currently, only float is supported as model input, actually, there
are other data types, this patch adds uint8.
Signed-off-by: Guo, Yejun
---
libavfilter/dnn_backend_native.c | 4 +++-
libavfilter/dnn_backend_tf.c | 28
libavfilter/dnn_interface.h | 10
use TF_Tensor's cpu address to avoid extra memcpy.
Signed-off-by: Guo, Yejun
---
libavfilter/dnn_backend_tf.c | 36
libavfilter/vf_sr.c | 3 ---
2 files changed, 12 insertions(+), 27 deletions(-)
diff --git a/libavfilter/dnn_backend_tf.c
some models such as ssd, yolo have more than one output.
the clean up code in this patch is a little complex, it is because
that set_input_output_tf could be called for many times together
with ff_dnn_execute_model_tf, we have to clean resources for the
case that the two interfaces are called
remove the requirment that the name of DNN model input/output
should be "x"/"y",
Signed-off-by: Guo, Yejun
---
libavfilter/dnn_backend_native.c | 2 +-
libavfilter/dnn_backend_tf.c | 10 +-
libavfilter/dnn_interface.h | 2 +-
libavfilter/vf_sr.c | 4 ++--
4
otherwise, the following check will return error if layer_add_res
is randomly initialized.
Signed-off-by: Guo, Yejun
---
libavfilter/dnn_backend_tf.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavfilter/dnn_backend_tf.c b/libavfilter/dnn_backend_tf.c
index 5bc7f06..9e0c127 100644
---
Currently, within interface set_input_output, the dims/memory of the tensorflow
dnn model output is determined by executing the model with zero input,
actually, the output dims might vary with different input data for networks
such as object detection models faster-rcnn, ssd and yolo.
This patch
remove 'else' since there is always 'return' in 'if' scope,
so the code will be clean for later maintenance
Signed-off-by: Guo, Yejun
---
libavfilter/vf_sr.c | 143 ++--
1 file changed, 71 insertions(+), 72 deletions(-)
diff --git
Hi!
Attached patch allows to run fate from wsl for out-of-tree builds.
Please comment, Carl Eugen
From 8e8742a42a8e945c7d6c90810e6409cbb4210967 Mon Sep 17 00:00:00 2001
From: Carl Eugen Hoyos
Date: Thu, 25 Apr 2019 00:42:48 +0200
Subject: [PATCH] configure: Do not overwrite src symlink if it
2019-04-24 22:27 GMT+02:00, Paul B Mahol :
> On 4/23/19, Carl Eugen Hoyos wrote:
>> Hi!
>>
>> I failed to test attached patch but it seems like a more useful fix for
>> the following (past) warning:
>> libavfilter/vf_frei0r.c:130:17: warning: assignment to ‘char **’ from
>> incompatible pointer
On 4/16/2019 6:48 PM, Mark Thompson wrote:
> On 11/04/2019 04:10, James Almer wrote:
>> On 4/10/2019 3:30 PM, James Almer wrote:
>>> The spec defines the valid range of values to be INT32_MIN + 1 to
>>> INT32_MAX, inclusive.
>>>
>>> Signed-off-by: James Almer
>>> ---
>>> A good example of why
Fixes: Timeout (11sec -> 60ms)
Fixes:
14270/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_PAM_fuzzer-5734809634078720
Found-by: continuous fuzzing process
https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer
---
libavcodec/pnm_parser.c | 29
Fixes: OOM
Fixes:
14198/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AGM_fuzzer-5723579234123776
Found-by: continuous fuzzing process
https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer
---
libavcodec/agm.c | 3 +++
1 file changed, 3
On 4/23/19, Carl Eugen Hoyos wrote:
> Hi!
>
> I failed to test attached patch but it seems like a more useful fix for
> the following (past) warning:
> libavfilter/vf_frei0r.c:130:17: warning: assignment to ‘char **’ from
> incompatible pointer type ‘char *’
>
> Please comment, Carl Eugen
>
LGTM
On Wed, 24 Apr 2019, Tomas Härdin wrote:
mån 2019-04-22 klockan 19:15 +0200 skrev Marton Balint:
This affects the following samples:
samples/ffmpeg-bugs/roundup/issue1775/av_seek_frame_failure.mxf
samples/ffmpeg-bugs/trac/ticket1957/16ch.mxf
samples/ffmpeg-bugs/trac/ticket5016/r0.mxf
Hello,
James Almer:
> cbs_mpeg2_free_slice() calls av_buffer_unref() on extra_information_ref,
> meaning allocating with av_malloc() was not the intention.
>
> Signed-off-by: James Almer
> ---
> Couldn't find any mpeg2 sample containing these fields, so it's untested.
> The leak is obvious
cbs_mpeg2_free_slice() calls av_buffer_unref() on extra_information_ref,
meaning allocating with av_malloc() was not the intention.
Signed-off-by: James Almer
---
Couldn't find any mpeg2 sample containing these fields, so it's untested.
The leak is obvious regardless of that.
Hi Michael, I hope you are doing well,
> On Apr 16, 2019, at 4:03 PM, Michael Niedermayer
> wrote:
>
> Signed PGP part
> On Wed, Apr 10, 2019 at 06:54:53PM -0300, James Almer wrote:
>> On 4/10/2019 6:26 PM, Baptiste Coudurier wrote:
On Apr 9, 2019, at 6:27 PM, James Almer wrote:
On Sun, Apr 21, 2019 at 5:51 PM Jun Li wrote:
> Fix #7144.
> The current packet duration calculation is heuristic, which uses the
> historical durtion as current duration. This commit adds the option
> to buffer one packet and calcuate its duration when next packet arrives.
> Obviously it adds
On 24.04.2019 12:20, Timo Rothenpieler wrote:
> On 24/04/2019 01:38, Carl Eugen Hoyos wrote:
>> 2017-12-29 13:16 GMT+01:00, Cyber Sinh :
>>> Sorry for the diff instead of regular git patch. Here is the patch.
>>
>> Only saw this today after wondering
>> why nobody tried using wsl before...
>
> I'm
>
> From: avih [mailto:avih...@yahoo.com]
> Sent: Wednesday, April 24, 2019 9:22 PM
> To: FFmpeg development discussions and patches
> Cc: Guo, Yejun
> Subject: Re: [FFmpeg-devel] [PATCH V5 1/2] configure: sort
> decoder/encoder/filter/... names in alphabet order
>
> > print_in_columns() {
>
> print_in_columns() {
> - cols=$(expr $ncols / 24)
> - cat | tr ' ' '\n' | sort | pr -r "-$cols" -w $ncols -t
> + # the input should not contain chars such as '*',
> + # otherwise, '*' will be expanded to be all files in the current
> + # working directory which don't begin with a
> log_file(){
> - log BEGIN $1
> - pr -n -t $1 >> $logfile
> - log END $1
> + log BEGIN "$1"
> + log_file_i=1
> + while IFS= read -r log_file_line;do
> + printf '%5s %s\n' "${log_file_i}" "${log_file_line}"
> + log_file_i=$(($log_file_i+1))
> + done < "$1" >>
This patch is based on https://trac.ffmpeg.org/ticket/5680 provided by
Kylie McClain at Wed, 29 Jun 2016 16:37:20 -0400,
and have some changes.
contributor: Kylie McClain
contributor: avih
Signed-off-by: Guo, Yejun
---
configure | 10 +++---
1 file changed, 7 insertions(+), 3
take decoder names an example, with the default page length, shell command
'pr' needs two pages for all the decoder names. The names are firstly printed
in the first page, then in the second page. So, as a whole, the names are
sorted neither in column order nor in row order. It's a little
> From: avih [mailto:avih...@yahoo.com]
> Sent: Wednesday, April 24, 2019 8:16 PM
> To: Guo, Yejun ; FFmpeg development discussions and
> patches
> Subject: Re: [FFmpeg-devel] [PATCH V4 2/2] configure: replace 'pr' with printf
> since busybox does not support pr
>
> > how about add the keyword
> how about add the keyword local which makes the variables not global.
No. It's not portable and not standard, and will break with a lot of shells
which ffmpeg currently supports.
Different shells have different ways to limit a variable scope to the current
function, some don't have a way at
> -Original Message-
> From: Guo, Yejun
> Sent: Wednesday, April 24, 2019 7:02 PM
> To: Guo, Yejun
> Subject: Re: [FFmpeg-devel] [PATCH V4 2/2] configure: replace 'pr' with printf
> since busybox does not support pr
>
> > log_file(){
> > log BEGIN $1
> > -pr -n -t $1 >>
Has a slight speedup.
Can't be carried over to aarch64, since it has no shufps-like instruction.
>From 6b09809a05c0d0bd916f7f6de5b205965bf4b69a Mon Sep 17 00:00:00 2001
From: Lynne
Date: Wed, 24 Apr 2019 12:19:48 +0100
Subject: [PATCH] x86/opusdsp: replace loads with shuffles
Has a slight
./ffmpeg -f lavfi -i yuvtestsrc=duration=1:size=1200x1440 -sws_flags
fast_bilinear \
-s 2400x720 -f rawvideo -vframes 5 -pix_fmt abgr -nostats test.raw
4.27 speedup for hyscale_fast:
24796 UNITS in hyscale_fast,4096 runs, 0 skips
5797 UNITS in hyscale_fast,4096 runs,
On Sun, Apr 14, 2019 at 01:05:03PM +0200, Michael Niedermayer wrote:
> Fixes: Timeout (16sec -> 5sec)
> Fixes:
> 14128/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ARBC_fuzzer-5767365721063424
>
> Found-by: continuous fuzzing process
>
On Tue, Apr 16, 2019 at 06:31:28PM +0200, Paul B Mahol wrote:
> On 4/16/19, Michael Niedermayer wrote:
> > The function in case of n=0 would read more bytes than 0.
> > The end pointer could be beyond the allocated space, which
> > is undefined.
> >
> > Signed-off-by: Michael Niedermayer
> > ---
On Wed, Apr 17, 2019 at 12:21:54AM +0200, Michael Niedermayer wrote:
> Fixes: Assertion failure
> Fixes:
> clusterfuzz-testcase-minimized-media_pipeline_integration_fuzzer-5683096400822272
>
> Found-by: Clusterfuzz
> Reported-by: Dan Sanders
> Signed-off-by: Michael Niedermayer
> ---
>
On Mon, Apr 15, 2019 at 02:40:07AM +0200, Michael Niedermayer wrote:
> Fixes: Timeout (33sec -> 16 milli sec)
> Fixes:
> 14181/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DXTORY_fuzzer-5681840708386816
>
> Found-by: continuous fuzzing process
>
On Mon, Apr 15, 2019 at 07:28:18PM +0200, Michael Niedermayer wrote:
> Fixes: Timeout (158sec -> 36sec)
> Fixes:
> 14214/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DSICINVIDEO_fuzzer-5633569034076160
>
> This is untested with valid cin files as none of the files i found cover the
>
On Sun, Apr 07, 2019 at 01:21:14AM +0200, Michael Niedermayer wrote:
> Fixes: Timeout (46sec -> 7ms)
> Fixes:
> 14030/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AGM_fuzzer-5721258760601600
>
> Found-by: continuous fuzzing process
>
> -Original Message-
> From: ffmpeg-devel [mailto:ffmpeg-devel-boun...@ffmpeg.org] On Behalf Of
> avih
> Sent: Wednesday, April 24, 2019 5:22 PM
> To: FFmpeg development discussions and patches
> Subject: Re: [FFmpeg-devel] [PATCH V4 1/2] configure: sort
> decoder/encoder/filter/...
On 24/04/2019 01:38, Carl Eugen Hoyos wrote:
2017-12-29 13:16 GMT+01:00, Cyber Sinh :
Sorry for the diff instead of regular git patch. Here is the patch.
Only saw this today after wondering
why nobody tried using wsl before...
I'm using WSL since a while, but never bothered trying to setup
mån 2019-04-22 klockan 21:33 +0200 skrev Michael Niedermayer:
> Fixes: Timeout (16 sec - 7 sec)
> Fixes:
> 14237/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ZMBV_fuzzer-5693453897302016
>
> Found-by: continuous fuzzing process
>
mån 2019-04-22 klockan 19:15 +0200 skrev Marton Balint:
> This affects the following samples:
>
> samples/ffmpeg-bugs/roundup/issue1775/av_seek_frame_failure.mxf
> samples/ffmpeg-bugs/trac/ticket1957/16ch.mxf
> samples/ffmpeg-bugs/trac/ticket5016/r0.mxf
>
> No. It expands at `echo` because you didn't quote `$inputs`.
I meant the at the line `echo $inputs`. The shell expands it before it
executes `echo`. `echo` (or any other command) will get the expanded value,
which with `*` is all files in the current working directory which don't
begin with a
> btw, I tried read as following, looks that read also expands '*':
> print_in_columns() {
> read inputs
> echo $inputs
> }
No. It expands at `echo` because you didn't quote `$inputs`.
As I commented on your v4 part 2 patch, you _must_ quote all strings unless
you know for a fact that
> -Original Message-
> From: Guo, Yejun
> Sent: Wednesday, April 24, 2019 3:29 PM
> To: Guo, Yejun
> Subject: Re: [FFmpeg-devel] [PATCH V4 1/2] configure: sort
> decoder/encoder/filter/... names in alphabet order
>
> > print_in_columns() {
> > - cols=$(expr $ncols / 24)
> > - cat
---
libavcodec/mediacodecdec_common.c | 20 ++--
1 file changed, 10 insertions(+), 10 deletions(-)
diff --git a/libavcodec/mediacodecdec_common.c
b/libavcodec/mediacodecdec_common.c
index f90e2f1a23..1656cd6664 100644
--- a/libavcodec/mediacodecdec_common.c
+++
Avoids returning EAGAIN after signaling EOF to the codec in
ff_mediacodec_dec_send() so we can try to receive a frame before
returning in mediacodec_receive_frame().
This helps avoiding an extra round-trip between avcodec_send_frame() and
avcodec_receive_frame() while draining the remaining
---
libavcodec/mediacodecdec_common.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/mediacodecdec_common.c
b/libavcodec/mediacodecdec_common.c
index f7a06cdc6d..f90e2f1a23 100644
--- a/libavcodec/mediacodecdec_common.c
+++ b/libavcodec/mediacodecdec_common.c
> log_file(){
> log BEGIN $1
> - pr -n -t $1 >> $logfile
> + i=1
Shell variables are global. This means that any part of the code which ends
up calling `log_file` directly or through some other function (and not isolated
in a subshell) will get its `i` variable clobbered by `log_file`.
‐‐‐ Original Message ‐‐‐
On Wednesday, 24 de April de 2019 0:58, Michael Niedermayer
wrote:
> On Tue, Apr 23, 2019 at 11:55:59PM +0200, Michael Niedermayer wrote:
>
> > On Tue, Apr 23, 2019 at 07:16:07AM +, Andreas Håkon wrote:
> >
> > > ‐‐‐ Original Message ‐‐‐
> > >
> > >
> print_in_columns() {
> - cols=$(expr $ncols / 24)
> - cat | tr ' ' '\n' | sort | pr -r "-$cols" -w $ncols -t
> + set -- $(cat | tr ' ' '\n' | sort)
> + col_width=24
> + cols=$(($ncols / $col_width))
> + rows=$(($(($# + $cols - 1)) / $cols))
> + cols_seq=$(seq $cols)
> +
48 matches
Mail list logo
