[x265] rdcost: overflow check by integer

Steve Borho steve at borho.org
Mon May 26 04:48:17 CEST 2014


On Sat, May 24, 2014 at 10:48 PM, Satoshi Nakagawa <nakagawa424 at oki.com> wrote:
>
>> Aside: Most systems I've seen seem to do this in via -D<...> or whatever.
>
> It's simpler. Thanks.
>
> # HG changeset patch
> # User Satoshi Nakagawa <nakagawa424 at oki.com>
> # Date 1400989083 -32400
> #      Sun May 25 12:38:03 2014 +0900
> # Node ID 54a0dc2278494f7ec3b74f3f06a8521c805af79a
> # Parent  5e8cce428457f63fd9b8e18dafed2f8bed674d53
> add -D__STDC_LIMIT_MACROS=1
>
> diff -r 5e8cce428457 -r 54a0dc227849 source/CMakeLists.txt
> --- a/source/CMakeLists.txt     Fri May 23 09:11:15 2014 -0500
> +++ b/source/CMakeLists.txt     Sun May 25 12:38:03 2014 +0900
> @@ -103,6 +103,8 @@
>      endif()
>  endif(MSVC)
>
> +add_definitions(-D__STDC_LIMIT_MACROS=1)
> +
>  check_include_files(inttypes.h HAVE_INT_TYPES_H)
>  if(HAVE_INT_TYPES_H)
>      add_definitions(-DHAVE_INT_TYPES_H=1)

Is this still necessary given the way I modified compat/msvc/stdint.h?
 If so, for what compiler?

-- 
Steve Borho


More information about the x265-devel mailing list