summaryrefslogtreecommitdiff
AgeCommit message (Expand)AuthorFilesLines
2018-01-10i965/fs: Lower gles mediump floats into 16-bits16_bit_glesTopi Pohjolainen1-0/+3
2018-01-10glsl: HACK: Lower all temporary float variables to 16-bitsTopi Pohjolainen1-0/+11
2018-01-10glsl: HACK: Lower builtin float outputs to 16-bits by defaultTopi Pohjolainen1-0/+11
2018-01-10glsl: HACK: Treat input varyings as 16-bits by conversionTopi Pohjolainen1-1/+25
2018-01-10glsl: HACK: Force texture return into 16-bitsTopi Pohjolainen1-0/+19
2018-01-10glsl: Lower float conversions to mediumpTopi Pohjolainen1-0/+26
2018-01-10glsl: Use 16-bit constants if operation is otherwise 16-bitTopi Pohjolainen1-1/+42
2018-01-10glsl: WIP: Add lowering pass for treating mediump as float16Topi Pohjolainen4-0/+275
2018-01-10intel/compiler/fs: WIP: Use 32-bit slots for 16-bit uniformsTopi Pohjolainen1-1/+5
2018-01-10i965: WIP: Support for uploading 16-bit uniforms from 32-bit storeTopi Pohjolainen19-1/+101
2018-01-10intel/compiler/eu: Take stride into account in 16-bit opsTopi Pohjolainen1-0/+3
2018-01-10intel/compiler/fs: Prepare 16-bit and/or/xor for 32-bit srcTopi Pohjolainen1-0/+21
2018-01-10intel/compiler/fs: Consider logic ops on 16-bit booleansTopi Pohjolainen1-1/+69
2018-01-10intel/compiler/fs: Use original reg size when retyping nir srcTopi Pohjolainen1-2/+3
2018-01-10intel/compiler/fs: Consider original sizes when retyping alu opsTopi Pohjolainen1-6/+24
2018-01-10intel/compiler/fs: Prepare nir_emit_if() for 16-bit sourcesTopi Pohjolainen1-3/+8
2018-01-10intel/compiler/fs: Pad 16-bit payload loweringTopi Pohjolainen2-1/+17
2018-01-10intel/compiler/fs: Pad 16-bit const loads into full regsTopi Pohjolainen1-1/+2
2018-01-10intel/compiler/fs: Pad 16-bit nir intrinsic dest into full regTopi Pohjolainen1-1/+1
2018-01-10intel/compiler/fs: Pad 16-bit nir vec* components into full regTopi Pohjolainen2-1/+19
2018-01-10intel/compiler/fs: Pad 16-bit output (store/fb write) payloadsTopi Pohjolainen1-1/+1
2018-01-10intel/compiler/fs: Pad 16-bit texture return payloadsTopi Pohjolainen1-2/+19
2018-01-10intel/compiler/fs: Add register padding supportTopi Pohjolainen6-11/+33
2018-01-10intel/compiler/fs: Use component_size() instead of open codedTopi Pohjolainen2-3/+3
2018-01-10intel/compiler/fs: Set tex type for generator to flag fp16Topi Pohjolainen1-1/+9
2018-01-10intel/compiler/fs: Set 16-bit sampler return formatTopi Pohjolainen1-0/+3
2018-01-10intel/compiler/fs: Support for combining 16-bit immediatesTopi Pohjolainen1-13/+71
2018-01-10intel/compiler: Add support for negating 16-bit floatsTopi Pohjolainen1-1/+2
2018-01-10intel/compiler: Prepare for 16-bit 3-src opsTopi Pohjolainen3-0/+27
2018-01-10intel/compiler/fs: Use 16-bit null dest with 16-bit compareTopi Pohjolainen1-3/+13
2018-01-10intel/compiler/fs: Use 16-bit null dest with 16-bit mathTopi Pohjolainen1-1/+6
2018-01-10intel/compiler/fs: Use two SIMD8 instructions for 16-bit mathTopi Pohjolainen1-0/+18
2018-01-10intel/compiler/fs: Add helpers for 16-bit null regsTopi Pohjolainen1-0/+12
2018-01-10intel/compiler: Allow 16-bit mathTopi Pohjolainen1-2/+4
2018-01-10intel/compiler: Prepare for glsl mediump float uniformsTopi Pohjolainen2-1/+22
2018-01-10intel/compiler: Move type_size_scalar() into brw_shader.cppTopi Pohjolainen2-48/+48
2018-01-10intel/compiler: Add support for loading 16-bit constantsTopi Pohjolainen1-0/+5
2018-01-10intel/compiler/fs: Support for dumping 16-bit IMM valuesTopi Pohjolainen1-0/+5
2018-01-10intel/compiler/disasm: Print fp16 also for sampler messagesTopi Pohjolainen1-0/+5
2018-01-10intel/compiler/disasm: Print 16-bit IMM valuesTopi Pohjolainen1-1/+2
2018-01-10intel/compiler: Lower 16-bit lrp()Topi Pohjolainen1-0/+1
2018-01-10nir: Lower also 16-bit lrp() if neededTopi Pohjolainen2-0/+2
2018-01-10glsl: Enable 16-bit texturing in nir-conversionTopi Pohjolainen1-0/+3
2018-01-10glsl: Allow 16-bit lrp()Topi Pohjolainen1-0/+2
2018-01-10glsl: Allow 16-bit mathTopi Pohjolainen1-2/+5
2018-01-10glsl: Allow 16-bit neg() and dot()Topi Pohjolainen1-3/+6
2018-01-10glsl: Add more conversion ops to/from 16-bit floatsTopi Pohjolainen5-2/+56
2018-01-10glsl: Add conversion ops to/from 16-bit floatsTopi Pohjolainen6-0/+28
2018-01-10glsl: Add support for 16-bit float constants in nir-conversionTopi Pohjolainen1-0/+9
2018-01-10nir: Print 16-bit constantsTopi Pohjolainen1-0/+5
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-15 13:34:47 +0000