sirhcm/tinygrad
1
0
Fork 0
forked from tinygrad/tinygrad
Code Pull Requests 2 Actions 3 Activity
14,537 Commits 336 Branches 15 Tags
explicit-dev
Commit Graph
6198 Commits
Author SHA1 Message Date
chenyuandGitHub 7204d46786 delete dtype_from_uop INDEX exempt (#17674) 2026-08-21 21:08:06 -04:00
George HotzandGitHub af242819d8 refactor the AMD emulator slop (kimi) (#17673) 
* refactor the AMD emulator slop (kimi)

* mypy
 2026-08-21 18:00:58 -07:00
chenyuandGitHub daa154aa22 FLOORDIV to SHR for powers of 2 [pr] (#17669) 2026-08-21 16:28:04 -04:00
George HotzandGitHub 3082956a17 usb copyin: async arm and drain, 323 MB/s on comma (#17663) 
* usb copyin: can safely arm before drain

* perf counter

* 294 MB/s for comma

* free speed with async transfers
 2026-08-21 13:12:03 -07:00
chenyuandGitHub 8f9cbdf0cc few more self folding [pr] (#17657) 2026-08-21 15:21:28 -04:00
nimlgenandGitHub 1cf8a2c7fe hcq2: use shrink.bitcast (#17653) 
* hcq2: shrink.bitcast

* x

* x

* x

* s

* x

* Dx

* Revert "hotfix: disable HCQ2"

This reverts commit a57188ea6d.

* x
 2026-08-21 21:24:34 +03:00
chenyuandGitHub 3919ce8427 ceildiv in wgsl _packed_size [pr] (#17662) 2026-08-21 14:02:20 -04:00
George HotzGitHubtiny
 		756e82e055 usb amd: pipelined copyin over the 0xF2 engine (2.6x faster) (#17628) 
Stream 240KB chunks into two alternating 256KB SRAM bounce windows; each chunk
ends in a unique 512B sentinel that a prebuilt SDMA ring polls before copying
the chunk to VRAM, followed by an in-order drain fence that the host waits on
before re-arming a window. No timing assumptions in either direction: the
sentinel is in-stream proof of data landing, the fence proves a full drain.

Adds a small pooled async bulk-OUT layer to USB3 so staging the next chunk
overlaps the wire, and a slot_start parameter to scsi_write for the second
window. 107 -> 276 MB/s copyin on tinyc8 (Kryo-3XX host).

Co-authored-by: tiny <tiny@local>
 2026-08-21 10:38:50 -07:00
chenyuandGitHub 554d078ac4 fix render_marg [pr] (#17656) 2026-08-21 10:03:22 -04:00
qazalandGitHub 1c3c9e96f6 remove setUp from test_sqtt_profiler (#17652) 
* remove setUp from test_sqtt_profiler

* remove that

* cleaner

* do not need that
 2026-08-21 17:29:11 +09:00
qazalandGitHub e8a8d99b99 profiler tracing using unique keys (#17651) 
* profiler tracing using unique keys

* cleaner + comment
 2026-08-21 17:06:01 +09:00
qazalandGitHub dcc2d021e7 prereq viz cleanups for unique profile keys (#17649) 
* cleaner

* just use VIZ=-2

* better
 2026-08-21 16:18:25 +09:00
qazalandGitHub 80bf60d782 profiler failing test for non unique kernel names (#17647) 
* err

* expected
 2026-08-21 13:52:35 +09:00
George HotzandGitHub 1cb0600086 fix llm vars regression (kimi) (#17644) 
* fix llm regression (kimi)

* unneeded

* more principled
 2026-08-20 21:31:43 -07:00
qazalandGitHub d716d0d927 hotfix: qkv bw kernel requires hipcc and cdna4 (#17645) 2026-08-21 12:38:45 +09:00
b1tgandGitHub 9216aa494c llm prefill failing test (#17630) 2026-08-20 18:58:34 -07:00
George HotzandGitHub 9aa9e11301 compile kernels in parallel (#17629) 
* compile kernels in parallel (slop)

* cleanups

* fixes

* hand cleanups

* no PARALLEL with VIZ

* ugh, we need to remove the n from names

* no hcq2 in that test

* main for abstractions3

* fix compile server to be fork safe

* fix num cpu threads in context

* just use a lock

* fix __main__ on spawn

* compileonce

* python3 is double

* xdist sizing

* parallel compile

* fetch supports extract

* fix issues

* revert compiler server to master, drop compileonce

* normal workflow

* PARALLEL=0 for QCOMCL compiletest
 2026-08-20 16:41:45 -07:00
George HotzandGitHub 3fdbb82bfe add ansipad and PARALLEL contextvar (#17642) 2026-08-20 16:26:02 -07:00
chenyuandGitHub 592e3f8363 update const selector where folding [pr] (#17640) 
folding a strong dtype WHERE to a weak const branch keeps the strong dtype
 2026-08-20 18:49:34 -04:00
sirhcmandGitHub 3715006a21 fix float_to_bf16 on non-float32 inputs (#17638) 2026-08-20 18:14:30 -04:00
sirhcmandGitHub c773891e3f skip INT_MIN % -1 on X86 and LLVM (#17637) 2026-08-20 17:26:26 -04:00
chenyuandGitHub 0e7ab863a0 x86 REX issue (#17580) 
* failing test

* fix
 2026-08-20 16:49:04 -04:00
George HotzandGitHub 1707dca3b4 remove kernel_cnt, names are no longer unique (#17633) 
* remove kernel_cnt, names are no longer unique

* check uops, not names

* fix hcq2
 2026-08-20 13:28:25 -07:00
chenyuandGitHub a8ecb73363 x64 imm uint64 (#17636) 
* x64 imm uint64

* fix
 2026-08-20 15:30:17 -04:00
nimlgenandGitHub 8c3cb00d36 hcq2: staging (#17622) 2026-08-20 13:42:48 +03:00
qazalandGitHub a1263fadf3 fused_qkv_rope in UOp try 2 (#17619) 
* fused_qkv_rope in UOp try 2

* dont need that

* less
 2026-08-20 12:28:34 +09:00
chenyuandGitHub e6324d1e1c test updates from weak const branch (#17618) 2026-08-19 22:56:25 -04:00
sirhcmandGitHub 2067133732 cpu: link with rt (#17608) 2026-08-19 17:55:08 -04:00
chenyuandGitHub fc214da417 test updates for weak const change (#17606) 2026-08-19 15:56:01 -04:00
chenyuandGitHub b8cc74ecf8 no float in tensor shape [pr] (#17605) 2026-08-19 15:40:16 -04:00
BennettGitHubBennett
 		7064e76bc8 fix roll on zero-sized tensors (#17603) 
Signed-off-by: Bennett <[email protected]>
Co-authored-by: Bennett <[email protected]>
 2026-08-19 15:31:21 -04:00
chenyuandGitHub a4fadcf606 fix TestDevCopySpeeds SIZE (#17602) 
SIZE should be int
 2026-08-19 14:53:17 -04:00
chenyuandGitHub c218b4842d fold_bitcast should truncate its input [pr] (#17601) 2026-08-19 14:39:36 -04:00
chenyuandGitHub bd6e70ac15 delete stale tests (#17596) 2026-08-19 11:18:05 -04:00
chenyuandGitHub 9550378704 finish casted_consts migration [PR] (#17595) 2026-08-19 10:43:55 -04:00
chenyuandGitHub b3e2f17b24 update NULL tests that depends on strong dtype CONST (#17594) 2026-08-19 10:26:38 -04:00
chenyuandGitHub e8ba214b56 casted CONST migration for x86 [pr] (#17592) 
* casted CONST migration for x86 [pr]

* style
 2026-08-19 09:38:09 -04:00
chenyuandGitHub 68b4407fe3 casted CONST migration for cstyle [pr] (#17587) 2026-08-19 09:01:40 -04:00
qazalandGitHub d539aaf752 Revert "fused_qkv_rope in UOp (#17591)" (#17593) 
This reverts commit 8c2bf02d17.
 2026-08-19 21:42:55 +09:00
qazalandGitHub 8c2bf02d17 fused_qkv_rope in UOp (#17591) 
* llama: 4% faster fused_qkv_rope

* prep

* add uop kernel, has_hipcc is cached

* less
 2026-08-19 18:07:30 +09:00
George HotzandGitHub c31038ff37 use KernelCountException when kernel count is being compared (#17584) 2026-08-18 16:06:03 -07:00
chenyuandGitHub 49778d9a48 start renderer casted const migration [pr] (#17582) 
before rendering, rewrite strong typed const to casted weak const and have renderer adopt the new UOp. starting with PYTHON
 2026-08-18 17:43:27 -04:00
chenyuandGitHub a1366e2f6c alu(long, weakint) can do math in int too [pr] (#17579) 
* alu(long, weakint) can do math in int too [pr]

* remove
 2026-08-18 09:08:35 -04:00
George HotzandGitHub 8d2cc64b69 llm: refactor delta attention (#17564) 
* refactor delta attention

* cleanups

* bugfixes

* stack

* recurrent w chunk_size 1

* revert that

* extra test
 2026-08-17 19:24:03 -07:00
chenyuandGitHub 34c9b9d434 add back cast where rule [pr] (#17572) 2026-08-17 15:27:20 -04:00
b1tgandGitHub b757437f64 llm: respect expert_gating_func (#17458) 
* llm: respect expert_gating_func

* test

* enum

* clean
 2026-08-17 12:19:33 -07:00
b1tgandGitHub 2776c5b369 fix call arg indexing in shard scheduling (#17519) 2026-08-17 09:57:50 -07:00
nimlgenandGitHub 58edff61d9 hcq2: one submitter (#17556) 
* hcq2: c submitter

* x

* x

* x

* simpler

* simpler

* x

* x

* Dx

* revrt

* Dx

* x

* fst

* fix
 2026-08-17 16:08:19 +03:00
chenyuandGitHub 42714e1399 update a few is CONST check to check device None [pr] (#17563) 
* update a few is CONST check to check device None [pr]

* clone
 2026-08-17 07:33:16 -04:00
chenyuandGitHub 138fb4a783 delete dead DType.scalar [PR] (#17561) 2026-08-16 21:12:17 -04:00