[MPlayer-dev-eng] [PATCH] Maybe fix gcc4 for problems on amd64 without breaking other arches

[İsmail Dönmez]

Guillaume POIRIER wrote:
> Hi,
> 
> On 7/7/05, İsmail Dönmez <ismail at kde.org.tr> wrote:
> 
>>Reimar Döffinger wrote:
>>
>>>Hi,
>>>On Thu, Jul 07, 2005 at 07:36:19AM +0300, Ismail Donmez wrote:
>>>
>>>
>>>>#endif
>>>>     long xInc_shr16 = (long) (xInc >> 16);
>>>>+     long xDstWidth = (long) dstWidth;
>>>>     int xInc_mask = xInc & 0xffff;
>>>>     asm volatile(
>>>>             "xor %%"REG_a", %%"REG_a"       \n\t" // i
>>>>@@ -2547,7 +2548,7 @@ FUNNY_UV_CODE
>>>>             "cmp %2, %%"REG_a"              \n\t"
>>>>             " jb 1b                         \n\t"
>>>>
>>>>-             :: "m" (src1), "m" (dst), "m" ((long)dstWidth), "m" (xInc_shr16), "m" (xInc_mask),
>>>>+             :: "m" (src1), "m" (dst), "m" (xDstWidth), "m" (xInc_shr16), "m" (xInc_mask),
>>>
>>>
>>>I tried something like that (making the dstWidth argument long), and it
>>>breaks with gcc 3.3. Did you test that?
>>
>>No but I don't really see any reason for it to break with gcc-3.3. Diego
>>  I think you got gcc 3.3 around, can you please test this patch?
> 
> 
> I did (on IA-32):
> SwScaler: using unscaled Planar YV12 -> BGRA special converter
> V:   0.1   2/  2 ??% ??% ??,?% 0 0
> 
> MPlayer interrupted by signal 11 in module: decode_video
> - MPlayer crashed by bad usage of CPU/FPU/RAM.
>   Recompile MPlayer with --enable-debug and make a 'gdb' backtrace and
>   disassembly. Details in DOCS/HTML/en/bugreports_what.html#bugreports_crash.
> - MPlayer crashed. This shouldn't happen.
> 
> Holly crap, is gcc-3.3 that stupid?

Seems so :-(. time for an ugly ifdef amd64 ? :/


Regards,
ismail