search

wuying225 / softart

Automatically exported from code.google.com/p/softart
Other
0 stars 0 forks source link

Add host implemented intrinsic function to SASL #121

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
Description:
 1. Support host-optimized function load.
 2. C implementation and SIMD implementation.

Original issue reported on code.google.com by wuye9036 on 18 Nov 2011 at 4:13

GoogleCodeExporter commented 9 years ago 
Performance warning:
  Function call is about 40% slower than intrinsic.
  Return by first parameter is about 25% slower than return by value directly.

Original comment by wuye9036 on 18 Nov 2011 at 6:43

GoogleCodeExporter commented 9 years ago 
This issue was updated by revision 63bc38ffe993.

EFLIB:
  Add instruction detector to eflib.
  Now cpuinfo could detect SSE instruction set family.

Original comment by wuye9036 on 19 Nov 2011 at 4:47

GoogleCodeExporter commented 9 years ago 
This issue was updated by revision 341934ef7997.

EFLIB:
  Rename instruction_sets to cpu_features
  Add AVX detection on x86/x64
  Refine display of cpu features.

Original comment by wuye9036 on 19 Nov 2011 at 4:47

GoogleCodeExporter commented 9 years ago 
This issue was updated by revision 31ee642a417b.

SASL:
  Add an empty test for check ABI support.

Original comment by wuye9036 on 19 Nov 2011 at 4:47

GoogleCodeExporter commented 9 years ago 
This issue was updated by revision 5a73e60eab12.

SASL:
  Add inline_hint to function_t.
  Add sqrt( f32/f32v ) support.
  Add sqrt test.
  Add test into project.

Original comment by wuye9036 on 19 Nov 2011 at 4:02

GoogleCodeExporter commented 9 years ago

Original comment by wuye9036 on 21 Nov 2011 at 1:40

GoogleCodeExporter commented 9 years ago

Original comment by wuye9036 on 15 Dec 2011 at 3:25

GoogleCodeExporter commented 9 years ago

Original comment by wuye9036 on 7 Mar 2012 at 3:08

GoogleCodeExporter commented 9 years ago 
Intrinsic functions in HLSL4:
abs acos all any asfloat asin asint asuint atan atan2 ceil clamp clip cos cosh 
cross
ddx ddy degrees determinant distance dot exp exp2 faceforward floor fmod frac 
frexp 
fwidth GetRenderTargetSampleCount GetRenderTargetSamplePosition isfinite isinf 
isnan 
ldexp length lerp lit log log10 log2 max min modf mul noise normalize pow 
radians reflect refract round rsqrt saturate sign sin sincos sinh smoothstep 
sqrt step tan 
tanh transpose trunc

Original comment by minmin.gong on 20 Mar 2012 at 5:48

GoogleCodeExporter commented 9 years ago 
Lots of intrinsic was supported by SISD mode. others we will support it later, 
and list them in other issues.

Original comment by wuye9036 on 13 Sep 2012 at 8:22