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

• cae1c41755 test case of softmax backward kernel count (#7022) chenyuandGitHub 2024-10-12 15:46:32 -04:00
• 5ce224ceb3 handle arbitrary multiplication case (#7019) George HotzandGitHub 2024-10-12 23:16:27 +08:00
• 23faeacb23 remove outdated comments (#7018) chenyuandGitHub 2024-10-12 10:51:07 -04:00
• 85a45164fb remove pyint [pr] (#7016) George HotzandGitHub 2024-10-12 22:36:24 +08:00
• 38d45dfba5 hotfix: no rng in test/external/external_benchmark_schedule.py geohot 2024-10-12 22:03:04 +08:00
• ed1ed9e4ff bert use BS=72 (#7015) chenyuandGitHub 2024-10-12 09:41:56 -04:00
• cba4b9a058 clean up ops file [pr] (#7013) George HotzandGitHub 2024-10-12 19:53:52 +08:00
• 746a1f8c86 prep uoping diff for big graph [pr] (#7014) qazalandGitHub 2024-10-12 14:09:32 +03:00
• 334f499e6a consistent render of recip in cuda with CStyleLanguage (#6980) ignaciosicaandGitHub 2024-10-12 07:56:47 -03:00
• a71bb09ec3 remove symbolic file [pr] (#7012) George HotzandGitHub 2024-10-12 18:44:44 +08:00
• 16271189ea hotfix: don't spend lines on a (broken) favicon geohot 2024-10-12 18:20:19 +08:00
• b737ee5bac move to_indexed_uops to uops (#7011) George HotzandGitHub 2024-10-12 18:20:57 +08:00
• 5ae2de9845 UOp.variable (#7010) George HotzandGitHub 2024-10-12 18:20:44 +08:00
• f79e05cac0 add types in all nn/init.py classes (#7002) Bhavya GadaandGitHub 2024-10-11 23:42:14 -07:00
• 2bb6b95e9f refactor _make_hip_code_for_op into pm rules (#7001) ignaciosicaandGitHub 2024-10-12 01:46:22 -03:00
• 5c9f76e274 hotfix: openpilot compile3 compare to i==1 geohot 2024-10-12 09:44:24 +08:00
• 36056e0760 update mlperf systems and copy 4.1 to 5.0 (#7004) chenyuandGitHub 2024-10-11 16:20:34 -04:00
• 8831c691e2 Add slice parameter type checking to disallow Tensor usage for slices (#6967) Markiian NovosadandGitHub 2024-10-11 22:20:21 +02:00
• b0dd407cdd ops_cuda: add optional dynamic smem parameter (#6956) Francis LamandGitHub 2024-10-11 11:51:06 -07:00
• 0e42662f2a log seed at the right place for bert (#7000) chenyuandGitHub 2024-10-11 10:39:40 -04:00
• da42d422f8 add run results and update software versions mlperf_training_v4.1 chenyu 2024-10-11 09:50:42 -04:00
• 5496a36536 update red mlperf bert readme (#6969) nimlgenandGitHub 2024-10-11 13:08:06 +03:00
• feb0bcb58b qcom bench bind to perf cluster (#6996) nimlgenandGitHub 2024-10-11 12:21:52 +03:00
• 7451812bbf delete AST_REWRITE ctx var (#6995) qazalandGitHub 2024-10-11 11:33:16 +03:00
• 7988547df2 start changes from big graph (#6993) qazalandGitHub 2024-10-11 11:13:46 +03:00
• e7a0ffe46a break out linearization [pr] (#6994) George HotzandGitHub 2024-10-11 15:27:33 +08:00
• f319530191 don't track simplify [pr] (#6992) George HotzandGitHub 2024-10-11 15:03:03 +08:00
• e441794c4b remove custom op support, we waste time maintaining this (#6991) George HotzandGitHub 2024-10-11 14:31:09 +08:00
• c08521e823 minor cleanups from toonygrad (#6990) George HotzandGitHub 2024-10-11 14:19:10 +08:00
• f50d0e0ee0 cloud device [pr] (#6964) George HotzandGitHub 2024-10-11 12:24:06 +08:00
• 23c09f4b4c add support for padding='same' in nn.conv (#6975) Bhavya GadaandGitHub 2024-10-10 20:39:07 -07:00
• 54dcea235d viz auto recenter on out of view graph [pr] (#6986) qazalandGitHub 2024-10-11 02:40:06 +03:00
• 159ee04489 include qcom in view_supported_devices (#6985) nimlgenandGitHub 2024-10-11 01:10:51 +03:00
• f9d454aed5 correct kernargs alignment (#6984) nimlgenandGitHub 2024-10-11 00:06:28 +03:00
• 2b17279d4e viz don't default open the browser [pr] (#6983) qazalandGitHub 2024-10-10 22:12:18 +03:00
• 4f60252210 reduce scheduler process replay overhead [pr] (#6981) qazalandGitHub 2024-10-10 20:03:38 +03:00
• 859d6d0407 Fix mypy examples/beautiful_*.py (#6978) 3 people 2024-10-10 17:34:29 +02:00
• 4ef5310039 track viz context even if rewrite errors [pr] (#6976) qazalandGitHub 2024-10-10 18:33:15 +03:00
• 592e5f1df2 skip test_viz test_no_dedup_different_opts (#6979) chenyuandGitHub 2024-10-10 11:10:24 -04:00
• e3dc10f8f6 improve fold_unrolled_divs (#6977) chenyuandGitHub 2024-10-10 10:52:05 -04:00
• 3481468702 bring viz to core (#6970) qazalandGitHub 2024-10-10 16:56:26 +03:00
• fad575ec76 qcom tiny cleanups (#6973) nimlgenandGitHub 2024-10-10 12:26:41 +03:00
• 3724a66716 move test_viz to test/, prereq for tinygrad/viz [pr] (#6972) qazalandGitHub 2024-10-10 11:40:46 +03:00
• 960c495755 added beautiful fashion mnist and example (#6961) 3 people 2024-10-10 00:01:07 -04:00
• b5546912e2 10% more TRAIN_STEPS for bert (#6971) chenyuandGitHub 2024-10-09 19:21:43 -04:00
• f90d8493cc add HCQDEV_WAIT_TIMEOUT_MS (#6968) nimlgenandGitHub 2024-10-09 19:50:00 +03:00
• 35cf48659b limit beam param for bert on green (#6966) chenyuandGitHub 2024-10-09 11:48:18 -04:00
• 0e8bcda07e get readable error from wait_check (#6965) 3 people 2024-10-09 22:28:58 +08:00
• 20d3c2d113 unify UOps.SHAPETRACKER and UOps.SWIZZLE with UOps.VIEW (#6955) qazalandGitHub 2024-10-08 21:00:17 +03:00
• 137ad5519f amd fix cwsr for gfx11 (#6950) nimlgenandGitHub 2024-10-08 17:44:29 +03:00
• 0d526e251e nv sync on gpu before local update (#6954) nimlgenandGitHub 2024-10-08 17:43:58 +03:00
• 2800520dd5 even smaller process_replay.py [pr] (#6941) qazalandGitHub 2024-10-08 15:43:22 +03:00
• 851f39653a rename to BUFFER_VIEW + MetaOps cleanup (#6953) qazalandGitHub 2024-10-08 15:09:22 +03:00
• 1ff2c98f8a fix logfile name for bert red (#6952) chenyuandGitHub 2024-10-08 05:37:52 -04:00
• 08bfa8632b embedding shape (#6930) czhuandGitHub 2024-10-07 23:42:20 -07:00
• 20a9683403 Make self.fd Optional. (#6855) vladovandGitHub 2024-10-07 22:25:34 -07:00
• a78c96273a update bert epoch logging (#6940) chenyuandGitHub 2024-10-08 00:34:06 -04:00
• 0498e846a5 break out metaops (#6948) George HotzandGitHub 2024-10-08 12:08:54 +08:00
• 42609300ff hcq no timeline signals in init (#6944) nimlgenandGitHub 2024-10-07 23:36:19 +03:00
• 0ecc417dd2 prep for viz move to core [pr] (#6938) qazalandGitHub 2024-10-07 18:24:04 +03:00
• e4c0743188 failed example for logcumsumexp (#6936) chenyuandGitHub 2024-10-07 10:55:45 -04:00
• 102dfe5510 back to 2**10 for bert loss scaler (#6934) chenyuandGitHub 2024-10-07 10:17:21 -04:00
• 9250452da4 no codegen import in ops [pr] (#6888) qazalandGitHub 2024-10-07 15:54:21 +03:00
• f7f94cd62f bitcast cleanup [pr] (#6933) George HotzandGitHub 2024-10-07 19:16:16 +08:00
• 0cf815a93a bert use BS=66 and update hparams (#6932) chenyuandGitHub 2024-10-07 05:08:27 -04:00
• 32ac24c45c Generic wmma rendering for cuda, ptx [run_process_replay] (#6838) ignaciosicaandGitHub 2024-10-07 05:36:36 -03:00
• b82023c97e process replay cleanup to generic _pmap [pr] (#6929) qazalandGitHub 2024-10-07 08:57:05 +03:00
• 16312b4c59 rip out old scheduler process replay stuff, diff pure UOps [pr] (#6927) qazalandGitHub 2024-10-07 08:20:35 +03:00
• 999e3780e9 dropout contiguous after >= p (#6892) chenyuandGitHub 2024-10-06 19:40:42 -04:00
• 9eb6eef441 seed in tensor (#6869) wozeparrotandGitHub 2024-10-06 14:46:58 -04:00
• f9e32f2bb2 clip device fix (#6924) Tobias FischerandGitHub 2024-10-06 12:47:32 -04:00
• 01a2d7316d dtype=float in bert log_softmax for loss and accuracy (#6916) chenyuandGitHub 2024-10-06 11:15:56 -04:00
• 19a7e41113 implement logcumsumexp (#6921) jeffzh4ngandGitHub 2024-10-06 10:45:36 -04:00
• f588169fdc hotfix: ad for DEBUG=2 in the mnist tutorial geohot 2024-10-06 21:04:06 +08:00
• 10ff1d6fb9 viz prep refactor for tracked scope decorator [pr] (#6920) qazalandGitHub 2024-10-06 16:02:09 +03:00
• 837f9c6832 new viz fuzz tests, track multiple contexts (#6913) qazalandGitHub 2024-10-06 14:58:15 +03:00
• 75d9dcf000 support dtype in softmax and log_softmax (#6914) chenyuandGitHub 2024-10-06 07:18:15 -04:00
• 718b959349 log epoch start and stop for bert (#6912) chenyuandGitHub 2024-10-06 06:39:46 -04:00
• b066ef2282 small changes from the viz_rewrite branch [pr] (#6907) qazalandGitHub 2024-10-06 12:00:55 +03:00
• 16c1fa4208 use BEAM=3 for red box bert runs (#6904) chenyuandGitHub 2024-10-05 09:21:12 -04:00
• 0e706227a2 add seed to bert result log filename (#6903) chenyuandGitHub 2024-10-05 09:15:24 -04:00
• 8ed3a00c9c ceildiv helper [pr] (#6899) George HotzandGitHub 2024-10-05 14:59:10 +08:00
• fd68b6dbc2 type annotation to round_up (#6898) chenyuandGitHub 2024-10-04 23:27:23 -04:00
• 3c12244cfc remove DTypeLike from lazy (#6897) chenyuandGitHub 2024-10-04 22:49:21 -04:00
• 0d6216aba1 bump the download cache (#6896) George HotzandGitHub 2024-10-05 10:23:18 +08:00
• 4058a99275 symbolic in ops 2 [pr] (#6895) George HotzandGitHub 2024-10-05 10:20:07 +08:00
• 08414d7b7c cleanup test_uop_symbolic.py (#6894) chenyuandGitHub 2024-10-04 20:53:10 -04:00
• 555bcb5e54 static access for code_for_op (#6889) ignaciosicaandGitHub 2024-10-04 20:38:01 -03:00
• 5f6b6162b3 Suppress warnings in transcendental tests. (#6891) vladovandGitHub 2024-10-04 16:37:17 -07:00
• 707c805a68 nv set localmem sm count to max (#6890) nimlgenandGitHub 2024-10-04 23:29:46 +03:00
• 4df5c7a4ef move lazy to engine [pr] (#6886) George HotzandGitHub 2024-10-04 23:19:26 +08:00
• 6b063450df move hcq device to runtime [pr] (#6879) George HotzandGitHub 2024-10-04 22:26:50 +08:00
• 5be2bd18a6 use UOps.BIND instead of ASSIGN, it's different (#6885) George HotzandGitHub 2024-10-04 22:26:33 +08:00
• 4c3895744e type annotation for layernorm (#6883) chenyuandGitHub 2024-10-04 09:03:56 -04:00
• 8ca506ee37 remove the magic methods for moving between devices [pr] (#6881) George HotzandGitHub 2024-10-04 20:27:52 +08:00
• 7c8849010a fix var_vals in MCTS (#6882) chenyuandGitHub 2024-10-04 08:19:35 -04:00
• a0cb16ac61 node cleanup + local metal test speed [pr] (#6880) George HotzandGitHub 2024-10-04 18:14:23 +08:00
• cdff1d75b6 things that are only used in one place don't belong in helpers [pr] (#6878) George HotzandGitHub 2024-10-04 17:27:38 +08:00
• f4ec39fe58 switch symbolic from old to uops, final PR (#6872) George HotzandGitHub 2024-10-04 16:42:27 +08:00
• 738a5794a9 last update for new symbolic [pr] (#6877) George HotzandGitHub 2024-10-04 14:58:51 +08:00