nimlgen/tinygrad
1
0
Fork 0
forked from tinygrad/tinygrad
Code Pull Requests Activity
10,931 Commits 334 Branches 15 Tags
6809ff8fe1654aae3b32d571cb97e893f89d14fd
Commit Graph
10931 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
geohot 6809ff8fe1 simplify priority 2025-11-06 07:57:59 -08:00
nimlgenandGitHub b9b68bf437 amd: add kern to sqtt event (#13126) 
* amd: add kern to sqtt event

* fix
 2025-11-06 22:02:02 +08:00
qazalandGitHub 88245d6579 qol improvements to sqtt decoder and timing tests (#13125) 2025-11-06 20:51:30 +08:00
nimlgenandGitHub dafdb4bfb1 test hcq open with pytest (#13124) 
* test hcq open with pytest

* fi
 2025-11-06 20:09:51 +08:00
nimlgenandGitHub 05e2ff4d87 system: fix flock on pcidevs (#13123) 
* system: fix locking of hcq devices

* rename and fullrun

* force ok

* fix

* fix
 2025-11-06 19:02:13 +08:00
qazalandGitHub 3126c89b84 viz: visible horizontal scrollbar in long texts (#13122) 2025-11-06 17:23:02 +08:00
George HotzandGitHub 91cc773397 add run count to toposort (#13119) 2025-11-05 22:29:34 -08:00
Adeeb ShihadehandGitHub dca7fb0a49 qcom: make priority configurable (#13120) 2025-11-05 22:27:54 -08:00
qazalandGitHub b2bb3af12a make range_color work in VIZ (#13121) 2025-11-06 14:26:48 +08:00
chenyuandGitHub f33c182393 test custom qkv kernel (#13118) 
adding the online softmax hits infinite loop so starting with this
 2025-11-05 23:32:13 -05:00
George HotzandGitHub c65e6d8887 add ranges to print_uops (#13116) 
* remove tuplize from linearizer

* try this

* simple priority

* add colored ranges to print_uops

* improve comments

* fix no const in src

* fix mypy

* fix define global

* fix var placement

* no prefer early load

* revert linearizer for now
 2025-11-05 20:26:56 -08:00
George HotzandGitHub 9b2b535fa4 fix issue with multi flip (#13115) 2025-11-05 15:28:50 -08:00
George HotzandGitHub 4027eef264 fix test warnings (#13114) 
* fix test warnings

* precommit passes

* ignore std_mean warning
 2025-11-05 15:06:29 -08:00
George HotzandGitHub bcfe42937f move permute/flip/shrink to mixins (#13113) 
* move permute to mixins

* move more stuff

* two more

* fix local mypy

* fix tests

* fix shrink
 2025-11-05 14:14:15 -08:00
George HotzandGitHub 2d4f01fda0 move mixins to mixin dir (#13105) 
* move mixins to mixin dir

* math
 2025-11-05 10:18:33 -08:00
chenyuandGitHub 52f0081e77 use where instead of mul in Embedding (#13112) 2025-11-05 12:49:01 -05:00
b1tgandGitHub edc4e1aede ignore trailing nops in llvm-objdump output (#13110) 2025-11-06 01:10:51 +08:00
chenyuandGitHub 03ee0cfe45 minor fast_idiv cleanup [pr] (#13109) 2025-11-05 11:44:36 -05:00
chenyuandGitHub 18d4ecc1f3 lower nv test_gemm_4096 target (#13107) 2025-11-05 11:05:16 -05:00
nimlgenandGitHub eff80beeed amd: props in device not sqtt (#13106) 
* amd: props in device not sqtt

* fix

* f

* fix

* fix
 2025-11-05 23:43:20 +08:00
nimlgenandGitHub 757ceab2a2 system: allow using vidmem for uc mem (#13104) 2025-11-05 19:12:59 +08:00
qazalandGitHub 8119d9f082 sqtt: decode each instruction exec (#13093) 
* sqtt: decode each instruction exec

* start tests

* run_asm

* capture sqtt per kernel

* chaining vgprs

* test things

* inst_execs in viz

* can also configure l and g

* 1l + cleanup

* test_sleep

* test_wmma

* work

* test sleep with llvm builtin
 2025-11-05 17:30:27 +08:00
chenyuandGitHub 54141e9cb9 DISABLE_COMPILER_CACHE=1 in speed_v_theoretical (#13096) 2025-11-04 11:28:18 -05:00
chenyuandGitHub 1c9f720654 remove unused type ignore [pr] (#13095) 2025-11-04 10:08:07 -05:00
nimlgenandGitHub c857dc5af0 autogen: try/except in try_dlopen (#13094) 
* autogen: try/except in try_dlopen

* ugh
 2025-11-04 22:51:53 +08:00
nimlgenandGitHub eaf7cbc178 amd: flush sqtt after each kernel (#13092) 
* amd: flush sqtt after each kernel

* merge for rgp
 2025-11-04 22:12:48 +08:00
qazalandGitHub 96417665e8 show sqtt decoder errs in viz (#13088) 
* show sqtt decoder errs in viz

* don't touch roc.py

* give hljs a default language

* work from tinyr9

* work
 2025-11-04 22:05:06 +08:00
nimlgenandGitHub 49191ada77 roc: install sqtt decoder (#13091) 
* roc: install?

* msg

* 0.1.4
 2025-11-04 18:56:01 +08:00
nimlgenandGitHub 16f1f644ba amd: remove sqtt=2 (#13090) 2025-11-04 18:29:24 +08:00
nimlgenandGitHub 2e97eaa866 roc: no nullptr when no wave instructions (#13087) 2025-11-04 17:32:14 +08:00
wozeparrotandGitHub 9c00c0688a tk fa: use 16x64 tiles (#13086) 2025-11-03 18:25:38 -08:00
wozeparrotandGitHub 4ed0f216b5 fix: make max_matmul run again (#13085) 2025-11-03 18:09:09 -08:00
chenyuandGitHub ca17718b6d remove symbolic_flat (#13083) 
* remove symbolic_flat

some kernels are different but sometimes it's better so not clear, will merge as long as benchmark passes

* test_location
 2025-11-03 17:25:21 -05:00
chenyuandGitHub fda720e013 simpler _is_balanced [pr] (#13082) 
returns False earlier
 2025-11-03 16:47:14 -05:00
chenyuandGitHub ddf01fdb15 revert mlperf.yml setting (#13080) 2025-11-03 15:24:13 -05:00
qazalandGitHub 6df34a5887 lint sqtt parser with mypy (#13079) 
* llvm address table errs

* mypy likes annotated dicts

* unwrap nullable
 2025-11-04 00:53:59 +08:00
qazalandGitHub 2d2040bc92 viz: tabulate sqtt (#13078) 
* viz: tabulate sqtt

* nomore asdict
 2025-11-04 00:03:15 +08:00
nimlgenandGitHub dfde3f54d9 rocprof: use llvm disasm (#13077) 
* rocprof: use llvm disasm

* rm
 2025-11-03 23:58:58 +08:00
qazalandGitHub 27d42fd575 sqtt decoder print behind DEBUG>=5 (#13076) 
* sqtt decoder print behind DEBUG>=5

* gfx version stuff also behind 5
 2025-11-03 23:20:03 +08:00
George HotzandGitHub 416b15cc59 improve uop matmul syntax (#13074) 
* improve uop matmul syntax

* store takes const

* copy

* cleanups

* faster and simpler

* label them reduce

* better syntax

* touchup
 2025-11-03 21:34:26 +08:00
nimlgenandGitHub 08855c162b amd: correct sqtt_read for several xccs (#13075) 
* amd: correct sqtt_read for several xccs

* default mask
 2025-11-03 19:59:56 +08:00
qazalandGitHub 1c0d4f1cd2 viz: counters loader (#12987) 
* standalone custom loader

* first iteration on the ui

* work

* add center helper

* add edge offsets

* enumerate all edge types

* try dagre layout algorithm

* simpler spec

* bring back double edges

* more work on edge paths

* aesthetics

* custom edges also works

* dimmer inactive links

* cleanup

* cleanup

* split out the ncu layout

* this is just a k/v map now

* rm that

* more cleanup and comments

* do work

* also this work

* simpler start

* rm that

* sqtt work

* view sqtt

* sqtt

* --custom is just in profile

* wrap c call

* from tinygrad install

* eg. module not found
 2025-11-03 19:42:36 +08:00
George HotzandGitHub 1e3d6e49a6 index slicing + allclose (#13071) 
* continue work on slicing+allclose

* Revert "Revert "slicing + allclose""

This reverts commit 6c7a12f21c.

* fix tests + better syntax

* forgot an after

* slot is an integer
 2025-11-03 13:01:48 +08:00
geohot 6c7a12f21c Revert "slicing + allclose" 
This reverts commit c9a1e35b1e.
 2025-11-03 12:05:44 +08:00
geohot c9a1e35b1e slicing + allclose 2025-11-03 12:00:45 +08:00
chenyuandGitHub a317d6e625 extra/amdpci/setup_python_cap.sh (#13070) 2025-11-02 19:19:36 -05:00
chenyuandGitHub ad501ce50a mlperf cron install tqdm (#13069) 
one more...
 2025-11-02 18:09:27 -05:00
chenyuandGitHub 2c8d619147 mlperf cron install influxdb3-python (#13068) 2025-11-02 17:55:40 -05:00
chenyuandGitHub 4c22f089fc mlperf cron install tensorflow try 2 (#13067) 2025-11-02 17:11:01 -05:00
chenyuandGitHub c58cf91850 mlperf cron install tensorflow (#13066) 2025-11-02 16:48:05 -05:00