| Message ID | 44213cf8-c2dd-02b6-fe9d-08c8c8261ee9@carpalis.nl |
|---|---|
| State | Accepted |
| Commit | c0402e1e30c2318359af1bce1d652f4616b21566 |
| Headers | show |
On Fri, May 18, 2018 at 05:06:23PM +0200, Jerome Borsboom wrote: > For interlace field pictures s->mb_height indicates the height of the full > picture in MBs, i.e. the two fields combined. A single field is half this > size. When calculating mquant for interlace field pictures, the bottom edge > is the last MB row of the field. > > Signed-off-by: Jerome Borsboom <jerome.borsboom@carpalis.nl> > --- > libavcodec/vc1_block.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) for patch 1 you list a file that it improves. Are there testcase known that the other patches improve ? (such testcase would be usefull for fate for example) [...]
2018-05-20 0:01 GMT+02:00, Michael Niedermayer <michael@niedermayer.cc>: > On Fri, May 18, 2018 at 05:06:23PM +0200, Jerome Borsboom wrote: >> For interlace field pictures s->mb_height indicates the height of the >> full >> picture in MBs, i.e. the two fields combined. A single field is half this >> size. When calculating mquant for interlace field pictures, the bottom >> edge >> is the last MB row of the field. >> >> Signed-off-by: Jerome Borsboom <jerome.borsboom@carpalis.nl> >> --- >> libavcodec/vc1_block.c | 3 ++- >> 1 file changed, 2 insertions(+), 1 deletion(-) > > for patch 1 you list a file that it improves. The file - afaict - is related to the patchset as a whole. The current version of the patchset does not fix the file though, so I believe we should wait for a new version of the patchset. The sample is >20MB, most of the reference files are very small, so perhaps we can avoid using this one. Carl Eugen
>> On Fri, May 18, 2018 at 05:06:23PM +0200, Jerome Borsboom wrote: >>> For interlace field pictures s->mb_height indicates the height of the >>> full >>> picture in MBs, i.e. the two fields combined. A single field is half this >>> size. When calculating mquant for interlace field pictures, the bottom >>> edge >>> is the last MB row of the field. >>> >>> Signed-off-by: Jerome Borsboom <jerome.borsboom at carpalis.nl> >>> --- >>> libavcodec/vc1_block.c | 3 ++- >>> 1 file changed, 2 insertions(+), 1 deletion(-) >> >> for patch 1 you list a file that it improves. > > The file - afaict - is related to the patchset as a whole. > The current version of the patchset does not fix the > file though, so I believe we should wait for a new > version of the patchset. > > The sample is >20MB, most of the reference files are > very small, so perhaps we can avoid using this one. > > Carl Eugen The issue that remains is not related to this patch set. In my view, there is no need to wait for a new patch set. I will just submit a new patch for that issue. Regards, Jerome
2018-05-20 11:00 GMT+02:00, Jerome Borsboom <jerome.borsboom@carpalis.nl>: >>> for patch 1 you list a file that it improves. >> >> The file - afaict - is related to the patchset as a whole. >> The current version of the patchset does not fix the >> file though, so I believe we should wait for a new >> version of the patchset. > The issue that remains is not related to this patch set. In > my view, there is no need to wait for a new patch set. Thank you for clarifying this! Carl Eugen
2018-05-18 17:06 GMT+02:00, Jerome Borsboom <jerome.borsboom@carpalis.nl>: > For interlace field pictures s->mb_height indicates the height of the full > picture in MBs, i.e. the two fields combined. A single field is half this > size. When calculating mquant for interlace field pictures, the bottom edge > is the last MB row of the field. > > Signed-off-by: Jerome Borsboom <jerome.borsboom@carpalis.nl> > --- > libavcodec/vc1_block.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/libavcodec/vc1_block.c b/libavcodec/vc1_block.c > index daf30fdbfe..aa2ea5024e 100644 > --- a/libavcodec/vc1_block.c > +++ b/libavcodec/vc1_block.c > @@ -181,7 +181,8 @@ static void vc1_put_signed_blocks_clamped(VC1Context *v) > mquant = -v->altpq; \ > if ((edges&4) && s->mb_x == (s->mb_width - 1)) \ > mquant = -v->altpq; \ > - if ((edges&8) && s->mb_y == (s->mb_height - 1)) \ > + if ((edges&8) && \ > + s->mb_y == ((s->mb_height >> v->field_mode) - 1)) \ > mquant = -v->altpq; \ > if (!mquant || mquant > 31) { \ > av_log(v->s.avctx, AV_LOG_ERROR, \ Patch applied. Thank you, Carl Eugen
diff --git a/libavcodec/vc1_block.c b/libavcodec/vc1_block.c index daf30fdbfe..aa2ea5024e 100644 --- a/libavcodec/vc1_block.c +++ b/libavcodec/vc1_block.c @@ -181,7 +181,8 @@ static void vc1_put_signed_blocks_clamped(VC1Context *v) mquant = -v->altpq; \ if ((edges&4) && s->mb_x == (s->mb_width - 1)) \ mquant = -v->altpq; \ - if ((edges&8) && s->mb_y == (s->mb_height - 1)) \ + if ((edges&8) && \ + s->mb_y == ((s->mb_height >> v->field_mode) - 1)) \ mquant = -v->altpq; \ if (!mquant || mquant > 31) { \ av_log(v->s.avctx, AV_LOG_ERROR, \
For interlace field pictures s->mb_height indicates the height of the full picture in MBs, i.e. the two fields combined. A single field is half this size. When calculating mquant for interlace field pictures, the bottom edge is the last MB row of the field. Signed-off-by: Jerome Borsboom <jerome.borsboom@carpalis.nl> --- libavcodec/vc1_block.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)