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

• 8cc2dff4d8 only float Tensors have gradient [pr] (#10475) chenyuandGitHub 2025-05-22 21:02:11 -04:00
• 147f7747f2 remove the map from create_schedule_with_vars [pr] (#10472) George HotzandGitHub 2025-05-22 15:58:25 -07:00
• 6d5f87a18a lshift/rshift reverse is broken [pr] (#10467) George HotzandGitHub 2025-05-22 13:01:48 -07:00
• 209d4401f8 Merge SimpleMathTrait and MathTrait (#10463) Mike AshcroftandGitHub 2025-05-22 14:47:22 -04:00
• 0d39bb5de1 rename to get_kernelize_map (#10465) George HotzandGitHub 2025-05-22 11:44:44 -07:00
• 1e0a59aca4 fix: handle buffer size calculation in to_movement_ops and add scalar assignment test in torch_backend (#10464) XingyuandGitHub 2025-05-23 01:54:13 +08:00
• 577a0b4cfa openpilot compile4 (wip) (#10407) George HotzandGitHub 2025-05-22 10:47:34 -07:00
• ab591fa4dd make schedule explicit about kernels [pr] (#10462) George HotzandGitHub 2025-05-22 09:32:16 -07:00
• c46edbf262 hotfix: add note to relu geohot 2025-05-22 09:13:38 -07:00
• c6cbf0145a check that arg on copy is only used on multi [pr] (#10461) George HotzandGitHub 2025-05-22 09:08:43 -07:00
• f69722dc2a refactor cuda disassemble (#10449) Ignacio SicaandGitHub 2025-05-22 12:58:24 -03:00
• 5c4cfbc22c remove merge_views from kernel grouping rewrite [pr] (#10457) qazalandGitHub 2025-05-22 18:36:54 +03:00
• 035dffb00c nv: refactor qmd from ctypes (#10459) nimlgenandGitHub 2025-05-22 17:20:11 +03:00
• 12285e926a fix: apply ip version fixes during AMDIP creation (#10454) wozeparrotandGitHub 2025-05-22 00:14:48 -07:00
• 5e6b96a1be align 16 in ptx, metal, cuda and amd (#10450) Ignacio SicaandGitHub 2025-05-21 18:38:54 -03:00
• 570cb89652 amd: handle all exceptions (#10448) nimlgenandGitHub 2025-05-21 16:51:44 +03:00
• 475a7583b3 usbgpu: tiny changes (#10445) nimlgenandGitHub 2025-05-21 16:20:35 +03:00
• 7720c1aef1 hotfix: remove viz_sz.py [pr] (#10446) qazalandGitHub 2025-05-21 14:17:42 +03:00
• 7bfb20757c fix tensor int floor div (#10327) chenyuandGitHub 2025-05-21 06:46:54 -04:00
• 2b4375f36d Correct divmod folding behind flag (#10433) Sieds LyklesandGitHub 2025-05-21 12:46:13 +02:00
• df4cbb69e9 move fuzz_schedule.py to extra [pr] (#10444) qazalandGitHub 2025-05-21 10:07:24 +03:00
• 29624af872 skip commavq in external_model_benchmark (#10439) chenyuandGitHub 2025-05-21 01:45:33 -04:00
• 03e7a99ca8 add edge cases found by codex [pr] (#10423) George HotzandGitHub 2025-05-20 14:53:18 -07:00
• 2895198c36 am: download regs (#10419) nimlgenandGitHub 2025-05-20 18:59:56 +03:00
• 965f9e0696 amd: refactor amdreg (#10427) nimlgenandGitHub 2025-05-20 15:23:28 +03:00
• 0b65c367f5 hotfix: make vfio not default (#10429) nimlgenandGitHub 2025-05-20 15:08:49 +03:00
• cfa5c1cac6 am: disable idle d3 (#10428) nimlgenandGitHub 2025-05-20 13:50:58 +03:00
• 252c1dc737 am: close flock in fini (#10426) nimlgenandGitHub 2025-05-20 13:15:48 +03:00
• ceb9d94eab Update AGENTS.md George HotzandGitHub 2025-05-19 17:59:59 -07:00
• 9389edf7ac hotfix: add AGENTS.md geohot 2025-05-19 17:47:31 -07:00
• ec9955c956 Use REAL_DEV for test skips (#10420) uuuvnandGitHub 2025-05-20 05:32:14 +05:00
• 9a199ccd81 am: try to modprobe vfio (#10418) nimlgenandGitHub 2025-05-19 23:46:50 +03:00
• 67d1364106 update LOGMLPERF in red resnet run_and_time (#10416) chenyuandGitHub 2025-05-20 01:23:33 +08:00
• db09676250 Dont simplify gate in gate, fix FUSE_ARANGE=1 python test/test_ops.py TestOps.test_scatter_add (#10411) Sieds LyklesandGitHub 2025-05-19 19:16:21 +02:00
• 116d9e6306 run mlperf resnet on red box (#10413) chenyuandGitHub 2025-05-20 00:48:36 +08:00
• f1fe1f93c1 hotfix: 14000 lines geohot 2025-05-19 09:40:53 -07:00
• 90eb3c0e5d add MobileNetV2 benchmark to comma CI (#10250) qazalandGitHub 2025-05-19 18:22:50 +03:00
• f9a5ad24c5 faster viz to_program [pr] (#10410) qazalandGitHub 2025-05-19 12:27:49 +03:00
• cc8dda1d75 move multi_map to grouper rewrite pass (#10409) qazalandGitHub 2025-05-19 10:44:06 +03:00
• b06291077c no amdgpu kernel driver (#10408) George HotzandGitHub 2025-05-18 20:52:39 -07:00
• 4b1f1a47bb hotfix: allow ModuleNotFoundError in metal llvm import geohot 2025-05-18 20:46:31 -07:00
• 485e80da69 run_and_time for resnet ci (#10405) chenyuandGitHub 2025-05-19 11:39:57 +08:00
• d1eeb19437 count viz javascript in lines (#10403) qazalandGitHub 2025-05-19 05:34:00 +03:00
• 260d194523 merge insert_fuse and do_fuse [pr] (#10406) qazalandGitHub 2025-05-19 04:44:36 +03:00
• 0071d864f8 run_and_time for resnet ci resnet-ci Chen-Yu Yang 2025-05-18 20:29:27 -04:00
• 33cf33902a Slightly less slow remote copyin (#10404) uuuvnandGitHub 2025-05-19 04:20:43 +05:00
• e55ee28b29 little smaller viz/worker.js [pr] (#10402) qazalandGitHub 2025-05-18 23:44:46 +03:00
• 8a6fb37560 move viz /prof to extra [pr] (#10401) qazalandGitHub 2025-05-18 23:25:59 +03:00
• 411392dfb7 move files into uop dir (#10399) George HotzandGitHub 2025-05-18 11:38:28 -07:00
• 0f825e12f2 Remote fixedvars (#10371) 3 people 2025-05-18 21:57:13 +05:00
• 27c12be471 amd mockgpu graph support (#10385) uuuvnandGitHub 2025-05-18 21:43:16 +05:00
• a3308e145d hotfix: remote print -> DEBUG=3 geohot 2025-05-18 09:09:04 -07:00
• 04b23087d8 grouper tests from fuse_arange_default [pr] (#10394) qazalandGitHub 2025-05-18 18:42:43 +03:00
• 17f0f5e764 add v_rcp_f32_e64 to remu (#10393) qazalandGitHub 2025-05-18 17:08:21 +03:00
• 9e2089dcd4 don't raise Exception in process replay [pr] (#10392) qazalandGitHub 2025-05-18 11:23:23 +03:00
• 9b4e2a75cd symlink datasets in mlperf workflow (#10391) chenyuandGitHub 2025-05-18 15:26:05 +08:00
• f20c5aac1f Use itertools.count instead of manual increment in remote (#10389) uuuvnandGitHub 2025-05-18 12:15:37 +05:00
• 0294bfe507 simpler can_pad (#10364) qazalandGitHub 2025-05-18 10:00:07 +03:00
• c91f2c4580 use float32 for sgd momentum (#10387) George HotzandGitHub 2025-05-17 21:56:44 -07:00
• 305a3231c4 fix beam none if buf is optimized out (#10388) George HotzandGitHub 2025-05-17 21:50:33 -07:00
• 6f77b938d7 Move getbits tests into test_helpers (#10382) George HotzandGitHub 2025-05-17 17:04:00 -07:00
• 6ebfb505e9 docs: fix crossentropy name (#10377) George HotzandGitHub 2025-05-17 16:39:14 -07:00
• 0b733ba75e multi device training with GPT2 [pr] (#10375) George HotzandGitHub 2025-05-17 15:33:56 -07:00
• 6ec88d94df add tests for multi ram usage [pr] (#10376) George HotzandGitHub 2025-05-17 15:33:40 -07:00
• 5a18eab908 Fix __del__ in remote program (#10372) uuuvnandGitHub 2025-05-17 23:29:44 +05:00
• 2453d99050 rms matching pytorch implementation (#10319) 3 people 2025-05-17 20:53:11 +05:30
• da2b1834b4 hotfix: metal graph var vals (#10370) nimlgenandGitHub 2025-05-17 17:22:55 +03:00
• e054b53a75 kernel count tests for pad [pr] (#10369) qazalandGitHub 2025-05-17 17:20:46 +03:00
• 90c4bb10c0 fixedvars in all graphs (#10365) nimlgenandGitHub 2025-05-17 16:18:52 +03:00
• efa8dfe7fb test cron job to run resnet (#10368) chenyuandGitHub 2025-05-17 20:57:02 +08:00
• 2c706d363e Remote higher timeout and overridable via REMOTE_TIMEOUT (#10367) uuuvnandGitHub 2025-05-17 17:30:49 +05:00
• 4fa1837916 metal: do not require icb fix on m3+ (#10366) nimlgenandGitHub 2025-05-17 15:30:40 +03:00
• 286b0f4051 Add equal function implementation and corresponding test (#10351) XingyuandGitHub 2025-05-17 14:39:49 +08:00
• e13f2a3092 multi is O(1) (#10183) George HotzandGitHub 2025-05-16 23:14:23 -07:00
• e1a40e8040 add hcq fixedvars support [pr] (#10356) George HotzandGitHub 2025-05-16 22:05:53 -07:00
• 11b5895c85 hotfix: schedule timing in tensor.py geohot 2025-05-16 20:09:41 -07:00
• 64409a8bda Remote beam (#10357) uuuvnandGitHub 2025-05-17 06:59:22 +05:00
• 7cc35a031b don't use UOp.multi in Tensor.rand (#10362) George HotzandGitHub 2025-05-16 16:09:36 -07:00
• 7703dbef99 view substitute [pr] (#10360) George HotzandGitHub 2025-05-16 15:08:24 -07:00
• de2b323d97 Fixed a typo in "simplify" (#10358) Elnur RakhmatullinandGitHub 2025-05-17 02:45:14 +05:00
• ee5258328a You never want multiple backends (#10354) Harald SchäferandGitHub 2025-05-16 13:10:39 -07:00
• 876d2275a1 changes from new multi (#10353) George HotzandGitHub 2025-05-16 13:07:29 -07:00
• 66e00c04dd fix: skip kernel timing tests on ci cuda (#10348) wozeparrotandGitHub 2025-05-16 11:48:06 -07:00
• a54fd745c3 simpler barrier match in remu (#10339) Ignacio SicaandGitHub 2025-05-16 08:40:58 -03:00
• e9e5b54e43 grouper cleanups and merge with insert_kernels [pr] (#10349) qazalandGitHub 2025-05-16 14:39:56 +03:00
• caded2f413 llvm diagnostic error (#10267) 4 people 2025-05-16 14:03:20 +08:00
• a4a25720b2 add test_multitensor_jit_input [pr] (#10347) George HotzandGitHub 2025-05-15 20:47:57 -07:00
• c798f2f427 brew --quiet to suppress already installed warnings (#10346) chenyuandGitHub 2025-05-16 11:31:18 +08:00
• 12a1ccc680 clean: double import (#10345) wozeparrotandGitHub 2025-05-15 20:15:09 -07:00
• 1ed04f993b move benchmark stat tracking to influxdb (#10185) wozeparrotandGitHub 2025-05-15 16:14:56 -07:00
• f59ecf2116 fix: mockgpu cuda timing (#10343) wozeparrotandGitHub 2025-05-15 14:14:14 -07:00
• a825608dc2 hcq: fix progs' __del__ when shutdown (#10341) nimlgenandGitHub 2025-05-15 23:26:48 +03:00
• 47b3055fe2 set fail-fast behavior (#10336) Ignacio SicaandGitHub 2025-05-15 15:24:45 -03:00
• c2bf2c6bb0 Remote offset (#10311) uuuvnandGitHub 2025-05-15 23:20:01 +05:00
• 3c453e96a9 add ds_load_b96 and ds_store_b96 instructions (#10338) Ignacio SicaandGitHub 2025-05-15 12:11:08 -03:00
• be8202b293 add s_abs_i32 instruction to remu (#10334) qazalandGitHub 2025-05-15 16:47:58 +03:00
• 5efbe1c947 print offset only for subbuf (#10332) nimlgenandGitHub 2025-05-15 15:35:19 +03:00
• 7cfe367c07 failing test for slow embedding kernel with FUSE_ARANGE=1 [pr] (#10330) qazalandGitHub 2025-05-15 14:58:11 +03:00
• 5f03688280 usbgpu: remove max_read_len (#10328) nimlgenandGitHub 2025-05-15 14:49:58 +03:00
• 27b3dbe67e remove FUSE_ARANGE_UINT [pr] (#10324) qazalandGitHub 2025-05-15 14:39:54 +03:00