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

• 73c75d6ee1 DEFINE_LOCAL variable names start from temp0, not temp1 (#8870) Ali LadjevardiandGitHub 2025-02-03 18:20:38 +03:30
• b6c617272a New schedule.py Order [pr] (#8874) qazalandGitHub 2025-02-03 07:59:11 -05:00
• b075aefc12 hotfix: revert llvm host_arch geohot 2025-02-03 16:46:19 +08:00
• a5753095dc llvm cleanups [pr] (#8867) George HotzandGitHub 2025-02-03 15:32:41 +08:00
• f484db0e63 dsp cleanups [pr] (#8866) George HotzandGitHub 2025-02-03 15:18:53 +08:00
• af2c2837f6 hotfix: skip broken test, add KERNEL Op geohot 2025-02-03 14:02:55 +08:00
• 565c37c681 start simplifying the scheduler context [pr] (#8830) qazalandGitHub 2025-02-02 11:11:36 -05:00
• d64af3c884 reorder simplifier and grouper logic in scheduler [pr] (#8861) qazalandGitHub 2025-02-02 10:19:52 -05:00
• 83a904aaad just schedule in test_recursive_pad [pr] (#8860) qazalandGitHub 2025-02-02 08:01:24 -05:00
• 6dadb60c93 LLVM JIT (+autogen llvm instead of llvmlite) (#8486) 3 people 2025-02-02 13:52:42 +02:00
• 66306b5321 Fix disk tensor assignment (#8855) 3 people 2025-02-02 15:50:34 +04:00
• 6e523e4d17 Remove size arg from DEFINE_LOCAL [pr] (#8845) Ali LadjevardiandGitHub 2025-02-02 15:17:32 +03:30
• 7841852870 hcq pci signal fuzzer (#8854) nimlgenandGitHub 2025-02-01 23:42:27 +03:00
• dc34a4146f better process_replay context print [pr] (#8856) qazalandGitHub 2025-02-01 14:50:23 -05:00
• 5b1fc4dcb2 push cast to branches in UOp where (#8850) chenyuandGitHub 2025-02-01 13:55:24 -05:00
• 73ee2d74c0 raise RuntimeError for int base pow (#8852) chenyuandGitHub 2025-02-01 12:11:57 -05:00
• 72e1f41f8e add unbind_vars pattern matcher (#8851) qazalandGitHub 2025-02-01 11:25:44 -05:00
• b3fa76419a am: move queues to gpus (#8848) nimlgenandGitHub 2025-02-01 18:02:52 +03:00
• 42d7c800a1 hotfix: add missing tinychat fonts + other assets geohot 2025-02-01 09:32:33 +08:00
• 431a86615d fix multi Ops.CONTIGUOUS_BACKWARD [pr] (#8843) George HotzandGitHub 2025-02-01 09:21:31 +08:00
• 07d3676019 weights_only=False (#8839) Ahmed HarmoucheandGitHub 2025-01-31 23:16:47 +01:00
• 741bbc900d Revert "am: queues allocated on gpus (#8836)" (#8837) nimlgenandGitHub 2025-01-31 22:53:41 +03:00
• 7bbb568dec am: queues allocated on gpus (#8836) nimlgenandGitHub 2025-01-31 22:14:43 +03:00
• 1f730ae8f8 remove retain_graph in Tensor.backward [pr] (#8835) chenyuandGitHub 2025-01-31 13:41:26 -05:00
• 0a59db936a raise RuntimeError in schedule_step if not Tensor.training [pr] (#8834) chenyuandGitHub 2025-01-31 12:03:04 -05:00
• af4f9d1aa9 use matchers to verify AST shape [pr] (#8828) qazalandGitHub 2025-01-31 02:17:42 -05:00
• 643c09a6c6 tensor uop spec should be in spec.py [pr] (#8827) George HotzandGitHub 2025-01-31 13:54:04 +08:00
• a78f0f85d3 remove support for checking tensor uops in FUSE_ARANGE [pr] (#8829) qazalandGitHub 2025-01-31 00:48:28 -05:00
• 2a33750e4c simpler group_realizes + ScheduleItem construction [pr] (#8825) qazalandGitHub 2025-01-30 23:34:53 -05:00
• e63d160376 hotfix: sched comment geohot 2025-01-31 12:10:04 +08:00
• 1fce864a6d delete multi output support (#8822) qazalandGitHub 2025-01-30 22:45:50 -05:00
• 7647cd8428 [bounty] Stride is flip (#8792) 4 people 2025-01-31 08:04:10 +05:30
• 0513b0c17d lower green test_gemm_8192 tflops to 125 [pr] (#8820) chenyuandGitHub 2025-01-30 17:30:08 -05:00
• f0924e0857 fix and test (#8814) 3 people 2025-01-30 18:35:53 -03:00
• f5da275f46 simpler remove_movement_ops [pr] (#8818) qazalandGitHub 2025-01-30 16:32:52 -05:00
• c8d878a5c1 remove r.lazydata.buf_uop_view [pr] (#8817) qazalandGitHub 2025-01-30 16:14:36 -05:00
• 530961f7d5 realized only exists on base (#8815) qazalandGitHub 2025-01-30 16:02:25 -05:00
• 7cdc607544 add max as associative (#8816) Sieds LyklesandGitHub 2025-01-30 22:01:42 +01:00
• 5643429c17 give BUFFER UOp a ShapeTracker [pr] (#8811) qazalandGitHub 2025-01-30 15:33:32 -05:00
• 5527f86a8f skip tests in test_indexing that set stride with lazydata.view [pr] (#8813) chenyuandGitHub 2025-01-30 15:17:35 -05:00
• a2faa5e49b am: fix pt free (#8810) nimlgenandGitHub 2025-01-30 15:14:55 +03:00
• 9df8e34160 prereqs for giving BUFFER UOps a ShapeTracker [pr] (#8809) qazalandGitHub 2025-01-30 06:30:24 -05:00
• 78c0455c7a Better stable sigmoid (#8806) 3 people 2025-01-29 22:08:53 +01:00
• cac2b4e8b6 bump db version (#8805) chenyuandGitHub 2025-01-29 15:34:43 -05:00
• a2d55fb644 am: unset CGCG override (#8804) nimlgenandGitHub 2025-01-29 22:28:06 +03:00
• 7f606fbde4 remove DEBUG=5 in windows ci test [pr] (#8803) chenyuandGitHub 2025-01-29 14:18:17 -05:00
• 260df1a17f tc_select noop (#8801) Ignacio SicaandGitHub 2025-01-29 15:53:23 -03:00
• ec120ce6b9 Fix allocator memory alignment (#8800) FICTURE7andGitHub 2025-01-29 22:03:17 +04:00
• 50ba2bb642 am: move ring to host mem (#8802) nimlgenandGitHub 2025-01-29 20:56:11 +03:00
• c7ca7959e6 set DISABLE_DROPOUT=1 in bert script for now (#8799) chenyuandGitHub 2025-01-29 10:51:29 -05:00
• 199a36d079 add pagination to viz [pr] (#8794) qazalandGitHub 2025-01-28 21:21:53 -05:00
• ba17786068 do not construct unmasked VALID (#8759) 3 people 2025-01-28 13:51:21 -05:00
• 3417bc1814 fix ShapeTracker spec for const [pr] (#8791) qazalandGitHub 2025-01-28 12:53:36 -05:00
• 801ec9e697 am: no hardcoded clocks (#8788) nimlgenandGitHub 2025-01-28 20:18:46 +03:00
• e724af74d7 allow VIEW source in DEFINE_VAR spec [pr] (#8790) qazalandGitHub 2025-01-28 10:42:14 -05:00
• da464d039f fix windows ci cache (#8787) 3 people 2025-01-28 19:22:15 +08:00
• e8be8a5835 support lowering CONST(VIEW) in lowerer (#8785) qazalandGitHub 2025-01-28 05:04:41 -05:00
• 80089536e5 Revert "move llvm_bf16_cast to renderer for CLANG and LLVM [pr] (#8720)" (#8786) George HotzandGitHub 2025-01-28 18:59:02 +09:00
• 5d62aa28dc Support CLANG backend on Windows (#8768) 4 people 2025-01-28 17:19:34 +08:00
• af0452f116 move llvm_bf16_cast to renderer for CLANG and LLVM [pr] (#8720) 4 people 2025-01-28 17:16:43 +08:00
• d66680b17e hotfix: am: fix hang (#8783) nimlgenandGitHub 2025-01-28 11:54:19 +03:00
• a65d2917cb remove unused fields from viz uop_to_json [pr] (#8782) qazalandGitHub 2025-01-28 03:50:11 -05:00
• aefbc2637f test fixups from unmasked valid deletion [pr] (#8776) qazalandGitHub 2025-01-28 02:23:30 -05:00
• ed672881b0 remove additions/deletion in pr + check uops are equal [pr] (#8779) qazalandGitHub 2025-01-28 01:57:34 -05:00
• 2c71c60719 opt arg is int or tuple (#8780) Ignacio SicaandGitHub 2025-01-27 23:02:32 -03:00
• 62655e4999 move multi into engine [pr] (#8778) George HotzandGitHub 2025-01-28 09:15:29 +09:00
• 299fa8f37b am: unset high clocks for sleep (#8775) nimlgenandGitHub 2025-01-28 01:15:56 +03:00
• c99ae81f63 update default resnet LOSS_SCALER to 256 [pr] (#8774) chenyuandGitHub 2025-01-27 16:59:05 -05:00
• 1c608ae34f am_smi: better spacing (#8773) nimlgenandGitHub 2025-01-27 23:01:02 +03:00
• b240f12593 [TIP-9] rename Opt's amt to arg 2 (#8770) 3 people 2025-01-27 16:19:04 -03:00
• 9760688e7f use IGNORE_BEAM_CACHE in search [pr] (#8772) chenyuandGitHub 2025-01-27 13:41:01 -05:00
• ed1b573868 ignore beam cache in test_tiny for stateless beam (#8771) Ignacio SicaandGitHub 2025-01-27 14:56:30 -03:00
• 3ed146a5ff Revert "rename Opt amt to arg (#8767)" (#8769) George HotzandGitHub 2025-01-27 23:46:37 +09:00
• bf041659a5 rename Opt amt to arg (#8767) Ignacio SicaandGitHub 2025-01-27 11:36:47 -03:00
• 96bff0b4f7 contiguous is no longer needed in SGD [pr] (#8760) George HotzandGitHub 2025-01-27 15:19:11 +09:00
• efc7971090 add windows test to ci (#8761) 3 people 2025-01-27 13:53:21 +08:00
• a9d9f98d05 hotfix: those tests fail locally on mac due to buffer count geohot 2025-01-27 07:53:48 +09:00
• 2454bf01c3 hotfix: remove shapetracker spam in viz geohot 2025-01-27 07:20:21 +09:00
• d488bbb1ec share merge_views/valid creation for CONST/DEFINE_VAR (#8758) qazalandGitHub 2025-01-26 10:41:54 -05:00
• bbb2dd8141 move VALID creation after merging the views (#8757) qazalandGitHub 2025-01-26 09:58:05 -05:00
• a6e496b195 remove Function class [pr] (#8753) George HotzandGitHub 2025-01-26 18:58:02 +09:00
• ac70f63d4b tensor_map cleanups [pr] (#8754) qazalandGitHub 2025-01-26 04:41:54 -05:00
• b53fe7c2fc remove unused ctx [pr] (#8751) George HotzandGitHub 2025-01-26 17:59:15 +09:00
• 06b58aa7ec move unneeded fields out of ScheduleContext [pr] (#8752) qazalandGitHub 2025-01-26 03:36:15 -05:00
• 1b4618e257 gradient cleanup (#8750) George HotzandGitHub 2025-01-26 09:30:55 +09:00
• b4bf6a7dea switch backward to use gradient [pr] (#8235) George HotzandGitHub 2025-01-26 09:12:16 +09:00
• 0ffd572e1e fix multi with no real srcs (#8749) George HotzandGitHub 2025-01-26 08:41:00 +09:00
• 0e42befc6e viz cleanups 2 [pr] (#8748) qazalandGitHub 2025-01-25 12:41:57 -05:00
• c74c5901a8 am disable bind (#8747) nimlgenandGitHub 2025-01-25 19:06:35 +03:00
• a037201168 test_viz cleanups + move to /unit directory (#8746) qazalandGitHub 2025-01-25 07:33:31 -05:00
• e2b380b743 make UOp.multi real a tuple instead of list [pr] (#8744) chenyuandGitHub 2025-01-24 20:47:27 -05:00
• cb0978b377 add Ops.CONTIGUOUS_BACKWARD (#8743) George HotzandGitHub 2025-01-25 07:28:43 +09:00
• 2f06eccf1d am: script and vfio msg (#8742) nimlgenandGitHub 2025-01-25 00:33:00 +03:00
• 0c759e1ff6 add bert to bechmark ci (#8741) chenyuandGitHub 2025-01-24 14:45:11 -05:00
• e0e176efbc failed test case for multi rand_like [pr] (#8740) chenyuandGitHub 2025-01-24 13:56:51 -05:00
• dc10187fc0 am: add am_smi (#8739) nimlgenandGitHub 2025-01-24 20:16:19 +03:00
• 7a2223a6c6 add merge views to ops_folding [pr] (#8051) 3 people 2025-01-25 00:45:11 +09:00
• 0814a79cb4 cleanup the merge_views upats [pr] (#8738) qazalandGitHub 2025-01-24 09:49:54 -05:00
• 07069b9988 rename to tensor_uop [pr] (#8737) qazalandGitHub 2025-01-24 06:42:25 -05:00
• e82ba1454b MultiLazyBuffer is UOp [pr] (#8662) George HotzandGitHub 2025-01-24 13:28:55 +09:00