[x265] [PATCH] param: update calculated maxCLL, maxFALL values to contentLightLevelInfo
Deepthi Nandakumar
deepthi at multicorewareinc.com
Wed Aug 26 13:34:24 CEST 2015
Agreed that's the general philosophy. But this works better if the user
wants to regenerate headers, because the encoder->param has now been
reset. Else, the user is responsible for
a) copying stats->maxCLL to user Param
b) doing an encoder reconfig, and copying user param to encoder->param
c) generate headers
What we have here avoids step 1 and 2.
On Wed, Aug 26, 2015 at 4:53 PM, Steve Borho <steve at borho.org> wrote:
> On 08/26, kavitha at multicorewareinc.com wrote:
> > # HG changeset patch
> > # User Kavitha Sampath <kavitha at multicorewareinc.com>
> > # Date 1440572877 -19800
> > # Wed Aug 26 12:37:57 2015 +0530
> > # Node ID 73b94b3431f98bc94c2113d98f7788783453197e
> > # Parent a28a863393994d8fb1d58c721352d9b4ec8c46ee
> > param: update calculated maxCLL, maxFALL values to contentLightLevelInfo
>
> the encoder is generally supposed to treat the param structure as
> read-only (the encoder keeps a seperate copy from what the cli/user
> passed into the encoder creation function).
>
> > diff -r a28a86339399 -r 73b94b3431f9 source/encoder/encoder.cpp
> > --- a/source/encoder/encoder.cpp Mon Aug 24 14:04:32 2015 +0530
> > +++ b/source/encoder/encoder.cpp Wed Aug 26 12:37:57 2015 +0530
> > @@ -39,6 +39,10 @@
> >
> > #include "x265.h"
> >
> > +#if _MSC_VER
> > +#pragma warning(disable: 4996) // POSIX functions are just fine, thanks
> > +#endif
> > +
> > namespace X265_NS {
> > const char g_sliceTypeToChar[] = {'B', 'P', 'I'};
> > }
> > @@ -1046,6 +1050,14 @@
> >
> > stats->maxCLL = m_analyzeAll.m_maxCLL;
> > stats->maxFALL = (uint16_t)(m_analyzeAll.m_maxFALL /
> m_analyzeAll.m_numPics);
> > + if (m_param->contentLightLevelInfo)
> > + {
> > + free((char*)m_param->contentLightLevelInfo);
> > +
> > + char value[16];
> > + sprintf(value, "%hu,%hu", stats->maxCLL, stats->maxFALL);
> > + m_param->contentLightLevelInfo = strdup(value);
> > + }
> > }
> >
> > /* If new statistics are added to x265_stats, we must check here
> whether the
> > _______________________________________________
> > x265-devel mailing list
> > x265-devel at videolan.org
> > https://mailman.videolan.org/listinfo/x265-devel
>
> --
> Steve Borho
> _______________________________________________
> x265-devel mailing list
> x265-devel at videolan.org
> https://mailman.videolan.org/listinfo/x265-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20150826/a942f3be/attachment-0001.html>
More information about the x265-devel
mailing list