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

• 22af7132cd fix test_dev_jitter_matrix (#14255) nimlgenandGitHub 2026-01-20 20:07:51 +03:00
• c7fbd177d4 USBGPU: debug script for comma chestnut (#14252) Robbe DerksandGitHub 2026-01-20 16:52:25 +01:00
• 26f8b12e01 Whisper audio helpers (mel filters in tinygrad) (#13478) 3 people 2026-01-20 17:50:02 +02:00
• dc82856084 tbgpu: shim binary + remote apl pci dev (#14124) nimlgenandGitHub 2026-01-20 16:15:18 +03:00
• 4548fcc1b8 amd/sqtt: add rdna4 and cdna sqtt examples (#14251) qazalandGitHub 2026-01-20 07:11:48 -05:00
• 2dc281b32a assembly/amd: test helpers for arch to gfx target mapping (#14250) qazalandGitHub 2026-01-20 06:35:09 -05:00
• 823e88c0d0 nv: request bar 3 (#14249) nimlgenandGitHub 2026-01-20 13:52:38 +03:00
• dddd0e384f ALLOW_DEVICE_USAGE=0 in codegen (#14238) qazalandGitHub 2026-01-20 01:15:16 -05:00
• 0243f4a0f1 clear wins from ucode branch (#14243) George HotzandGitHub 2026-01-20 15:11:09 +09:00
• 5e24643889 minor import speedups (#14244) George HotzandGitHub 2026-01-20 15:05:36 +09:00
• d60a155e48 defer compilation of upats (#14242) George HotzandGitHub 2026-01-20 13:50:00 +09:00
• 56c8926d32 import speedups: refactor validate to late import (#14241) George HotzandGitHub 2026-01-20 13:23:39 +09:00
• 9d3b1cf1e7 simpler _cached_to_python_const (#14236) chenyuandGitHub 2026-01-19 23:10:53 -05:00
• b1c5a242b7 Revert "move is_dtype_supported logic to renderer (#14188)" (#14237) qazalandGitHub 2026-01-19 22:19:14 -05:00
• 1f89eaf790 tk: fa bert mask fix + some numerical stability improvements (#14214) wozeparrotandGitHub 2026-01-19 19:18:07 -08:00
• 9ea63d7d52 failed test case for onnx IF with jit (#14235) chenyuandGitHub 2026-01-19 18:10:05 -05:00
• b65dc9fd8e refactor: use generic type for ContextVar [pr] (#13998) 3 people 2026-01-19 10:37:54 -08:00
• 7010c176cf pre commit: fix path to test_assign.py (#14231) Martin SzewieczekandGitHub 2026-01-19 19:36:30 +01:00
• 34f6192739 look for cuda in /opt/cuda (#14230) sirhcmandGitHub 2026-01-19 08:51:00 -08:00
• 0f61cbd51f viz: draw shapes directly on the canvas (#14229) qazalandGitHub 2026-01-19 10:57:06 -05:00
• acb0045ba0 system: alloc_sysmem is part of interface (#14226) nimlgenandGitHub 2026-01-19 18:15:54 +03:00
• ab426cb671 viz: simplify row line logic (#14227) qazalandGitHub 2026-01-19 10:00:28 -05:00
• 01653db4fd nv: GPPut is mmiointerface (#14225) nimlgenandGitHub 2026-01-19 17:36:26 +03:00
• 7cb7abeeb0 amd: fix scratch_wave64_lane_byte_size (#14223) nimlgenandGitHub 2026-01-19 15:21:39 +03:00
• 979ce211f7 amd: missing self in aql's exec (#14224) nimlgenandGitHub 2026-01-19 14:27:54 +03:00
• 31bcbed6bb AMD_DISABLE_SDMA for testing with -n12 (#14216) George HotzandGitHub 2026-01-19 16:10:30 +09:00
• 578a4a50d3 viz: row lines in timeline (#14213) qazalandGitHub 2026-01-18 23:01:43 -05:00
• 161fee9a48 move is_dtype_supported logic to renderer (#14188) sirhcmandGitHub 2026-01-18 19:37:04 -08:00
• 7abe9b020f viz: add border colors to pkts timeline (#14211) qazalandGitHub 2026-01-18 21:37:46 -05:00
• 67d9712ef6 jit copy aliased output if it's read later (#14210) chenyuandGitHub 2026-01-18 18:48:59 -05:00
• 97333b1954 jit footguns test case on assign with same buffer outputs (#14209) chenyuandGitHub 2026-01-18 16:01:09 -05:00
• e7c2df9113 improve consecutive Tensor indexing (#14208) chenyuandGitHub 2026-01-18 15:14:33 -05:00
• c7b8f6496f remove dtypes.index_like and dtypes.fields [pr] (#14207) chenyuandGitHub 2026-01-18 11:49:01 -05:00
• e27a0002c5 viz: only keep the sqtt bytes for pkts (#14203) qazalandGitHub 2026-01-18 03:04:26 -05:00
• d8f87ae2f2 SQTT packets to assembly mapper (#14198) qazalandGitHub 2026-01-18 02:32:13 -05:00
• 1eb110cd7d fix memory corruption in NIR, reenable process replay (#14204) sirhcmandGitHub 2026-01-17 23:05:12 -08:00
• a51e0a86db assembly/amd: clean up disasm.py + add CDNA support (#14200) George HotzandGitHub 2026-01-18 14:48:44 +09:00
• 4b18c92bc5 simpler Context.__enter__ [pr] (#14201) chenyuandGitHub 2026-01-18 00:38:59 -05:00
• feaa804158 skip lvp process replay in CI [pr] (#14202) qazalandGitHub 2026-01-17 23:25:04 -05:00
• b12a9fea80 runtime int call instead of cast(int) (#14183) chenyuandGitHub 2026-01-17 20:34:45 -05:00
• 79c1559f69 amd asm can still be simpler (#14199) George HotzandGitHub 2026-01-17 18:40:10 +09:00
• a5afdcc79b add lds double buffering to amd_asm_matmul lds_double_buffering geohot 2026-01-17 16:26:07 +09:00
• 5e6a72c33f new Onnx Gather (#14187) chenyuandGitHub 2026-01-16 22:24:07 -05:00
• 9f7f2f0e0c MAX_SQTT_PKTS geohot 2026-01-17 12:05:36 +09:00
• 50554115ee fix VALU_SALU / IMMED_MASK and improve amd_asm_matmul (#14196) George HotzandGitHub 2026-01-17 11:58:34 +09:00
• ab244c7f81 onnx Gather should not assume indices to be const (#14185) chenyuandGitHub 2026-01-16 20:55:00 -05:00
• a879b54234 tk: fa jit fix (#14170) wozeparrotandGitHub 2026-01-16 16:38:45 -08:00
• a8ae9757dd viz: put alts in the same row, LDS color (#14194) qazalandGitHub 2026-01-16 19:36:14 -05:00
• 5aa71f437b viz: precise clock cycles in PKTS (#14179) qazalandGitHub 2026-01-16 19:08:13 -05:00
• eafcd44d95 fix OSX image pitch (#14193) sirhcmandGitHub 2026-01-16 16:07:33 -08:00
• 3960e2758c suppress_finalizing in hip (#14189) sirhcmandGitHub 2026-01-16 15:56:29 -08:00
• 9302ab003a viz: show ALT/OTHER packets on second lane (#14192) qazalandGitHub 2026-01-16 18:55:24 -05:00
• 551454f476 viz: fix wave sort, show message if sqtt trace is empty (#14190) qazalandGitHub 2026-01-16 18:01:26 -05:00
• 8a2549d42b improve amd_asm_matmul + minor VIZ PKTS improvements (#14186) George HotzandGitHub 2026-01-17 06:56:59 +09:00
• 7d1d9d4568 assembly/amd: remove IMG instruction support and asm.py (#14163) George HotzandGitHub 2026-01-17 06:21:50 +09:00
• dc4ae7dd08 lower ASSERT_MIN_STEP_TIME for driving_policy to 3ms (#14184) chenyuandGitHub 2026-01-16 15:04:53 -05:00
• 0a14e1fcd4 fix some type ignore (#14182) chenyuandGitHub 2026-01-16 13:56:45 -05:00
• fc10470883 add UOp.__index__ (#14181) chenyuandGitHub 2026-01-16 12:28:33 -05:00
• 6790165ef8 minor _apply_uop cleanup (#14180) chenyuandGitHub 2026-01-16 11:27:55 -05:00
• e855ec8ee3 tbgpu: refactor dext to support user mappings (#14177) nimlgenandGitHub 2026-01-16 15:55:57 +03:00
• bbc55962ee viz: color SQTT INST Ops like UOps (#14175) qazalandGitHub 2026-01-16 07:24:43 -05:00
• 3751b29a3d viz: skip OTHER_ SQTT packets (#14178) qazalandGitHub 2026-01-16 06:37:19 -05:00
• 7c1f1cb2bc viz: fix INST packets coloring (#14176) qazalandGitHub 2026-01-16 04:46:13 -05:00
• 1696991988 viz: add PKTS group to sqtt trace (#14173) qazalandGitHub 2026-01-16 03:29:47 -05:00
• a021b84604 autogen: fix enum (#14171) sirhcmandGitHub 2026-01-15 22:30:11 -08:00
• fa5475307c viz: collapse wave packets in one row, 1 clk per packet (#14169) qazalandGitHub 2026-01-15 23:52:07 -05:00
• 5abc262e22 fix dll.bind caching (#14168) sirhcmandGitHub 2026-01-15 17:25:42 -08:00
• f9ca072b61 cuda compilers disassemble properly (#14166) sirhcmandGitHub 2026-01-15 16:02:40 -08:00
• 14e9a71a41 move test_assign to unit (#14165) chenyuandGitHub 2026-01-15 17:10:13 -05:00
• a0dd9d2146 tbgpu: correct com.apple.developer.driverkit.transport.pci entitlements (#14164) nimlgenandGitHub 2026-01-15 20:56:39 +03:00
• 32e1c267ee viz: SQTT timeline with our decoder (#14139) qazalandGitHub 2026-01-15 06:45:16 -05:00
• 0cb024a5bb remove ctypes.Structure (#13651) sirhcmandGitHub 2026-01-15 02:06:22 -08:00
• 255e0573b1 assembly/amd: clean up asm/disasm (#14158) George HotzandGitHub 2026-01-15 17:45:40 +09:00
• 164bc678a6 scheduler: sched_cache bugfix for different Tensor.custom_kernel schedules (#14161) qazalandGitHub 2026-01-15 00:59:14 -05:00
• b46da603fe codegen/custom_kernel: do not attach KernelInfo to user program (#14160) qazalandGitHub 2026-01-15 00:01:48 -05:00
• fd60626ea1 assembly/amd: refactor to use op_bits/op_regs (#14156) George HotzandGitHub 2026-01-15 11:20:21 +09:00
• add7da268f multiple slice assign test (#14157) chenyuandGitHub 2026-01-14 21:08:03 -05:00
• e9ce12028e assembly/amd: amdxml cleanups, remove broken SDWA/DPP, merge in pdf.py (#14154) George HotzandGitHub 2026-01-15 09:23:19 +09:00
• 7e5687f6a3 more fa multi fix (#14152) wozeparrotandGitHub 2026-01-14 13:57:11 -08:00
• 1381daac06 many more failed assign tests (#14153) chenyuandGitHub 2026-01-14 16:20:28 -05:00
• 8c55ef4f01 amd: cleanup props (#14145) nimlgenandGitHub 2026-01-14 20:27:41 +03:00
• 899a56446e failed assign test cases with write before read (#14148) chenyuandGitHub 2026-01-14 10:30:50 -05:00
• 986e865830 fix TINY_BACKEND=1 cumsum (#14138) chenyuandGitHub 2026-01-14 09:54:49 -05:00
• 434dbafab5 optional Estimates in KernelInfo (#14147) qazalandGitHub 2026-01-14 08:55:03 -05:00
• 76b577ee76 viz: only SIMD name in sqtt timeline rows (#14146) qazalandGitHub 2026-01-14 06:13:27 -05:00
• e5500ae4ad add ALU stuff to default perf counters (#14135) George HotzandGitHub 2026-01-14 19:47:59 +09:00
• 86708ccac5 hip_ioctl: dump aql (#14142) nimlgenandGitHub 2026-01-14 13:15:10 +03:00
• f9147422a3 ci: add setcap (#14143) nimlgenandGitHub 2026-01-14 13:15:01 +03:00
• 62c1a014a6 amd: rename to be consistent (#14141) nimlgenandGitHub 2026-01-14 11:41:04 +03:00
• e0eea0d833 autogen: verify all files in CI (#14140) sirhcmandGitHub 2026-01-13 23:35:54 -08:00
• 2a2c1eacf6 disable fast_idiv on metal (#14137) chenyuandGitHub 2026-01-13 21:40:40 -05:00
• a92778aa0c tk: fa multi fix (#14134) wozeparrotandGitHub 2026-01-13 17:22:15 -08:00
• 2ab18ea7e3 assembly/amd: use xml instead of pdf (#14118) George HotzandGitHub 2026-01-14 10:03:37 +09:00
• 002ea39da7 assembly/amd: use Tensor.custom_kernel to run assembly (#14125) qazalandGitHub 2026-01-13 18:29:25 -05:00
• fe00682502 clean up svd tests (#14133) chenyuandGitHub 2026-01-13 16:32:21 -05:00
• 84b88a0a31 more doc of newly added functions (#14132) chenyuandGitHub 2026-01-13 15:48:45 -05:00
• e610821c52 Tensor.cummin and Tensor.nonzero (#14131) chenyuandGitHub 2026-01-13 15:09:56 -05:00
• 176a934ddd Tensor.diagonal support offset and dims (#14130) chenyuandGitHub 2026-01-13 14:49:06 -05:00
• 2a217ba206 tinybackend isin and log10 (#14120) chenyuandGitHub 2026-01-13 14:14:09 -05:00
• 79d00521f8 viz: fix cfg err when endpgm is in the middle of stream (#14128) qazalandGitHub 2026-01-13 12:00:34 -05:00