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

• d32f5e9f3a improve rendering of shapes in viz + investigate symbolic [pr] (#10091) George HotzandGitHub 2025-04-28 16:44:09 -04:00
• dbb7aee02e Split constant in div with negative x (#10088) Sieds LyklesandGitHub 2025-04-28 22:24:14 +02:00
• 610ee79b22 cherry pick mlperf5.0 branch to master (#10089) chenyuandGitHub 2025-04-28 15:36:56 -04:00
• 459a223202 simpler Literal annotation in code_for_workitem [pr] (#10087) chenyuandGitHub 2025-04-28 14:59:25 -04:00
• 20ce1b5a33 branch mlperf_training_v5.0 Chen-Yu Yang 2025-04-28 14:42:44 -04:00
• 12c6573cf6 update for 5.0 Chen-Yu Yang 2025-04-28 14:32:08 -04:00
• dcd9a633c3 am: load minimum fw (#10083) nimlgenandGitHub 2025-04-28 21:28:05 +03:00
• edd4fa3b36 tinybox green mlperf submission Francis LataandChen-Yu Yang 2025-04-27 21:13:38 -07:00
• 8acf215660 tiny15 Chen-Yu Yang 2025-04-28 11:20:57 -07:00
• 46778ed9f3 tiny13 Chen-Yu Yang 2025-04-28 11:12:47 -07:00
• e795af38ea tiny10 Chen-Yu Yang 2025-04-28 11:08:20 -07:00
• ecff82a698 fixing single kernel softmax: resolve (#10086) George HotzandGitHub 2025-04-28 13:46:20 -04:00
• 4c242b0483 hotfix: tests all pass on metal local geohot 2025-04-28 12:09:00 -04:00
• 690dac79b5 don't modify the ranges on reduce rewrite (#10062) 3 people 2025-04-28 12:01:19 -04:00
• 5130759605 Make sure clang always inline batched functions (#10037) quortusandGitHub 2025-04-28 16:48:24 +02:00
• c4a50f9d89 fix full shape in kernel.py [pr] (#10085) George HotzandGitHub 2025-04-28 09:32:58 -04:00
• ac37510f60 remu: only write v_cmp result if exec is set (#10084) qazalandGitHub 2025-04-28 15:31:52 +03:00
• d6b436a815 remu bugfix with -0.0 negation (#10082) qazalandGitHub 2025-04-28 10:46:42 +03:00
• 15e4302784 am: optimize zeroing out boot structs (#10081) nimlgenandGitHub 2025-04-28 10:15:32 +03:00
• 68e5ab8552 am: fix typo in fw loading (#10080) nimlgenandGitHub 2025-04-28 09:45:00 +03:00
• e996584685 olmoe in mac benchmark (#10077) chenyuandGitHub 2025-04-27 21:07:02 -04:00
• 732e172961 don't require contiguous after fuse (#10074) George HotzandGitHub 2025-04-27 13:17:22 -04:00
• 1aed04ec12 cpu is ground truth in VALIDATE_WITH_CPU=1 [pr] (#10067) qazalandGitHub 2025-04-27 20:14:21 +03:00
• 129bddde74 lin failure from SINGLE_KERNEL_SOFTMAX (#10073) George HotzandGitHub 2025-04-27 13:02:10 -04:00
• b341296304 hotfix: save sdxl ram geohot 2025-04-27 12:09:45 -04:00
• 68c5f7ba80 load fast in sdxl (#10072) George HotzandGitHub 2025-04-27 11:58:51 -04:00
• 768eb94c3e disable debug for load_state_dict [pr] (#10070) George HotzandGitHub 2025-04-27 11:11:56 -04:00
• 4b8ef6ce78 hotfix: sdxl corealize geohot 2025-04-27 10:41:46 -04:00
• b6d2effaf5 assign is contiguous (#10066) George HotzandGitHub 2025-04-27 08:40:33 -04:00
• 1253819151 make beautiful indexing use a Variable (#10063) George HotzandGitHub 2025-04-27 08:22:38 -04:00
• a13a43c4fe yolo 416 to 640 res (#10047) Rory ClearandGitHub 2025-04-27 01:45:58 +01:00
• 4c1ce1a299 don't simplify if div folding resulted in negative numerator (#10064) chenyuandGitHub 2025-04-26 17:01:18 -04:00
• 75250cac56 green Chen-Yu Yang 2025-04-26 10:01:23 -07:00
• 1805403821 fix rand arange folding (#10060) George HotzandGitHub 2025-04-26 12:24:05 -04:00
• d13c100981 don't sort dims in verify_sink_dims [pr] (#10059) qazalandGitHub 2025-04-26 18:24:30 +03:00
• c80fe6d5fc handle some fancier reduces (#10057) George HotzandGitHub 2025-04-26 11:20:15 -04:00
• e08270c1ba nv: fix program init for no-args kernels (#10058) nimlgenandGitHub 2025-04-26 18:08:53 +03:00
• d99555f0bc mlperf 5.0 Chen-Yu Yang 2025-04-26 14:07:19 +00:00
• 11113c9d07 reduce_unparented (#10056) George HotzandGitHub 2025-04-26 09:48:16 -04:00
• ea5dddc537 reduce collapse generic (#10045) George HotzandGitHub 2025-04-26 09:13:24 -04:00
• 5cdc96409e Update outdated renderer.render calls (#10044) quortusandGitHub 2025-04-26 13:35:19 +02:00
• e055b9422f am: fix mmap failures (#10054) nimlgenandGitHub 2025-04-26 14:21:28 +03:00
• e1d2b64e92 remu new instructions (#10050) qazalandGitHub 2025-04-26 02:04:12 +03:00
• bba5d0a3e4 remu refactors (#10028) qazalandGitHub 2025-04-25 23:31:10 +03:00
• 0fc85a2b0a hcqfuzz: init (#10049) nimlgenandGitHub 2025-04-25 23:19:21 +03:00
• b30050e287 fix amdgpu_disassemble on mac [pr] (#10042) qazalandGitHub 2025-04-25 10:23:11 +03:00
• a197aa4ef3 upat reduce syntax [pr] (#10040) George HotzandGitHub 2025-04-24 22:05:28 -04:00
• 76a86735c0 hotfix amd bf16 is supported case (#10039) Ignacio SicaandGitHub 2025-04-24 21:29:27 -03:00
• b4f823acbe fix helper_tc_allclose (#9606) Ignacio SicaandGitHub 2025-04-24 18:36:40 -03:00
• 3a189fa561 More yolo processing in tinygrad (#9928) 3 people 2025-04-24 21:21:46 +01:00
• 74c6cf8be3 lint mlperf model_train (#10038) chenyuandGitHub 2025-04-24 16:19:44 -04:00
• 51ca19d061 set test_tensor_cores_padded_amd to expectedFailure (#10036) Ignacio SicaandGitHub 2025-04-24 17:11:40 -03:00
• 914d89fa0b fix tensor cores for gfx1201 (#9838) 4 people 2025-04-25 02:57:41 +08:00
• 779aa1e2e9 Enable image tests on cloud if clouddev supports image (#9903) 3 people 2025-04-24 23:30:12 +05:00
• 29a12b19ea Add macos CLOUD tests (#10033) uuuvnandGitHub 2025-04-24 23:14:13 +05:00
• 55942a8d8e [Bounty] moved index_tensor off cpu in torch_backend (#9916) Nishant RajadhyakshaandGitHub 2025-04-24 11:12:37 -07:00
• 373ca59b7f use is_dtype_supported to check dtype support in tc tests (#10035) Ignacio SicaandGitHub 2025-04-24 14:59:14 -03:00
• 93a1e9eeb9 improve bf16 case for is_dtype_supported [pr] (#10034) Ignacio SicaandGitHub 2025-04-24 14:03:57 -03:00
• 754d789f51 Fix and enable jit tests on CLOUD (#10031) uuuvnandGitHub 2025-04-24 20:39:31 +05:00
• 0b482fb824 add RDNA3 parser to remu (#10025) qazalandGitHub 2025-04-24 16:34:07 +03:00
• 0d903c9495 Print clouddev instead of cloudev's renderer (#10023) uuuvnandGitHub 2025-04-24 18:32:08 +05:00
• aec75f51ef fixup some slow CI tests [pr] (#10027) George HotzandGitHub 2025-04-24 14:20:49 +01:00
• c990aac2b1 skip flaky test_transcribe_file1_OOB (#10026) qazalandGitHub 2025-04-24 16:09:43 +03:00
• 4e2ccfddc6 ci refactor to split AMD/NVIDIA [pr] (#10024) George HotzandGitHub 2025-04-24 13:59:54 +01:00
• 0c68e44d6f Cloud properties (#10021) uuuvnandGitHub 2025-04-24 17:17:01 +05:00
• db00d88415 hotfix: handle bad z3 install like z3 import fail geohot 2025-04-24 08:09:40 -04:00
• e75be6eafc [bounty] [pr] index validation with z3 (#9981) 3 people 2025-04-24 14:06:08 +02:00
• 9e49721c47 CPUGraph support for clang (#10014) 3 people 2025-04-24 13:52:35 +02:00
• c3ad7b2a84 create randperm and support pytorch backend (#10019) Park JunandGitHub 2025-04-24 20:29:02 +09:00
• b545338e59 isin_Tensor_out added (#10018) Matthew DaiterandGitHub 2025-04-24 19:26:51 +08:00
• a25abf55e3 retinanet only call postprocess_detections with RUNMLPERF (#10017) chenyuandGitHub 2025-04-23 20:45:38 -04:00
• 7f53e80db9 hotfix: amd mmio on mi300 (#10016) nimlgenandGitHub 2025-04-24 01:08:18 +03:00
• 1c5e353249 am: use mmio iface (#10012) nimlgenandGitHub 2025-04-24 00:27:04 +03:00
• 65faa1d94b explicit device in mlperf scripts (#10015) chenyuandGitHub 2025-04-23 17:11:52 -04:00
• a3f938dbee remove retinanet INITMLPERF from beam script (#10011) chenyuandGitHub 2025-04-23 14:32:54 -04:00
• cc52b9c528 am: add entry() to PT (#10010) nimlgenandGitHub 2025-04-23 20:45:52 +03:00
• c952cb965e amd: use mmio iface (#9997) nimlgenandGitHub 2025-04-23 20:13:00 +03:00
• 5542aeb0e4 RetinaNet MLPerf flag updates (#10009) Francis LataandGitHub 2025-04-23 13:00:34 -04:00
• de0504276b pop 0 is slow [pr] (#10007) George HotzandGitHub 2025-04-23 17:00:59 +01:00
• d3a8d5c128 print postprocess_detections time in retinanet eval (#10005) chenyuandGitHub 2025-04-23 11:39:56 -04:00
• 2ed3acd767 toposort is a function [pr] (#10004) George HotzandGitHub 2025-04-23 16:25:03 +01:00
• 0730ff0e50 Skip test that requires lru if device's allocator isn't lru (#10003) uuuvnandGitHub 2025-04-23 20:12:56 +05:00
• 954cb06957 deepwalk without recursion [pr] (#10002) George HotzandGitHub 2025-04-23 15:57:50 +01:00
• 9de73ccc22 Skip test that requires python 3.12 on older versions (#10001) uuuvnandGitHub 2025-04-23 19:09:26 +05:00
• 71ecc7fa1a use a pattern matcher for upcast [pr] (#10000) George HotzandGitHub 2025-04-23 14:24:23 +01:00
• cc1087d2ec move simplify into views_to_indexed_uops (#9999) George HotzandGitHub 2025-04-23 13:50:27 +01:00
• c39128133c retinanet green scripts (#9996) chenyuandGitHub 2025-04-23 08:28:03 -04:00
• a4a5f2d54a faster block order [pr] (#9998) George HotzandGitHub 2025-04-23 13:11:30 +01:00
• 61bfd23881 update mlperf-logging version (#9995) chenyuandGitHub 2025-04-22 19:32:39 -04:00
• dbbd755cba FP8s truncate (#9937) 3 people 2025-04-23 01:12:49 +02:00
• 58180caad3 schedule linearize small cleanups [pr] (#9994) qazalandGitHub 2025-04-23 00:42:29 +03:00
• f4ec57baff new schedule linearizer enqueues KERNEL UOps [pr] (#9993) qazalandGitHub 2025-04-23 00:17:58 +03:00
• d1f6701eb7 hotfix: lower amd threshold + improve block reorder test geohot 2025-04-22 20:43:22 +01:00
• db51133537 rename HWInterface -> FileIOInterface (#9989) nimlgenandGitHub 2025-04-22 22:18:57 +03:00
• c1539b0319 putting add first orders loads as expected (#9991) George HotzandGitHub 2025-04-22 20:12:05 +01:00
• bd580d8ea4 hcq: use mmio interface in nv (#9986) nimlgenandGitHub 2025-04-22 21:58:12 +03:00
• feee6986c9 faster block reorder (#9990) George HotzandGitHub 2025-04-22 19:18:57 +01:00
• 6cb2d18c03 refactor schedule linearize to defaultdict [pr] (#9984) qazalandGitHub 2025-04-22 19:00:23 +03:00
• 9e5e371999 make DISABLE_COMPILER_CACHE a ContextVar [pr] (#9983) chenyuandGitHub 2025-04-22 10:32:54 -04:00
• bbc324f5dc remove CAST_AFTER_EXPAND (#9980) qazalandGitHub 2025-04-22 16:06:11 +03:00