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

autogen-comgr-2

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

ci-deps

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

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

explicit-dev

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

gitea-test-ci

gl_dims

gldims

glm_flash

grad_outerworld

grok_x86_novec

group_div_rules

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-chat-template

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

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

onnx-cpu-count

outer_range

outerworld_work

pad_invalid

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

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-lp

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

view_after_const

view_in_codegen

viz_cbroadcast

warp_fun

wgpu-f16-packed

working_postopt

x86_moves

x86_numel

#1

#2

#3

#4

#5

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

• 9759899cb7 actually make the file readable geohot 2025-10-08 19:14:02 +08:00
• c5a1f9f5f9 no ShapeTrackers in multi.py (#12521) qazalandGitHub 2025-10-08 14:04:05 +03:00
• 29509a1a57 Merge branch 'master' into rtoposort George HotzandGitHub 2025-10-08 18:48:17 +08:00
• 8c3242bb45 disable failing test geohot 2025-10-08 18:47:27 +08:00
• b264772ae4 touch up tests + add comments geohot 2025-10-08 18:38:48 +08:00
• ee0382ad99 remove ShapeTracker.invert (#12520) chenyuandGitHub 2025-10-08 18:37:34 +08:00
• 3f927e7bb3 move to indexing.py geohot 2025-10-08 18:32:21 +08:00
• e131a5ef0f mstack/mselect no hacks geohot 2025-10-08 18:20:36 +08:00
• d5058427ea remove ShapeTracker.real_size (#12519) chenyuandGitHub 2025-10-08 18:15:29 +08:00
• fc01a7cc15 tests geohot 2025-10-08 18:02:58 +08:00
• 6f26603f06 delete swizzler.py (#12518) qazalandGitHub 2025-10-08 13:02:34 +03:00
• 04b7b68242 ending ranges fix geohot 2025-10-08 18:00:21 +08:00
• f09f46d028 put in rangeify map geohot 2025-10-08 17:32:40 +08:00
• 7e0b14243e delete grouper and kernelize (#12517) qazalandGitHub 2025-10-08 12:27:26 +03:00
• 9e884403e9 don't delete those shape ops geohot 2025-10-08 17:20:10 +08:00
• 942022c309 smaller LLAMA_LAYER in Test llama 3 training (#12516) chenyuandGitHub 2025-10-08 17:10:51 +08:00
• cd8942cc77 tests passing geohot 2025-10-08 17:06:43 +08:00
• e701106a64 remove FUSE_ARANGE (#12511) chenyuandGitHub 2025-10-08 16:54:07 +08:00
• 291a19650b move Kernel dataclass to rangeify (#12510) qazalandGitHub 2025-10-08 11:30:06 +03:00
• ad49f8148b switch process_replay to rangeify (#12509) qazalandGitHub 2025-10-08 11:26:43 +03:00
• da1f46ff3f remove RANGEIFY specific test jobs (#12507) chenyuandGitHub 2025-10-08 16:12:04 +08:00
• e3f073b957 Merge branch 'master' into rtoposort George HotzandGitHub 2025-10-08 16:04:43 +08:00
• 91d60d83e0 FAST geohot 2025-10-08 16:04:33 +08:00
• 1e567a5cf8 make RANGEIFY=1 the default (#12161) 5 people 2025-10-08 15:46:09 +08:00
• c9ab0a0e26 pad fix geohot 2025-10-08 15:31:40 +08:00
• 9e7103647d amd: rename cmd_id to sqtt_next_cmd_id (#12503) nimlgenandGitHub 2025-10-08 15:16:19 +08:00
• b4eb5fcac5 Merge branch 'master' into rtoposort George HotzandGitHub 2025-10-08 15:14:43 +08:00
• 985c0c20a9 progress geohot 2025-10-08 14:33:06 +08:00
• 4a756a37d8 amd: support rocm7 (#12502) nimlgenandGitHub 2025-10-08 14:30:39 +08:00
• 5042e0b5af should work geohot 2025-10-08 14:06:51 +08:00
• 53eb2af4ce fast rangeify works for mnist geohot 2025-10-08 12:58:40 +08:00
• 60b6dca5ba update some tests instead of expect_rangeify_fails (#12500) qazalandGitHub 2025-10-08 07:42:31 +03:00
• e35152a004 work geohot 2025-10-08 12:34:07 +08:00
• d823a6d9d5 fast rangeify geohot 2025-10-08 12:19:43 +08:00
• 84597ed53c early assert for device mistmatched asts in rangeify (#12499) qazalandGitHub 2025-10-08 07:19:36 +03:00
• 2e19354c1c viz: reorder timeline graphs (#12498) qazalandGitHub 2025-10-08 07:10:23 +03:00
• d06226b575 fix SPEC and all_tensors iterator (#12496) George HotzandGitHub 2025-10-08 11:18:17 +08:00
• a7cb80bfab use recursive_property in UOp device (#12477) qazalandGitHub 2025-10-08 06:15:05 +03:00
• d23324779c rtoposort is fast, can replace rangeify with this geohot 2025-10-08 11:06:12 +08:00
• a6d59a0b45 backward_slice to get srcs recursively (#12494) George HotzandGitHub 2025-10-08 10:31:42 +08:00
• eb3bc277b3 remove ASSERT_MIN_STEP_TIME in external_benchmark_openpilot (#12495) chenyuandGitHub 2025-10-08 10:13:42 +08:00
• 239f9a3029 update viz to not use children [pr] (#12493) qazalandGitHub 2025-10-08 04:35:01 +03:00
• b465c17b56 Revert "UOp.factor and add chain sorting (#12413)" (#12492) Sieds LyklesandGitHub 2025-10-08 03:20:23 +02:00
• 945cc46475 delete children tracking from uop (#12491) George HotzandGitHub 2025-10-08 09:04:14 +08:00
• 940b88f5c5 Merge branch 'master' into new_linearizer ttomsaandGitHub 2025-10-08 01:10:31 +01:00
• 332293cfa7 terrible ttomsa 2025-10-08 00:35:53 +01:00
• 4687d000a7 fix this ttomsa 2025-10-07 21:21:53 +01:00
• 381df7f45f cpu should work now ttomsa 2025-10-07 21:05:44 +01:00
• 0047700ac0 this is the correct algorithm ttomsa 2025-10-07 20:18:46 +01:00
• 648e5bb223 hcq: do not raise when fini (#12487) nimlgenandGitHub 2025-10-07 23:27:03 +08:00
• a2345787b9 parents is faster than sparents (#12490) George HotzandGitHub 2025-10-07 21:31:50 +08:00
• f082cbcb36 respect the 8x8 tiles clone_tg geohot 2025-10-07 17:51:24 +08:00
• 12c4963489 add more rangeify pm tests (#12488) George HotzandGitHub 2025-10-07 17:45:38 +08:00
• 5ad62f130d split on tile_dim geohot 2025-10-07 17:40:00 +08:00
• 9de19ef89f add more rangeify pm tests more_rangeify_pm_tests geohot 2025-10-07 17:20:39 +08:00
• 403fdfcfd4 check spec in test, cleanup vectorize render (#12484) George HotzandGitHub 2025-10-07 17:05:50 +08:00
• f129d75ee5 fix on cpu geohot 2025-10-07 16:43:52 +08:00
• 22674798df assert correctness in test_permuted_assignment [pr] (#12483) qazalandGitHub 2025-10-07 11:42:22 +03:00
• 7e4cf39929 check spec in test, cleanup vectorize render spec_ren geohot 2025-10-07 16:36:24 +08:00
• 75ce11593c test_reshape_match should match (#12479) George HotzandGitHub 2025-10-07 16:07:21 +08:00
• 51f3a5cbb4 gpu geohot 2025-10-07 16:00:10 +08:00
• 1d7a8b33c1 gemm works on pc geohot 2025-10-07 15:52:00 +08:00
• fe774a4319 more skip WINO on benchmark (#12482) chenyuandGitHub 2025-10-07 15:43:51 +08:00
• 8ad5f9e74f skip slow benchmarks (#12481) chenyuandGitHub 2025-10-07 15:28:56 +08:00
• ea7672931f fix test_matmul_relu_cat (#12478) George HotzandGitHub 2025-10-07 14:32:23 +08:00
• 2650f15c4b fix test_matmul_relu_cat fix_rc geohot 2025-10-07 14:15:31 +08:00
• 3fae886aa9 Merge branch 'master' into clone_tg George HotzandGitHub 2025-10-07 14:02:36 +08:00
• 514d2a0774 merge tagless reshapes (#12474) George HotzandGitHub 2025-10-07 13:57:58 +08:00
• 7b48f3cc45 failed test case repro for openpilot model (#12475) chenyuandGitHub 2025-10-07 13:46:43 +08:00
• 3f44ef699f Merge branch 'master' into clone_tg George HotzandGitHub 2025-10-07 13:08:35 +08:00
• fa23f37e33 clone thunderkittens in uops geohot 2025-10-07 13:06:58 +08:00
• 284db26a12 cleanup geohot 2025-10-07 12:13:01 +08:00
• 0a0cb0b9e8 merge tagless reshapes geohot 2025-10-07 12:11:11 +08:00
• a5484b767e remove skipping cast in simplify_valid [pr] (#12472) chenyuandGitHub 2025-10-07 12:10:04 +08:00
• b4509fba31 thundermittens (#12471) George HotzandGitHub 2025-10-07 11:47:39 +08:00
• 0f25b4b289 move frontend dir to nn [pr] (#12470) George HotzandGitHub 2025-10-07 10:42:22 +08:00
• f664bcc8bd use recursive_property in UOp tracing (#12469) qazalandGitHub 2025-10-06 21:10:52 +03:00
• e4ec4d2c51 remove noops add_local_buffer geohot 2025-10-06 23:21:48 +08:00
• 1d0b114a7b flip that geohot 2025-10-06 19:25:05 +08:00
• 51301c3b22 no locals geohot 2025-10-06 19:18:26 +08:00
• 17644fc304 gate pipeline geohot 2025-10-06 19:02:21 +08:00
• bf59379741 pipeline works on tc geohot 2025-10-06 18:59:51 +08:00
• 1af05dae77 fix rangeify in compile4.py (#12467) qazalandGitHub 2025-10-06 13:37:46 +03:00
• 97f122b591 tensor core works geohot 2025-10-06 18:34:03 +08:00
• afe31cc92a it works geohot 2025-10-06 17:54:05 +08:00
• 76e8a3250c rangeify: late zero folding (#12464) qazalandGitHub 2025-10-06 12:52:33 +03:00
• 3444e414f6 Merge branch 'master' into add_local_buffer George HotzandGitHub 2025-10-06 16:17:08 +08:00
• 0c015a24fe use recursive_property to prevent RecursionError (#12465) George HotzandGitHub 2025-10-06 15:59:18 +08:00
• a1881b0c17 update test_chicken (#12466) chenyuandGitHub 2025-10-06 15:58:44 +08:00
• 39d8459ff2 comments geohot 2025-10-06 14:13:31 +08:00
• fdc0489e18 pipelining wip geohot 2025-10-06 12:31:20 +08:00
• 1b1978b9c0 early copy fixup (#12463) qazalandGitHub 2025-10-06 06:38:29 +03:00
• c1e85f699c multi test case for sharded ring allreduce (#12462) chenyuandGitHub 2025-10-06 11:18:24 +08:00
• 1823a5043f don't check MAX_BUFFER_SIZE on NULL (#12461) chenyuandGitHub 2025-10-06 10:09:29 +08:00
• 46e8ea15c1 split pm_substitute_recurse (#12460) George HotzandGitHub 2025-10-06 09:35:50 +08:00
• 061a17ee4d split pm_substitute_recurse split_sub geohot 2025-10-06 09:22:01 +08:00
• df1b379a36 Merge branch 'master' into add_local_buffer George HotzandGitHub 2025-10-06 08:58:46 +08:00
• 5a9acee9b7 fixes ttomsa 2025-10-05 22:07:09 +01:00
• 1ffff0a801 draft ttomsa 2025-10-05 20:22:50 +01:00
• 1216fff781 remote: raise runtimeerror in checkz (#12453) nimlgenandGitHub 2025-10-05 21:22:53 +08:00