Dean 417c7d02a7 Merge pull request #4586 from ee6aba54/fix/agc-lookahead-write-data ...

fix(agc): count every CP data write of a dispatch group as a write of the run

2026-10-13 15:37:38 +13:00

..

AlphaToCoverage.cpp

test(agc): cover UNORM16 and SNORM16 exports in the alpha-to-coverage GPU test

2026-10-11 08:00:05 -03:00

ArrayLayerTarget.cpp

test(agc): verify display XOR array attachment transport

2026-10-12 22:38:59 +03:00

BdaSpanReads.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

BindlessImages.cpp

fix(tests): keep bindless image selection within one host subgroup

2026-10-08 19:04:45 +03:00

BitfieldMask.cpp

test(shader): skip the bitfield mask execution test below 32-lane subgroups

2026-10-10 02:14:00 +02:00

BranchPastEndpgm.cpp

fix(shader): keep decoding past an s_endpgm while a branch targets later code

2026-10-04 17:03:31 +02:00

BufferAtomicCsub.cpp

fix(shader): stop decoding gfx6/7-only VOP2 opcodes on RDNA

2026-10-09 12:30:40 -04:00

BufferAtomics.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

BufferAtomicsFloat.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

BufferAtomicsIncDec64.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

BufferAtomicsOrSwap64.cpp

test(agc): check device capabilities by name in the execution tests

2026-10-08 08:29:12 -04:00

BufferDescriptorRuntime.cpp

perf(recompiler): keep compute shaders specialized; only prebuilt vertex and pixel shaders take the registration form

2026-10-12 09:50:41 +02:00

BufferFormat.cpp

fix(shader): convert unorm and snorm buffer format loads with correct rounding

2026-10-04 18:49:49 +02:00

BufferIntegerAtomics.cpp

test(shader): execute the untested buffer byte/short/dwordx2/x3 loads and stores and the 32-bit integer buffer atomics against values measured on RDNA

2026-10-04 18:37:45 +02:00

BufferLoadStore.cpp

fix(shader): use guest wave lane for ADD_TID on narrow subgroups

2026-10-08 19:48:08 +03:00

BufferUnalignedBase.cpp

test(agc): skip unsupported int64 atomic subcase

2026-10-09 11:23:21 +02:00

Bvh64IntersectRay.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

CacheControl.cpp

feat(shader): decode cache invalidations and performance hints as no-ops

2026-10-01 22:45:25 +02:00

Color1555Target.cpp

test(agc): draw into a 1_5_5_5 target and read the 16 bits back

2026-10-12 07:59:29 +02:00

ComputeUserDataCalls.cpp

feat(shader): link compute shaders' user-data callees at dispatch

2026-10-11 09:59:27 +02:00

ConservativeRasterization.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

ConstantStore.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

CopiedWriterCopy.cpp

perf(agc): list a dispatch or draw as a copied writer only while it has a CPU write-back

2026-10-10 17:10:57 +02:00

CrossWorkgroupFlag.cpp

test(shader): register the cross-workgroup test block as GPU-accessible

2026-10-11 02:46:00 -05:00

D16Memory.cpp

fix(tests): correct buffer descriptors in memory and wave execution tests

2026-10-08 17:37:32 +03:00

DccKeysFirstWrite.cpp

test(agc): keep key bytes reused after a first write through the write-back

2026-10-11 12:51:36 +02:00

DepthBiasPerFace.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

DepthCompareCopy.cpp

test(agc): pin the depth copy path, its refresh on reuse, R16 and several mips and layers; drop APS5_NO_DEPTH_COPY

2026-10-11 04:52:36 +02:00

DepthFastClear.cpp

fix(agc): take a memory value fill only when the hardware stores the raw value

2026-10-13 12:40:31 +13:00

DepthFastClearHarness.hpp

perf(recompiler): read depth-compare state and export mappings at run time when modules are built at registration

2026-10-12 09:50:33 +02:00

DepthFastClearSubmit.cpp

fix(agc): an HTILE fill covers a depth surface only over its whole pipe-aligned metadata

2026-10-11 10:19:53 +02:00

DepthPlaneGuestLayers.cpp

perf(agc): skip the depth plane copy when no slice changed

2026-10-13 14:33:40 +13:00

DepthPlaneSampling.cpp

fix(agc): first-slice view of depth plane arrays, refuse plane copies of another host format

2026-10-13 12:18:50 +13:00

DepthStorageReuse.cpp

test(agc): distinguish GPU depth from reused storage bytes

2026-10-11 21:29:59 -03:00

DeviceSelection.cpp

feat(agc): choose the Vulkan device by name with ANYPS5_GPU

2026-10-06 14:55:12 +03:00

DivFixupF16.cpp

fix(shader): quiet signaling NaNs only in IEEE mode

2026-10-09 00:33:02 +02:00

Division.cpp

fix(shader): flush v_div_scale_f32 input denormals

2026-10-11 01:14:31 +02:00

DivisionOutputModifiers.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

DivisionResultModifiers.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

Dpp8.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

Dpp8PkFmac.cpp

fix(shader): accept DPP8 on v_pk_fmac_f16

2026-10-11 13:16:11 -04:00

Dpp16InactiveSource.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

Dpp16Operands.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

Dpp16RowShareXmask.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

DppNoForm.cpp

fix(shader): refuse DPP16 on opcodes with no DPP form

2026-10-05 03:33:06 +02:00

DrawCacheReuse.cpp

test(agc): require a repeated draw to reuse the cached draw

2026-10-10 21:24:11 +02:00

DrawOffsetIndices.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

DsAppendConsume.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

DsPermute.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

DsSwizzle.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

DsWrxchg2.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

DsWrxchg2B64.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

DualSourceBlend.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

DwordPatternFill.cpp

fix(agc): take the dword pattern-fill kernel only through a 32-bit output format

2026-10-12 17:09:31 +02:00

Dx10Clamp.cpp

feat(shader): follow DX10_CLAMP in the clamp modifier

2026-10-09 00:25:07 +02:00

EmulatedColorCompare.cpp

fix(shader): return 0 for 8-bit UNORM color comparison

2026-10-12 12:00:21 -07:00

ExecHostLanes.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

F16DenormalFlush.cpp

fix(shader): flush f16 denormals in unary, frexp and conversion ops per the float mode

2026-10-11 14:27:58 -04:00

F16MinMaxNan.cpp

fix(shader): flush f16 min3/max3/med3 denormal inputs

2026-10-10 14:21:31 -04:00

F16PackTowardZero.cpp

perf(shader): convert v_cvt_pkrtz_f16_f32 from the native nearest-even conversion

2026-10-09 14:11:02 -04:00

F32ArithmeticNan.cpp

feat(shader): choose f32 arithmetic NaN results on the bits

2026-10-09 09:14:44 +02:00

F32CompareConstant.cpp

perf(shader): fold the denormal flush into f32 compares against constants

2026-10-09 13:38:56 -04:00

F32ConstantFactorTininess.cpp

perf(shader): check f32 product tininess in one multiply to nearest even

2026-10-11 01:54:16 +02:00

F32DenormalFlush.cpp

test(shader): accept the split fma sum in F32DenormalFlush

2026-10-11 19:10:52 +02:00

F32Dot2Rounding.cpp

test(agc): skip the f32 dot2 rounding test on devices without shaderFloat64

2026-10-12 10:32:54 -04:00

F32MinMaxNan.cpp

feat(shader): follow IEEE_MODE in f32 min, max, min3, max3 and med3

2026-10-08 23:09:31 +02:00

F32OutputModifier.cpp

fix(shader): apply the output modifier only in the float modes where the hardware does

2026-10-08 23:58:08 +02:00

F32PackRounding.cpp

perf(shader): round v_cvt_pknorm and v_cvt_pk_u8 in f32

2026-10-09 14:05:07 -04:00

F32ResultClamp.cpp

fix(tests): correct buffer OOB modes in F32, F64 and division tests

2026-10-08 17:37:30 +03:00

F64Arithmetic.cpp

fix(shader): quiet f64 signaling NaNs only in IEEE mode

2026-10-10 09:40:09 +02:00

F64Conversions.cpp

fix(shader): quiet f64 signaling NaNs only in IEEE mode

2026-10-10 09:40:09 +02:00

F64Division.cpp

fix(shader): quiet f64 signaling NaNs only in IEEE mode

2026-10-10 09:40:09 +02:00

F64DivisionModifiers.cpp

fix(shader): quiet f64 signaling NaNs only in IEEE mode

2026-10-10 09:40:09 +02:00

F64FmaSubnormals.cpp

test(shader): run f64 denorm cases without preserve mode

2026-10-11 03:09:34 +00:00

F64NanConversions.cpp

fix(shader): quiet f64 signaling NaNs only in IEEE mode

2026-10-10 09:40:09 +02:00

F64Rounding.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

F64Transcendental.cpp

fix(shader): quiet f64 signaling NaNs only in IEEE mode

2026-10-10 09:40:09 +02:00

FillCoverKeys.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

FlatApertures.cpp

test(agc): split agc_driver_flat_apertures by shader instead of by row

2026-10-11 14:05:40 -03:00

FlatCacheBits.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

FlatD16Loads.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

FlatGlobalLoadsRegion.hpp

test(shader): split the flat and global loads test to keep it within CI's time limit

2026-10-05 17:32:46 +02:00

FlatLoads.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

FlatLoadsX2X3.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

FlatLoadsX4.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

FlatStore.cpp

test(agc): flat store stamped blocks skip when host imports are compared, not watched

2026-10-11 12:54:02 +05:30

Float16Dot.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

Float16FmaRounding.cpp

fix(shader): round the f16 fused multiply-adds once

2026-10-09 09:28:33 +00:00

Float16MinMaxBits.cpp

perf(shader): pick f16 min, max and med3 results on the bits

2026-10-11 04:59:08 -04:00

Float16Misc.cpp

fix(tests): correct buffer OOB modes in half, FMA and packed arithmetic tests

2026-10-08 17:37:31 +03:00

Float16Modifiers.cpp

fix(shader): apply the output modifier only in the float modes where the hardware does

2026-10-08 23:58:08 +02:00

Float16ResultModifiers.cpp

fix(tests): correct buffer OOB modes in half, FMA and packed arithmetic tests

2026-10-08 17:37:31 +03:00

Float16Rounding.cpp

fix(shader): flush v_ldexp_f16 denormals per the f16 denormal mode

2026-10-11 04:24:00 -04:00

Float16TernaryClamp.cpp

fix(tests): correct buffer OOB modes in half, FMA and packed arithmetic tests

2026-10-08 17:37:31 +03:00

FmaLegacy.cpp

fix(shader): round the v_mad_legacy_f32 product before the add

2026-10-10 08:18:55 +02:00

FmaRounding.cpp

test(shader): check v_fma_mix_f32 in FmaRounding

2026-10-09 05:20:42 +03:00

FormatD16.cpp

feat(shader): move buffer accesses to runtime descriptors via bda

2026-10-08 17:25:14 +03:00

FormatD16Store.cpp

feat(shader): move buffer accesses to runtime descriptors via bda

2026-10-08 17:25:14 +03:00

FormattedStore.cpp

feat(shader): move buffer accesses to runtime descriptors via bda

2026-10-08 17:25:14 +03:00

Fp16Overflow.cpp

test(agc): skip fp16 overflow without shaderFloat64

2026-10-11 04:44:29 +00:00

GatedDispatchCapture.cpp

test(agc): split the gated dispatch producer spin into short dispatches

2026-10-11 19:01:47 +02:00

GdsAppendConsume.cpp

fix(tests): correct buffer bounds modes in GDS LDS and scalar tests

2026-10-08 18:20:08 +03:00

GlobalAtomicCsub.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalAtomics64.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalAtomics.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalAtomicsFloat.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalAtomicsLanes.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalLoadAddtid.cpp

fix(shader): stop decoding gfx6/7-only VOP2 opcodes on RDNA

2026-10-09 12:30:40 -04:00

GlobalLoads.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalLoadsX2X3.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalLoadsX4.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalStoreAddtid.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GlobalVccBase.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

GuardedNullPointer.cpp

feat(agc): implement deferred faults for descriptors read through null SRT pointers

2026-10-09 00:21:00 +02:00

HalfExpansion.cpp

test(shader): cover half expansion bit patterns

2026-10-08 12:19:59 +02:00

HostImportAtMap.cpp

fix(agc): import GPU-mapped guest memory when it is mapped

2026-10-11 00:23:23 +02:00

HostImportTeardown.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

ImageAddressDimension.cpp

test(agc): avoid ambiguous row-boundary filtering

2026-10-11 11:34:47 +00:00

ImageAtomics64.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

ImageAtomics.cpp

feat(shader): select image and sampler resources at runtime

2026-10-08 17:26:34 +03:00

ImageAtomicsBitwise.cpp

fix(tests): correct buffer OOB modes in image and BVH tests

2026-10-08 17:37:33 +03:00

ImageAtomicsFloat.cpp

feat(shader): select image and sampler resources at runtime

2026-10-08 17:26:34 +03:00

ImageBy.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

ImageCacheBits.cpp

fix(tests): correct buffer OOB modes in image and BVH tests

2026-10-08 17:37:33 +03:00

ImageConvertedUnorm.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

ImageGatherOneDimensional.cpp

fix(tests): correct buffer OOB modes in image and BVH tests

2026-10-08 17:37:33 +03:00

ImageGatherOneDimensionalArray.cpp

fix(tests): correct raw buffer bounds modes in image tests

2026-10-08 18:21:50 +03:00

ImageLoadPacked.cpp

perf(shader): eliminate prepared-path overhead in data, descriptors, dispatch, bda

2026-10-08 17:29:17 +03:00

ImageOneDimensionalArray.cpp

fix(tests): correct raw buffer bounds modes in image tests

2026-10-08 18:21:50 +03:00

ImagePacked16Load.cpp

fix(agc): read X for a DST_SEL W of 5_6_5 textures as measured on gfx1035

2026-10-10 11:22:07 -04:00

ImagePackedHorizontalGather.cpp

test(shader): use raw buffer bounds in the packed horizontal gather test

2026-10-08 20:42:14 +02:00

ImagePckN.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

ImagePipeBankXor.cpp

feat(agc): detile XOR-swizzled textures whose base carries a pipe/bank XOR

2026-10-08 14:30:52 -05:00

ImagePlaneAddressOnLine.cpp

fix(shader): read 2D image accesses through 1D T#s as measured

2026-10-10 09:35:34 +02:00

ImagePlaneAddressOnLineImplicitLod.cpp

test(agc): enter whole quad mode in the implicit-LOD 1D sample test

2026-10-12 07:35:46 +13:00

ImageSampleAnisoOverride.cpp

fix(agc): decide ANISO_OVERRIDE single-level images from the T#'s own levels

2026-10-10 10:57:02 +02:00

ImageSampleBiasA16.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

ImageSampleBorderColor.cpp

fix(agc): pick float or integer sampler border colours from the paired images

2026-10-10 21:57:09 -04:00

ImageSampleDerivatives.cpp

fix(tests): correct buffer OOB modes in image and BVH tests

2026-10-08 17:37:33 +03:00

ImageSampleForceDegamma.cpp

fix(agc): keep FORCE_DEGAMMA paired when ANISO_OVERRIDE rebuilds a sampler

2026-10-11 13:50:25 -04:00

ImageSampleLodClamp.cpp

Merge pull request #1558 from krishnaraya4/test/agc-target-has-capability

2026-10-09 21:03:59 +13:00

ImageSampleReduction.cpp

fix(tests): correct buffer OOB modes in image and BVH tests

2026-10-08 17:37:33 +03:00

ImageSampleSharedSampler.cpp

fix(tests): correct raw buffer bounds modes in image tests

2026-10-08 18:21:50 +03:00

ImageSampleUnnormalized.cpp

fix(agc): read the base level of a mipmapped view through unnormalized coordinates only where the hardware does

2026-10-11 11:07:25 +02:00

ImageSint8.cpp

fix(tests): correct raw buffer bounds modes in image tests

2026-10-08 18:21:50 +03:00

ImageSrgbLoad.cpp

fix(shader): keep sRGB comparisons in the emulated runtime mode

2026-10-11 12:54:54 +00:00

ImageStoreLineThroughPlane.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

ImageStorePacked.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

ImageStoreSint.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

ImageTexelStatus.cpp

fix(tests): correct buffer OOB modes in image and BVH tests

2026-10-08 17:37:33 +03:00

ImageVolumeAddressOnPlane.cpp

fix(shader): bind 2D textures to 3D sample and load accesses as measured

2026-10-10 05:15:30 +02:00

ImageVolumeAddressPlanarTexture.cpp

fix(shader): read 2D-array T#s through 3D image accesses

2026-10-10 23:42:50 +00:00

ImageXorSwizzleVolume.cpp

fix(agc): lay out 3D SW_64KB_D_X textures as thick volumes

2026-10-08 18:28:19 -04:00

IndirectArgumentsWrittenAhead.cpp

fix(agc): stop the dispatch look-ahead before indirect arguments an earlier dispatch may write

2026-10-10 14:58:18 -05:00

IntegerDot.cpp

fix(tests): correct buffer OOB modes in half, FMA and packed arithmetic tests

2026-10-08 17:37:31 +03:00

IntegerToF32Rounding.cpp

fix(shader): honor directed round32 in integer to f32 conversions

2026-10-10 16:34:07 -04:00

InterpolationF16.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

InterpolationModes.cpp

feat(shader): follow the float mode in 16-bit interpolation

2026-10-08 23:36:04 +02:00

LdexpF16InlineConstants.cpp

fix(shader): read v_ldexp_f16 src1 float inline constants as f16

2026-10-09 05:53:32 +03:00

LdexpF32Denormal.cpp

fix(shader): keep f32 subnormals in v_ldexp_f32 when denormals are kept

2026-10-11 02:36:48 -04:00

LdsAddtid.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

LdsAtomics64.cpp

fix(shader): LDS float atomics quiet signalling NaNs and flush denormals as the hardware does

2026-10-10 09:40:41 +02:00

LdsAtomics.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

LdsBeyondDeviceLimit.cpp

feat(shader): implement compute LDS beyond the device's shared memory limit

2026-10-10 10:56:56 +02:00

LdsCondxchg.cpp

fix(tests): correct buffer bounds modes in GDS LDS and scalar tests

2026-10-08 18:20:08 +03:00

LdsFloatAtomicModes.cpp

fix(shader): flush LDS float atomic inputs for FP_DENORM_FLUSH_IN as well

2026-10-10 09:40:42 +02:00

LdsIntegerAtomics.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

LdsMixedWidthAtomics.cpp

test(agc-driver): add a split-lane case with per-lane operands to the LDS mixed-width atomic test

2026-10-11 01:53:37 +03:00

LdsReadWrite.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

LdsSrc2.cpp

fix(shader): refuse the f32 DS src2 operations until LDS float denormals are modelled

2026-10-08 23:13:23 +02:00

LogNearOne.cpp

fix(shader): correct v_log_f32 near 1.0 when the driver's log2 strays from a series

2026-10-11 01:33:44 +02:00

LookBackScan.cpp

test(agc): check device capabilities by name in the execution tests

2026-10-08 08:29:12 -04:00

LoopEndingExits.cpp

feat(recompiler): structurize loops whose exits reach different program ends

2026-10-06 00:35:47 +02:00

MacLegacy.cpp

fix(shader): flush v_mac_legacy_f32 denormals as measured on gfx1010

2026-10-10 10:54:23 +02:00

ManyBuffers.cpp

fix(tests): correct raw buffer bounds in DCC many-buffer and interlock tests

2026-10-08 18:24:15 +03:00

MaskedGlobalStore.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

MixPrecision.cpp

fix(shader): give an exact zero mix f16 sum -0 when rounding toward -infinity

2026-10-11 14:59:56 +02:00

MsaaEnableSingleSample.cpp

fix(agc): accept MSAA_ENABLE on a single-sample target sampled at the pixel center

2026-10-12 19:07:15 +13:00

NanQuieting.cpp

fix(shader): quiet f64 signaling NaNs only in IEEE mode

2026-10-10 09:40:09 +02:00

NggVertexWaveInfo.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

NggVertexWaveReduction.cpp

test(agc): run a wave64 NGG full-wave OR on a 32-lane device

2026-10-11 13:24:14 +02:00

OversizedDescriptorSets.cpp

fix(agc): allocate oversized descriptor sets from a dedicated pool

2026-10-07 22:37:10 +02:00

OversizedDispatch.cpp

test(agc): word the tessellation and oversized dispatch skips as device limitations, as check_skips expects

2026-10-12 15:06:12 +13:00

PackedAlu.cpp

fix(shader): clamp packed f16 min/max signaling NaN results to zero

2026-10-09 00:43:35 -04:00

PackedFloat16Native.cpp

test(agc): report the packed f16 native skip as a device limitation, as check_skips expects

2026-10-12 14:51:21 +13:00

PackedIntegerClamp.cpp

fix(tests): correct buffer OOB modes in half, FMA and packed arithmetic tests

2026-10-08 17:37:31 +03:00

PackedIntegerNegate.cpp

fix(shader): apply VOP3P neg_lo/neg_hi to packed 16-bit integer sources

2026-10-12 04:57:52 +02:00

PackedMipTailTarget.cpp

test(agc): render into packed mip tails through the guest-memory fallback

2026-10-12 03:17:32 -03:00

PackedUnormColorTarget.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PatternBufferFill.cpp

fix(agc): run a partial last thread group through the shader and validate every axis

2026-10-11 05:05:55 +02:00

PendingExactSurface.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

Permlane16.cpp

fix(tests): correct buffer descriptors in memory and wave execution tests

2026-10-08 17:37:32 +03:00

PipelineOptimizedReadback.cpp

test(agc): verify optimized pipeline draws and readback

2026-10-12 00:16:28 +02:00

PixelAncillaryWord.cpp

feat(shader): give reads the masked-select pass keeps the whole pixel ancillary word

2026-10-13 13:06:21 +13:00

PixelBarycentricEmulation.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelFixedPointPosition.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelHelperLanes.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelInterlock.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelLds.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelLdsLaneSlots.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelUint16Export.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelValidMask.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

PixelWaveReduction.cpp

fix(shader): match half-wave reductions over relatively addressed vector registers

2026-10-12 02:07:50 +02:00

PooledBufferRange.cpp

fix(agc): give pooled buffers a VkBuffer of exactly the requested size

2026-10-10 15:25:21 +01:00

PreparedExports.cpp

perf(recompiler): read depth-compare state and export mappings at run time when modules are built at registration

2026-10-12 09:50:33 +02:00

PreparedGraphics.cpp

perf(agc): prebuild each stage's library alone and keep two descriptor sets with pipeline libraries

2026-10-12 09:50:41 +02:00

PreparedPublication.cpp

perf(agc-driver): prepare shader ABI variants before taking the preparation lock

2026-10-11 14:54:48 +02:00

PreparedShaders.cpp

test(agc): cover unaligned shader ABI headers

2026-10-12 22:28:55 +00:00

PreparedUnregisteredGraphics.cpp

fix(agc): read unregistered graphics programs only while no shader is registered

2026-10-10 19:12:14 -05:00

ProvokingVertex.cpp

perf(recompiler): read depth-compare state and export mappings at run time when modules are built at registration

2026-10-12 09:50:33 +02:00

ReadFirstLane.cpp

fix(tests): correct buffer descriptors in memory and wave execution tests

2026-10-08 17:37:32 +03:00

ReadLane.cpp

fix(tests): correct buffer descriptors in memory and wave execution tests

2026-10-08 17:37:32 +03:00

RecordedKeptBytes.cpp

perf(recompiler): read depth-compare state and export mappings at run time when modules are built at registration

2026-10-12 09:50:33 +02:00

RegistrationPrepareShutdown.cpp

fix(agc): drain the registration preparation pool before the device is reset

2026-10-11 01:34:05 +02:00

RelativeAncillary.cpp

test(agc): a relative read of the whole ancillary word gets the composed word

2026-10-13 13:06:27 +13:00

RelativeIndexing.cpp

test(recompiler): pin that an unwritten window register reads 0

2026-10-11 18:36:09 +00:00

RelativeStageInputs.cpp

test(recompiler): keep vector window fixtures observable

2026-10-11 19:23:00 -04:00

RuntimeImages.cpp

perf(shader): eliminate prepared-path overhead in data, descriptors, dispatch, bda

2026-10-08 17:29:17 +03:00

SampledCacheBudget.cpp

test(agc): keep sampled cache fixture registrations separate

2026-10-12 00:12:50 +00:00

ScalarBitCount.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarGetreg.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarProgramEnd.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarProgramEndSaved.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarProgramFlow.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarRelativeIndexing.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarSethalt.cpp

fix(tests): correct buffer bounds modes in GDS LDS and scalar tests

2026-10-08 18:20:08 +03:00

ScalarSetpc.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarSop1.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarSop2.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarSopkCompare.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarSopkMisc.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarSopkWaitcnt.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarStaticCall.cpp

test(shader): report effective call targets

2026-10-08 20:23:47 +02:00

ScalarSwappc.cpp

fix(tests): correct buffer OOB modes in DPP, DS, LDS and scalar tests

2026-10-08 17:37:28 +03:00

ScalarSwappcFetch.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

ScalarWideLoads.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

SdwaCvtF16U16.cpp

fix(recompiler): accept byte source selects on v_cvt_f16_u16 SDWA

2026-10-10 16:17:56 +00:00

SdwaDestination.cpp

fix(recompiler): accept SDWA byte sources on v_cvt_f16_u16 and v_cvt_f16_i16

2026-10-10 13:11:10 -05:00

SdwaFloatSelectors.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

SdwaIntegerClamp.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

SdwaIntegerSelectors.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

SdwaLdexpF16.cpp

fix(shader): quiet signaling NaNs only in IEEE mode

2026-10-09 00:33:02 +02:00

SdwaMoveSelectors.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

ShaderClock.cpp

test(agc): compare llvmpipe shader clocks modulo 2^32

2026-10-09 10:30:28 +00:00

SinCosNearZero.cpp

fix(agc): accept 2 ULP at the cosine's maximum in agc_driver_sin_cos_near_zero

2026-10-10 09:39:02 +02:00

SinCosReduction.cpp

fix(shader): reduce v_sin_f32 and v_cos_f32 inputs exactly

2026-10-09 06:34:59 +03:00

Srgb8ColorTarget.cpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

StencilRefExport.cpp

fix(agc): export the stencil reference from pixel shaders with VK_EXT_shader_stencil_export

2026-10-12 19:44:41 +13:00

StorageCacheBudget.cpp

fix(agc): evict resource-cache images with storage entries

2026-10-11 19:08:29 +08:00

StorageOneSliceAlias.cpp

test(agc): pass the GPU flag to GuestAllocations Add in the one-slice alias test

2026-10-11 05:41:07 +02:00

StoreIdleResults.cpp

test(agc): keep queue zero alive after an interrupted storage idle wait

2026-10-12 18:46:24 +00:00

SubvectorLoop.cpp

feat(shader): implement s_subvector_loop_begin and s_subvector_loop_end

2026-10-11 22:03:26 +02:00

SwapAltBlend.cpp

test(agc): reject a split SWAP_ALT draw whose vertex shader stores to memory

2026-10-11 22:44:57 +02:00

TbufferFormatD16.cpp

feat(shader): move buffer accesses to runtime descriptors via bda

2026-10-08 17:25:14 +03:00

Tessellation.cpp

test(agc): word the tessellation and oversized dispatch skips as device limitations, as check_skips expects

2026-10-12 15:06:12 +13:00

TypedHeaps.cpp

perf(recompiler): keep compute shaders specialized; only prebuilt vertex and pixel shaders take the registration form

2026-10-12 09:50:41 +02:00

Uint8x4ColorTarget.cpp

feat(agc): support 64 KiB display XOR color targets

2026-10-12 22:38:59 +03:00

UnimportedLabelOrder.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

VCmpEqF16Clamp.cpp

fix(shader): accept the clamp bit on VOP3 float compares

2026-10-11 11:04:38 -04:00

VCmpF32Denormals.cpp

fix(shader): keep f32 denormals in compares when denormals are kept

2026-10-11 18:51:37 -04:00

VCvtI16F16OutputModifiers.cpp

test(shader): cover clamp alone and inf/NaN/fraction rows in v_cvt_i16_f16 output modifiers

2026-10-11 19:58:44 +02:00

VCvtPknormF16SourceModifiers.cpp

fix(shader): accept abs, neg and op_sel source modifiers on v_cvt_pknorm_i16_f16 and v_cvt_pknorm_u16_f16

2026-10-11 11:45:26 -04:00

VectorWindowDirectAccess.cpp

test(shader): make the read-only window tests pass in the full Linux build

2026-10-13 14:09:05 +13:00

VertexFetch.cpp

perf(agc): keep oversized shaders specialized and prebuild libraries on their own thread

2026-10-12 09:50:41 +02:00

ViewPastLastMip.cpp

fix(agc): address-based builds map only memory mapped for the GPU

2026-10-10 14:34:34 +05:30

Vop1Control.cpp

test(shader): cover the operand-less decoding of v_nop, v_pipeflush and v_clrexcp

2026-10-02 17:50:53 +02:00

Vop1FloatUnary.cpp

fix(shader): quiet signaling NaNs only in IEEE mode

2026-10-09 00:33:02 +02:00

Vop1Vop2Integer.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop2ModifierForms.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3AddSubClamp.cpp

test(shader): pin VOP3 integer add/sub clamp results measured on RDNA2

2026-10-08 16:55:06 -06:00

Vop3Aliases.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3AlignOpSel.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3CarryOut.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3CompareModifiers.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3CubeClamp.cpp

fix(shader): quiet signaling NaN cube results in IEEE mode

2026-10-10 09:13:00 +02:00

Vop3Integer16TernaryOpSel.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3Integer.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3IntegerAlu.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3IntegerClamp.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3IntegerClampWide.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3LegacySourceModifiers.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3Lerp.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3MadClamp.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3MadI64.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3PackLdexp.cpp

fix(shader): quiet signaling NaN halves in v_pack_b32_f16 in IEEE mode

2026-10-10 08:52:03 +02:00

Vop3QuadSad.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

Vop3Sad.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

VopcCompare.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

VopcCompareF64.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

VopcCompareTable.cpp

fix(tests): correct buffer OOB modes in VOP and SDWA tests

2026-10-08 17:37:29 +03:00

VulkanTestDevice.hpp

perf(agc): give each graphics stage a fixed push constant slot with pipeline libraries

2026-10-10 12:19:31 +02:00

Wave32Subgroup.cpp

fix(tests): correct buffer descriptors in memory and wave execution tests

2026-10-08 17:37:32 +03:00

WaveIdInWorkgroup.cpp

feat(shader): implement s_get_waveid_in_workgroup

2026-10-02 18:08:46 +02:00

WaveSleep.cpp

feat(shader): decode s_wakeup as a no-op

2026-10-02 18:10:07 +02:00

WaveUniform.cpp

fix(tests): correct buffer descriptors in memory and wave execution tests

2026-10-08 17:37:32 +03:00

WriteDataWrittenAhead.cpp

fix(agc): count every CP data write of a dispatch group as a write of the run

2026-10-12 18:09:14 +02:00

WriteLane.cpp

fix(tests): correct buffer descriptors in memory and wave execution tests

2026-10-08 17:37:32 +03:00

ZeroPositionVertex.cpp

test(agc): give the formatted buffer vertex case clip distances and the zero-position pixel shader its push slot

2026-10-12 09:59:23 +02:00

ZeroStrideVertexFetch.cpp

perf(recompiler): read vertex attribute component selectors at run time when modules are built at registration

2026-10-12 09:50:41 +02:00