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

• 52301fe68e move Buffer refcount increment out of schedule.py (#9564) qazalandGitHub 2025-03-25 12:08:27 +08:00
• 0d10c7ae2f working on kernel 15 geohot 2025-03-25 12:00:16 +08:00
• 262f5a2bd3 hotfix: replace link in viz/readme (#9568) qazalandGitHub 2025-03-25 10:24:49 +08:00
• 6427272bf6 minor update to rand [pr] (#9566) chenyuandGitHub 2025-03-24 18:49:50 -04:00
• b0e070e737 remove MOCKGPU workaround in rand (#9565) chenyuandGitHub 2025-03-24 17:49:45 -04:00
• d7c754ce49 failing test for UOp buffer ref count (#9563) qazalandGitHub 2025-03-25 00:10:48 +08:00
• f90001e1a6 amd llvm render (no_comgr prereq) (#9543) 3 people 2025-03-24 22:50:51 +08:00
• 4f5e03bd60 better fix inplace detach (#9557) Priyank PatelandGitHub 2025-03-24 07:50:28 -07:00
• 1c40873962 show buffer info in memory viz (#9562) qazalandGitHub 2025-03-24 22:12:30 +08:00
• efaee75656 start viz of memory usage (#9561) qazalandGitHub 2025-03-24 19:05:35 +08:00
• 1cfe6d02fe refactor uop_to_json to return a dict [pr] (#9560) qazalandGitHub 2025-03-24 16:38:17 +08:00
• 3cab6a3d4a 3x3 geohot 2025-03-24 16:36:53 +08:00
• 22a56cbaea something for 8 geohot 2025-03-24 16:16:59 +08:00
• afd61730b4 kernel 5 geohot 2025-03-24 16:10:08 +08:00
• 536556434b padding ish geohot 2025-03-24 15:57:03 +08:00
• 52bff5f39d more geohot 2025-03-24 15:33:48 +08:00
• 64d0f14d3d broken geohot 2025-03-24 15:21:43 +08:00
• edf9e1bf8d am: move out soc21 to a sep module (#9551) nimlgenandGitHub 2025-03-24 14:17:42 +07:00
• 1b61cc6ec3 unaligned geohot 2025-03-24 15:13:57 +08:00
• 6f792e8045 vmemu geohot 2025-03-24 15:02:40 +08:00
• b1f8018bf4 unaligned load geohot 2025-03-24 14:54:11 +08:00
• 2eb9241329 better conv geohot 2025-03-24 13:07:14 +08:00
• 554a490751 Merge branch 'master' into dsp_search George HotzandGitHub 2025-03-24 12:29:22 +08:00
• 74d98eafb8 add onnx frontend stub [pr] (#9558) George HotzandGitHub 2025-03-24 12:24:34 +08:00
• de7d6cec3a hotfix: DEBUG 5 prints the ast geohot 2025-03-24 11:43:11 +08:00
• 651c678edf work geohot 2025-03-24 09:49:53 +08:00
• ba41076e94 update embedding test to not use dtypes.long [pr] (#9556) chenyuandGitHub 2025-03-23 21:33:38 -04:00
• c965f4c20b update bert config (#9555) chenyuandGitHub 2025-03-23 16:14:41 -04:00
• d734e24c01 minor WEBGPU_PATH cleanup [pr] (#9552) chenyuandGitHub 2025-03-23 09:10:02 -04:00
• 7ce7fe0574 Refactor webgpu_dawn lib finding (#9547) Ahmed HarmoucheandGitHub 2025-03-23 13:23:29 +01:00
• c631c72f22 HCQ: Increment timeline signal before submitting (#9550) uuuvnandGitHub 2025-03-23 16:30:38 +05:00
• d5667419af am: move out pte creation logic (#9548) nimlgenandGitHub 2025-03-23 18:29:10 +07:00
• 3274bd2d81 output geohot 2025-03-23 15:13:00 +08:00
• 309afa20b7 add Tensor.max_unpool2d (#9518) geohotstanandGitHub 2025-03-23 00:11:33 +08:00
• 30f4d64148 rules geohot 2025-03-22 19:17:16 +08:00
• 2634975d5a 5 and 8 geohot 2025-03-22 19:14:04 +08:00
• fd73ec2b1b knum geohot 2025-03-22 18:59:54 +08:00
• e1d2bec4a4 opt geohot 2025-03-22 18:52:56 +08:00
• 1b4e9f5e91 more opt rules geohot 2025-03-22 18:07:31 +08:00
• 25c023bcbe more geohot 2025-03-22 17:49:34 +08:00
• 07abf9e6bc multi_add_int32 geohot 2025-03-22 17:33:56 +08:00
• 26b02a037c fix 33 geohot 2025-03-22 17:17:47 +08:00
• 5089a601c6 name it geohot 2025-03-22 14:44:01 +08:00
• 6b49a63c48 linearizer workaround geohot 2025-03-22 14:18:02 +08:00
• bdd44d4255 Fix DSP transcendentals (#9542) quortusandGitHub 2025-03-22 04:08:18 +01:00
• dca95428a5 touch geohot 2025-03-22 11:05:36 +08:00
• eddafb84e5 Bugfix for TC=3 (#9464) Ignacio SicaandGitHub 2025-03-21 16:43:42 -07:00
• 6da78164f9 assert Kernel ast.op to be Ops.SINK [pr] (#9539) chenyuandGitHub 2025-03-21 18:09:44 -04:00
• c33679c47b increase size in test_multinomial_counterexample (#9540) chenyuandGitHub 2025-03-21 17:46:52 -04:00
• 1a1087e3a0 cleanups on losses and dataset tests (#9538) Francis LataandGitHub 2025-03-21 17:03:18 -04:00
• 8cbe4009fc RetinaNet losses (#9536) Francis LataandGitHub 2025-03-21 15:52:54 -04:00
• e6389184c5 update comment for retinanet dataloader implementations (#9534) 3 people 2025-03-21 15:07:45 -04:00
• ee3d313b34 Revert "update ruff to 0.11.2 (#9531)" (#9535) chenyuandGitHub 2025-03-21 14:52:25 -04:00
• b46b8ee15e add a flag to log when beam surpassed max limit [pr] (#9533) chenyuandGitHub 2025-03-21 13:37:02 -04:00
• eb95825eea RetinaNet dataloader (#9442) Francis LataandGitHub 2025-03-21 13:36:41 -04:00
• 58206fa8a9 add amd llvm compiler (#9519) 4 people 2025-03-21 23:13:27 +08:00
• d8d65e2747 update ruff to 0.11.2 (#9531) chenyuandGitHub 2025-03-21 10:32:59 -04:00
• 8a477ba4e1 knum 3 geohot 2025-03-21 20:36:18 +08:00
• 264dd91b8a 70 GFLOPS geohot 2025-03-21 20:31:14 +08:00
• bdf716b915 mul work geohot 2025-03-21 20:05:29 +08:00
• cf41c803d0 fast 13 geohot 2025-03-21 18:10:59 +08:00
• 3cf9224df5 a scale and b scale geohot 2025-03-21 18:07:53 +08:00
• af94addb3a ish geohot 2025-03-21 17:46:45 +08:00
• ee3ed73ed1 add reorder_view matcher to scheduler [pr] (#9528) qazalandGitHub 2025-03-21 17:46:20 +08:00
• dc1469a188 double reduce geohot 2025-03-21 17:33:48 +08:00
• 0416b0998d revert those geohot 2025-03-21 17:15:38 +08:00
• c715c25420 Merge branch 'master' into dsp_search George HotzandGitHub 2025-03-21 17:13:10 +08:00
• 8e555c586c switch quantization to unsigned/unsigned + add Ops.REDUCE (#9527) George HotzandGitHub 2025-03-21 17:02:37 +08:00
• f66b03f0a6 dsp ish geohot 2025-03-21 16:28:08 +08:00
• 2729a46ca6 don't do that geohot 2025-03-21 16:04:21 +08:00
• a35b0a88bf am: just rename and reorder ip init funcs (#9504) nimlgenandGitHub 2025-03-21 15:57:32 +08:00
• 8a131ab271 am: allow allocations as small as a page (#9523) nimlgenandGitHub 2025-03-21 15:53:32 +08:00
• dbb50e4a00 knum 4 geohot 2025-03-21 15:48:50 +08:00
• 71c7c455a6 quantize geohot 2025-03-21 14:55:29 +08:00
• ff3438be4e fast geohot 2025-03-21 13:04:18 +08:00
• bc5e23061b diasm geohot 2025-03-21 11:22:40 +08:00
• 5ce951fb34 l2 geohot 2025-03-21 11:14:12 +08:00
• 4a49d05a3f Merge branch 'master' into dsp_search George HotzandGitHub 2025-03-21 10:26:38 +08:00
• c3c85c64ee simpler geohot 2025-03-21 09:24:33 +08:00
• 3ad3ac4d1e Change dtypes.int to dtypes.ints (#9517) Sieds LyklesandGitHub 2025-03-20 22:24:26 +01:00
• b9fab9b914 pin ruff to 0.11.0 in CI (#9520) chenyuandGitHub 2025-03-20 13:12:50 -04:00
• 61c02ca634 cleanups geohot 2025-03-20 23:27:06 +08:00
• 325044bcaf okay that should actually prefetch geohot 2025-03-20 22:59:59 +08:00
• 91ac508878 prefetch geohot 2025-03-20 22:56:38 +08:00
• 2ed30f5366 correct flops geohot 2025-03-20 21:46:13 +08:00
• d0b9c7e7ca fast like nascar? geohot 2025-03-20 21:27:26 +08:00
• f6ed8f4a27 8 folds geohot 2025-03-20 21:20:46 +08:00
• 87718170d2 more generic geohot 2025-03-20 21:14:33 +08:00
• b67af4049c knum 20 geohot 2025-03-20 20:59:06 +08:00
• 16e425a4c0 work geohot 2025-03-20 20:24:21 +08:00
• c867a48ab4 custom geohot 2025-03-20 20:02:35 +08:00
• 2dc82c0604 should be fast geohot 2025-03-20 19:49:04 +08:00
• e7402e6643 KNUM=13 will be fast like roadrunner geohot 2025-03-20 18:45:53 +08:00
• e5ccd9e846 work geohot 2025-03-20 15:20:03 +08:00
• 624197f169 swizzle better geohot 2025-03-20 12:41:24 +08:00
• d42350a401 simple test geohot 2025-03-20 12:37:29 +08:00
• 3c5161b4cb add validation of the bounds of Ops.INDEX (#9503) 3 people 2025-03-20 12:15:55 +08:00
• 0b20f91ce7 remove move_mask from the devectorizer (#9511) qazalandGitHub 2025-03-20 11:53:12 +08:00
• 9302738263 hotfix: more consistent wgsl.py spacing + cleanups [pr] (#9515) qazalandGitHub 2025-03-20 11:07:15 +08:00
• 223feb2118 Merge branch 'master' into dsp_search George HotzandGitHub 2025-03-20 10:52:30 +08:00