* nak works
* TestOps::test_add works
* testop has no crashes
* fix bool casts
* fix typo
* add disassemble
* RANGE and locals/regs
* simplify NAKCompiler
* disass cleanup
* cleanup nir codegen
* almost all tests passing
* cleanup notes in extra/
* old notes
* only import nak if NIR=1
* fix new SPECIAL syntax
* fix local/shared memory
* more tests passing
* add DEFINE_VAR support
* llvmpipe kinda works
* diskcache
* some mypy stuff
* lvp passing test_ops.py
* fix imports
* actually fix imports
* remove 'stdout'
* fix llvm import
* fix mypy issues
* nicer errors
* simpler test_dtype skips
* test lvp in CI
* fix github action syntax
* fix more actions typos
* switch to mesa 25.1.0
* diskcache_put
* better generation for lvp nir_options
* b64encode shader blobs
* Revert diskcache changes
This reverts commits 930fa3de8a and 8428c694b3.
* general cleanup
* better error messages
* fix llvm import
* fix windows tests
* link with libm and libgcc_s
* fix some errors
* dont check for 'float4'
* NIR uses pointer arithmetic
* use tinymesa
* bump tinymesa
* bump tinymesa again
* update lvp nir_options
* print nir shader with DEBUG
* simplify LVPCompiler
* more tests
* "gated" STORE
* NAK is cacheable
* more tests
* all tests pass locally for NAK
* test autogen in CI
* autogen deps
* more deps
* fix uop_gc
* fix macos
* mypy
* save 2 lines
* save two more lines
* save 1 line
* save 4 lines
* save more lines
* Revert "save more lines"
This reverts commit dd3a720c5a.
* save more lines
* fix LVP on windows
* refactor
* reorganize some code
* refactor lib_gpu
* move LVP check
* out of order loads
* remove support.mesa
* bump tinymesa version
* simplify LVP jit
* macos
* macos ci
* shell: bash
* testing
* more testing
* compute brew prefix
* stupid typo
* actually fix
* lib
* stdout on macos
* inline gallivm_compile_module
* Revert "inline gallivm_compile_module"
This reverts commit b65983b151.
* elf macos
* semicolon
* inherit from CPULLVMCompiler
* ruff
* disas test
* fix libm linking
* default is fine actually
* arm works
* add elf loader link test
* fix NAK beam
* pylint is too smart by half
---------
Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
Co-authored-by: nimlgen <138685161+nimlgen@users.noreply.github.com>
* WebGPU f16 support
* Don't enable f16 yet
* dtype tests passing after bitcast fix
* Maybe all WebGPU green?
* Require shader-f16 in examples
* Minor wgsl touchup
* 1 line shorter
* Simpler
* Add transcendetal support
* log2 nan location mismatch on Vulkan
* Nan skips
* init mockcuda
* run gpu ocelot
* fix
* sfixes
* disable broken tests
* linter
* these fails as well
* pylint
* myypy
* this fails on real platforms as well
* mypy please
* [WIP] Added an approximated implementation of Sin(FP32, FP64) passing all tests on Clang runtime
* Map nan/-inf/inf as 1.0 in order to avoid doing as_const(math.inf)
* [WIP] Added a support for LLVM IR
* cleaned up the code for the mypy and linter
* [WIP] Updated fp64 supports (bitwise shift causes the compilation error), fixed linter issue.
* [Add] added fast=true mode which disables the payne-hanek reduction which is slow
* [Fix] fails to compute elements when shape includes zero
* [WIP] Added BinaryOps.ADD/BinaryOps.OR to assembly
* [wip] update the assembly for ptx
* Enables fast=True when device is one of PTX, NV, CUDA, to avoid slow bitwise ops (as lv3 reduction is not required).
* [WIP] Added an approximation of LOG2/EXP2 (FP32, FP64)
* [Fix] Cyclic dependencies existing in xlog2
* [Fix] Cycle dependency in the graph of exp2, and log2. (passing test_symbolic_ops.py)
* [Fix] keep using higher precision for exp2, but cycle graph issue remained to be fixed...
* [Refactor] removed is_metal option. xsin does not rely on fp64 when fp32 mode.
* [WIP] fp16 xsin implementation passing all tests. (still needs to be refactored)
* [WIP] Added fp16 exp2 implementation
* [WIP] Increased the precision of Log2 from 3.5 ULP to 1.0 ULP, and added FP16 Log2 approximation.
* stashed the changes for FP16 sin
* [Fix] Patch for FP16 Sin/Exp2. (updated the dtype_via, fp32_p, and lower)
* [Refactor] migration to fastmath.py, some code simplification, renamed apis in fastmath, et al.
* [Refactor] Added the function polyN to clean-up N-terms polynomial approximation.
* [Patch] Increase fp64 precision when ldexp3k if possible, and patch for fp16 exp2
* [Patch] added bitcast_forward option
* [Patch] resolved cycle graph
* patch fix cycle graph
* set bitcast_forward=True in ilogb2k
* bitcast_forward for multi.py
* E501
* Break into multiple small PRs
* [Patch] FP16 -> FP64 upcast is not anymore required since xlog2 use quad precision polyN
* [Patch] NV still required FP64 for xlog2
* updated schedule test
* updated the count of kernels
* [Update] Removed all bitwise ops (SHL/SHR), tweaked the nan manipulation of log2, passing all tests except for AMD.
* Bitcast: make them api-compatible
* [update] force to use bitcast
* updated the count of constant folding
* [Patch] Creating a mask for exp2 using x <= Inf satisfies True as long as x is a real value
* [Update] isNaN(x) Free log2 algorithm, passing PTX tests, METAL with fastmath enabled is able to handle nan well, amd backend will not crash.
* xsin is reluctant to call payne_hanek_reduction which is slow to compile, passing stable diffusion compilation in a realistic time
* some minor simplification to payne hanek reduction
* [refactor] refactored some rebundant parts existing in payne hanek
* [refactor] more readable payne hanek impl
* [refactor] improved the code consistency of payne hanek
* [experiment] topological sort when doing _recursive_group (i dunno if this is good but at least it works.)
* Revert "[experiment] topological sort when doing _recursive_group (i dunno if this is good but at least it works.)"
This reverts commit 0eee08b87c.
* use allow_buffer_view
* lets support multilazytensor
* updated the count of kernels
* [test] added the jit tests for approx ops
* keep failed constant folding tests tested, added expectedFailure
* explict the timeout deadline when testing approx jit timeout
* [WIP] Simplified the implementation of xsin, never timeouts
* [Refactor] Improved the consistency of approx sin implementation, passing time out tests
* integrated xexp2_base into xexp2
* Set switch_over=39800.0
* delete: is_buffer_fastmath_supported
* sin: compute against abs(x)
* some cleanups
* fix typo
* removed the space between param and dtype
* allow 514 kernels on CI for sd
* [refactor] no need to upcast ad ldexp3k
* [refactor] added some comments, references to help understanding the code.
* [Fix] 1.0 ULP Sine Approximation for FP16
* [update] assume e != 0
* use pow2if instead of ldexp3k to fuse payne_hanek reduction into one
* check if approximated sin/log2/exp are fused into one
* clean up changes
* test amd exp
* some code cleanup and test sigmoid
* fix: enabled payne_hanek for fp16 to achieve higher acc
* fix: payne_hanek always accumlates the value with uint64, and fp16 sin is fused to a single kernel
* [Refactor] Rename: fastmath -> transcendental
* [Refactor] Added TRANSCENDENTAL, Moved the gate function to function.py
* updated const folding tests
* TRANSCENDENTAL as a ContextVar, removed old test of cody waite reduction, added assertions, et al.
* Add: unittest.main()
* Import TRANSCENDENTAL instead of getenv
* Refactor: Added dtype check when TRANSCENDENTAL=2, more context var
* Patch: xlog2, break expt(2, 32) x 2 -> expt(2, 16) x 4 for fp16 math
---------
Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
Co-authored-by: chenyu <chenyu@fastmail.com>
