ffmpeg | branch: master | Martin Storsjö | Sat Feb 25
00:20:25 2017 +0200| [32e273c111d8700dde895b80741622afc285ad3c] | committer:
Martin Storsjö
arm: vp9itxfm16: Avoid reloading the idct32 coefficients
Keep the idct32 coefficients in narrow form in q6-q7, and idct16
coefficients in
ffmpeg | branch: master | Martin Storsjö | Mon Jan 9
00:04:19 2017 +0200| [21c89f3a26bb1331381b90e653277585447cfbb3] | committer:
Martin Storsjö
arm/aarch64: vp9: Fix vertical alignment
Align the second/third operands as they usually are.
Due to the wildly varying sizes of the written out
ffmpeg | branch: master | Martin Storsjö | Sun Feb 26
22:13:10 2017 +0200| [70317b25aa35c0907720e4d2b7686408588c07aa] | committer:
Martin Storsjö
arm/aarch64: vp9itxfm: Skip loading the min_eob pointer when it won't be used
In the half/quarter cases where we don't use the min_eob ar
ffmpeg | branch: master | Martin Storsjö | Sun Feb 26
00:24:50 2017 +0200| [25ced1eb1c6cf23099a0fa07ba503b4581713561] | committer:
Martin Storsjö
aarch64: vp9itxfm16: Fix a typo in a comment
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Martin Storsjö | Fri Feb 24
17:39:00 2017 +0200| [eabc5abf949bb8cadafe33df73adacf87ee4c5e0] | committer:
Martin Storsjö
arm: vp9itxfm16: Do a simpler half/quarter idct16/idct32 when possible
This work is sponsored by, and copyright, Google.
This avoids loading and
ffmpeg | branch: master | Martin Storsjö | Fri Feb 24
16:49:12 2017 +0200| [d564c9018f8a45c1f4c38f02844186545582531d] | committer:
Martin Storsjö
aarch64: vp9itxfm16: Move the load_add_store macro out from the itxfm16 pass2
function
This allows reusing the macro for a separate implementation
ffmpeg | branch: master | Martin Storsjö | Fri Feb 24
16:02:23 2017 +0200| [0ea603203d1a46ea36cbaa3fb53d6fc69f5367ad] | committer:
Martin Storsjö
arm: vp9itxfm16: Make the larger core transforms standalone functions
This work is sponsored by, and copyright, Google.
This reduces the code size
ffmpeg | branch: master | Martin Storsjö | Sun Feb 26
00:28:12 2017 +0200| [d613251622d6807fd685452770612a743d7f20a0] | committer:
Martin Storsjö
aarch64: vp9itxfm16: Avoid .irp when it doesn't save any lines
This makes the code a bit more readable.
Signed-off-by: Martin Storsjö
&
ffmpeg | branch: master | Martin Storsjö | Sun Feb 26
00:38:48 2017 +0200| [61b8a9ea2930130a1fecf6c06a391a18d8b95d83] | committer:
Martin Storsjö
aarch64: vp9itxfm16: Do a simpler half/quarter idct16/idct32 when possible
This work is sponsored by, and copyright, Google.
This avoids loading
ffmpeg | branch: master | Martin Storsjö | Sat Feb 25
00:07:22 2017 +0200| [c1619318e540a214c730c6a300ebee0a4f450ba2] | committer:
Martin Storsjö
arm: vp9itxfm16: Fix vertical alignment
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Martin Storsjö | Sun Feb 26
13:43:10 2017 +0200| [b76533f105cc01f6fb64199309fab84ba22da725] | committer:
Martin Storsjö
aarch64: vp9itxfm16: Restructure the idct32 store macros
This avoids concatenation, which can't be used if the whole macro
is wrapped w
ffmpeg | branch: master | Martin Storsjö | Fri Feb 24
16:10:25 2017 +0200| [0f2705e66b1f7f9ae900667c400e46fa0e4f15a7] | committer:
Martin Storsjö
aarch64: vp9itxfm16: Make the larger core transforms standalone functions
This work is sponsored by, and copyright, Google.
This reduces the code
ffmpeg | branch: master | Martin Storsjö | Fri Feb 24
17:36:05 2017 +0200| [b46d37e93ab1d38281ce2cc54b70de2c422d06e5] | committer:
Martin Storsjö
arm: vp9itxfm16: Use the right lane size
This makes the code slightly clearer, but doesn't make any functional
difference.
Signed-off-by: M
ffmpeg | branch: master | Martin Storsjö | Thu Sep 15
14:21:38 2016 +0300| [df3795025337479a639cb3cd26c93a4e82ccd4db] | committer:
Martin Storsjö
rtsp: Fix a crash with the RTSP muxer
This was introduced in bc2a32969e.
The whole block that the statement was added to is only
relevant when
ffmpeg | branch: master | Martin Storsjö | Fri Sep 30
12:05:55 2016 +0300| [9b2ccafb480c94fd09cfb24306d5296dc013cf5b] | committer:
Martin Storsjö
aarch64: Add missing sign extension in ff_h264_idct8_add_neon
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git
ffmpeg | branch: master | Martin Storsjö | Thu Oct 13
15:24:54 2016 +0300| [7395784ba72742b6daa62d35db4028e09f3fdf06] | committer:
Martin Storsjö
rtmpproto: Check the return from ff_amf_read_string
If this failed, we used to continue with an uninitialized
filename buffer.
CC: libav-sta
ffmpeg | branch: master | Martin Storsjö | Thu Oct 13
15:25:33 2016 +0300| [d6ded94036e43a04889f4ff2813a7f7dd60b82fe] | committer:
Martin Storsjö
rtmpproto: Lengthen the filename buffer when receiving streams
Some applications such as Adobe FME append lots of parameters
here, making it easily
ffmpeg | branch: master | Martin Storsjö | Thu Oct 13
16:17:11 2016 +0300| [8b5e0d17e70400eaf5dc3845b5c1df8b2b88d830] | committer:
Martin Storsjö
rtmpproto: Send chunk size on the network channel
This makes sure that e.g. Adobe FME actually reacts to it. As long
as the value we've
ffmpeg | branch: master | Martin Storsjö | Fri Oct 14
15:55:52 2016 +0300| [7d8d726be7dc46343ab1c98c339c1ed44bcb07c1] | committer:
Martin Storsjö
rtmpproto: Don't include a client version in the unencrypted C1 handshake
According to the public RTMP specification, these 4 bytes should
be
ffmpeg | branch: master | Martin Storsjö | Fri Oct 14
11:30:12 2016 +0300| [9f23f77a532ca9c2b7dc4b5328bc413e4f6f5b56] | committer:
Martin Storsjö
rtmpproto: Don't include the libavformat version as "clientid"
When acting as server, the server can include a "clientid"
ffmpeg | branch: master | Martin Storsjö | Fri Sep 30
11:49:47 2016 +0300| [a05cc56124b4f1237f6355784de821e3290ddb44] | committer:
Martin Storsjö
checkasm: arm/aarch64: Fix the amount of space reserved for stack parameters
Even if MAX_ARGS - 2 (for arm) or MAX_ARGS - 7 (for aarch64
ffmpeg | branch: master | Martin Storsjö | Fri Sep 30
11:39:22 2016 +0300| [f1b3e131385176c3c9d9783b25047856a0dcebf6] | committer:
Martin Storsjö
checkasm: aarch64: Clobber the stack before calling functions
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git
ffmpeg | branch: master | Martin Storsjö | Fri Sep 30
12:06:49 2016 +0300| [c91d6a33f872574c95c8784277cf60ffcf6bff4f] | committer:
Martin Storsjö
checkasm: aarch64: Add filler args to make sure all parameters are passed on
the stack
This, combined with clobbering the stack space prior to the
ffmpeg | branch: master | Martin Storsjö | Tue Oct 18
13:48:56 2016 +0300| [dd5d4a0e1e3a30a254d1a57ecbdcedf230c6014b] | committer:
Martin Storsjö
checkasm: aarch64: Don't clobber x29 in checkasm_stack_clobber
x29 (FP) is a callee saved register and should be restored on
return. Inste
ffmpeg | branch: master | Martin Storsjö | Fri Oct 21
12:40:05 2016 +0300| [016387fe0fe3eff1a03ec0673bf4d2967f6cad94] | committer:
Martin Storsjö
rtmpdh: Don't use the OpenSSL DH struct
Instead use our own struct, which we already use when using
gcrypt and gnutls.
In OpenSSL 1.1, t
ffmpeg | branch: master | Martin Storsjö | Tue Oct 25
21:33:34 2016 +0300| [f22363c72968f1a1fc4881d8695ec7068b0aa03c] | committer:
Martin Storsjö
openssl: Avoid double semicolons after the GET_BIO_DATA macro
When the macro is expanded with a semicolon following it and the
macro itself
ffmpeg | branch: master | Martin Storsjö | Thu Nov 3
09:12:02 2016 +0200| [2e55e26b40e269816bba54da7d0e03955731b8fe] | committer:
Martin Storsjö
vp9: Flip the order of arguments in MC functions
This makes it match the pattern already used for VP8 MC functions.
This also makes the signature
ffmpeg | branch: master | Martin Storsjö | Thu Nov 3
09:12:08 2016 +0200| [ffbd1d2b0002576ef0d976a41ff959c635373fdc] | committer:
Martin Storsjö
arm: vp9: Add NEON optimizations of VP9 MC functions
This work is sponsored by, and copyright, Google.
The filter coefficients are signed values
ffmpeg | branch: master | Martin Storsjö | Thu Nov 3
10:39:24 2016 +0200| [1a469a5e423bdad779b8534247dea8cc86169b88] | committer:
Martin Storsjö
options_table: Remove a now unnecessary include of config.h
The include of config.h was added in 2012 in 1d9c2dc8, due to
the use of
ffmpeg | branch: master | Martin Storsjö | Fri Nov 4
09:16:22 2016 +0200| [392caa65df3efa8b2d48a80f08a6af4892c61c08] | committer:
Martin Storsjö
arm: vp9mc: Insert a literal pool at the middle of the file
This fixes errors like this when building non-pic binaries with armv6
as baseline
ffmpeg | branch: master | Martin Storsjö | Thu Nov 3
14:14:12 2016 +0200| [d1ef1b9eaa45043ea5df5a004fb37243e05da61d] | committer:
Martin Storsjö
configure: Silence lld-link when getting the version number
In recent lld-link versions, this command prints the version to
stdout, but also prints
ffmpeg | branch: master | Martin Storsjö | Sat Nov 5
13:18:53 2016 +0200| [a4cfcddcb0f76e837d5abc06840c2b26c0e8aefc] | committer:
Martin Storsjö
vp9: Make the subpel filters non-static
Make them aligned, to allow efficient access to them from simd.
Signed-off-by: Martin Storsjö
> h
ffmpeg | branch: master | Martin Storsjö | Wed Nov 9
11:47:57 2016 +0200| [c44a8a3eabcd6acd2ba79f32ec8a432e6ebe552c] | committer:
Martin Storsjö
aarch64: Add an offset parameter to the movrel macro
With apple tools, the linker fails with errors like these, if the
offset is negative:
ld: in
ffmpeg | branch: master | Martin Storsjö | Thu Nov 10
11:06:26 2016 +0200| [383d96aa2229f644d9bd77b821ed3a309da5e9fc] | committer:
Martin Storsjö
aarch64: vp9: Add NEON optimizations of VP9 MC functions
This work is sponsored by, and copyright, Google.
These are ported from the ARM version
ffmpeg | branch: master | Martin Storsjö | Thu Nov 10
11:07:39 2016 +0200| [557c1675cf0e803b2fee43b4c8b58433842c84d0] | committer:
Martin Storsjö
arm: vp9mc: Minor adjustments from review of the aarch64 version
This work is sponsored by, and copyright, Google.
The speedup for the large
ffmpeg | branch: master | Martin Storsjö | Wed Nov 9
11:54:25 2016 +0200| [6a62795d4051f435a9a2c59395d96913693922f8] | committer:
Martin Storsjö
aarch64: h264idct: Use the offset parameter to movrel
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Martin Storsjö | Thu Nov 10
13:23:38 2016 +0200| [824e8c284054f323f854892d1b4739239ed1fdc7] | committer:
Martin Storsjö
arm: Clear the gp register alias at the end of functions
We reset .Lpic_gp to zero at the start of each function, which means
that the logic within
ffmpeg | branch: master | Martin Storsjö | Thu Nov 10
13:25:36 2016 +0200| [11623217e3c9b859daee544e31acdd0821b61039] | committer:
Martin Storsjö
arm: vp9mc: Use a different helper register for PIC loads
This fixes crashes since 557c1675cf in linux PIC builds.
Previously, movrelx silently
ffmpeg | branch: master | Martin Storsjö | Sat Oct 8
22:36:18 2016 +0300| [a67ae67083151f2f9595a1f2d17b601da19b939e] | committer:
Martin Storsjö
arm: vp9: Add NEON itxfm routines
This work is sponsored by, and copyright, Google.
For the transforms up to 8x8, we can fit all the data
ffmpeg | branch: master | Martin Storsjö | Mon Oct 10
09:48:03 2016 +0300| [dd299a2d6d4d1af9528ed35a8131c35946be5973] | committer:
Martin Storsjö
arm: vp9: Add NEON loop filters
This work is sponsored by, and copyright, Google.
The implementation tries to have smart handling of cases
where
ffmpeg | branch: master | Martin Storsjö | Sun Nov 13
23:53:08 2016 +0200| [3c9546dfafcdfe8e7860aff9ebbf609318220f29] | committer:
Martin Storsjö
aarch64: vp9: Add NEON itxfm routines
This work is sponsored by, and copyright, Google.
These are ported from the ARM version; thanks to the
ffmpeg | branch: master | Martin Storsjö | Sat Nov 12
21:25:50 2016 +0200| [52d196fb30fb6628921b5f1b31e7bd11eb7e1d9a] | committer:
Martin Storsjö
arm: vp9itxfm: Simplify txfm string comparisons
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Martin Storsjö | Mon Nov 14
00:07:27 2016 +0200| [9d2afd1eb8c5cc0633062430e66326dbf98c99e0] | committer:
Martin Storsjö
aarch64: vp9: Implement NEON loop filters
This work is sponsored by, and copyright, Google.
These are ported from the ARM version; thanks to the
ffmpeg | branch: master | Martin Storsjö | Mon Nov 14
23:44:06 2016 +0200| [81d7f0bbca837afda1f7e60d3ae52ab1360ab44b] | committer:
Martin Storsjö
checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately
The dc-only mode is already checked to work correctly above, but this
ffmpeg | branch: master | Martin Storsjö | Fri Nov 18
12:09:06 2016 +0200| [4d960a11855f4212eb3a4e470ce890db7f01df29] | committer:
Martin Storsjö
aarch64: vp9itxfm: Use w3 instead of x3 for the int eob parameter
The clobbering tests in checkasm are only invoked when testing
correctness, so
ffmpeg | branch: master | Martin Storsjö | Tue May 21
10:21:37 2013 +0300| [286ab878bd39b56008035638227b3ecb8ec5bbb7] | committer:
Martin Storsjö
fate.sh: Allow setting other make flags for running tests
If makeopts_fate is set, these makeopts are used for running the
tests instead of the
ffmpeg | branch: master | Martin Storsjö | Fri Nov 18
10:09:12 2016 +0200| [effc1430b2fe5997d9d55bf28dc507c27125eb27] | committer:
Martin Storsjö
Revert "checkasm: vp9dsp: Benchmark the dc-only version of idct_idct separately"
This reverts commit 81d7f0bbca837afda1f7e60d3ae52a
ffmpeg | branch: master | Martin Storsjö | Tue Nov 22
11:32:25 2016 +0200| [721bc37522c5c1d6a8c3cea5e9c3fcde8d256c05] | committer:
Martin Storsjö
arm/aarch64: vp9itxfm: Fix indentation of macro arguments
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=com
ffmpeg | branch: master | Martin Storsjö | Tue Nov 22
13:52:55 2016 +0200| [79566ec8c77969d5f9be533de04b1349834cca62] | committer:
Martin Storsjö
arm: vp9itxfm: Rename a macro parameter to fit better
Since the same parameter is used for both input and output,
the name inout is more fitting
ffmpeg | branch: master | Martin Storsjö | Tue Nov 22
15:47:17 2016 +0200| [2f99117f6ff24ce5be2abb9e014cb8b86c2aa0e0] | committer:
Martin Storsjö
aarch64: vp9itxfm: Don't repeatedly set x9 when nothing overwrites it
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gi
ffmpeg | branch: master | Martin Storsjö | Wed Nov 23
23:27:26 2016 +0200| [537b5b773b317af79d3a5b576ee9683e15ed84f6] | committer:
Martin Storsjö
rtmpdh: Do global initialization before running the test
The rtmpdh code can use crypto libraries which may require
a process global init. (gcrypt
ffmpeg | branch: master | Martin Storsjö | Thu Sep 13
12:55:32 2012 +0300| [233d50b275dd7cf6cc0656851e670e1b2dfba56f] | committer:
Diego Biurrun
qt-faststart: Do not try to use fancy 64-bit seeking functions on mingw32ce
These functions are not available on mingw32ce.
Signed-off-by: Diego
ffmpeg | branch: master | Martin Storsjö | Mon Nov 28
11:05:18 2016 +0200| [3c87039a404c5659ae9bf7454a04e186532eb40b] | committer:
Martin Storsjö
arm: vp9itxfm: Only reload the idct coeffs for the iadst_idct combination
This avoids reloading them if they haven't been clobbered, if the
ffmpeg | branch: master | Martin Storsjö | Fri Nov 18
11:37:16 2016 +0200| [9c8bc74c2b40537b0997f646c87c008042d788c2] | committer:
Martin Storsjö
arm: vp9itxfm: Skip empty slices in the first pass of idct_idct 16x16 and 32x32
This work is sponsored by, and copyright, Google.
Previously all
ffmpeg | branch: master | Martin Storsjö | Fri Nov 18
12:26:04 2016 +0200| [cad42fadcd2c2ae1b3676bb398844a1f521a2d7b] | committer:
Martin Storsjö
aarch64: vp9itxfm: Skip empty slices in the first pass of idct_idct 16x16 and
32x32
This work is sponsored by, and copyright, Google.
Previously
ffmpeg | branch: master | Martin Storsjö | Thu Dec 15
09:45:49 2016 +0200| [cdcfa97dc49d83b5eefd0a651db6bb0a6f98e8f2] | committer:
Martin Storsjö
libavformat: Fix a faulty api deprecation guard in prepare_input_packet
This seems to have been added by mistake in 11de006b, by not
noticing the
ffmpeg | branch: master | Martin Storsjö | Mon Dec 19
10:38:33 2016 +0200| [a0c443a3980dc22eb02b067ac4cb9ffa2f9b04d2] | committer:
Martin Storsjö
aarch64: vp9itxfm: Use the offset parameter to movrel
This fixes build failures for iOS, broken since cad42fadcd2c.
Signed-off-by: Martin Storsjö
ffmpeg | branch: master | Martin Storsjö | Thu Dec 15
10:24:20 2016 +0200| [131644677970a3c4a0096270ea2a5b5d437c2e63] | committer:
Martin Storsjö
http: Check for negative chunk sizes
A negative chunk size is illegal and would end up used as
length for memcpy, where it would lead to memory
ffmpeg | branch: master | Martin Storsjö | Thu Dec 29
00:15:17 2016 +0200| [5c83b4d550ea42653fece092987bab56ccc32ead] | committer:
Martin Storsjö
fate: Unset the sig variable if ignoring a test failure
Otherwise the .rep file would still contain a signal instead of a
zero, even if the process
ffmpeg | branch: master | Martin Storsjö | Thu Dec 29
14:20:51 2016 +0200| [eef860dd92538764f4ab7872812914ff10384268] | committer:
Martin Storsjö
fate: Tweak printing of ignored tests
Use a tab instead of two spaces, skip the fate prefix for the test name.
This makes IGNORE line fit in even
ffmpeg | branch: master | Martin Storsjö | Fri Dec 9
12:36:28 2016 +0200| [c536e5e8698110c139b1c17938998a5547550aa3] | committer:
Martin Storsjö
arm: vp9mc: Fix vertical alignment of operands
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Martin Storsjö | Fri Dec 16
22:59:45 2016 +0200| [85ad5ea72ce3983947a3b07e4b35c66cb16dfaba] | committer:
Martin Storsjö
aarch64: vp9mc: Fix a comment to refer to a register with the right name
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.
ffmpeg | branch: master | Martin Storsjö | Thu Dec 15
12:00:18 2016 +0200| [65074791e8f8397600aacc9801efdd1eb6e3] | committer:
Martin Storsjö
aarch64: vp9dsp: Fix vertical alignment in the init file
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=com
ffmpeg | branch: master | Martin Storsjö | Fri Dec 30
21:45:51 2016 +0200| [4e62b57ee03928c12a3119dcaf78ffa1f4d6985f] | committer:
Martin Storsjö
fate: Skip the checkasm test if CONFIG_STATIC is disabled
When building DLLs with MSVC, CONFIG_STATIC is disabled (see
d66c52c2b3694 for a more
ffmpeg | branch: master | Martin Storsjö | Wed Jan 25
15:11:23 2017 +0200| [f96d07f4ec4193fb5293d7ac8f1324aac3c3ea07] | committer:
Martin Storsjö
configure: Add quotes around a variable which might be empty
If we only have a target compiler but no host compiler, the $type
variable will be
ffmpeg | branch: master | Martin Storsjö | Wed May 31
12:53:32 2017 +0300| [47c43ce36f0c48b800b5367674839c16f10dc597] | committer:
Hendrik Leppkes
configure: Fix the msvcrt version check for mingw32
This was actually broken when committed in 46e3936fb04; the
test never succeeded, and thus
ffmpeg | branch: release/3.3 | Martin Storsjö | Wed May 31
12:53:32 2017 +0300| [1cbeb16187c87655e1affaae075a9646aeb465e6] | committer:
Hendrik Leppkes
configure: Fix the msvcrt version check for mingw32
This was actually broken when committed in 46e3936fb04; the
test never succeeded, and
ffmpeg | branch: master | Martin Storsjö | Fri Jun 9
12:05:00 2017 +0300| [3125a4a8a8fc7dde938bff6a668f53b26bf47860] | committer: wm4
d3d11va: Link directly to dxgi.dll and d3d11.dll functions if LoadLibrary is
unavailable
When targeting the UWP API subset, the LoadLibrary function is not
ffmpeg | branch: master | Martin Storsjö | Fri Jun 30
12:49:49 2017 +0300| [d086e40459abe96a0765ee7f29fc297bf446135d] | committer:
Derek Buitenhuis
lavf: Remove codec_tag from dashenc and smoothstreamingenc
Skip the codec_tag altogether here, to let the user (try to) set
whichever codec/tag
ffmpeg | branch: master | Martin Storsjö | Tue Jul 4
20:03:00 2017 +0300| [9042402ec78ad050eb7ad129610450ef90371df4] | committer: wm4
d3d11va: Check WINAPI_FAMILY instead of HAVE_LOADLIBRARY
If using the winstore compat library, a fallback LoadLibrary
function does exist, that only calls
ffmpeg | branch: master | Martin Storsjö | Tue Jan 31
15:47:00 2017 +0200| [a1a143adb0fd11c474221431417cff25db7d920f] | committer:
Martin Storsjö
rtmp: Rename packet types to closer match the spec
Also rename comments and log messages accordingly,
and add clarifying comments for some
ffmpeg | branch: master | Martin Storsjö | Tue Jan 31
16:15:56 2017 +0200| [15a92e0c402c830b607f905d6bf203b6cfb4fa8c] | committer:
Martin Storsjö
rtmp: Correctly handle the Window Acknowledgement Size packets
This swaps which field is set when the Window Acknowledgement Size
and Set Peer BW
ffmpeg | branch: master | Martin Storsjö | Sat Feb 4
22:16:09 2017 +0200| [3bc5b28d5a191864c54bba60646933a63da31656] | committer:
Martin Storsjö
arm: vp9itxfm: Avoid .irp when it doesn't save any lines
This makes it more readable.
Signed-off-by: Martin Storsjö
> http://git.vide
ffmpeg | branch: master | Martin Storsjö | Thu Dec 1
11:10:19 2016 +0200| [58d87e0f49bcbbc6f426328f53b657bae7430cd2] | committer:
Martin Storsjö
aarch64: vp9itxfm: Restructure the idct32 store macros
This avoids concatenation, which can't be used if the whole macro
is wrapped within an
ffmpeg | branch: master | Martin Storsjö | Tue Feb 7
00:25:19 2017 +0200| [57ec83e4246b21c2f0c068b9151d806737d4497f] | committer:
Martin Storsjö
omx: Use the EOS flag to handle flushing at the end
This avoids having to count the number of frames sent to the codec
and the number of output
ffmpeg | branch: master | Martin Storsjö | Wed Nov 23
10:56:12 2016 +0200| [0331c3f5e8cb6e6b53fab7893e91d1be1bfa979c] | committer:
Martin Storsjö
arm: vp9itxfm: Make the larger core transforms standalone functions
This work is sponsored by, and copyright, Google.
This reduces the code size
ffmpeg | branch: master | Martin Storsjö | Sun Feb 5
22:55:20 2017 +0200| [47b3c2c18d1897f3c753ba0cec4b2d7aa24526af] | committer:
Martin Storsjö
arm: vp9itxfm: Move the load_add_store macro out from the itxfm16 pass2 function
This allows reusing the macro for a separate implementation of the
ffmpeg | branch: master | Martin Storsjö | Tue Nov 22
11:07:38 2016 +0200| [5eb5aec475aabc884d083566f902876ecbc072cb] | committer:
Martin Storsjö
arm: vp9itxfm: Do a simpler half/quarter idct16/idct32 when possible
This work is sponsored by, and copyright, Google.
This avoids loading and
ffmpeg | branch: master | Martin Storsjö | Wed Nov 23
14:03:05 2016 +0200| [115476018d2c97df7e9b4445fe8f6cc7420ab91f] | committer:
Martin Storsjö
aarch64: vp9itxfm: Make the larger core transforms standalone functions
This work is sponsored by, and copyright, Google.
This reduces the code
ffmpeg | branch: master | Martin Storsjö | Tue Jan 3
16:46:17 2017 +0200| [3dd7827258ddaa2e51085d0c677d6f3b1be3572f] | committer:
Martin Storsjö
aarch64: vp9itxfm: Use the right lane sizes in 8x8 for improved readability
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.
ffmpeg | branch: master | Martin Storsjö | Tue Jan 3
23:11:51 2017 +0200| [8476eb0d3ab1f7a52317b23346646389c08fb57a] | committer:
Martin Storsjö
aarch64: vp9itxfm: Update a comment to refer to a register with a different name
Signed-off-by: Martin Storsjö
> http://git.videolan.
ffmpeg | branch: master | Martin Storsjö | Tue Jan 3
16:39:41 2017 +0200| [4da4b2b87f08a1331650c7e36eb7d4029a160776] | committer:
Martin Storsjö
aarch64: vp9itxfm: Share instructions for loading idct coeffs in the 8x8
function
Signed-off-by: Martin Storsjö
> http://git.videolan.
ffmpeg | branch: master | Martin Storsjö | Tue Jan 3
16:38:56 2017 +0200| [3933b86bb93aca47f29fbd493075b0f110c1e3f5] | committer:
Martin Storsjö
arm: vp9itxfm: Share instructions for loading idct coeffs in the 8x8 function
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.
ffmpeg | branch: master | Martin Storsjö | Sun Feb 5
22:53:55 2017 +0200| [79d332ebbde8c0a3e9da094dcfd10abd33ba7378] | committer:
Martin Storsjö
aarch64: vp9itxfm: Move the load_add_store macro out from the itxfm16 pass2
function
This allows reusing the macro for a separate implementation
ffmpeg | branch: master | Martin Storsjö | Tue Jan 3
14:55:46 2017 +0200| [ed8d293306e12c9b79022d37d39f48825ce7f2fa] | committer:
Martin Storsjö
aarch64: vp9itxfm: Use a single lane ld1 instead of ld1r where possible
The ld1r is a leftover from the arm version, where this trick is
beneficial
ffmpeg | branch: master | Martin Storsjö | Tue Nov 22
22:58:35 2016 +0200| [a63da4511d0fee66695ff4afd264ba1dbf1e812d] | committer:
Martin Storsjö
aarch64: vp9itxfm: Do separate functions for half/quarter idct16 and idct32
This work is sponsored by, and copyright, Google.
This avoids loading
ffmpeg | branch: master | Martin Storsjö | Tue Jan 3
16:11:56 2017 +0200| [0c0b87f12d48d4e7f0d3d13f9345e828a3a5ea32] | committer:
Martin Storsjö
aarch64: vp9itxfm: Fix incorrect vertical alignment
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Martin Storsjö | Fri Feb 10
09:20:39 2017 +0200| [bc2589763042dc2384b724b203ec778f35bcebad] | committer:
Martin Storsjö
utvideodec: Add a missing include
This was missing from 77c23704c76, fixing building.
Signed-off-by: Martin Storsjö
> http://git.videolan.
ffmpeg | branch: master | Martin Storsjö | Sat Dec 17
00:55:41 2016 +0200| [5e0c2158fbc774f87d3ce4b7b950ba4d42c4a7b8] | committer:
Martin Storsjö
aarch64: vp9mc: Simplify the extmla macro parameters
Fold the field lengths into the macro.
This makes the macro invocations much more readable
ffmpeg | branch: master | Martin Storsjö | Sat Dec 17
13:14:38 2016 +0200| [388e0d2515bc6bbc9d0c9af1d230bd16cf945fe7] | committer:
Martin Storsjö
aarch64: vp9mc: Calculate less unused data in the 4 pixel wide horizontal filter
No measured speedup on a Cortex A53, but other cores might benefit
ffmpeg | branch: master | Martin Storsjö | Wed Jan 4
13:08:51 2017 +0200| [a76bf8cf1277ef6feb1580b578f5e6ca327e713c] | committer:
Martin Storsjö
arm: vp9itxfm: Optimize 16x16 and 32x32 idct dc by unrolling
This work is sponsored by, and copyright, Google.
Before
ffmpeg | branch: master | Martin Storsjö | Wed Jan 4
12:57:56 2017 +0200| [3fcf788fbbccc4130868e7abe58a88990290f7c1] | committer:
Martin Storsjö
aarch64: vp9itxfm: Optimize 16x16 and 32x32 idct dc by unrolling
This work is sponsored by, and copyright, Google.
Before
ffmpeg | branch: master | Martin Storsjö | Sat Dec 17
13:09:50 2016 +0200| [fea92a4b57d1c328b1de226a5f213a629ee63754] | committer:
Martin Storsjö
arm: vp9mc: Calculate less unused data in the 4 pixel wide horizontal filter
Before:Cortex A7 A8 A9 A53
ffmpeg | branch: master | Martin Storsjö | Fri Jan 13
23:42:28 2017 +0200| [435cd7bc99671bf561193421a50ac6e9d63c4266] | committer:
Martin Storsjö
arm: vp9lpf: Use orrs instead of orr+cmp
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Martin Storsjö | Thu Jan 12
16:52:33 2017 +0200| [e1f9de86f454861b69b199ad801adc2ec6c3b220] | committer:
Martin Storsjö
arm/aarch64: vp9lpf: Calculate !hev directly
Previously we first calculated hev, and then negated it.
Since we were able to schedule the negation
ffmpeg | branch: master | Martin Storsjö | Tue Jan 10
16:49:13 2017 +0200| [e18c39005ad1dbb178b336f691da1de91afd434e] | committer:
Martin Storsjö
arm: vp9lpf: Interleave the start of flat8in into the calculation above
This adds lots of extra .ifs, but speeds it up by a couple cycles,
by
ffmpeg | branch: master | Martin Storsjö | Wed Jan 11
11:58:02 2017 +0200| [07b5136c481d394992c7e951967df0cfbb346c0b] | committer:
Martin Storsjö
aarch64: vp9lpf: Fix broken indentation/vertical alignment
Signed-off-by: Martin Storsjö
> http://git.videolan.org/gitweb.cgi/ffmpeg.git
ffmpeg | branch: master | Martin Storsjö | Tue Jan 10
22:08:50 2017 +0200| [b0806088d3b27044145b20421da8d39089ae0c6a] | committer:
Martin Storsjö
aarch64: vp9lpf: Interleave the start of flat8in into the calculation above
This adds lots of extra .ifs, but speeds it up by a couple cycles,
by
ffmpeg | branch: master | Martin Storsjö | Wed Feb 15
11:06:17 2017 +0200| [030de53e9cc225dc767458aedcc87efd457b4f3b] | committer:
Martin Storsjö
libopenh264dec: Let the framework use the h264_mp4toannexb bitstream filter
This avoids a lot of boilerplate code within the decoder wrapper itself
ffmpeg | branch: master | Martin Storsjö | Thu Feb 16
09:18:25 2017 +0200| [8847eeaa14189885038140fb2b8a7adc7100] | committer:
Martin Storsjö
aarch64: Add parentheses around the offset parameter in movrel
This fixes building with clang for linux with PIC enabled.
Signed-off-by: Martin
1 - 100 of 1069 matches
Mail list logo
