summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2018-01-19glsl: add support for min/max 64glsl_arb_gpu_shader_fp64Dave Airlie7-52/+396
removes some dmin/dmax opcodes
2018-01-18XXX: Remove ir_binop_lequal and ir_binop_greaterElie Tournier6-8275/+8060
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float rsq()Elie Tournier3-1/+27
2018-01-18XXX: rcp using fp32 divElie Tournier2-1601/+350
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18XXX: do not check MOD_TO_FLOORElie Tournier1-3/+2
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18XXX: use sqrt32Elie Tournier2-4789/+1541
2018-01-18remove any_nequalElie Tournier1-8/+8
2018-01-18st_glsl_to_tgsi: track range for writes in a if/else/endif blocks. (v2)Dave Airlie1-14/+109
This overhauls the copy prop and dead code passes to avoid major CPU overhead in some corner cases trigged by the fp64 patches v2: fix level not being updated correctly
2018-01-18glsl: Add a lowering pass for 64-bit float div()Elie Tournier1-0/+1
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float frac()Elie Tournier1-0/+25
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float ceil()Elie Tournier1-2/+29
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float floor()Elie Tournier3-3/+34
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float max()Elie Tournier1-0/+21
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float min()Elie Tournier3-0/+30
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float nequal()Elie Tournier1-1/+3
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float greater()Elie Tournier1-1/+3
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float gequal()Elie Tournier1-1/+3
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl/lower_64bit: lower d2b using comparisonDave Airlie1-1/+21
This just does a compare to 0 and inverts the result to lower d2b. Not 100% sure this is always correct, but it passes piglit
2018-01-18glsl/lower_64bit: handle any/all operationsDave Airlie1-1/+60
This just splits them out and combines the results. Signed-off-by: Dave Airlie <airlied@redhat.com>
2018-01-18glsl: Add a lowering pass for 64-bit float rcp()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float round()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float trunc()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float sqrt()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float f2d()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float d2f()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float i2d()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float d2i()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float u2d()Elie Tournier3-10/+43
Handle non 64bit sources (airlied) Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float d2u()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float mul()Elie Tournier2-2/+6
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float add()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float less()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float lequal()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float equal()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float sign()Elie Tournier2-2/+6
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float neg()Elie Tournier3-1/+10
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add a lowering pass for 64-bit float abs()Elie Tournier5-16/+91
Squashed with: glsl/lower_64bit: fix return type conversion (airlied) Only do conversion for the 64-bit types, add a path to do result merging without conversion. Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do rcp(fp64)Elie Tournier5-0/+1847
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do round(fp64)Elie Tournier5-0/+274
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do trunc(fp64)Elie Tournier5-0/+152
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do sqrt(fp64)Elie Tournier5-0/+4438
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do fp32_to_fp64(fp32)Elie Tournier5-0/+357
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do fp64_to_fp32(fp64)Elie Tournier5-0/+610
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do int_to_fp64(int)Elie Tournier5-0/+365
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do fp64_to_int(fp64)Elie Tournier5-0/+337
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do uint_to_fp64(uint)Elie Tournier5-0/+293
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do fp64_to_uint(fp64)Elie Tournier5-0/+522
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do mul(fp64, fp64)Elie Tournier5-0/+2630
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do add(fp64, fp64)Elie Tournier5-0/+9276
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
2018-01-18glsl: Add "built-in" functions to do lt(fp64, fp64)Elie Tournier5-0/+206
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-24 19:21:06 +0000