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

• 1ccd987e6a simpler tc permaxis in fixup_ast.fix_st [run_process_replay] (#5502) chenyuandGitHub 2024-07-15 21:35:32 -04:00
• 9d4c3c553c prepare expand to support multiexpand [run_process_replay] (#5503) George HotzandGitHub 2024-07-15 18:21:24 -07:00
• fd43d33b7d shave some lines from transcend math [run_process_replay] (#5500) chenyuandGitHub 2024-07-15 21:02:24 -04:00
• 63990705b5 test kernel opts case for 4 local and 4 groups (#5499) chenyuandGitHub 2024-07-15 20:09:38 -04:00
• 13e200b437 add strict mkdocs check (#5497) Alessandro BenettiandGitHub 2024-07-15 23:21:37 +02:00
• 8dfd11c1d8 docs: hcq add types (#5495) nimlgenandGitHub 2024-07-15 22:14:48 +03:00
• aab1e8c6dc uniform init to match torch (#5494) George HotzandGitHub 2024-07-15 12:07:44 -07:00
• 338b7590b9 hotfix: docs for BatchNorm geohot 2024-07-15 12:04:17 -07:00
• c9ec7ce070 start hcq docs (#5411) nimlgenandGitHub 2024-07-15 21:31:11 +03:00
• 9a7d5a148e move colorize_float to helpers.py (#5490) Edward WangandGitHub 2024-07-15 14:29:03 -04:00
• a347d91e0e remove outdated thread local aliases (#5493) P4ssengerandGitHub 2024-07-15 15:28:11 -03:00
• ac08f0eb00 reshape rawbufs in test_linearizer (#5492) qazalandGitHub 2024-07-16 00:14:38 +08:00
• ae4cb7994e run process replay with DEBUG=0 (#5491) qazalandGitHub 2024-07-15 21:30:57 +08:00
• e219103677 Add Pad to Pooling (#5488) Tobias FischerandGitHub 2024-07-15 00:50:20 -04:00
• eef43c9f49 include dims in kernel/nv invalid err msg (#5487) chenyuandGitHub 2024-07-14 22:51:30 -04:00
• c80801c266 len(full_shape)-ki.upcasted -> first_upcasted (#5485) chenyuandGitHub 2024-07-14 20:21:18 -04:00
• 5849130cbb gather negative dim fix (#5486) Tobias FischerandGitHub 2024-07-14 20:20:53 -04:00
• 3c378efcb6 process replay docs improvements (#5481) qazalandGitHub 2024-07-15 05:09:28 +08:00
• 613a1dbeed render lidx starting with 0 (#5478) 3 people 2024-07-14 16:34:04 -04:00
• 671779f280 limit process replay diff to ~20% of kernels (#5480) 3 people 2024-07-15 04:10:08 +08:00
• f8a47608cc test dtype.min and dtype.max (#5479) chenyuandGitHub 2024-07-14 15:31:37 -04:00
• a9f5a764dc make BatchNorm work for 2D and 3D (#5477) George HotzandGitHub 2024-07-14 11:39:58 -07:00
• e41ab66653 use is to compare types (#5476) chenyuandGitHub 2024-07-14 14:26:41 -04:00
• aade18d20c beautiful_mnist in torch geohot 2024-07-14 11:09:58 -07:00
• 604fb60143 docs: fix link to jit in env_vars (#5474) nimlgenandGitHub 2024-07-14 16:08:16 +03:00
• 61822d1a14 nv fix timeline signal rollover on copy queue (#5473) nimlgenandGitHub 2024-07-14 16:06:12 +03:00
• 8835d6c49a cleanup nv/amd program (#5449) nimlgenandGitHub 2024-07-14 14:08:35 +03:00
• 0b3a34e3b1 vectorize folding [run_process_replay] (#5470) qazalandGitHub 2024-07-14 14:41:48 +08:00
• cdf63e41bf mnist mlx example uses compile to be fair to tinyjit geohot 2024-07-13 18:14:45 -07:00
• 8940530290 add mlx beautiful_mnist example geohot 2024-07-13 17:55:47 -07:00
• 28972418c4 s/get_linearizer/get_kernel [run_process_replay] (#5467) chenyuandGitHub 2024-07-13 20:32:22 -04:00
• 0345577032 UNet3D dataloader shared memory fix (#5465) Francis LataandGitHub 2024-07-13 20:26:00 -04:00
• ef578b4de8 new UOp style patterns [run_process_replay] (#5444) Carson PowersandGitHub 2024-07-13 20:21:08 -04:00
• 942c58be90 BEAM_COMPARE=2 validates the correctness of BEAM kernels (#5458) George HotzandGitHub 2024-07-13 13:53:43 -07:00
• 6943ea5f29 nv remove copy_from_cpu command (#5459) nimlgenandGitHub 2024-07-13 23:08:49 +03:00
• 67f70cef02 amd better allocation error messages (#5462) nimlgenandGitHub 2024-07-13 22:55:09 +03:00
• 2427f149a3 threefry as pattern matcher (#5371) wozeparrotandGitHub 2024-07-13 18:59:03 +00:00
• 487ceff825 hotfix: ASSERT_PROCESS_REPLAY sometimes doesn't exist (#5456) qazalandGitHub 2024-07-14 02:15:40 +08:00
• de6ab56458 clean up transcend math with uop syntactic sugar [run_process_replay] (#5455) chenyuandGitHub 2024-07-13 14:00:14 -04:00
• 40ec9410f9 simpler process replay (#5452) qazalandGitHub 2024-07-14 00:55:06 +08:00
• d2933d3548 simplify transcend math [run_process_replay] (#5454) chenyuandGitHub 2024-07-13 12:43:31 -04:00
• 23b907efbb restore process replay runs by their id (#5453) qazalandGitHub 2024-07-14 00:32:34 +08:00
• b8c9298164 verify_lazyop in for WMMA and group_for_reduces (#5448) qazalandGitHub 2024-07-13 23:06:19 +08:00
• 955e1179fb move compile tests and merge (#5451) George HotzandGitHub 2024-07-13 08:04:46 -07:00
• e638b0084f smaller multitensor resnet test (#5450) George HotzandGitHub 2024-07-13 07:31:28 -07:00
• 03c3b14cc2 docs: addded JIT description to dos/env_vars.md (#5445) Simone MargaritelliandGitHub 2024-07-13 16:07:11 +02:00
• bb1a9ebf78 run process replay in parallel (#5443) qazalandGitHub 2024-07-13 16:29:36 +08:00
• 3ebf569f04 relax fuzz transend math threshold a bit (#5442) chenyuandGitHub 2024-07-13 03:31:21 -04:00
• e398734890 fuzz test transcend math (#5383) chenyuandGitHub 2024-07-13 01:54:52 -04:00
• 3a7262d923 [Patch] Fixed an invaild value of fp64 xlog(DBL_MIN) (#5441) hiketteiandGitHub 2024-07-13 14:11:53 +09:00
• 90f0e2fc49 db in wal mode (#5388) wozeparrotandGitHub 2024-07-13 03:43:36 +00:00
• 414aa6ee98 minor improvments to matcher speed [run_process_replay] (#5439) George HotzandGitHub 2024-07-12 20:41:41 -07:00
• 4df63da190 clean up rest of the loadop [run_process_replay] (#5440) chenyuandGitHub 2024-07-12 23:38:51 -04:00
• 0795139f30 Fix TRANSCENDENTAL=2 fp64 sin (#5385) 3 people 2024-07-13 12:15:04 +09:00
• fb3011ac61 improve matcher speed [run_process_replay] (#5438) George HotzandGitHub 2024-07-12 20:02:19 -07:00
• 03c2dc8bd7 lowerer is kernel [run_process_replay] (#5437) George HotzandGitHub 2024-07-12 18:50:55 -07:00
• b8342fb085 independent lowerer [run_process_replay] (#5434) George HotzandGitHub 2024-07-12 18:08:43 -07:00
• 9a187e6102 fix handcode_opt script (#5435) chenyuandGitHub 2024-07-12 20:52:28 -04:00
• b80fd7d23c allow benchmarking forward only (#5436) wozeparrotandGitHub 2024-07-13 00:37:49 +00:00
• 00813a92a0 update Tensor.eye api to match torch (#5433) chenyuandGitHub 2024-07-12 20:25:12 -04:00
• cddfd8e25d bugfix: group for reduce should check all dimensions (#5431) George HotzandGitHub 2024-07-12 17:02:40 -07:00
• fbaf040baf compute full_shape from LazyOp [run_process_replay] (#5429) George HotzandGitHub 2024-07-12 16:47:08 -07:00
• 870dc8c350 s/Linearizer/Lowerer [run_process_replay] (#5428) George HotzandGitHub 2024-07-12 15:54:07 -07:00
• 6707c778d0 scheduleitem is not Tuple [run_process_replay] (#5425) George HotzandGitHub 2024-07-12 15:13:19 -07:00
• 4cd1de038a smaller reshape_and_permute arg in shift_to (#5426) chenyuandGitHub 2024-07-12 17:46:48 -04:00
• 94599c0637 fixup ast in kernel to be MetaOps.SINK [run_process_replay] (#5424) George HotzandGitHub 2024-07-12 14:01:03 -07:00
• b055ece550 hotfix: bump to cache gpuocelot George Hotz 2024-07-12 13:54:14 -07:00
• d37056f3b1 pass Renderer.global_max / local_max into get_grouped_dims (#5423) chenyuandGitHub 2024-07-12 16:49:27 -04:00
• 4aefb1595d MetaOps.SINK [run_process_replay] (#5422) George HotzandGitHub 2024-07-12 13:37:30 -07:00
• f6ef283e6a s/loadops/metaops [run_process_replay] (#5421) George HotzandGitHub 2024-07-12 13:26:50 -07:00
• f4944ced09 tiny amd cleanups (#5420) nimlgenandGitHub 2024-07-12 22:54:42 +03:00
• b17e4adb3a add -c advice.detachedHead=false to process replay git checkout (#5419) chenyuandGitHub 2024-07-12 15:13:26 -04:00
• d1cbd6bb95 unity handcode_resnet_opt and handcode_bert_opt (#5418) wozeparrotandGitHub 2024-07-12 19:05:01 +00:00
• a0dbe20dbd skip some redundant and slow tests in ci (#5416) chenyuandGitHub 2024-07-12 14:43:13 -04:00
• 76125c07be make some grouped_dim test work (#5415) chenyuandGitHub 2024-07-12 14:22:50 -04:00
• b7cc75a9df usage summary in handcode opt (#5414) wozeparrotandGitHub 2024-07-12 18:21:18 +00:00
• 3cb94a0a15 Rename tinygrad/runtime/driver to support (#5413) uuuvnandGitHub 2024-07-12 21:06:42 +03:00
• 6604d2b2c3 amd/nv respect visible devs (#5409) nimlgenandGitHub 2024-07-12 20:02:12 +03:00
• b18aa00bba refactor: consolidate replace [run_process_replay] (#5403) Roelof van DijkandGitHub 2024-07-12 16:36:57 +02:00
• 497274f663 add float64 to test_dtype_alu dtypes_float (#5410) chenyuandGitHub 2024-07-12 10:21:32 -04:00
• 31fcc516dc more process replay tooling (#5407) qazalandGitHub 2024-07-12 18:11:34 +08:00
• 6ec7dbc287 ci: parallelize uops tests (#5405) Roelof van DijkandGitHub 2024-07-12 10:22:41 +02:00
• e22b377839 generalize FUSE_AS_ONE_KERNEL in the scheduler (#5397) qazalandGitHub 2024-07-12 15:23:16 +08:00
• 6e0a523078 repro slow resnet kernel with 4 global dims (#5402) chenyuandGitHub 2024-07-11 23:31:15 -04:00
• 8390feb7b9 optim.OptimizerGroup in hlb_cifar (#5401) George HotzandGitHub 2024-07-11 20:14:36 -07:00
• 01fbd18209 metal compile fail geohot 2024-07-11 19:27:05 -07:00
• 3a2b5a75d2 improve single kernel indexing (#5398) 3 people 2024-07-11 19:00:57 -07:00
• c24d495ef9 metadata in handcode_opt (#5400) wozeparrotandGitHub 2024-07-12 00:45:34 +00:00
• c60838594c metadata in graph (#5399) wozeparrotandGitHub 2024-07-12 00:02:12 +00:00
• c2da4454cd indexing getting better (#5389) George HotzandGitHub 2024-07-11 16:41:51 -07:00
• 9712d9ffb6 pass lowering errors if not asserting process replay (#5395) qazalandGitHub 2024-07-12 07:09:12 +08:00
• a02b38c0ac download openimages by running it (#5396) wozeparrotandGitHub 2024-07-11 23:06:13 +00:00
• 0421f5d83e hotfix: compare test_var_multireduce against numpy (#5394) qazalandGitHub 2024-07-12 06:57:08 +08:00
• b91a0ccdc3 make [run_process_replay] [no_assert] the default (#5390) qazalandGitHub 2024-07-12 03:36:59 +08:00
• e8191479a3 add bigint type for indexing [run_process_replay] (#5387) George HotzandGitHub 2024-07-11 11:37:10 -07:00
• 5232e405ce hotfix: add BS to beautiful_mnist geohot 2024-07-11 10:55:05 -07:00
• 3e40211e45 add UOP_IS_SYMBOLIC [run_process_replay] [no_assert] (#5386) George HotzandGitHub 2024-07-11 10:48:45 -07:00
• b3790b759b nv cleanup gpfifo setup (#5382) nimlgenandGitHub 2024-07-11 17:50:52 +03:00
• 416f838a1a hotfix tqdm respects total=0 if set (#5380) chenyuandGitHub 2024-07-11 10:30:12 -04:00
• 2ba96d4c29 nv use mv_address (#5381) nimlgenandGitHub 2024-07-11 16:45:03 +03:00