkasm/svq1enc: Use proper range for input

[Andreas Rheinhardt]

Andreas Rheinhardt:
> ssd_int8_vs_int16 is only called from encode_block()
> in svq1enc.c; it calls it in stages: At stage 0,
> the int16_t array contains the difference of two
> uint16_t. At each of the following stages, the
> int16_t array is filled by subtracting an int8_t from
> the current stage's int16_t array. The maximum stage
> is five, so the int16_t are in the range
> (-255 + 5 * 127)..(255 + 5 * 128).
> 
> This commit modifies the checkasm test to only use
> values from this range, fixing (undefined) integer overflow
> in the test.
> 
> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
> ---
>  tests/checkasm/svq1enc.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/checkasm/svq1enc.c b/tests/checkasm/svq1enc.c
> index 1a6f531141..f9abdcbff8 100644
> --- a/tests/checkasm/svq1enc.c
> +++ b/tests/checkasm/svq1enc.c
> @@ -18,7 +18,6 @@
>   * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
>   */
>  
> -#include "libavutil/mem.h"
>  #include "libavutil/mem_internal.h"
>  
>  #include "libavcodec/svq1encdsp.h"
> @@ -26,11 +25,13 @@
>  #include "checkasm.h"
>  
>  #define BUF_SIZE 1024
> +#define MIN_VAL (-255 - 5 * 127)
> +#define MAX_VAL ( 255 + 5 * 128)
>  
>  #define randomize(buf, len) \
>      do { \
>          for (int i = 0; i < len; i++) \
> -            buf[i] = ((rnd() % 65281) - 32641); \
> +            buf[i] = ((rnd() % (MAX_VAL - MIN_VAL + 1)) + MIN_VAL); \
>      } while (0)
>  
>  static void test_ssd_int8_vs_int16(SVQ1EncDSPContext *s) {

Will apply tomorrow unless there are objections.

- Andreas

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH] tests/checkasm/svq1enc: Use proper range for input

[Tomas Härdin]

mån 2024-05-06 klockan 17:41 +0200 skrev Andreas Rheinhardt:
> ssd_int8_vs_int16 is only called from encode_block()
> in svq1enc.c; it calls it in stages: At stage 0,
> the int16_t array contains the difference of two
> uint16_t. At each of the following stages, the
> int16_t array is filled by subtracting an int8_t from
> the current stage's int16_t array. The maximum stage
> is five, so the int16_t are in the range
> (-255 + 5 * 127)..(255 + 5 * 128).
> 
> This commit modifies the checkasm test to only use
> values from this range, fixing (undefined) integer overflow
> in the test.
> 
> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
> ---
>  tests/checkasm/svq1enc.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/checkasm/svq1enc.c b/tests/checkasm/svq1enc.c
> index 1a6f531141..f9abdcbff8 100644
> --- a/tests/checkasm/svq1enc.c
> +++ b/tests/checkasm/svq1enc.c
> @@ -18,7 +18,6 @@
>   * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
>   */
>  
> -#include "libavutil/mem.h"
>  #include "libavutil/mem_internal.h"
>  
>  #include "libavcodec/svq1encdsp.h"
> @@ -26,11 +25,13 @@
>  #include "checkasm.h"
>  
>  #define BUF_SIZE 1024
> +#define MIN_VAL (-255 - 5 * 127)
> +#define MAX_VAL ( 255 + 5 * 128)
>  
>  #define randomize(buf, len) \
>      do { \
>          for (int i = 0; i < len; i++) \
> -            buf[i] = ((rnd() % 65281) - 32641); \
> +            buf[i] = ((rnd() % (MAX_VAL - MIN_VAL + 1)) + MIN_VAL);

This is why formal methods are a good thing

Looks OK of course

/Tomas
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".