[FFmpeg-devel] [PATCH] avocdec/flac_parser: another fix

[FFmpeg-devel] [PATCH] avocdec/flac_parser: another fix

[Paul B Mahol]

[-- Attachment #1: Type: text/plain, Size: 16 bytes --]

Patch attached.

[-- Attachment #2: 0001-avcodec-flac_parser-add-missed-opportunity-to-check-.patch --]
[-- Type: text/x-patch, Size: 1759 bytes --]

From 01e931d2c7f50727d8893268cdc3ae0dbe75c250 Mon Sep 17 00:00:00 2001
From: Paul B Mahol <onemda@gmail.com>
Date: Mon, 5 Sep 2022 20:16:13 +0200
Subject: [PATCH] avcodec/flac_parser: add missed opportunity to check crc

Fixes #9621

Signed-off-by: Paul B Mahol <onemda@gmail.com>
---
 libavcodec/flac_parser.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/libavcodec/flac_parser.c b/libavcodec/flac_parser.c
index 11cd5540cf..dd70721696 100644
--- a/libavcodec/flac_parser.c
+++ b/libavcodec/flac_parser.c
@@ -455,7 +455,7 @@ static int check_header_mismatch(FLACParseContext  *fpc,
                                  int                log_level_offset)
 {
     FLACFrameInfo  *header_fi = &header->fi, *child_fi = &child->fi;
-    int deduction, deduction_expected = 0, i;
+    int check_crc, deduction, deduction_expected = 0, i;
     deduction = check_header_fi_mismatch(fpc, header_fi, child_fi,
                                          log_level_offset);
     /* Check sample and frame numbers. */
@@ -491,8 +491,15 @@ static int check_header_mismatch(FLACParseContext  *fpc,
                    "sample/frame number mismatch in adjacent frames\n");
     }
 
+    if ((fpc->last_fi.frame_or_sample_num + 1 == header_fi->frame_or_sample_num) ||
+        (fpc->last_fi.frame_or_sample_num + fpc->last_fi.blocksize == header_fi->frame_or_sample_num)) {
+        check_crc = 0;
+    } else {
+        check_crc = !deduction && !deduction_expected;
+    }
+
     /* If we have suspicious headers, check the CRC between them */
-    if (deduction && !deduction_expected) {
+    if (check_crc || (deduction && !deduction_expected)) {
         FLACHeaderMarker *curr;
         int read_len;
         uint8_t *buf;
-- 
2.37.2


[-- Attachment #3: Type: text/plain, Size: 251 bytes --]

_______________________________________________
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] avocdec/flac_parser: another fix

[Mattias Wadman]

On Mon, Sep 5, 2022 at 8:16 PM Paul B Mahol <onemda@gmail.com> wrote:

> Patch attached.
>

Thanks and can confirm that the patch produces the same samples as the flac
reference decoder for the original file in
https://trac.ffmpeg.org/ticket/9621 that I could not share.

But I'm not sure I follow how the patch works. I read it as we skip
checking CRC if the current frame or sample number is the expected next one?

+    if ((fpc->last_fi.frame_or_sample_num + 1 ==
header_fi->frame_or_sample_num) ||
+        (fpc->last_fi.frame_or_sample_num + fpc->last_fi.blocksize ==
header_fi->frame_or_sample_num)) {

Would it make sense to look at the blocking strategy bit to know which one
to check?
_______________________________________________
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] avocdec/flac_parser: another fix

[Paul B Mahol]

On Wed, Sep 7, 2022 at 12:31 PM Mattias Wadman <mattias.wadman@gmail.com>
wrote:

> On Mon, Sep 5, 2022 at 8:16 PM Paul B Mahol <onemda@gmail.com> wrote:
>
> > Patch attached.
> >
>
> Thanks and can confirm that the patch produces the same samples as the flac
> reference decoder for the original file in
> https://trac.ffmpeg.org/ticket/9621 that I could not share.
>
> But I'm not sure I follow how the patch works. I read it as we skip
> checking CRC if the current frame or sample number is the expected next
> one?
>

It always force CRC check if they do not match, previously code would do
that sometimes.


>
> +    if ((fpc->last_fi.frame_or_sample_num + 1 ==
> header_fi->frame_or_sample_num) ||
> +        (fpc->last_fi.frame_or_sample_num + fpc->last_fi.blocksize ==
> header_fi->frame_or_sample_num)) {
>
> Would it make sense to look at the blocking strategy bit to know which one
> to check?
>

 Could be.

_______________________________________________
> 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".
>
_______________________________________________
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".