mehrdadh commented on PR #13600:
URL: https://github.com/apache/tvm/pull/13600#issuecomment-1372614310
Given the time that this PR was open I will merge it.
@leandron @areusch thanks for the reviews!
--
This is an automated message from the Apache Git Service.
To respond to the mess
mehrdadh commented on code in PR #13690:
URL: https://github.com/apache/tvm/pull/13690#discussion_r1062762449
##
3rdparty/mlperftiny/api/submitter_implemented.h:
##
@@ -0,0 +1,85 @@
+/*
+Copyright 2020 EEMBC and The MLPerf Authors. All Rights Reserved.
+Licensed under the Apache
tqchen commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062785880
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.
tqchen commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062785880
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.
tqchen commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062785880
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.
tqchen commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062785880
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.
tqchen commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062785880
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.
Mousius commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062784681
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements
tqchen commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062711599
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.
tqchen commented on PR #96:
URL: https://github.com/apache/tvm-rfcs/pull/96#issuecomment-1372526217
Thanks for the RFC. The main thing that I think worth doing is scoping, so
that the relation of the proposed module can be reasonably understandable by
the overall community. Here are two thi
tqchen commented on code in PR #96:
URL: https://github.com/apache/tvm-rfcs/pull/96#discussion_r1002466312
##
rfcs/0096-embedded-rust-interface-api.md:
##
@@ -0,0 +1,274 @@
+- Feature Name: embedded_rust_interface
+- Start Date: 2022-10-04
+- RFC PR: [apache/tvm-rfcs#](https
tqchen commented on code in PR #13707:
URL: https://github.com/apache/tvm/pull/13707#discussion_r1062711599
##
src/target/source/interface_rust.cc:
##
@@ -0,0 +1,380 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.
tvm-bot commented on PR #13707:
URL: https://github.com/apache/tvm/pull/13707#issuecomment-1372253319
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Mousius opened a new pull request, #13707:
URL: https://github.com/apache/tvm/pull/13707
(Builds on #13706)
This is based on the older API in the RFC, we're landing this now so as to
upstream it iteratively and allow others to see it in action 😸
There'll be a follow up, as note
tvm-bot commented on PR #13706:
URL: https://github.com/apache/tvm/pull/13706#issuecomment-1372227690
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Mousius opened a new pull request, #13706:
URL: https://github.com/apache/tvm/pull/13706
Adding transformations to Rust-style alongside the existing C-style
transforms.
Co-authored-by: Ashutosh Parkhi
--
This is an automated message from the Apache Git Service.
To respond to the m
Mousius commented on PR #96:
URL: https://github.com/apache/tvm-rfcs/pull/96#issuecomment-137596
New APIs are now documented and I've raised
https://github.com/apache/tvm/issues/13705 😸
--
This is an automated message from the Apache Git Service.
To respond to the message, please log
Mousius opened a new issue, #13705:
URL: https://github.com/apache/tvm/issues/13705
### This issue is to track progress for Rust Embedded API
- [ ] P1. Add Rust name transforms
- [ ] P2. Add Rust interface generator
- [ ] P3. Initial Rust MLF generation and testing
- [ ] P4. Add R
masahi merged PR #13693:
URL: https://github.com/apache/tvm/pull/13693
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
Hzfengsy merged PR #13692:
URL: https://github.com/apache/tvm/pull/13692
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.
tvm-bot commented on PR #13704:
URL: https://github.com/apache/tvm/pull/13704#issuecomment-1372139749
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Mousius opened a new pull request, #13704:
URL: https://github.com/apache/tvm/pull/13704
Following from #12789, this adds support for determining the output tensor
name from the input model within the MLF metadata json.
Co-authored-by: Ashutosh Parkhi
--
This is an automated messa
heliqi commented on code in PR #13701:
URL: https://github.com/apache/tvm/pull/13701#discussion_r1062402600
##
tests/python/frontend/paddlepaddle/test_forward.py:
##
@@ -1681,5 +1687,15 @@ def forward(self, inputs, prev_h):
)
+@tvm.testing.uses_gpu
+def test_forward
jzh18 closed pull request #13703: Load separate func sos
URL: https://github.com/apache/tvm/pull/13703
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: c
tvm-bot commented on PR #13703:
URL: https://github.com/apache/tvm/pull/13703#issuecomment-1372130954
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
ashutosh-arm commented on PR #13363:
URL: https://github.com/apache/tvm/pull/13363#issuecomment-1372053762
> @Mousius any update on this?
We have started looking at it. I am able to reproduce the hang locally:
`test_arm_mprofile_dsp.py::test_conv1d[int16-data_shape_nwc2-3-5-1-0]` that
wangzy0327 commented on issue #13666:
URL: https://github.com/apache/tvm/issues/13666#issuecomment-1372045435
> Okay reproduced.
How to solve this problem?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
masahi commented on issue #13666:
URL: https://github.com/apache/tvm/issues/13666#issuecomment-1372042675
Okay reproduced.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
T
wangzy0327 commented on issue #13666:
URL: https://github.com/apache/tvm/issues/13666#issuecomment-1372001733
Remove NetworkData stuff.The "minimal" test which compare two outputs
between rocm and opencl as follow.
onnx_rocm.py
```
from pyexpat import model
import onn
woobinw commented on code in PR #13701:
URL: https://github.com/apache/tvm/pull/13701#discussion_r1062278664
##
tests/python/frontend/paddlepaddle/test_forward.py:
##
@@ -1681,5 +1687,15 @@ def forward(self, inputs, prev_h):
)
+@tvm.testing.uses_gpu
+def test_forwar
masahi commented on issue #13666:
URL: https://github.com/apache/tvm/issues/13666#issuecomment-1371979718
I said "minimal" test. Remove `NetworkData` stuff which is not even defined.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Gi
masahi commented on PR #13450:
URL: https://github.com/apache/tvm/pull/13450#issuecomment-1371974855
Please remove typos and obvious grammar errors. I found many just looking
over the code briefly. Please go through the change.
Also remove `TODO(mbs)` stuff.
--
This is an automat
heliqi commented on code in PR #13701:
URL: https://github.com/apache/tvm/pull/13701#discussion_r1062261690
##
tests/python/frontend/paddlepaddle/test_forward.py:
##
@@ -1681,5 +1687,15 @@ def forward(self, inputs, prev_h):
)
+@tvm.testing.uses_gpu
+def test_forward
heliqi commented on code in PR #13701:
URL: https://github.com/apache/tvm/pull/13701#discussion_r1062261690
##
tests/python/frontend/paddlepaddle/test_forward.py:
##
@@ -1681,5 +1687,15 @@ def forward(self, inputs, prev_h):
)
+@tvm.testing.uses_gpu
+def test_forward
krishnaraj36 commented on PR #13450:
URL: https://github.com/apache/tvm/pull/13450#issuecomment-1371912072
> @masahi :
I have resolved all merge conflicts on this PR, can you please review and
merge this PR.
--
This is an automated message from the Apache Git Service.
To respo
zhaojinxi commented on PR #13693:
URL: https://github.com/apache/tvm/pull/13693#issuecomment-1371902392
> I mean, can you add that as a test, probably in
`unittest/test_runtime_graph_debug.py`.
done
--
This is an automated message from the Apache Git Service.
To respond to the mess
masahi commented on PR #13693:
URL: https://github.com/apache/tvm/pull/13693#issuecomment-1371876572
I mean, can you add that as a test.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specif
zhaojinxi commented on PR #13693:
URL: https://github.com/apache/tvm/pull/13693#issuecomment-1371869902
@masahi Try the code below, with the old wrong code, it will print "p0 has
shape: [1, 3, 4, 16]", actually it should print "p0 has shape: [3, 3, 1, 1]"
import tvm
import nump
tvm-bot commented on PR #13702:
URL: https://github.com/apache/tvm/pull/13702#issuecomment-1371868332
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
multiverstack-intellif opened a new pull request, #13702:
URL: https://github.com/apache/tvm/pull/13702
Add an arith helper interface to analyze common sub expression.
Also modify InsertVectorToSortedSemanticComputations to be aware of the
occurrence.
--
This is an automated messag
masahi merged PR #13642:
URL: https://github.com/apache/tvm/pull/13642
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
masahi commented on PR #13450:
URL: https://github.com/apache/tvm/pull/13450#issuecomment-1371775072
Not sure and I'm not familiar with it either. But I can take care of this.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub an
srkreddy1238 commented on PR #13450:
URL: https://github.com/apache/tvm/pull/13450#issuecomment-1371767541
@masahi any moderator for Collage ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the
tvm-bot commented on PR #13701:
URL: https://github.com/apache/tvm/pull/13701#issuecomment-1371741543
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
woobinw opened a new pull request, #13701:
URL: https://github.com/apache/tvm/pull/13701
There are some operation in PaddlePaddle that output a tensor which shape is
[1], meanwhile, the slice op have decrease_axis attr, but it should not be
squeezed.
@heliqi
@jiangjiajun
multiverstack-intellif opened a new pull request, #13700:
URL: https://github.com/apache/tvm/pull/13700
Much of the index computation is duplicated, so this PR is to do common
subexpression analyze when doing cache_index.
A threshold number is needed when identifying a common sub expr, si
tvm-bot commented on PR #13700:
URL: https://github.com/apache/tvm/pull/13700#issuecomment-1371724928
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Qianshui-Jiang commented on code in PR #13642:
URL: https://github.com/apache/tvm/pull/13642#discussion_r1062065916
##
apps/microtvm/zephyr/template_project/qemu-hack/qemu-system-arm:
##
@@ -1 +0,0 @@
-qemu-system-i386
Review Comment:
@masahi by misoperation, already remove
wrongtest-intellif commented on issue #13695:
URL: https://github.com/apache/tvm/issues/13695#issuecomment-1371673160
> In this case
For my curiosity, why we have divergent dtype value support in
`BufferStore`? If this is the expected supporting behavior in backend, it seems
that we
lightzhan-intellif commented on code in PR #13692:
URL: https://github.com/apache/tvm/pull/13692#discussion_r1062035156
##
python/tvm/script/ir_builder/tir/ir.py:
##
@@ -1532,7 +1532,7 @@ def wrapped(*args, **kwargs):
nearbyint = _op_wrapper(_tir_op.nearbyint)
nextafter = _op_
lightzhan-intellif commented on issue #13695:
URL: https://github.com/apache/tvm/issues/13695#issuecomment-1371648780
I am not very sure which way we should go for. But from the python view, if
we can do auto casting from low precision datatype to the higher precision
datatype and throw an
junrushao commented on PR #13534:
URL: https://github.com/apache/tvm/pull/13534#issuecomment-1371598225
@vinx13 for context, @Hzfengsy and I are having some issues where
synchronization might potentially be missing, so we are not proceeding with the
PR yet
--
This is an automated message
masahi merged PR #13698:
URL: https://github.com/apache/tvm/pull/13698
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
masahi commented on code in PR #13642:
URL: https://github.com/apache/tvm/pull/13642#discussion_r1061991180
##
apps/microtvm/zephyr/template_project/qemu-hack/qemu-system-arm:
##
@@ -1 +0,0 @@
-qemu-system-i386
Review Comment:
Is this change relevant? If not remove it.
##
masahi commented on PR #13693:
URL: https://github.com/apache/tvm/pull/13693#issuecomment-1371579370
Can you come up a test case for this?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the spec
masahi merged PR #13696:
URL: https://github.com/apache/tvm/pull/13696
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
masahi merged PR #13699:
URL: https://github.com/apache/tvm/pull/13699
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
masahi merged PR #13682:
URL: https://github.com/apache/tvm/pull/13682
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
mehrdadh commented on PR #13363:
URL: https://github.com/apache/tvm/pull/13363#issuecomment-1371548710
@Mousius any update on this?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific co
mkatanbaf commented on code in PR #13690:
URL: https://github.com/apache/tvm/pull/13690#discussion_r1061971885
##
3rdparty/mlperftiny/api/submitter_implemented.h:
##
@@ -0,0 +1,85 @@
+/*
+Copyright 2020 EEMBC and The MLPerf Authors. All Rights Reserved.
+Licensed under the Apach
guberti commented on code in PR #13690:
URL: https://github.com/apache/tvm/pull/13690#discussion_r1061967618
##
3rdparty/mlperftiny/api/submitter_implemented.h:
##
@@ -0,0 +1,85 @@
+/*
+Copyright 2020 EEMBC and The MLPerf Authors. All Rights Reserved.
+Licensed under the Apache
mehrdadh commented on code in PR #13627:
URL: https://github.com/apache/tvm/pull/13627#discussion_r1061966967
##
gallery/how_to/work_with_microtvm/micro_tflite.py:
##
@@ -26,109 +26,36 @@
"""
##
-# .. note::
mehrdadh commented on code in PR #13627:
URL: https://github.com/apache/tvm/pull/13627#discussion_r1061965740
##
apps/microtvm/pyproject.toml:
##
@@ -98,22 +95,23 @@ onnxoptimizer = { version = "==0.2.6", optional = true }
onnxruntime = { version = "==1.9.0", optional = true }
mehrdadh commented on code in PR #13627:
URL: https://github.com/apache/tvm/pull/13627#discussion_r1061965045
##
gallery/how_to/work_with_microtvm/micro_aot.py:
##
@@ -30,6 +30,29 @@
or on Zephyr platform on a microcontroller/board supported by Zephyr.
"""
+#
mehrdadh commented on code in PR #13627:
URL: https://github.com/apache/tvm/pull/13627#discussion_r1061964453
##
apps/microtvm/poetry.lock:
##
@@ -1,28 +1,9 @@
-# Licensed to the Apache Software Foundation (ASF) under one
-# or more contributor license agreements. See the NOTIC
mkatanbaf commented on code in PR #13690:
URL: https://github.com/apache/tvm/pull/13690#discussion_r1061961376
##
3rdparty/mlperftiny/api/submitter_implemented.h:
##
@@ -0,0 +1,85 @@
+/*
+Copyright 2020 EEMBC and The MLPerf Authors. All Rights Reserved.
+Licensed under the Apach
guberti commented on code in PR #13690:
URL: https://github.com/apache/tvm/pull/13690#discussion_r1061953843
##
apps/microtvm/zephyr/template_project/src/mlperftiny/main.cc:
##
@@ -0,0 +1,37 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contri
vinx13 commented on issue #13695:
URL: https://github.com/apache/tvm/issues/13695#issuecomment-1371476901
`T.cast` should be used to cast explicitly. In this case, without cast we
should throw error in tvmscript parser instead
--
This is an automated message from the Apache Git Service.
T
AndrewZhaoLuo closed pull request #13532: [Packed Func] Pack args fp16 support
URL: https://github.com/apache/tvm/pull/13532
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
AndrewZhaoLuo commented on PR #13532:
URL: https://github.com/apache/tvm/pull/13532#issuecomment-1371470841
Can confirm that this was a CSE issue which is now fixed for all the use
cases I care about. masahi has a good summary in the above of what occured. I
am closing for now to avoid delv
AndrewZhaoLuo merged PR #13694:
URL: https://github.com/apache/tvm/pull/13694
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.ap
adstraw commented on PR #13699:
URL: https://github.com/apache/tvm/pull/13699#issuecomment-1371383197
@tvm-bot rerun
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsu
tvm-bot commented on PR #13699:
URL: https://github.com/apache/tvm/pull/13699#issuecomment-1371372299
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
junrushao commented on code in PR #13692:
URL: https://github.com/apache/tvm/pull/13692#discussion_r1061812640
##
python/tvm/script/ir_builder/tir/ir.py:
##
@@ -1532,7 +1532,7 @@ def wrapped(*args, **kwargs):
nearbyint = _op_wrapper(_tir_op.nearbyint)
nextafter = _op_wrapper(_
driazati closed issue #13337: [ci] Split up Jenkins into platform-specific jobs
URL: https://github.com/apache/tvm/issues/13337
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
driazati commented on issue #13337:
URL: https://github.com/apache/tvm/issues/13337#issuecomment-1371279563
This is done now
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
mehrdadh commented on code in PR #13514:
URL: https://github.com/apache/tvm/pull/13514#discussion_r1061753588
##
python/tvm/contrib/micro/meta_schedule/test_autotune_ms.py:
##
@@ -0,0 +1,181 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor
mehrdadh commented on code in PR #13690:
URL: https://github.com/apache/tvm/pull/13690#discussion_r1061743014
##
3rdparty/tiny/api/internally_implemented.cpp:
##
@@ -0,0 +1,325 @@
+/*
Review Comment:
I matched it with their repo name, but I agree with you. I'll also add a
r
mehrdadh merged PR #13610:
URL: https://github.com/apache/tvm/pull/13610
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.
tkonolige merged PR #13697:
URL: https://github.com/apache/tvm/pull/13697
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache
tkonolige merged PR #13599:
URL: https://github.com/apache/tvm/pull/13599
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache
Mousius commented on code in PR #13643:
URL: https://github.com/apache/tvm/pull/13643#discussion_r1061728070
##
apps/microtvm/cmsisnn/Makefile:
##
@@ -69,6 +69,18 @@ $(BUILD_DIR)/crt_backend_api.o:
$(STANDALONE_CRT_PATH)/src/runtime/crt/common/cr
$(QUIET)mkdir -p $(@D)
tvm-bot commented on PR #13698:
URL: https://github.com/apache/tvm/pull/13698#issuecomment-1371172912
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Lunderberg opened a new pull request, #13698:
URL: https://github.com/apache/tvm/pull/13698
Previously, this line caused a warning for `-Wmaybe-uninitialized` when
compiling in g++ 11.3.0.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log o
adstraw commented on PR #13681:
URL: https://github.com/apache/tvm/pull/13681#issuecomment-1371169894
LGTM as well. Thanks @janetsc
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific
Lunderberg merged PR #13681:
URL: https://github.com/apache/tvm/pull/13681
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apach
Lunderberg commented on issue #10667:
URL: https://github.com/apache/tvm/issues/10667#issuecomment-1371017046
Ooh, I like it. That's definitely a cleaner way that I had been thinking.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
Lunderberg commented on PR #13697:
URL: https://github.com/apache/tvm/pull/13697#issuecomment-1371015141
@tkonolige This is a follow-up to #12839. I noticed that ccache wasn't
being used in my dev environment, and it looks like there was a different
variable to be used.
--
This is an au
tvm-bot commented on PR #13697:
URL: https://github.com/apache/tvm/pull/13697#issuecomment-1371013178
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
tvm-bot commented on PR #13697:
URL: https://github.com/apache/tvm/pull/13697#issuecomment-1371013175
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Lunderberg opened a new pull request, #13697:
URL: https://github.com/apache/tvm/pull/13697
Previously, when using `set(USE_CCACHE AUTO)`, the cmake config would set
variables `CXX_COMPILER_LAUNCHER` and `C_COMPILER_LAUNCHER`. While there are
the target-specific properties named
[`_COMPILE
wangzy0327 commented on issue #13666:
URL: https://github.com/apache/tvm/issues/13666#issuecomment-1370942424
@mashi
The minimal test case as follow.
onnx_rocm.py
```
from csv import writer
from pyexpat import model
import onnx
#from tvm.driver import tvmc
masahi commented on issue #13666:
URL: https://github.com/apache/tvm/issues/13666#issuecomment-1370777170
Does
https://github.com/apache/tvm/blob/main/apps/topi_recipe/gemm/cuda_gemm_square.py
run successfully for you?
Also, please make your test case minimal. It is not obvious what
tvm-bot commented on PR #13696:
URL: https://github.com/apache/tvm/pull/13696#issuecomment-1370766418
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
srkreddy1238 opened a new pull request, #13696:
URL: https://github.com/apache/tvm/pull/13696
Various benchmarks enabled for CLML BYOC backend for Adreno GPU.
Networks resnet-18, resnet-34, resnet-50, densenet-121, inception_v3,
mobilenetv1, squeezenet_v1.0, squeezenet_v1.1 are added wit
lightzhan-intellif closed issue #13695: [Bug] [TVMScript] Weird output of my
script
URL: https://github.com/apache/tvm/issues/13695
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comme
lightzhan-intellif opened a new issue, #13695:
URL: https://github.com/apache/tvm/issues/13695
Hi, all
I have a program like this:
```python
import numpy as np
from tvm.script import tir as T
import tvm
@T.prim_func
def test_case(A: T.Buffer[(1, ), "uint8"], B: T.Buff
masahi commented on issue #13691:
URL: https://github.com/apache/tvm/issues/13691#issuecomment-1370695229
For rocm 5, we need the change to `contrib/rocm.py` in this commit
https://github.com/masahi/tvm/commit/26d2701b7823ab4d93b8d980bc8689e9c03b2ee1
Unfortunately the rocm backend is
masahi closed issue #13691: [Bug] Which tvm version currently supports rocm?
URL: https://github.com/apache/tvm/issues/13691
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
lightzhan-intellif commented on PR #13692:
URL: https://github.com/apache/tvm/pull/13692#issuecomment-1370546578
Do you mean that users can use both `tir.pow` and `tir.power`? I think it
looks like a little bit confusing and redundant. How about we do not make any
change to `tir.power`?
901 - 1000 of 19571 matches
Mail list logo