* global -> group
* allow None for local_size in custom function
* lil local
* comment on shape
* fix cuda
* smart local cast
* better local heuristic
* fix ptx, and work_dim cleanup
* fix metal
* fix ops test
* fix openpilot jit
* no more optlocal
* might fix metal tests
* try metal now
* see generated metal code
* test free removal. REVERT THIS
* mergable
* matrix strategy
* push env to GITHUB_ENV
* use printf instead of echo
* use temp helper function for cross os paths
* use path join
* switched to using temp helper function
* skip test on windows due to memory limit
* small fix
* removed semi
* touchups
* clean up
* seperate tests
* test changes to test_utils on windows
* small refactor
* more cleanups
* undo helpers change
* only skip if in CI and WINDOWS
* e2e testing
* min failure
* no affine on bn, still fails
* why did i think i could detach that?
* allow more kernels for bn
* some test issue i don't understand
* fix binop, other tests failure
* that was a bad idea
* better layernorm
* inference kernel count tests
* new style reshape pushing
* fixup replacement
* 199 kernels is okay. fix flops
* push reshape through unaryops only
* GRAPH=2 draws the phantom ops
* found resnet issue
* non working test
* mul is cheaper than div
* OPT inflation
* SHUFFLE_PAD_OPS in OPT=2
* runs one metal kernel
* conv2d works
* ops tests are passing
* const folding
* all ops work
* pre commit always passes
* torch works
* working still
* fix graph test
* tests passing
* image almost works
* image conv works
* most images
* fix custom
* fix assignment
* fix compile enet
* clean up comments
* fix realize return value
* include shapetracker in LB repr
* copy should make a copy
* reenable method cache
* fix lna
* dtypes in graph
* forward only for IMAGE=2
* simple realize
* getting close
* fixup new api, it's good except the kernel count
* back to 197 kernels
* tests should pass
* go to a real float
* no type_on_cpu
* fix the docs
* put shapetracker back in it's proper place
* add dtype class
* dtypes
* buffers are lazy
* dtype is tracked by lazybuffer and GenericShape
* fix types in llvm
* llvm store
* dtype tests
* fix tests maybe
* fix flop counter
* fix CI
* CI fix and check format
* fix dtype and dtype check
* fix custom test
* fix test graph
* behavior is correct without VALIDHACKS
* simple div and mod
* fix tests
* no negative variables
* alt form is correct
* still correct
* bug in mulnode
* at least validhacks works now
* cleanups
* test validhacks, and to_image_idx
* cache compare key
* tests and __neg__
* conv2d is an hlop
* shorter conv
* KOPT=-1
* alt imp
* MULACC
* smarter mulacc
* pop conv
* 7x7 -> 5x5
* didn't fix, that's not going to work
* this is faster and matches old behavior
* oh, non lazy just won't work with mulacc
* mulacc in torch
* bool types were creeping in
* optimizer is actually better with hlop conv
* fix pushing permutes issue
* refactor einsum_mulacc
* fix up readme
* update readme
* _image_conv2d
* fix bias addition location
* pushing permutes gets back to 200 kernels
* conv cleanup
* disable hlop conv
* don't hide that in helpers
* start clang backend
* mostly working
* no group for reduce w clang
* it compiles
* compiles
* a11y
* minor fixups
* formatting
* add a test
* rename test
* add image
* load + store + boring stuff:
* image tests pass
* thneed print GFLOPS
* op conv test
* more debugging
* hack for multiview image
* shapetracker creates less views
* disable image tests
* working better
* ugh, lkey not key
* print in DEBUG, and allow views
* works
* simple padding conv2d
* use index for image
* that was bad code
* debug print
* fix types
* less lines
* save lines
* bringing back reshape and permute
* done with E701
* 4x4 works in generic way
* max and sum not vectorizing...
* special case single float
* support comparing to MPS
* improve matmul speed, consider generic principles
* GlobalCounter
* fix op tracking
* faster
* comment that out for now
* err, it needs that
* fix minor issues
* fix global_mem
* chonker will make llvm fast
* work
* better speed tests, we will make them fast
* with the cache add is the same speed
* relu and neg are fast
* fix sum speed
* maximum maxnum?
* hack for gemm opt
* gemm very slow
* zeros like
* test_permute
* shapetracker returns self
* fix shapetracker factorization
* err, int strides
* permutes are faster now in tinygrad than pytorch
* support -1 in expand
* gemm unrolled
* improve final test case
* WIP GEMM
* why isn't GEMM fast?
* revert cache dim
* ffp contract works on clang, not llvm?
* ignore llvm ir
* this makes fma work at least, but no faster
* USE_4x4
* 63 GFLOPS
* 87 GFLOPS
* that wasn't matmul, 44 GFLOPS now
* 82 GFLOPS permuted
* this permute too
* a little speed for the convs
* 45 GFLOPS
* speed tests pass again
* clean up prints
* fix FMA WHAT A WASTE OF TIME
* colors
* moar fair
* GPU
* useless on chonker
* cleanups
* improve factorized shapetracker
* better threshold
* label conv
* work
* ops test pass again
* hot load the index
* run the last view, no need to create
* ZeroView needs a repr for the key to work
* fix segfault on out of bounds
* one more test
* start amx, and llvm.initialize_native_asmparser
* amx works
* nice AMX class
* nicer AMX class
* refactor get_idxs
* amx working
* is slower...
* useless flip
* cache
* SZ_X
* AMX_SZ_X/Y work alone
* Contiguous mlop
* test gemm packed
* PREPARE in packed
* use_amx factor
* prefetch isn't faster
* loop
* same 3ms
* 2.24 ms
* allow double on store in TG
* amx reduce is the same speed as non amx reduce
* include memory bandwidth
* clean up shapetracker
* flip returns stride
* prepare for upstream
* Update ops_llvm.py (#426)
* permutes are yellow and green now
* faster conv
* llvm cleanups
* Show optimised IR under debug 4 (#428)
* ASTKernel class
* Make tinygrad work with older python version (#427)
* Make tinygrad work with older python version
* Use partialmethod instead of partial
* smiple chonker is chonking
* remove junk from test speed vs torch
* fix linker and types
* AMX is only here now
* add LLVM tests, it's a valid backend now
* oops, run llvm test
* contiguous_op
* fix loadops compare
* dedup reduceops
Co-authored-by: calledit <1573053+calledit@users.noreply.github.com>
* working exec ast
* exec_ast is staticmethod
* GenericExecAST
* fold that sometimes
* ExplicitExecAST
* exec_ast for GPU
* gpu working
* get_lazyop_shape
* now gpubuffer is ExplicitExecAST
* dedup
* add a type
* RESHAPE in opencl code
* fix linter
* that too for linter
* cleanups
* remove dead code
* GenericShape is less lines
* add ALLOWED_KERNEL_COUNT to tests
* fix mypy
* that's gotta be recursive
* fix opencl shape processing
* remove unneeded lambda
