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

• a5ede535ef NOp field name [run_process_replay] (#5742) kormannandGitHub 2024-07-27 00:45:59 +02:00
• 0d7d4dd731 UOp._min_max for MUL and MOD (#5741) chenyuandGitHub 2024-07-26 18:38:10 -04:00
• c50e374bb6 multiple locals + get_kernel_modifier + fix valid (#5739) George HotzandGitHub 2024-07-26 15:10:10 -07:00
• f6c0e17a2c optimize symbolic-related updates in graphs (#5727) nimlgenandGitHub 2024-07-27 00:57:59 +03:00
• dc7483ee6f UOp simple div folding (#5740) chenyuandGitHub 2024-07-26 17:14:32 -04:00
• 671259417f reuse UOp __repr__ for NOp (#5738) chenyuandGitHub 2024-07-26 16:59:55 -04:00
• b0c1dba299 named UOp class "NOP" [run_process_replay] (#5728) 3 people 2024-07-26 22:25:53 +02:00
• 4df46eac67 clean up tensor cores [run_process_replay] (#5736) George HotzandGitHub 2024-07-26 13:21:23 -07:00
• 94d578396f separate process replay main loop (#5734) qazalandGitHub 2024-07-27 02:43:08 +08:00
• 9838c1a6ff update import style in runtime (#5735) chenyuandGitHub 2024-07-26 14:00:23 -04:00
• a4e9ebc68a update test_uop_symbolic (#5733) chenyuandGitHub 2024-07-26 13:46:09 -04:00
• 5c688560bc move CUDA/HIP compilers to their own files [run_process_replay] (#5732) George HotzandGitHub 2024-07-26 10:00:15 -07:00
• 2cc55a3095 UOp simple mul add div fold (#5726) chenyuandGitHub 2024-07-25 22:00:30 -04:00
• 78f75aa80d remove redundant symbolic mod rule [run_process_replay] (#5725) chenyuandGitHub 2024-07-25 21:21:02 -04:00
• 5521b6d437 UOp simple mul-add-lt fold (#5721) chenyuandGitHub 2024-07-25 20:49:38 -04:00
• 1b53207b4f revert isolated dags scheduling (#5724) qazalandGitHub 2024-07-26 07:45:12 +08:00
• 845b0d1c9d UOp more generic div folding (#5722) chenyuandGitHub 2024-07-25 17:49:14 -04:00
• fb8148077e hcq do not update the same signal (#5719) nimlgenandGitHub 2024-07-26 00:24:45 +03:00
• 6ec9ea9ddd hcq update_exec with optional params (#5708) nimlgenandGitHub 2024-07-26 00:04:57 +03:00
• 8b34ee2f52 remove global_size and local_size from Kernel class [run_process_replay] (#5720) George HotzandGitHub 2024-07-25 13:55:08 -07:00
• 142b7fb22f faster beam [run_process_replay] (#5718) George HotzandGitHub 2024-07-25 11:58:41 -07:00
• eff7c5fd2c halve kernel counts in metal Fuzz Test linearizer (#5716) chenyuandGitHub 2024-07-25 14:35:11 -04:00
• e877ed9688 cleaner uop expand [run_process_replay] (#5715) George HotzandGitHub 2024-07-25 11:29:53 -07:00
• a82815262c more test_pattern_matcher fixups (#5714) chenyuandGitHub 2024-07-25 14:12:21 -04:00
• b8b5411845 move Function to Developer section of docs geohot 2024-07-25 11:04:03 -07:00
• f02124ffa0 rename to realize_reduceop (#5713) qazalandGitHub 2024-07-26 01:57:33 +08:00
• 05e02ddfb3 fixup test_pattern_matcher (#5712) chenyuandGitHub 2024-07-25 13:48:52 -04:00
• 9ceb3a3d1f beautiful_mnist -4.3% kernels (#5709) qazalandGitHub 2024-07-26 01:30:49 +08:00
• 92eefab4b0 method alu (#5711) 3 people 2024-07-25 19:25:38 +02:00
• 76877df518 map groupable children (#5710) qazalandGitHub 2024-07-26 00:27:48 +08:00
• 1e2eac755d Fix repr upat (#5705) kormannandGitHub 2024-07-25 18:05:48 +02:00
• 1c992de257 hotfix: compare_schedule defaults to false (#5707) qazalandGitHub 2024-07-25 22:08:28 +08:00
• 489cda827a more scheduler process replay tooling (#5706) qazalandGitHub 2024-07-25 20:47:18 +08:00
• 4e070a2c89 start work on indexing fusion (#5590) qazalandGitHub 2024-07-25 18:23:38 +08:00
• 08f47d7dc3 more info on failure 41 (#5704) nimlgenandGitHub 2024-07-25 12:14:28 +03:00
• 69d4f474d8 amd resnet pf (#5703) nimlgenandGitHub 2024-07-25 11:21:22 +03:00
• 1038482a66 enable hip tc (#5702) nimlgenandGitHub 2024-07-25 11:12:11 +03:00
• 5b38ff8679 shorter llvm and ptx rendering [run_process_replay] (#5686) qazalandGitHub 2024-07-25 15:42:25 +08:00
• 46e1151c02 UOp more generic mul -> mod folding (#5698) chenyuandGitHub 2024-07-24 21:41:25 -04:00
• 66a9c372af UOp mod reduction (#5697) chenyuandGitHub 2024-07-24 20:36:00 -04:00
• 489a5b99a5 hotfix: triton_nv_matmul touchups geohot 2024-07-24 23:24:23 +00:00
• 8648fb2636 UOp vmin/vmax on ADD (#5689) chenyuandGitHub 2024-07-24 19:09:42 -04:00
• e2e70bd90b bring unbind back in Varaible const (#5687) qazalandGitHub 2024-07-25 06:37:00 +08:00
• b026312a31 nv ptx print log (#5691) nimlgenandGitHub 2024-07-24 21:40:58 +03:00
• bf24be4c8c triton gets 163 TFLOPS on 4090 geohot 2024-07-24 18:31:32 +00:00
• 85710e86cb UOps div folding (#5690) chenyuandGitHub 2024-07-24 14:21:44 -04:00
• fb1b51811b unify UOp min/max default [run_process_replay] (#5688) chenyuandGitHub 2024-07-24 13:05:26 -04:00
• 33d44f00ae first fold, then expand (#5673) 3 people 2024-07-24 09:43:09 -07:00
• b7b4c7844f shorter BufferOps.LOAD creation (#5685) qazalandGitHub 2024-07-24 23:53:07 +08:00
• 365e7afd4d make fusion deterministic (#5684) qazalandGitHub 2024-07-24 23:37:31 +08:00
• 2ea54176e2 docs: add more info on HCQProgram (#5683) nimlgenandGitHub 2024-07-24 17:20:18 +03:00
• baface413a nv better nvdisasm fail message (#5682) nimlgenandGitHub 2024-07-24 16:19:26 +03:00
• 37347528bf shorter BufferOps.CONST creation (#5681) qazalandGitHub 2024-07-24 19:33:04 +08:00
• 6dcdff3bfd share fusion behavior for r3 kernels (#5680) qazalandGitHub 2024-07-24 19:07:10 +08:00
• 3ffb1059a0 scheduling infra for isolated dags (#5679) qazalandGitHub 2024-07-24 17:14:26 +08:00
• e6e2d86fcf replace RANGE max fold with generic max fold (#5676) chenyuandGitHub 2024-07-24 03:15:39 -04:00
• a7a77dfd83 UOp mul lt fold (#5677) chenyuandGitHub 2024-07-24 02:49:25 -04:00
• 67b036bdfd generic UOp max folding (#5675) chenyuandGitHub 2024-07-24 01:30:32 -04:00
• d1d81b359f UOp compute min and max in one call [run_process_replay] (#5674) chenyuandGitHub 2024-07-24 00:51:23 -04:00
• 4e85761d40 UOp mod folding (#5668) chenyuandGitHub 2024-07-24 00:10:47 -04:00
• f46ba37f8f increase amount of float2/float4 folding (#5672) George HotzandGitHub 2024-07-23 20:52:56 -07:00
• 053550c3f3 remove MERGE opt, cleanup wmma upcast (#5669) George HotzandGitHub 2024-07-23 20:43:42 -07:00
• 918eebb1b1 simple TC change [run_process_replay] (#5671) George HotzandGitHub 2024-07-23 20:11:31 -07:00
• 3060e0be4f add vmin vmax of SPECIAL (#5670) chenyuandGitHub 2024-07-23 22:55:54 -04:00
• 5de43e7073 Revert "this fails too" geohot 2024-07-24 02:20:53 +00:00
• df20c4602a this fails too geohot 2024-07-24 02:19:55 +00:00
• fa14f7b4fd switch contract arg to match expand arg [run_process_replay] (#5667) George HotzandGitHub 2024-07-23 18:08:33 -07:00
• ea99efe815 remove UOps lt pattern of booleans (#5666) chenyuandGitHub 2024-07-23 20:11:21 -04:00
• e196640d71 more generic lt folding (#5665) chenyuandGitHub 2024-07-23 19:50:59 -04:00
• 7c8fe0fe47 skip interpolate tests for PYTHON=1 (#5664) chenyuandGitHub 2024-07-23 18:47:15 -04:00
• a85493bdbe multiaxis contract test geohot 2024-07-23 15:09:15 -07:00
• e3f00ac77d Fix cuda tc emu test (#5663) George HotzandGitHub 2024-07-23 15:04:25 -07:00
• c34f9db0f7 remove ptx PTXRenderer.gdim gid lid [run_process_replay] (#5662) chenyuandGitHub 2024-07-23 17:33:20 -04:00
• 16c27ae400 update UOp.SPECIAL arg spec [run_process_replay] (#5661) chenyuandGitHub 2024-07-23 16:58:12 -04:00
• 4d47968580 fix acc folding for NV tensor cores (#5658) George HotzandGitHub 2024-07-23 13:03:02 -07:00
• 01fe00e055 skip test_failure_39 in CI (#5660) chenyuandGitHub 2024-07-23 14:47:05 -04:00
• fdc72ba102 reorder UOps.DEFINE_VAR in runtime [run_process_replay] (#5659) chenyuandGitHub 2024-07-23 14:32:10 -04:00
• 199b3bf02b simple UOp lt/ge folding (#5657) chenyuandGitHub 2024-07-23 14:11:05 -04:00
• b0fc5a4c6f start scheduler process replay (#5656) qazalandGitHub 2024-07-24 01:02:51 +08:00
• e210c87b4a uop mod-mod simplification (#5650) chenyuandGitHub 2024-07-23 12:33:55 -04:00
• 1384f08cd4 hcq profile tests (#5654) nimlgenandGitHub 2024-07-23 18:40:33 +03:00
• 5f394fc9c6 more work toward non-blocking process replay (#5653) qazalandGitHub 2024-07-23 19:26:31 +08:00
• a93982ef42 hcq move out program call to base class (#5638) nimlgenandGitHub 2024-07-23 14:25:38 +03:00
• 7cb67e6fb2 merge gated stores spec (#5652) qazalandGitHub 2024-07-23 18:53:27 +08:00
• 4dcca0a6d4 amd tiny cleanups (#5651) nimlgenandGitHub 2024-07-23 13:06:23 +03:00
• 7c4b177e3a add tests for uops stats (#5649) George HotzandGitHub 2024-07-22 21:57:03 -07:00
• 4f83da626e uop symbolic simple mul mod (#5648) chenyuandGitHub 2024-07-22 23:17:41 -04:00
• 4042bc2399 hotfix: put that space back in DEBUG=2 geohot 2024-07-22 20:11:15 -07:00
• 2a436fa5c6 memory estimate of cache also (#5646) George HotzandGitHub 2024-07-22 19:56:36 -07:00
• efc7bf37a2 reuse UOp.sparents in UOps.vars [run_process_replay] (#5647) chenyuandGitHub 2024-07-22 22:50:19 -04:00
• f2d2afdaa4 dumb linearizer example that max is not simplified (#5644) chenyuandGitHub 2024-07-22 18:37:26 -04:00
• fe17ea5c88 typo in ops_amd invalidate_caches (#5643) chenyuandGitHub 2024-07-22 18:37:11 -04:00
• ed2ee52b8b fix arange 4096 with more folding rules (#5641) George HotzandGitHub 2024-07-22 14:21:11 -07:00
• 24505199fb UOp.const(x.dtype, y) -> x.const(y) [run_process_replay] (#5642) chenyuandGitHub 2024-07-22 17:09:40 -04:00
• 97b116bb1d UOp mul div simplification (#5637) chenyuandGitHub 2024-07-22 16:14:12 -04:00
• ee633c1988 hcq move out synchronize to base class (#5634) nimlgenandGitHub 2024-07-22 20:36:04 +03:00
• 26fc4610a0 amd more accurate cache managment (#5631) nimlgenandGitHub 2024-07-22 19:07:01 +03:00
• fe6f9b2048 more actionable verify_lazyop assert (#5635) qazalandGitHub 2024-07-23 00:06:11 +08:00
• edc58e6b6e hcq: remove duplicate allocation of kernel args by abstracting (#5633) Vyacheslav PachkovandGitHub 2024-07-22 18:29:41 +03:00
• 08a9c0ae5e hcq cache invalidation for beam (#5630) nimlgenandGitHub 2024-07-22 18:13:17 +03:00