| Message ID | 1537277849-14671-1-git-send-email-mypopydev@gmail.com |
|---|---|
| State | Accepted |
| Headers | show |
On Tue, Sep 18, 2018 at 09:37:29PM +0800, Jun Zhao wrote: > Fixes ticket #7441. > > Signed-off-by: Jun Zhao <mypopydev@gmail.com> > --- > libavfilter/vf_deshake.c | 12 +++++++----- > 1 files changed, 7 insertions(+), 5 deletions(-) this doesnt look correct the blocks that will be used are not going out of array? so the contrast calculated from them also should not [...]
On Wed, Sep 19, 2018 at 7:07 PM Michael Niedermayer <michael@niedermayer.cc> wrote: > > On Tue, Sep 18, 2018 at 09:37:29PM +0800, Jun Zhao wrote: > > Fixes ticket #7441. > > > > Signed-off-by: Jun Zhao <mypopydev@gmail.com> > > --- > > libavfilter/vf_deshake.c | 12 +++++++----- > > 1 files changed, 7 insertions(+), 5 deletions(-) > > this doesnt look correct > > the blocks that will be used are not going out of array? so the contrast > calculated from them also should not > Do you mean the correct way is changing the pos calculated from: pos = (y - i) * stride + (x - j); to pos = (y + i) * stride + (x + j) ?
On Thu, Sep 20, 2018 at 10:39:37AM +0800, mypopy@gmail.com wrote: > On Wed, Sep 19, 2018 at 7:07 PM Michael Niedermayer > <michael@niedermayer.cc> wrote: > > > > On Tue, Sep 18, 2018 at 09:37:29PM +0800, Jun Zhao wrote: > > > Fixes ticket #7441. > > > > > > Signed-off-by: Jun Zhao <mypopydev@gmail.com> > > > --- > > > libavfilter/vf_deshake.c | 12 +++++++----- > > > 1 files changed, 7 insertions(+), 5 deletions(-) > > > > this doesnt look correct > > > > the blocks that will be used are not going out of array? so the contrast > > calculated from them also should not > > > Do you mean the correct way is changing the pos calculated from: > pos = (y - i) * stride + (x - j); to pos = (y + i) * stride + (x + j) ? Iam not the author of vf_deshake, but whatever aligns the checked blocks correctly, should be the correct solution [...]
1) The crash is caused by accessing un-allocated memory area. 2) You can't compute the contrast outside of search window (by default, rx=ry=16).
On Sat, Sep 22, 2018 at 6:52 AM Colin NG <colin_ng@hotmail.com> wrote: > > 1) The crash is caused by accessing un-allocated memory area. > 2) You can't compute the contrast outside of search window (by default, rx=ry=16). > Yes, so I give other fix in https://patchwork.ffmpeg.org/patch/10426/, can you help to review or comment? Thanks. > ________________________________ > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> on behalf of Michael Niedermayer <michael@niedermayer.cc> > Sent: September 21, 2018 3:24 PM > To: FFmpeg development discussions and patches > Subject: Re: [FFmpeg-devel] [PATCH] lavfi/deshake: fix deshake crash issue. > > On Thu, Sep 20, 2018 at 10:39:37AM +0800, mypopy@gmail.com wrote: > > On Wed, Sep 19, 2018 at 7:07 PM Michael Niedermayer > > <michael@niedermayer.cc> wrote: > > > > > > On Tue, Sep 18, 2018 at 09:37:29PM +0800, Jun Zhao wrote: > > > > Fixes ticket #7441. > > > > > > > > Signed-off-by: Jun Zhao <mypopydev@gmail.com> > > > > --- > > > > libavfilter/vf_deshake.c | 12 +++++++----- > > > > 1 files changed, 7 insertions(+), 5 deletions(-) > > > > > > this doesnt look correct > > > > > > the blocks that will be used are not going out of array? so the contrast > > > calculated from them also should not > > > > > Do you mean the correct way is changing the pos calculated from: > > pos = (y - i) * stride + (x - j); to pos = (y + i) * stride + (x + j) ? > > Iam not the author of vf_deshake, but whatever aligns the checked blocks > correctly, should be the correct solution > > [...] > -- > Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB > > Modern terrorism, a quick summary: Need oil, start war with country that > has oil, kill hundread thousand in war. Let country fall into chaos, > be surprised about raise of fundamantalists. Drop more bombs, kill more > people, be surprised about them taking revenge and drop even more bombs > and strip your own citizens of their rights and freedoms. to be continued > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
On Sat, Sep 22, 2018 at 3:25 AM Michael Niedermayer <michael@niedermayer.cc> wrote: > > On Thu, Sep 20, 2018 at 10:39:37AM +0800, mypopy@gmail.com wrote: > > On Wed, Sep 19, 2018 at 7:07 PM Michael Niedermayer > > <michael@niedermayer.cc> wrote: > > > > > > On Tue, Sep 18, 2018 at 09:37:29PM +0800, Jun Zhao wrote: > > > > Fixes ticket #7441. > > > > > > > > Signed-off-by: Jun Zhao <mypopydev@gmail.com> > > > > --- > > > > libavfilter/vf_deshake.c | 12 +++++++----- > > > > 1 files changed, 7 insertions(+), 5 deletions(-) > > > > > > this doesnt look correct > > > > > > the blocks that will be used are not going out of array? so the contrast > > > calculated from them also should not > > > > > Do you mean the correct way is changing the pos calculated from: > > pos = (y - i) * stride + (x - j); to pos = (y + i) * stride + (x + j) ? > > Iam not the author of vf_deshake, but whatever aligns the checked blocks > correctly, should be the correct solution Agree, so I submitted V2 patch for this issue, Thanks. > > [...] > -- > Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB > > Modern terrorism, a quick summary: Need oil, start war with country that > has oil, kill hundread thousand in war. Let country fall into chaos, > be surprised about raise of fundamantalists. Drop more bombs, kill more > people, be surprised about them taking revenge and drop even more bombs > and strip your own citizens of their rights and freedoms. to be continued > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
diff --git a/libavfilter/vf_deshake.c b/libavfilter/vf_deshake.c index 55ce5e1..5d2c7fa 100644 --- a/libavfilter/vf_deshake.c +++ b/libavfilter/vf_deshake.c @@ -196,11 +196,13 @@ static int block_contrast(uint8_t *src, int x, int y, int stride, int blocksize) for (i = 0; i <= blocksize * 2; i++) { // We use a width of 16 here to match the sad function for (j = 0; j <= 15; j++) { - pos = (y - i) * stride + (x - j); - if (src[pos] < lowest) - lowest = src[pos]; - else if (src[pos] > highest) { - highest = src[pos]; + if (y >= i) { + pos = (y - i) * stride + (x - j); + if (src[pos] < lowest) + lowest = src[pos]; + else if (src[pos] > highest) { + highest = src[pos]; + } } } }
Fixes ticket #7441. Signed-off-by: Jun Zhao <mypopydev@gmail.com> --- libavfilter/vf_deshake.c | 12 +++++++----- 1 files changed, 7 insertions(+), 5 deletions(-)