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

• 3d6e84051b gpt make it better? geohot 2026-07-10 14:07:48 -07:00
• 601509b60e remove vec from x86 backend (glm) [pr] geohot 2026-07-10 20:56:14 +00:00
• 52e060fd7a hcq2: link cache (#16965) nimlgenandGitHub 2026-07-10 23:47:16 +03:00
• df50e0814c explicit error for unbound Variable in program (#16971) chenyuandGitHub 2026-07-10 16:23:59 -04:00
• 2fda6b3888 fix shape broadcast for symbolic (#16970) chenyuandGitHub 2026-07-10 15:13:01 -04:00
• 3519da4599 updates geohot 2026-07-10 11:53:16 -07:00
• 43aed9a827 Merge branch 'master' into no_vec_x86_4 George HotzandGitHub 2026-07-10 11:46:43 -07:00
• c6a2823935 remove vec from x86 backend (gpt) [pr] geohot 2026-07-10 11:37:36 -07:00
• 77823056d4 remove dead DSP code and pre_matcher (#16968) George HotzandGitHub 2026-07-10 11:27:50 -07:00
• 3964eee64f failing tests for Tensor(Variable) (#16967) chenyuandGitHub 2026-07-10 14:17:03 -04:00
• 1e55cef493 remove vec from llvm (#16966) George HotzandGitHub 2026-07-10 10:47:31 -07:00
• 2b7c298aaf hotfix: remove SPEC=2 timeout to match other jobs geohot 2026-07-10 09:35:48 -07:00
• bc3bf1988a asm gemm test cleanups try 2 (#16964) qazalandGitHub 2026-07-10 15:00:24 +08:00
• 34452efa22 remove dtypes.vec from cstyle (#16961) George HotzandGitHub 2026-07-09 23:11:11 -07:00
• 88c6f02abe remove late_buffer_view, try 2 (#16963) sirhcmandGitHub 2026-07-10 00:46:26 -04:00
• 5cc31e23e6 gptoss: scripts (#16962) wozeparrotandGitHub 2026-07-09 23:50:20 -04:00
• 414995d1f5 Revert "remove late buffer view" (#16960) sirhcmandGitHub 2026-07-09 21:57:42 -04:00
• 6e6977efe8 16.4 seconds cifar_30 geohot 2026-07-10 00:55:06 +00:00
• 8d546f55e9 dtypes.weakfloat (#16956) chenyuandGitHub 2026-07-09 19:32:21 -04:00
• d80c971ea2 param + cast dtype in arg (#16955) George HotzandGitHub 2026-07-09 16:32:01 -07:00
• 1b195e76ac 20 seconds geohot 2026-07-09 23:19:04 +00:00
• 78223d690a update SPECIAL spec (#16952) chenyuandGitHub 2026-07-09 19:05:34 -04:00
• 3856c334a8 cifar 30 seconds geohot 2026-07-09 22:15:55 +00:00
• 95681f17ee remove extra dtypes from tests (#16954) George HotzandGitHub 2026-07-09 14:59:56 -07:00
• 1df49a7bf5 dtype doesn't need to be explicit in most places now (glm) [PR] (#16951) George HotzandGitHub 2026-07-09 14:41:34 -07:00
• 851e5727d2 remove late buffer view (#16941) sirhcmandGitHub 2026-07-09 17:32:35 -04:00
• 93338df753 real dtypes.index (#16950) chenyuandGitHub 2026-07-09 16:28:39 -04:00
• 8d4c9d1058 make the upat compiler match the spec (glm) [PR] (#16948) George HotzandGitHub 2026-07-09 12:30:39 -07:00
• ba2c68b1ed start dtypes.index [PR] (#16949) chenyuandGitHub 2026-07-09 15:12:23 -04:00
• da435c719b tinyfs: use assign (#16947) sirhcmandGitHub 2026-07-09 14:52:42 -04:00
• 8b96b95af5 tighter pm_no_weakints [PR] (#16946) chenyuandGitHub 2026-07-09 14:31:16 -04:00
• 904b51a783 dtype fixups (grok) [PR] (#16940) George HotzandGitHub 2026-07-09 11:03:22 -07:00
• c9f6b2e42b hcq2: prereq for linker (#16945) nimlgenandGitHub 2026-07-09 17:21:21 +03:00
• 6dbc35b8e0 fix am_smi for non interactive shell (#16942) qazalandGitHub 2026-07-09 12:32:59 +08:00
• 2971343a60 fa: support D=64 and attn sink (#16935) wozeparrotandGitHub 2026-07-08 23:03:47 -04:00
• 00d01d978d add dtype production rule checked with SPEC=2 (#16939) George HotzandGitHub 2026-07-08 18:04:34 -07:00
• 7f8bbe5407 hlb_cifar free speed (#16938) George HotzandGitHub 2026-07-08 15:49:48 -07:00
• e2fc928d85 use get_idx and get_valid in coalese [PR] (#16937) chenyuandGitHub 2026-07-08 17:20:54 -04:00
• 8085bd57ec remove old stale metadata codes [pr] (#16936) chenyuandGitHub 2026-07-08 17:11:15 -04:00
• 52b84adc2a hcq2: tiny changes (#16934) nimlgenandGitHub 2026-07-08 23:17:28 +03:00
• fdffc6c0c8 remove dtypes base (#16931) George HotzandGitHub 2026-07-08 11:48:11 -07:00
• 63cb1369cb remove ImageDType (glm) (#16918) George HotzandGitHub 2026-07-08 11:27:52 -07:00
• 6358f939e7 param arg is always the shape (#16914) George HotzandGitHub 2026-07-08 10:51:56 -07:00
• 6e05cbdbcb remove the dtype.scalar calls everywhere [pr] (#16930) George HotzandGitHub 2026-07-08 10:10:03 -07:00
• e4d0d634d4 remove a bunch of dtype.count [PR] (#16929) George HotzandGitHub 2026-07-08 09:34:46 -07:00
• a951650865 lil cleanups from reduce and expand [PR] (#16928) George HotzandGitHub 2026-07-08 08:54:19 -07:00
• e0c70b6b82 mixin/op.py [PR] (#16926) chenyuandGitHub 2026-07-07 21:45:52 -04:00
• ab131c2086 gptoss: zero-1 optim (#16916) wozeparrotandGitHub 2026-07-07 21:34:47 -04:00
• 4f7d4e95d7 fix <= 2 case (#16923) 3 people 2026-07-07 21:17:58 -04:00
• c26468c5d0 keccak and hash to mixin [PR] (#16925) chenyuandGitHub 2026-07-07 21:12:46 -04:00
• c43a3fdebb bitcast to mixin [PR] (#16924) chenyuandGitHub 2026-07-07 20:03:29 -04:00
• 9f388d42b7 padded maxpool with -inf as done in pytorch (#16858) DivijandGitHub 2026-07-08 04:40:44 +05:30
• 4a51047146 remove unused pm_syntactic_sugar (#16920) George HotzandGitHub 2026-07-07 15:40:52 -07:00
• 82ec1ba572 NOOP index in mop_cleanup noop_index geohot 2026-07-07 14:33:28 -07:00
• 2aebb6f6c4 move mop cleanup [pr] (#16915) George HotzandGitHub 2026-07-07 14:30:32 -07:00
• 38610b5953 movement.py move_mop_cleanup geohot 2026-07-07 14:19:48 -07:00
• 79de8a2d45 revert geohot 2026-07-07 14:11:10 -07:00
• 12f15cd7dd more mop_cleanups geohot 2026-07-07 14:06:57 -07:00
• b61c33d972 those are mop cleanups geohot 2026-07-07 14:01:56 -07:00
• fc1d1878d1 move mop cleanup [pr] geohot 2026-07-07 13:59:36 -07:00
• 3f248070b2 add movement op rendering to pyrender (#16910) qazalandGitHub 2026-07-08 04:48:48 +08:00
• 2fc7e5341b final removal of PtrDType (glm) (#16913) George HotzandGitHub 2026-07-07 13:37:33 -07:00
• 27de6c6db0 use more UOp.valid method [PR] (#16912) chenyuandGitHub 2026-07-07 16:27:37 -04:00
• 39f24c4773 stack is fine reg_coalese geohot 2026-07-07 13:19:44 -07:00
• d8e4ad4f37 allow coalese for reg geohot 2026-07-07 13:13:27 -07:00
• fb4a781fca fix a few imports [PR] (#16909) chenyuandGitHub 2026-07-07 14:21:13 -04:00
• d8fbbff260 EXPAND adds dims to the front (glm) (#16908) George HotzandGitHub 2026-07-07 11:03:06 -07:00
• 91cefdb52a hcq2: rewrite patch split (#16907) nimlgenandGitHub 2026-07-07 20:43:52 +03:00
• 0035bb6fa8 only front reduce (#16901) George HotzandGitHub 2026-07-07 09:55:07 -07:00
• 6f1a983493 Ops.GROUP has no shape, like sink (#16906) George HotzandGitHub 2026-07-07 08:20:38 -07:00
• 357b7544d6 Allow non-const alt values for wgsl gated load (#16885) Ben WaldronandGitHub 2026-07-07 15:18:02 +00:00
• f0a06fae4c viz: readable REWRITE_ERROR (#16903) qazalandGitHub 2026-07-07 16:25:42 +08:00
• 9454e2ddbf viz: inf loop early cutoff (#16902) qazalandGitHub 2026-07-07 16:00:45 +08:00
• a805ce03b1 gptoss: model train (#16884) wozeparrotandGitHub 2026-07-07 01:07:50 -04:00
• 6e44176cfe fix flash attention example (#16900) George HotzandGitHub 2026-07-06 20:51:32 -07:00
• d94ad4444e fix spec for amd_copy_matmul + remove SHAPED_WMMA (#16897) George HotzandGitHub 2026-07-06 19:00:10 -07:00
• 40d112d4d6 remove ptr=True from index (#16898) George HotzandGitHub 2026-07-06 18:34:18 -07:00
• 6b8b2f5aeb remove PtrDtype from tests (#16896) George HotzandGitHub 2026-07-06 18:22:22 -07:00
• eb3cb16cc1 clean up pm_remove_invalid [PR] (#16895) chenyuandGitHub 2026-07-06 20:10:52 -04:00
• 4faed79216 remove placeholder from is_ptr (#16893) George HotzandGitHub 2026-07-06 16:55:56 -07:00
• ccd3428aad remove ptrdtype from kernel graph (#16892) George HotzandGitHub 2026-07-06 14:59:07 -07:00
• 9512dc30f4 remove vec from const (#16889) George HotzandGitHub 2026-07-06 12:23:07 -07:00
• 47f9677f67 hcq2: shuffle passes (#16886) nimlgenandGitHub 2026-07-06 21:56:55 +03:00
• 465c689d1f reduce removes axes now geohot 2026-07-06 11:52:11 -07:00
• e86c56f00f upstream decouple isa reg property from uop (#16804) RaineandGitHub 2026-07-06 14:28:07 -04:00
• deb5232413 skip test_index_fused_out_of_bounds if CL (#16888) sirhcmandGitHub 2026-07-06 13:53:58 -04:00
• a330bfffc9 gptoss: remove contiguous (#16882) wozeparrotandGitHub 2026-07-05 20:48:32 -04:00
• 35522af0df gptoss: use quantize_mxfp8 (#16881) wozeparrotandGitHub 2026-07-05 20:48:24 -04:00
• 62cde59cb5 hcq2: tiny changes (#16880) nimlgenandGitHub 2026-07-06 00:24:55 +03:00
• 4462ec3013 remove sintify and sgep [PR] (#16879) chenyuandGitHub 2026-07-05 13:28:23 -04:00
• 17e4f9b6f9 hcq2: simpler (#16878) nimlgenandGitHub 2026-07-05 18:26:19 +03:00
• a398b678cd skip test_index_fused_out_of_bounds (#16877) chenyuandGitHub 2026-07-05 09:30:37 -04:00
• dd6aa77fd1 delete Ops.UNROLL and Ops.CONTRACT (#16872) chenyuandGitHub 2026-07-05 08:28:40 -04:00
• 96b94bbf1d skip sqtt tests that unpickle UOps (#16875) qazalandGitHub 2026-07-05 17:52:31 +08:00
• 0cff3d74ca do not use Ops.UNROLL and Ops.CONTRACT [PR] (#16870) chenyuandGitHub 2026-07-04 19:37:13 -04:00
• c92a4c0442 move devectorizer remains into coalese (#16869) George HotzandGitHub 2026-07-04 15:36:45 -07:00
• af0abe4032 fix flaky move_devec geohot 2026-07-04 15:29:27 -07:00
• 964df8ec0e move devectorizer remains into coalese geohot 2026-07-04 15:27:12 -07:00
• 43c1bbceb1 remove pm_render (#16868) George HotzandGitHub 2026-07-04 15:18:43 -07:00
• a145fdce4b remove the .gep method (#16867) George HotzandGitHub 2026-07-04 14:07:09 -07:00