[FFmpeg-devel] [PATCH] Avoid-compiler-warning-unary-minus-operator-applied

[Nicolas George]

Le septidi 27 germinal, an CCXX, Robert Nagy a écrit :
> Now that you mention that error codes are large unsigned int. The entire

You misunderstood me: -MKTAG makes a large unsigned int, as in larger than
INT_MAX, but obviously not larger than UINT_MAX.

> libavfitler library seems to assume that error codes fit into "int", which
> isn't a problem for 64 bit builds, but 32 bit builds can be undefined?

No, 64- and 32-bits builds have the same size for ints.

The problem is for architectures that do not work in two's complement for
negative integers. Since AFAIK these architectures are all but dead, it does
not matter much.

The real problem is for sign extension. See the patch I just sent.

Regards,

-- 
  Nicolas George
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20120415/2ce0c168/attachment.asc>