v1.1.0

Metal v1.1.0

Diff since v1.0.0

Merged pull requests:

• Add resize! (#279) (@mtfishman)
• Initial MTLTexture support (#280) (@christiangnrd)
• Avoid redundant pointer conversions for threadgroup memory. (#283) (@maleadt)
• Re-implement metallib generation in Julia. (#284) (@maleadt)
• CompatHelper: add new compat entry for SHA at version 0.7, (keep existing compat) (#286) (@github-actions[bot])
• Support more of the metallib format (#288) (@maleadt)
• Address potentiallly buggy mtl behaviour. (#290) (@christiangnrd)
• CompatHelper: add new compat entry for CodecBzip2 at version 0.8, (keep existing compat) (#292) (@github-actions[bot])
• Remove an unneeded pointer method. (#293) (@maleadt)
• Use NSAutoreleasePool to clean up memory. (#294) (@maleadt)
• adapt_storage-related improvements (#296) (@christiangnrd)
• CompatHelper: bump compat for ObjectiveC to 2, (keep existing compat) (#297) (@github-actions[bot])
• Add support for signposts (#300) (@maleadt)
• Retain NSError we rethrow to avoid an UAF. (#302) (@maleadt)
• Minor mapreduce improvements (#303) (@maleadt)
• Specialize broadcast to avoid integer divisions. (#304) (@maleadt)
• Better Support for Unified Memory (#305) (@tgymnich)
• Add 1.11 CI (#306) (@christiangnrd)
• Remove unused files (#307) (@tgymnich)
• Skip profiling tests on macOS 14.4/M1. (#310) (@maleadt)
• Increase test timeout limit to accomodate 1.8 (#311) (@christiangnrd)
• Test all storage modes (#314) (@christiangnrd)
• Fix doctests (#315) (@christiangnrd)
• Fix KernelAbstractions for Unified Memory (#316) (@tgymnich)
• CompatHelper: add new compat entry for Preferences at version 1, (keep existing compat) (#318) (@github-actions[bot])
• Minor cleanup (#319) (@christiangnrd)
• Create MtlArray using memory allocated by Array (#320) (@christiangnrd)
• Re-enable profiling tests on M1/14.4 when using Xcode 15.3. (#322) (@maleadt)
• Small typo and doc fixup (#325) (@christiangnrd)
• BFloat16s.jl extension and related improvements (#326) (@christiangnrd)
• Support for Julia 1.11 (#327) (@maleadt)

Closed issues:

• Validation-related back-end crash on macOS Ventura (#34)
• slow broadcast copy in 2D (#41)
• Poor performance of mapreduce (#46)
• Multiplication with SubArrays (#47)
• Add support to creating MtlArray using a memory allocated by Array (#62)
• Improve use of unified memory (#86)
• Use Autoreleasepools with Metal (#103)
• Unknown RFLT tag generated by macOS 13 Metal compiler (#167)
• mapreduce allocates a lot on the CPU (#211)
• Legalization errors with vectorized code (#257)
• Compilation Failure due to undefined symbols (#276)
• resize!, append! not defined (#277)
• tag new version (#278)
• Panic during profiling tests on 14.4 beta (#281)
• M3 backend cannot handle atomics with complicated pointer conversions (#282)
• Int128 does not compile (#287)
• Two suspicious mtl-related behaviours (#289)
• LU factorization: add allowsingular keyword argument (#299)
• Autorelease changes lead to use after free with errors (#301)
• Reductions don't work on Shared Arrays (#312)

v1.0.0

Metal v1.0.0

Diff since v0.5.1

Merged pull requests:

• Matrix batches (#158) (@tgymnich)
• Add 1.10 CI. (#256) (@maleadt)
• Update manifest (#258) (@github-actions[bot])
• CompatHelper: bump compat for GPUCompiler to 0.25, (keep existing compat) (#259) (@github-actions[bot])
• Bump actions/checkout from 3 to 4 (#260) (@dependabot[bot])
• Update manifest (#261) (@github-actions[bot])
• CompatHelper: bump compat for CEnum to 0.5, (keep existing compat) (#262) (@github-actions[bot])
• Update manifest (#263) (@github-actions[bot])
• CompatHelper: add new compat entry for Artifacts at version 1, (keep existing compat) (#264) (@github-actions[bot])
• Reduce launch overhead by generating code to encode arguments. (#265) (@maleadt)
• Remove unused function argument (#266) (@tgymnich)
• Introduce application tracing profiler (#267) (@maleadt)
• Remove content(::MTLBuffer), use convert intead. (#268) (@maleadt)
• Allow more kwargs syntax with kernel launches (#269) (@maleadt)
• Don't re-use the IO object when shelling out to Python. (#271) (@maleadt)
• Preserve storage mode when broadcasting. (#273) (@maleadt)

Closed issues:

• Support for macOS Sonoma (#201)
• Error with Julia 1.10 (#274)

v0.5.1

Metal v0.5.1

Diff since v0.5.0

Merged pull requests:

• MPSMatrix improvements (#157) (@tgymnich)
• Update manifest (#221) (@github-actions[bot])
• Update manifest (#222) (@github-actions[bot])
• Update manifest (#224) (@github-actions[bot])
• Update manifest (#227) (@github-actions[bot])
• CompatHelper: bump compat for ObjectiveC to 1, (keep existing compat) (#228) (@github-actions[bot])
• Update manifest (#230) (@github-actions[bot])
• Fix argument types in sincos (#232) (@fjebaker)
• Update manifest (#233) (@github-actions[bot])
• Improve docs (#235) (@christiangnrd)
• Remove linear algebra section of MPS docs (#237) (@christiangnrd)
• CompatHelper: bump compat for GPUCompiler to 0.22, (keep existing compat) (#238) (@github-actions[bot])
• Port openlibm log1pf as log1p (#239) (@sotlampr)
• Port openlibm erf (#240) (@tgymnich)
• Remove 1.6-era override mechanism. (#241) (@maleadt)
• CompatHelper: add new compat entry for Requires at version 1, (keep existing compat) (#242) (@github-actions[bot])
• Update manifest (#243) (@github-actions[bot])
• enable dependabot for GitHub actions (#244) (@ranocha)
• Bump actions/checkout from 2 to 3 (#245) (@dependabot[bot])
• Bump peter-evans/create-pull-request from 3 to 5 (#246) (@dependabot[bot])
• Show METAL_CAPTURE_ENABLED in Metal.versioninfo() when the environment variable is set (#248) (@christiangnrd)
• Update manifest (#249) (@github-actions[bot])
• Adapt to GPUCompiler.jl, and other small updates. (#250) (@maleadt)
• Switch to GPUArrays buffer management. (#251) (@maleadt)
• Update manifest (#252) (@github-actions[bot])
• Update manifest (#253) (@github-actions[bot])
• Bump GPUCompiler (#255) (@maleadt)

Closed issues:

• Random access indexing into MtlArray views cause scalar indexing (#149)
• Q: How to debug kernels - KA.@print? (#223)
• Crash during MTLDispatchListApply (#225)
• Unable to compile trig functions through ForwardDiff (#229)
• symbol multiply defined! Bug/crash on Julia master, fine on 1.10 (#231)
• log1p fails on MtlArray{Float32} (#234)
• When precompiling, UndefVarError: CompilerConfig not defined (#247)

v0.5.0

Metal v0.5.0

Diff since v0.4.1

Metal.jl 0.5 is a feature release, bringing initial support for atomic operations (#168).
Low-level atomics that mimic Metal C are supported (atomic_store_explicit,
atomic_load_explicit, etc), as well as a higher-level Metal.@atomic that can be used to
update array values similar to how CUDA.jl's @atomic works. This uses native atomics when
supported, and falls back to a compare-exchange loop otherwise.

Minor changes include an update for the @device_code_agx disassembler, the addition of a
type variable to MtlArray encoding the storage mode (#194), and support for MPSVector
(#199) which should accelerate matrix/vector multiplications.

Also note that Metal.jl now disallows the construction of Float64 arrays, as these are not
support by the Metal libraries.

Closed issues:

• Support for atomics (#79)
• Make MtlArray storage mode a type parameter (#190)
• Long stacktrace when trying to create Float64 rand arrays (#205)
• allowscalar equivalent for Metal.jl (#206)
• Define map! ? (#219)

Merged pull requests:

• Implement atomics using compiler intrinsics (#168) (@maleadt)
• Parameterize MtlArray storage mode (#194) (@christiangnrd)
• Implement MPSVector (#199) (@tgymnich)
• Update manifest (#200) (@github-actions[bot])
• Add Metal 3.1 to MTLLanguageVersion (#202) (@christiangnrd)
• Update manifest (#203) (@github-actions[bot])
• CompatHelper: bump compat for GPUCompiler to 0.21, (keep existing compat) (#204) (@github-actions[bot])
• Update manifest (#207) (@github-actions[bot])
• Disallow Float64 arrays entirely. (#209) (@maleadt)
• Adapt to LLVM.jl 6. (#213) (@maleadt)
• Update manifest (#215) (@github-actions[bot])
• Bump disassembler. (#216) (@maleadt)

v0.4.1

Metal v0.4.1

Diff since v0.4.0

Closed issues:

• Command buffer callbacks can cause bus error during thread adoption (#138)
• how to set up Project.toml (#185)
• Metal.rand() creates a CPU array (#187)
• fill! for Int8 errors when the value is negative (#192)

Merged pull requests:

• Refactor matmatmul code for faster load time (#186) (@dkarrasch)
• Add *.DS_Store to .gitignore (#188) (@christiangnrd)
• Add GPUArrays out-of-place random methods (#189) (@tgymnich)
• Revert "Don't rely on thread adoption for command buffer callbacks." (#191) (@maleadt)
• Fix fill! with negative Int8 values (#193) (@christiangnrd)
• disambiguate gemm_wrapper! with LinAlg.jl (#195) (@dkarrasch)
• Add type annotations for character args in matmatmul (#196) (@dkarrasch)
• Handle missing adjoint case. (#197) (@maleadt)
• Fix transposed matmul. (#198) (@maleadt)

v0.4.0

Metal v0.4.0

Diff since v0.3.0

Closed issues:

• Restore mtlcall (#17)
• mapreduce has poor performance (#87)
• Native code reflection (#95)
• rand! with Bools sometimes fails in tests in 1.9 (#141)
• LLVM assertion failures (#153)
• Time macro similar to CUDA.@time (#160)
• bug in rand!? (#162)
• Why not support threadIdx().x, blockIdx().x, blockDim().x etc? (#163)
• Incorrect(?) darwin version in 1.8 with Metal.versioninfo() (#179)

Merged pull requests:

• Add native code reflection. (#96) (@maleadt)
• Move MPSKernels into a dedicated file (#155) (@tgymnich)
• [LU decomposition] Fix types (#156) (@tgymnich)
• Update manifest (#161) (@github-actions[bot])
• Implement Time macro (#164) (@christiangnrd)
• Fix some references to CUDA (#165) (@christiangnrd)
• Fix GPUArrays RNG interface implementation. (#166) (@maleadt)
• Bump the LLVM back-end. (#169) (@maleadt)
• Update manifest (#170) (@github-actions[bot])
• Update manifest (#171) (@github-actions[bot])
• Update manifest (#172) (@github-actions[bot])
• Bump GPUCompiler to v0.20 (#173) (@christiangnrd)
• Detect mapreduce threadgroup limits instead of guessing. (#176) (@maleadt)
• Remove reference to no longer used library in README.md (#177) (@christiangnrd)
• Report package versions as part of versioninfo() (#180) (@christiangnrd)
• Fix Darwin version indentification (#181) (@christiangnrd)
• Topk for MPSMatrix (#182) (@christiangnrd)
• Update manifest (#183) (@github-actions[bot])
• Don't rely on thread adoption for command buffer callbacks. (#184) (@maleadt)

v0.3.0

Metal v0.3.0

Diff since v0.2.0

Closed issues:

• Migrate to metal C++? (#2)
• Improved errors when calling device functions on CPU (#90)
• Improve Objective-C interfacing (#104)
• Rename grid to groups (#116)
• Add functionality check helper (#121)
• inputing non-isbits types (#128)
• @metal docstring out-of-date (#129)
• mapreduce kernel uses too many threads (#132)
• Powers don't work with complex floats (#142)

Merged pull requests:

• Add contributing documentation (#93) (@max-Hawkins)
• Reduce multiple consecutive values in each thread to improve efficiency (#112) (@maxwindiff)
• Remove libcmt, use native ObjectiveC FFI (#117) (@maleadt)
• Rename grid to groups (#119) (@habemus-papadum)
• Audit MRR (#122) (@maleadt)
• Faster in-place reduction by using broadcasting to initialize partial… (#123) (@maxwindiff)
• Add MPS matrix decompositions (#124) (@tgymnich)
• Minor documentation formatting (#125) (@asinghvi17)
• Switch default mode to private storage (#126) (@christiangnrd)
• Update manifest (#127) (@github-actions[bot])
• Add some MtlArray docs (#130) (@christiangnrd)
• Port MetalKernels (#131) (@maxwindiff)
• Adapt to GPUCompiler 0.18. (#134) (@maleadt)
• Support passing non-isbits arguments, as long as they're unused. (#135) (@maleadt)
• Do not change grain size after pipeline creation (#136) (@maxwindiff)
• Bump GPUArrays. (#137) (@maleadt)
• Specialize GPUArrays' global_size query. (#139) (@maleadt)
• Catch errors that happen during command buffer callbacks. (#140) (@maleadt)
• Call the correct current_device() in reflection (#143) (@maxwindiff)
• Error when calling device functions on CPU (#144) (@christiangnrd)
• Implement MTLGPUFamily and use it to validate gpu (#146) (@christiangnrd)
• Add functional() (#147) (@christiangnrd)
• Update manifest (#148) (@github-actions[bot])
• CompatHelper: add new compat entry for StaticArrays at version 1, (keep existing compat) (#151) (@github-actions[bot])
• Update to LLVM.jl 5 and GPUCompiler 0.19. (#154) (@maleadt)

v0.2.0

Metal v0.2.0

Diff since v0.1.2

Closed issues:

• Threadgroup memory breaks on small datatypes (#26)
• Int64 not supported on AMD GPUs? (#38)
• Base.unsafe_convert is ambiguous (#42)
• Support for multiple devices (#44)
• Add CITATION file (#55)
• XGBoost on Metal.jl (#82)
• first try at metal (#84)
• Copysign intrinsic possibly wrong (#89)
• Metal.jl fails to precompile on Linux (#97)
• Silent failure with unsupported(?) Intel Iris Graphics (#109)
• I have 2 question about Metal.jl and Flux.jl (#110)

Merged pull requests:

• Update manifest (#57) (@github-actions[bot])
• Add GPU profiling capabilities (#58) (@max-Hawkins)
• Automatically detect if we need cmt build from source. (#59) (@maleadt)
• Update manifest (#60) (@github-actions[bot])
• Add queue kernel launch argument (#61) (@tgymnich)
• Update manifest (#63) (@github-actions[bot])
• Switch pipeline to juliaecosystem (#64) (@vchuravy)
• Update manifest (#65) (@github-actions[bot])
• Add a function for setting the current device (#66) (@maxwindiff)
• Add documentation webpage (#67) (@max-Hawkins)
• Wrap simdgroup matrix functions (#70) (@maxwindiff)
• Support loading/saving simdgroup matrix from threadgroup memory (#71) (@maxwindiff)
• Conditionalize the MtlDeviceArray element-type workaround. (#72) (@maleadt)
• Add basic SIMD shuffle up/down (#73) (@max-Hawkins)
• Update manifest (#74) (@github-actions[bot])
• Optimize warp reduction for mapreduce (#75) (@max-Hawkins)
• Specialize GPUArrays.global_index() to improve broadcast performance (#76) (@maxwindiff)
• Update manifest (#78) (@github-actions[bot])
• Add initial performance shader support (matmul) (#80) (@max-Hawkins)
• Use Ninja to build cmt. (#81) (@maleadt)
• Update manifest (#83) (@github-actions[bot])
• Support Julia 1.9 (#85) (@maleadt)
• Add queue parameter to unsafe_copyto (#88) (@tgymnich)
• Update manifest (#91) (@github-actions[bot])
• Add MPS tests. (#92) (@maleadt)
• Support for writing binary archives (#94) (@maleadt)
• Support precompilation and loading on non-Apple hardware (#98) (@maleadt)
• Update manifest (#99) (@github-actions[bot])
• Improve reduce performance by passing CartesianIndices and length statically (#100) (@maxwindiff)
• Do not release objects that are autoreleased. (#102) (@habemus-papadum)
• Fix path the cmt in Hacking Section of the Readme (#105) (@habemus-papadum)
• Add example showing Metal and Gtk4 integration (#106) (@habemus-papadum)
• Fix memory leak. (#107) (@habemus-papadum)
• Add a mtl function for simple recursive data conversions. (#114) (@maleadt)
• Write profile trace in the current folder. (#115) (@maleadt)

v0.1.2

Metal v0.1.2

Diff since v0.1.1

Closed issues:

• installation issue (libz.1.dylib not found) [+workaround] (#51)
• Optimally choosing threads and grid (#54)

Merged pull requests:

• Use Base.active_project. (#43) (@maleadt)
• Update manifest (#45) (@github-actions[bot])
• Add aliases MtlVector and MtlMatrix (#48) (@amontoison)
• Update manifest (#49) (@github-actions[bot])
• Wrap at-metal's output in a let block. (#50) (@maleadt)
• Update manifest (#52) (@github-actions[bot])
• Update manifest (#56) (@github-actions[bot])

v0.1.1

Metal v0.1.1

Diff since v0.1.0

Closed issues:

• Super slow broadcast (#39)

Merged pull requests:

• Fix typos in unified memory example (#37) (@pitmonticone)
• Fix the launch heuristic. (#40) (@maleadt)