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

• 6755a9254f constant fold pattern match (#3696) George HotzandGitHub 2024-03-12 08:48:07 -07:00
• dd1a1c12df rocm path in autogen (#3697) nimlgenandGitHub 2024-03-12 14:06:43 +03:00
• 971d7f5d7c O(n) arange attempt (#3530) 4 people 2024-03-11 19:09:20 -04:00
• a5d023dff8 reciprocal mlop (#3694) George HotzandGitHub 2024-03-11 16:08:46 -07:00
• 3af1c1051a Revert "bring reciprocal back (#3687)" (#3692) George HotzandGitHub 2024-03-11 15:55:14 -07:00
• ef44c8959b Revert "rewrite recip to div (#3690)" (#3691) George HotzandGitHub 2024-03-11 15:54:58 -07:00
• 2b089bfd18 rewrite recip to div (#3690) George HotzandGitHub 2024-03-11 15:52:24 -07:00
• aec4c4f01b linearizer ast as a tuple of lazyops (#3689) qazalandGitHub 2024-03-12 00:39:04 +02:00
• d0bcc9a66b replace all if dim < 0: dim += self.ndim with _resolve_dim (#3688) chenyuandGitHub 2024-03-11 17:33:36 -04:00
• bcf6fbd3b2 bring reciprocal back (#3687) George HotzandGitHub 2024-03-11 14:19:54 -07:00
• 9f13960f72 search: catch RuntimeError when timing acted_lins (#3664) Francis LamandGitHub 2024-03-11 13:14:03 -07:00
• 490c5a3ec3 examples/stable_diffusion: support model checkpoints without alphas_cumprod key (#3681) 3 people 2024-03-11 20:05:52 +00:00
• 3219a527d6 search: add a tool that beam searches one or more kernels (#3685) Francis LamandGitHub 2024-03-11 13:02:17 -07:00
• b68fbd7d81 View.__add__ to merge_view (#3686) chenyuandGitHub 2024-03-11 15:52:34 -04:00
• 76ade20b89 hsa driver tiny cleanups (#3684) nimlgenandGitHub 2024-03-11 22:32:43 +03:00
• d69170e27e add llama 2 70B in ci and verify output (#3682) chenyuandGitHub 2024-03-11 12:48:22 -04:00
• e10ee2ed3f llama beam tinybox ci (#3680) chenyuandGitHub 2024-03-11 01:35:39 -04:00
• 3415b0ee54 hotfix: mixtral copies norms together for 2% speed geohot 2024-03-11 01:28:03 +00:00
• e8c350fdac fix: make Tensor.rand produce correct values for float16 (#3654) 3 people 2024-03-11 00:48:00 +02:00
• bad6adaf8c add mixtral and 6 gpus cifar to tinybox ci (#3676) chenyuandGitHub 2024-03-10 18:25:31 -04:00
• 44a67bf783 constant folding (#3675) George HotzandGitHub 2024-03-10 14:47:24 -07:00
• 25aede6fd9 truncate for exec_alu (#3674) George HotzandGitHub 2024-03-10 14:19:04 -07:00
• 957ae9b594 Fix Tensor's __repr__ for printing out grad (#3673) Francis LataandGitHub 2024-03-10 17:04:29 -04:00
• 0f16729023 RDNA3: restore launch bounds (#3672) George HotzandGitHub 2024-03-10 10:27:52 -07:00
• d7452c2a20 clean up llvmir builder (#3671) chenyuandGitHub 2024-03-09 21:19:36 -05:00
• 1143c62519 tensor.py touchups (#3667) George HotzandGitHub 2024-03-09 16:12:20 -08:00
• 69ca7f7bf9 changes for teenygrad (#3665) George HotzandGitHub 2024-03-09 15:30:34 -08:00
• 89b8b5d549 Fix missing import. (#3666) Quentin WachandGitHub 2024-03-09 23:55:23 +01:00
• 8ae85b2cf5 add inference_mode context manager with decorator support (#3621) Maximilian WolfandGitHub 2024-03-09 17:38:26 +01:00
• b5cbf1792a Fix Tensor.cumsum when axis of length 0 is selected (#3473) Obada KhaliliandGitHub 2024-03-09 18:26:41 +02:00
• 915f98791c use custom KernelOptError in kernel opt (#3661) chenyuandGitHub 2024-03-08 15:36:16 -05:00
• ac02e7347d ptx timing vs cuda timing (#3659) George HotzandGitHub 2024-03-08 10:17:49 -08:00
• daa4034e80 No more metal flakiness (#3643) uuuvnandGitHub 2024-03-08 18:54:44 +02:00
• e25879d50e don't get new var_val for the same ast in fuzz_linearizer (#3657) chenyuandGitHub 2024-03-08 09:49:24 -05:00
• 1130c73844 add FUZZ_NTH to fuzz_linearizer (#3656) chenyuandGitHub 2024-03-08 09:16:49 -05:00
• 9f66dcf718 PolynomialDecayWithWarmup + tests (#3649) David HouandGitHub 2024-03-07 15:53:36 -08:00
• 57df8e8d82 update fuzz_linearizer (#3648) chenyuandGitHub 2024-03-07 18:41:22 -05:00
• b282a45e39 fix direct store float4 with same vin (#3652) chenyuandGitHub 2024-03-07 18:11:50 -05:00
• a66ffec6d3 update kernel dataset to exclude the disktensor ones (#3651) chenyuandGitHub 2024-03-07 17:35:19 -05:00
• fcf4a5ccf2 fix example that calls Tensor.__bool__ (#3650) chenyuandGitHub 2024-03-07 16:59:26 -05:00
• 6e50582e62 working to improve ptx (#3647) George HotzandGitHub 2024-03-07 12:39:31 -08:00
• 1853ec9a02 add tests for bfloat16 on HIP (#3638) ZafferandGitHub 2024-03-07 13:45:36 -05:00
• 0cef284aac fix typing FlopCounter.flops can be sint (#3646) chenyuandGitHub 2024-03-07 12:49:17 -05:00
• 906cc3a69b cleanup tests Device[Device.DEFAULT] is always Compiled (#3645) chenyuandGitHub 2024-03-07 11:15:42 -05:00
• bdd62c7fd8 make the bf16 include dynamic (#3642) qazalandGitHub 2024-03-07 17:31:35 +02:00
• 4552248c84 fix Tensor.to preserves grad.data (#3636) chenyuandGitHub 2024-03-06 21:44:49 -05:00
• d33311ebe0 remove parens of ALU if it has associative property (#3635) chenyuandGitHub 2024-03-06 21:12:11 -05:00
• fe6b6e38c1 remove parentheses of GEP if it's from SSA (#3634) chenyuandGitHub 2024-03-06 20:22:46 -05:00
• 0afaf70d57 lars optimizer + tests (#3631) 3 people 2024-03-06 15:11:01 -08:00
• b2e92d44fa skip METAL sin test in test_dtype_alu (#3633) chenyuandGitHub 2024-03-06 17:29:19 -05:00
• 8f10bfa2ff ban __bool__ on Tensor (#3632) chenyuandGitHub 2024-03-06 17:12:35 -05:00
• 81baf3eed3 bring ptx back (#3623) George HotzandGitHub 2024-03-06 13:34:21 -08:00
• c270d54c32 update test_dtype_alu for METAL (#3629) chenyuandGitHub 2024-03-06 14:55:19 -05:00
• abc5f3a6a0 hip bf16 hotfix (#3630) qazalandGitHub 2024-03-06 21:42:30 +02:00
• bc2a13a5f7 test case to show clang and python doing math in double (#3628) chenyuandGitHub 2024-03-06 13:49:03 -05:00
• 568353fa84 hotfix: bump line count to 6500 geohot 2024-03-06 07:52:18 -08:00
• a1507c7fd4 Fix Tensor.dropout() with multigpu (#3619) Elias WahlandGitHub 2024-03-06 00:26:21 +01:00
• e5ee6bb2bd fix outdated url in showcase doc (#3624) Jungwan WooandGitHub 2024-03-06 07:44:40 +09:00
• 8500265561 this mem fault still happening (#3620) George HotzandGitHub 2024-03-05 10:39:32 -08:00
• 3c3f846c45 tinybox benchmark with HSA (#3603) chenyuandGitHub 2024-03-05 11:03:52 -05:00
• f500be1313 out of bounds access caused by launch bounds (#3615) George HotzandGitHub 2024-03-05 06:34:00 -08:00
• eb83e2d3a0 decouple buffer mutability from cstyle (#3617) qazalandGitHub 2024-03-05 16:20:59 +02:00
• 3275260c98 Revert "test: add failing bfloat16 test case for metal backend (#3481)" (#3618) chenyuandGitHub 2024-03-05 09:08:42 -05:00
• 1e12a2ae80 test: add failing bfloat16 test case for metal backend (#3481) SkoshandGitHub 2024-03-05 15:44:54 +02:00
• 957e9800f1 llama + beam to mac benchmark, full cifar to nvidia benchmark (#3612) chenyuandGitHub 2024-03-04 21:35:57 -05:00
• 282bbd5acb check the input length into argfix (#3610) chenyuandGitHub 2024-03-04 19:50:17 -05:00
• 7db6dd725d multilazybuffer fix (#3609) Elias WahlandGitHub 2024-03-04 23:36:23 +01:00
• c3b8d285aa cleanup uops (#3605) chenyuandGitHub 2024-03-04 11:03:14 -05:00
• 94679322a3 simpler float4 direct store and locals support (#3592) qazalandGitHub 2024-03-04 16:28:28 +02:00
• 3db826e195 hsa in lin opts (#3602) nimlgenandGitHub 2024-03-04 17:17:32 +03:00
• 7c90005c65 search: hotfix to make sure TC behavior is all in applied_opts (#3598) Francis LamandGitHub 2024-03-03 18:44:38 -08:00
• 8e5d60a322 add more gpt2 variant in mac/nvidia benchmark (#3599) chenyuandGitHub 2024-03-03 17:55:30 -05:00
• 968d109453 apply more create_lt_node (#3597) chenyuandGitHub 2024-03-03 16:12:39 -05:00
• bc562c4747 Python div alu behavior differs slightly from others (#3596) 3 people 2024-03-03 13:48:25 -05:00
• 56d21d77b3 Fix two bugs concerning Tensor.to. (#3593) Marcin SłowikandGitHub 2024-03-03 17:48:56 +01:00
• 0082300a59 Fix symbolic negative floordiv (#3594) 3 people 2024-03-03 11:40:52 -05:00
• e09619ab6c explicitly create_lt_node when used in shapetracker _expr_view (#3561) chenyuandGitHub 2024-03-03 10:08:21 -05:00
• 640dc0fc51 hsa flush hdp (#3591) nimlgenandGitHub 2024-03-03 15:55:07 +03:00
• 660df3cff1 Add test for .softmax.argmax (#3559) 4 people 2024-03-02 23:51:52 -05:00
• ee41fafdab use operator instead of lambda in python_alu (#3590) chenyuandGitHub 2024-03-02 19:33:21 -05:00
• a89afd4ffa Directly store float4 nodes (#3564) 3 people 2024-03-03 01:58:20 +02:00
• 770707b376 hotfix: gpuocelot no rebuild geohot 2024-03-02 15:57:38 -08:00
• 74c9acddb0 simple python ALU (#3589) George HotzandGitHub 2024-03-02 15:50:58 -08:00
• 162dfb07d9 fuzz_linearizer: fix uops and add to test.yml (#3588) Francis LamandGitHub 2024-03-02 15:03:42 -08:00
• 8978488565 add sanity tests for bufs_from_lin (#3586) Jovan SardinhaandGitHub 2024-03-02 14:17:43 -08:00
• aa9b013d79 add constant folding for WHERE in uops (#3584) George HotzandGitHub 2024-03-02 10:37:14 -08:00
• 3b7e3fa2e4 fix sync in hsa graph (#3582) nimlgenandGitHub 2024-03-02 18:37:51 +03:00
• 6c36264790 Improve type hints for optimizer (#3583) Szymon OżógandGitHub 2024-03-02 16:35:44 +01:00
• 83530a585f add quick external data select test geohot 2024-03-02 05:38:32 -08:00
• 9a37273d36 consts don't have nodes in the graph (#3579) George HotzandGitHub 2024-03-02 04:19:11 -08:00
• 41f0a25b53 lazy.py: cache consts (#3577) George HotzandGitHub 2024-03-02 03:50:05 -08:00
• fb8acd1851 Don't touch UOps.DEFINE_GLOBAL (#3575) uuuvnandGitHub 2024-03-02 13:30:05 +02:00
• 50e1445e60 Revert "allow overriding weight init for Linear (#3569)" (#3576) George HotzandGitHub 2024-03-02 03:17:13 -08:00
• 2d0973a852 allow overriding weight init for Linear (#3569) David HouandGitHub 2024-03-02 03:16:04 -08:00
• 9642a8f547 search: add BEAM UPCAST/LOCAL params and loosen TC criteria during BEAM (#3563) Francis LamandGitHub 2024-03-02 03:11:25 -08:00
• ba6c041eab fix SCE ignore_index with label_smoothing (#3574) David HouandGitHub 2024-03-01 19:19:45 -08:00
• e17f1821a7 wmma: add CUDA tensor core and fix test_speed_v_torch failure (#3544) Francis LamandGitHub 2024-03-01 17:51:02 -08:00
• b3cdc11a58 label_smoothing in sparse_cat_crossentropy (#3568) David HouandGitHub 2024-03-01 17:02:46 -08:00
• 6b29c70b3d Refactor to UOpGraph class (#3566) George HotzandGitHub 2024-03-01 15:14:48 -08:00
• b7e555f6c0 run test_linearizer_failures on PYTHON backend (#3565) chenyuandGitHub 2024-03-01 17:00:18 -05:00