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

• cc37c99ae4 tiny hcq touchups (#5964) nimlgenandGitHub 2024-08-07 21:03:20 +03:00
• 8d8704af2d fix amd exec_update for locals (#5966) nimlgenandGitHub 2024-08-07 21:02:56 +03:00
• 0ddcd005f5 fix priority width and give more space for src (#5509) ignaciosicaandGitHub 2024-08-07 14:48:18 -03:00
• 0c4e9dbe71 retrieve defined opencl error codes (#5792) tyoc213andGitHub 2024-08-07 11:46:24 -06:00
• 4b48f166ec Refactor render_kernel for NV [run_process_replay] (#5965) 3 people 2024-08-07 14:36:04 -03:00
• d6f4a61c42 graph LBScheduleItem [run_process_replay] (#5960) qazalandGitHub 2024-08-08 00:59:11 +08:00
• 0a8668cf30 improvements to docs geohot 2024-08-07 09:57:24 -07:00
• 7677361d90 test pushing through different expands in 1 kernel (#5963) qazalandGitHub 2024-08-08 00:33:18 +08:00
• 564a352194 nv unify _gpu_free (#5961) nimlgenandGitHub 2024-08-07 18:18:17 +03:00
• 39c8c9c00a Add docs (#5942) 3 people 2024-08-07 10:38:51 -04:00
• 39dda3d042 rename prescheduled items to lsi [run_process_replay] (#5959) qazalandGitHub 2024-08-07 19:31:50 +08:00
• 728b7e189e diff_schedule tests [run_process_replay] (#5958) qazalandGitHub 2024-08-07 18:50:27 +08:00
• a7163b80d8 lower test_transcendental fuzz test threshold for sin float64 (#5956) chenyuandGitHub 2024-08-07 02:04:37 -04:00
• fa3a36e576 fancier UOp div gcd folding (#5953) chenyuandGitHub 2024-08-07 02:04:25 -04:00
• aa7fd7ef74 Use (-self).lt(-x+1) for UOp.ge (#5955) chenyuandGitHub 2024-08-07 01:31:27 -04:00
• 3d445039c2 hotfix: 8800 lines for AMX+intel tc geohot 2024-08-06 17:50:26 -07:00
• 658d58784b embedding doesn't cast (#5952) George HotzandGitHub 2024-08-06 17:49:14 -07:00
• 30d0cb2a82 fix: fix transcendental flakyness on exp float with 9.96875 (#5951) wozeparrotandGitHub 2024-08-06 17:32:13 -07:00
• 3a0515ea22 hotfix: process_replay/diff_schedule.py to LBScheduleItem geohot 2024-08-06 16:59:37 -07:00
• aee737bd9e divide by gcd in UOp div folding (#5949) chenyuandGitHub 2024-08-06 20:00:57 -04:00
• 6d1fdcfce2 don't reduce the same thing in a vector (#5950) George HotzandGitHub 2024-08-06 16:59:15 -07:00
• d5d7f4e7b8 more TestIndexing correctness asserts [run_process_replay] (#5948) qazalandGitHub 2024-08-07 06:50:42 +08:00
• 7f062929e8 start all cached scheduler functions with buf, st [run_process_replay] (#5946) qazalandGitHub 2024-08-07 06:24:22 +08:00
• 794796256c UOp.const_factor [run_process_replay] (#5945) chenyuandGitHub 2024-08-06 18:18:29 -04:00
• c9862e17d4 MLPERF BERT submission scripts (#5931) Elias WahlandGitHub 2024-08-07 00:09:18 +02:00
• 73d4d51845 add LBScheduleItem type [run_process_replay] (#5944) George HotzandGitHub 2024-08-06 14:49:40 -07:00
• 1dab75ae37 clean up mlperf dataloader import (#5940) chenyuandGitHub 2024-08-06 17:10:08 -04:00
• 7b6496f2e6 fix the reduceops cache breaking beautiful_mnist (#5938) qazalandGitHub 2024-08-07 05:02:54 +08:00
• 1417cc8df1 can reenable that test now (#5914) George HotzandGitHub 2024-08-06 13:38:21 -07:00
• 75154d7ae2 add some types to the scheduler [run_process_replay] (#5941) George HotzandGitHub 2024-08-06 12:23:54 -07:00
• e077bc7baf move memory planner to realize (#5937) George HotzandGitHub 2024-08-06 10:41:29 -07:00
• 489575c3be more UOp sum div with gcd tests (#5936) chenyuandGitHub 2024-08-06 12:50:10 -04:00
• 81ae9fadc8 Float4 support for CLANG (#5915) ignaciosicaandGitHub 2024-08-06 11:50:12 -03:00
• a7db4c3ee9 show timings for DIFF_ARANGE=1 (#5935) qazalandGitHub 2024-08-06 22:20:38 +08:00
• 102a8c184b diff fused arange schedules with ARANGE_DIFF=1 (#5934) qazalandGitHub 2024-08-06 21:52:26 +08:00
• f7761245aa save_schedule pre toposort [run_process_replay] (#5933) qazalandGitHub 2024-08-06 20:10:01 +08:00
• 895e062723 nv remove useless init (#5932) nimlgenandGitHub 2024-08-06 14:41:40 +03:00
• 3d4742dd2e override output shape in fused assign (#5930) qazalandGitHub 2024-08-06 18:28:50 +08:00
• 341c394c89 amd save exec offsets (#5928) nimlgenandGitHub 2024-08-06 12:11:46 +03:00
• 5808e8a30f mockgpu remu changes (#5925) wozeparrotandGitHub 2024-08-05 19:26:58 -07:00
• 09b7722637 UOp generic div folding (#5896) chenyuandGitHub 2024-08-05 21:38:43 -04:00
• 3e1336957d test arange with all opts (#5923) George HotzandGitHub 2024-08-05 18:38:25 -07:00
• 2e7adb529f don't run kernels with 1000x more compute (fix BEAM with FUSE_ARANGE) (#5924) George HotzandGitHub 2024-08-05 16:28:09 -07:00
• 5d17f54e3c fast mnist indexing (#5921) George HotzandGitHub 2024-08-05 13:55:15 -07:00
• e81c18f494 make the arange test check correctness [run_process_replay] (#5920) George HotzandGitHub 2024-08-05 13:41:06 -07:00
• 8d1c884e78 capture the const pattern in both directions (#5919) George HotzandGitHub 2024-08-05 12:15:38 -07:00
• 42f599870c unroll arange is broken (#5918) George HotzandGitHub 2024-08-05 12:15:07 -07:00
• 6740a0a6a0 hip_ioctl changes (#5917) wozeparrotandGitHub 2024-08-05 11:58:38 -07:00
• 70949ea7e6 test cstyle compile error for max with inline const (#5838) qazalandGitHub 2024-08-06 00:02:16 +08:00
• 98df648a79 metal sync queues in transfer (#5308) nimlgenandGitHub 2024-08-05 18:43:22 +03:00
• 6a70c69167 hotfix: TC renders nv_bfloat16 (#5913) qazalandGitHub 2024-08-05 23:40:31 +08:00
• 8ce9e6e693 Fix vectorized dtype rendering bug in CLANG (#5911) P4ssengerandGitHub 2024-08-05 11:43:26 -03:00
• e0c6520138 check arange fusing with VIEW and COPY (#5912) qazalandGitHub 2024-08-05 22:09:21 +08:00
• 590b9ebb34 hcq copy queue is optional (#5909) nimlgenandGitHub 2024-08-05 14:03:25 +03:00
• 159ac06b5b remove unused reduce rules + improve unparented (#5908) George HotzandGitHub 2024-08-04 18:18:27 -07:00
• d7387d31bf remove useless reduce cases [run_process_replay] (#5907) George HotzandGitHub 2024-08-04 17:11:08 -07:00
• 94917521ee fix: sqlite on pypy (#5906) wozeparrotandGitHub 2024-08-04 23:40:59 +00:00
• be8958e26b use CONTRACT before REDUCE (#5903) George HotzandGitHub 2024-08-04 16:17:33 -07:00
• f33950f454 tracemeta fixups (#5904) wozeparrotandGitHub 2024-08-04 23:15:06 +00:00
• adba5efc64 enable llama 2 70B in tinybox green CI (#5905) chenyuandGitHub 2024-08-04 18:48:46 -04:00
• 4a65010de8 remove CUDACPU flag in tests [run_process_replay] (#5902) chenyuandGitHub 2024-08-04 16:06:38 -04:00
• 996ff0c135 pow(2) -> square in RMSNorm [run_process_replay] (#5901) chenyuandGitHub 2024-08-04 14:21:31 -04:00
• aad9234e52 test fused precompute_freqs_cis (#5900) qazalandGitHub 2024-08-05 02:01:05 +08:00
• c67e9887f7 support using str to specify dtype (#5897) chenyuandGitHub 2024-08-04 12:56:28 -04:00
• 4f9221e8dd remove useless _ensure_shared_time_base (#5899) nimlgenandGitHub 2024-08-04 17:01:54 +03:00
• 4c5ef2cc4f setitem with arange fusion 1 (#5898) qazalandGitHub 2024-08-04 21:09:21 +08:00
• 59315ffc78 minor cleanup to UOp mod folding [run_process_replay] (#5895) chenyuandGitHub 2024-08-03 21:38:44 -04:00
• dad8e72ee9 hcq graph refactor (#5887) nimlgenandGitHub 2024-08-03 23:35:33 +03:00
• da61dea1b2 simple failed UOp sub symbolic test case (#5894) chenyuandGitHub 2024-08-03 14:27:23 -04:00
• 937bf5fe12 better hparam (#5891) Elias WahlandGitHub 2024-08-03 18:38:53 +02:00
• 37cc87ea75 save lines in the scheduler [run_process_replay] (#5890) qazalandGitHub 2024-08-03 19:20:11 +08:00
• 56ef9e453e pad reduceops to the max of each dimension (#5889) qazalandGitHub 2024-08-03 19:03:30 +08:00
• 65fa86901a indexing fusion 2 (#5888) qazalandGitHub 2024-08-03 18:13:39 +08:00
• af59b2eea9 tests from the indexing fusion branch (#5886) qazalandGitHub 2024-08-03 16:56:48 +08:00
• a77eab89ca UOp mod folding cleanup (#5885) chenyuandGitHub 2024-08-02 22:56:32 -04:00
• d5de44340e UOp add mod folding (#5862) chenyuandGitHub 2024-08-02 18:31:46 -04:00
• 714d00f325 hotfix: median > mean for sampling clock jitter geohot 2024-08-02 22:02:39 +00:00
• 7348c40d9d sampling time sync (8700 lines) (#5843) George HotzandGitHub 2024-08-02 14:44:35 -07:00
• 41bbd3f4c1 update UOp mod reduction patterns (#5883) chenyuandGitHub 2024-08-02 17:43:40 -04:00
• acadccf344 comma benchmark (#5518) wozeparrotandGitHub 2024-08-02 14:36:54 -07:00
• b4709d294a hotfix: hcq profiler use mid point for deps flow (#5882) nimlgenandGitHub 2024-08-02 23:53:10 +03:00
• 4a114756f6 New BERT dataloader (#5881) Elias WahlandGitHub 2024-08-02 21:12:23 +02:00
• 2777784b91 add dependency viewer to hcq profiler (#5874) nimlgenandGitHub 2024-08-02 22:07:01 +03:00
• 23e8c39288 get program fields in __post_init__ [run_process_replay] (#5878) George HotzandGitHub 2024-08-02 09:57:12 -07:00
• 8611fa6c99 apply opts.extra_matcher in process replay [run_process_replay] (#5877) qazalandGitHub 2024-08-02 23:07:58 +08:00
• 2a791f7924 fuzz uops is simpler with List[UOp] [run_process_replay] (#5875) qazalandGitHub 2024-08-02 22:28:15 +08:00
• 3995f1ddf1 move ops lds estimate to Program [run_process_replay] (#5872) George HotzandGitHub 2024-08-01 19:12:07 -07:00
• 877e0b4ba0 define global only has the index [run_process_replay] (#5869) George HotzandGitHub 2024-08-01 19:01:15 -07:00
• f27f949a5d Revert "revert some UOp IDIV bound (#5863)" (#5871) chenyuandGitHub 2024-08-01 21:38:31 -04:00
• df138bc558 Revert "revert a mod pattern (#5864)" (#5870) chenyuandGitHub 2024-08-01 20:44:26 -04:00
• 1b0314d9ef Revert "remove one more UOp mod pattern (#5865)" (#5868) chenyuandGitHub 2024-08-01 20:28:35 -04:00
• d73bc85ba9 UOpGraph not in renderer or Program [run_process_replay] (#5867) George HotzandGitHub 2024-08-01 16:20:30 -07:00
• b392b8edc3 increase atol and rtol test_gemm_fp16 (#5866) chenyuandGitHub 2024-08-01 19:09:58 -04:00
• b03b8e18c2 remove one more UOp mod pattern (#5865) chenyuandGitHub 2024-08-01 18:29:04 -04:00
• 5c8de2d044 revert a mod pattern (#5864) chenyuandGitHub 2024-08-01 17:24:26 -04:00
• 34168a64e3 optimize nv profiler (#5856) nimlgenandGitHub 2024-08-01 23:57:45 +03:00
• 2d3c7e4d4e some TestPickleJIT tests (#5860) George HotzandGitHub 2024-08-01 12:39:59 -07:00
• e347f10d33 hotfix: print which opencl device we are using geohot 2024-08-01 12:38:48 -07:00
• 0c8d202348 revert some UOp IDIV bound (#5863) chenyuandGitHub 2024-08-01 15:09:06 -04:00
• 53fcac9e80 hotfix: increase time on flaky NV test geohot 2024-08-01 10:19:22 -07:00