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

• 9c533e5c38 hcq: cpu prereq (#11296) nimlgenandGitHub 2025-07-21 13:35:18 +03:00
• e87a42e243 hcq: prepare for windows (#11293) nimlgenandGitHub 2025-07-21 13:08:56 +03:00
• df3ba0a7c0 autogen: fix imports in libusb (#11294) nimlgenandGitHub 2025-07-21 13:04:27 +03:00
• dd6a2d432f hcq: default timestamp metrics is ns (#11295) nimlgenandGitHub 2025-07-21 12:56:30 +03:00
• 53345ef4e2 feat: make ops_disk work on block devices (#11291) wozeparrotandGitHub 2025-07-20 14:39:50 -07:00
• 3002c63b1e process replay: optionally pass tinygrad import error (#11289) qazalandGitHub 2025-07-20 22:57:56 +03:00
• 9e3a593313 minor kernel.py cleanups [pr] (#11286) chenyuandGitHub 2025-07-20 10:15:31 -04:00
• 5f17927a87 Shorten UOp.load method (#11285) quortusandGitHub 2025-07-20 12:48:04 +02:00
• 54924f9969 type remove Union and Optional [pr] (#11283) chenyuandGitHub 2025-07-19 14:05:52 -04:00
• 2f72be5055 nv_smi: init basic insmod/rmmod/reset cmds (#11282) nimlgenandGitHub 2025-07-19 15:43:03 +03:00
• 577e581943 fix typo in sqtt/readme (#11281) qazalandGitHub 2025-07-19 15:10:24 +03:00
• 188ed38315 replace from_mv with lightweight mv_address (#11280) nimlgenandGitHub 2025-07-19 13:50:51 +03:00
• 1a25e27f32 Do not produce out of spec intermediate UOp in gated LOAD/STORE folding (#11207) 3 people 2025-07-18 21:42:55 +02:00
• ec3efd2919 move upcast before reduce (#11250) chenyuandGitHub 2025-07-18 14:42:15 -04:00
• be2f4336e6 use onnx 1.18.0 in DSP test (#11279) chenyuandGitHub 2025-07-18 14:09:23 -04:00
• 9a88bd841c hcq: refactor into peer_groups (#11277) nimlgenandGitHub 2025-07-18 16:34:18 +03:00
• f432eef708 hcq: rename CPU -> KICK in graph for kickoff signal (#11278) nimlgenandGitHub 2025-07-18 15:54:35 +03:00
• 52bbd9900b [pr] Stable tensor order in _find_all_tensors_for_uops (#11276) quortusandGitHub 2025-07-18 12:12:01 +02:00
• c5a5d74642 Revert "image_dot of 2 half inputs returns half (#11007)" (#11274) chenyuandGitHub 2025-07-17 17:34:18 -04:00
• fa8e08f922 image_dot of 2 half inputs returns half (#11007) 3 people 2025-07-18 02:17:22 +05:30
• 536b254df4 Bump onnx to 1.18.0 (#11266) geohotstanandGitHub 2025-07-18 03:35:41 +08:00
• 1606491b1c viz: refactor to generic shape spec (#11272) qazalandGitHub 2025-07-17 20:25:15 +03:00
• cfb229473f hcq: refactor buffer mapping (#11271) nimlgenandGitHub 2025-07-17 15:16:49 +03:00
• e68af3b336 disable flaky assert in test_cpu_profile (#11270) qazalandGitHub 2025-07-17 06:50:39 +03:00
• 60ffe00172 remove Kernel.first_reduce [pr] (#11269) chenyuandGitHub 2025-07-16 18:30:14 -04:00
• 522dc72f08 remove Kernel.local_dims [pr] (#11268) chenyuandGitHub 2025-07-16 17:46:19 -04:00
• d8c783f65f remove Kernel.global_dims [pr] (#11267) chenyuandGitHub 2025-07-16 17:16:49 -04:00
• 6f0ddcc24c Remote cross-host graph (#11229) uuuvnandGitHub 2025-07-17 01:27:54 +05:00
• 6aa20c607d nv: graceful shutdown to cold state (#11265) nimlgenandGitHub 2025-07-16 19:49:35 +03:00
• 59b52d49d7 remove .global_dims that are for locating GLOBAL [pr] (#11264) chenyuandGitHub 2025-07-16 11:19:31 -04:00
• e6c016ddd0 move check axis < shape_len to real_axis [pr] (#11263) chenyuandGitHub 2025-07-16 10:15:44 -04:00
• 924bc7c9ae Fix test_uop_spec (#11259) quortusandGitHub 2025-07-16 10:02:31 +02:00
• c8e5c4d7c3 insert_before -> insert_at [pr] (#11257) chenyuandGitHub 2025-07-15 17:44:34 -04:00
• b32d9321fb feat: more keccak cleanup + more explicit shape (#11256) wozeparrotandGitHub 2025-07-15 13:57:47 -07:00
• 9f79079cbe update KernelInfo dims to return list of dims [pr] (#11255) chenyuandGitHub 2025-07-15 15:01:39 -04:00
• 629fa21b6b remove final range in heuristic [pr] (#11251) chenyuandGitHub 2025-07-15 11:39:15 -04:00
• d7adc24083 remove Kernel.first_upcast [pr] (#11248) chenyuandGitHub 2025-07-15 10:21:34 -04:00
• 197d345804 nv: print rpc msg with DEBUG>=3 (#11247) nimlgenandGitHub 2025-07-15 16:39:58 +03:00
• 034e51bd36 remove first_reduce used for locate real_axis [pr] (#11245) chenyuandGitHub 2025-07-15 09:19:38 -04:00
• 0e2422d216 Kernel.axes_of helper [pr] (#11243) chenyuandGitHub 2025-07-14 22:17:43 -04:00
• 968f6b2a2e remove hasattr(self, 'axis_types') checks in dims property [pr] (#11242) chenyuandGitHub 2025-07-14 20:59:51 -04:00
• 557ca7d757 testing SimpleTokenizer against OASST1 (#11214) leopfandGitHub 2025-07-15 02:09:31 +02:00
• 5878b189b8 don't const fold shape changing bitcast (#11236) wozeparrotandGitHub 2025-07-14 16:42:16 -07:00
• b6662096cb remove more first_reduce [pr] (#11239) chenyuandGitHub 2025-07-14 19:13:44 -04:00
• eb8e17ef59 remove most of the first_upcast [pr] (#11238) chenyuandGitHub 2025-07-14 16:54:24 -04:00
• c78b1cbae7 viz profiler cleanups (#11234) qazalandGitHub 2025-07-14 19:06:33 +03:00
• 36ce883c7d update heuristic to use k.upcastable_dims and k.unrollable_dims [pr] (#11233) chenyuandGitHub 2025-07-14 11:10:30 -04:00
• c0c695dd89 viz: remove extra transform (#11232) qazalandGitHub 2025-07-14 16:51:47 +03:00
• da219199f5 minor hcopt cleanup [pr] (#11231) chenyuandGitHub 2025-07-14 09:36:25 -04:00
• 756ba1a5f9 nv: support ampere in nvpci (#11230) nimlgenandGitHub 2025-07-14 15:35:44 +03:00
• b2cc6cfa1b JIT_BATCH_SIZE is a ContextVar (#11228) uuuvnandGitHub 2025-07-14 11:03:45 +00:00
• c4a920d95c nv: use last signature (#11227) nimlgenandGitHub 2025-07-14 13:00:39 +03:00
• a830d37881 nv: check wpr2 is inited (#11226) nimlgenandGitHub 2025-07-14 11:46:14 +03:00
• 0387bb9630 clean up image upcast in hcopt [pr] (#11220) chenyuandGitHub 2025-07-13 18:06:43 -04:00
• 85ddd72038 simpler grouptop in hcopt (#11219) chenyuandGitHub 2025-07-13 16:06:09 -04:00
• 40847ca29c viz: prune out of screen rects (#11217) qazalandGitHub 2025-07-13 21:49:59 +03:00
• 674dc28505 remove Kernel.full_unupcasted_shape [pr] (#11215) chenyuandGitHub 2025-07-13 13:56:23 -04:00
• 9575cf6c6e shave more hcopt [pr] (#11213) chenyuandGitHub 2025-07-13 12:43:58 -04:00
• 4ef6b46b34 hcq: reduce launch overhead (#11193) Alisher ZhubanyshevandGitHub 2025-07-13 21:25:50 +05:00
• 1cc2b3f845 nv: use wait_cond (#11212) nimlgenandGitHub 2025-07-13 19:25:20 +03:00
• 6cce3a5d58 generic wait_cond (#11210) nimlgenandGitHub 2025-07-13 16:59:21 +03:00
• e11ccf2342 update float4 condition in hcopt (#11211) chenyuandGitHub 2025-07-13 09:51:45 -04:00
• 55c54d9745 nv: sync after gpfifo setup (#11209) nimlgenandGitHub 2025-07-13 14:40:11 +03:00
• d90d837013 clean up hcopt [pr] (#11205) chenyuandGitHub 2025-07-12 23:10:27 -04:00
• 2b48b961be fix a few broken AMX tests (#11204) chenyuandGitHub 2025-07-12 21:42:38 -04:00
• 667c7a9fa6 clean: keccak cleanups + explicit shapes (#11202) wozeparrotandGitHub 2025-07-12 18:17:14 -07:00
• a0438012af remove Kernel.get_program [pr] (#11203) chenyuandGitHub 2025-07-12 20:50:29 -04:00
• d67c8e7b42 local metal on metal in uop syntax (#11185) George HotzandGitHub 2025-07-12 16:31:19 -07:00
• 40da5f0c81 fix silent mypy failure in ci (#11201) uuuvnandGitHub 2025-07-12 19:12:25 +00:00
• 73caa5dd1b remove Kernel.membufs [pr] (#11200) chenyuandGitHub 2025-07-12 14:48:47 -04:00
• 5ce278b245 OnnxRunner file as input (#10789) 3 people 2025-07-13 02:27:46 +08:00
• 110cff3f2e fix device arg to Tensor.randn (#11194) nimlgenandGitHub 2025-07-12 20:51:59 +03:00
• 6283d50224 DEPRECATED_linearize -> to_program [pr] (#11198) chenyuandGitHub 2025-07-12 13:46:20 -04:00
• 770a558585 lil cleanups from uop branch [pr] (#11197) George HotzandGitHub 2025-07-12 09:46:28 -07:00
• 5625e1904b axis types in KernelInfo (#11196) George HotzandGitHub 2025-07-12 09:36:20 -07:00
• ea7f2f779c hcq: p2p nv-amd (#11195) nimlgenandGitHub 2025-07-12 18:53:34 +03:00
• 6a9f059b21 viz: early convert to cpu time (#11192) qazalandGitHub 2025-07-12 17:19:41 +03:00
• 12b04efd69 remove a TODO prod(k.full_shape[k.first_upcast:]) (#11191) chenyuandGitHub 2025-07-12 10:16:56 -04:00
• 6f5250d158 nv: fix typing in rpc_rm_control (#11189) nimlgenandGitHub 2025-07-12 16:09:42 +03:00
• c0a5490c72 viz: minor profiler cleanup (#11190) qazalandGitHub 2025-07-12 14:18:24 +03:00
• fdcc25e392 some noop hand_coded_optimizations cleanup [pr] (#11188) chenyuandGitHub 2025-07-12 00:09:23 -04:00
• 1ad852a892 break up Kernel.reshape_and_permute [pr] (#11187) chenyuandGitHub 2025-07-11 18:08:08 -04:00
• d11b20129d DMARef infra (#10753) 3 people 2025-07-11 21:09:47 +00:00
• b072be0e2d hotfix whisper main script (#11184) chenyuandGitHub 2025-07-11 12:34:00 -04:00
• 0b7e9b5db7 viz: bugfix for multiple rewrites with the same name (#11182) qazalandGitHub 2025-07-11 18:26:12 +03:00
• f9e4c4e57a nv: nvpci blackwell support (#11127) nimlgenandGitHub 2025-07-11 17:02:09 +03:00
• 1d85323572 viz: absolute scaling of memory graph (#11181) qazalandGitHub 2025-07-11 16:39:11 +03:00
• c7f6b617b4 nv: do not hardcode lv0 pd size (#11180) nimlgenandGitHub 2025-07-11 16:26:18 +03:00
• 27922c986a nv: generic mmu impl (#11179) nimlgenandGitHub 2025-07-11 16:26:09 +03:00
• d3ec63a5c3 viz: add base class for unittests (#11178) qazalandGitHub 2025-07-11 13:58:03 +03:00
• b791ea117d viz: enable scrolling in profiler (#11169) qazalandGitHub 2025-07-11 11:30:13 +03:00
• b219e47bef remove Kernel.upcasted_axis [pr] (#11175) chenyuandGitHub 2025-07-10 23:19:21 -04:00
• ccd382bc6f use axis_types more [pr] (#11172) George HotzandGitHub 2025-07-10 15:05:13 -07:00
• fb278c6a02 do not recreate Compiled.profile_events in helper_collect_profile (#11171) nimlgenandGitHub 2025-07-10 23:55:12 +03:00
• 5c5eb92ed4 tc unroll after upcast [pr] (#11170) George HotzandGitHub 2025-07-10 13:43:50 -07:00
• 05613c8cac use shape str for tensor cores upcast/reduce [pr] (#11168) George HotzandGitHub 2025-07-10 13:10:58 -07:00
• cc6ed30f4f nv: relative lv addressing in NVPageTableEntry (#11164) nimlgenandGitHub 2025-07-10 22:35:50 +03:00
• 439d033af9 update the README matmul example (#11167) chenyuandGitHub 2025-07-10 14:47:29 -04:00
• bde80c0cdf record GraphEvents in metal graph (#11145) qazalandGitHub 2025-07-10 21:32:06 +03:00
• 8ce3d5906b use shape_str for tensor cores (#11165) George HotzandGitHub 2025-07-10 09:10:36 -07:00