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

• c64e9591e3 replace gates in uopgraph [run_process_replay] (#5632) qazalandGitHub 2024-07-22 20:56:55 +08:00
• dc21e63bd2 test: put conv in one reduce (#4441) 3 people 2024-07-22 02:16:13 -07:00
• 386fb5e7f8 folding without UNMUL (#5628) George HotzandGitHub 2024-07-21 20:14:44 -07:00
• 583829ab44 helpers: remove duplicate data64 helpers in amd/nv (#5627) Vyacheslav PachkovandGitHub 2024-07-22 02:50:59 +03:00
• 6c6d74d922 parallel mcts (#5626) George HotzandGitHub 2024-07-21 14:53:23 -07:00
• c56c9c7519 move ufix inside UOp [run_process_replay] (#5621) chenyuandGitHub 2024-07-21 17:30:37 -04:00
• ef179087a4 mcts exit condition wasn't right, also use it with BEAM>=100 (#5619) George HotzandGitHub 2024-07-21 10:16:47 -07:00
• a823759dc5 simpler pattern matcher rules [run_process_replay] (#5620) chenyuandGitHub 2024-07-21 04:05:01 -04:00
• 0f67ef4674 mcts graph and dedup support (#5618) George HotzandGitHub 2024-07-20 23:29:14 -07:00
• 7f5282b2f5 tests if the linearizer is generating dumb code (#5611) George HotzandGitHub 2024-07-20 20:36:32 -07:00
• eddc5bcfd7 MCTS tweaks (#5616) chenyuandGitHub 2024-07-20 22:45:59 -04:00
• b399ccd6ef BEAM bugfix, kernels dedup now (#5617) George HotzandGitHub 2024-07-20 19:43:50 -07:00
• 92e7e65712 one more test case for symbolic mod mul (#5615) chenyuandGitHub 2024-07-20 17:23:06 -04:00
• d71308ed68 copy mlperf 4.0 to mlperf 4.1 (#5614) chenyuandGitHub 2024-07-20 16:12:00 -04:00
• 1113e47f96 print best in MCTS + light up the winner in hcopt geohot 2024-07-20 09:39:36 -07:00
• 0de5812032 hcq move map to allocator (#5610) nimlgenandGitHub 2024-07-20 19:02:45 +03:00
• ac99ecd94e use statistics.median for timing (#5606) George HotzandGitHub 2024-07-20 08:37:32 -07:00
• 3ab5fe4e1b test argmax multireduce failure (#5609) qazalandGitHub 2024-07-20 21:33:03 +08:00
• a96b5e3abb small input_st reorder (#5608) qazalandGitHub 2024-07-20 20:50:34 +08:00
• 646bdc1c0e elf loader touchups (#5607) nimlgenandGitHub 2024-07-20 12:30:18 +03:00
• 7ca2c48b64 hcq simpler _gpu2cpu_time (#5605) nimlgenandGitHub 2024-07-20 11:10:25 +03:00
• 32b0c07d5a docs: fix synchronization example in hcq (#5604) nimlgenandGitHub 2024-07-20 10:52:06 +03:00
• 06e336bccb mcts search (#5598) George HotzandGitHub 2024-07-19 21:38:39 -07:00
• b991097d41 move UPat and PatternMatcher from uopgraph.py to uops.py (#5597) chenyuandGitHub 2024-07-19 19:28:24 -04:00
• 72da3fe7e6 added clip vision model (#5595) 3 people 2024-07-19 18:35:51 -04:00
• a1af5a79ad remove obsolete code (#5596) P4ssengerandGitHub 2024-07-19 19:12:03 -03:00
• a02998472b fix no locals behavior (#5593) George HotzandGitHub 2024-07-19 14:35:09 -07:00
• 2e617ca59e lowerer img index (#5592) George HotzandGitHub 2024-07-19 14:22:02 -07:00
• 3acd8559f4 doc: variable names in abstractions2.py (#5591) chenyuandGitHub 2024-07-19 17:06:15 -04:00
• 00c01f6f4d correct IDIV dtype check error msg (#5589) chenyuandGitHub 2024-07-19 16:36:47 -04:00
• b1782e3fef hcq refactor signal into class (#5575) nimlgenandGitHub 2024-07-19 23:23:05 +03:00
• 2dc100c565 fix typo in runtime overview docs (#5588) Francis LataandGitHub 2024-07-19 15:00:15 -04:00
• d0ab20a5e5 careful memory counting (with tests to specify behavior) (#5587) George HotzandGitHub 2024-07-19 11:37:34 -07:00
• 37dd233650 always reverse global dim (#5586) chenyuandGitHub 2024-07-19 13:58:05 -04:00
• 10be05aae5 push contract through cast to fix test_float2_acc (try 2) (#5585) George HotzandGitHub 2024-07-19 10:34:43 -07:00
• 51892c8fac Revert "push contract through cast to fix test_float2_acc (#5581)" (#5583) George HotzandGitHub 2024-07-19 09:44:30 -07:00
• 6bade4d419 save the uops in their own file (#5582) George HotzandGitHub 2024-07-19 09:30:37 -07:00
• ddda9420be push contract through cast to fix test_float2_acc (#5581) George HotzandGitHub 2024-07-19 09:30:26 -07:00
• 3f590c3b31 some limit_dims to limit global merging (#5489) chenyuandGitHub 2024-07-19 12:17:46 -04:00
• e04704faff put acc first again (#5580) George HotzandGitHub 2024-07-19 08:55:19 -07:00
• fc5b9f8dc9 Kernel.required_optimizations and Kernel.hand_coded_optimizations returns self (#5576) chenyuandGitHub 2024-07-19 10:55:14 -04:00
• da34e1f617 scheduler refactors from the fuse_index branch (#5579) qazalandGitHub 2024-07-19 21:23:31 +08:00
• ecf88bb775 move assign_targets assignment (#5578) qazalandGitHub 2024-07-19 20:29:50 +08:00
• 0ad87021e2 move acc to end (#5568) George HotzandGitHub 2024-07-19 03:06:52 -07:00
• 2de82b8a5d remove get_lazyop_info (#5570) George HotzandGitHub 2024-07-19 03:05:33 -07:00
• 9d7edc9269 hcq rename HCQCompat -> HCQ (#5577) nimlgenandGitHub 2024-07-19 11:34:17 +03:00
• 2b2f8ad18c failed example of float2 acc no long applies (#5573) chenyuandGitHub 2024-07-19 02:40:04 -04:00
• efccb1c3ba swap global for size 3 too (#5567) chenyuandGitHub 2024-07-18 23:31:15 -04:00
• abe29a05b0 swap first and last global in hcopt / hc tc path (#5566) chenyuandGitHub 2024-07-18 18:54:44 -04:00
• 946da97820 swap action (#5565) George HotzandGitHub 2024-07-18 15:19:40 -07:00
• e7a057c20f retire replay_schedule (#5563) qazalandGitHub 2024-07-19 04:07:02 +08:00
• 50aba32ea8 hotfix: don't assert process replay in master. (#5562) qazalandGitHub 2024-07-19 03:05:00 +08:00
• 223d9283ee fix float4 acc by moving contracts (#5559) George HotzandGitHub 2024-07-18 11:30:16 -07:00
• c41cd55556 remove vectorized alu in expander [run_process_replay] (#5561) George HotzandGitHub 2024-07-18 11:27:40 -07:00
• c951bc99af fix abstracions2 printout (#5557) kormannandGitHub 2024-07-18 20:21:45 +02:00
• a7fec05acc fix broken store rule [run_process_replay] (#5558) George HotzandGitHub 2024-07-18 11:07:34 -07:00
• f5af98c450 failed test case that DEFINE_ACC no long uses float4 (#5555) chenyuandGitHub 2024-07-18 13:55:59 -04:00
• 923e0fe0b8 fix half4 folding (#5556) George HotzandGitHub 2024-07-18 10:47:39 -07:00
• 12e6771209 failed test case for unrolled half4 (#5552) chenyuandGitHub 2024-07-18 13:05:52 -04:00
• d1a7279605 indexing fold with casted bool (#5551) George HotzandGitHub 2024-07-18 10:02:29 -07:00
• fdfc0015a7 [run_process_replay] for opencl/openpilot (#5009) qazalandGitHub 2024-07-19 00:42:33 +08:00
• 2c4add6844 pretty print lazy op per default (#5505) kormannandGitHub 2024-07-18 18:34:08 +02:00
• c30092e56d amd remove useless barrier (#5550) nimlgenandGitHub 2024-07-18 18:05:33 +03:00
• 4e9d2b1615 nv memory_barrier command (#5548) nimlgenandGitHub 2024-07-18 16:23:11 +03:00
• 6d7cd34250 more save_schedule tooling (#5547) qazalandGitHub 2024-07-18 20:59:53 +08:00
• 0ad1672d5f fuse indexing (LazyOp creation) (#5506) qazalandGitHub 2024-07-18 19:09:17 +08:00
• 6ccb2390c3 feat: update_benchmark_staging (#5529) wozeparrotandGitHub 2024-07-18 03:40:57 +00:00
• e569c927cf remove Kernel.shape_offsets [run_process_replay] (#5544) chenyuandGitHub 2024-07-17 23:16:47 -04:00
• fa7e734b49 MetaOps.KERNEL (#5543) George HotzandGitHub 2024-07-17 19:41:23 -07:00
• d3b098299d add failing regression test for image (#5540) George HotzandGitHub 2024-07-17 17:27:18 -07:00
• 218e157f00 benchmark on update_benchmark_staging (#5541) wozeparrotandGitHub 2024-07-18 00:11:52 +00:00
• 8845a5dbfd feat: begin immediate (#5539) wozeparrotandGitHub 2024-07-17 23:11:21 +00:00
• a6e70f8a71 clean up expand function [run_process_replay] (#5538) George HotzandGitHub 2024-07-17 15:02:00 -07:00
• 61ee02e93d start multireduce lowerer work (var/std) (#5537) 3 people 2024-07-18 04:43:46 +08:00
• 67ea4af01f depth first recurse_reduceops (#5536) qazalandGitHub 2024-07-18 04:27:53 +08:00
• c4eb30a04c test/test_linearizer_failures: add a new beautiful_mnist one (#5531) 3 people 2024-07-17 13:27:04 -07:00
• 0259d76183 use Context only in replaying Kernel [run_process_replay] (#5535) qazalandGitHub 2024-07-18 03:46:14 +08:00
• 1a68854766 PatternMatcher add (#5532) George HotzandGitHub 2024-07-17 12:44:42 -07:00
• d3c137d478 utility for computing reduceop output_shape (#5534) qazalandGitHub 2024-07-18 03:40:07 +08:00
• 0a7872a62f use exec_alu in uops flop counting (#5511) qazalandGitHub 2024-07-18 03:39:27 +08:00
• a7706e05f9 option to [skip_process_replay] (#5533) qazalandGitHub 2024-07-18 03:30:46 +08:00
• 4193095f67 fix handcode_opt.py with DEBUG=2 (#5530) chenyuandGitHub 2024-07-17 14:50:47 -04:00
• 466555cd17 touchup Tensor.interpolate (#5525) chenyuandGitHub 2024-07-17 13:35:57 -04:00
• 1242b302fa expand UOps with rewrite rules (#5501) George HotzandGitHub 2024-07-17 10:17:50 -07:00
• 158221b36b expand tests from uop_expander [run_process_replay] (#5524) George HotzandGitHub 2024-07-17 09:22:36 -07:00
• 42c25cc961 fix fixup_ast (#5523) George HotzandGitHub 2024-07-17 08:52:21 -07:00
• fbe0233be3 infra for multi reduce asts (#5522) qazalandGitHub 2024-07-17 22:23:46 +08:00
• dcd462860f elf loader (#5508) nimlgenandGitHub 2024-07-17 17:09:34 +03:00
• 661da32aff nv do not map regions twice (#5521) nimlgenandGitHub 2024-07-17 11:20:02 +03:00
• 2d53abb04a test/external/fuzz_linearizer: fix for new AST changes (#5519) Francis LamandGitHub 2024-07-16 21:08:07 -07:00
• 85d4ca7caa FID Inception Model (#5516) Tobias FischerandGitHub 2024-07-16 23:12:03 -04:00
• 4ad83d032e remove Kernel.lazyops [run_process_replay] (#5517) chenyuandGitHub 2024-07-16 19:47:42 -04:00
• 1c1d6d3a4a feat: show caller when tracemeta >= 2 (#5514) wozeparrotandGitHub 2024-07-16 22:06:02 +00:00
• 5aad043522 cleanup fixup_ast local shape long line [run_process_replay] (#5513) chenyuandGitHub 2024-07-16 17:29:38 -04:00
• 6e405b0a2b add 0d tensor to trunc/floor/ceil/round tests (#5512) chenyuandGitHub 2024-07-16 16:48:25 -04:00
• 0afcbfae84 docs: add Tensor.interpolate to doc page (#5510) chenyuandGitHub 2024-07-16 14:17:19 -04:00
• 87a2ef2bc2 Add Interpolate Function (#5482) 3 people 2024-07-16 12:44:01 -04:00
• 203161c75d refactor VECTORIZE/GEP rules (#5507) gswanggandGitHub 2024-07-16 09:41:23 -07:00
• 173064c69c (re)start multireduce in codegen/* (#5391) qazalandGitHub 2024-07-16 19:20:48 +08:00
• 07ff4b7d24 test_failure_33 ast that has UOps.UNMUL after linearize (#5504) chenyuandGitHub 2024-07-15 22:54:23 -04:00