On 1/24/2021 8:30 PM, Andreas Rheinhardt wrote:
James Almer:
On 1/24/2021 8:14 PM, Andreas Rheinhardt wrote:
James Almer:
It's been unused for almost three years now.
Signed-off-by: James Almer <jamr...@gmail.com>
---
doc/APIchanges | 4 ++++
libavcodec/avcodec.h | 7 ++++---
libavcodec/pthread_frame.c | 4 ++++
libavcodec/version.h | 10 +++++-----
4 files changed, 17 insertions(+), 8 deletions(-)
diff --git a/doc/APIchanges b/doc/APIchanges
index bbf56a5385..51f97b2759 100644
--- a/doc/APIchanges
+++ b/doc/APIchanges
@@ -15,6 +15,10 @@ libavutil: 2017-10-21
API changes, most recent first:
+2021-01-xx - xxxxxxxxxx - lavc 58.118.100 - avcodec.h
+ deprecate AVCodecContext.debug_mv, FF_DEBUG_VIS_MV_P_FOR,
FF_DEBUG_VIS_MV_B_FOR,
+ FF_DEBUG_VIS_MV_B_BACK
+
2021-01-11 - xxxxxxxxxx - lavc 58.116.100 - avcodec.h
Add FF_PROFILE_VVC_MAIN_10 and FF_PROFILE_VVC_MAIN_10_444.
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index dc8738c819..eafc2e84e3 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -2141,15 +2141,16 @@ typedef struct AVCodecContext {
*/
int seek_preroll;
+#if FF_API_DEBUG_MV
/**
- * debug motion vectors
- * - encoding: Set by user.
- * - decoding: Set by user.
+ * @deprecated unused
*/
+ attribute_deprecated
int debug_mv;
#define FF_DEBUG_VIS_MV_P_FOR 0x00000001 //visualize forward
predicted MVs of P frames
#define FF_DEBUG_VIS_MV_B_FOR 0x00000002 //visualize forward
predicted MVs of B frames
#define FF_DEBUG_VIS_MV_B_BACK 0x00000004 //visualize backward
predicted MVs of B frames
+#endif
/**
* custom intra quantization matrix
diff --git a/libavcodec/pthread_frame.c b/libavcodec/pthread_frame.c
index a570e25e0d..3abb691ac7 100644
--- a/libavcodec/pthread_frame.c
+++ b/libavcodec/pthread_frame.c
@@ -342,7 +342,11 @@ static int
update_context_from_user(AVCodecContext *dst, AVCodecContext *src)
dst->opaque = src->opaque;
dst->debug = src->debug;
+#if FF_API_DEBUG_MV
+FF_DISABLE_DEPRECATION_WARNINGS
dst->debug_mv = src->debug_mv;
+FF_ENABLE_DEPRECATION_WARNINGS
+#endif
If debug_mv is unused, then why don't you just remove the assignment?
It's a "set by user" field, so even if whatever value it may contain is
never used by libavcodec, removing this would be a change in behavior.
Admittedly harmless, but a change nonetheless, so I'm inclined to just
leave it there.
What user-visible behaviour would change?
Oh, these don't have the user context as dst (I recall now that Hendrik
implemented a separate callback for that purpose on his LAVFilters
fork). Changed locally to remove it, then.
dst->slice_flags = src->slice_flags;
dst->flags2 = src->flags2;
diff --git a/libavcodec/version.h b/libavcodec/version.h
index 9f80caa9e0..f3974c2cb3 100644
--- a/libavcodec/version.h
+++ b/libavcodec/version.h
@@ -28,8 +28,8 @@
#include "libavutil/version.h"
#define LIBAVCODEC_VERSION_MAJOR 58
-#define LIBAVCODEC_VERSION_MINOR 117
-#define LIBAVCODEC_VERSION_MICRO 101
+#define LIBAVCODEC_VERSION_MINOR 118
+#define LIBAVCODEC_VERSION_MICRO 100
#define LIBAVCODEC_VERSION_INT
AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \
LIBAVCODEC_VERSION_MINOR, \
@@ -54,9 +54,6 @@
#ifndef FF_API_LOWRES
#define FF_API_LOWRES (LIBAVCODEC_VERSION_MAJOR < 59)
#endif
-#ifndef FF_API_DEBUG_MV
-#define FF_API_DEBUG_MV (LIBAVCODEC_VERSION_MAJOR < 58)
-#endif
Why is this hunk not in the preceding patch?
I figured I'd leave the actual macro in place instead of removing it and
then re-adding it in a following commit, but i can change it if you
prefer that way.
Keeping it would make sense if you just replaced 58 by 60; but not if
you are moving this to the end anyway.
I don't see how one way makes sense but the other doesn't. It's all
cosmetic. But sure, changed locally to remove it in the first patch.
#ifndef FF_API_AVCTX_TIMEBASE
#define FF_API_AVCTX_TIMEBASE (LIBAVCODEC_VERSION_MAJOR < 59)
#endif
@@ -150,5 +147,8 @@
#ifndef FF_API_THREAD_SAFE_CALLBACKS
#define FF_API_THREAD_SAFE_CALLBACKS (LIBAVCODEC_VERSION_MAJOR < 60)
#endif
+#ifndef FF_API_DEBUG_MV
+#define FF_API_DEBUG_MV (LIBAVCODEC_VERSION_MAJOR < 60)
+#endif
#endif /* AVCODEC_VERSION_H */
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@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-requ...@ffmpeg.org with subject "unsubscribe".
