Jason Ekstrand da1c49171d spirv: Use a simpler and more correct implementaiton of tanh() ...

The new implementation is more correct because it clamps the incoming value
to 10 to avoid floating-point overflow.  It also uses a much reduced
version of the formula which only requires 1 exp() rather than 2.  This
fixes all of the dEQP-VK.glsl.builtin.precision.tanh.* tests.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "13.0" <mesa-dev@lists.freedesktop.org>

2016-12-09 18:38:21 -08:00

..

GLSL.std.450.h

spirv: Move to compiler/

2016-04-14 10:28:47 -07:00

nir_spirv.h

spirv: Move to compiler/

2016-04-14 10:28:47 -07:00

spirv2nir.c

nir/spirv: improve mmap() error handling

2016-10-01 15:27:46 -07:00

spirv_info.c

nir/spirv: Add string lookup tables for a couple of SPIR-V enums

2016-06-03 19:29:28 -07:00

spirv_info.h

nir/spirv: Add string lookup tables for a couple of SPIR-V enums

2016-06-03 19:29:28 -07:00

spirv_to_nir.c

nir: Get rid of nir_constant_data

2016-12-02 10:53:32 -08:00

spirv.h

spirv: Move to compiler/

2016-04-14 10:28:47 -07:00

vtn_alu.c

nir/spirv: implement ordered / unordered floating point comparisons properly

2016-11-23 08:07:44 +01:00

vtn_cfg.c

nir/spirv/cfg: Use a nop intrinsic for tagging the ends of blocks

2016-10-03 16:17:12 -07:00

vtn_glsl450.c

spirv: Use a simpler and more correct implementaiton of tanh()

2016-12-09 18:38:21 -08:00

vtn_private.h

spirv: Handle the InputAttachmentIndex decoration

2016-11-22 13:02:35 -08:00

vtn_variables.c

nir: Remove some unused fields from nir_variable

2016-12-05 15:40:10 -08:00