github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-04-29 03:00:14 -04:00
Code Issues Packages Projects Releases Wiki Activity
12,455 Commits 248 Branches 14 Tags
5ff278446cf4a413be5ae0e66ea1097072cbbe43
Commit Graph

1661 Commits

Author SHA1 Message Date
qazal
b8a55d5f68 sqtt: new packet types, add discovery script (#14960) 2026-02-28 04:27:27 +09:00
qazal
448e997be4 gemm/asm: cleanup custom function args (#15007) 2026-02-25 22:05:56 +09:00
wozeparrot
8d9545e09e llama3: correctly shard wqkv (#14978) 2026-02-23 23:57:10 -08:00
wozeparrot
25565b2410 fa: test for mp (#14907) 2026-02-22 21:47:36 -08:00
qazal
d6145736c7 sqtt: examples generator changes from inst_discovery (#14961) 
* sqtt examples generator changes from inst_discovery

* rdna4

* rdna3

* cdna

* sad reality for mi300x
 2026-02-23 14:42:48 +09:00
George Hotz
8ef5544e4a realized PYTHON copies (#14934) 
* realized PYTHON copies

* comment that out

* fix that test

* append afters

* contig

* disk copies

* should be 124

* 332
 2026-02-21 20:29:31 +08:00
George Hotz
55d3a5def9 preallocate all realized buffers (#14823) 
* preallocate all realized buffers

* contiguous

* work

* comment that out

* move to schedule

* better

* correct fix

* just buffer

* disk bufs

* fixes disk tensor stuff

* fix symbolic stuff

* fix multi

* 162 failures

* bugfixes

* don't check that anymore

* fix schedule tests

* mnist should be contiguious

* type and buffer

* fix tests

* shrink axis correction

* mypy fixes

* tests skips

* same 37 failures

* dedup

* no shrink in the graph

* 29 failures

* skips

* fix custom kernel

* fix training

* those optimizations aren't supported currently

* simpler

* more correct

* tests

* 14 failures

* works

* fix that test

* broken

* 11 failures

* only kernel counts left

* fixes

* all tests pass

* remove tensor_map

* op test

* 200 -> 230

* test fixes

* fixes

* revert test_tiny thing

* guard

* revert that

* test tiny passes

* no contigs there

* base realize back

* Revert "no contigs there"

This reverts commit c45bb9fcfd.

* revert that

* chop many assigns

* 12 failures

* fix tests

* tests

* apply after

* pre-commit

* remove old code

* delete that

* fix types

* remove extra contig

* fix dataloader

* torch fix

* disk fix

* update kernel fusion numbres

* runs on amd

* restore kernel count

* add that rule back

* that

* disable that

* wrong

* add the correct rule for that folding

* more tests

* guard c1.arg

* no newlines

* realize those

* split into a different file

* remove detach/contig back

* skip 2

* update that
 2026-02-20 20:05:54 +08:00
qazal
32f569b573 viz/sqtt: decoder fixes pre rdna4/cdna4 work (#14900) 
* viz/sqtt: decoder fixes pre rdna4/cdna4 work

* fix

* branch_inst + more tests

* smaller
 2026-02-20 12:10:15 +09:00
wozeparrot
9317e96881 fa: explicitly pass shapes (#14857) 2026-02-19 05:26:16 -08:00
nimlgen
3b95fa0ed4 am_smi: enable mem usage back (#14858) 2026-02-18 19:27:27 +03:00
wozeparrot
6d301ad2c4 feat: llama wqkv (#14841) 2026-02-17 23:01:33 -08:00
wozeparrot
95e97ec341 seperate llama optim (#14810) 2026-02-17 13:02:35 -08:00
qazal
f8e485ee9e nvcc/nvdisasm macos shim (#14822) 
* move to backend

* and arch

* setup_nvcc_osx

* blackwell

* min test

* now getting dumb assert is_ptx

* support cubin.

* work

* remove that

* simpler
 2026-02-17 20:07:05 +09:00
qazal
f590564bf7 gemm multiple is only for cdna4 asm (#14814) 
* gemm multiple is only for cdna4 asm

* move to backend

* and arch

* path
 2026-02-17 14:00:02 +09:00
George Hotz
5bd2862d1a late compile the cdna gemm (#14783) 
* late compile the cdna gemm

* remove old things

* finalize inplace

---------

Co-authored-by: qazal <qazal.software@gmail.com>
 2026-02-17 13:04:22 +09:00
George Hotz
f081f154ae parameterize the CDNA asm gemm (#14813) 
* parameterize the CDNA asm gemm

* fix llama test

* fix

* add more gemmt ests

* confirm all match

* test these asm gemms
 2026-02-17 11:35:18 +08:00
nimlgen
131bbbbfd8 am: smu_v13_0_12 (#14800) 2026-02-16 22:58:10 +03:00
wozeparrot
45aebe1572 hipkittens fa backward (#14723) 2026-02-16 00:38:44 -08:00
qazal
c7a4dbf918 viz: get program binary from the UOp (#14787) 
* viz: get program binary from the UOp

* remove that

* less

* rename View Program to View Source

* two words

* fix
 2026-02-16 15:46:58 +09:00
George Hotz
dff9cf35c2 amd asm emulator fixes + run it in CI (#14786) 
* amd asm fix, try 2

* fix tests
 2026-02-16 13:24:21 +08:00
qazal
55a4dfa2e0 cdna4 asm_gemm tests in CI on the null backend (#14785) 
* cdna4 asm_gemm tests in CI on the null backend

* no .numpy() in null

* better

* gemm/asm: device comes from renderer
 2026-02-16 14:06:23 +09:00
George Hotz
ac079e43d7 ElementwiseMixin (#14777) 2026-02-16 08:50:47 +08:00
qazal
33b31d9cd6 tinykittens flash attention dtype fix, add CI (#14770) 
* don't hardcdoe amd device

* add failing tests, ci too

* fix: fix for dtype mixin

* bump to rocm 7.1

---------

Co-authored-by: Woze Parrot <wozeparrot@gmail.com>
 2026-02-16 01:15:11 +09:00
qazal
9bb6014900 keep existing profile trace in viz cli (#14757) 2026-02-15 13:16:32 +09:00
nimlgen
4ab51b55bd stream pma decoder (#14746) 2026-02-14 17:40:18 +03:00
George Hotz
c0de4f75b1 improve mmapeak, print names with sqtt (#14726) 2026-02-13 16:07:06 +08:00
wozeparrot
0613c0ac0c hipkittens fa forward (#14692) 2026-02-12 20:16:43 -08:00
George Hotz
4088d686b2 remove llvm requirement from amd (#14717) 
* remove llvm requirement from amd

* tests pass

* test

* sink kernarg_size

* move stuff

* amd_asm_matmul to new style

* default type

* fix tests, simpler

* cu mode is faster and simpler

* darken
 2026-02-13 10:50:12 +08:00
chenyu
557134e1c7 model/test fix that failed with WEBGPU=1 DEBUG=2 (#14706) 2026-02-12 09:08:16 -05:00
George Hotz
4680247e35 renderer/amd: move in tree (#14702) 
* renderer/amd: move in tree

* fix paths in tests

* 24000 lines

* no delete for amd files
 2026-02-12 18:09:16 +08:00
George Hotz
d5fc3ea1ba assembly/amd: mypy+ruff passes (#14701) 
* assembly/amd: mypy+ruff passes

* touchups
 2026-02-12 16:59:42 +08:00
George Hotz
025049c521 clean up sqtt / update src formatting in viz (#14696) 
* update src formatting in viz

* rename to RDNA3/RDNA4 in sqtt

* wrap

* move sqttmap

* update readme

* why did that change?

* cdna

* that's just for test
 2026-02-12 14:27:14 +08:00
George Hotz
befc1e800c assembly/amd: disasm is test only (#14694) 
* assembly/amd: disasm is test only

* viz uses str
 2026-02-12 12:33:46 +08:00
George Hotz
c331798201 move tests to test/backend (#14691) 
* move tests to test/backend

* fix imports

* fix CI

* revert that one

* Fix formatting in README for test command
 2026-02-12 11:09:44 +08:00
George Hotz
3fab43c57c add cache to asm gemm (#14675) 2026-02-11 08:26:30 +08:00
wozeparrot
69574542ab fix: use correct fa implementation in eval (#14651) 2026-02-09 18:20:44 -08:00
qazal
80b0119cef llama: add new asm gemm shape (#14611) 
* llama: add new asm gemm shape

* work

* cleanup

* half dtype

* more comment
 2026-02-10 00:34:29 +09:00
nimlgen
e087c58ae0 print tables in llama/profile.sh (#14639) 2026-02-09 12:32:54 +03:00
nimlgen
01a4ee4d66 do not hive_reset when amdgpu (#14624) 2026-02-08 19:14:13 +03:00
George Hotz
183d38b128 remove CUSTOM_KERNEL / directly construct it (#14604) 
* remove CUSTOM_KERNEL / directly construct it

* clean that up

* simpler multi

* custom kernel spec

* remove Kernel

* fix multi

* use sharded shape

* explicit regression test
 2026-02-08 18:43:33 +08:00
nimlgen
e29a88ca09 hive_reset respects lock (#14618) 2026-02-08 10:47:25 +03:00
wozeparrot
d87ae1c84c feat: tinyfs load test in benchmark (#14602) 2026-02-06 18:00:00 -08:00
nimlgen
fbb67a3f95 am_smi: fix after regen (#14594) 2026-02-06 20:57:41 +03:00
qazal
b7e3fbe07e llama: add VIZ=-1 to dev_run (#14583) 
* llama: add VIZ=-1 to dev_run

* readme

* cleaner

* add profile.sh script

* better grouping of options

* add other row

* readme edits

* work
 2026-02-06 22:59:22 +09:00
nimlgen
fbeb978170 diff devices for sdma (#14589) 
* start

* x

* fix

* sdma

* c

* clean

* x

* hm

* cleaer
 2026-02-06 16:39:12 +03:00
qazal
cf73d7e2a7 hotfix: disable slower asm gemm shape from llama seqlen 8192 (#14582) 2026-02-06 15:05:19 +09:00
qazal
be77873974 llama: contig backward for wk / wv matmul backward (#14581) 2026-02-06 14:54:00 +09:00
wozeparrot
f73468d516 fa: block skipping for fa kv bwd (#14569) 2026-02-05 16:13:53 -08:00
chenyu
41a179f542 fix test_xlm_roberta_large (#14564) 
onnxruntime does not allow symlink that's outside model dir. update snapshot_download to use local_dir instead of cache_dir. some ad hoc migration step to copy the existing model too
 2026-02-05 14:56:06 -05:00
qazal
190042358f llama: faster bf16 matmul / rope backward (#14558) 2026-02-05 23:57:25 +09:00