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

• a3652e6ddc minor cleanups to test_ops (#3290) chenyuandGitHub 2024-01-31 19:01:25 -05:00
• 77251336d5 fix handcode_resnet50_opt.py (#3289) chenyuandGitHub 2024-01-31 19:01:08 -05:00
• 9b8c1a0408 Tensor.batchnorm works more than 2d and reuse in onnx (#3284) chenyuandGitHub 2024-01-30 19:02:45 -05:00
• 7816c3b692 onnx update for trilu and argmax (#3283) chenyuandGitHub 2024-01-30 18:39:16 -05:00
• 5b46b0ff3d Simple RDNA3 emulator (#2974) 3 people 2024-01-30 20:39:28 +02:00
• 247a8a2a6c add canonicalization to View.create (#3280) George HotzandGitHub 2024-01-30 10:26:48 -08:00
• d8f6280ffb hotfix: add CHECK_NEQ to fuzz_shapetracker_math geohot 2024-01-30 10:07:54 -08:00
• 09f2952dc3 reintroduce merge views in update benchmark (#3279) George HotzandGitHub 2024-01-30 09:47:20 -08:00
• d298916232 Revert "take merge views from corsix branch" (#3278) George HotzandGitHub 2024-01-30 09:34:28 -08:00
• b57a16aa89 take merge views from corsix branch (#3273) George HotzandGitHub 2024-01-30 09:25:16 -08:00
• 6a4a5dc79d fix pad 0 size (#3277) George HotzandGitHub 2024-01-30 08:58:10 -08:00
• b0a755288f cifar EVAL_BS set default value to BS (#3274) chenyuandGitHub 2024-01-29 17:37:12 -05:00
• 861d5ac224 wmma: fix the upcasts after WMMA to be hcopt ordering invariant (#3250) Francis LamandGitHub 2024-01-29 11:51:57 -08:00
• af4ca85594 MultiLazyBuffer.reshape new_axis without real_strides (#3272) chenyuandGitHub 2024-01-28 23:53:52 -05:00
• 34c7621556 HIP=1 NOCLANG=1 for tinybox external_model_benchmark (#3270) chenyuandGitHub 2024-01-28 22:05:26 -05:00
• 085dc87bed winograd should be 4 kernels (#3268) George HotzandGitHub 2024-01-28 09:21:26 -08:00
• f48b6aca77 long running beam pool (#3267) George HotzandGitHub 2024-01-28 08:06:03 -08:00
• 9e17378b60 Fix metal tests (#3266) George HotzandGitHub 2024-01-27 18:09:42 -08:00
• 86748f4a8c fix bbox format to be a list (#3265) Francis LataandGitHub 2024-01-27 20:54:19 -05:00
• 67a78615e5 uoptimizer (#3262) George HotzandGitHub 2024-01-27 10:26:04 -08:00
• 3ae811af21 tests for Tensor init data dtype and resulting dtype (#3247) 3 people 2024-01-27 10:13:42 +02:00
• 3c728d1082 compiler support (#3260) George HotzandGitHub 2024-01-26 23:36:40 -08:00
• 4273aabe31 extra/gemm: add a simple_conv.py along with correctness check (#3236) Francis LamandGitHub 2024-01-26 19:06:57 -08:00
• 0aad8d238b rebuild ocelot (#3259) George HotzandGitHub 2024-01-26 18:46:36 -08:00
• 473935125a use comgr to compile (#3248) George HotzandGitHub 2024-01-26 18:27:49 -08:00
• c4d870db0d fix jit realize issue (#3258) George HotzandGitHub 2024-01-26 18:27:35 -08:00
• 4197ef17c4 const cleanup with dtype.Scalar (#3257) chenyuandGitHub 2024-01-26 21:16:22 -05:00
• 03a6bc59c1 move autogen to runtime/autogen (#3254) George HotzandGitHub 2024-01-26 12:44:19 -08:00
• a3869ffd46 move gpuctypes in tree (#3253) George HotzandGitHub 2024-01-26 12:25:03 -08:00
• bc92c4cc32 onnx Einsum, CumSum, DepthToSpace, SpaceToDepth (#3252) chenyuandGitHub 2024-01-26 10:47:53 -05:00
• e45ffdb6cf cleanup onnx (#3249) chenyuandGitHub 2024-01-25 20:39:59 -05:00
• 168b1f879c Fix hip_matmul gemm in extra (#3241) Ahmed HarmoucheandGitHub 2024-01-26 01:03:04 +01:00
• 7feeb118e6 hip launch speed (#3246) George HotzandGitHub 2024-01-25 15:13:55 -08:00
• cb372b053f add device speed test (#3244) George HotzandGitHub 2024-01-25 12:01:22 -08:00
• d0e116c6d6 fix maximum/where Scalar casting (#3194) geohotstanandGitHub 2024-01-26 01:26:04 +08:00
• 3628bea910 fix: big round even rounder round (#3242) geohotstanandGitHub 2024-01-26 01:24:15 +08:00
• da5e27968c failed test cases for Tensor.round (#3240) chenyuandGitHub 2024-01-25 02:12:50 -05:00
• b0b5eba535 fix _round in onnx_ops to look more like new Tensor.round (#3239) geohotstanandGitHub 2024-01-25 14:18:58 +08:00
• aa0d1b6330 hotfix: don't use noqa: E702 that's just dumb geohot 2024-01-24 20:01:00 -08:00
• b92945c98d hotfix: DEBUG >= 2 for kernels geohot 2024-01-24 23:55:17 +00:00
• a8fbb03438 minor hip cleanups (#3237) George HotzandGitHub 2024-01-24 15:13:38 -08:00
• 3205fd8481 fix cuda device var rewrite (#3233) nimlgenandGitHub 2024-01-25 00:57:49 +03:00
• ed8a32722a hip mutex signal (#3234) George HotzandGitHub 2024-01-24 13:23:09 -08:00
• 47f9887ce4 hip events work (#3229) George HotzandGitHub 2024-01-24 11:49:53 -08:00
• de7a3a56ff save lines in llvm (#3231) George HotzandGitHub 2024-01-24 11:40:53 -08:00
• 83d614295e reduce lines (#3230) George HotzandGitHub 2024-01-24 10:35:59 -08:00
• afeadbedc9 touch up Tensor.round and Tensor.neg (#3228) chenyuandGitHub 2024-01-24 12:29:37 -05:00
• 0e103b4aa0 implement Tensor.round (#3225) Obada KhaliliandGitHub 2024-01-24 18:49:17 +02:00
• 842053873d fix neg logical_not inconsistencies (#3222) geohotstanandGitHub 2024-01-25 00:48:40 +08:00
• e2e4632aea LoadOps SYNC (#3223) George HotzandGitHub 2024-01-23 21:59:18 -08:00
• 2f4b3ab1c0 shard and to should preserve requires_grad (#3224) chenyuandGitHub 2024-01-24 00:15:10 -05:00
• 23b084e70a add device name to device, all are constructed (#3221) George HotzandGitHub 2024-01-23 20:34:56 -08:00
• 91a1b2bd7a the runner does the build (#3220) George HotzandGitHub 2024-01-23 18:45:43 -08:00
• 9e5409be6c cifar move GlobalCounters.reset() before shard (#3217) chenyuandGitHub 2024-01-23 16:07:43 -05:00
• 595d05a250 test: fix test_linearizer to use the correct tc_dims (#3218) Francis LamandGitHub 2024-01-23 13:07:31 -08:00
• 3c179cc27c cifar only shuffle data at epoch start (#3216) chenyuandGitHub 2024-01-23 14:41:22 -05:00
• 4a07ea355d buffer options should work (#3211) George HotzandGitHub 2024-01-22 19:23:55 -08:00
• a06f34ae42 remove dead lines from cstyle (#3212) George HotzandGitHub 2024-01-22 18:59:19 -08:00
• 8465938d29 minor hlb_cifar cleanups (#3208) chenyuandGitHub 2024-01-22 12:38:39 -05:00
• 3378625773 name upcast variables (#3200) David HouandGitHub 2024-01-22 08:37:28 -08:00
• 827b7a3c64 cleanup pad_reflect and make_square_mask in hlb_cifar (#3206) chenyuandGitHub 2024-01-22 11:30:46 -05:00
• 99884f4c98 cifar flags for RANDOM_CROP, RANDOM_FLIP, and CUTMIX (#3204) chenyuandGitHub 2024-01-22 01:12:51 -05:00
• 53afec2841 add HALF to handcode_resnet50_opt.py (#3202) chenyuandGitHub 2024-01-21 23:03:59 -05:00
• 836883fedc comment out cutmix in hlb_cifar (#3201) chenyuandGitHub 2024-01-21 22:24:53 -05:00
• e6c71f1b26 fix device of Tensor.arange inside Tensor.one_hot (#3199) chenyuandGitHub 2024-01-21 21:03:50 -05:00
• f7d1c42239 cleanup noop prefixes in _pool (#3198) chenyuandGitHub 2024-01-21 20:03:32 -05:00
• 640e5c36ad Fix metal tests broken by 3f56d1a (#3196) uuuvnandGitHub 2024-01-21 18:53:25 +02:00
• b9d27636aa cleanup test_ops.py (#3192) chenyuandGitHub 2024-01-20 20:08:56 -05:00
• 3f56d1a5e8 add operator.lt and operator.eq to test_dtype_alu (#3191) chenyuandGitHub 2024-01-20 14:54:02 -05:00
• c4b5661146 fuzz length for multitensor reduce test case (#3190) chenyuandGitHub 2024-01-20 00:44:38 -05:00
• fdb1c2b1d9 move reduce over 0 len axis logic to lazy.py (#3188) chenyuandGitHub 2024-01-20 00:13:03 -05:00
• 485332935e ring copy example (#3185) chenyuandGitHub 2024-01-19 23:34:30 -05:00
• 254a7372fe buffer copy refactor (#3187) George HotzandGitHub 2024-01-19 20:21:24 -08:00
• fb4bd2a57d reenable padto to search action (#3183) chenyuandGitHub 2024-01-19 14:17:53 -05:00
• cb4cfc078a parameterize multitensor tests for reduce (#3181) chenyuandGitHub 2024-01-19 14:03:01 -05:00
• 5097d5b808 fix padto when with late reduce (#3180) nimlgenandGitHub 2024-01-19 22:01:44 +03:00
• 729a01bf3e complex PRs will not be merged geohot 2024-01-19 10:58:47 -08:00
• f87ecbb0f3 fuzzer validates outputs + (partially) oob accesses (#3178) nimlgenandGitHub 2024-01-19 21:34:51 +03:00
• b2571d586c hypothesis.st -> hypothesis.strat (#3179) chenyuandGitHub 2024-01-19 11:55:26 -05:00
• c4faedebf3 add test cases for negative entry max allreduce (#3177) chenyuandGitHub 2024-01-18 22:26:51 -05:00
• ab1b7c4d09 fix allreduce for max (#3175) chenyuandGitHub 2024-01-18 20:25:35 -05:00
• c51c90bcd4 more sync in transfer (#3174) George HotzandGitHub 2024-01-18 17:17:03 -08:00
• 28dcbf0e00 test case sharded batchnorm has different ast on devices (#3172) chenyuandGitHub 2024-01-18 18:12:15 -05:00
• a60d50487d disable padto, seems to have bug in gpt2 (#3173) chenyuandGitHub 2024-01-18 18:09:30 -05:00
• c80884884e event driven hip (#3160) George HotzandGitHub 2024-01-18 14:35:18 -08:00
• d2aab65958 remove unused expr node (#3170) George HotzandGitHub 2024-01-18 14:18:43 -08:00
• 097b1390ec touchup test_indexing (#3169) chenyuandGitHub 2024-01-18 14:32:43 -05:00
• a04e4d0442 inline clang renderer (#3168) George HotzandGitHub 2024-01-18 11:17:34 -08:00
• efbe4788d1 indexing: Final cleanup (#3156) geohotstanandGitHub 2024-01-19 03:08:03 +08:00
• e139ae550d smaller limit_dims_to_max (#3167) chenyuandGitHub 2024-01-18 13:02:20 -05:00
• 992067399e clean up exceptions in __del__ everywhere (#3165) nimlgenandGitHub 2024-01-18 19:34:09 +03:00
• 0338903429 Update kits19.py (#3166) Max-WeandGitHub 2024-01-18 17:33:50 +01:00
• 67bc2ddfd8 JIT cleanups (#3164) George HotzandGitHub 2024-01-17 23:39:57 -08:00
• f0c178b7e9 move get_contraction to helpers (#3162) George HotzandGitHub 2024-01-17 19:13:11 -08:00
• e52a609240 make WINO a context var, and LATEWINO in hlb_cifar (#3161) chenyuandGitHub 2024-01-17 20:21:26 -05:00
• ee83505fcc fix test extra issue (#3159) George HotzandGitHub 2024-01-17 11:58:08 -08:00
• 9cc2577a08 use hip events (#3157) George HotzandGitHub 2024-01-17 10:39:57 -08:00
• 1b508e0f71 fix fuzz_linearizer toCPU to as_buffer (#3158) chenyuandGitHub 2024-01-17 13:18:46 -05:00
• 743b36f0ce hotfix: copy size is in bytes geohot 2024-01-17 16:44:15 +00:00
• 2e6162b281 graph cleanup (#3155) George HotzandGitHub 2024-01-16 20:57:31 -08:00