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

• cdc3b95729 if you don't appreciate a 15 second timeout, you get a 10 second timeout geohot 2023-11-28 17:44:09 -08:00
• d87a246439 move to new cached fetch (#2493) George HotzandGitHub 2023-11-28 17:36:55 -08:00
• ab5d14d4ba MEM -> LOAD (#2492) George HotzandGitHub 2023-11-28 16:46:37 -08:00
• a739c6646e fp16 in gpt2 attention (#2491) chenyuandGitHub 2023-11-28 19:27:03 -05:00
• 847f0a02b1 non-simplifiable mod should result in ModNode (#2490) chenyuandGitHub 2023-11-28 16:52:19 -05:00
• 3f137b134a jax parallel matmul example geohot 2023-11-28 13:48:11 -08:00
• ddb6a33ae5 improve test assertions for jit cache len with graph executor (#2476) mmmkkaaayyandGitHub 2023-11-27 23:02:45 -08:00
• 28a67106ca enable symbolic ops tests for hip (#2485) chenyuandGitHub 2023-11-28 01:33:41 -05:00
• 7f01dd04f0 Apply ruff linting rules to tests (#2473) Christopher Mauri MilanandGitHub 2023-11-27 21:24:06 -08:00
• 136dbd8b36 HIP CI that compiles (to RDNA3) but doesn't have to run (#2482) 3 people 2023-11-28 12:17:06 +07:00
• 756b01f46f why were these ever called buffer (#2483) George HotzandGitHub 2023-11-27 21:02:07 -08:00
• acbe6d1b53 Revert "HIP compilation on CI targeting RDNA3 (#2459)" (#2481) George HotzandGitHub 2023-11-27 20:41:21 -08:00
• cb507a9389 Remove the toCPU copy (#2445) qtkiteandGitHub 2023-11-28 15:37:13 +11:00
• d275ff930a HIP compilation on CI targeting RDNA3 (#2459) Davi SilvaandGitHub 2023-11-28 11:33:11 +07:00
• d43485ae9e Fix graph_uops (#2457) Yingbo MaandGitHub 2023-11-27 21:42:48 -05:00
• 98cd9e8926 Add assertion to prevent nonsense mod values (#2474) Paul GustafsonandGitHub 2023-11-27 18:37:44 -08:00
• 186ac77ec3 Update hip_matmul.py (#2480) Davi SilvaandGitHub 2023-11-28 09:36:19 +07:00
• 7f9a4c1285 fp16 and noshow flags for gpt2 (#2470) chenyuandGitHub 2023-11-27 16:23:03 -05:00
• e267a93124 reset seed on every run (#2468) qazalandGitHub 2023-11-27 15:55:54 -05:00
• 9e07824542 move device to device.py (#2466) George HotzandGitHub 2023-11-27 11:34:37 -08:00
• 262cd26d28 Simplify openpilot kernel (#2460) qazalandGitHub 2023-11-27 13:02:27 -05:00
• 61a80a0675 asserts LtNodes of SumNode with MulNode of Nodes (#2465) chenyuandGitHub 2023-11-27 12:56:59 -05:00
• c4dfde761e remove the commented import (#2463) chenyuandGitHub 2023-11-27 11:50:41 -05:00
• a031afb2f6 Update display_name in resnet50 example (#2454) Akshay KashyapandGitHub 2023-11-26 19:07:36 -05:00
• 1d89c018fa Add isinstance check before gcd call in SumNode.__lt__ (#2450) 3 people 2023-11-26 13:05:04 -08:00
• 58b1dd463e Add error code to type: ignore (#2451) 3 people 2023-11-26 13:04:10 -08:00
• 8e9cdef61f clean up the buffers (#2447) George HotzandGitHub 2023-11-26 11:02:29 -08:00
• f6f712e609 split out the three steps of exec_ast (#2446) George HotzandGitHub 2023-11-26 09:07:37 -08:00
• 511310737e test_linearizer_failures to run on all backends (#2443) chenyuandGitHub 2023-11-26 01:17:29 -05:00
• c42d2c4731 strip whitespace geohot 2023-11-25 14:09:06 -08:00
• 9eb2746d62 fix copy issue + add regression test (#2441) George HotzandGitHub 2023-11-25 14:06:08 -08:00
• 259a869fc1 Fix UnicodeDecodeError when debugging on Intel APU (#2421) andresgitandGitHub 2023-11-25 22:30:50 +02:00
• 7170a9a057 coder.py can write and run code (#2439) George HotzandGitHub 2023-11-25 12:27:54 -08:00
• df41a57e09 Fix: missing n_kv_heads for smaller models from huggingface (#2438) Davi SilvaandGitHub 2023-11-26 01:29:04 +07:00
• 96c12fdeab multibatch gpt2 (#2432) George HotzandGitHub 2023-11-24 18:10:10 -08:00
• 5326bbc9a6 fix causal mask in Tensor class (#2425) Tobias FischerandGitHub 2023-11-24 18:38:18 -05:00
• 9a5d0e70de Device.DEFAULT instead of getenv to exclude tests (#2429) chenyuandGitHub 2023-11-24 17:10:24 -05:00
• 6223f8894d clean up ast_parse (#2428) chenyuandGitHub 2023-11-24 16:43:32 -05:00
• 8ff2e13550 From teeny (#2426) George HotzandGitHub 2023-11-24 12:50:56 -08:00
• 9ae83fba04 flatten instead of reduce, improve type inference (#2423) chenyuandGitHub 2023-11-24 13:19:22 -05:00
• 7169de57e2 Update VITS to use fetch helper (#2422) Francis LataandGitHub 2023-11-24 11:50:03 -05:00
• 857d440ea7 fail means fail (#2391) George HotzandGitHub 2023-11-24 08:27:39 -08:00
• 1f4231a8f9 global pipefail geohot 2023-11-24 08:03:49 -08:00
• e68aebfff9 bring hip graph back (#2385) 3 people 2023-11-24 18:53:44 +03:00
• 46b05daf7c simple display_name (#2416) George HotzandGitHub 2023-11-23 19:50:23 -08:00
• 12023b6824 onnx ops cleanup (#2413) George HotzandGitHub 2023-11-23 18:39:49 -08:00
• 8f89e21fca torch and numpy don't share ops anymore (#2412) George HotzandGitHub 2023-11-23 16:58:10 -08:00
• 193be14b6c that had bugs, force an order (#2411) George HotzandGitHub 2023-11-23 15:52:16 -08:00
• 65f4e6971b beautiful_mnist.py link geohot 2023-11-23 14:58:22 -08:00
• 1b3b8de5e2 update readme examples geohot 2023-11-23 14:54:52 -08:00
• 5bb720a777 Cocoa is no longer used geohot 2023-11-23 14:31:21 -08:00
• 095e2ced61 add name support to fetch (#2407) George HotzandGitHub 2023-11-23 14:16:17 -08:00
• 397c093656 fix wait in jit (#2408) nimlgenandGitHub 2023-11-24 00:54:13 +03:00
• b927942d58 Move HIP render logic to its dedicated place (#2394) 3 people 2023-11-23 16:03:29 -05:00
• 6d672785db Update Whisper to use fetch helper (#2401) Francis LataandGitHub 2023-11-23 15:59:59 -05:00
• 0505c5ea50 remove force_wait, refactor to graph (#2405) George HotzandGitHub 2023-11-23 12:46:07 -08:00
• c5d585ea35 Fix Triton README broken link (#2406) Ivan BeňovicandGitHub 2023-11-23 21:38:17 +01:00
• b27c845531 minor cleanup for View strides (#2404) chenyuandGitHub 2023-11-23 13:40:01 -05:00
• 64aa2f4156 clean up to_shape_strides (#2402) chenyuandGitHub 2023-11-23 13:04:00 -05:00
• e4026dc197 don't pass lazybuffer to rawbuffer (#2400) George HotzandGitHub 2023-11-23 09:40:28 -08:00
• aefa97a962 Remove runtime imports in realize (#2157) 3 people 2023-11-24 01:17:04 +08:00
• 4f8f0ac139 minor cleanups, remove dead files (#2398) George HotzandGitHub 2023-11-23 09:01:50 -08:00
• 66c75f30c6 remove triton (#2396) George HotzandGitHub 2023-11-23 07:40:59 -08:00
• 8656eebb42 jit doesn't use named tensors (#2393) George HotzandGitHub 2023-11-23 00:13:18 -08:00
• 80e4ad8bf5 faster get_recursive_parents (#2392) George HotzandGitHub 2023-11-22 20:37:19 -08:00
• 8798d120bb autopad shapetracker for BEAM (#2375) chenyuandGitHub 2023-11-22 21:05:25 -05:00
• 162db466c3 hotfix: fix hip WMMA casting hack Tiny Box 2023-11-22 17:58:08 -08:00
• 6ceecc961e hotfix: scalar geohot 2023-11-22 17:48:24 -08:00
• 0eda545946 dtypes.float.vec(sz) (#2386) qazalandGitHub 2023-11-22 20:43:14 -05:00
• cbb8486779 ResNet training changes (update benchmark) (#2390) George HotzandGitHub 2023-11-22 17:41:12 -08:00
• 2dec86970a hotfix: default remains gen 1 llama geohot 2023-11-21 14:43:02 -08:00
• 7f0cc4a4e8 whisper: support audio >30s (#2378) mmmkkaaayyandGitHub 2023-11-21 14:37:51 -08:00
• 7220f5c9fc fixed hf convert and now it's working with tinyllama (#2374) Oleg RybalkoandGitHub 2023-11-22 01:36:52 +03:00
• d0f966b320 add a segfault linearizer test case (#2383) chenyuandGitHub 2023-11-21 15:06:41 -05:00
• 9eeba968cd fix the variable arg order (#2382) chenyuandGitHub 2023-11-21 12:02:31 -05:00
• c5f429a40a Fix linearizer cache (#2371) nimlgenandGitHub 2023-11-21 18:58:35 +03:00
• 0da72119bb Readable and Faster Union of Vars (#2380) Umut ZenginandGitHub 2023-11-21 17:45:19 +03:00
• 15c316b9b1 add marker (#2379) qazalandGitHub 2023-11-21 09:44:15 -05:00
• fb0d650b25 feat: don't optimize buffers when its not an astrunner (#2377) wozeparrotandGitHub 2023-11-21 01:07:31 -05:00
• abbcc7aefa missed cleanup from cache_id removal (#2376) wozeparrotandGitHub 2023-11-21 01:03:43 -05:00
• 179551a55c remove file writing in metal ops (#2369) 3 people 2023-11-21 10:24:39 +07:00
• c4cc4966ed update some test_tensor.py cases with 0 in shape (#2368) chenyuandGitHub 2023-11-19 20:35:05 -05:00
• 6add808f6a support tuple shape input for rand and empty (#2367) chenyuandGitHub 2023-11-19 20:20:39 -05:00
• e9847be790 remove whisper +1-1 hack (#2360) chenyuandGitHub 2023-11-19 17:56:36 -05:00
• a0890f4e6c move fetch to helpers (#2363) George HotzandGitHub 2023-11-19 12:29:51 -08:00
• 03968622a2 Pretty multinomial (#2365) chenyuandGitHub 2023-11-19 15:10:10 -05:00
• 0eb0defa6f remove unused key properties (#2359) Friedrich Carl EichenrothandGitHub 2023-11-19 08:30:21 +01:00
• b3a21eee7d just new types (#2358) Friedrich Carl EichenrothandGitHub 2023-11-19 08:29:46 +01:00
• f203d37258 retry test_webgpu.js 3 times (#2362) chenyuandGitHub 2023-11-18 21:24:47 -05:00
• 08d09eb666 Enable whisper test in CI for more backends (#2355) mmmkkaaayyandGitHub 2023-11-18 14:52:50 -08:00
• d7d078c7f9 Node.vars() returns a set and properly dedup (#2356) chenyuandGitHub 2023-11-18 17:44:52 -05:00
• 0443cbfbb9 fix shm path test on macos (#2357) chenyuandGitHub 2023-11-18 17:37:42 -05:00
• f02e17a967 Variable.num -> NumNode (#2354) chenyuandGitHub 2023-11-18 15:45:52 -05:00
• 40246d35bc ops_shm removed (#2351) George HotzandGitHub 2023-11-18 11:41:58 -08:00
• 9b58d4cb37 cleanup unused movement ops (#2353) George HotzandGitHub 2023-11-18 09:19:02 -08:00
• c4d97bba8c simplify Node.sum, remove factorize method (#2352) chenyuandGitHub 2023-11-18 11:55:48 -05:00
• e35c31c8e5 xid for hip, device in time linearizer (#2348) 3 people 2023-11-17 20:50:07 -08:00
• 6e44a798df update fixed linearizer test (#2347) chenyuandGitHub 2023-11-17 23:46:37 -05:00
• c8c5212dce a lil more beautiful_mnist geohot 2023-11-17 19:53:06 -08:00
• c7b38b324b A beautiful MNIST training example (#2272) George HotzandGitHub 2023-11-17 19:42:43 -08:00