Re: FireWire DV/HDV input device using libiec61883

On date Saturday 2012-06-30 16:35:25 +0200, Georg Lippitsch encoded:
> Hi,
> 
> since it's quite a long time ago I posted my first version of the
> FireWire input patch, I start a new thread.
> Here is a new version of the patch, which uses the new Linux kernel
> FireWire stack and libiec61883 instead of the old deprecated one.
> 
> I fixed most of the issues discussed in the former thread, and also
> rewrote the capture loop using threads. This allows stable
> operation, even if iec61883_read_packet() is not called frequently
> enough. If incoming frames are not processed fast enough, they can
> be buffered up to a specified limit.
> 
> 
> Regards,
> 
> Georg

> From 0551eba2b5164e9c277a07d0b9a11f8ee47b4c3d Mon Sep 17 00:00:00 2001
> From: Georg Lippitsch <georg.lippitsch <at> gmx.at>
> Date: Mon, 23 Apr 2012 16:01:17 +0200
> Subject: [PATCH] FireWire DV/HDV input device using libiec61883
> 
> ---
>  configure                |    3 +
>  doc/indevs.texi          |   49 +++++
>  libavdevice/Makefile     |    1 +
>  libavdevice/alldevices.c |    1 +
>  libavdevice/iec61883.c   |  461 ++++++++++++++++++++++++++++++++++++++++++++++

[PATCH 0/2] lavf/file: small cleanup

Hi,

  while reading the commit diffs I found df531b0 to be a bit hard
and unusual to read. Other than checking compilation I did not test
my changes besides trying to make sure the logic is not changed.

  The patches can be sqashed into one patch if both get approved.
That would require to merge the commit messages. I will be happy
to do both if requested.

Alexander Strasser (2):
  lavf/file: cosmetic: Remove superfluous ternary operator
  lavf/file: cosmetic: Use "not" instead of "equals zero"

 libavformat/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--

-- 
1.7.10.2.552.gaa3bb87

[PATCH 1/2] lavf/file: cosmetic: Remove superfluous ternary operator

---
 libavformat/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libavformat/file.c b/libavformat/file.c
index 35c36a5..b49f84f 100644
--- a/libavformat/file.c
+++ b/libavformat/file.c
 <at>  <at>  -91,7 +91,7  <at>  <at>  static int file_open(URLContext *h, const char *filename, int flags)
         return AVERROR(errno);
     h->priv_data = (void *) (intptr_t) fd;

-    h->is_streamed = (0==fstat(fd, &st) && S_ISFIFO(st.st_mode)) ? 1 : 0;
+    h->is_streamed = 0==fstat(fd, &st) && S_ISFIFO(st.st_mode);

     return 0;
 }
--

-- 
1.7.10.2.552.gaa3bb87

Re: [PATCH]Fix --disable-optimizations --disable-neon (arm)

On Sat, Jun 30, 2012 at 02:00:08PM +0000, Carl Eugen Hoyos wrote:
> Carl Eugen Hoyos <cehoyos <at> ag.or.at> writes:
> 
> > It seems attached patch is necessary for --disable-optimizations 
> > --disable-neon, fixes ticket #1241.
> 
> Ping.

this patch does not fix compilation with the quoted paramaters above

In file included from libavutil/intmath.h:36:0,
                 from libavutil/common.h:102,
                 from libavutil/avutil.h:305,
                 from libavutil/mem.h:31,
                 from libavdevice/timefilter.c:25:
libavutil/arm/intmath.h: In function ‘av_clip_uintp2_arm’:
libavutil/arm/intmath.h:83:5: warning: asm operand 2 probably doesn’t match constraints
libavutil/arm/intmath.h:83:5: error: impossible constraint in ‘asm’
make: *** [libavdevice/timefilter.o] Error 1
make: *** Waiting for unfinished jobs....
In file included from libavutil/intmath.h:36:0,
                 from libavutil/common.h:102,
                 from libavutil/avutil.h:305,
                 from libavdevice/avdevice.h:44,
                 from libavdevice/alldevices.c:22:

--

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Good people do not need laws to tell them to act responsibly, while bad

Re: [PATCH] ffplay: only configure video filters after we got the first frame

On Sat, Jun 30, 2012 at 07:30:48PM +0200, Marton Balint wrote:
> 
> 
> On Sun, 24 Jun 2012, Marton Balint wrote:
> 
> >Otherwise the codec width, height and pixel format values may not be set.
> >
> >Signed-off-by: Marton Balint <cus <at> passwd.hu>
> >---
> >ffplay.c |   33 ++++++++++++++-------------------
> >1 files changed, 14 insertions(+), 19 deletions(-)
> 
> Michael, please merge from ffplay stable for this patch.

merged

thanks

[...]
--

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Into a blind darkness they enter who follow after the Ignorance,
they as if into a greater darkness enter who devote themselves
to the Knowledge alone. -- Isha Upanishad

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> ffmpeg.org

[PATCH 2/2] lavf/file: cosmetic: Use "not" instead of "equals zero"

  This is more similar to the "overall" FFmpeg coding style.
---

   Assuming the context member is_streamed is initialized
properly to zero 

    if (!fstat(fd, &st))
      h->is_streamed = S_ISFIFO(st.st_mode);

would be another possibility. I personally find use of !
in places like this a bit odd, but OTOH we use it every
where else like this (including !strcmp(a, b)).

 libavformat/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libavformat/file.c b/libavformat/file.c
index b49f84f..c54ec28 100644
--- a/libavformat/file.c
+++ b/libavformat/file.c
 <at>  <at>  -91,7 +91,7  <at>  <at>  static int file_open(URLContext *h, const char *filename, int flags)
         return AVERROR(errno);
     h->priv_data = (void *) (intptr_t) fd;

-    h->is_streamed = 0==fstat(fd, &st) && S_ISFIFO(st.st_mode);
+    h->is_streamed = !fstat(fd, &st) && S_ISFIFO(st.st_mode);

     return 0;
 }
--

-- 

[PATCH] wtvdec: *_to_iso8601: Behave less surprising

  Do not violate the (implicit) contract that is indicated by
the arguments of those functions. In other words if buf_size
is passed as zero do not try to write to the buffer's first
element.
---

   I did not check if it is actually possible at any current
invocation places for the size of the passed buffer to be
be zero. The places are also limited to that file as the
functions are declared to have file scope. I just thought it
litters up the codebase little by little to have surprising
code like that, so I thought it is better to clean it up now.

   I did test nothing more than compilation on my machine.

 libavformat/wtvdec.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/libavformat/wtvdec.c b/libavformat/wtvdec.c
index 338eff8..d356dc3 100644
--- a/libavformat/wtvdec.c
+++ b/libavformat/wtvdec.c
 <at>  <at>  -375,7 +375,7  <at>  <at>  static void filetime_to_iso8601(char *buf, int buf_size, int64_t value)
     struct tm *tm = gmtime(&t);
     if (tm)
         strftime(buf, buf_size, "%Y-%m-%d %H:%M:%S", gmtime(&t));
-    else
+    else if (buf_size >= 1)
         buf[0] = '\0';
 }

Re: [PATCH 1/2] lavf/file: cosmetic: Remove superfluous ternary operator

On Sun, Jul 01, 2012 at 02:42:27AM +0200, Alexander Strasser wrote:
> ---
>  libavformat/file.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

LGTM

[...]
--

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

In a rich man's house there is no place to spit but his face.
-- Diogenes of Sinope

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [PATCH 2/2] lavf/file: cosmetic: Use "not" instead of "equals zero"

On Sun, Jul 01, 2012 at 02:50:06AM +0200, Alexander Strasser wrote:
>   This is more similar to the "overall" FFmpeg coding style.
> ---
> 
>    Assuming the context member is_streamed is initialized
> properly to zero 
> 
>     if (!fstat(fd, &st))
>       h->is_streamed = S_ISFIFO(st.st_mode);
> 
> would be another possibility. I personally find use of !
> in places like this a bit odd, but OTOH we use it every
> where else like this (including !strcmp(a, b)).
> 
>  libavformat/file.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/libavformat/file.c b/libavformat/file.c
> index b49f84f..c54ec28 100644
> --- a/libavformat/file.c
> +++ b/libavformat/file.c
>  <at>  <at>  -91,7 +91,7  <at>  <at>  static int file_open(URLContext *h, const char *filename, int flags)
>          return AVERROR(errno);
>      h->priv_data = (void *) (intptr_t) fd;
>  
> -    h->is_streamed = 0==fstat(fd, &st) && S_ISFIFO(st.st_mode);
> +    h->is_streamed = !fstat(fd, &st) && S_ISFIFO(st.st_mode);

whatever people prefer ...

Re: [PATCH] wtvdec: *_to_iso8601: Behave less surprising

On Sun, Jul 01, 2012 at 02:55:57AM +0200, Alexander Strasser wrote:
>   Do not violate the (implicit) contract that is indicated by
> the arguments of those functions. In other words if buf_size
> is passed as zero do not try to write to the buffer's first
> element.
> ---
> 
>    I did not check if it is actually possible at any current
> invocation places for the size of the passed buffer to be
> be zero. The places are also limited to that file as the
> functions are declared to have file scope. I just thought it
> litters up the codebase little by little to have surprising
> code like that, so I thought it is better to clean it up now.
> 
>    I did test nothing more than compilation on my machine.
> 
>  libavformat/wtvdec.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

LGTM, iam not maintainer of wtv though

[...]
--

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

There will always be a question for which you do not know the correct awnser.

_______________________________________________
ffmpeg-devel mailing list