github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-14 09:28:04 -05:00
Code Issues Packages Projects Releases Wiki Activity
10,089 Commits 203 Branches 14 Tags
delete_kernel_py
Commit Graph

52 Commits

Author SHA1 Message Date
qazal
da61b40604 some viz tests don't need track_rewrites (#12010) 2025-09-04 23:59:32 +03:00
qazal
be364a1adb viz: add default tracing group (#12009) 
This enables seeing rewrites in unit tests like `VIZ=1 python3 test/test_uop_graph.py TestUOpGraph.test_in_bounds_access_gated_local` that call graph_rewrite directly.

`@track_rewrites` keeps existing as an optional helper to organize larger traces.
 2025-09-04 23:29:56 +03:00
qazal
4996bb668b load all traces before asserting in test_viz (#12004) 2025-09-04 21:34:48 +03:00
qazal
f750c15965 viz: add python marker (#11952) 
* viz: add python marker

* remove duplicate
 2025-09-02 23:44:00 +03:00
qazal
0a53e72f70 viz: fix trace duration in python test decoder (#11949) 2025-09-01 14:32:25 +03:00
qazal
27c9ed5a84 viz: more consistent naming of events (#11948) 
* s/shapes/events in test_viz

* s/bufs/events in the memory packer
 2025-09-01 14:16:47 +03:00
qazal
c27b99d68f viz: refactor to indexed rewrite traces (#11923) 2025-08-30 20:01:47 +03:00
qazal
bf0d055b39 viz: color by name (#11919) 2025-08-30 16:04:58 +03:00
qazal
a1f6823060 viz: memory layout in client side (#11830) 
* viz: memory layout in client side

* update test_viz
 2025-08-25 14:49:33 +03:00
qazal
0d86288bd7 viz: calculate timeline fixed points in client side (#11805) 
* viz: calculate timeline fixed points in client side

* 26 bytes / event

* math
 2025-08-24 01:44:40 +03:00
qazal
2407fecdae viz bytepack format (#11792) 
* viz bytepack format

Training a 1B llama yields ~20M profiler events.

With JSON serialization, the browser tries to load 6GB to memory. This OOMs since each tab is limited to <3-4GB memory usage. Using a packed format, we only need ~600MB.

**Design decisions:**

- Timestamps are in microseconds relative to start time. They're stored in u32, which can express up to ~1 hr of trace events.
- Strings (kernel names, metadata, etc) are deduped.
- Buffer sizes are in u64 nbytes.

More optimization possible:

- The string lookup is a JSON dumped array, we can compress this.
- Can store less for memory by moving the layout to client.

**Results**

|  | Events | JSON | bytepack |
|----------------|---------|-------------|-------------|
| DP=8 llama 1B train (`command: [1]`) | 24M | 5.8GB | 640MB |
| examples/beautiful_mnist.py | 16K | 3.7MB | 745KB |
| examples/gpt2.py | 55K | 12.54MB | 1.40MB |

`[1]`: `VIZ=1 FAKEDATA=1 OFFLOAD_OPTIM=1 DP=8 BS=8 GRADIENT_ACC_STEPS=2 BLOCK_REORDER=0 LR=3e-4 TRAIN_ON_VAL=1 DEFAULT_FLOAT=bfloat16 OPTIM_DTYPE=bfloat16 LLAMA3_SIZE=1B WARMUP_STEPS=36 DECAY_STEPS=360 SEQLEN=8192 PYTHONPATH=. AMD=1 AMD_LLVM=0 MODEL=llama3 python3 examples/mlperf/model_train.py`

* python reference decoder

* 27 bytes / event, 1hr hard limit
 2025-08-23 23:50:21 +03:00
qazal
b12d1d866c count bytes per kernel in test_viz (#11801) 
Currently at ~100 bytes/kernel with JSON.
 2025-08-23 23:35:27 +03:00
qazal
b975830424 add profile loader helper in test_viz (#11797) 2025-08-23 19:20:29 +03:00
qazal
9ff03680ba viz: store relative timestamps (#11787) 
* viz: store relative timestamps

* err

* update test
 2025-08-22 19:30:21 +03:00
qazal
2e0eb88549 viz: add metadata to UOp tracing (#11772) 
* viz: add metadata to UOp tracing

* place after tag

* optional field

* err, refcount of root must be 0
 2025-08-22 00:18:45 +03:00
qazal
d762edd694 viz: define tracks in python (#11701) 
* viz: defines tracks in python

* update unittests

* figuring it out

* works

* diff cleanup

* math

* y axis is back
 2025-08-17 18:19:13 +03:00
qazal
c8ba48b223 show rewrite errors in viz (#11684) 2025-08-15 19:09:47 +03:00
George Hotz
22bdf48cdd render ranges in viz, name gbufs with sizes. changes from rangeify (#11656) 
* render ranges in viz, name gbufs with sizes. changes from rangeify

* fix unit test dtypes
 2025-08-13 12:46:16 -07:00
qazal
960cc6533a pass through name function args in track_rewrites (#11572) 2025-08-08 02:28:52 +03:00
qazal
846a2826ab viz: remove TracingKey.fmt (#11482) 
* viz: remove TracingKey.fmt

* remove from test too
 2025-08-05 00:00:03 +03:00
qazal
fa66d9772d viz: show const node when it's root (#11456) 2025-08-01 01:01:58 +03:00
qazal
d3ec63a5c3 viz: add base class for unittests (#11178) 2025-07-11 13:58:03 +03:00
qazal
5c1d215b41 viz: add Graph stream (#11144) 
* viz: stack an event for the entire batch

* multi

* whitespace

* work

* multi graph, Graph gets its own row
 2025-07-09 20:56:46 +03:00
qazal
3dfc0ff887 move cpu_profile and shared ProfileEvents from device.py to helpers [pr] (#11126) 
* move cpu_profile and shared ProfileEvents to helpers [pr]

* TestProfiler.test_cpu_profile

* update test_viz.py

* TestProfiler.test_profile_multiops ordering, it's different streams now
 2025-07-08 12:14:03 +03:00
qazal
81781dc12b viz: renames and spacing changes to tracing (#11102) 2025-07-05 18:40:39 +03:00
qazal
4fcfaa0ef7 viz: switch to TracingKey (#11100) 
* viz: switch to TracingKey

* tuple

* order is name, keys, fmt

* add test_tracing_key
 2025-07-05 17:46:18 +03:00
qazal
b695e8c4d6 viz: remove support for naming with self (#11076) 2025-07-03 17:29:14 +03:00
qazal
8b0871ac31 viz: test for no lockup on infinite loop (#11041) 
* viz: add test infinite loop fallback

* assert

* continue til the end

* work

* bring that back

* fallback to nop
 2025-07-01 17:44:20 +03:00
qazal
2ea4737930 viz: fix newlines breaking label colors (#11030) 
* viz: fix newlines breaking label colors

* TestViz.test_colored_label

* TestWordWrap
 2025-06-30 13:39:44 +03:00
qazal
4c8d2a0383 buffer viz (#10960) 
* add mem_layout

* ui

* cleanup

* work

* debugLine work and expander

* tooltip style

* real expand device

* wheel does one thing

* diff

* shows llama oom

* add y axis

* mypy chill

* work

* unittests for the memory layout
 2025-06-28 21:50:32 +03:00
qazal
a39343e39f viz: move timeline layout to python (#10998) 
* viz: move timeline layout to python

* DevEvent has a device and a name
 2025-06-27 13:06:00 +03:00
qazal
1127302c46 move perfetto to extra (#10994) 
* move perfetto to extra

* update TestViz and fix tests

* remove perfetto.html from viz directory

* work

* mypy
 2025-06-27 01:53:54 +03:00
qazal
ac39f27ae6 viz: non blocking UOp tracing (#10913) 
* viz: non blocking UOp tracing

* u.arg

* no if Ops.KENREL

* drop replace

* switch to weakref.WeakKeyDictionary

* back

* remove ram usage skips, viz works here

* cache on reconstruct
 2025-06-23 19:59:28 +03:00
qazal
9201224e0b viz: remove Kernel check [pr] (#10920) 
* viz: remove Kernel check [pr]

* TestVizIntegration

* test/unit allows opening of devices

* kernel -> Kernel
 2025-06-22 20:47:54 +03:00
qazal
c7ec913210 viz: cleanup unit tests (#10909) 
* cleanup test_viz

* tree view
 2025-06-21 12:35:09 +03:00
qazal
00d0071b36 simpler viz naming [pr] (#10874) 
* simpler viz naming [pr]

* n2
 2025-06-19 12:10:47 +03:00
qazal
5839542fc8 viz: one name arg in track_rewrites [pr] (#10873) 
* viz: one name arg in track_rewrites [pr]

* other test
 2025-06-19 03:34:56 +03:00
Sieds Lykles
ac27c46104 fix UPat get_location after mathtraits refactor (#10814) 
* fix UPat get_location

* fold line
 2025-06-15 12:47:55 -07:00
George Hotz
a38947b4bb move symbolic and transcendental to uop [pr] (#10771) 2025-06-10 20:51:22 -07:00
qazal
cb61774ab6 move shared viz fields out of serve.py [pr] (#10684) 
* move shared viz fields out [pr]

* update javascript

* update test_viz
 2025-06-07 17:18:18 +03:00
qazal
b515d796fb inline viz get_name [pr] (#10682) 
* inline viz get_name [pr]

* changing name_fxn makes this simpler

* waitUntil dom
 2025-06-07 11:16:16 +03:00
George Hotz
411392dfb7 move files into uop dir (#10399) 
* move files into uop dir [pr]

* tinygrad.uop is a thing

* fix uop docs, no pr

* fix viz
 2025-05-18 11:38:28 -07:00
qazal
043efc6ec4 do not require self for track_rewrites [pr] (#10302) 2025-05-14 18:23:32 +03:00
qazal
cbf7347cd6 display viz rewrites with tabbing if they are subrewrites (#10097) 
* display viz rewrites with tabbing if they are subrewrites

* update viz api
 2025-04-29 17:57:21 +08:00
qazal
498a2bf738 add err handling tests to viz + cleanups (#9825) 
* cleanup

* add err handling tests to viz + cleanups

* lint
 2025-04-10 14:05:05 +08:00
qazal
bf94924d5a fix viz with nested graph_rewrite (#9595) 2025-03-27 13:14:28 +08:00
qazal
e5ff7b23d7 refactor to @track_matches + add failing test_nested_rewrite (#9592) 
* test_nested_rewrite

* refactor to track_matches

* positional arg
 2025-03-27 11:11:56 +08:00
qazal
d12efc95d4 support custom name function in viz [pr] (#9219) 
* support custom name function in viz [pr]

* title case

* assert name count in test_track_rewrites_name_fxn
 2025-02-24 03:03:25 +02:00
George Hotz
1bf66d62cf symbolic gets its own file [pr] (#9132) 2025-02-17 18:55:21 +08:00
qazal
2d04a75a40 start tracking bottom_up_rewrite in viz [pr] (#9071) 
* start tracking bottom_up_rewrite in viz [pr]

* use the tracking matcher in test_viz
 2025-02-14 00:28:10 +01:00