wozeparrot and GitHub
36f6d1b064
gemm: fix bf16 atb for mp sharding ( #16637 )
2026-06-16 15:58:47 -07:00
wozeparrot and GitHub
c38d6a7e3a
mxfp8 part 2 ( #16561 )
2026-06-10 23:36:11 -07:00
wozeparrot and GitHub
2bdc360606
gemm: mxfp8 hipkittens gemm ( #16541 )
...
* gemm: mxfp8 hipkittens gemm
* feat: update hipkittens
* feat: kernel signature
* clean: just kernel
* feat: from tinygrad
* feat: test
* fix: add back utils
* clean: no diff
* clean: no diff
2026-06-09 15:20:05 -07:00
sirhcm and GitHub
172f9493e1
move is_dtype_supported to renderer ( #16226 )
2026-05-20 21:19:37 -04:00
chenyu and GitHub
8631b6f17d
remove use of requires_grad in test/ ( #16237 )
2026-05-16 17:21:07 -04:00
chenyu and GitHub
0ddc50d050
do not gate backward on requires_grad ( #16230 )
...
DETACH is filtered in _deepwalk. instead of None, it gets 0 grad now
2026-05-16 12:29:49 -04:00
qazal and GitHub
ebcb7b7cc0
fp8 gemm tests with scale args ( #16231 )
...
* update atol
* update fp8 path
* more work
* update profile.sh
2026-05-16 20:47:58 +09:00
chenyu and GitHub
409bb0c9ad
requires_grad cannot be None ( #16212 )
...
final goal is to remove requires_grad, first change the default to True, and don't allow None
2026-05-15 02:01:04 -04:00
chenyu and GitHub
7a1adfd2aa
update Tensor.allclose to return Tensor ( #15904 )
...
matches jax
2026-04-24 08:27:17 -04:00
sirhcm and GitHub
6adf4c3cd9
MOCKGPU interfaces ( #15796 )
2026-04-17 21:56:29 -04:00
qazal and GitHub
39a029ec55
remove ASM_GEMM context var ( #15645 )
2026-04-08 18:02:40 +09:00
wozeparrot and GitHub
70dbd35023
llama: move custom_kernel into flat_llama ( #15643 )
2026-04-08 00:19:14 -07:00
7e54992bf6
fp8 llama ( #15588 )
...
Co-authored-by: qazal <[email protected] >
2026-04-04 18:24:57 -07:00
645d45d968
DEV has arch ( #15577 )
...
Co-authored-by: Comma Device <[email protected] >
2026-04-03 19:17:19 -04:00
sirhcm and GitHub
0ed8d9271d
Renderers accept Target or nothing ( #15590 )
2026-04-03 01:09:41 -04:00
qazal and GitHub
fefb0ebc2a
gemm/asm: fp8 cleanups ( #15580 )
...
* normal gemm here
* s/dtypes.fp8e4m3/FP8_DTYPE
* gemm_bw
* device UOp stays NULL
2026-04-02 19:02:38 +09:00
qazal and GitHub
8feb8edc68
gemm/asm: add fp8 support to cdna asm_gemm ( #15542 )
...
* work
* hmm, mixins
* rhs_transposed
* also fix the dtype
* check for hipcc
* Exception
* select dev
* default
2026-03-31 19:32:54 +09:00
qazal and GitHub
f88e255cea
gemm/asm: split and parameterize dtype in llama gemm tests ( #15408 )
...
* gemm/asm: more tests for emulator, parameterize llama gemm tests
* bf16 atol
2026-03-31 17:12:44 +09:00
sirhcm and GitHub
bc180a963c
deprecate <dev>=1 in favor of DEV=<dev> ( #15467 )
...
* start work on target
* add test
* update actions to use DEV
* update docs
* update readmes
* tests need that too
* update example
* update tests (comments)
* fix that test
* ruff
* mypy
* oops
* remove getenvs
* don't add Target yet
* and the test
* lint
* and docs
* more stuff
* assert
* few more fixes
* test assert
2026-03-26 03:48:03 -04:00
176ad47d7d
cdna4 emulator testing ASM_GEMM in CI ( #15373 )
...
* cdna emulator work
* accvgprs
* cdna passes most tests
* ruff
* add cdna4 to tests
* cdna emu
* crash
* pass?
* work
* gen
* clean up wave_size access
* asm_gemm passes
* remove acc from dsl.py, emulator can keep its different reg file
it's purely an encoding here, the ASM_GEMM already encodes acc srcs with v[], this can
be cleaned up later, but not functionally required for emulator.
* split asm_gemm tests to ones fast on the emulator
* don't do that
* 124 stays null on rdna
* the segfault was because of hw regs, not this
* Revert "clean up wave_size access", it's explicitly tested
This reverts commit 1202ff5787 .
* nullcopyout
---------
Co-authored-by: George Hotz <[email protected] >
Co-authored-by: George Hotz <[email protected] >
2026-03-20 05:51:30 +09:00
qazal and GitHub
33bd33e783
sqtt: add CDNA ops enum, show in viz ( #15140 )
2026-03-17 09:38:42 +09:00
qazal and GitHub
5cd1daa3bc
cdna asm_gemm in one file, remove old rdna3 asm ( #15281 )
2026-03-16 04:32:30 +09:00
wozeparrot and GitHub
c35de9bd68
asm_gemm: support more sharding ( #15002 )
2026-03-02 23:16:37 -08:00
qazal and GitHub
5b6fcd1cda
gemm/asm: smallest cdna4 asm gemm test ( #14925 )
2026-02-21 11:56:05 +09:00
qazal and GitHub
f590564bf7
gemm multiple is only for cdna4 asm ( #14814 )
...
* gemm multiple is only for cdna4 asm
* move to backend
* and arch
* path
2026-02-17 14:00:02 +09:00