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

• 0a45cd0cbe grouper: merge views in fuse elementwise (#10325) qazalandGitHub 2025-05-15 13:17:09 +03:00
• 89d8d5b25e add dims check in FUSE_ARANGE (#10323) qazalandGitHub 2025-05-15 11:33:21 +03:00
• 8fad0f0124 grouper: check for unsafe PAD in FUSE (#10322) qazalandGitHub 2025-05-15 10:53:44 +03:00
• f008e5f233 test_dtype_alu should cast bf16 input (#10320) chenyuandGitHub 2025-05-15 13:11:39 +08:00
• 568d6d96e7 small changes from new multi [pr] (#10318) George HotzandGitHub 2025-05-14 20:50:59 -07:00
• f6cf25fce4 cleanup test_conv2d_ceildiv_edge_case [pr] (#10317) chenyuandGitHub 2025-05-15 11:35:28 +08:00
• 50d7162acd Add conv2d ceildiv edge case (#10303) Kirill R.andGitHub 2025-05-15 05:50:23 +03:00
• e5639b7788 Remote finalize (#10314) 3 people 2025-05-15 07:40:44 +05:00
• bfc30fa6ea hotfix: typo in shm_name geohot 2025-05-14 19:34:52 -07:00
• 2bc54b3e22 manually handle OSX geohot 2025-05-14 19:17:51 -07:00
• ab460486d7 Revert "resnet dataloader osx (#10316)" geohot 2025-05-14 19:15:07 -07:00
• 7b4f27a219 Remote .q(..., wait=True) (#10313) uuuvnandGitHub 2025-05-15 07:07:20 +05:00
• 50181ab09f hotfix: bump to 13500 lines geohot 2025-05-14 18:49:59 -07:00
• aef336930a resnet dataloader osx (#10316) George HotzandGitHub 2025-05-14 18:31:26 -07:00
• 9b14e8c3cd feat: tag 0.10.3 (#10310) v0.10.3 wozeparrotandGitHub 2025-05-14 15:45:13 -07:00
• 18f532d110 small changes from O(1) multi [pr] (#10309) George HotzandGitHub 2025-05-14 15:34:07 -07:00
• 9bbc2bc2a7 hotfix: filter_too_much (#10308) wozeparrotandGitHub 2025-05-14 15:31:51 -07:00
• fc8ef63194 multi doesn't need tuple arg anymore [pr] (#10307) George HotzandGitHub 2025-05-14 15:16:40 -07:00
• 7a3d4de59a hotfix: add GRAPH_ONE_KERNEL=1 to UsbGPU openpilot test geohot 2025-05-14 14:49:48 -07:00
• 2df2ec6640 feat: unpin hypothesis (#10306) wozeparrotandGitHub 2025-05-14 14:26:28 -07:00
• b52452d69f Remote multi (graph) (#9902) 3 people 2025-05-14 22:54:23 +05:00
• 42e70193c9 multi: instead of real, just copy (#10289) George HotzandGitHub 2025-05-14 10:36:55 -07:00
• 043efc6ec4 do not require self for track_rewrites [pr] (#10302) qazalandGitHub 2025-05-14 18:23:32 +03:00
• dd816d0237 All MultiGraphRunners can graph transfers (#10301) uuuvnandGitHub 2025-05-14 19:23:02 +05:00
• e00679dc92 am_smi: fix layout with sleep mode (#10300) nimlgenandGitHub 2025-05-14 15:44:42 +03:00
• fbaa26247a randn_like in minrf (#10298) chenyuandGitHub 2025-05-14 07:59:50 -04:00
• 0788659d08 usbgpu: fast cold boot (#10260) nimlgenandGitHub 2025-05-14 14:58:55 +03:00
• d342f7688d remove some skips in test_schedule + use assertRaisesRegex [pr] (#10296) qazalandGitHub 2025-05-14 14:54:07 +03:00
• 40f4ce3390 enable AMD CI for TestRandomness.test_multinomial [pr] (#10295) qazalandGitHub 2025-05-14 14:32:22 +03:00
• 792853b9e2 usbgpu: enable cache for compute queue (#10294) nimlgenandGitHub 2025-05-14 13:05:36 +03:00
• 1218fc2230 usbgpu: enable cache for 64bit addresses (#10293) nimlgenandGitHub 2025-05-14 12:37:39 +03:00
• 1770e00c41 only CAPTURE_PROCESS_REPLAY=1 + add filterwarnings back [pr] (#10292) qazalandGitHub 2025-05-14 11:58:42 +03:00
• 1c97338be5 enable process replay assert for schedule [pr] (#10280) qazalandGitHub 2025-05-14 11:10:47 +03:00
• f1130ab3d3 openpilot benchmark test (#10290) George HotzandGitHub 2025-05-13 22:49:28 -07:00
• f726f79a9e Remote multi (transfer) (#10285) uuuvnandGitHub 2025-05-14 06:26:32 +05:00
• 7bc4864bc4 Make dev a property of Allocator (#10286) uuuvnandGitHub 2025-05-14 05:01:01 +05:00
• ec46f658d7 openpilot llvm test [pr] (#10288) George HotzandGitHub 2025-05-13 16:51:41 -07:00
• 453b268342 Factor out remote connection and cache it (#10282) uuuvnandGitHub 2025-05-14 03:02:06 +05:00
• ddff9857b8 Remote properties is a dataclass (#10283) uuuvnandGitHub 2025-05-13 23:56:58 +05:00
• ba87eca0f1 Remote multi (basic) (#10269) uuuvnandGitHub 2025-05-13 21:52:47 +05:00
• 5f64bbc63d improve multi tests + add support for fixedvars [pr] (#10281) George HotzandGitHub 2025-05-13 09:27:00 -07:00
• 8a906cb124 Tensor.randn_like (#10276) chenyuandGitHub 2025-05-13 11:53:59 -04:00
• eab71d70ba usbgpu: rescan pci bus every run (#10279) nimlgenandGitHub 2025-05-13 18:31:42 +03:00
• c4988bc07b only run test_u32_to_f16 if it supports fp16 (#10277) chenyuandGitHub 2025-05-13 11:16:14 -04:00
• 9924c7d0e4 usbgpu: rebar (#10275) nimlgenandGitHub 2025-05-13 17:25:51 +03:00
• 1900c3c68a Metal multi in ci is fine actually (#10274) uuuvnandGitHub 2025-05-13 19:07:35 +05:00
• 6f42bf8b54 usbgpu: 10 steps in benchmark to hit cache (#10273) nimlgenandGitHub 2025-05-13 17:06:50 +03:00
• ad5cb2717d FUSE_ARANGE=1 in bert bench (#10263) 3 people 2025-05-13 09:12:19 -04:00
• a2d6b0afe0 fix FUSE pushing through SHRINK (#10271) qazalandGitHub 2025-05-13 11:38:53 +03:00
• 1c4ab6b991 ONNX add tests against ORT (#10270) geohotstanandGitHub 2025-05-13 16:03:52 +08:00
• bb31cc4582 usbgpu: check hash in patcher (#10266) nimlgenandGitHub 2025-05-12 21:08:53 +03:00
• 94907d02c8 Move session to RemoteRequest (#10264) uuuvnandGitHub 2025-05-12 22:06:09 +05:00
• 02208565de add check (#10257) Sieds LyklesandGitHub 2025-05-12 17:03:01 +02:00
• 08ab184dfd usbgpu: copyin over 100mb/s (#10259) nimlgenandGitHub 2025-05-12 16:52:43 +03:00
• 4c7c139102 Use cmod/cdiv in sym_infer (#10258) 3 people 2025-05-12 16:07:28 +03:00
• 0015b3921f sleep more in CI Remove amdgpu (#10261) chenyuandGitHub 2025-05-12 08:13:44 -04:00
• 95c6a736a9 fix FUSE_ARANGE=1 for bert (#10255) qazalandGitHub 2025-05-12 14:44:05 +03:00
• 7c4b381fbf Extra simplify valid test [pr] (#10256) Sieds LyklesandGitHub 2025-05-12 13:32:03 +02:00
• 7eeb35ba6f fix AMD LLVM compile error for bf16 cifar (#10254) 3 people 2025-05-12 13:57:07 +08:00
• a0ed1ec1ae Faster remote server (#10235) uuuvnandGitHub 2025-05-12 07:15:05 +05:00
• 41f5ece877 add nsw flag (#10249) 3 people 2025-05-12 10:14:32 +08:00
• 8864ff894b hotfix: that repeat_kv belongs outside the if geohot 2025-05-11 18:41:57 -07:00
• 98c84a711d min rectified flow example [pr] (#10252) George HotzandGitHub 2025-05-11 18:36:44 -07:00
• 70c797b107 train bert tests (#10248) chenyuandGitHub 2025-05-11 08:42:08 -04:00
• b2df4cb696 add support for amdgpu-flat-work-group-size to AMD LLVM IR (#10246) 3 people 2025-05-10 19:11:10 -07:00
• 9210280811 add v_fmac_f16 vop3 instruction to remu (#10247) qazalandGitHub 2025-05-10 23:48:25 +03:00
• 697259a8a1 amd_comgr_action_info_set_options was deprecated [pr] (#10245) George HotzandGitHub 2025-05-10 11:59:04 -07:00
• 2e0990c4e9 even spacing in viz nodes (#10168) 3 people 2025-05-10 01:35:10 -06:00
• d0e9b74f40 minor div_and_mod_folding cleanup [pr] (#10243) chenyuandGitHub 2025-05-09 22:42:01 -04:00
• a28ca0680f update dead link (#10242) Adam Van YmerenandGitHub 2025-05-09 16:59:52 -07:00
• 2145bce3f9 usbgpu: copyin size is 16k (#10240) nimlgenandGitHub 2025-05-09 22:12:54 +03:00
• 74e40aafa0 use cdiv in div and mod folding (#10216) 3 people 2025-05-09 18:37:24 +02:00
• 8da9c070ca take gcd out of trunc div (#10238) Sieds LyklesandGitHub 2025-05-09 18:08:10 +02:00
• e2292f6663 TRACEMETA>=2 displays UOp metadata in VIZ (#10237) qazalandGitHub 2025-05-09 17:42:00 +03:00
• d5686f33a9 delete KernelContext dataclass [pr] (#10236) qazalandGitHub 2025-05-09 17:36:21 +03:00
• 467daf8d4c remap UOp metadata in graph_rewrite_map [pr] (#10234) qazalandGitHub 2025-05-09 17:20:53 +03:00
• 4c75b124b6 usb: copy into mv is faster (#10233) nimlgenandGitHub 2025-05-09 14:53:36 +03:00
• d08ce62553 hcq: do not reread signal in wait (#10232) nimlgenandGitHub 2025-05-09 14:38:36 +03:00
• 0464a31000 usbgpu: no overrun check needed (#10231) nimlgenandGitHub 2025-05-09 14:20:24 +03:00
• 116390083f nvme speed write example (#10230) nimlgenandGitHub 2025-05-09 14:20:01 +03:00
• 9846435c2e fix test_div_numerator_negative (#10229) chenyuandGitHub 2025-05-09 06:19:59 -04:00
• cba508c8c3 update uop symbolic tests (#10228) chenyuandGitHub 2025-05-09 01:55:53 -04:00
• 56def6c319 better bound for mod negative number (#10227) chenyuandGitHub 2025-05-09 01:19:47 -04:00
• 99f6d89dfb tighter idiv bound for symbolic denominator (#10226) chenyuandGitHub 2025-05-08 22:38:56 -04:00
• 82a6160ff7 Detect metal paravirtualization bug via device name instead of CI (#10225) uuuvnandGitHub 2025-05-09 07:31:47 +05:00
• a21369d039 Enhance tensor random functions with dtype support (#10214) XingyuandGitHub 2025-05-09 08:48:07 +08:00
• b6904bbf83 Revert "split grouper into insert and finalize stages [pr] (#10222)" (#10224) qazalandGitHub 2025-05-09 03:02:38 +03:00
• 2594e4db15 split grouper into insert and finalize stages [pr] (#10222) qazalandGitHub 2025-05-09 02:36:22 +03:00
• 0b7e3e86d0 single device copy [pr] (#10221) George HotzandGitHub 2025-05-08 15:23:22 -07:00
• 1d0f239df7 use Tensor.train() in schedule test + typo [pr] (#10220) qazalandGitHub 2025-05-08 23:46:42 +03:00
• ff2aa6d0b2 buffer in create_kernel is optional [pr] (#10218) qazalandGitHub 2025-05-08 22:35:55 +03:00
• 40560e77c2 minor grouper + viz fixup [pr] (#10217) qazalandGitHub 2025-05-08 21:39:44 +03:00
• 0411b09763 small changes from new multi [pr] (#10213) George HotzandGitHub 2025-05-08 07:04:27 -07:00
• a0580e8d3c Cleanup in div_and_mod_folding [pr] (#10178) 3 people 2025-05-08 15:25:32 +02:00
• 267ba9b592 usbgpu: better names in copy speed benchmark (#10212) nimlgenandGitHub 2025-05-08 16:12:37 +03:00
• 7b4f05fd00 Add test for correctness of Infinity in WebGPU (#10201) hoovedandGitHub 2025-05-08 08:20:05 -04:00
• e24fe1c746 usbgpu: pci cache (#10207) nimlgenandGitHub 2025-05-08 14:31:01 +03:00
• 7d6ed1b1e9 hotfix: mac ci (#10210) nimlgenandGitHub 2025-05-08 14:13:23 +03:00
• ba52fce4b2 usbgpu: benchmark in ci (#10208) nimlgenandGitHub 2025-05-08 12:02:04 +03:00
• d0e3449992 remove view_supported_devices, check allocator instead [pr] (#10209) qazalandGitHub 2025-05-08 11:45:02 +03:00