Displaying 5 results from an estimated 5 matches for "opus_val64".
Did you mean:
opus_val16
2015 Nov 13
2
[Aarch64 00/11] Patches to enable Aarch64
...__aarch64__ to OPUS_FAST_INT64, and replacing
MULT16_32_Q15 with something like:
#define MULT16_32_Q15(a,b)
((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),15))
and MULT16_32_Q16 with:
#define MULT16_32_Q16(a,b)
((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),16))
There isn't an "opus_val64" or I would have used it (it would be defined
as "float" for floating point).
I think the problem here is that "fixed_generic.h" needs to be updated
for 64-bit machines the way "macros.h" was. x86 64-bit machines will
perform just as poorly using the current m...
2015 Nov 13
2
[Aarch64 00/11] Patches to enable Aarch64
...and replacing MULT16_32_Q15 with something like:
>> #define MULT16_32_Q15(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),15))
>> and MULT16_32_Q16 with:
>> #define MULT16_32_Q16(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),16))
>> There isn't an "opus_val64" or I would have used it (it would be defined as "float" for floating point).
>>
>> I think the problem here is that "fixed_generic.h" needs to be updated for 64-bit machines the way "macros.h" was. x86 64-bit machines will perform just as poorly using...
2015 Nov 12
2
[Aarch64 00/11] Patches to enable Aarch64
One other minor thing: I notice that in the inline assembly the result
(rd) is constrained as an earlyclobber operand. What was the reason for
that?
2015 Nov 13
0
[Aarch64 00/11] Patches to enable Aarch64
...PUS_FAST_INT64, and replacing MULT16_32_Q15 with something like:
> #define MULT16_32_Q15(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),15))
> and MULT16_32_Q16 with:
> #define MULT16_32_Q16(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),16))
> There isn't an "opus_val64" or I would have used it (it would be defined as "float" for floating point).
>
> I think the problem here is that "fixed_generic.h" needs to be updated for 64-bit machines the way "macros.h" was. x86 64-bit machines will perform just as poorly using the cu...
2015 Nov 16
0
[Aarch64 00/11] Patches to enable Aarch64
...for __aarch64__ to OPUS_FAST_INT64, and replacing MULT16_32_Q15 with something like:
#define MULT16_32_Q15(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),15))
and MULT16_32_Q16 with:
#define MULT16_32_Q16(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),16))
There isn't an "opus_val64" or I would have used it (it would be defined as "float" for floating point).
I think the problem here is that "fixed_generic.h" needs to be updated for 64-bit machines the way "macros.h" was. x86 64-bit machines will perform just as poorly using the current macr...