Age | Commit message (Collapse) | Author |
|
and checkasm test
|
|
float_near_abs_eps() failures
This will make understanding failures and adjusting EPS easier
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
|
|
grainextract
grainmerge
average
extremity
negation
|
|
Signed-off-by: Josh de Kock <josh@itanimul.li>
|
|
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
|
|
|
|
Should fix https://ffmpeg.org/pipermail/ffmpeg-devel/2018-February/225058.html
Signed-off-by: Muhammad Faiz <mfcc64@gmail.com>
|
|
|
|
|
|
This reverts commit adff97be5e2ff51c0bb66080c2f904ed40b6c571.
It currently fails on Windows targets.
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
|
|
|
|
This fixes fate-checkasm-hevc_mc on ARMCC 5.0 after adding
NEON HEVC MC assembly.
Signed-off-by: Martin Storsjö <martin@martin.st>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
|
|
|
|
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
|
|
|
|
Tested for 10000 iterations on x86-32
Fixes: Ticket6848
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
|
|
|
|
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
* commit '516c479172755c63063180b0c0953b68b670cdbd':
checkasm: Test more h264 idct variants
See 2d263188ba87b0d1e1120f00eecae4bb0e163f47
Merged-by: James Almer <jamrial@gmail.com>
|
|
* commit '547db1eaecd597031165a2bf637acaaacde52788':
checkasm: Test more h264 idct variants
Merged-by: James Almer <jamrial@gmail.com>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
add_median_pred
add_left_pred : add two func one with acc 0, and one with random acc
add_left_pred16
|
|
* commit 'd05c9cde0e87c23ca42957646bea483dfc09d6bf':
checkasm: aarch64: Specify alignment for the register_init const array
Merged-by: James Almer <jamrial@gmail.com>
|
|
* commit 'e00db9f78bb475ed5103364f61892f4e75ef89ba':
checkasm: hevc: Add a hevc_ prefix to the add_residual functions
Merged-by: James Almer <jamrial@gmail.com>
|
|
* commit '7cb1d9e2dbbe5bf4652be5d78cdd68e956fa3d63':
build: Fine-grained link-time dependency settings
Also included are bug fix commits 5ff3b5cafcc685b6936d16602b0f80aa09a95870,
d9da7151eef7fc469787e7298196cea291acfd82 and
5e27ef800bfa2be17a6353ddedac6b7400e4624f.
Merged-by: James Almer <jamrial@gmail.com>
|
|
Also modify the required alignment, to 32 instead of 16
for several codecs
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
Signed-off-by: Martin Storsjö <martin@martin.st>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
On ARM platforms, accessing the PMU registers requires special user
access permissions. Since there is no other way to get accurate timers,
the current implementation of timers in FFmpeg rely on these registers.
Unfortunately, enabling user access to these registers on Linux is not
trivial, and generally involve compiling a random and unreliable github
kernel module, or patching somehow your kernel.
Such module is very unlikely to reach the upstream anytime soon. Quoting
Robin Murphin from ARM:
> Say you do give userspace direct access to the PMU; now run two or more
> programs at once that believe they can use the counters for their own
> "minimal-overhead" profiling. Have fun interpreting those results...
>
> And that's not even getting into the implications of scheduling across
> different CPUs, CPUidle, etc. where the PMU state is completely beyond
> userspace's control. In general, the plan to provide userspace with
> something which might happen to just about work in a few corner cases,
> but is meaningless, misleading or downright broken in all others, is to
> never do so.
As a result, the alternative is to use the Performance Monitoring Linux
API which makes use of these registers internally (assuming the PMU of
your ARM board is supported in the kernel, which is definitely not a
given...).
While the Linux API is obviously cross platform, it does have a
significant overhead which needs to be taken into account. As a result,
that mode is only weakly enabled on ARM platforms exclusively.
Note on the non flexibility of the implementation: the timers (native
FFmpeg vs Linux API) are selected at compilation time to prevent the
need of function calls, which would result in a negative impact on the
cycle counters.
|
|
This reverts commit 547db1eaecd597031165a2bf637acaaacde52788.
This commit wasn't supposed to be pushed (yet) since it hasn't
been reviewed.
Signed-off-by: Martin Storsjö <martin@martin.st>
|
|
|
|
grainextract
This was missing from f8d0689d3f.
Fixes checkasm.
|
|
aacpsdsp
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
The function returns a float.
This fixes the test in x86_32 targets.
Signed-off-by: James Almer <jamrial@gmail.com>
|
|
|
|
Missed in the last commit.
|