aboutsummaryrefslogtreecommitdiffstats
path: root/src/amd
Commit message (Expand)AuthorAgeFilesLines
* ac: add ac_choose_spi_color_formats() to common codeSamuel Pitoiset2020-06-153-112/+141
* radv: update internal referenceErik Faye-Lund2020-06-131-1/+1
* aco: fix shared subdword loadsDaniel Schürmann2020-06-121-1/+1
* radv/llvm: implement radv_enable_mrt_output_nan_fixup workaroundSamuel Pitoiset2020-06-121-0/+24
* aco: implement radv_enable_mrt_output_nan_fixup workaroundSamuel Pitoiset2020-06-121-0/+20
* radv: add new drirc option radv_enable_mrt_output_nan_fixupSamuel Pitoiset2020-06-124-0/+12
* radv: set DB_SHADER_CONTROL.CONSERVATIVE_Z_EXPORT correctlySamuel Pitoiset2020-06-123-0/+9
* android: aco: add aco_ir.cpp to Makefile.sourcesMauro Rossi2020-06-111-0/+1
* ac/surface: don't free dcc_retile_map on failureMarek Olšák2020-06-111-9/+4
* ac/surface: enable DCC for the first level in the mip tail on gfx10Marek Olšák2020-06-111-1/+10
* ac/surface: require that gfx8 doesn't have DCC in order to be displayableMarek Olšák2020-06-111-3/+4
* ac/surface: don't set is_displayable if displayable DCC is missingMarek Olšák2020-06-111-3/+6
* amd/addrlib: fix the C++ one definition rule violationMarek Olšák2020-06-112-2/+2
* ac/surface: cache DCC retile maps (v2)Marek Olšák2020-06-102-47/+199
* ac/surface: add a wrapper structure to hold ADDR_HANDLEMarek Olšák2020-06-105-33/+50
* amd/addrlib: remove unused members of ADDR2_COMPUTE_DCC_ADDRFROMCOORD_INPUTMarek Olšák2020-06-102-6/+0
* amd/addrlib: don't recompute DCC info for every ComputeDccAddrFromCoord callMarek Olšák2020-06-104-115/+91
* ac/surface: don't recompute the DCC retile map for imported texturesMarek Olšák2020-06-101-61/+61
* aco: fix moving sub-dword values out of a register for a fixed definitionRhys Perry2020-06-101-30/+31
* aco: use Info::definition_size instead of definition's regclassRhys Perry2020-06-102-4/+5
* aco: add Info::{operand_size,definition_size}Rhys Perry2020-06-103-0/+44
* aco: prefer 4-byte aligned definitionsRhys Perry2020-06-101-30/+30
* aco: allow reading/writing upper halves/bytes when possibleRhys Perry2020-06-106-69/+424
* aco: p_extract_vector in 64-bit u2f16/i2f16Rhys Perry2020-06-101-0/+4
* aco: validate instructions reading/writing upper halves/bytesRhys Perry2020-06-104-12/+229
* ac/surface: fix epitch when modifying surf_pitchPierre-Eric Pelloux-Prayer2020-06-101-2/+6
* ac/surface: set SCANOUT if surf->is_displayablePierre-Eric Pelloux-Prayer2020-06-101-0/+3
* ac/nir: fix integer comparisons with pointersSamuel Pitoiset2020-06-101-0/+11
* radv/aco: enable shaderInt8 and VK_KHR_shader_float16_int8 on GFX6-GFX7Samuel Pitoiset2020-06-093-5/+3
* radv/aco: enable shaderInt16 on GFX6-GFX7Samuel Pitoiset2020-06-092-5/+5
* radv/aco: enable 8-bit/16-bit storage on GFX6-GFX7Samuel Pitoiset2020-06-093-10/+10
* aco: remove unnecessary split- and create_vector instructions for subdword loadsDaniel Schürmann2020-06-091-79/+66
* aco: fix alignment of vectors with 4 elementsSamuel Pitoiset2020-06-091-2/+5
* aco: implement 8-bit/16-bit conversions on GFX6-GFX7Samuel Pitoiset2020-06-091-13/+17
* aco: optimize packing of 16bit subdword registers on GFX6/7Daniel Schürmann2020-06-091-0/+51
* aco: skip partial copies on first iteration when lowering to hwDaniel Schürmann2020-06-091-27/+15
* aco: coalesce copies more aggressively when lowering to hwDaniel Schürmann2020-06-091-0/+20
* aco: add and use scratch SGPR to lower subdword p_create_vector on GFX6/7Daniel Schürmann2020-06-092-8/+25
* aco: adjust GFX6 subdword lowering workarounds for 8bitDaniel Schürmann2020-06-091-17/+21
* aco: Workarounds subdword lowering on GFX6/7Daniel Schürmann2020-06-091-0/+43
* aco: use full-register instructions to implement subdword packing on GFX6/7Daniel Schürmann2020-06-091-1/+25
* aco: simplify statistics collection for copiesDaniel Schürmann2020-06-091-9/+5
* aco: fix register assignment for p_create_vector on GFX6/7Daniel Schürmann2020-06-091-24/+32
* radeonsi: add support for Sienna CichlidMarek Olšák2020-06-095-1/+16
* ac: align num_vgprs for gfx10.3Marek Olšák2020-06-095-4/+22
* radeonsi: don't set any XNACK options on gfx10.3Marek Olšák2020-06-091-2/+2
* radeonsi: implement R9G9B9E5 render target and image store support on gfx10.3Marek Olšák2020-06-091-2/+29
* radeonsi: enable larger SDMA clears and copies on gfx10.3Marek Olšák2020-06-091-3/+4
* ac/surface: add displayable DCC code for gfx10.3Marek Olšák2020-06-091-7/+20
* ac,radeonsi: start adding support for gfx10.3Marek Olšák2020-06-094-9/+42
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 07:46:14 +0000