summaryrefslogtreecommitdiff
Commit message (Expand)AuthorAgeFilesLines
* glsl: dfloor_to_arith WIPfp64_floorTapani Pälli2015-01-093-1/+115
* glsl: do DSQRT_TO_FSQRT only for doublesTapani Pälli2014-12-311-1/+1
* topin fiksi, fs_visitor::visit(ir_constant *ir)Tapani Pälli2014-12-311-1/+1
* i965/fs: Support for u2d, i2dTapani Pälli2014-11-202-4/+9
* i965/fs: Allow d2i and d2uTapani Pälli2014-11-202-11/+15
* dbg/fs: Dump instructions before double loweringTopi Pohjolainen2014-11-111-0/+4
* i965/eu: Use simd8 for treating hi/lo 32-bits of 16-width doublesTopi Pohjolainen2014-11-111-0/+2
* i965/eu: Allow 3-src float ops with doublesTopi Pohjolainen2014-11-111-6/+14
* i965/vec4: Make sure math isn't tried with doublesTopi Pohjolainen2014-11-111-0/+5
* i965/vec4: Treat double precision floats as any other basic typeTopi Pohjolainen2014-11-111-1/+1
* i965/fs: Double precision to single conversion supportTopi Pohjolainen2014-11-112-2/+2
* lower dfTopi Pohjolainen2014-11-111-0/+40
* i965: Add lowering pass splitting df-ops into twoTopi Pohjolainen2014-11-114-0/+194
* i965/fs: Generator support for packing d2f resultsTopi Pohjolainen2014-11-111-0/+27
* i965: Add new opcode for packing d2f resultsTopi Pohjolainen2014-11-112-0/+14
* i965/fs: Add support for ir_unop_unpack_double_2x32Topi Pohjolainen2014-11-112-1/+13
* i965/fs: Add unpack_double_2x32 virtual opcodeTopi Pohjolainen2014-11-114-0/+93
* i965/fs: Lower double precision scalars into vectorsTopi Pohjolainen2014-11-111-4/+39
* i965: Add helper telling if uniform is double and requires special loadTopi Pohjolainen2014-11-112-0/+38
* i965/fs/gen7: Add generator support for loading double precision uniformsTopi Pohjolainen2014-11-114-0/+39
* i965/gen7: Add support for loading double float scalars in 16-widthTopi Pohjolainen2014-11-112-0/+28
* i965/gen7: Add support for double precision constant operandsTopi Pohjolainen2014-11-111-0/+5
* i965/gen8: Add support for double precision constant operandsTopi Pohjolainen2014-11-115-0/+26
* i965/fs: Add support for ir_unop_pack_double_2x32Topi Pohjolainen2014-11-113-0/+25
* i965/fs: Add pack_double_2x32 virtual opcodeTopi Pohjolainen2014-11-114-0/+85
* i965: Add helper telling if a register is scalarTopi Pohjolainen2014-11-111-0/+8
* i965/fs: Add support for double precision uniform loadingTopi Pohjolainen2014-11-111-7/+9
* i965/fs: Prepare live interval analysis for double precisionTopi Pohjolainen2014-11-111-1/+6
* i965/fs: Take double float into account in register offsetsTopi Pohjolainen2014-11-112-3/+9
* i965/fs: Prepare virtual registers for double precision floatsTopi Pohjolainen2014-11-111-1/+4
* i965: Tell backend register about double precision typeTopi Pohjolainen2014-11-111-1/+2
* i965: Determine size of double precision float registerTopi Pohjolainen2014-11-111-0/+2
* i965: Lower DFSQRTTopi Pohjolainen2014-11-111-1/+2
* glsl: Add lowering pass approximating sqrt(double) with sqrt(float)Topi Pohjolainen2014-11-112-0/+213
* i965: Lower DFRACEXP/DLDEXPTopi Pohjolainen2014-11-111-1/+2
* glsl: add a lowering pass for frexp/ldexp with double argumentsIlia Mirkin2014-11-112-2/+280
* glsl: handle d2f/f2d in mat to vec visitorIlia Mirkin2014-11-111-0/+2
* st_glsl_to_tgsi: handle fp64 constants betterDave Airlie2014-11-111-32/+51
* tgsi: add support for flt64 constantsDave Airlie2014-11-118-6/+85
* st/glsl: add st fp64 support (v2)Dave Airlie2014-11-112-139/+539
* tgsi: expose doubles for softpipe.Dave Airlie2014-11-111-1/+1
* gallium: add double opcodes and TGSI execution (v2)Dave Airlie2014-11-113-14/+779
* i965: just avoid warnings with fp64Dave Airlie2014-11-114-0/+38
* glsl: lower double optional passesDave Airlie2014-11-112-0/+211
* glsl: implement double builtin functionsDave Airlie2014-11-111-259/+492
* glsl/lower_instructions: add double lowering passesDave Airlie2014-11-111-0/+83
* glsl: enable/disable certain lowering passes for doublesDave Airlie2014-11-111-3/+3
* SQUASH: support double when constructing matrix from scalarTapani Pälli2014-11-112-9/+21
* SQUASH: set correct size for double array uniformsTapani Pälli2014-11-112-2/+4
* SQUASH: support double in 420pack vector constructorTapani Pälli2014-11-111-2/+2
View Lorry Controller Status