SEV: [ROCm] MI350 CI jobs will have longer queue times due to CI migration (#187193)

pytorch/pytorch : main

Job filter:

Time	SHA	Commit	PR	Author	linux ▸	win ▸	mac ▸	inductor ▸	Lint	pull / linux-jammy-py3.8-gcc11 / build	pull / linux-jammy-py3.8-gcc11 / test	pull / linux-focal-cuda12.1 / build	pull / linux-focal-cuda12.1 / test	pull / linux-focal-cuda12.4 / build	pull / linux-focal-cuda12.4 / test	pull / linux-focal-rocm6.2 / build	pull / win-vs2019-cpu / build	pull / win-vs2019-cpu / test	pull / macos-py3-arm64 / build	inductor / A100 / test	pull / manylinux / build	pull / linux-jammy-py3.12 / build	trunk / linux-focal-py3.11 / test	CRCR ▸	intel / build (L3)	intel / test-xpu (L3)	ascend / build (L3)	ascend / test-npu (L3)	cambricon / build (L4)
11:59 am	a1b2c3d	Improve conv2d backward pass performance	#183512	contributor42	O linux: 8/8 passed	? win: 1/2 complete	? mac: 0/1 complete	? inductor: running	O	O	O	O	X pull / linux-focal-cuda12.1 / test FAIL test_sparse (SparseTestCUDA) Duration: 45m 12s	O	O	O	O	?	?	?	?	O	~	? CRCR: 3/5 passed, 2 running intel [L3]: 1/2 (1 running) ascend [L3]: 1/2 (1 running) cambricon [L4]: 1/1 passed	O intel / build [L3] — success Duration: 2m 14s	? intel / test-xpu [L3] — running Queue: 8.2s	O ascend / build [L3] — success Duration: 3m 5s	? ascend / test-npu [L3] — running Queue: 11.3s	O cambricon / build [L4] — success Duration: 1m 52s
11:42 am	e4f5a6b	Fix memory leak in autograd engine	#183498	pytorchbot	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O CRCR: 5/5 passed intel [L3]: 2/2, ascend [L3]: 2/2, cambricon [L4]: 1/1	O	O	O	O	O
10:38 am	c7d8e9f	[dynamo] Add support for torch.vmap	#183475	dynamodev	O	O	O	O	O	O	O	O	O	O	F	O	O	O	O	O	O	O	O	O CRCR: 5/5 passed intel [L3]: 2/2, ascend [L3]: 2/2, cambricon [L4]: 1/1	O	O	O	O	O
10:38 am	f0a1b2c	Refactor inductor scheduler pass	#183460	inductor_fan	O	O	O	X	O	O	O	O	O	O	O	O	O	X	O	X	O	O	O	X CRCR: 4/5 passed, 1 failed intel [L3]: 1/2 (test-xpu failed) ascend [L3]: 2/2, cambricon [L4]: 1/1	O	X intel / test-xpu [L3] — failure Duration: 14m 3s Tests: 890/1210 passed	O	O	O
8:37 am	b6c7d8e	Enable flex attention for inference	#183420	attndev	X	O	O	X	O	O	X	O	X	O	O	O	O	O	O	X	O	O	O	X CRCR: 3/5 passed, 2 failed intel [L3]: 1/2 (test-xpu failed) ascend [L3]: 1/2 (test-npu failed) cambricon [L4]: 1/1	O	X intel / test-xpu [L3] — failure Duration: 11m 48s Tests: 920/1210 passed	O	X ascend / test-npu [L3] — failure Duration: 8m 22s Tests: 310/380 passed	O
8:33 am	d3e4f5a	Update FBGEMM to latest nightly	#183398	fbgemmbot	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O CRCR: 5/5 passed intel [L3]: 2/2, ascend [L3]: 2/2, cambricon [L4]: 1/1	O	O	O	O	O
8:08 am	e9f0a1b	[compile] Graph break on side effects	#183380	compiledev	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	X	O	O CRCR: 5/5 passed intel [L3]: 2/2, ascend [L3]: 2/2, cambricon [L4]: 1/1	O	O	O	O	O
6:34 am	a2b3c4d	Optimize batch norm backward CUDA	#183362	cudadev	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	X CRCR: 4/5 passed, 1 failed intel [L3]: 2/2 ascend [L3]: 1/2 (test-npu failed) cambricon [L4]: 1/1	O	O	O	X ascend / test-npu [L3] — failure Duration: 9m 11s Tests: 298/380 passed	O
6:34 am	b4c5d6e	[export] Fix dynamic shapes serialization	#183341	exportdev	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O	O CRCR: 5/5 passed intel [L3]: 2/2, ascend [L3]: 2/2, cambricon [L4]: 1/1	O	O	O	O	O
6:28 am	c6d7e8f	Add torch.linalg.svdvals decomposition	#183320	linalgfan	O	O	O	X	O	O	O	O	O	O	O	X	O	O	O	X	O	O	O	O CRCR: 5/5 passed intel [L3]: 2/2, ascend [L3]: 2/2, cambricon [L4]: 1/1	O	O	O	O	O
5:54 am	d7e8f9a	[distributed] Fix NCCL timeout handling	#183301	nccldev	O	X	X	O	O	O	O	X	O	O	X	O	O	O	O	O	O	?	F	? CRCR: 3 passed, 2 running	O	O	?	?	O
5:22 am	e8f9a0b	Upgrade cuDNN to 9.1	#183288	cudnnbot	O	?	O	X	O	O	~	O	O	O	O	O	O	X	O	O	O	O	O	X CRCR: 3 passed, 2 failed	O	X	X	O	O
5:22 am	f9a0b1c	[quantize] Add int4 weight-only quant	#183275	quantdev	?	O	X	O	~	O	X	O	X	~	O	O	O	O	O	O	X	X	X	X CRCR: 3 passed, 2 failed	X	O	O	O	X
3:47 am	a0b1c2d	Fix flaky test_matmul on ROCm	#183260	rocmfix	?	O	O	?	O	O	O	O	F	O	?	O	O	?	O	?	O	O	F	? CRCR: 4 passed, 1 running	O	O	O	?	O
3:35 am	b1c2d3e	[PT2] Support custom triton ops	#183248	tritondev	O	O	X	O	O	O	O	~	O	O	O	O	~	O	O	O	O	O	O	X CRCR: 2 passed, 3 failed	O	X	O	X	X
3:29 am	c2d3e4f	Deprecate torch.cuda.amp autocast	#183235	ampdev	O	?	?	?	X	X	O	O	O	O	O	O	O	O	X	~	O	X	F	X CRCR: 4 passed, 1 failed	O	O	O	X	O
3:00 am	d3e4f5a	[MPS] Add conv3d backward kernel	#183220	mpsdev	O	O	O	O	O	X	O	O	O	O	O	X	O	O	O	O	X	X	O	X CRCR: 3 passed, 1 failed, 1 running	?	X	O	O	O
2:42 am	e4f5a6b	Update XLA pin to latest release	#183208	xlabot	O	O	O	O	O	~	O	O	O	O	O	X	~	O	O	~	O	O	O	X CRCR: 3 passed, 2 failed	X	X	O	O	O
2:39 am	f5a6b7c	[dynamo] Inline closures in graph	#183195	dynamodev2	O	O	O	O	~	O	~	O	F	O	O	O	O	O	~	O	O	O	O	O CRCR: 5 passed	O	O	O	O	O
2:34 am	a6b7c8d	Fix autograd saved tensor hooks	#183180	agdev	O	O	O	?	O	~	O	X	X	O	X	O	O	X	O	O	O	O	O	? CRCR: 4 passed, 1 running	O	O	?	O	O
2:06 am	b7c8d9e	[inductor] Vectorize reduction ops	#183168	vecdev	O	O	O	?	X	~	O	O	O	O	O	O	O	X	O	O	O	O	O	X CRCR: 3 passed, 1 failed, 1 running	X	O	?	O	O
2:03 am	c8d9e0f	Add support for structured sparsity	#183155	sparsedev	X	O	?	X	O	O	O	O	O	O	O	O	O	O	X	O	O	O	~	X CRCR: 1 passed, 2 failed, 2 running	?	?	X	O	X
1:58 am	d9e0f1a	[CI] Pin nightly CUDA toolkit version	#183140	cibot	O	O	O	?	O	O	O	F	O	O	O	X	O	O	O	O	O	O	O	X CRCR: 3 passed, 1 failed, 1 running	?	X	O	O	O
1:44 am	e0f1a2b	Improve TorchScript type inference	#183128	tsdev	O	O	O	O	X	~	O	X	O	O	X	~	O	O	O	X	O	O	O	? CRCR: 4 passed, 1 running	O	O	O	?	O
1:40 am	f1a2b3c	[fx] Add pattern matcher for concat	#183115	fxdev	O	O	O	O	O	O	O	O	O	?	O	O	O	O	F	O	O	O	O	O CRCR: 5 passed	O	O	O	O	O

Page 1: Prev | Next

This page automatically updates.