Commit Graph

Select branches

Hide Pull Requests

abstractions_4

add_local_buffer

after_qol

algebraic_upat

all_bv_no_idx

all_sink

amd_isel

amd_refac

amd_smu_reset_patch

amd_sqtt

amd_uop

amd_uop_cleanups

assign_copy

assign_try_3

before2

better_spec

better_syn

bitcast_spec

broadcast_shape_2

bt_ei

buf_heur

bump_nv_2404

cache_reshape

call_inline

call_is_kernels

callify

cast_vec_as_cast

cat_mop

changes_from_rangify

check_clsetarg

check_dims

cifar_30

cifar_30_clean

clean_load

clean_spec

clone_tg

codegen_try_2

contig_on_after

continue_mat

deepsek

default_late_allreduce

delete_kernel_py

delete_ones

delete_slow_rangeify

delta_attention

delta_attention_scan

demote_op

disable_rules

dsp_search

dsp_search_merged

dtype_shape

earlier_gater

emu_cdna

emulate_ctx

end_all_ranges

endrange

er_prioity

execution_unit

failing_assign

fancy_usb

fast_llm

fix_fa_fwd_clang_22

fix_fuse

fix_llvm_cast

fix_range_merging_sd

fix_rangeify_tests

fix_rc

fix_sched_copy

fix_slice_store

fix_src_spec

fixup_fix_kernel

full_spec

fun_w_egraphs

gate-on-load-store

gate_rewrite

gemma4_gpt

gen_pdf_fast

gh-pages

gl_dims

gldims

glm_flash

grad_outerworld

grok_x86_novec

group_div_rules

hcq2_vram_copies

heur_4096

heur_group

hk_gemm

http_threading

hwtest_fixes

id_store

if_range_cleanups

ignore_oob

image_idx_xy

image_in_mc

image_no_vec

image_refactors_stack

index_to_slice

invalid_try_3

jinja_template

kernel4

kernel_is_call

kernelless

kimi_slop

kitten_matmul

late_Locals

late_add_load

late_allreduce

late_codegen

late_decomp_is_late

late_lower_idx

late_view

late_weakint

lds_double_buffering

less_wmma_arg

lil_dg_change

lil_rangeify_td

llama_trainer

llm_262k_minimal

llm_flash_attn

llm_qwen

llm_serve_fixes

load_view_late

local_cache

locals_have_device

loop_end_op

loop_is_rng

lowerer_hacks

mac_pytest

manual_kernel_3

master

mega_lowerer

min_new_lin

min_outer_range

minor_views

mixins7

mlperf_training_v4.1

mlperf_training_v5.0

mlperf_training_v6.0

modern_scan

more_apps

more_early_comps

more_folding_rules

more_kernelcount

more_pcode

more_rangeify_pm_tests

more_ren_clean

more_uop_gc_test

more_uop_mm

more_vmap

move_devec

move_dtype_decomps

move_expander

move_gates_to_load_store

move_mop_cleanup

move_schedule

move_simp_views

moveleftright

mt_typing

multi_range

multioutput

new_ck

new_devec

new_invalid_and_image

new_linearizer

new_shape

new_test

no_assign_2

no_buf_0

no_decimals

no_late_simplify_marg

no_merge_ends

no_merge_views

no_mstack_in_allreduce

no_one_ranges

no_sip_usbgpu

no_tsink_base

no_tuplize

no_user_ck_contig

no_vec_x86_4

no_weakint_variable

noop_index

novecimg

num_batches_tracked_shape

olmo3_support

one_compiler

one_hcopt

only_assign_buffers

only_loop_backends

only_reg_emu

only_reg_emu2

outer_range

outerworld_work

pad_invalid_work

pad_mixin

param_call_is_function

parse_sqtt_more

pci_iommu

port_nir

postrange_hcopts

precompile_backward

prefill_qwen_fast

print_special_ops

ps_clean

pyremu_try

python_prg

python_speed

q4k_support

qcom-openpilot-13ms

qcom_830

qcom_mmapeak

qcom_speed_codex

qol_string

qwen36_27b_amd_900

qwen36_27b_amd_minimal

qwen36_amd_bugfixes

qwen_mergable

r_cleanups

r_new_opt

ranged_call

rangeify_2

rangeify_const_folding

rdna3_vibes

rdna4_asm

rdna4_gemm

real_pyrender

recurse_sub

reg_coalese

relu_pattern

remove_assign

remove_cast_hack

remove_ck_contig

remove_define_var

remove_forced_re

remove_on_stack

remove_programspec

remove_pylint

remove_slow_tests

remove_vec_2

remove_vec_3

renderers_use_shapes_2

replace_if_with_range

reprocess_node_experiments

reshape_trait

resnet-ci

revert-11547-fix_mismatch_red

revert-11715-revert-11711-amd_llvm_tc_cast

revert-11809-no_ki_in_gpudims

revert-11811-better_div_nesting

revert-11876-memplan_vbig

revert-12104-simplify_valid_in_symbolic

revert-12308-bufferize_early

revert-12731-jitlink_init_check

revert-12813-after_cleanups

revert-13046-better_cleanup_arange_buffers

revert-13311-prof_link_sqtt

revert-13583-rocr_desc_2

revert-14734-tload

revert-17141-full_rewrite_to_sink

revert-17408-_broadcasted-weak

revert-9438-simple_matmul_np_rand

revert_wrong_ck_fix

rewrite_rangeify2

rtoposort

scan_assign

scan_op_8

sched_cache

sched_lin

sd_load_simple

self_type

set_testpath

shrink_in_render

simp_spec

simple_gidxs

simple_pool

simple_priority

simple_rangeify

simpler-sd-compile

simpler_end

simpler_fusion

simpler_image

simpler_postrange

simpler_reduce

slice_like_shrink

spec_3

spec_cleanup

spec_ren

speed_one_linear

split_buf_idx

split_decomp_pass

split_sub

stack_0

store_is_void

string_ranges

support_opts_in_contig

swizzle_in_kernel

sym_llm

sym_work

symbolic-chunk-split

szdiff_bot_test

tc_is_heur

tensor_graph_single_sink

test_amd_crash

test_fa

test_from_postopt

test_hgemm

test_unit

tiny_scan

tmr

tool_call_support

topoprop

ttomsa-new_linearizer

tuplegtuple

two_stage_remove

typed_checks

typing_work

unused_permute_arg

uop_prg

upd_test4

update_benchmark

update_benchmark_search

update_mlperf

user_failure

vecless_x86_glm

verify_sink_dims_clean

vf_ish

view_after_const

view_in_codegen

viz_cbroadcast

warp_fun

wgpu-f16-packed

working_postopt

x86_moves

x86_numel

v0.10.0

v0.10.1

v0.10.2

v0.10.3

v0.11.0

v0.12.0

v0.13.0

v0.4.0

v0.5.0

v0.6.0

v0.7.0

v0.8.0

v0.9.0

v0.9.1

v0.9.2

Mono Color

• 86439561fa a lil more geohot 2026-01-28 13:29:27 +08:00
• c49aca3bc0 wmma refactor geohot 2026-01-28 13:24:56 +08:00
• 0afa914063 Fmt matters it seems geohot 2026-01-28 13:14:14 +08:00
• 067e27857e nested composite actions don't work (#14393) sirhcmandGitHub 2026-01-27 21:13:30 -08:00
• 5c33ff6536 correct fix + tests geohot 2026-01-28 12:48:42 +08:00
• 8eaa6b593e fix tests geohot 2026-01-28 12:40:25 +08:00
• cd12f54982 smem test geohot 2026-01-28 12:35:40 +08:00
• 9dddf3d478 don't save caches for PRs, try 2 (#14391) sirhcmandGitHub 2026-01-27 20:30:17 -08:00
• 4ae8cf9184 junk remove geohot 2026-01-28 12:24:48 +08:00
• 68fe5d8b36 Revert "don't save caches for PRs (#14389)" (#14390) sirhcmandGitHub 2026-01-27 20:22:26 -08:00
• 4ab228b498 don't save caches for PRs (#14389) sirhcmandGitHub 2026-01-27 20:21:31 -08:00
• d697f7073a bits junk remove geohot 2026-01-28 12:21:23 +08:00
• 5e36482314 decompose long to ints where unsupported, try 2 (#14383) sirhcmandGitHub 2026-01-27 20:20:43 -08:00
• 09709b082a remove bad helpers geohot 2026-01-28 12:17:01 +08:00
• 68abb5bf2d write clean code geohot 2026-01-28 12:07:00 +08:00
• 313897eb2e dup code geohot 2026-01-28 12:04:38 +08:00
• 8e1fbe2f5b use the correct apis geohot 2026-01-28 12:00:47 +08:00
• 2a9e0af9b4 no ctx geohot 2026-01-28 11:51:36 +08:00
• e496547720 llama3 gradacc (#14291) wozeparrotandGitHub 2026-01-27 19:48:10 -08:00
• d7b2033009 flat geohot 2026-01-28 11:46:07 +08:00
• ef82390499 merge geohot 2026-01-28 11:23:49 +08:00
• 6d45afd1ec simpler geohot 2026-01-28 11:19:19 +08:00
• 88bc5ee212 assembly/amd: rename to better names (#14384) George HotzandGitHub 2026-01-28 10:00:54 +08:00
• 065b95cfb0 Revert "add retry to fetch (#14370)" (#14385) George HotzandGitHub 2026-01-28 09:35:37 +08:00
• dc4d7f2d55 add retry to fetch (#14370) Eitan TurokandGitHub 2026-01-28 00:04:25 +02:00
• 8d1f3c8885 fix copysign for inf input (#14381) chenyuandGitHub 2026-01-27 16:45:48 -05:00
• 289a3e415e also skip test_nonoverlapping_shrink_assignment (#14382) sirhcmandGitHub 2026-01-27 13:26:26 -08:00
• f34efc1ad1 DISABLE_FAST_IDIV actually works as a ContextVar (#14378) sirhcmandGitHub 2026-01-27 13:12:42 -08:00
• 8c899e4aaf fix copysign for -0 (#14380) chenyuandGitHub 2026-01-27 15:44:58 -05:00
• 62884585a7 failed test case for copysign -0.0 (#14379) chenyuandGitHub 2026-01-27 14:37:17 -05:00
• ec1b28bc2c am: exit early in case of failures (#14376) nimlgenandGitHub 2026-01-27 22:10:02 +03:00
• cd22ee9ed0 add InvalidType to ConstType [pr] (#14373) chenyuandGitHub 2026-01-27 14:09:34 -05:00
• 5b42a1357b SCACHE=0 works with DEBUG (#14377) sirhcmandGitHub 2026-01-27 10:12:43 -08:00
• db010a31be IGNORE_OOB -> CHECK_OOB [pr] (#14374) chenyuandGitHub 2026-01-27 12:20:59 -05:00
• c22667b0c4 also skip test_overlapping_shrink_assignment_reverse (#14375) chenyuandGitHub 2026-01-27 12:20:39 -05:00
• e52d58b041 autogen: update amd (#14372) nimlgenandGitHub 2026-01-27 19:53:14 +03:00
• cbf94a0a95 nv: exit early in case of failures (#14363) nimlgenandGitHub 2026-01-27 19:16:22 +03:00
• ec691cb299 am: print sq intrs (#14366) nimlgenandGitHub 2026-01-27 18:28:13 +03:00
• a5f3d46423 hcq: do not assume kernel names are unique (#14371) qazalandGitHub 2026-01-27 09:03:15 -05:00
• e5df7e640b fix branches in amd_asm_matmul (#14369) George HotzandGitHub 2026-01-27 20:48:42 +08:00
• 0ced258726 HOTFIX: skip crashing assign test geohot 2026-01-27 20:35:17 +08:00
• 131ae604de force_transcendental on sqrt (#14368) George HotzandGitHub 2026-01-27 20:24:41 +08:00
• 14574c68fa Add ContextVar to disable the scheduler cache (#14257) 3 people 2026-01-27 06:55:29 -05:00
• bfc88bcfb8 assembly/amd: emu refactors + enable PYTHON_REMU by default (#14361) George HotzandGitHub 2026-01-27 17:42:24 +08:00
• 2e72625652 Revert "decompose dtypes.long to ints where unsupported (#14261)" (#14362) sirhcmandGitHub 2026-01-26 23:04:59 -08:00
• f866b2a513 mfma loop in asm dsl (#14349) qazalandGitHub 2026-01-26 21:11:37 -05:00
• 0793319929 decompose dtypes.long to ints where unsupported (#14261) sirhcmandGitHub 2026-01-26 15:34:13 -08:00
• a987a4abc3 feat: llama8b dev_beam.sh (#14358) wozeparrotandGitHub 2026-01-26 14:51:23 -08:00
• c9c533fc78 libclang path is homebrew on macos (#14357) sirhcmandGitHub 2026-01-26 14:32:09 -08:00
• d641e63189 improve min/max for AND (#14356) chenyuandGitHub 2026-01-26 15:44:18 -05:00
• f16372487a fix assign hazard on shrink (#14355) chenyuandGitHub 2026-01-26 14:46:30 -05:00
• 145df879c1 find_permutes -> fix_assign_hazard [pr] (#14354) chenyuandGitHub 2026-01-26 14:05:19 -05:00
• e152f1b0f5 llama: use ALL2ALL (#14353) nimlgenandGitHub 2026-01-26 22:01:53 +03:00
• 3f25eb3026 am: ih (#14346) nimlgenandGitHub 2026-01-26 20:11:04 +03:00
• d8d2ad92e7 Merge branch 'master' into pyremu_try George HotzandGitHub 2026-01-27 01:05:07 +08:00
• 823bc17fb5 failed test case for shrink overlap assigns (#14350) chenyuandGitHub 2026-01-26 11:58:45 -05:00
• b7031aa381 assembly/amd: set PYTHON_REMU=1 geohot 2026-01-27 00:48:58 +08:00
• 204f51e739 assembly/amd: bug fixes for PYTHON_REMU (#14347) George HotzandGitHub 2026-01-27 00:48:22 +08:00
• 231305603d remove REAL_DEV [pr] (#14337) chenyuandGitHub 2026-01-26 10:08:16 -05:00
• 9cbe99348a func meshgrid: change param index to type str (#14331) Martin SzewieczekandGitHub 2026-01-26 16:07:56 +01:00
• 3b43d26f10 assembly/amd: emu speed (#14344) George HotzandGitHub 2026-01-26 22:21:34 +08:00
• 774a454bb5 assembly/amd: fix scratch SVE (#14340) George HotzandGitHub 2026-01-26 21:03:51 +08:00
• 2d91fe6310 use amdgpu dsl in mmapeak (#14342) qazalandGitHub 2026-01-26 08:03:43 -05:00
• b2e2ace85b viz: remove ci check, it's VIZ=-1/-2 (#14343) qazalandGitHub 2026-01-26 06:36:23 -05:00
• be23776ba7 assembly/amd: replace pcode with ucode (#14002) George HotzandGitHub 2026-01-26 18:04:29 +08:00
• 984cdc4840 add wrapper class for the -0.0 != 0.0 issue (#14339) George HotzandGitHub 2026-01-26 16:52:37 +08:00
• 92bfe92138 assembly/amd: fix cdna mfma xml (#14329) qazalandGitHub 2026-01-26 03:51:26 -05:00
• 6c109f4d75 LLVM: CPU threading support (#14320) 3 people 2026-01-25 21:12:39 -08:00
• cc49e47ea2 tinygrad changes from ucode (#14336) George HotzandGitHub 2026-01-26 11:30:18 +08:00
• dd48f6a111 Merge remote-tracking branch 'upstream/master' into new_x86_backend ttomsa 2026-01-26 02:34:46 +00:00
• 1fe4185e89 start new scheduler ttomsa 2026-01-26 02:30:38 +00:00
• 8477368d07 generic LLVMRenderer class for CPU and AMD (#14321) 3 people 2026-01-25 17:11:49 -08:00
• 11ce1e847d llama train: null device support geohot 2026-01-26 08:53:05 +08:00
• e3601788fa update torch backend function (#14333) chenyuandGitHub 2026-01-25 16:39:34 -05:00
• 9865f51e39 cupti: ref collector (#14330) nimlgenandGitHub 2026-01-25 20:35:21 +03:00
• 21ab23ae18 nv: add pma for ada (#14328) nimlgenandGitHub 2026-01-25 17:33:37 +03:00
• 49db266b96 ReprEnum for repr roundtrips (#14327) George HotzandGitHub 2026-01-25 18:58:31 +08:00
• bf2d9d138f viz: simplify amdgpu cfg (#14326) qazalandGitHub 2026-01-25 01:21:45 -05:00
• 647e527a7e viz: replace llvm disasm with our disasm (#14325) qazalandGitHub 2026-01-24 23:56:56 -05:00
• 4280a8eef2 am: update fw (#14323) nimlgenandGitHub 2026-01-25 01:08:47 +03:00
• 7e41da1ae8 fix generate_dataset.sh (#14324) chenyuandGitHub 2026-01-24 16:47:10 -05:00
• 311bfd91d6 clean up where_on_load [pr] (#14322) chenyuandGitHub 2026-01-24 14:43:43 -05:00
• 8b282ba6d2 memory: reserved vram (#14318) nimlgenandGitHub 2026-01-24 19:39:24 +03:00
• 00e9ba0b82 update type for split_uop and where_on_load [pr] (#14319) chenyuandGitHub 2026-01-24 11:17:41 -05:00
• cb69b7b2b2 comment out fold_where_closure (#14316) chenyuandGitHub 2026-01-24 10:15:42 -05:00
• d74587f16d fa multi fix 2 (#14314) wozeparrotandGitHub 2026-01-23 23:35:02 -08:00
• d9f0ad1d87 update return type for Tensor.tolist (#14313) chenyuandGitHub 2026-01-23 23:21:49 -05:00
• 807bc40931 assembly/amd: dsl and disasm cleanup (#14311) qazalandGitHub 2026-01-23 21:36:12 -05:00
• e782d44918 WEBGPU/NIR truncates ints (#14307) sirhcmandGitHub 2026-01-23 16:28:06 -08:00
• 26220a472e no core_id (#14265) nimlgenandGitHub 2026-01-23 21:30:12 +03:00
• e65bc7a7c5 where closure folding (#14304) chenyuandGitHub 2026-01-23 10:55:13 -05:00
• d5a3b02a9c clean up xpow (#14295) chenyuandGitHub 2026-01-23 10:19:47 -05:00
• b913c910c5 assembly/amd: rdna4 passing test_roundtrip (#14300) 3 people 2026-01-23 07:33:53 -05:00
• f3b0e42863 remove extra sqtt pickles in gfx1200 (#14302) qazalandGitHub 2026-01-23 06:13:48 -05:00
• d116312b1a get cdna sqtt working (#14301) George HotzandGitHub 2026-01-23 18:46:15 +08:00
• a5c4fa39d1 RDNA4 support in SQTT (#14299) George HotzandGitHub 2026-01-23 16:16:45 +08:00
• 963c59ebdb fix: pull fixes from gradacc branch (#14296) wozeparrotandGitHub 2026-01-22 23:07:54 -08:00
• 68668b8f28 fix WEBGPU NEG (#14298) sirhcmandGitHub 2026-01-22 22:44:52 -08:00
• 3b8a7bb8c9 use existing roc.py infra for sqtt tests (#14297) qazalandGitHub 2026-01-23 00:07:11 -05:00
• 5f32f7a06b fix winograd padding order (#14294) chenyuandGitHub 2026-01-22 23:00:14 -05:00