[libav-commits] Announce new releases: 10.4, 10.7
Module: libav.org Branch: master Commit: 805241cb26acc0eafcd5d8ed261d9f8959e94ee5 Author:Reinhard Tartler siret...@tauware.de Committer: Reinhard Tartler siret...@tauware.de Date: Sun May 31 13:29:39 2015 -0400 Announce new releases: 10.4, 10.7 --- src/download | 44 ++-- src/index| 20 ++-- src/news | 23 +++ 3 files changed, 55 insertions(+), 32 deletions(-) diff --git a/src/download b/src/download index fb1ca0e..904a556 100644 --- a/src/download +++ b/src/download @@ -29,45 +29,45 @@ this is the only place in which bnew development/b happens. a name=release_11/ah2Libav 11 One Louder/h2 p -11.3 was released on 2015-03-11. It is the latest point release from the 11 +11.4 was released on 2015-05-31. It is the latest point release from the 11 release series. 11 was published 2014-09-08, the 11 branch was cut 2014-08-20. Please give us feedback and use our a href=https://bugzilla.libav.org;Bugzilla/a for filing bugs. /p p -a href=releases/libav-11.3.tar.xzDownload XZ tarball/anbsp;nbsp; -a href=releases/libav-11.3.tar.xz.md5MD5/a -a href=releases/libav-11.3.tar.xz.sha1SHA1/a -a href=releases/libav-11.3.tar.xz.ascPGP signature/abr / -a href=releases/libav-11.3.tar.gzDownload gzip tarball/anbsp;nbsp; -a href=releases/libav-11.3.tar.gz.md5MD5/a -a href=releases/libav-11.3.tar.gz.sha1SHA1/a -a href=releases/libav-11.3.tar.gz.ascPGP signature/abr / -a href=releases/libav-11.3.changelogChangelog/abr / -a href=releases/libav-11.3.releaseRelease Notes/a +a href=releases/libav-11.4.tar.xzDownload XZ tarball/anbsp;nbsp; +a href=releases/libav-11.4.tar.xz.md5MD5/a +a href=releases/libav-11.4.tar.xz.sha1SHA1/a +a href=releases/libav-11.4.tar.xz.ascPGP signature/abr / +a href=releases/libav-11.4.tar.gzDownload gzip tarball/anbsp;nbsp; +a href=releases/libav-11.4.tar.gz.md5MD5/a +a href=releases/libav-11.4.tar.gz.sha1SHA1/a +a href=releases/libav-11.4.tar.gz.ascPGP signature/abr / +a href=releases/libav-11.4.changelogChangelog/abr / +a href=releases/libav-11.4.releaseRelease Notes/a /p a name=release_10/ah2Libav 10 Eks/h2 p -10.6 was released on 2015-03-11. It is the latest point release from the 10 +10.7 was released on 2015-05-31. It is the latest point release from the 10 release series. 10 was published 2014-03-23, the 10 branch was cut 2014-03-23. Please give us feedback and use our a href=//bugzilla.libav.orgBugzilla/a for filing bugs. /p p -a href=releases/libav-10.6.tar.xzDownload XZ tarball/anbsp;nbsp; -a href=releases/libav-10.6.tar.xz.md5MD5/a -a href=releases/libav-10.6.tar.xz.sha1SHA1/a -a href=releases/libav-10.6.tar.xz.ascPGP signature/abr / -a href=releases/libav-10.6.tar.gzDownload gzip tarball/anbsp;nbsp; -a href=releases/libav-10.6.tar.gz.md5MD5/a -a href=releases/libav-10.6.tar.gz.sha1SHA1/a -a href=releases/libav-10.6.tar.gz.ascPGP signature/abr / -a href=releases/libav-10.6.changelogChangelog/abr / -a href=releases/libav-10.6.releaseRelease Notes/a +a href=releases/libav-10.7.tar.xzDownload XZ tarball/anbsp;nbsp; +a href=releases/libav-10.7.tar.xz.md5MD5/a +a href=releases/libav-10.7.tar.xz.sha1SHA1/a +a href=releases/libav-10.7.tar.xz.ascPGP signature/abr / +a href=releases/libav-10.7.tar.gzDownload gzip tarball/anbsp;nbsp; +a href=releases/libav-10.7.tar.gz.md5MD5/a +a href=releases/libav-10.7.tar.gz.sha1SHA1/a +a href=releases/libav-10.7.tar.gz.ascPGP signature/abr / +a href=releases/libav-10.7.changelogChangelog/abr / +a href=releases/libav-10.7.releaseRelease Notes/a /p a name=release_9/ah2Libav 9 plain 9/h2 diff --git a/src/index b/src/index index 1bb9c6f..1059631 100644 --- a/src/index +++ b/src/index @@ -35,30 +35,30 @@ with the latest developments by subscribing to both the div class=panel-headingLatest Releases/div div class=list-group div href=# class=list-group-item -span class=badge alert-successMarch 11, 2015/span -h4 class=list-group-item-headingLibav 11.3/h4 +span class=badge alert-successMay 31, 2015/span +h4 class=list-group-item-headingLibav 11.4/h4 p class=list-group-item-text clearfix span class=pull-right a class=label label-info - href=releases/libav-11.3.releaseRelease Notes/a + href=releases/libav-11.4.releaseRelease Notes/a a class=label label-success - href=releases/libav-11.3.tar.xzxz/a + href=releases/libav-11.4.tar.xzxz/a a class=label label-primary - href=releases/libav-11.3.tar.gzgzip/a + href=releases/libav-11.4.tar.gzgzip/a /span /p /div div href=# class=list-group-item -span class=badge alert-successMarch 11, 2015/span -h4 class=list-group-item-headingLibav 10.6/h4 +span class=badge
[libav-commits] Tag v11.4 : 11.4 Release
Module: libav Branch: refs/tags/v11.4 Tag:3b7d396359134d26cb8175b707cd0a6a9cdd2734 Tagger: Reinhard Tartler siret...@tauware.de Date: Sun May 31 12:03:47 2015 -0400 11.4 Release ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] Tag v10.7 : 10.7 Release
Module: libav Branch: refs/tags/v10.7 Tag:d233127b0e75737d5561a4796f9f8df5e15eb2c3 Tagger: Reinhard Tartler siret...@tauware.de Date: Sun May 31 12:57:48 2015 -0400 10.7 Release ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] x86: Put COPY3_IF_LT under HAVE_6REGS
Module: libav Branch: release/10 Commit: 987a8f8514e07aba8136d491949d0b8d6bb183f9 Author:Luca Barbato lu_z...@gentoo.org Committer: Reinhard Tartler siret...@tauware.de Date: Mon Mar 16 11:26:48 2015 +0100 x86: Put COPY3_IF_LT under HAVE_6REGS It uses 6 registers, unbreaks building on hardened x86 system. Bug-Id: gentoo/541930 CC: libav-sta...@libav.org (cherry picked from commit 2af720fe5f0418612a8fc26b0147a0e10414fcbe) Signed-off-by: Reinhard Tartler siret...@tauware.de --- libavcodec/x86/mathops.h |5 + 1 file changed, 5 insertions(+) diff --git a/libavcodec/x86/mathops.h b/libavcodec/x86/mathops.h index a62094e..2c04d9d 100644 --- a/libavcodec/x86/mathops.h +++ b/libavcodec/x86/mathops.h @@ -23,7 +23,9 @@ #define AVCODEC_X86_MATHOPS_H #include config.h + #include libavutil/common.h +#include libavutil/x86/asm.h #if HAVE_INLINE_ASM @@ -88,6 +90,7 @@ static inline av_const int mid_pred(int a, int b, int c) return i; } +#if HAVE_6REGS #define COPY3_IF_LT(x, y, a, b, c, d)\ __asm__ volatile(\ cmpl %0, %3 \n\t\ @@ -97,6 +100,8 @@ __asm__ volatile(\ : +r (x), +r (a), +r (c)\ : r (y), r (b), r (d)\ ); +#endif /* HAVE_6REGS */ + #endif /* HAVE_I686 */ #define MASK_ABS(mask, level) \ ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mov: Fix little endian audio detection
Module: libav
Branch: release/10
Commit: 8b9d0f5d3aceb67d472d190db53a21177fea9275
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Fri Mar 13 19:45:14 2015 +
mov: Fix little endian audio detection
Set this field to TRUE if the audio component is to operate on
little-endian data, and FALSE otherwise.
However TRUE and FALSE are not defined. Since this flag is just a boolean,
interpret all values except for 0 as little endian.
Sample-Id: 64bit_FLOAT_Little_Endian.mov
(cherry picked from commit 8ae4d4e117626313e0b7df746e82de84d00d160a)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavformat/mov.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/mov.c b/libavformat/mov.c
index 5ef343e..4f0062d 100644
--- a/libavformat/mov.c
+++ b/libavformat/mov.c
@@ -816,7 +816,7 @@ static int mov_read_enda(MOVContext *c, AVIOContext *pb,
MOVAtom atom)
return 0;
st = c-fc-streams[c-fc-nb_streams-1];
-little_endian = avio_rb16(pb);
+little_endian = !!avio_rb16(pb);
av_dlog(c-fc, enda %d\n, little_endian);
if (little_endian == 1) {
switch (st-codec-codec_id) {
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] nut: Make sure to clean up on read_header failure
Module: libav
Branch: release/10
Commit: c34d1099db6f9bc49f30eb3141f87c42b01a3714
Author:Luca Barbato lu_z...@gentoo.org
Committer: Reinhard Tartler siret...@tauware.de
Date: Wed Apr 29 21:29:49 2015 +0200
nut: Make sure to clean up on read_header failure
Based on Andreas Cadhalpun andreas.cadhal...@googlemail.com work.
CC: libav-sta...@libav.org
(cherry picked from commit 1f64b018cbec018fa66a4a20f79958d9707913de)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavformat/nutdec.c | 39 ++-
1 file changed, 22 insertions(+), 17 deletions(-)
diff --git a/libavformat/nutdec.c b/libavformat/nutdec.c
index 0da7c4b..56bffe0 100644
--- a/libavformat/nutdec.c
+++ b/libavformat/nutdec.c
@@ -659,6 +659,20 @@ fail:
return ret;
}
+static int nut_read_close(AVFormatContext *s)
+{
+NUTContext *nut = s-priv_data;
+int i;
+
+av_freep(nut-time_base);
+av_freep(nut-stream);
+ff_nut_free_sp(nut);
+for (i = 1; i nut-header_count; i++)
+av_freep(nut-header[i]);
+
+return 0;
+}
+
static int nut_read_header(AVFormatContext *s)
{
NUTContext *nut = s-priv_data;
@@ -674,7 +688,7 @@ static int nut_read_header(AVFormatContext *s)
pos = find_startcode(bc, MAIN_STARTCODE, pos) + 1;
if (pos 0 + 1) {
av_log(s, AV_LOG_ERROR, No main startcode found.\n);
-return AVERROR_INVALIDDATA;
+goto fail;
}
} while (decode_main_header(nut) 0);
@@ -684,7 +698,7 @@ static int nut_read_header(AVFormatContext *s)
pos = find_startcode(bc, STREAM_STARTCODE, pos) + 1;
if (pos 0 + 1) {
av_log(s, AV_LOG_ERROR, Not all stream headers found.\n);
-return AVERROR_INVALIDDATA;
+goto fail;
}
if (decode_stream_header(nut) = 0)
initialized_stream_count++;
@@ -698,7 +712,7 @@ static int nut_read_header(AVFormatContext *s)
if (startcode == 0) {
av_log(s, AV_LOG_ERROR, EOF before video frames\n);
-return AVERROR_INVALIDDATA;
+goto fail;
} else if (startcode == SYNCPOINT_STARTCODE) {
nut-next_startcode = startcode;
break;
@@ -721,6 +735,11 @@ static int nut_read_header(AVFormatContext *s)
ff_metadata_conv_ctx(s, NULL, ff_nut_metadata_conv);
return 0;
+
+fail:
+nut_read_close(s);
+
+return AVERROR_INVALIDDATA;
}
static int decode_frame_header(NUTContext *nut, int64_t *pts, int *stream_id,
@@ -987,20 +1006,6 @@ static int read_seek(AVFormatContext *s, int stream_index,
return 0;
}
-static int nut_read_close(AVFormatContext *s)
-{
-NUTContext *nut = s-priv_data;
-int i;
-
-av_freep(nut-time_base);
-av_freep(nut-stream);
-ff_nut_free_sp(nut);
-for (i = 1; i nut-header_count; i++)
-av_freep(nut-header[i]);
-
-return 0;
-}
-
AVInputFormat ff_nut_demuxer = {
.name = nut,
.long_name = NULL_IF_CONFIG_SMALL(NUT),
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] avpacket: Check for and return errors in ff_interleave_add_packet()
Module: libav
Branch: release/10
Commit: e3c4eb87e4d9650d76f6d8d790a9b749b325e973
Author:Nidhi Makhijani nidhim...@gmail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Mon Jul 14 12:10:44 2014 +0530
avpacket: Check for and return errors in ff_interleave_add_packet()
Signed-off-by: Diego Biurrun di...@biurrun.de
(cherry picked from commit 324ff59444ff5470bb325ff1e2be7c4b054fc944)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavformat/audiointerleave.c |8 +---
libavformat/internal.h|4 ++--
libavformat/mux.c | 20 +++-
3 files changed, 22 insertions(+), 10 deletions(-)
diff --git a/libavformat/audiointerleave.c b/libavformat/audiointerleave.c
index e4cde9d..e49c77f 100644
--- a/libavformat/audiointerleave.c
+++ b/libavformat/audiointerleave.c
@@ -99,7 +99,7 @@ int ff_audio_rechunk_interleave(AVFormatContext *s, AVPacket
*out, AVPacket *pkt
int (*get_packet)(AVFormatContext *, AVPacket *,
AVPacket *, int),
int (*compare_ts)(AVFormatContext *, AVPacket *,
AVPacket *))
{
-int i;
+int i, ret;
if (pkt) {
AVStream *st = s-streams[pkt-stream_index];
@@ -116,7 +116,8 @@ int ff_audio_rechunk_interleave(AVFormatContext *s,
AVPacket *out, AVPacket *pkt
// rewrite pts and dts to be decoded time line position
pkt-pts = pkt-dts = aic-dts;
aic-dts += pkt-duration;
-ff_interleave_add_packet(s, pkt, compare_ts);
+if ((ret = ff_interleave_add_packet(s, pkt, compare_ts)) 0)
+return ret;
}
pkt = NULL;
}
@@ -126,7 +127,8 @@ int ff_audio_rechunk_interleave(AVFormatContext *s,
AVPacket *out, AVPacket *pkt
if (st-codec-codec_type == AVMEDIA_TYPE_AUDIO) {
AVPacket new_pkt;
while (interleave_new_audio_packet(s, new_pkt, i, flush))
-ff_interleave_add_packet(s, new_pkt, compare_ts);
+if ((ret = ff_interleave_add_packet(s, new_pkt, compare_ts))
0)
+return ret;
}
}
diff --git a/libavformat/internal.h b/libavformat/internal.h
index 0e7eb36..2824436 100644
--- a/libavformat/internal.h
+++ b/libavformat/internal.h
@@ -87,8 +87,8 @@ void ff_program_add_stream_index(AVFormatContext *ac, int
progid, unsigned int i
* Add packet to AVFormatContext-packet_buffer list, determining its
* interleaved position using compare() function argument.
*/
-void ff_interleave_add_packet(AVFormatContext *s, AVPacket *pkt,
- int (*compare)(AVFormatContext *, AVPacket *,
AVPacket *));
+int ff_interleave_add_packet(AVFormatContext *s, AVPacket *pkt,
+ int (*compare)(AVFormatContext *, AVPacket *,
AVPacket *));
void ff_read_frame_flush(AVFormatContext *s);
diff --git a/libavformat/mux.c b/libavformat/mux.c
index 59f9c42..8466605 100644
--- a/libavformat/mux.c
+++ b/libavformat/mux.c
@@ -477,12 +477,15 @@ int av_write_frame(AVFormatContext *s, AVPacket *pkt)
return ret;
}
-void ff_interleave_add_packet(AVFormatContext *s, AVPacket *pkt,
- int (*compare)(AVFormatContext *, AVPacket *,
AVPacket *))
+int ff_interleave_add_packet(AVFormatContext *s, AVPacket *pkt,
+ int (*compare)(AVFormatContext *, AVPacket *,
AVPacket *))
{
+int ret;
AVPacketList **next_point, *this_pktl;
this_pktl = av_mallocz(sizeof(AVPacketList));
+if (!this_pktl)
+return AVERROR(ENOMEM);
this_pktl-pkt = *pkt;
#if FF_API_DESTRUCT_PACKET
FF_DISABLE_DEPRECATION_WARNINGS
@@ -490,7 +493,11 @@ FF_DISABLE_DEPRECATION_WARNINGS
FF_ENABLE_DEPRECATION_WARNINGS
#endif
pkt-buf = NULL;
-av_dup_packet(this_pktl-pkt); // duplicate the packet if it uses
non-alloced memory
+// Duplicate the packet if it uses non-allocated memory
+if ((ret = av_dup_packet(this_pktl-pkt)) 0) {
+av_free(this_pktl);
+return ret;
+}
if (s-streams[pkt-stream_index]-last_in_packet_buffer) {
next_point =
(s-streams[pkt-stream_index]-last_in_packet_buffer-next);
@@ -515,6 +522,8 @@ next_non_null:
s-streams[pkt-stream_index]-last_in_packet_buffer =
*next_point = this_pktl;
+
+return 0;
}
static int interleave_compare_dts(AVFormatContext *s, AVPacket *next,
@@ -535,10 +544,11 @@ int ff_interleave_packet_per_dts(AVFormatContext *s,
AVPacket *out,
{
AVPacketList *pktl;
int stream_count = 0;
-int i;
+int i, ret;
if (pkt) {
-ff_interleave_add_packet(s, pkt, interleave_compare_dts);
+if ((ret = ff_interleave_add_packet(s, pkt, interleave_compare_dts))
0)
+return ret;
}
if (s-max_interleave_delta 0 s-packet_buffer !flush) {
___
libav-commits mailing list
[libav-commits] h264: Make sure reinit failures mark the context as not initialized
Module: libav
Branch: release/10
Commit: 964fef3f3ced60e67831549df223bc177e1537c9
Author:Luca Barbato lu_z...@gentoo.org
Committer: Reinhard Tartler siret...@tauware.de
Date: Mon May 25 22:30:10 2015 +0200
h264: Make sure reinit failures mark the context as not initialized
Bug-Id: CVE-2015-3417
CC: libav-sta...@libav.org
(cherry picked from commit 3b69f245dbe6e2016659a45c4bfe284f6c5ac57e)
Signed-off-by: Reinhard Tartler siret...@tauware.de
Conflicts:
libavcodec/h264_slice.c
---
libavcodec/h264.c |1 +
1 file changed, 1 insertion(+)
diff --git a/libavcodec/h264.c b/libavcodec/h264.c
index 857a949..c327113 100644
--- a/libavcodec/h264.c
+++ b/libavcodec/h264.c
@@ -3489,6 +3489,7 @@ static int decode_slice_header(H264Context *h,
H264Context *h0)
(h-width != h-avctx-coded_width ||
h-height != h-avctx-coded_height ||
needs_reinit)) {
+h-context_initialized = 0;
if (h != h0) {
av_log(h-avctx, AV_LOG_ERROR, changing width/height on
slice %d\n, h0-current_slice + 1);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] Update Changelog for 10.7
Module: libav Branch: release/10 Commit: e863f4cd2a5ea05c74b8beb92e55d7ed10908d64 Author:Reinhard Tartler siret...@tauware.de Committer: Reinhard Tartler siret...@tauware.de Date: Sun May 31 11:39:39 2015 -0400 Update Changelog for 10.7 --- Changelog | 28 RELEASE |2 +- 2 files changed, 29 insertions(+), 1 deletion(-) diff --git a/Changelog b/Changelog index adaf30e..290d791 100644 --- a/Changelog +++ b/Changelog @@ -1,6 +1,34 @@ Entries are sorted chronologically from oldest to youngest within each release, releases are sorted from youngest to oldest. +version 10.7: +- h264: Make sure reinit failures mark the context as not initialized (CVE-2015-3417) +- msrle: Use FFABS to determine the frame size in msrle_decode_pal4 (CVE-2015-3395) +- x86: cavs: Remove an unneeded scratch buffer +- configure: Disable i686 for i586 and lower CPUs (debian/783082) +- mjpegenc: Fix JFIF header byte ordering (bug/808) +- nut: Make sure to clean up on read_header failure +- png: Set the color range as full range +- avi: Validate sample_size +- nut: Check chapter creation in decode_info_header +- alac: Reject rice_limit 0 if compression is used +- ape: Support _ files with nblock smaller than 64 +- mux: Do not leave stale side data pointers in ff_interleave_add_packet() +- avpacket: Check for and return errors in ff_interleave_add_packet() +- mpegts: Update the PSI/SI table only if the version change +- rtsp: Make sure we don't write too many transport entries into a fixed-size array +- rtpenc_jpeg: Handle case of picture dimensions not dividing by 8 +- mov: Fix little endian audio detection +- x86: Put COPY3_IF_LT under HAVE_6REGS (gentoo/541930) +- roqvideoenc: set enc-avctx in roq_encode_init +- libvpx: Fix mixed use of av_malloc() and av_reallocp() +- (libav.org/release/10) alsdec: only adapt order for positive max_order +- alsdec: check sample pointer range in revert_channel_correlation +- aacpsy: correct calculation of minath in psy_3gpp_init +- alsdec: limit avctx-bits_per_raw_sample to 32 +- aasc: return correct buffer size from aasc_decode_frame +- matroskadec: fix crash when parsing invalid mkv + version 10.6: - utvideodec: Handle slice_height being zero (CVE-2014-9604) - tiff: Check that there is no aliasing in pixel format selection (CVE-2014-8544) diff --git a/RELEASE b/RELEASE index 12c0281..4e74f9f 100644 --- a/RELEASE +++ b/RELEASE @@ -1 +1 @@ -10.6 +10.7 ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] x86: cavs: Remove an unneeded scratch buffer
Module: libav
Branch: release/10
Commit: 60852532ba0971c4d253b755a3596e0a004a8bd9
Author:Michael Niedermayer michae...@gmx.at
Committer: Reinhard Tartler siret...@tauware.de
Date: Thu May 28 12:38:35 2015 +0200
x86: cavs: Remove an unneeded scratch buffer
Simplifies the code and makes it build on certain compilers
running out of registers on x86.
CC: libav-sta...@libav.org
Reported-By: mudler
(cherry picked from commit e4610300de6869bd6b3b00e76cfeabb6d7653dcd)
Signed-off-by: Luca Barbato lu_z...@gentoo.org
(cherry picked from commit 4dc0fbb13c33b4e5bdb766652f4daf900ccc952f)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavcodec/x86/cavsdsp.c | 28 +---
1 file changed, 13 insertions(+), 15 deletions(-)
diff --git a/libavcodec/x86/cavsdsp.c b/libavcodec/x86/cavsdsp.c
index bc9cbf7..fef299c 100644
--- a/libavcodec/x86/cavsdsp.c
+++ b/libavcodec/x86/cavsdsp.c
@@ -140,9 +140,7 @@ static void cavs_idct8_add_mmx(uint8_t *dst, int16_t
*block, int stride)
DECLARE_ALIGNED(8, int16_t, b2)[64];
for(i=0; i2; i++){
-DECLARE_ALIGNED(8, uint64_t, tmp);
-
-cavs_idct8_1d(block+4*i, ff_pw_4.a);
+cavs_idct8_1d(block + 4 * i, ff_pw_4.a);
__asm__ volatile(
psraw $3, %%mm7 \n\t
@@ -153,20 +151,20 @@ static void cavs_idct8_add_mmx(uint8_t *dst, int16_t
*block, int stride)
psraw $3, %%mm2 \n\t
psraw $3, %%mm1 \n\t
psraw $3, %%mm0 \n\t
-movq %%mm7,%0 \n\t
+movq %%mm7, (%0) \n\t
TRANSPOSE4( %%mm0, %%mm2, %%mm4, %%mm6, %%mm7 )
-movq %%mm0, 8(%1) \n\t
-movq %%mm6, 24(%1) \n\t
-movq %%mm7, 40(%1) \n\t
-movq %%mm4, 56(%1) \n\t
-movq%0,%%mm7 \n\t
+movq %%mm0, 8(%0) \n\t
+movq %%mm6, 24(%0) \n\t
+movq %%mm7, 40(%0) \n\t
+movq %%mm4, 56(%0) \n\t
+movq(%0), %%mm7 \n\t
TRANSPOSE4( %%mm7, %%mm5, %%mm3, %%mm1, %%mm0 )
-movq %%mm7, (%1) \n\t
-movq %%mm1, 16(%1) \n\t
-movq %%mm0, 32(%1) \n\t
-movq %%mm3, 48(%1) \n\t
-: =m(tmp)
-: r(b2+32*i)
+movq %%mm7, (%0) \n\t
+movq %%mm1, 16(%0) \n\t
+movq %%mm0, 32(%0) \n\t
+movq %%mm3, 48(%0) \n\t
+:
+: r(b2 + 32 * i)
: memory
);
}
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] png: Set the color range as full range
Module: libav
Branch: release/10
Commit: f29cf578923d2d4d2cf891dfc0c7ac45c641a5b7
Author:wm4 nfx...@googlemail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Fri May 8 17:01:50 2015 +0200
png: Set the color range as full range
The format uses full range for the gray formats.
CC: libav-sta...@libav.org
(cherry picked from commit 0f50c53cfb959162f2bccc1a2c2e066d35723595)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavcodec/pngdec.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/libavcodec/pngdec.c b/libavcodec/pngdec.c
index e646296..eac4bd7 100644
--- a/libavcodec/pngdec.c
+++ b/libavcodec/pngdec.c
@@ -629,6 +629,8 @@ static av_cold int png_dec_init(AVCodecContext *avctx)
{
PNGDecContext *s = avctx-priv_data;
+avctx-color_range = AVCOL_RANGE_JPEG;
+
s-prev = av_frame_alloc();
if (!s-prev)
return AVERROR(ENOMEM);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] rtsp: Make sure we don' t write too many transport entries into a fixed-size array
Module: libav Branch: release/10 Commit: 4c4cc9b27b69a86e405fd7612aa0a62f3b62b027 Author:Martin Storsjö mar...@martin.st Committer: Reinhard Tartler siret...@tauware.de Date: Fri Apr 24 12:38:09 2015 +0300 rtsp: Make sure we don't write too many transport entries into a fixed-size array CC: libav-sta...@libav.org Signed-off-by: Martin Storsjö mar...@martin.st (cherry picked from commit f77c9d71615e17414aacbb1720693b800a5a32d3) Signed-off-by: Reinhard Tartler siret...@tauware.de --- libavformat/rtsp.c |2 ++ 1 file changed, 2 insertions(+) diff --git a/libavformat/rtsp.c b/libavformat/rtsp.c index b95be46..6f7d775 100644 --- a/libavformat/rtsp.c +++ b/libavformat/rtsp.c @@ -904,6 +904,8 @@ static void rtsp_parse_transport(RTSPMessageHeader *reply, const char *p) p++; reply-nb_transports++; +if (reply-nb_transports = RTSP_MAX_TRANSPORTS) +break; } } ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] configure: Disable i686 for i586 and lower CPUs
Module: libav Branch: release/10 Commit: 9eba675e549c48860601a4018df7756b57623e93 Author:Mikulas Patocka miku...@artax.karlin.mff.cuni.cz Committer: Reinhard Tartler siret...@tauware.de Date: Mon Sep 15 05:11:21 2014 -0700 configure: Disable i686 for i586 and lower CPUs (cherry picked from commit b37bfbfbe53917820d1f97312fa0b2e8c7a15217) Signed-off-by: Reinhard Tartler siret...@tauware.de --- configure |1 + 1 file changed, 1 insertion(+) diff --git a/configure b/configure index 35f9d5c..b4ba1e0 100755 --- a/configure +++ b/configure @@ -3121,6 +3121,7 @@ elif enabled x86; then case $cpu in i[345]86|pentium) cpuflags=-march=$cpu +disable i686 disable mmx ;; # targets that do NOT support nopl and conditional mov (cmov) ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] nut: Check chapter creation in decode_info_header
Module: libav
Branch: release/10
Commit: 3bebca9634f05ea5da7624e3a3f35ec95341e250
Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Tue Apr 28 20:57:59 2015 +0200
nut: Check chapter creation in decode_info_header
This fixes a segmentation fault when accessing the metadata.
Signed-off-by: Luca Barbato lu_z...@gentoo.org
CC: libav-sta...@libav.org
(cherry picked from commit 21b21aed797b5e636adcf2df811f96a95f208930)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavformat/nutdec.c |4
1 file changed, 4 insertions(+)
diff --git a/libavformat/nutdec.c b/libavformat/nutdec.c
index 6328549..0da7c4b 100644
--- a/libavformat/nutdec.c
+++ b/libavformat/nutdec.c
@@ -474,6 +474,10 @@ static int decode_info_header(NUTContext *nut)
nut-time_base[chapter_start %
nut-time_base_count],
start, start + chapter_len, NULL);
+if (!chapter) {
+av_log(s, AV_LOG_ERROR, Could not create chapter.\n);
+return AVERROR(ENOMEM);
+}
metadata = chapter-metadata;
} else if (stream_id_plus1) {
st = s-streams[stream_id_plus1 - 1];
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] msrle: Use FFABS to determine the frame size in msrle_decode_pal4
Module: libav
Branch: release/10
Commit: 48c7fe5b5834a197f10a6eb56cbe7cda8ee32407
Author:Luca Barbato lu_z...@gentoo.org
Committer: Reinhard Tartler siret...@tauware.de
Date: Mon May 25 21:53:26 2015 +0200
msrle: Use FFABS to determine the frame size in msrle_decode_pal4
As done in msrle_decode_8_16_24_32.
Bug-Id: CVE-2015-3395
CC: libav-sta...@libav.org
---
libavcodec/msrledec.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/msrledec.c b/libavcodec/msrledec.c
index af2a247..370d9bd 100644
--- a/libavcodec/msrledec.c
+++ b/libavcodec/msrledec.c
@@ -39,7 +39,7 @@ static int msrle_decode_pal4(AVCodecContext *avctx, AVPicture
*pic,
unsigned int pixel_ptr = 0;
int row_dec = pic-linesize[0];
int row_ptr = (avctx-height - 1) * row_dec;
-int frame_size = row_dec * avctx-height;
+int frame_size = FFABS(row_dec) * avctx-height;
int i;
while (row_ptr = 0) {
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] avi: Validate sample_size
Module: libav
Branch: release/10
Commit: 0654518597e6ee2947e4a81c26f03f9aec7ef656
Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Wed May 6 02:26:57 2015 +0200
avi: Validate sample_size
And either error out or set it to 0 if it is negative.
CC: libav-sta...@libav.org
Signed-off-by: Luca Barbato lu_z...@gentoo.org
(cherry picked from commit a55a70644872027fdf76a75edf12a09c9008880f)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavformat/avidec.c | 17 +
1 file changed, 17 insertions(+)
diff --git a/libavformat/avidec.c b/libavformat/avidec.c
index e851f0b..0423da2 100644
--- a/libavformat/avidec.c
+++ b/libavformat/avidec.c
@@ -569,6 +569,23 @@ static int avi_read_header(AVFormatContext *s)
av_log(s, AV_LOG_ERROR, unknown stream type %X\n, tag1);
goto fail;
}
+
+if (ast-sample_size 0) {
+if (s-error_recognition AV_EF_EXPLODE) {
+av_log(s, AV_LOG_ERROR,
+ Invalid sample_size %d at stream %d\n,
+ ast-sample_size,
+ stream_index);
+goto fail;
+}
+av_log(s, AV_LOG_WARNING,
+ Invalid sample_size %d at stream %d
+ setting it to 0\n,
+ ast-sample_size,
+ stream_index);
+ast-sample_size = 0;
+}
+
if (ast-sample_size == 0)
st-duration = st-nb_frames;
ast-frame_offset = ast-cum_len;
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mjpegenc: Fix JFIF header byte ordering
Module: libav Branch: release/10 Commit: 5e886756ee687be5840867d3810991efe17a66c3 Author:Shiina Hideaki shi...@yndrd.com Committer: Reinhard Tartler siret...@tauware.de Date: Thu May 7 01:46:55 2015 +0100 mjpegenc: Fix JFIF header byte ordering The header had a wrong version description. Bug-Id: 808 Signed-off-by: Shiina Hideaki shi...@yndrd.com Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com (cherry picked from commit 5549f693d2181b3211427f65e48eaa2f4fc5a402) Signed-off-by: Reinhard Tartler siret...@tauware.de Conflicts: libavcodec/mjpegenc_common.c --- libavcodec/mjpegenc.c |5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/libavcodec/mjpegenc.c b/libavcodec/mjpegenc.c index 30433c3..cf0f510 100644 --- a/libavcodec/mjpegenc.c +++ b/libavcodec/mjpegenc.c @@ -142,7 +142,10 @@ static void jpeg_put_comments(AVCodecContext *avctx, PutBitContext *p) put_marker(p, APP0); put_bits(p, 16, 16); avpriv_put_string(p, JFIF, 1); /* this puts the trailing zero-byte too */ -put_bits(p, 16, 0x0201); /* v 1.02 */ +/* The most significant byte is used for major revisions, the least + * significant byte for minor revisions. Version 1.02 is the current + * released revision. */ +put_bits(p, 16, 0x0102); put_bits(p, 8, 0); /* units type: 0 - aspect ratio */ put_bits(p, 16, avctx-sample_aspect_ratio.num); put_bits(p, 16, avctx-sample_aspect_ratio.den); ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] alac: Reject rice_limit 0 if compression is used
Module: libav
Branch: release/10
Commit: 1dcb68c02a55648facb82c0b2b6a1eb3dc18f988
Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Fri Apr 24 00:01:43 2015 +0200
alac: Reject rice_limit 0 if compression is used
If in compression mode rice_limit = 0 leads to call
`show_bits(gb, k)` in `decode_scalar` with k = 0.
Request a sample in case it is valid and it should be accepted.
Signed-off-by: Luca Barbato lu_z...@gentoo.org
CC: libav-sta...@libav.org
(cherry picked from commit cb5324200ccdc693dd5b28dcd7d4b722fad83ea2)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavcodec/alac.c |6 ++
1 file changed, 6 insertions(+)
diff --git a/libavcodec/alac.c b/libavcodec/alac.c
index f972531..7c2b925 100644
--- a/libavcodec/alac.c
+++ b/libavcodec/alac.c
@@ -305,6 +305,12 @@ static int decode_element(AVCodecContext *avctx, AVFrame
*frame, int ch_index,
int lpc_quant[2];
int rice_history_mult[2];
+if (!alac-rice_limit) {
+avpriv_request_sample(alac-avctx,
+ Compression with rice limit 0);
+return AVERROR(ENOSYS);
+}
+
decorr_shift = get_bits(alac-gb, 8);
decorr_left_weight = get_bits(alac-gb, 8);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mux: Do not leave stale side data pointers in ff_interleave_add_packet()
Module: libav
Branch: release/10
Commit: 296d70f58517d6702bfda32178c1e395bca9665e
Author:Michael Niedermayer michae...@gmx.at
Committer: Reinhard Tartler siret...@tauware.de
Date: Fri May 1 23:55:42 2015 +0100
mux: Do not leave stale side data pointers in ff_interleave_add_packet()
Signed-off-by: Michael Niedermayer michae...@gmx.at
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
(cherry picked from commit 386e80610de282c92ad5897683ccaf2675766ac5)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavformat/mux.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/libavformat/mux.c b/libavformat/mux.c
index 8466605..e4b6144 100644
--- a/libavformat/mux.c
+++ b/libavformat/mux.c
@@ -493,6 +493,8 @@ FF_DISABLE_DEPRECATION_WARNINGS
FF_ENABLE_DEPRECATION_WARNINGS
#endif
pkt-buf = NULL;
+pkt-side_data = NULL;
+pkt-side_data_elems = 0;
// Duplicate the packet if it uses non-allocated memory
if ((ret = av_dup_packet(this_pktl-pkt)) 0) {
av_free(this_pktl);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] roqvideoenc: set enc-avctx in roq_encode_init
Module: libav
Branch: release/10
Commit: c7a983762ab40bca4f85be64995d5398a4307aa5
Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Mon Mar 9 19:24:09 2015 +0100
roqvideoenc: set enc-avctx in roq_encode_init
So far it is only set in roq_encode_frame, but it is used in
roq_encode_end to free the coded_frame. This currently segfaults if
roq_encode_frame is not called between roq_encode_init and
roq_encode_end.
CC:libav-sta...@libav.org
Signed-off-by: Andreas Cadhalpun andreas.cadhal...@googlemail.com
Signed-off-by: Anton Khirnov an...@khirnov.net
(cherry picked from commit 9f6c36d961d27283808310e3ca1d8390b55fce9b)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavcodec/roqvideoenc.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/libavcodec/roqvideoenc.c b/libavcodec/roqvideoenc.c
index af0089f..871371b 100644
--- a/libavcodec/roqvideoenc.c
+++ b/libavcodec/roqvideoenc.c
@@ -959,6 +959,8 @@ static av_cold int roq_encode_init(AVCodecContext *avctx)
av_lfg_init(enc-randctx, 1);
+enc-avctx = avctx;
+
enc-framesSinceKeyframe = 0;
if ((avctx-width 0xf) || (avctx-height 0xf)) {
av_log(avctx, AV_LOG_ERROR, Dimensions must be divisible by 16\n);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] ape: Support _0000 files with nblock smaller than 64
Module: libav
Branch: release/10
Commit: 9726f3007196a9c3589c4f09a81c0d75e1d97f2c
Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Wed Apr 29 20:39:22 2015 +0200
ape: Support _ files with nblock smaller than 64
The decode_array_ assumed that 64 is the minimal block size
while it is not.
CC: libav-sta...@libav.org
Signed-off-by: Luca Barbato lu_z...@gentoo.org
(cherry picked from commit ac1660509ecfbeca7b63eb5ab8360011180e705b)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavcodec/apedec.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/apedec.c b/libavcodec/apedec.c
index 8669db8..d107d2b 100644
--- a/libavcodec/apedec.c
+++ b/libavcodec/apedec.c
@@ -590,12 +590,12 @@ static void decode_array_(APEContext *ctx,
GetBitContext *gb,
int ksummax, ksummin;
rice-ksum = 0;
-for (i = 0; i 5; i++) {
+for (i = 0; i FFMIN(blockstodecode, 5); i++) {
out[i] = get_rice_ook(ctx-gb, 10);
rice-ksum += out[i];
}
rice-k = av_log2(rice-ksum / 10) + 1;
-for (; i 64; i++) {
+for (; i FFMIN(blockstodecode, 64); i++) {
out[i] = get_rice_ook(ctx-gb, rice-k);
rice-ksum += out[i];
rice-k = av_log2(rice-ksum / ((i + 1) * 2)) + 1;
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] libvpx: Fix mixed use of av_malloc() and av_reallocp()
Module: libav
Branch: release/10
Commit: 891ed1184e49bc7944311302d9ece01e87c188a8
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Sun Mar 8 21:08:16 2015 +
libvpx: Fix mixed use of av_malloc() and av_reallocp()
This buffer is resized when vpx_codec_get_cx_data() returns a
VPX_CODEC_STATS_PKT packet.
CC: libav-sta...@libav.org
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
(cherry picked from commit 7244cefd6e6ba7258cb022dfd7a284099d88a3e8)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavcodec/libvpxenc.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/libavcodec/libvpxenc.c b/libavcodec/libvpxenc.c
index dc1ddc3..ba14b1d 100644
--- a/libavcodec/libvpxenc.c
+++ b/libavcodec/libvpxenc.c
@@ -291,7 +291,7 @@ static av_cold int vpx_init(AVCodecContext *avctx,
if (enccfg.g_pass == VPX_RC_FIRST_PASS)
enccfg.g_lag_in_frames = 0;
else if (enccfg.g_pass == VPX_RC_LAST_PASS) {
-int decode_size;
+int decode_size, ret;
if (!avctx-stats_in) {
av_log(avctx, AV_LOG_ERROR, No stats file for second pass\n);
@@ -299,12 +299,12 @@ static av_cold int vpx_init(AVCodecContext *avctx,
}
ctx-twopass_stats.sz = strlen(avctx-stats_in) * 3 / 4;
-ctx-twopass_stats.buf = av_malloc(ctx-twopass_stats.sz);
-if (!ctx-twopass_stats.buf) {
+ret = av_reallocp(ctx-twopass_stats.buf, ctx-twopass_stats.sz);
+if (ret 0) {
av_log(avctx, AV_LOG_ERROR,
Stat buffer alloc (%zu bytes) failed\n,
ctx-twopass_stats.sz);
-return AVERROR(ENOMEM);
+return ret;
}
decode_size = av_base64_decode(ctx-twopass_stats.buf, avctx-stats_in,
ctx-twopass_stats.sz);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] rtpenc_jpeg: Handle case of picture dimensions not dividing by 8
Module: libav
Branch: release/10
Commit: 1f1686615c20982a68b2d8db3ff109834a260d27
Author:Andrey Utkin andrey.krieger.ut...@gmail.com
Committer: Reinhard Tartler siret...@tauware.de
Date: Sat Apr 11 00:54:10 2015 +0300
rtpenc_jpeg: Handle case of picture dimensions not dividing by 8
This fixes the calculation of the number of needed blocks to make
sure that ALL pixels are represented by the result.
Signed-off-by: Martin Storsjö mar...@martin.st
(cherry picked from commit 4415d0f3bbaeb287327ef101ae98d727a69d9af1)
Signed-off-by: Reinhard Tartler siret...@tauware.de
---
libavformat/rtpenc_jpeg.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavformat/rtpenc_jpeg.c b/libavformat/rtpenc_jpeg.c
index 04df658..0ceb091 100644
--- a/libavformat/rtpenc_jpeg.c
+++ b/libavformat/rtpenc_jpeg.c
@@ -40,8 +40,8 @@ void ff_rtp_send_jpeg(AVFormatContext *s1, const uint8_t
*buf, int size)
s-timestamp = s-cur_timestamp;
/* convert video pixel dimensions from pixels to blocks */
-w = s1-streams[0]-codec-width 3;
-h = s1-streams[0]-codec-height 3;
+w = (s1-streams[0]-codec-width + 7) 3;
+h = (s1-streams[0]-codec-height + 7) 3;
/* check if pixel format is not the normal 420 case */
if (s1-streams[0]-codec-pix_fmt == AV_PIX_FMT_YUVJ422P) {
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] network: prevent SIGPIPE on OSX
Module: libav
Branch: master
Commit: 881b80b3294483696a21b21a69fa5eee30f33037
Author:wm4 nfx...@googlemail.com
Committer: Martin Storsjö mar...@martin.st
Date: Thu May 28 18:23:27 2015 +0200
network: prevent SIGPIPE on OSX
OSX does not know MSG_NOSIGNAL. BSD (which OSX is based on) has got
the socket option SO_NOSIGPIPE (even if modern BSDs also support
MSG_NOSIGNAL).
Signed-off-by: Martin Storsjö mar...@martin.st
---
libavformat/network.c |4
1 file changed, 4 insertions(+)
diff --git a/libavformat/network.c b/libavformat/network.c
index a12e70d..2c34b4a 100644
--- a/libavformat/network.c
+++ b/libavformat/network.c
@@ -153,6 +153,10 @@ int ff_socket(int af, int type, int proto)
fcntl(fd, F_SETFD, FD_CLOEXEC);
#endif
}
+#ifdef SO_NOSIGPIPE
+if (fd != -1)
+setsockopt(fd, SOL_SOCKET, SO_NOSIGPIPE, (int){1}, sizeof(int));
+#endif
return fd;
}
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mpegvideo: Move tables to a separate file
Module: libav
Branch: master
Commit: 378a00087fdadcc9b34165c05cd10a1a15f3fe61
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Vittorio Giovara vittorio.giov...@gmail.com
Date: Fri May 29 19:44:03 2015 +0100
mpegvideo: Move tables to a separate file
---
libavcodec/Makefile|3 +-
libavcodec/flvdec.c|1 +
libavcodec/flvenc.c|1 +
libavcodec/h261enc.c |1 +
libavcodec/intelh263dec.c |1 +
libavcodec/ituh263dec.c|1 +
libavcodec/ituh263enc.c|1 +
libavcodec/mpeg12.c|1 +
libavcodec/mpeg12dec.c |1 +
libavcodec/mpeg4videodec.c |1 +
libavcodec/mpegvideo.c | 84 +---
libavcodec/mpegvideo.h |8
libavcodec/mpegvideo_enc.c |1 +
libavcodec/mpegvideodata.c | 102
libavcodec/mpegvideodata.h | 33 ++
libavcodec/msmpeg4.c |1 +
libavcodec/rv10.c |1 +
libavcodec/rv20enc.c |1 +
18 files changed, 151 insertions(+), 92 deletions(-)
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 0107460..96cc258 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -76,7 +76,8 @@ OBJS-$(CONFIG_MPEGAUDIODSP)+= mpegaudiodsp.o
\
mpegaudiodsp_fixed.o \
mpegaudiodsp_float.o
OBJS-$(CONFIG_MPEGVIDEO) += mpegvideo.o mpegvideodsp.o rl.o \
- mpegvideo_motion.o mpegutils.o
+ mpegvideo_motion.o mpegutils.o \
+ mpegvideodata.o
OBJS-$(CONFIG_MPEGVIDEOENC)+= mpegvideo_enc.o mpeg12data.o \
motion_est.o ratecontrol.o\
mpegvideoencdsp.o
diff --git a/libavcodec/flvdec.c b/libavcodec/flvdec.c
index 0a6f268..7161794 100644
--- a/libavcodec/flvdec.c
+++ b/libavcodec/flvdec.c
@@ -23,6 +23,7 @@
#include flv.h
#include h263.h
#include mpegvideo.h
+#include mpegvideodata.h
void ff_flv2_decode_ac_esc(GetBitContext *gb, int *level, int *run, int *last)
{
diff --git a/libavcodec/flvenc.c b/libavcodec/flvenc.c
index e14a105..f52f66a 100644
--- a/libavcodec/flvenc.c
+++ b/libavcodec/flvenc.c
@@ -21,6 +21,7 @@
#include flv.h
#include h263.h
#include mpegvideo.h
+#include mpegvideodata.h
void ff_flv_encode_picture_header(MpegEncContext *s, int picture_number)
{
diff --git a/libavcodec/h261enc.c b/libavcodec/h261enc.c
index f24e590..ae61fc2 100644
--- a/libavcodec/h261enc.c
+++ b/libavcodec/h261enc.c
@@ -31,6 +31,7 @@
#include mpegvideo.h
#include h263.h
#include h261.h
+#include mpegvideodata.h
int ff_h261_get_picture_format(int width, int height)
{
diff --git a/libavcodec/intelh263dec.c b/libavcodec/intelh263dec.c
index e34da5c..3f37489 100644
--- a/libavcodec/intelh263dec.c
+++ b/libavcodec/intelh263dec.c
@@ -20,6 +20,7 @@
#include mpegvideo.h
#include h263.h
+#include mpegvideodata.h
/* don't understand why they choose a different header ! */
int ff_intel_h263_decode_picture_header(MpegEncContext *s)
diff --git a/libavcodec/ituh263dec.c b/libavcodec/ituh263dec.c
index 1f9ac27..8f69f0f 100644
--- a/libavcodec/ituh263dec.c
+++ b/libavcodec/ituh263dec.c
@@ -41,6 +41,7 @@
#include unary.h
#include flv.h
#include mpeg4video.h
+#include mpegvideodata.h
// The defines below define the number of bits that are read at once for
// reading vlc values. Changing these may improve speed and data cache needs
diff --git a/libavcodec/ituh263enc.c b/libavcodec/ituh263enc.c
index 3afc238..952e15c 100644
--- a/libavcodec/ituh263enc.c
+++ b/libavcodec/ituh263enc.c
@@ -32,6 +32,7 @@
#include libavutil/attributes.h
#include avcodec.h
#include mpegvideo.h
+#include mpegvideodata.h
#include h263.h
#include mathops.h
#include mpegutils.h
diff --git a/libavcodec/mpeg12.c b/libavcodec/mpeg12.c
index 900761d..69c6d0a 100644
--- a/libavcodec/mpeg12.c
+++ b/libavcodec/mpeg12.c
@@ -32,6 +32,7 @@
#include error_resilience.h
#include mpeg12.h
#include mpeg12data.h
+#include mpegvideodata.h
#include bytestream.h
#include thread.h
diff --git a/libavcodec/mpeg12dec.c b/libavcodec/mpeg12dec.c
index c66d6aa..d4c517d 100644
--- a/libavcodec/mpeg12dec.c
+++ b/libavcodec/mpeg12dec.c
@@ -41,6 +41,7 @@
#include mpeg12data.h
#include mpegutils.h
#include mpegvideo.h
+#include mpegvideodata.h
#include thread.h
#include version.h
#include xvmc_internal.h
diff --git a/libavcodec/mpeg4videodec.c b/libavcodec/mpeg4videodec.c
index 7e36170..eb71a86 100644
--- a/libavcodec/mpeg4videodec.c
+++ b/libavcodec/mpeg4videodec.c
@@ -25,6 +25,7 @@
#include internal.h
#include mpegutils.h
#include mpegvideo.h
+#include mpegvideodata.h
#include mpeg4video.h
#include h263.h
#include thread.h
diff --git a/libavcodec/mpegvideo.c
[libav-commits] mpegvideo: wmv2: Move function declarations
Module: libav
Branch: master
Commit: 2f15846ad7ad57beb0bca99c624affa0facf284b
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Vittorio Giovara vittorio.giov...@gmail.com
Date: Fri May 29 19:44:01 2015 +0100
mpegvideo: wmv2: Move function declarations
---
libavcodec/h263dec.c |1 +
libavcodec/mpegvideo.c|1 +
libavcodec/mpegvideo.h| 11 ---
libavcodec/mpegvideo_enc.c|1 +
libavcodec/mpegvideo_motion.c |1 +
libavcodec/msmpeg4.h |1 -
libavcodec/msmpeg4dec.c |1 +
libavcodec/wmv2.h | 14 ++
8 files changed, 19 insertions(+), 12 deletions(-)
diff --git a/libavcodec/h263dec.c b/libavcodec/h263dec.c
index 0e08a8d..3d0d1de 100644
--- a/libavcodec/h263dec.c
+++ b/libavcodec/h263dec.c
@@ -39,6 +39,7 @@
#include msmpeg4.h
#include qpeldsp.h
#include thread.h
+#include wmv2.h
static enum AVPixelFormat h263_get_format(AVCodecContext *avctx)
{
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index 4294eb6..49f2e66 100644
--- a/libavcodec/mpegvideo.c
+++ b/libavcodec/mpegvideo.c
@@ -44,6 +44,7 @@
#include qpeldsp.h
#include xvmc_internal.h
#include thread.h
+#include wmv2.h
#include limits.h
static const uint8_t ff_default_chroma_qscale_table[32] = {
diff --git a/libavcodec/mpegvideo.h b/libavcodec/mpegvideo.h
index 4eff4d9..9c226cc 100644
--- a/libavcodec/mpegvideo.h
+++ b/libavcodec/mpegvideo.h
@@ -851,17 +851,6 @@ int ff_msmpeg4_decode_picture_header(MpegEncContext * s);
int ff_msmpeg4_decode_ext_header(MpegEncContext * s, int buf_size);
int ff_msmpeg4_decode_init(AVCodecContext *avctx);
int ff_msmpeg4_encode_init(MpegEncContext *s);
-int ff_wmv2_decode_picture_header(MpegEncContext * s);
-int ff_wmv2_decode_secondary_picture_header(MpegEncContext * s);
-void ff_wmv2_add_mb(MpegEncContext *s, int16_t block[6][64], uint8_t *dest_y,
uint8_t *dest_cb, uint8_t *dest_cr);
-void ff_mspel_motion(MpegEncContext *s,
- uint8_t *dest_y, uint8_t *dest_cb, uint8_t
*dest_cr,
- uint8_t **ref_picture, op_pixels_func
(*pix_op)[4],
- int motion_x, int motion_y, int h);
-int ff_wmv2_encode_picture_header(MpegEncContext * s, int picture_number);
-void ff_wmv2_encode_mb(MpegEncContext * s,
- int16_t block[6][64],
- int motion_x, int motion_y);
int ff_mpeg_ref_picture(AVCodecContext *avctx, Picture *dst, Picture *src);
void ff_mpeg_unref_picture(AVCodecContext *avctx, Picture *picture);
diff --git a/libavcodec/mpegvideo_enc.c b/libavcodec/mpegvideo_enc.c
index 15cca64..a477818 100644
--- a/libavcodec/mpegvideo_enc.c
+++ b/libavcodec/mpegvideo_enc.c
@@ -56,6 +56,7 @@
#include mpeg4video.h
#include internal.h
#include bytestream.h
+#include wmv2.h
#include limits.h
#define QUANT_BIAS_SHIFT 8
diff --git a/libavcodec/mpegvideo_motion.c b/libavcodec/mpegvideo_motion.c
index 95414d1..a25b251 100644
--- a/libavcodec/mpegvideo_motion.c
+++ b/libavcodec/mpegvideo_motion.c
@@ -31,6 +31,7 @@
#include mjpegenc.h
#include msmpeg4.h
#include qpeldsp.h
+#include wmv2.h
#include limits.h
static void gmc1_motion(MpegEncContext *s,
diff --git a/libavcodec/msmpeg4.h b/libavcodec/msmpeg4.h
index 0a8ecd9..23138dc 100644
--- a/libavcodec/msmpeg4.h
+++ b/libavcodec/msmpeg4.h
@@ -54,7 +54,6 @@ int ff_msmpeg4_decode_block(MpegEncContext * s, int16_t *
block,
int n, int coded, const uint8_t *scan_table);
int ff_msmpeg4_pred_dc(MpegEncContext *s, int n,
int16_t **dc_val_ptr, int *dir_ptr);
-int ff_wmv2_decode_mb(MpegEncContext *s, int16_t block[6][64]);
#define CONFIG_MSMPEG4_DECODER (CONFIG_MSMPEG4V1_DECODER || \
CONFIG_MSMPEG4V2_DECODER || \
diff --git a/libavcodec/msmpeg4dec.c b/libavcodec/msmpeg4dec.c
index 2fc2de1..bc3cd76 100644
--- a/libavcodec/msmpeg4dec.c
+++ b/libavcodec/msmpeg4dec.c
@@ -32,6 +32,7 @@
#include mpeg4video.h
#include msmpeg4data.h
#include vc1data.h
+#include wmv2.h
#define DC_VLC_BITS 9
#define V2_INTRA_CBPC_VLC_BITS 3
diff --git a/libavcodec/wmv2.h b/libavcodec/wmv2.h
index 03a8514..b77dd98 100644
--- a/libavcodec/wmv2.h
+++ b/libavcodec/wmv2.h
@@ -56,4 +56,18 @@ typedef struct Wmv2Context {
void ff_wmv2_common_init(Wmv2Context *w);
+int ff_wmv2_decode_mb(MpegEncContext *s, int16_t block[6][64]);
+int ff_wmv2_encode_picture_header(MpegEncContext * s, int picture_number);
+void ff_wmv2_encode_mb(MpegEncContext * s, int16_t block[6][64],
+ int motion_x, int motion_y);
+int ff_wmv2_decode_picture_header(MpegEncContext * s);
+int ff_wmv2_decode_secondary_picture_header(MpegEncContext * s);
+void ff_wmv2_add_mb(MpegEncContext *s, int16_t block[6][64],
+uint8_t *dest_y, uint8_t *dest_cb, uint8_t *dest_cr);
+
+void ff_mspel_motion(MpegEncContext *s,
+ uint8_t *dest_y,
[libav-commits] mpegvideo: Move MotionEstContext and function declarations to a separate header
Module: libav
Branch: master
Commit: 149fa0b7ac180fe1df48a2e379c560813555bf57
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Vittorio Giovara vittorio.giov...@gmail.com
Date: Fri May 29 19:44:05 2015 +0100
mpegvideo: Move MotionEstContext and function declarations to a separate header
---
libavcodec/motion_est.c |1 +
libavcodec/motion_est.h | 128 +++
libavcodec/mpegvideo.h | 88 +---
3 files changed, 130 insertions(+), 87 deletions(-)
diff --git a/libavcodec/motion_est.c b/libavcodec/motion_est.c
index c51be8d..d0c0439 100644
--- a/libavcodec/motion_est.c
+++ b/libavcodec/motion_est.c
@@ -34,6 +34,7 @@
#include avcodec.h
#include internal.h
#include mathops.h
+#include motion_est.h
#include mpegutils.h
#include mpegvideo.h
diff --git a/libavcodec/motion_est.h b/libavcodec/motion_est.h
new file mode 100644
index 000..cbfbbe8
--- /dev/null
+++ b/libavcodec/motion_est.h
@@ -0,0 +1,128 @@
+/*
+ * Motion estimation
+ *
+ * This file is part of Libav.
+ *
+ * Libav is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * Libav is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with Libav; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#ifndef AVCODEC_MOTIONEST_H
+#define AVCODEC_MOTIONEST_H
+
+#include stdint.h
+
+#include avcodec.h
+#include hpeldsp.h
+#include qpeldsp.h
+
+struct MpegEncContext;
+
+#define MAX_MV 2048
+
+/**
+ * Motion estimation context.
+ */
+typedef struct MotionEstContext {
+AVCodecContext *avctx;
+int skip; /// set if ME is skipped for the current
MB
+int co_located_mv[4][2];/// mv from last P-frame for direct mode
ME
+int direct_basis_mv[4][2];
+uint8_t *scratchpad;/** data area for the ME algo, so that
+ * the ME does not need to malloc/free. */
+uint8_t *best_mb;
+uint8_t *temp_mb[2];
+uint8_t *temp;
+int best_bits;
+uint32_t *map; /// map to avoid duplicate evaluations
+uint32_t *score_map;/// map to store the scores
+unsigned map_generation;
+int pre_penalty_factor;
+int penalty_factor; /** an estimate of the bits required to
+ * code a given mv value, e.g. (1,0) takes
+ * more bits than (0,0). We have to
+ * estimate whether any reduction in
+ * residual is worth the extra bits. */
+int sub_penalty_factor;
+int mb_penalty_factor;
+int flags;
+int sub_flags;
+int mb_flags;
+int pre_pass; /// = 1 for the pre pass
+int dia_size;
+int xmin;
+int xmax;
+int ymin;
+int ymax;
+int pred_x;
+int pred_y;
+uint8_t *src[4][4];
+uint8_t *ref[4][4];
+int stride;
+int uvstride;
+/* temp variables for picture complexity calculation */
+int mc_mb_var_sum_temp;
+int mb_var_sum_temp;
+int scene_change_score;
+
+op_pixels_func(*hpel_put)[4];
+op_pixels_func(*hpel_avg)[4];
+qpel_mc_func(*qpel_put)[16];
+qpel_mc_func(*qpel_avg)[16];
+uint8_t (*mv_penalty)[MAX_MV * 2 + 1]; /// bit amount needed to encode a
MV
+uint8_t *current_mv_penalty;
+int (*sub_motion_search)(struct MpegEncContext *s,
+ int *mx_ptr, int *my_ptr, int dmin,
+ int src_index, int ref_index,
+ int size, int h);
+} MotionEstContext;
+
+static inline int ff_h263_round_chroma(int x)
+{
+//FIXME static or not?
+static const uint8_t h263_chroma_roundtab[16] = {
+// 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
+0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1,
+};
+return h263_chroma_roundtab[x 0xf] + (x 3);
+}
+
+int ff_init_me(struct MpegEncContext *s);
+
+void ff_estimate_p_frame_motion(struct MpegEncContext *s, int mb_x, int mb_y);
+void ff_estimate_b_frame_motion(struct MpegEncContext *s, int mb_x, int mb_y);
+
+int ff_pre_estimate_p_frame_motion(struct MpegEncContext *s,
+ int mb_x, int mb_y);
+
+int ff_epzs_motion_search(struct MpegEncContext *s, int *mx_ptr, int *my_ptr,
+ int P[10][2], int src_index, int ref_index,
+ int16_t
[libav-commits] mpegvideo: mpeg12: Move function declarations
Module: libav Branch: master Commit: b2b766914a49c4e3537df3a585e97b98d432edd2 Author:Vittorio Giovara vittorio.giov...@gmail.com Committer: Vittorio Giovara vittorio.giov...@gmail.com Date: Fri May 29 19:44:04 2015 +0100 mpegvideo: mpeg12: Move function declarations --- libavcodec/mpeg12.h|6 ++ libavcodec/mpegvideo.h |7 --- 2 files changed, 6 insertions(+), 7 deletions(-) diff --git a/libavcodec/mpeg12.h b/libavcodec/mpeg12.h index 9132dc3..4c1578f 100644 --- a/libavcodec/mpeg12.h +++ b/libavcodec/mpeg12.h @@ -71,4 +71,10 @@ int ff_mpeg1_decode_block_intra(MpegEncContext *s, int16_t *block, int n); void ff_mpeg1_clean_buffers(MpegEncContext *s); int ff_mpeg1_find_frame_end(ParseContext *pc, const uint8_t *buf, int buf_size, AVCodecParserContext *s); +void ff_mpeg1_encode_picture_header(MpegEncContext *s, int picture_number); +void ff_mpeg1_encode_mb(MpegEncContext *s, int16_t block[8][64], +int motion_x, int motion_y); +void ff_mpeg1_encode_init(MpegEncContext *s); +void ff_mpeg1_encode_slice_header(MpegEncContext *s); + #endif /* AVCODEC_MPEG12_H */ diff --git a/libavcodec/mpegvideo.h b/libavcodec/mpegvideo.h index 7fd4f11..bb27d06 100644 --- a/libavcodec/mpegvideo.h +++ b/libavcodec/mpegvideo.h @@ -817,13 +817,6 @@ int ff_epzs_motion_search(MpegEncContext * s, int *mx_ptr, int *my_ptr, int ff_get_mb_score(MpegEncContext * s, int mx, int my, int src_index, int ref_index, int size, int h, int add_rate); -void ff_mpeg1_encode_picture_header(MpegEncContext *s, int picture_number); -void ff_mpeg1_encode_mb(MpegEncContext *s, -int16_t block[8][64], -int motion_x, int motion_y); -void ff_mpeg1_encode_init(MpegEncContext *s); -void ff_mpeg1_encode_slice_header(MpegEncContext *s); - extern const uint8_t ff_aic_dc_scale_table[32]; extern const uint8_t ff_h263_chroma_qscale_table[32]; ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] vp9: Parse subsampling and report missing feature
Module: libav
Branch: master
Commit: 3f38d4b816b2aeca15c0b9ea7ed40377eb0c
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Vittorio Giovara vittorio.giov...@gmail.com
Date: Wed May 27 15:06:57 2015 +0100
vp9: Parse subsampling and report missing feature
---
libavcodec/vp9.c | 24 +++-
libavcodec/vp9.h |2 ++
2 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/libavcodec/vp9.c b/libavcodec/vp9.c
index 50b84ae..b9397f5 100644
--- a/libavcodec/vp9.c
+++ b/libavcodec/vp9.c
@@ -219,7 +219,29 @@ static int decode_frame_header(AVCodecContext *avctx,
return AVERROR_INVALIDDATA;
}
s-fullrange = get_bits1(s-gb);
-// for profile 1, here follows the subsampling bits
+
+// subsampling bits
+if (s-profile == 1 || s-profile == 3) {
+s-sub_x = get_bits1(s-gb);
+s-sub_y = get_bits1(s-gb);
+if (s-sub_x s-sub_y) {
+av_log(avctx, AV_LOG_ERROR,
+ 4:2:0 color not supported in profile 1 or 3\n);
+return AVERROR_INVALIDDATA;
+}
+if (get_bits1(s-gb)) { // reserved bit
+av_log(avctx, AV_LOG_ERROR, Reserved bit should be zero\n);
+return AVERROR_INVALIDDATA;
+}
+} else {
+s-sub_x = s-sub_y = 1;
+}
+if (!s-sub_x || !s-sub_y) {
+avpriv_report_missing_feature(avctx, Subsampling %d:%d,
+ s-sub_x, s-sub_y);
+return AVERROR_PATCHWELCOME;
+}
+
s-refreshrefmask = 0xff;
w = get_bits(s-gb, 16) + 1;
h = get_bits(s-gb, 16) + 1;
diff --git a/libavcodec/vp9.h b/libavcodec/vp9.h
index 724288d..c6f395e 100644
--- a/libavcodec/vp9.h
+++ b/libavcodec/vp9.h
@@ -277,6 +277,8 @@ typedef struct VP9Context {
uint8_t use_last_frame_mvs;
uint8_t errorres;
uint8_t colorspace;
+uint8_t sub_x;
+uint8_t sub_y;
uint8_t fullrange;
uint8_t intraonly;
uint8_t resetctx;
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] dxva2_hevc: fix 32x32 scaling lists
Module: libav Branch: master Commit: 35818b8aaf4d669ed0bbb05c3fd06e619f2db01a Author:Hendrik Leppkes h.lepp...@gmail.com Committer: Luca Barbato lu_z...@gentoo.org Date: Thu May 28 18:25:36 2015 +0200 dxva2_hevc: fix 32x32 scaling lists Signed-off-by: Luca Barbato lu_z...@gentoo.org --- libavcodec/dxva2_hevc.c |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavcodec/dxva2_hevc.c b/libavcodec/dxva2_hevc.c index 9d130d9..a2abb97 100644 --- a/libavcodec/dxva2_hevc.c +++ b/libavcodec/dxva2_hevc.c @@ -209,12 +209,12 @@ static void fill_scaling_lists(AVDXVAContext *ctx, const HEVCContext *h, DXVA_Qm qm-ucScalingLists2[i][j] = sl-sl[2][i][pos]; if (i 2) -qm-ucScalingLists3[i][j] = sl-sl[3][i][pos]; +qm-ucScalingLists3[i][j] = sl-sl[3][i * 3][pos]; } qm-ucScalingListDCCoefSizeID2[i] = sl-sl_dc[0][i]; if (i 2) -qm-ucScalingListDCCoefSizeID3[i] = sl-sl_dc[1][i]; +qm-ucScalingListDCCoefSizeID3[i] = sl-sl_dc[1][i * 3]; } } ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] x86: check for AV_CPU_FLAG_AVXSLOW where useful
Module: libav
Branch: master
Commit: d68c05380cebf563915412182643a8be04ef890b
Author:James Almer jamr...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Tue May 26 14:29:08 2015 -0300
x86: check for AV_CPU_FLAG_AVXSLOW where useful
Signed-off-by: James Almer jamr...@gmail.com
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
libavcodec/x86/dcadsp_init.c |4 ++--
libavcodec/x86/dct_init.c |2 +-
libavcodec/x86/fft_init.c |2 +-
libavfilter/x86/af_volume_init.c |2 +-
libavresample/x86/audio_convert_init.c |4 +++-
libavresample/x86/audio_mix_init.c |4 +++-
libavresample/x86/dither_init.c|4 ++--
libavutil/x86/float_dsp_init.c |2 +-
libavutil/x86/lls_init.c |2 +-
9 files changed, 15 insertions(+), 11 deletions(-)
diff --git a/libavcodec/x86/dcadsp_init.c b/libavcodec/x86/dcadsp_init.c
index 9acb818..7c2bec1 100644
--- a/libavcodec/x86/dcadsp_init.c
+++ b/libavcodec/x86/dcadsp_init.c
@@ -98,10 +98,10 @@ av_cold void ff_synth_filter_init_x86(SynthFilterContext *s)
if (EXTERNAL_SSE2(cpu_flags)) {
s-synth_filter_float = synth_filter_sse2;
}
-if (EXTERNAL_AVX(cpu_flags)) {
+if (EXTERNAL_AVX_FAST(cpu_flags)) {
s-synth_filter_float = synth_filter_avx;
}
-if (EXTERNAL_FMA3(cpu_flags)) {
+if (EXTERNAL_FMA3(cpu_flags) !(cpu_flags AV_CPU_FLAG_AVXSLOW)) {
s-synth_filter_float = synth_filter_fma3;
}
#endif /* HAVE_YASM */
diff --git a/libavcodec/x86/dct_init.c b/libavcodec/x86/dct_init.c
index 7bda5e8..ca9fbc7 100644
--- a/libavcodec/x86/dct_init.c
+++ b/libavcodec/x86/dct_init.c
@@ -34,6 +34,6 @@ av_cold void ff_dct_init_x86(DCTContext *s)
s-dct32 = ff_dct32_float_sse;
if (EXTERNAL_SSE2(cpu_flags))
s-dct32 = ff_dct32_float_sse2;
-if (EXTERNAL_AVX(cpu_flags))
+if (EXTERNAL_AVX_FAST(cpu_flags))
s-dct32 = ff_dct32_float_avx;
}
diff --git a/libavcodec/x86/fft_init.c b/libavcodec/x86/fft_init.c
index 7ca72c5..5c0273d 100644
--- a/libavcodec/x86/fft_init.c
+++ b/libavcodec/x86/fft_init.c
@@ -48,7 +48,7 @@ av_cold void ff_fft_init_x86(FFTContext *s)
s-fft_calc= ff_fft_calc_sse;
s-fft_permutation = FF_FFT_PERM_SWAP_LSBS;
}
-if (EXTERNAL_AVX(cpu_flags) s-nbits = 5) {
+if (EXTERNAL_AVX_FAST(cpu_flags) s-nbits = 5) {
/* AVX for SB */
s-imdct_half = ff_imdct_half_avx;
s-fft_calc= ff_fft_calc_avx;
diff --git a/libavfilter/x86/af_volume_init.c b/libavfilter/x86/af_volume_init.c
index c59e0ed..26605fb 100644
--- a/libavfilter/x86/af_volume_init.c
+++ b/libavfilter/x86/af_volume_init.c
@@ -52,7 +52,7 @@ av_cold void ff_volume_init_x86(VolumeContext *vol)
vol-scale_samples = ff_scale_samples_s32_ssse3_atom;
vol-samples_align = 4;
}
-if (EXTERNAL_AVX(cpu_flags)) {
+if (EXTERNAL_AVX_FAST(cpu_flags)) {
vol-scale_samples = ff_scale_samples_s32_avx;
vol-samples_align = 8;
}
diff --git a/libavresample/x86/audio_convert_init.c
b/libavresample/x86/audio_convert_init.c
index d85ca84..ae6c319 100644
--- a/libavresample/x86/audio_convert_init.c
+++ b/libavresample/x86/audio_convert_init.c
@@ -226,11 +226,13 @@ av_cold void ff_audio_convert_init_x86(AudioConvert *ac)
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_FLT, AV_SAMPLE_FMT_FLTP,
6, 16, 4, SSE4,
ff_conv_fltp_to_flt_6ch_sse4);
}
-if (EXTERNAL_AVX(cpu_flags)) {
+if (EXTERNAL_AVX_FAST(cpu_flags)) {
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_FLT, AV_SAMPLE_FMT_S32,
0, 32, 16, AVX, ff_conv_s32_to_flt_avx);
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_S32, AV_SAMPLE_FMT_FLT,
0, 32, 32, AVX, ff_conv_flt_to_s32_avx);
+}
+if (EXTERNAL_AVX(cpu_flags)) {
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_S16, AV_SAMPLE_FMT_S16P,
2, 16, 16, AVX,
ff_conv_s16p_to_s16_2ch_avx);
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_S16, AV_SAMPLE_FMT_S16P,
diff --git a/libavresample/x86/audio_mix_init.c
b/libavresample/x86/audio_mix_init.c
index 7fc530e..e14a540 100644
--- a/libavresample/x86/audio_mix_init.c
+++ b/libavresample/x86/audio_mix_init.c
@@ -195,11 +195,13 @@ av_cold void ff_audio_mix_init_x86(AudioMix *am)
ff_audio_mix_set_func(am, AV_SAMPLE_FMT_S16P, AV_MIX_COEFF_TYPE_FLT,
1, 2, 16, 8, SSE4,
ff_mix_1_to_2_s16p_flt_sse4);
}
-if (EXTERNAL_AVX(cpu_flags)) {
+if (EXTERNAL_AVX_FAST(cpu_flags)) {
ff_audio_mix_set_func(am, AV_SAMPLE_FMT_FLTP, AV_MIX_COEFF_TYPE_FLT,
2, 1, 32, 16, AVX, ff_mix_2_to_1_fltp_flt_avx);
ff_audio_mix_set_func(am, AV_SAMPLE_FMT_FLTP, AV_MIX_COEFF_TYPE_FLT,
[libav-commits] ppc: Clarify and extend the cpuid check
Module: libav Branch: master Commit: 3058872c293e239e3b51e86fe18cfbe720aadff1 Author:Luca Barbato lu_z...@gentoo.org Committer: Luca Barbato lu_z...@gentoo.org Date: Sun May 10 00:49:51 2015 +0200 ppc: Clarify and extend the cpuid check Add POWER entries. --- libavutil/ppc/cpu.c | 23 +++ 1 file changed, 19 insertions(+), 4 deletions(-) diff --git a/libavutil/ppc/cpu.c b/libavutil/ppc/cpu.c index 2ceb88a..27a2853 100644 --- a/libavutil/ppc/cpu.c +++ b/libavutil/ppc/cpu.c @@ -94,15 +94,30 @@ out: close(fd); return ret; #elif CONFIG_RUNTIME_CPUDETECT +#define PVR_G4_7400 0x000C +#define PVR_G5_970 0x0039 +#define PVR_G5_970FX 0x003C +#define PVR_G5_970MP 0x0044 +#define PVR_G5_970GX 0x0045 +#define PVR_POWER6 0x003E +#define PVR_POWER7 0x003F +#define PVR_POWER8 0x004B +#define PVR_CELL_PPU 0x0070 + int proc_ver; // Support of mfspr PVR emulation added in Linux 2.6.17. __asm__ volatile(mfspr %0, 287 : =r (proc_ver)); proc_ver = 16; if (proc_ver 0x8000 || -proc_ver == 0x000c || -proc_ver == 0x0039 || proc_ver == 0x003c || -proc_ver == 0x0044 || proc_ver == 0x0045 || -proc_ver == 0x0070) +proc_ver == PVR_G4_7400 || +proc_ver == PVR_G5_970 || +proc_ver == PVR_G5_970FX || +proc_ver == PVR_G5_970MP || +proc_ver == PVR_G5_970GX || +proc_ver == PVR_POWER6 || +proc_ver == PVR_POWER7 || +proc_ver == PVR_POWER8 || +proc_ver == PVR_CELL_PPU) return AV_CPU_FLAG_ALTIVEC; return 0; #else ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] ppc: avutil: Use the abriged vector types
Module: libav
Branch: master
Commit: 72cebae0d981dde144340cf51f3c323f01e215e5
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 10 04:37:50 2015 +0200
ppc: avutil: Use the abriged vector types
---
libavutil/ppc/float_dsp_altivec.c | 18 +-
libavutil/ppc/util_altivec.h | 16
2 files changed, 17 insertions(+), 17 deletions(-)
diff --git a/libavutil/ppc/float_dsp_altivec.c
b/libavutil/ppc/float_dsp_altivec.c
index fee4e7c..f6918e1 100644
--- a/libavutil/ppc/float_dsp_altivec.c
+++ b/libavutil/ppc/float_dsp_altivec.c
@@ -25,7 +25,7 @@ void ff_vector_fmul_altivec(float *dst, const float *src0,
const float *src1,
int len)
{
int i;
-vector float d0, d1, s, zero = (vector float)vec_splat_u32(0);
+vec_f d0, d1, s, zero = (vec_f)vec_splat_u32(0);
for (i = 0; i len - 7; i += 8) {
d0 = vec_ld( 0, src0 + i);
s = vec_ld( 0, src1 + i);
@@ -40,15 +40,15 @@ void ff_vector_fmul_altivec(float *dst, const float *src0,
const float *src1,
void ff_vector_fmul_window_altivec(float *dst, const float *src0,
const float *src1, const float *win, int
len)
{
-vector float zero, t0, t1, s0, s1, wi, wj;
-const vector unsigned char reverse = vcprm(3, 2, 1, 0);
+vec_f zero, t0, t1, s0, s1, wi, wj;
+const vec_u8 reverse = vcprm(3, 2, 1, 0);
int i, j;
dst += len;
win += len;
src0 += len;
-zero = (vector float)vec_splat_u32(0);
+zero = (vec_f)vec_splat_u32(0);
for (i = -len * 4, j = len * 4 - 16; i 0; i += 16, j -= 16) {
s0 = vec_ld(i, src0);
@@ -75,9 +75,9 @@ void ff_vector_fmul_add_altivec(float *dst, const float *src0,
int len)
{
int i;
-vector float d, s0, s1, s2, t0, t1, edges;
-vector unsigned char align = vec_lvsr(0,dst),
- mask = vec_lvsl(0, dst);
+vec_f d, s0, s1, s2, t0, t1, edges;
+vec_u8 align = vec_lvsr(0, dst);
+vec_u8 mask = vec_lvsl(0, dst);
for (i = 0; i len - 3; i += 4) {
t0 = vec_ld(0, dst + i);
@@ -98,8 +98,8 @@ void ff_vector_fmul_reverse_altivec(float *dst, const float
*src0,
const float *src1, int len)
{
int i;
-vector float d, s0, s1, h0, l0,
-s2, s3, zero = (vector float) vec_splat_u32(0);
+vec_f d, s0, s1, h0, l0, s2, s3;
+vec_f zero = (vec_f)vec_splat_u32(0);
src1 += len-4;
for(i = 0; i len - 7; i += 8) {
diff --git a/libavutil/ppc/util_altivec.h b/libavutil/ppc/util_altivec.h
index 5624ac5..8b31327 100644
--- a/libavutil/ppc/util_altivec.h
+++ b/libavutil/ppc/util_altivec.h
@@ -47,14 +47,14 @@
#define WORD_s2 0x18,0x19,0x1a,0x1b
#define WORD_s3 0x1c,0x1d,0x1e,0x1f
-#define vcprm(a,b,c,d) (const vector unsigned char){WORD_ ## a, WORD_ ## b,
WORD_ ## c, WORD_ ## d}
+#define vcprm(a,b,c,d) (const vec_u8){WORD_ ## a, WORD_ ## b, WORD_ ## c,
WORD_ ## d}
// Transpose 8x8 matrix of 16-bit elements (in-place)
#define TRANSPOSE8(a,b,c,d,e,f,g,h) \
do { \
-vector signed short A1, B1, C1, D1, E1, F1, G1, H1; \
-vector signed short A2, B2, C2, D2, E2, F2, G2, H2; \
+vec_s16 A1, B1, C1, D1, E1, F1, G1, H1; \
+vec_s16 A2, B2, C2, D2, E2, F2, G2, H2; \
\
A1 = vec_mergeh (a, e); \
B1 = vec_mergel (a, e); \
@@ -87,11 +87,11 @@ do { \
/** @brief loads unaligned vector @a *src with offset @a offset
and returns it */
-static inline vector unsigned char unaligned_load(int offset, uint8_t *src)
+static inline vec_u8 unaligned_load(int offset, uint8_t *src)
{
-register vector unsigned char first = vec_ld(offset, src);
-register vector unsigned char second = vec_ld(offset+15, src);
-register vector unsigned char mask = vec_lvsl(offset, src);
+register vec_u8 first = vec_ld(offset, src);
+register vec_u8 second = vec_ld(offset + 15, src);
+register vec_u8 mask = vec_lvsl(offset, src);
return vec_perm(first, second, mask);
}
@@ -102,7 +102,7 @@ static inline vector unsigned char unaligned_load(int
offset, uint8_t *src)
static inline vec_u8 load_with_perm_vec(int offset, uint8_t *src, vec_u8
perm_vec)
{
vec_u8 a = vec_ld(offset, src);
-vec_u8 b = vec_ld(offset+15, src);
+vec_u8 b = vec_ld(offset + 15, src);
return vec_perm(a, b, perm_vec);
}
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] x86: add AV_CPU_FLAG_AVXSLOW flag
Module: libav
Branch: master
Commit: f7cafb5d02aa3f26c185f6f9851413ad77a73872
Author:James Almer jamr...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Tue May 26 14:29:06 2015 -0300
x86: add AV_CPU_FLAG_AVXSLOW flag
Signed-off-by: James Almer jamr...@gmail.com
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
doc/APIchanges |3 +++
libavutil/cpu.c |3 +++
libavutil/cpu.h |1 +
libavutil/version.h |4 ++--
libavutil/x86/cpu.c | 17 ++---
5 files changed, 23 insertions(+), 5 deletions(-)
diff --git a/doc/APIchanges b/doc/APIchanges
index 5d39ec6..2c443b0 100644
--- a/doc/APIchanges
+++ b/doc/APIchanges
@@ -13,6 +13,9 @@ libavutil: 2014-08-09
API changes, most recent first:
+2015-xx-xx - xxx - lavu 54.14.0 - cpu.h
+ Add AV_CPU_FLAG_AVXSLOW.
+
2015-xx-xx - xxx - lavc 56.23.0
Add av_vda_default_init2.
diff --git a/libavutil/cpu.c b/libavutil/cpu.c
index 4e8ef61..e24b9dd 100644
--- a/libavutil/cpu.c
+++ b/libavutil/cpu.c
@@ -86,6 +86,7 @@ int av_parse_cpu_flags(const char *s)
#define CPUFLAG_SSE4 (AV_CPU_FLAG_SSE4 | CPUFLAG_SSSE3)
#define CPUFLAG_SSE42(AV_CPU_FLAG_SSE42| CPUFLAG_SSE4)
#define CPUFLAG_AVX (AV_CPU_FLAG_AVX | CPUFLAG_SSE42)
+#define CPUFLAG_AVXSLOW (AV_CPU_FLAG_AVXSLOW | CPUFLAG_AVX)
#define CPUFLAG_XOP (AV_CPU_FLAG_XOP | CPUFLAG_AVX)
#define CPUFLAG_FMA3 (AV_CPU_FLAG_FMA3 | CPUFLAG_AVX)
#define CPUFLAG_FMA4 (AV_CPU_FLAG_FMA4 | CPUFLAG_AVX)
@@ -108,6 +109,7 @@ int av_parse_cpu_flags(const char *s)
{ sse4.1 , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPUFLAG_SSE4
},.unit = flags },
{ sse4.2 , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPUFLAG_SSE42
},.unit = flags },
{ avx , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPUFLAG_AVX
},.unit = flags },
+{ avxslow , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPUFLAG_AVXSLOW
},.unit = flags },
{ xop , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPUFLAG_XOP
},.unit = flags },
{ fma3, NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPUFLAG_FMA3
},.unit = flags },
{ fma4, NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPUFLAG_FMA4
},.unit = flags },
@@ -219,6 +221,7 @@ static const struct {
{ AV_CPU_FLAG_SSE4, sse4.1 },
{ AV_CPU_FLAG_SSE42, sse4.2 },
{ AV_CPU_FLAG_AVX, avx},
+{ AV_CPU_FLAG_AVXSLOW, avxslow},
{ AV_CPU_FLAG_XOP, xop},
{ AV_CPU_FLAG_FMA3, fma3 },
{ AV_CPU_FLAG_FMA4, fma4 },
diff --git a/libavutil/cpu.h b/libavutil/cpu.h
index d7fdf26..67a1abc 100644
--- a/libavutil/cpu.h
+++ b/libavutil/cpu.h
@@ -45,6 +45,7 @@
#define AV_CPU_FLAG_SSE4 0x0100 /// Penryn SSE4.1 functions
#define AV_CPU_FLAG_SSE420x0200 /// Nehalem SSE4.2 functions
#define AV_CPU_FLAG_AVX 0x4000 /// AVX functions: requires OS
support even if YMM registers aren't used
+#define AV_CPU_FLAG_AVXSLOW 0x800 /// AVX supported, but slow when
using YMM registers (e.g. Bulldozer)
#define AV_CPU_FLAG_XOP 0x0400 /// Bulldozer XOP functions
#define AV_CPU_FLAG_FMA4 0x0800 /// Bulldozer FMA4 functions
#define AV_CPU_FLAG_CMOV 0x1000 /// i686 cmov
diff --git a/libavutil/version.h b/libavutil/version.h
index 13bb6f0..c3342cd 100644
--- a/libavutil/version.h
+++ b/libavutil/version.h
@@ -54,8 +54,8 @@
*/
#define LIBAVUTIL_VERSION_MAJOR 54
-#define LIBAVUTIL_VERSION_MINOR 13
-#define LIBAVUTIL_VERSION_MICRO 1
+#define LIBAVUTIL_VERSION_MINOR 14
+#define LIBAVUTIL_VERSION_MICRO 0
#define LIBAVUTIL_VERSION_INT AV_VERSION_INT(LIBAVUTIL_VERSION_MAJOR, \
LIBAVUTIL_VERSION_MINOR, \
diff --git a/libavutil/x86/cpu.c b/libavutil/x86/cpu.c
index 8be6d94..098ccf7 100644
--- a/libavutil/x86/cpu.c
+++ b/libavutil/x86/cpu.c
@@ -167,6 +167,7 @@ int ff_get_cpu_flags_x86(void)
if (ext_caps (1 22))
rval |= AV_CPU_FLAG_MMXEXT;
+if (!strncmp(vendor.c, AuthenticAMD, 12)) {
/* Allow for selectively disabling SSE2 functions on AMD processors
with SSE2 support but not SSE4a. This includes Athlon64, some
Opteron, and some Sempron processors. MMX, SSE, or 3DNow! are faster
@@ -174,9 +175,19 @@ int ff_get_cpu_flags_x86(void)
AV_CPU_FLAG_SSE2 and AV_CPU_FLAG_SSE2SLOW are both set in this case
so that SSE2 is used unless explicitly disabled by checking
AV_CPU_FLAG_SSE2SLOW. */
-if (!strncmp(vendor.c, AuthenticAMD, 12)
-rval AV_CPU_FLAG_SSE2 !(ecx 0x0040)) {
-rval |= AV_CPU_FLAG_SSE2SLOW;
+if (rval AV_CPU_FLAG_SSE2 !(ecx 0x0040))
+rval |= AV_CPU_FLAG_SSE2SLOW;
+
+/* Similar to the above but for AVX functions on AMD processors.
+ This is
[libav-commits] ppc: avutil: Drop a potentially dangerous workaround
Module: libav Branch: master Commit: 254eb5b6faebb7bcfc3cefc1edced6652fe9d5be Author:Luca Barbato lu_z...@gentoo.org Committer: Luca Barbato lu_z...@gentoo.org Date: Sun May 10 22:48:30 2015 +0200 ppc: avutil: Drop a potentially dangerous workaround The compiler is free to optimize such expressions in any sort of way. --- libavutil/ppc/intreadwrite.h | 14 -- 1 file changed, 14 deletions(-) diff --git a/libavutil/ppc/intreadwrite.h b/libavutil/ppc/intreadwrite.h index 4471c6a..8f8078d 100644 --- a/libavutil/ppc/intreadwrite.h +++ b/libavutil/ppc/intreadwrite.h @@ -24,20 +24,6 @@ #include stdint.h #include config.h -/* - * -O0 would compile the packed struct version, which is used by - * default, in an overly verbose fashion, so we override it here. - */ -#if HAVE_BIGENDIAN -#define AV_RB64(p) (*(const uint64_t *)(p)) -#define AV_WB64(p, v) (*(uint64_t *)(p) = (v)) - -#else -#define AV_RL64(p) (*(const uint64_t *)(p)) -#define AV_WL64(p, v) (*(uint64_t *)(p) = (v)) - -#endif - #if HAVE_XFORM_ASM #if HAVE_BIGENDIAN ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] ppc: pixblockdsp: Use the abriged vector types
Module: libav
Branch: master
Commit: 7014b65995b1fe6188fb53447bf61b08e3963355
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Tue May 12 00:45:37 2015 +0200
ppc: pixblockdsp: Use the abriged vector types
---
libavcodec/ppc/pixblockdsp.c | 43 --
1 file changed, 20 insertions(+), 23 deletions(-)
diff --git a/libavcodec/ppc/pixblockdsp.c b/libavcodec/ppc/pixblockdsp.c
index 698d655..a500a70 100644
--- a/libavcodec/ppc/pixblockdsp.c
+++ b/libavcodec/ppc/pixblockdsp.c
@@ -39,24 +39,22 @@ static void get_pixels_altivec(int16_t *restrict block,
const uint8_t *pixels,
int line_size)
{
int i;
-vector unsigned char perm = vec_lvsl(0, pixels);
-const vector unsigned char zero =
-(const vector unsigned char) vec_splat_u8(0);
+vec_u8 perm = vec_lvsl(0, pixels);
+const vec_u8 zero = (const vec_u8)vec_splat_u8(0);
for (i = 0; i 8; i++) {
/* Read potentially unaligned pixels.
* We're reading 16 pixels, and actually only want 8,
* but we simply ignore the extras. */
-vector unsigned char pixl = vec_ld(0, pixels);
-vector unsigned char pixr = vec_ld(7, pixels);
-vector unsigned char bytes = vec_perm(pixl, pixr, perm);
+vec_u8 pixl = vec_ld(0, pixels);
+vec_u8 pixr = vec_ld(7, pixels);
+vec_u8 bytes = vec_perm(pixl, pixr, perm);
// Convert the bytes into shorts.
-vector signed short shorts = (vector signed short) vec_mergeh(zero,
- bytes);
+vec_s16 shorts = (vec_s16)vec_mergeh(zero, bytes);
// Save the data to the block, we assume the block is 16-byte aligned.
-vec_st(shorts, i * 16, (vector signed short *) block);
+vec_st(shorts, i * 16, (vec_s16 *)block);
pixels += line_size;
}
@@ -66,22 +64,21 @@ static void diff_pixels_altivec(int16_t *restrict block,
const uint8_t *s1,
const uint8_t *s2, int stride)
{
int i;
-vector unsigned char perm1 = vec_lvsl(0, s1);
-vector unsigned char perm2 = vec_lvsl(0, s2);
-const vector unsigned char zero =
-(const vector unsigned char) vec_splat_u8(0);
-vector signed short shorts1, shorts2;
+vec_u8 perm1 = vec_lvsl(0, s1);
+vec_u8 perm2 = vec_lvsl(0, s2);
+const vec_u8 zero = (const vec_u8)vec_splat_u8(0);
+vec_s16 shorts1, shorts2;
for (i = 0; i 4; i++) {
/* Read potentially unaligned pixels.
* We're reading 16 pixels, and actually only want 8,
* but we simply ignore the extras. */
-vector unsigned char pixl = vec_ld(0, s1);
-vector unsigned char pixr = vec_ld(15, s1);
-vector unsigned char bytes = vec_perm(pixl, pixr, perm1);
+vec_u8 pixl = vec_ld(0, s1);
+vec_u8 pixr = vec_ld(15, s1);
+vec_u8 bytes = vec_perm(pixl, pixr, perm1);
// Convert the bytes into shorts.
-shorts1 = (vector signed short) vec_mergeh(zero, bytes);
+shorts1 = (vec_s16)vec_mergeh(zero, bytes);
// Do the same for the second block of pixels.
pixl = vec_ld(0, s2);
@@ -89,13 +86,13 @@ static void diff_pixels_altivec(int16_t *restrict block,
const uint8_t *s1,
bytes = vec_perm(pixl, pixr, perm2);
// Convert the bytes into shorts.
-shorts2 = (vector signed short) vec_mergeh(zero, bytes);
+shorts2 = (vec_s16)vec_mergeh(zero, bytes);
// Do the subtraction.
shorts1 = vec_sub(shorts1, shorts2);
// Save the data to the block, we assume the block is 16-byte aligned.
-vec_st(shorts1, 0, (vector signed short *) block);
+vec_st(shorts1, 0, (vec_s16 *)block);
s1+= stride;
s2+= stride;
@@ -112,7 +109,7 @@ static void diff_pixels_altivec(int16_t *restrict block,
const uint8_t *s1,
bytes = vec_perm(pixl, pixr, perm1);
// Convert the bytes into shorts.
-shorts1 = (vector signed short) vec_mergeh(zero, bytes);
+shorts1 = (vec_s16)vec_mergeh(zero, bytes);
// Do the same for the second block of pixels.
pixl = vec_ld(0, s2);
@@ -120,13 +117,13 @@ static void diff_pixels_altivec(int16_t *restrict block,
const uint8_t *s1,
bytes = vec_perm(pixl, pixr, perm2);
// Convert the bytes into shorts.
-shorts2 = (vector signed short) vec_mergeh(zero, bytes);
+shorts2 = (vec_s16)vec_mergeh(zero, bytes);
// Do the subtraction.
shorts1 = vec_sub(shorts1, shorts2);
// Save the data to the block, we assume the block is 16-byte aligned.
-vec_st(shorts1, 0, (vector signed short *) block);
+vec_st(shorts1, 0, (vec_s16 *)block);
s1+= stride;
s2+= stride;
[libav-commits] ppc: vsx: Implement diff_pixels and get_pixels
Module: libav
Branch: master
Commit: d0bf20a4f25ac5de021c860a0c8ad05638ee2078
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Tue May 12 02:38:56 2015 +0200
ppc: vsx: Implement diff_pixels and get_pixels
Use a macro to abstract the endianness.
---
libavcodec/ppc/pixblockdsp.c | 44 ++
libavutil/ppc/util_altivec.h | 12
2 files changed, 56 insertions(+)
diff --git a/libavcodec/ppc/pixblockdsp.c b/libavcodec/ppc/pixblockdsp.c
index 2c28e29..9cac70e 100644
--- a/libavcodec/ppc/pixblockdsp.c
+++ b/libavcodec/ppc/pixblockdsp.c
@@ -133,6 +133,40 @@ static void diff_pixels_altivec(int16_t *restrict block,
const uint8_t *s1,
#endif /* HAVE_ALTIVEC */
+#if HAVE_VSX
+static void get_pixels_vsx(int16_t *restrict block, const uint8_t *pixels,
+ int line_size)
+{
+int i;
+for (i = 0; i 8; i++) {
+vec_s16 shorts = vsx_ld_u8_s16(0, pixels);
+
+vec_vsx_st(shorts, i * 16, block);
+
+pixels += line_size;
+}
+}
+
+static void diff_pixels_vsx(int16_t *restrict block, const uint8_t *s1,
+const uint8_t *s2, int stride)
+{
+int i;
+vec_s16 shorts1, shorts2;
+for (i = 0; i 8; i++) {
+shorts1 = vsx_ld_u8_s16(0, s1);
+shorts2 = vsx_ld_u8_s16(0, s2);
+
+shorts1 = vec_sub(shorts1, shorts2);
+
+vec_vsx_st(shorts1, 0, block);
+
+s1+= stride;
+s2+= stride;
+block += 8;
+}
+}
+#endif /* HAVE_VSX */
+
av_cold void ff_pixblockdsp_init_ppc(PixblockDSPContext *c,
AVCodecContext *avctx,
unsigned high_bit_depth)
@@ -147,4 +181,14 @@ av_cold void ff_pixblockdsp_init_ppc(PixblockDSPContext *c,
c-get_pixels = get_pixels_altivec;
}
#endif /* HAVE_ALTIVEC */
+
+#if HAVE_VSX
+if (!PPC_VSX(av_get_cpu_flags()))
+return;
+
+c-diff_pixels = diff_pixels_vsx;
+
+if (!high_bit_depth)
+c-get_pixels = get_pixels_vsx;
+#endif /* HAVE_VSX */
}
diff --git a/libavutil/ppc/util_altivec.h b/libavutil/ppc/util_altivec.h
index 8b31327..02cff18 100644
--- a/libavutil/ppc/util_altivec.h
+++ b/libavutil/ppc/util_altivec.h
@@ -111,4 +111,16 @@ static inline vec_u8 load_with_perm_vec(int offset,
uint8_t *src, vec_u8 perm_ve
#endif /* HAVE_ALTIVEC */
+#if HAVE_VSX
+#if HAVE_BIGENDIAN
+#define vsx_ld_u8_s16(off, p) \
+((vec_s16)vec_mergeh((vec_u8)vec_splat_u8(0), \
+ (vec_u8)vec_vsx_ld((off), (p
+#else
+#define vsx_ld_u8_s16(off, p) \
+((vec_s16)vec_mergeh((vec_u8)vec_vsx_ld((off), (p)),\
+ (vec_u8)vec_splat_u8(0)))
+#endif /* HAVE_BIGENDIAN */
+#endif /* HAVE_VSX */
+
#endif /* AVUTIL_PPC_UTIL_ALTIVEC_H */
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] ppc: Restrict some Altivec implementations to Big Endian
Module: libav
Branch: master
Commit: da60b99a8857d5ca236f32c1799a066e0135a866
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 10 23:22:17 2015 +0200
ppc: Restrict some Altivec implementations to Big Endian
In Little Endian the vec_ld/vec_st operations work as
expected only for byte-vectors.
---
libavcodec/ppc/apedsp_altivec.c |4 ++--
libavcodec/ppc/audiodsp.c |4 ++--
libavcodec/ppc/dct-test.c |2 +-
libavcodec/ppc/fdctdsp.c |4 ++--
libavcodec/ppc/fft_init.c |4 ++--
libavcodec/ppc/fmtconvert_altivec.c |4 ++--
libavcodec/ppc/h264chroma_init.c |4 ++--
libavcodec/ppc/h264dsp.c |4 ++--
libavcodec/ppc/h264qpel.c |4 ++--
libavcodec/ppc/hpeldsp_altivec.c |4 ++--
libavcodec/ppc/huffyuvdsp_altivec.c |4 ++--
libavcodec/ppc/idctdsp.c |4 ++--
libavcodec/ppc/me_cmp.c |4 ++--
libavcodec/ppc/mpegaudiodsp_altivec.c |4 ++--
libavcodec/ppc/mpegvideo_altivec.c|4 ++--
libavcodec/ppc/mpegvideodsp.c |4 ++--
libavcodec/ppc/mpegvideoencdsp.c |4 ++--
libavcodec/ppc/pixblockdsp.c |4 ++--
libavcodec/ppc/svq1enc_altivec.c |4 ++--
libavcodec/ppc/vc1dsp_altivec.c |4 ++--
libavcodec/ppc/vorbisdsp_altivec.c|4 ++--
libavcodec/ppc/vp3dsp_altivec.c |4 ++--
libavcodec/ppc/vp8dsp_altivec.c |4 ++--
libavutil/ppc/float_dsp_init.c|3 ++-
libswscale/ppc/swscale_altivec.c |2 +-
25 files changed, 48 insertions(+), 47 deletions(-)
diff --git a/libavcodec/ppc/apedsp_altivec.c b/libavcodec/ppc/apedsp_altivec.c
index d8bf4bd..3b9d045 100644
--- a/libavcodec/ppc/apedsp_altivec.c
+++ b/libavcodec/ppc/apedsp_altivec.c
@@ -29,7 +29,7 @@
#include libavutil/ppc/types_altivec.h
#include libavcodec/apedsp.h
-#if HAVE_ALTIVEC
+#if HAVE_ALTIVEC HAVE_BIGENDIAN
static int32_t scalarproduct_and_madd_int16_altivec(int16_t *v1,
const int16_t *v2,
const int16_t *v3,
@@ -73,7 +73,7 @@ static int32_t scalarproduct_and_madd_int16_altivec(int16_t
*v1,
av_cold void ff_apedsp_init_ppc(APEDSPContext *c)
{
-#if HAVE_ALTIVEC
+#if HAVE_ALTIVEC HAVE_BIGENDIAN
if (!PPC_ALTIVEC(av_get_cpu_flags()))
return;
diff --git a/libavcodec/ppc/audiodsp.c b/libavcodec/ppc/audiodsp.c
index 36506ce..2a0a6d8 100644
--- a/libavcodec/ppc/audiodsp.c
+++ b/libavcodec/ppc/audiodsp.c
@@ -35,7 +35,7 @@
#include libavutil/ppc/util_altivec.h
#include libavcodec/audiodsp.h
-#if HAVE_ALTIVEC
+#if HAVE_ALTIVEC HAVE_BIGENDIAN
static int32_t scalarproduct_int16_altivec(const int16_t *v1, const int16_t
*v2,
int order)
@@ -63,7 +63,7 @@ static int32_t scalarproduct_int16_altivec(const int16_t *v1,
const int16_t *v2,
av_cold void ff_audiodsp_init_ppc(AudioDSPContext *c)
{
-#if HAVE_ALTIVEC
+#if HAVE_ALTIVEC HAVE_BIGENDIAN
if (!PPC_ALTIVEC(av_get_cpu_flags()))
return;
diff --git a/libavcodec/ppc/dct-test.c b/libavcodec/ppc/dct-test.c
index 37fd8bb..2acbe2a 100644
--- a/libavcodec/ppc/dct-test.c
+++ b/libavcodec/ppc/dct-test.c
@@ -21,7 +21,7 @@
#include fdct.h
static const struct algo fdct_tab_arch[] = {
-#if HAVE_ALTIVEC
+#if HAVE_ALTIVEC HAVE_BIGENDIAN
{ altivecfdct, ff_fdct_altivec, FF_IDCT_PERM_NONE, AV_CPU_FLAG_ALTIVEC },
#endif
{ 0 }
diff --git a/libavcodec/ppc/fdctdsp.c b/libavcodec/ppc/fdctdsp.c
index 51417a5..89ceb47 100644
--- a/libavcodec/ppc/fdctdsp.c
+++ b/libavcodec/ppc/fdctdsp.c
@@ -29,7 +29,7 @@
#include libavcodec/fdctdsp.h
#include fdct.h
-#if HAVE_ALTIVEC
+#if HAVE_ALTIVEC HAVE_BIGENDIAN
#define vs16(v) ((vector signed short) (v))
#define vs32(v) ((vector signed int) (v))
@@ -465,7 +465,7 @@ void ff_fdct_altivec(int16_t *block)
av_cold void ff_fdctdsp_init_ppc(FDCTDSPContext *c, AVCodecContext *avctx,
unsigned high_bit_depth)
{
-#if HAVE_ALTIVEC
+#if HAVE_ALTIVEC HAVE_BIGENDIAN
if (!PPC_ALTIVEC(av_get_cpu_flags()))
return;
diff --git a/libavcodec/ppc/fft_init.c b/libavcodec/ppc/fft_init.c
index 8fcc033..9e79f70 100644
--- a/libavcodec/ppc/fft_init.c
+++ b/libavcodec/ppc/fft_init.c
@@ -39,7 +39,7 @@
void ff_fft_calc_altivec(FFTContext *s, FFTComplex *z);
void ff_fft_calc_interleave_altivec(FFTContext *s, FFTComplex *z);
-#if HAVE_GNU_AS HAVE_ALTIVEC
+#if HAVE_GNU_AS HAVE_ALTIVEC HAVE_BIGENDIAN
static void imdct_half_altivec(FFTContext *s, FFTSample *output, const
FFTSample *input)
{
int j, k;
@@ -143,7 +143,7 @@ static void imdct_calc_altivec(FFTContext *s, FFTSample
*output, const FFTSample
av_cold void ff_fft_init_ppc(FFTContext *s)
{
-#if HAVE_GNU_AS HAVE_ALTIVEC
+#if HAVE_GNU_AS HAVE_ALTIVEC HAVE_BIGENDIAN
if
[libav-commits] ppc: vsx: Implement float_dsp
Module: libav
Branch: master
Commit: eecd29b3fd7fee221580798346d6582b75c7ade4
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Mon May 11 02:44:10 2015 +0200
ppc: vsx: Implement float_dsp
---
arch.mak |1 +
libavutil/ppc/Makefile |2 +
libavutil/ppc/float_dsp_init.c | 26 +
libavutil/ppc/float_dsp_vsx.c | 117
libavutil/ppc/float_dsp_vsx.h | 38 +
5 files changed, 175 insertions(+), 9 deletions(-)
diff --git a/arch.mak b/arch.mak
index 4bfc883..720062a 100644
--- a/arch.mak
+++ b/arch.mak
@@ -5,6 +5,7 @@ OBJS-$(HAVE_VFP) += $(VFP-OBJS) $(VFP-OBJS-yes)
OBJS-$(HAVE_NEON)+= $(NEON-OBJS)$(NEON-OBJS-yes)
OBJS-$(HAVE_ALTIVEC) += $(ALTIVEC-OBJS) $(ALTIVEC-OBJS-yes)
+OBJS-$(HAVE_VSX) += $(VSX-OBJS) $(VSX-OBJS-yes)
OBJS-$(HAVE_MMX) += $(MMX-OBJS) $(MMX-OBJS-yes)
OBJS-$(HAVE_YASM)+= $(YASM-OBJS)$(YASM-OBJS-yes)
diff --git a/libavutil/ppc/Makefile b/libavutil/ppc/Makefile
index 4fd8d6d..a0febf8 100644
--- a/libavutil/ppc/Makefile
+++ b/libavutil/ppc/Makefile
@@ -2,3 +2,5 @@ OBJS += ppc/cpu.o
\
ppc/float_dsp_init.o\
ALTIVEC-OBJS += ppc/float_dsp_altivec.o \
+
+VSX-OBJS += ppc/float_dsp_vsx.o \
diff --git a/libavutil/ppc/float_dsp_init.c b/libavutil/ppc/float_dsp_init.c
index e81dc7a..4407962 100644
--- a/libavutil/ppc/float_dsp_init.c
+++ b/libavutil/ppc/float_dsp_init.c
@@ -24,18 +24,26 @@
#include libavutil/float_dsp.h
#include libavutil/ppc/cpu.h
#include float_dsp_altivec.h
+#include float_dsp_vsx.h
av_cold void ff_float_dsp_init_ppc(AVFloatDSPContext *fdsp, int bit_exact)
{
-if (!PPC_ALTIVEC(av_get_cpu_flags()))
-return;
-#if HAVE_BIGENDIAN
-fdsp-vector_fmul = ff_vector_fmul_altivec;
-fdsp-vector_fmul_add = ff_vector_fmul_add_altivec;
-fdsp-vector_fmul_reverse = ff_vector_fmul_reverse_altivec;
+if (HAVE_BIGENDIAN PPC_ALTIVEC(av_get_cpu_flags())) {
+fdsp-vector_fmul = ff_vector_fmul_altivec;
+fdsp-vector_fmul_add = ff_vector_fmul_add_altivec;
+fdsp-vector_fmul_reverse = ff_vector_fmul_reverse_altivec;
-if (!bit_exact) {
-fdsp-vector_fmul_window = ff_vector_fmul_window_altivec;
+if (!bit_exact) {
+fdsp-vector_fmul_window = ff_vector_fmul_window_altivec;
+}
+}
+if (PPC_VSX(av_get_cpu_flags())) {
+fdsp-vector_fmul = ff_vector_fmul_vsx;
+fdsp-vector_fmul_add = ff_vector_fmul_add_vsx;
+fdsp-vector_fmul_reverse = ff_vector_fmul_reverse_vsx;
+
+if (!bit_exact) {
+fdsp-vector_fmul_window = ff_vector_fmul_window_vsx;
+}
}
-#endif
}
diff --git a/libavutil/ppc/float_dsp_vsx.c b/libavutil/ppc/float_dsp_vsx.c
new file mode 100644
index 000..4135db1
--- /dev/null
+++ b/libavutil/ppc/float_dsp_vsx.c
@@ -0,0 +1,117 @@
+/*
+ * Copyright (c) 2015 Luca Barbato lu_z...@gentoo.org
+ *
+ * This file is part of Libav.
+ *
+ * Libav is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * Libav is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with Libav; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include util_altivec.h
+#include float_dsp_vsx.h
+
+void ff_vector_fmul_vsx(float *dst,
+const float *src0, const float *src1,
+int len)
+{
+int i;
+vec_f d0, d1, zero = (vec_f)vec_splat_u32(0);
+for (i = 0; i len - 7; i += 8) {
+d0 = vec_vsx_ld( 0, src0 + i);
+d1 = vec_vsx_ld(16, src0 + i);
+d0 = vec_madd(d0, vec_vsx_ld( 0, src1 + i), zero);
+d1 = vec_madd(d1, vec_vsx_ld(16, src1 + i), zero);
+vec_vsx_st(d0, 0, dst + i);
+vec_vsx_st(d1, 16, dst + i);
+}
+}
+
+void ff_vector_fmul_window_vsx(float *dst, const float *src0,
+ const float *src1, const float *win,
+ int len)
+{
+vec_f zero, t0, t1, s0, s1, wi, wj;
+const vec_u8 reverse = vcprm(3, 2, 1, 0);
+int i, j;
+
+dst += len;
+win += len;
+src0 += len;
+
+zero = (vec_f)vec_splat_u32(0);
+
+for (i = -len * 4, j = len * 4 - 16; i 0; i += 16, j -= 16) {
+s0 = vec_vsx_ld(i, src0);
+
[libav-commits] ppc: cpu: Add support for VSX and POWER8 extensions
Module: libav
Branch: master
Commit: 7d07ee5a9bd170a06d26fd967cf8de5d3b1ce331
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 10 04:15:59 2015 +0200
ppc: cpu: Add support for VSX and POWER8 extensions
---
libavutil/cpu.h |2 ++
libavutil/ppc/cpu.c | 20 +---
libavutil/ppc/cpu.h |2 ++
3 files changed, 21 insertions(+), 3 deletions(-)
diff --git a/libavutil/cpu.h b/libavutil/cpu.h
index 7ce..d7fdf26 100644
--- a/libavutil/cpu.h
+++ b/libavutil/cpu.h
@@ -54,6 +54,8 @@
#define AV_CPU_FLAG_BMI20x4 /// Bit Manipulation Instruction Set 2
#define AV_CPU_FLAG_ALTIVEC 0x0001 /// standard
+#define AV_CPU_FLAG_VSX 0x0002 /// ISA 2.06
+#define AV_CPU_FLAG_POWER8 0x0004 /// ISA 2.07
#define AV_CPU_FLAG_ARMV5TE (1 0)
#define AV_CPU_FLAG_ARMV6(1 1)
diff --git a/libavutil/ppc/cpu.c b/libavutil/ppc/cpu.c
index 27a2853..2981856 100644
--- a/libavutil/ppc/cpu.c
+++ b/libavutil/ppc/cpu.c
@@ -85,6 +85,14 @@ int ff_get_cpu_flags_ppc(void)
if (buf[i] == AT_HWCAP) {
if (buf[i + 1] PPC_FEATURE_HAS_ALTIVEC)
ret = AV_CPU_FLAG_ALTIVEC;
+#ifdef PPC_FEATURE_HAS_VSX
+if (buf[i + 1] PPC_FEATURE_HAS_VSX)
+ret |= AV_CPU_FLAG_VSX;
+#endif
+#ifdef PPC_FEATURE_ARCH_2_07
+if (buf[i + 1] PPC_FEATURE_HAS_POWER8)
+ret |= AV_CPU_FLAG_POWER8;
+#endif
goto out;
}
}
@@ -103,7 +111,7 @@ out:
#define PVR_POWER7 0x003F
#define PVR_POWER8 0x004B
#define PVR_CELL_PPU 0x0070
-
+int ret = 0;
int proc_ver;
// Support of mfspr PVR emulation added in Linux 2.6.17.
__asm__ volatile(mfspr %0, 287 : =r (proc_ver));
@@ -118,8 +126,14 @@ out:
proc_ver == PVR_POWER7 ||
proc_ver == PVR_POWER8 ||
proc_ver == PVR_CELL_PPU)
-return AV_CPU_FLAG_ALTIVEC;
-return 0;
+ret = AV_CPU_FLAG_ALTIVEC;
+if (proc_ver == PVR_POWER7 ||
+proc_ver == PVR_POWER8)
+ret |= AV_CPU_FLAG_VSX;
+if (proc_ver == PVR_POWER8)
+ret |= AV_CPU_FLAG_POWER8;
+
+return ret;
#else
// Since we were compiled for AltiVec, just assume we have it
// until someone comes up with a proper way (not involving signal hacks).
diff --git a/libavutil/ppc/cpu.h b/libavutil/ppc/cpu.h
index f8fae58..a8b823f 100644
--- a/libavutil/ppc/cpu.h
+++ b/libavutil/ppc/cpu.h
@@ -24,5 +24,7 @@
#include libavutil/cpu_internal.h
#define PPC_ALTIVEC(flags) CPUEXT(flags, ALTIVEC)
+#define PPC_VSX(flags) CPUEXT(flags, VSX)
+#define PPC_POWER8(flags) CPUEXT(flags, POWER8)
#endif /* AVUTIL_PPC_CPU_H */
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] ppc: configure: Support ISA 2.06 and later
Module: libav
Branch: master
Commit: f22ebd2555d15b59e109a5b630dd71374b076d0b
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 10 02:47:31 2015 +0200
ppc: configure: Support ISA 2.06 and later
POWER 7 and POWER 8 support VSX and ldbrx.
POWER 8 supports additional extended VSX instructions introduced
with ISA 2.07.
---
configure | 28 +++-
1 file changed, 27 insertions(+), 1 deletion(-)
diff --git a/configure b/configure
index 863e33b..d59f5e4 100755
--- a/configure
+++ b/configure
@@ -276,6 +276,8 @@ Advanced options (experts only):
Optimization options (experts only):
--disable-asmdisable all assembly optimizations
--disable-altivecdisable AltiVec optimizations
+ --disable-vsxdisable VSX optimizations
+ --disable-power8 disable POWER8 optimizations
--disable-amd3dnow disable 3DNow! optimizations
--disable-amd3dnowextdisable 3DNow! extended optimizations
--disable-mmxdisable MMX optimizations
@@ -1339,7 +1341,9 @@ ARCH_EXT_LIST_PPC=
altivec
dcbzl
ldbrx
+power8
ppc4xx
+vsx
ARCH_EXT_LIST_X86=
@@ -1709,6 +1713,8 @@ map 'eval ${v}_inline_deps=inline_asm' $ARCH_EXT_LIST_ARM
altivec_deps=ppc
ppc4xx_deps=ppc
+vsx_deps=altivec
+power8_deps=vsx
cpunop_deps=i686
x86_64_select=i686
@@ -3323,19 +3329,28 @@ elif enabled ppc; then
;;
g4|745*|ppc745*|powerpc745*)
cpuflags=-mcpu=7450
+disable vsx
;;
74*|ppc74*|powerpc74*)
cpuflags=-mcpu=7400
+disable vsx
;;
g5|970|ppc970|powerpc970)
cpuflags=-mcpu=970
+disable vsx
;;
-power[3-7]*)
+power[3-6]*)
cpuflags=-mcpu=$cpu
+disable vsx
+;;
+power[7-8]*)
+cpuflags=-mcpu=$cpu
+enable ldbrx
;;
cell)
cpuflags=-mcpu=cell
enable ldbrx
+disable vsx
;;
e500mc)
cpuflags=-mcpu=e500mc
@@ -4012,6 +4027,15 @@ EOF
enabled altivec || warn Altivec disabled, possibly missing --cpu flag
fi
+if enabled vsx; then
+check_cflags -mvsx
+check_builtin vec_vsx_ld altivec.h __builtin_vec_vsx_ld || disable
vsx
+fi
+
+if enabled power8; then
+check_cpp_condition altivec.h defined(_ARCH_PWR8) || disable power8
+fi
+
elif enabled x86; then
check_builtin rdtscintrin.h __rdtsc()
@@ -4706,6 +4730,8 @@ if enabled arm; then
fi
if enabled ppc; then
echo AltiVec enabled ${altivec-no}
+echo VSX enabled ${vsx-no}
+echo POWER8 enabled${power8-no}
echo PPC 4xx optimizations ${ppc4xx-no}
echo dcbzl available ${dcbzl-no}
fi
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] x86: Add helper macros to check for slow cpuflags
Module: libav Branch: master Commit: cae39851201b7781f1262e1c23627b45e6e80bb4 Author:James Almer jamr...@gmail.com Committer: Luca Barbato lu_z...@gentoo.org Date: Tue May 26 14:29:07 2015 -0300 x86: Add helper macros to check for slow cpuflags Signed-off-by: James Almer jamr...@gmail.com Signed-off-by: Luca Barbato lu_z...@gentoo.org --- libavutil/cpu_internal.h | 10 ++ libavutil/x86/cpu.h | 18 ++ 2 files changed, 28 insertions(+) diff --git a/libavutil/cpu_internal.h b/libavutil/cpu_internal.h index 3bfe8a8..49c7b07 100644 --- a/libavutil/cpu_internal.h +++ b/libavutil/cpu_internal.h @@ -24,7 +24,17 @@ #define CPUEXT_SUFFIX(flags, suffix, cpuext)\ (HAVE_ ## cpuext ## suffix ((flags) AV_CPU_FLAG_ ## cpuext)) +#define CPUEXT_SUFFIX_FAST(flags, suffix, cpuext) \ +(HAVE_ ## cpuext ## suffix ((flags) AV_CPU_FLAG_ ## cpuext) \ + !((flags) AV_CPU_FLAG_ ## cpuext ## SLOW)) + +#define CPUEXT_SUFFIX_SLOW(flags, suffix, cpuext) \ +(HAVE_ ## cpuext ## suffix ((flags) AV_CPU_FLAG_ ## cpuext) \ + ((flags) AV_CPU_FLAG_ ## cpuext ## SLOW)) + #define CPUEXT(flags, cpuext) CPUEXT_SUFFIX(flags, , cpuext) +#define CPUEXT_FAST(flags, cpuext) CPUEXT_SUFFIX_FAST(flags, , cpuext) +#define CPUEXT_SLOW(flags, cpuext) CPUEXT_SUFFIX_SLOW(flags, , cpuext) int ff_get_cpu_flags_aarch64(void); int ff_get_cpu_flags_arm(void); diff --git a/libavutil/x86/cpu.h b/libavutil/x86/cpu.h index 50da30e..0695436 100644 --- a/libavutil/x86/cpu.h +++ b/libavutil/x86/cpu.h @@ -32,11 +32,17 @@ #define X86_MMXEXT(flags) CPUEXT(flags, MMXEXT) #define X86_SSE(flags) CPUEXT(flags, SSE) #define X86_SSE2(flags) CPUEXT(flags, SSE2) +#define X86_SSE2_FAST(flags)CPUEXT_FAST(flags, SSE2) +#define X86_SSE2_SLOW(flags)CPUEXT_SLOW(flags, SSE2) #define X86_SSE3(flags) CPUEXT(flags, SSE3) +#define X86_SSE3_FAST(flags)CPUEXT_FAST(flags, SSE3) +#define X86_SSE3_SLOW(flags)CPUEXT_SLOW(flags, SSE3) #define X86_SSSE3(flags)CPUEXT(flags, SSSE3) #define X86_SSE4(flags) CPUEXT(flags, SSE4) #define X86_SSE42(flags)CPUEXT(flags, SSE42) #define X86_AVX(flags) CPUEXT(flags, AVX) +#define X86_AVX_FAST(flags) CPUEXT_FAST(flags, AVX) +#define X86_AVX_SLOW(flags) CPUEXT_SLOW(flags, AVX) #define X86_XOP(flags) CPUEXT(flags, XOP) #define X86_FMA3(flags) CPUEXT(flags, FMA3) #define X86_FMA4(flags) CPUEXT(flags, FMA4) @@ -48,11 +54,17 @@ #define EXTERNAL_MMXEXT(flags) CPUEXT_SUFFIX(flags, _EXTERNAL, MMXEXT) #define EXTERNAL_SSE(flags) CPUEXT_SUFFIX(flags, _EXTERNAL, SSE) #define EXTERNAL_SSE2(flags)CPUEXT_SUFFIX(flags, _EXTERNAL, SSE2) +#define EXTERNAL_SSE2_FAST(flags) CPUEXT_SUFFIX_FAST(flags, _EXTERNAL, SSE2) +#define EXTERNAL_SSE2_SLOW(flags) CPUEXT_SUFFIX_SLOW(flags, _EXTERNAL, SSE2) #define EXTERNAL_SSE3(flags)CPUEXT_SUFFIX(flags, _EXTERNAL, SSE3) +#define EXTERNAL_SSE3_FAST(flags) CPUEXT_SUFFIX_FAST(flags, _EXTERNAL, SSE3) +#define EXTERNAL_SSE3_SLOW(flags) CPUEXT_SUFFIX_SLOW(flags, _EXTERNAL, SSE3) #define EXTERNAL_SSSE3(flags) CPUEXT_SUFFIX(flags, _EXTERNAL, SSSE3) #define EXTERNAL_SSE4(flags)CPUEXT_SUFFIX(flags, _EXTERNAL, SSE4) #define EXTERNAL_SSE42(flags) CPUEXT_SUFFIX(flags, _EXTERNAL, SSE42) #define EXTERNAL_AVX(flags) CPUEXT_SUFFIX(flags, _EXTERNAL, AVX) +#define EXTERNAL_AVX_FAST(flags)CPUEXT_SUFFIX_FAST(flags, _EXTERNAL, AVX) +#define EXTERNAL_AVX_SLOW(flags)CPUEXT_SUFFIX_SLOW(flags, _EXTERNAL, AVX) #define EXTERNAL_XOP(flags) CPUEXT_SUFFIX(flags, _EXTERNAL, XOP) #define EXTERNAL_FMA3(flags)CPUEXT_SUFFIX(flags, _EXTERNAL, FMA3) #define EXTERNAL_FMA4(flags)CPUEXT_SUFFIX(flags, _EXTERNAL, FMA4) @@ -64,11 +76,17 @@ #define INLINE_MMXEXT(flags)CPUEXT_SUFFIX(flags, _INLINE, MMXEXT) #define INLINE_SSE(flags) CPUEXT_SUFFIX(flags, _INLINE, SSE) #define INLINE_SSE2(flags) CPUEXT_SUFFIX(flags, _INLINE, SSE2) +#define INLINE_SSE2_FAST(flags) CPUEXT_SUFFIX_FAST(flags, _INLINE, SSE2) +#define INLINE_SSE2_SLOW(flags) CPUEXT_SUFFIX_SLOW(flags, _INLINE, SSE2) #define INLINE_SSE3(flags) CPUEXT_SUFFIX(flags, _INLINE, SSE3) +#define INLINE_SSE3_FAST(flags) CPUEXT_SUFFIX_FAST(flags, _INLINE, SSE3) +#define INLINE_SSE3_SLOW(flags) CPUEXT_SUFFIX_SLOW(flags, _INLINE, SSE3) #define INLINE_SSSE3(flags) CPUEXT_SUFFIX(flags, _INLINE, SSSE3) #define INLINE_SSE4(flags) CPUEXT_SUFFIX(flags, _INLINE, SSE4) #define INLINE_SSE42(flags) CPUEXT_SUFFIX(flags, _INLINE, SSE42) #define INLINE_AVX(flags) CPUEXT_SUFFIX(flags, _INLINE, AVX) +#define INLINE_AVX_FAST(flags) CPUEXT_SUFFIX_FAST(flags, _INLINE, AVX) +#define INLINE_AVX_SLOW(flags)
[libav-commits] ppc: linux: Check altivec using the auxv
Module: libav
Branch: master
Commit: baa94563fede8959a638b0fa132dd2124acd93e8
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Fri Apr 3 15:31:32 2015 +0200
ppc: linux: Check altivec using the auxv
Should prevent trying to use altivec when it is disabled by the kernel.
---
libavutil/ppc/cpu.c | 31 +++
1 file changed, 31 insertions(+)
diff --git a/libavutil/ppc/cpu.c b/libavutil/ppc/cpu.c
index 50107e1..2ceb88a 100644
--- a/libavutil/ppc/cpu.c
+++ b/libavutil/ppc/cpu.c
@@ -18,6 +18,11 @@
#ifdef __APPLE__
#include sys/sysctl.h
+#elif defined(__linux__)
+#include asm/cputable.h
+#include linux/auxvec.h
+#include fcntl.h
+#include unistd.h
#elif defined(__OpenBSD__)
#include sys/param.h
#include sys/sysctl.h
@@ -62,6 +67,32 @@ int ff_get_cpu_flags_ppc(void)
if (err == 0)
return has_vu ? AV_CPU_FLAG_ALTIVEC : 0;
return 0;
+#elif defined(__linux__)
+// The linux kernel could have the altivec support disabled
+// even if the cpu has it.
+int i, ret = 0;
+int fd = open(/proc/self/auxv, O_RDONLY);
+unsigned long buf[64] = { 0 };
+ssize_t count;
+
+if (fd 0)
+return 0;
+
+while ((count = read(fd, buf, sizeof(buf))) 0) {
+for (i = 0; i count / sizeof(*buf); i += 2) {
+if (buf[i] == AT_NULL)
+goto out;
+if (buf[i] == AT_HWCAP) {
+if (buf[i + 1] PPC_FEATURE_HAS_ALTIVEC)
+ret = AV_CPU_FLAG_ALTIVEC;
+goto out;
+}
+}
+}
+
+out:
+close(fd);
+return ret;
#elif CONFIG_RUNTIME_CPUDETECT
int proc_ver;
// Support of mfspr PVR emulation added in Linux 2.6.17.
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] dxva2_hevc: re-write reference frame handling
Module: libav
Branch: master
Commit: 902a55f7161ffea10e13c1e7df14022e2ed90514
Author:Hendrik Leppkes h.lepp...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Thu May 28 18:25:37 2015 +0200
dxva2_hevc: re-write reference frame handling
The old logic required an explicit clearing of the lists first and was
prone to overflow the DXVA2 struct in some circumstances.
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
libavcodec/dxva2_hevc.c | 40 ++--
1 file changed, 22 insertions(+), 18 deletions(-)
diff --git a/libavcodec/dxva2_hevc.c b/libavcodec/dxva2_hevc.c
index a2abb97..5112bd2 100644
--- a/libavcodec/dxva2_hevc.c
+++ b/libavcodec/dxva2_hevc.c
@@ -57,7 +57,7 @@ static void fill_picture_parameters(const AVCodecContext
*avctx, AVDXVAContext *
DXVA_PicParams_HEVC *pp)
{
const HEVCFrame *current_picture = h-ref;
-int i, j, k;
+int i, j;
memset(pp, 0, sizeof(*pp));
@@ -155,30 +155,34 @@ static void fill_picture_parameters(const AVCodecContext
*avctx, AVDXVAContext *
pp-log2_parallel_merge_level_minus2 = h-pps-log2_parallel_merge_level -
2;
pp-CurrPicOrderCntVal = h-poc;
-// empty the lists
-memset(pp-RefPicList, 0xff, sizeof(pp-RefPicList));
-memset(pp-RefPicSetStCurrBefore, 0xff,
sizeof(pp-RefPicSetStCurrBefore));
-memset(pp-RefPicSetStCurrAfter, 0xff, sizeof(pp-RefPicSetStCurrAfter));
-memset(pp-RefPicSetLtCurr, 0xff, sizeof(pp-RefPicSetLtCurr));
-
// fill RefPicList from the DPB
-for (i = 0, j = 0; i FF_ARRAY_ELEMS(h-DPB); i++) {
-const HEVCFrame *frame = h-DPB[i];
-if (frame != current_picture (frame-flags
(HEVC_FRAME_FLAG_LONG_REF | HEVC_FRAME_FLAG_SHORT_REF))) {
-fill_picture_entry(pp-RefPicList[j],
ff_dxva2_get_surface_index(avctx, ctx, frame-frame), !!(frame-flags
HEVC_FRAME_FLAG_LONG_REF));
-pp-PicOrderCntValList[j] = frame-poc;
+for (i = 0, j = 0; i FF_ARRAY_ELEMS(pp-RefPicList); i++) {
+const HEVCFrame *frame = NULL;
+while (!frame j FF_ARRAY_ELEMS(h-DPB)) {
+if (h-DPB[j] != current_picture (h-DPB[j].flags
(HEVC_FRAME_FLAG_LONG_REF | HEVC_FRAME_FLAG_SHORT_REF)))
+frame = h-DPB[j];
j++;
}
+
+if (frame) {
+fill_picture_entry(pp-RefPicList[i],
ff_dxva2_get_surface_index(avctx, ctx, frame-frame), !!(frame-flags
HEVC_FRAME_FLAG_LONG_REF));
+pp-PicOrderCntValList[i] = frame-poc;
+} else {
+pp-RefPicList[i].bPicEntry = 0xff;
+pp-PicOrderCntValList[i] = 0;
+}
}
#define DO_REF_LIST(ref_idx, ref_list) { \
const RefPicList *rpl = h-rps[ref_idx]; \
-av_assert0(rpl-nb_refs = FF_ARRAY_ELEMS(pp-ref_list)); \
-for (j = 0, k = 0; j rpl-nb_refs; j++) { \
-if (rpl-ref[j]) { \
-pp-ref_list[k] = get_refpic_index(pp,
ff_dxva2_get_surface_index(avctx, ctx, rpl-ref[j]-frame)); \
-k++; \
-} \
+for (i = 0, j = 0; i FF_ARRAY_ELEMS(pp-ref_list); i++) { \
+const HEVCFrame *frame = NULL; \
+while (!frame j rpl-nb_refs) \
+frame = rpl-ref[j++]; \
+if (frame) \
+pp-ref_list[i] = get_refpic_index(pp,
ff_dxva2_get_surface_index(avctx, ctx, frame-frame)); \
+else \
+pp-ref_list[i] = 0xff; \
} \
}
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mov: Correctly allocate ctts_data
Module: libav Branch: master Commit: a8a90906fb0ad0f8d3c68a97f182528ee3b5198c Author:Luca Barbato lu_z...@gentoo.org Committer: Luca Barbato lu_z...@gentoo.org Date: Sun May 31 12:28:33 2015 +0200 mov: Correctly allocate ctts_data It can be reallocated. CC: libav-sta...@libav.org --- libavformat/mov.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavformat/mov.c b/libavformat/mov.c index 33e1054..fbbfef0 100644 --- a/libavformat/mov.c +++ b/libavformat/mov.c @@ -2035,7 +2035,7 @@ static int mov_read_ctts(MOVContext *c, AVIOContext *pb, MOVAtom atom) return 0; if (entries = UINT_MAX / sizeof(*sc-ctts_data)) return AVERROR_INVALIDDATA; -sc-ctts_data = av_malloc(entries * sizeof(*sc-ctts_data)); +sc-ctts_data = av_realloc(NULL, entries * sizeof(*sc-ctts_data)); if (!sc-ctts_data) return AVERROR(ENOMEM); ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mov: Fix two memleaks
Module: libav Branch: master Commit: f261a55d72753bcec33369954809732abae8f8f7 Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com Committer: Luca Barbato lu_z...@gentoo.org Date: Tue May 26 14:24:36 2015 +0100 mov: Fix two memleaks Signed-off-by: Andreas Cadhalpun andreas.cadhal...@googlemail.com Signed-off-by: Luca Barbato lu_z...@gentoo.org CC: libav-sta...@libav.org --- libavformat/mov.c |5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/libavformat/mov.c b/libavformat/mov.c index 70fa1e0..33e1054 100644 --- a/libavformat/mov.c +++ b/libavformat/mov.c @@ -1942,10 +1942,11 @@ static int mov_read_stsz(MOVContext *c, AVIOContext *pb, MOVAtom atom) sc-sample_count = i; +av_free(buf); + if (pb-eof_reached) return AVERROR_EOF; -av_free(buf); return 0; } @@ -2028,6 +2029,8 @@ static int mov_read_ctts(MOVContext *c, AVIOContext *pb, MOVAtom atom) av_log(c-fc, AV_LOG_TRACE, track[%i].ctts.entries = %i\n, c-fc-nb_streams-1, entries); +av_freep(sc-ctts_data); + if (!entries) return 0; if (entries = UINT_MAX / sizeof(*sc-ctts_data)) ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mov: reject zero bytes_per_frame with non-zero samples_per_frame
Module: libav
Branch: master
Commit: cffb9ea81bd16d2ed8bce187b6bf0279c07b3b9b
Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Tue May 26 14:24:37 2015 +0100
mov: reject zero bytes_per_frame with non-zero samples_per_frame
In this case the mov demuxer can return a large number of empty packets.
Signed-off-by: Andreas Cadhalpun andreas.cadhal...@googlemail.com
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
libavformat/mov.c |7 +++
1 file changed, 7 insertions(+)
diff --git a/libavformat/mov.c b/libavformat/mov.c
index fbbfef0..2cf6e8e 100644
--- a/libavformat/mov.c
+++ b/libavformat/mov.c
@@ -2263,6 +2263,13 @@ static void mov_build_index(MOVContext *mov, AVStream
*st)
AVIndexEntry *e;
unsigned size, samples;
+if (sc-samples_per_frame 1 !sc-bytes_per_frame) {
+avpriv_request_sample(mov-fc,
+ Zero bytes per frame, but %d samples per frame,
+ sc-samples_per_frame);
+return;
+}
+
if (sc-samples_per_frame = 160) { // gsm
samples = sc-samples_per_frame;
size = sc-bytes_per_frame;
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] mov: abort on EOF in ff_mov_read_chan
Module: libav
Branch: master
Commit: 8e9c39e81fe5ba34010a7ba05cbe4ae31f177d89
Author:Andreas Cadhalpun andreas.cadhal...@googlemail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Tue May 26 14:24:35 2015 +0100
mov: abort on EOF in ff_mov_read_chan
Otherwise the loop can take a lot of time if num_descr is very large.
Signed-off-by: Andreas Cadhalpun andreas.cadhal...@googlemail.com
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
libavformat/mov_chan.c |5 +
1 file changed, 5 insertions(+)
diff --git a/libavformat/mov_chan.c b/libavformat/mov_chan.c
index 2c54920..94b93cf 100644
--- a/libavformat/mov_chan.c
+++ b/libavformat/mov_chan.c
@@ -565,6 +565,11 @@ int ff_mov_read_chan(AVFormatContext *s, AVIOContext *pb,
AVStream *st,
label_mask = 0;
for (i = 0; i num_descr; i++) {
uint32_t label;
+if (pb-eof_reached) {
+av_log(s, AV_LOG_ERROR,
+ reached EOF while reading channel layout\n);
+return AVERROR_INVALIDDATA;
+}
label = avio_rb32(pb); // mChannelLabel
avio_rb32(pb); // mChannelFlags
avio_rl32(pb); // mCoordinates[0]
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] avio: Add avio_read wrapper to simplify error checking
Module: libav
Branch: master
Commit: bff0349d9da527084e1433167466d8afc9e25c7f
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Tue May 26 14:24:38 2015 +0100
avio: Add avio_read wrapper to simplify error checking
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
libavformat/avio_internal.h |7 +++
libavformat/aviobuf.c |8
2 files changed, 15 insertions(+)
diff --git a/libavformat/avio_internal.h b/libavformat/avio_internal.h
index c8630ab..aff5564 100644
--- a/libavformat/avio_internal.h
+++ b/libavformat/avio_internal.h
@@ -85,6 +85,13 @@ int ffio_rewind_with_probe_data(AVIOContext *s, unsigned
char *buf, int buf_size
uint64_t ffio_read_varlen(AVIOContext *bc);
+/**
+ * Read size bytes from AVIOContext into buf.
+ * Check that exactly size bytes have been read.
+ * @return number of bytes read or AVERROR
+ */
+int ffio_read_size(AVIOContext *s, unsigned char *buf, int size);
+
/** @warning must be called before any I/O */
int ffio_set_buf_size(AVIOContext *s, int buf_size);
diff --git a/libavformat/aviobuf.c b/libavformat/aviobuf.c
index 59f807c..d3e3452 100644
--- a/libavformat/aviobuf.c
+++ b/libavformat/aviobuf.c
@@ -496,6 +496,14 @@ int avio_read(AVIOContext *s, unsigned char *buf, int size)
return size1 - size;
}
+int ffio_read_size(AVIOContext *s, unsigned char *buf, int size)
+{
+int ret = avio_read(s, buf, size);
+if (ret != size)
+return AVERROR_INVALIDDATA;
+return ret;
+}
+
int ffio_read_indirect(AVIOContext *s, unsigned char *buf, int size, const
unsigned char **data)
{
if (s-buf_end - s-buf_ptr = size !s-write_flag) {
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] truemotion2: Check memory allocation
Module: libav
Branch: master
Commit: e1ea365f7e1477c78865b866a180712174536c20
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:55:27 2015 +0200
truemotion2: Check memory allocation
---
libavcodec/truemotion2.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/libavcodec/truemotion2.c b/libavcodec/truemotion2.c
index 122643d..094096e 100644
--- a/libavcodec/truemotion2.c
+++ b/libavcodec/truemotion2.c
@@ -171,6 +171,10 @@ static int tm2_build_huff_table(TM2Context *ctx, TM2Codes
*code)
huff.nums= av_mallocz(huff.max_num * sizeof(int));
huff.bits= av_mallocz(huff.max_num * sizeof(uint32_t));
huff.lens= av_mallocz(huff.max_num * sizeof(int));
+if (!huff.nums || !huff.bits || !huff.lens) {
+res = AVERROR(ENOMEM);
+goto out;
+}
res = tm2_read_tree(ctx, 0, 0, huff);
@@ -193,10 +197,16 @@ static int tm2_build_huff_table(TM2Context *ctx, TM2Codes
*code)
code-bits = huff.max_bits;
code-length = huff.max_num;
code-recode = av_malloc(code-length * sizeof(int));
+if (!code-recode) {
+res = AVERROR(ENOMEM);
+goto out;
+}
for (i = 0; i code-length; i++)
code-recode[i] = huff.nums[i];
}
}
+
+out:
/* free allocated memory */
av_free(huff.nums);
av_free(huff.bits);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] asv: Check memory allocation
Module: libav Branch: master Commit: e524f37356156893cae50fc46451bd4a6198703d Author:Vittorio Giovara vittorio.giov...@gmail.com Committer: Luca Barbato lu_z...@gentoo.org Date: Sun May 31 14:50:21 2015 +0200 asv: Check memory allocation --- libavcodec/asvenc.c |2 ++ 1 file changed, 2 insertions(+) diff --git a/libavcodec/asvenc.c b/libavcodec/asvenc.c index f8c52af..0879615 100644 --- a/libavcodec/asvenc.c +++ b/libavcodec/asvenc.c @@ -299,6 +299,8 @@ static av_cold int encode_init(AVCodecContext *avctx) avctx-global_quality / 2) / avctx-global_quality; avctx-extradata = av_mallocz(8); +if (!avctx-extradata) +return AVERROR(ENOMEM); avctx-extradata_size = 8; ((uint32_t *) avctx-extradata)[0] = av_le2ne32(a-inv_qscale); ((uint32_t *) avctx-extradata)[1] = av_le2ne32(AV_RL32(ASUS)); ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] pthread: Check memory allocation
Module: libav
Branch: master
Commit: 0994e142132200c706f704271b5a7ae81d128f5c
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:55:06 2015 +0200
pthread: Check memory allocation
---
libavcodec/pthread_frame.c |7 +++
1 file changed, 7 insertions(+)
diff --git a/libavcodec/pthread_frame.c b/libavcodec/pthread_frame.c
index effc9a5..d7f742b 100644
--- a/libavcodec/pthread_frame.c
+++ b/libavcodec/pthread_frame.c
@@ -584,8 +584,15 @@ int ff_frame_thread_init(AVCodecContext *avctx)
}
avctx-internal-thread_ctx = fctx =
av_mallocz(sizeof(FrameThreadContext));
+if (!fctx)
+return AVERROR(ENOMEM);
fctx-threads = av_mallocz(sizeof(PerThreadContext) * thread_count);
+if (!fctx-threads) {
+av_freep(avctx-internal-thread_ctx);
+return AVERROR(ENOMEM);
+}
+
pthread_mutex_init(fctx-buffer_mutex, NULL);
fctx-delaying = 1;
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] psymodel: Check memory allocation
Module: libav
Branch: master
Commit: 03927cb73399e6f07185fc7f8851d7612b4187b6
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:54:55 2015 +0200
psymodel: Check memory allocation
---
libavcodec/psymodel.c | 12
1 file changed, 12 insertions(+)
diff --git a/libavcodec/psymodel.c b/libavcodec/psymodel.c
index a2af611..5179ede 100644
--- a/libavcodec/psymodel.c
+++ b/libavcodec/psymodel.c
@@ -39,6 +39,12 @@ av_cold int ff_psy_init(FFPsyContext *ctx, AVCodecContext
*avctx, int num_lens,
ctx-group = av_mallocz(sizeof(ctx-group[0]) * num_groups);
ctx-bands = av_malloc (sizeof(ctx-bands[0]) * num_lens);
ctx-num_bands = av_malloc (sizeof(ctx-num_bands[0]) * num_lens);
+
+if (!ctx-ch || !ctx-group || !ctx-bands || !ctx-num_bands) {
+ff_psy_end(ctx);
+return AVERROR(ENOMEM);
+}
+
memcpy(ctx-bands, bands, sizeof(ctx-bands[0]) * num_lens);
memcpy(ctx-num_bands, num_bands, sizeof(ctx-num_bands[0]) * num_lens);
@@ -98,6 +104,8 @@ av_cold struct FFPsyPreprocessContext*
ff_psy_preprocess_init(AVCodecContext *av
int i;
float cutoff_coeff = 0;
ctx= av_mallocz(sizeof(FFPsyPreprocessContext));
+if (!ctx)
+return NULL;
ctx-avctx = avctx;
if (avctx-cutoff 0)
@@ -109,6 +117,10 @@ av_cold struct FFPsyPreprocessContext*
ff_psy_preprocess_init(AVCodecContext *av
cutoff_coeff, 0.0, 0.0);
if (ctx-fcoeffs) {
ctx-fstate = av_mallocz(sizeof(ctx-fstate[0]) * avctx-channels);
+if (!ctx-fstate) {
+av_free(ctx);
+return NULL;
+}
for (i = 0; i avctx-channels; i++)
ctx-fstate[i] = ff_iir_filter_init_state(FILT_ORDER);
}
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] nvenc: H264 and HEVC encoders
Module: libav
Branch: master
Commit: b08caa87c35a768ec0abb16b1e99c3a85f1df28e
Author:Luca Barbato lu_z...@gentoo.org
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sat May 30 16:40:13 2015 +0200
nvenc: H264 and HEVC encoders
Partially based on the work of Timo Rothenpieler t...@rothenpieler.org
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
Changelog |1 +
configure | 16 +
libavcodec/Makefile |3 +
libavcodec/allcodecs.c |2 +
libavcodec/nvenc.c | 1136 +++
libavcodec/nvenc.h | 135 ++
libavcodec/nvenc_h264.c | 110 +
libavcodec/nvenc_hevc.c | 108 +
libavcodec/version.h|2 +-
9 files changed, 1512 insertions(+), 1 deletion(-)
diff --git a/Changelog b/Changelog
index 41fe7c8..2c268ed 100644
--- a/Changelog
+++ b/Changelog
@@ -31,6 +31,7 @@ version next:
- Canopus HQ/HQA decoder
- Automatically rotate videos based on metadata in avconv
- improved Quickdraw compatibility
+- NVIDIA NVENC-accelerated H.264 and HEVC encoding support
version 11:
diff --git a/configure b/configure
index d59f5e4..c1e516c 100755
--- a/configure
+++ b/configure
@@ -220,6 +220,7 @@ External library support:
--enable-libxvid enable Xvid encoding via xvidcore,
native MPEG-4/Xvid encoder exists [no]
--enable-mmalenable decoding via MMAL [no]
+ --enable-nvenc enable encoding via NVENC [no]
--enable-openssl enable openssl [no]
--enable-x11grab enable X11 grabbing (legacy) [no]
--enable-zlibenable zlib [autodetect]
@@ -1191,6 +1192,7 @@ EXTERNAL_LIBRARY_LIST=
libxcb_xfixes
libxvid
mmal
+nvenc
openssl
x11grab
zlib
@@ -1483,6 +1485,7 @@ SYSTEM_FUNCS=
inet_aton
isatty
jack_port_get_latency_range
+LoadLibrary
localtime_r
mach_absolute_time
MapViewOfFile
@@ -1780,6 +1783,8 @@ mpegaudio_select=mpegaudiodsp
mpegaudiodsp_select=dct
mpegvideo_select=blockdsp hpeldsp idctdsp me_cmp videodsp
mpegvideoenc_select=me_cmp mpegvideo pixblockdsp qpeldsp
+nvenc_deps_any=dlopen LoadLibrary
+nvenc_extralibs='$ldl'
qsvdec_select=qsv
qsvenc_select=qsv
@@ -1849,11 +1854,13 @@ h263i_decoder_select=h263_decoder
h263p_encoder_select=h263_encoder
h264_decoder_select=cabac golomb h264chroma h264dsp h264pred h264qpel
startcode videodsp
h264_decoder_suggest=error_resilience
+h264_nvenc_encoder_deps=nvenc
h264_qsv_decoder_deps=libmfx
h264_qsv_decoder_select=h264_mp4toannexb_bsf h264_parser qsvdec
h264_qsv_hwaccel
h264_qsv_encoder_deps=libmfx
h264_qsv_encoder_select=qsvenc
hevc_decoder_select=bswapdsp cabac golomb videodsp
+hevc_nvenc_encoder_deps=nvenc
huffyuv_decoder_select=bswapdsp huffyuvdsp
huffyuv_encoder_select=bswapdsp huffman huffyuvencdsp
iac_decoder_select=imc_decoder
@@ -3822,6 +3829,7 @@ die_license_disabled gpl x11grab
die_license_disabled nonfree libfaac
die_license_disabled nonfree libfdk_aac
+die_license_disabled nonfree nvenc
die_license_disabled nonfree openssl
die_license_disabled version3 libopencore_amrnb
@@ -4188,6 +4196,7 @@ check_func_headers windows.h CoTaskMemFree -lole32
check_func_headers windows.h GetProcessAffinityMask
check_func_headers windows.h GetProcessTimes
check_func_headers windows.h GetSystemTimeAsFileTime
+check_func_headers windows.h LoadLibrary
check_func_headers windows.h MapViewOfFile
check_func_headers windows.h SetConsoleTextAttribute
check_func_headers windows.h Sleep
@@ -4340,6 +4349,13 @@ if enabled libdc1394; then
die ERROR: No version of libdc1394 found
fi
+if enabled nvenc; then
+check_header cuda.h || die ERROR: cuda.h not found.;
+check_header nvEncodeAPI.h || die ERROR: nvEncodeAPI.h not found.;
+check_cpp_condition nvEncodeAPI.h NVENCAPI_MAJOR_VERSION = 5 ||
+die ERROR: NVENC API version 4 or older is not supported;
+fi
+
if check_pkg_config sdl SDL_events.h SDL_PollEvent; then
check_cpp_condition SDL.h (SDL_MAJOR_VERSION16 | SDL_MINOR_VERSION8 |
SDL_PATCHLEVEL) = 0x010201 $sdl_cflags
check_cpp_condition SDL.h (SDL_MAJOR_VERSION16 | SDL_MINOR_VERSION8 |
SDL_PATCHLEVEL) 0x010300 $sdl_cflags
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 96cc258..629d25d 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -82,6 +82,7 @@ OBJS-$(CONFIG_MPEGVIDEOENC)+= mpegvideo_enc.o
mpeg12data.o \
motion_est.o ratecontrol.o\
mpegvideoencdsp.o
OBJS-$(CONFIG_PIXBLOCKDSP) += pixblockdsp.o
+OBJS-${CONFIG_NVENC} += nvenc.o
OBJS-$(CONFIG_QPELDSP) += qpeldsp.o
OBJS-$(CONFIG_QSV) += qsv.o
OBJS-$(CONFIG_QSVDEC) += qsvdec.o
@@ -228,11 +229,13 @@ OBJS-$(CONFIG_H264_DECODER)+= h264.o
h264_cabac.o h264_cavlc.o \
[libav-commits] jpegls: Check memory allocation
Module: libav
Branch: master
Commit: 7ca603f96f93d988e01d161d611f69a4ecaa3f02
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:54:11 2015 +0200
jpegls: Check memory allocation
---
libavcodec/jpeglsdec.c |6 ++
1 file changed, 6 insertions(+)
diff --git a/libavcodec/jpeglsdec.c b/libavcodec/jpeglsdec.c
index d9b08fb..3b93799 100644
--- a/libavcodec/jpeglsdec.c
+++ b/libavcodec/jpeglsdec.c
@@ -278,10 +278,16 @@ int ff_jpegls_decode_picture(MJpegDecodeContext *s, int
near,
int off = 0, stride = 1, width, shift, ret = 0;
zero = av_mallocz(s-picture_ptr-linesize[0]);
+if (!zero)
+return AVERROR(ENOMEM);
last = zero;
cur = s-picture_ptr-data[0];
state = av_mallocz(sizeof(JLSState));
+if (!state) {
+av_free(zero);
+return AVERROR(ENOMEM);
+}
/* initialize JPEG-LS state from JPEG parameters */
state-near = near;
state-bpp= (s-bits 2) ? 2 : s-bits;
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] ratecontrol: Check memory allocation
Module: libav
Branch: master
Commit: 42df71d9bbb1a5b4bce0bb34417692565c72d390
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:55:12 2015 +0200
ratecontrol: Check memory allocation
---
libavcodec/ratecontrol.c | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/libavcodec/ratecontrol.c b/libavcodec/ratecontrol.c
index 620c854..d0e13b92 100644
--- a/libavcodec/ratecontrol.c
+++ b/libavcodec/ratecontrol.c
@@ -163,6 +163,8 @@ av_cold int ff_rate_control_init(MpegEncContext *s)
return -1;
rcc-entry = av_mallocz(i * sizeof(RateControlEntry));
rcc-num_entries = i;
+if (!rcc-entry)
+return AVERROR(ENOMEM);
/* init all to skipped p frames
* (with b frames we might have a not encoded frame at the end FIXME)
*/
@@ -210,8 +212,10 @@ av_cold int ff_rate_control_init(MpegEncContext *s)
p = next;
}
-if (init_pass2(s) 0)
+if (init_pass2(s) 0) {
+ff_rate_control_uninit(s);
return -1;
+}
// FIXME maybe move to end
if ((s-avctx-flags CODEC_FLAG_PASS2) s-avctx-rc_strategy ==
FF_RC_STRATEGY_XVID) {
@@ -932,6 +936,11 @@ static int init_pass2(MpegEncContext *s)
qscale = av_malloc(sizeof(double) * rcc-num_entries);
blurred_qscale = av_malloc(sizeof(double) * rcc-num_entries);
+if (!qscale || !blurred_qscale) {
+av_free(qscale);
+av_free(blurred_qscale);
+return AVERROR(ENOMEM);
+}
toobig = 0;
for (step = 256 * 256; step 0.001; step *= 0.5) {
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] libtheora: Check memory allocation
Module: libav
Branch: master
Commit: 41658bc88553dab8499e4dfca311559dcbae2674
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:54:29 2015 +0200
libtheora: Check memory allocation
---
libavcodec/libtheoraenc.c |6 ++
1 file changed, 6 insertions(+)
diff --git a/libavcodec/libtheoraenc.c b/libavcodec/libtheoraenc.c
index 462bc94..097336b 100644
--- a/libavcodec/libtheoraenc.c
+++ b/libavcodec/libtheoraenc.c
@@ -101,6 +101,8 @@ static int get_stats(AVCodecContext *avctx, int eos)
if (!eos) {
h-stats = av_fast_realloc(h-stats, h-stats_size,
h-stats_offset + bytes);
+if (!h-stats)
+return AVERROR(ENOMEM);
memcpy(h-stats + h-stats_offset, buf, bytes);
h-stats_offset += bytes;
} else {
@@ -108,6 +110,8 @@ static int get_stats(AVCodecContext *avctx, int eos)
// libtheora generates a summary header at the end
memcpy(h-stats, buf, bytes);
avctx-stats_out = av_malloc(b64_size);
+if (!avctx-stats_out)
+return AVERROR(ENOMEM);
av_base64_encode(avctx-stats_out, b64_size, h-stats,
h-stats_offset);
}
return 0;
@@ -131,6 +135,8 @@ static int submit_stats(AVCodecContext *avctx)
}
h-stats_size = strlen(avctx-stats_in) * 3/4;
h-stats = av_malloc(h-stats_size);
+if (!h-stats)
+return AVERROR(ENOMEM);
h-stats_size = av_base64_decode(h-stats, avctx-stats_in,
h-stats_size);
}
while (h-stats_size - h-stats_offset 0) {
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] xsub: Check memory allocation
Module: libav
Branch: master
Commit: 522d971c7f5921bebcea2fc50e67056afdabb951
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:55:44 2015 +0200
xsub: Check memory allocation
---
libavcodec/xsubdec.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/libavcodec/xsubdec.c b/libavcodec/xsubdec.c
index d01b410..a7dd7ee 100644
--- a/libavcodec/xsubdec.c
+++ b/libavcodec/xsubdec.c
@@ -95,7 +95,13 @@ static int decode_frame(AVCodecContext *avctx, void *data,
int *data_size,
// allocate sub and set values
sub-rects = av_mallocz(sizeof(*sub-rects));
+if (!sub-rects)
+return AVERROR(ENOMEM);
sub-rects[0] = av_mallocz(sizeof(*sub-rects[0]));
+if (!sub-rects[0]) {
+av_freep(sub-rects);
+return AVERROR(ENOMEM);
+}
sub-num_rects = 1;
sub-rects[0]-x = x; sub-rects[0]-y = y;
sub-rects[0]-w = w; sub-rects[0]-h = h;
@@ -104,6 +110,13 @@ static int decode_frame(AVCodecContext *avctx, void *data,
int *data_size,
sub-rects[0]-pict.data[0] = av_malloc(w * h);
sub-rects[0]-nb_colors = 4;
sub-rects[0]-pict.data[1] = av_mallocz(AVPALETTE_SIZE);
+if (!sub-rects[0]-pict.data[0] || !sub-rects[0]-pict.data[1]) {
+av_freep(sub-rects[0]-pict.data[1]);
+av_freep(sub-rects[0]-pict.data[0]);
+av_freep(sub-rects[0]);
+av_freep(sub-rects);
+return AVERROR(ENOMEM);
+}
// read palette
for (i = 0; i sub-rects[0]-nb_colors; i++)
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] wma: Check memory allocation
Module: libav
Branch: master
Commit: 83797da6e36c1aadd85f41ca237dce823fc7bfa1
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:55:35 2015 +0200
wma: Check memory allocation
---
libavcodec/wmaenc.c |4
1 file changed, 4 insertions(+)
diff --git a/libavcodec/wmaenc.c b/libavcodec/wmaenc.c
index e801663..c176daa 100644
--- a/libavcodec/wmaenc.c
+++ b/libavcodec/wmaenc.c
@@ -62,11 +62,15 @@ static av_cold int encode_init(AVCodecContext *avctx)
flags2 = 1;
if (avctx-codec-id == AV_CODEC_ID_WMAV1) {
extradata = av_malloc(4);
+if (!extradata)
+return AVERROR(ENOMEM);
avctx-extradata_size = 4;
AV_WL16(extradata, flags1);
AV_WL16(extradata + 2, flags2);
} else if (avctx-codec-id == AV_CODEC_ID_WMAV2) {
extradata = av_mallocz(10);
+if (!extradata)
+return AVERROR(ENOMEM);
avctx-extradata_size = 10;
AV_WL32(extradata, flags1);
AV_WL16(extradata + 4, flags2);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] x264: Check memory allocation
Module: libav
Branch: master
Commit: 9fb483fede7207c1f3965e5ae1ac0ee82e4d7a4e
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:54:39 2015 +0200
x264: Check memory allocation
---
libavcodec/libx264.c |4
1 file changed, 4 insertions(+)
diff --git a/libavcodec/libx264.c b/libavcodec/libx264.c
index 841b824..d7d1df3 100644
--- a/libavcodec/libx264.c
+++ b/libavcodec/libx264.c
@@ -542,6 +542,8 @@ static av_cold int X264_init(AVCodecContext *avctx)
s = x264_encoder_headers(x4-enc, nal, nnal);
avctx-extradata = p = av_malloc(s);
+if (!p)
+return AVERROR(ENOMEM);
for (i = 0; i nnal; i++) {
/* Don't put the SEI in extradata. */
@@ -549,6 +551,8 @@ static av_cold int X264_init(AVCodecContext *avctx)
av_log(avctx, AV_LOG_INFO, %s\n, nal[i].p_payload+25);
x4-sei_size = nal[i].i_payload;
x4-sei = av_malloc(x4-sei_size);
+if (!x4-sei)
+return AVERROR(ENOMEM);
memcpy(x4-sei, nal[i].p_payload, nal[i].i_payload);
continue;
}
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] aacpsy: Check memory allocation
Module: libav
Branch: master
Commit: 074a1b37325bf6d1483fc23ebf1255e78d998339
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:45:54 2015 +0200
aacpsy: Check memory allocation
---
libavcodec/aacpsy.c |6 ++
1 file changed, 6 insertions(+)
diff --git a/libavcodec/aacpsy.c b/libavcodec/aacpsy.c
index ac58968..6cfe3e3 100644
--- a/libavcodec/aacpsy.c
+++ b/libavcodec/aacpsy.c
@@ -298,6 +298,8 @@ static av_cold int psy_3gpp_init(FFPsyContext *ctx) {
const float num_bark = calc_bark((float)bandwidth);
ctx-model_priv_data = av_mallocz(sizeof(AacPsyContext));
+if (!ctx-model_priv_data)
+return AVERROR(ENOMEM);
pctx = (AacPsyContext*) ctx-model_priv_data;
pctx-chan_bitrate = chan_bitrate;
@@ -349,6 +351,10 @@ static av_cold int psy_3gpp_init(FFPsyContext *ctx) {
}
pctx-ch = av_mallocz(sizeof(AacPsyChannel) * ctx-avctx-channels);
+if (!pctx-ch) {
+av_freep(pctx);
+return AVERROR(ENOMEM);
+}
lame_window_init(pctx, ctx-avctx);
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] svq1: Check memory allocation
Module: libav
Branch: master
Commit: 28fb80dcbf6f63eedfcfffc725a6bf0069d03fab
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:55:19 2015 +0200
svq1: Check memory allocation
---
libavcodec/svq1enc.c | 17 -
1 file changed, 16 insertions(+), 1 deletion(-)
diff --git a/libavcodec/svq1enc.c b/libavcodec/svq1enc.c
index 8677503..08d85b8 100644
--- a/libavcodec/svq1enc.c
+++ b/libavcodec/svq1enc.c
@@ -292,6 +292,8 @@ static int svq1_encode_plane(SVQ1EncContext *s, int plane,
s-motion_val16[plane] = av_mallocz((s-m.mb_stride *
(block_height + 2) + 1) *
2 * sizeof(int16_t));
+if (!s-motion_val8[plane] || !s-motion_val16[plane])
+return AVERROR(ENOMEM);
}
s-m.mb_type = s-mb_type;
@@ -549,6 +551,12 @@ static av_cold int svq1_encode_init(AVCodecContext *avctx)
s-y_block_height * sizeof(int32_t));
s-ssd_int8_vs_int16 = ssd_int8_vs_int16_c;
+if (!s-m.me.temp || !s-m.me.scratchpad || !s-m.me.map ||
+!s-m.me.score_map || !s-mb_type || !s-dummy) {
+svq1_encode_end(avctx);
+return AVERROR(ENOMEM);
+}
+
if (ARCH_PPC)
ff_svq1enc_init_ppc(s);
if (ARCH_X86)
@@ -612,8 +620,15 @@ static int svq1_encode_frame(AVCodecContext *avctx,
AVPacket *pkt,
s-frame_width / (i ? 4 : 1),
s-frame_height / (i ? 4 : 1),
pict-linesize[i],
- s-current_picture-linesize[i]) 0)
+ s-current_picture-linesize[i]) 0) {
+int j;
+for (j = 0; j i; j++) {
+av_freep(s-motion_val8[j]);
+av_freep(s-motion_val16[j]);
+}
+av_freep(s-scratchbuf);
return -1;
+}
// avpriv_align_put_bits(s-pb);
while (put_bits_count(s-pb) 31)
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] huffyuv: Check memory allocation
Module: libav Branch: master Commit: 7d2a6826912b629f17f3b0c5268e334fa123c4c9 Author:Vittorio Giovara vittorio.giov...@gmail.com Committer: Luca Barbato lu_z...@gentoo.org Date: Sun May 31 14:54:03 2015 +0200 huffyuv: Check memory allocation --- libavcodec/huffyuvenc.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavcodec/huffyuvenc.c b/libavcodec/huffyuvenc.c index 6b3ff76..175b256 100644 --- a/libavcodec/huffyuvenc.c +++ b/libavcodec/huffyuvenc.c @@ -155,7 +155,7 @@ static av_cold int encode_init(AVCodecContext *avctx) s-version = 2; avctx-coded_frame = av_frame_alloc(); -if (!avctx-coded_frame) +if (!avctx-extradata || !avctx-stats_out || !avctx-coded_frame) return AVERROR(ENOMEM); avctx-coded_frame-pict_type = AV_PICTURE_TYPE_I; ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] lcl: Check memory allocation
Module: libav Branch: master Commit: 8df5fbf0b0c0ba12b033e61c28fc240f4bccba47 Author:Vittorio Giovara vittorio.giov...@gmail.com Committer: Luca Barbato lu_z...@gentoo.org Date: Sun May 31 14:54:21 2015 +0200 lcl: Check memory allocation --- libavcodec/lclenc.c |2 ++ 1 file changed, 2 insertions(+) diff --git a/libavcodec/lclenc.c b/libavcodec/lclenc.c index 20841bc..2448b9e 100644 --- a/libavcodec/lclenc.c +++ b/libavcodec/lclenc.c @@ -135,6 +135,8 @@ static av_cold int encode_init(AVCodecContext *avctx) assert(avctx-width avctx-height); avctx-extradata= av_mallocz(8); +if (!avctx-extradata) +return AVERROR(ENOMEM); avctx-coded_frame = av_frame_alloc(); if (!avctx-coded_frame) ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] xvid: Check memory allocation
Module: libav
Branch: master
Commit: fef2f4722bcbfe1c3802898b0f5deb6356132c91
Author:Vittorio Giovara vittorio.giov...@gmail.com
Committer: Luca Barbato lu_z...@gentoo.org
Date: Sun May 31 14:54:48 2015 +0200
xvid: Check memory allocation
---
libavcodec/libxvid.c |6 ++
1 file changed, 6 insertions(+)
diff --git a/libavcodec/libxvid.c b/libavcodec/libxvid.c
index 46da773..97ff95b 100644
--- a/libavcodec/libxvid.c
+++ b/libavcodec/libxvid.c
@@ -276,6 +276,8 @@ static int xvid_strip_vol_header(AVCodecContext *avctx,
AVPacket *pkt,
/* We need to store the header, so extract it */
if (!avctx-extradata) {
avctx-extradata = av_malloc(vo_len);
+if (!avctx-extradata)
+return AVERROR(ENOMEM);
memcpy(avctx-extradata, pkt-data, vo_len);
avctx-extradata_size = vo_len;
}
@@ -594,11 +596,15 @@ static av_cold int xvid_encode_init(AVCodecContext *avctx)
if (avctx-intra_matrix) {
intra = avctx-intra_matrix;
x-intra_matrix = av_malloc(sizeof(unsigned char) * 64);
+if (!x-intra_matrix)
+return AVERROR(ENOMEM);
} else
intra = NULL;
if (avctx-inter_matrix) {
inter = avctx-inter_matrix;
x-inter_matrix = av_malloc(sizeof(unsigned char) * 64);
+if (!x-inter_matrix)
+return AVERROR(ENOMEM);
} else
inter = NULL;
___
libav-commits mailing list
libav-commits@libav.org
https://lists.libav.org/mailman/listinfo/libav-commits
[libav-commits] Update changelog for v11.4
Module: libav Branch: release/11 Commit: 0069d4597bda5723fbcae410784bbfa4750df706 Author:Reinhard Tartler siret...@tauware.de Committer: Reinhard Tartler siret...@tauware.de Date: Sun May 24 10:36:42 2015 +0200 Update changelog for v11.4 --- Changelog | 36 RELEASE |2 +- 2 files changed, 37 insertions(+), 1 deletion(-) diff --git a/Changelog b/Changelog index 6b8cabc..9fc1556 100644 --- a/Changelog +++ b/Changelog @@ -1,6 +1,42 @@ Entries are sorted chronologically from oldest to youngest within each release, releases are sorted from youngest to oldest. +version 11.4: + + - h264: Make sure reinit failures mark the context as not initialized (CVE-2015-3417) + - msrle: Use FFABS to determine the frame size in msrle_decode_pal4 (CVE-2015-3395) + - cavs: Remove an unneeded scratch buffer + - configure: Disable i686 for i586 and lower CPUs (debian/783082) + - mjpegenc: Fix JFIF header byte ordering (bug/808) + - nut: Make sure to clean up on read_header failure + - png: Set the color range as full range + - avi: Validate sample_size + - nut: Check chapter creation in decode_info_header + - alac: Reject rice_limit 0 if compression is used + - ape: Support _ files with nblock smaller than 64 + - mux: Do not leave stale side data pointers in ff_interleave_add_packet() + - avresample: Reallocate the internal buffer to the correct size (bug/825) + - mpegts: Update the PSI/SI table only if the version change + - rtsp: Make sure we don't write too many transport entries into a fixed-size array + - rtpenc_jpeg: Handle case of picture dimensions not dividing by 8 + - mov: Fix little endian audio detection + - x86: Put COPY3_IF_LT under HAVE_6REGS (gentoo/541930) + - roqvideoenc: set enc-avctx in roq_encode_init + - mp3: Properly use AVCodecContext API + - libvpx: Fix mixed use of av_malloc() and av_reallocp() + - Revert lavfi: always check av_expr_parse_and_eval() return value + - alsdec: only adapt order for positive max_order + - alsdec: check sample pointer range in revert_channel_correlation + - aacpsy: correct calculation of minath in psy_3gpp_init + - alsdec: limit avctx-bits_per_raw_sample to 32 + - aasc: return correct buffer size from aasc_decode_frame + - matroskadec: fix crash when parsing invalid mkv + - avconv: do not overwrite the stream codec context for streamcopy + - webp: ensure that each transform is only used once + - h264_ps: properly check cropping parameters against overflow + - hevc: zero the correct variables on invalid crop parameters + - hevc: make the crop sizes unsigned + version 11.3: - utvideodec: Handle slice_height being zero (CVE-2014-9604) diff --git a/RELEASE b/RELEASE index 8bb4222..1c7134d 100644 --- a/RELEASE +++ b/RELEASE @@ -1 +1 @@ -11.3 +11.4 ___ libav-commits mailing list libav-commits@libav.org https://lists.libav.org/mailman/listinfo/libav-commits
