spectrometerHBH commented on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-626110351
clang6.0 also works. But I will fix it under gcc5.4
This is an automated message from the Apache
icemelon9 commented on a change in pull request #5429:
URL: https://github.com/apache/incubator-tvm/pull/5429#discussion_r422452792
##
File path: python/tvm/relay/op/_transform.py
##
@@ -189,10 +192,9 @@ def _reshape_shape_func(data_shape, newshape, ndim):
spectrometerHBH commented on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-626098731
Both `release` and `debug` under gcc5.4 will show the same behavior. It is
not likely that O2 causes this problem.
spectrometerHBH edited a comment on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-626096064
I found the problem. It is due to the different behavior of gcc5.4 and gcc7.4
```c++
doc << "(" << Print(op->a) << OpString << Print(op->b) << ")";
spectrometerHBH edited a comment on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-626096064
I found the problem. It is due to the different behavior of gcc5.4 and gcc7.4
```c++
doc << "(" << Print(op->a) << OpString << Print(op->b) << ")";
spectrometerHBH commented on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-626096064
I found the problem. It is due to the different behavior of gcc5.4 and gcc
7.4
```c++
doc << "(" << Print(op->a) << OpString << Print(op->b) << ")";
```
tom-gall commented on pull request #5546:
URL: https://github.com/apache/incubator-tvm/pull/5546#issuecomment-626093626
I've updated the commit so the new location is:
tests/micro/test_runtime_micro_on_arm.py
I went ahead and committed the perhaps egregious act of doing a --force as
trevor-m commented on pull request #5549:
URL: https://github.com/apache/incubator-tvm/pull/5549#issuecomment-626088489
@masahi I see, thank you! I will change this PR to instead modify the
converter for `aten::max_pool2d_with_indices` to have a second dummy output.
tqchen commented on pull request #5382:
URL: https://github.com/apache/incubator-tvm/pull/5382#issuecomment-626078390
Thanks @yongfeng-nv @roastduck !
This is an automated message from the Apache Git Service.
To respond to
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from aded92d [Rust] Add first stage of updating and rewriting Rust
bindings. (#5526)
add 06efa7f [TE] Fix
trevor-m opened a new pull request #5549:
URL: https://github.com/apache/incubator-tvm/pull/5549
This bug occurs when an op has multiple outputs in pytorch, but only has a
single output in relay. For example, this bug occurs with
`aten::max_pool2d_with_indices` which has two outputs but
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git
The following commit(s) were added to refs/heads/master by this push:
new aded92d [Rust] Add first stage of
tqchen commented on pull request #5498:
URL: https://github.com/apache/incubator-tvm/pull/5498#issuecomment-626067165
ping @Hzfengsy
This is an automated message from the Apache Git Service.
To respond to the message,
binarybana commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422420158
##
File path: rust/tvm-sys/Cargo.toml
##
@@ -0,0 +1,35 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git
The following commit(s) were added to refs/heads/master by this push:
new 2175f6b [RPC] Improve RPCServer AsyncIO
kevinthesun commented on a change in pull request #5511:
URL: https://github.com/apache/incubator-tvm/pull/5511#discussion_r422409992
##
File path: topi/python/topi/mali/conv2d.py
##
@@ -138,20 +138,15 @@ def _schedule_spatial_pack(cfg, s, output, conv,
data_vec, kernel_vec):
kevinthesun commented on issue #5215:
URL: https://github.com/apache/incubator-tvm/issues/5215#issuecomment-626050953
@FrozenGene In https://github.com/apache/incubator-tvm/pull/5200 we
discussed another source of autotvm inaccurate measurement due to empty input
tensor. Do we have a
mbrookhart opened a new pull request #5548:
URL: https://github.com/apache/incubator-tvm/pull/5548
Can't find the behavior documented anywhere, but it showed up in an ONNX
model from Caffe2
@tmoreau89
This is an
areusch commented on pull request #5546:
URL: https://github.com/apache/incubator-tvm/pull/5546#issuecomment-626036189
aside from tianqi's comment, this looks good to me. thanks for adding this!
This is an automated message
ANSHUMAN87 opened a new pull request #5547:
URL: https://github.com/apache/incubator-tvm/pull/5547
Refer #5490
@jroesch , @tqchen , @zhiics : Please help review!
This is an automated message from the Apache Git
icemelon9 commented on a change in pull request #5144:
URL: https://github.com/apache/incubator-tvm/pull/5144#discussion_r422363192
##
File path: src/runtime/vm/vm.cc
##
@@ -535,6 +544,7 @@ void InstructionPrint(std::ostream& os, const Instruction&
instr) {
case
tqchen commented on pull request #5546:
URL: https://github.com/apache/incubator-tvm/pull/5546#issuecomment-626011102
Given that all the tests under unittests/python will be automatically
invoked by the CI.
Please put the tests under a different folder for now say `tests/micro`
robo-corg commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422360890
##
File path: rust/tvm-sys/src/context.rs
##
@@ -0,0 +1,285 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more
tom-gall commented on pull request #5546:
URL: https://github.com/apache/incubator-tvm/pull/5546#issuecomment-626006130
@weberlo @areusch - I've forked the test_runtime_micro.py unit test and
made one specifically for running native on arm hardware, in this case just the
STM32F7 (for
tom-gall opened a new pull request #5546:
URL: https://github.com/apache/incubator-tvm/pull/5546
A clone of test_runtime_micro.py, however modified to run specifically on
ARM hardware, currently just the STM32F746 discovery board.
Signed-off-by: Tom Gall
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git
The following commit(s) were added to refs/heads/master by this push:
new 15a4218 Load platform specific lib for
tqchen commented on pull request #5542:
URL: https://github.com/apache/incubator-tvm/pull/5542#issuecomment-625985407
Thanks @tobegit3hub !
This is an automated message from the Apache Git Service.
To respond to the message,
antinucleon commented on pull request #5545:
URL: https://github.com/apache/incubator-tvm/pull/5545#issuecomment-625962848
This is really nice to have. We can think of a unified wasm/web gpu
inference for the next step.
robo-corg commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422303593
##
File path: rust/tvm-sys/src/datatype.rs
##
@@ -0,0 +1,188 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more
tqchen commented on issue #5543:
URL: https://github.com/apache/incubator-tvm/issues/5543#issuecomment-625957768
Please open a new thread on https://discuss.tvm.ai/
This is an automated message from the Apache Git Service.
jroesch commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422297873
##
File path: rust/tvm-sys/build.rs
##
@@ -0,0 +1,117 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor
jroesch commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422296419
##
File path: rust/tvm-sys/build.rs
##
@@ -0,0 +1,117 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor
jroesch commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422295900
##
File path: rust/tvm-sys/src/context.rs
##
@@ -0,0 +1,285 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more
jroesch commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422295051
##
File path: rust/tvm-sys/src/context.rs
##
@@ -0,0 +1,285 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more
jroesch commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422293153
##
File path: rust/tvm-sys/src/context.rs
##
@@ -0,0 +1,285 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more
jroesch commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-tvm/pull/5526#discussion_r422292581
##
File path: rust/tvm-sys/src/datatype.rs
##
@@ -0,0 +1,188 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more
mbrookhart commented on pull request #5231:
URL: https://github.com/apache/incubator-tvm/pull/5231#issuecomment-625942189
@tqchen @mbaret care to take another look after I updated the code based on
your comments?
This is an
mbrookhart edited a comment on pull request #5231:
URL: https://github.com/apache/incubator-tvm/pull/5231#issuecomment-625942189
@tqchen @mbaret Thanks for the comments! Care to take another look after I
updated the code based on your comments?
mbrookhart commented on pull request #5231:
URL: https://github.com/apache/incubator-tvm/pull/5231#issuecomment-625934854
@zhiics Thanks for the comments! completed those refactors.
This is an automated message from the
tmoreau89 commented on a change in pull request #5544:
URL: https://github.com/apache/incubator-tvm/pull/5544#discussion_r422274176
##
File path: src/runtime/rpc/rpc_session.h
##
@@ -189,6 +197,98 @@ class RPCSession {
*/
virtual bool IsLocalSession() const = 0;
+ //
tqchen opened a new pull request #5545:
URL: https://github.com/apache/incubator-tvm/pull/5545
This PR introduces WebGPU support to tvm. The WebGPU runtime is directly
built in javascript(as WebGPU is a first-class js API) and exposes back to the
tvm's runtime via PackedFuncs.
One
tqchen commented on pull request #5545:
URL: https://github.com/apache/incubator-tvm/pull/5545#issuecomment-625930446
Depends on https://github.com/apache/incubator-tvm/pull/5544
cc @kazum @masahi @jroesch @jwfromm
This is an automated email from the ASF dual-hosted git repository.
moreau pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git
The following commit(s) were added to refs/heads/master by this push:
new 95540d2 [CRT]fix to reduce RAM size
wpan11nv commented on pull request #5498:
URL: https://github.com/apache/incubator-tvm/pull/5498#issuecomment-625917497
Minor fix-up and cleaned comments
This is an automated message from the Apache Git Service.
To respond
This is an automated email from the ASF dual-hosted git repository.
moreau pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 36d7fd9 Add Onnx Pad v11 (#5539)
add e54e760 fix restructured text (#5541)
No new revisions were
spectrometerHBH edited a comment on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-625889950
This is an automated message from the Apache Git Service.
To respond to the message, please
spectrometerHBH edited a comment on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-625889950
I can not reproduce the errors in CI now. The errors reported in CI look
strange. I will try to fix it.
spectrometerHBH commented on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-625889950
I can not reproduce the errors in CI.
This is an automated message from the Apache Git Service.
liangfu commented on a change in pull request #5492:
URL: https://github.com/apache/incubator-tvm/pull/5492#discussion_r422219093
##
File path: src/runtime/hexagon/sim/driver/CMakeLists.txt
##
@@ -0,0 +1,62 @@
+# Licensed to the Apache Software Foundation (ASF) under one
kparzysz-quic commented on pull request #5492:
URL: https://github.com/apache/incubator-tvm/pull/5492#issuecomment-625877725
Rebased to force a new CI build.
This is an automated message from the Apache Git Service.
To
kparzysz-quic commented on a change in pull request #5492:
URL: https://github.com/apache/incubator-tvm/pull/5492#discussion_r422164839
##
File path: src/runtime/hexagon/sim/driver/CMakeLists.txt
##
@@ -0,0 +1,62 @@
+# Licensed to the Apache Software Foundation (ASF) under one
kparzysz-quic commented on a change in pull request #5492:
URL: https://github.com/apache/incubator-tvm/pull/5492#discussion_r422164839
##
File path: src/runtime/hexagon/sim/driver/CMakeLists.txt
##
@@ -0,0 +1,62 @@
+# Licensed to the Apache Software Foundation (ASF) under one
Hzfengsy commented on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-625818864
> Binds in some pass function is not clean for round trip dump ir, how do we
deal with it?
Do you mean that `buffer_bind` is not necessary to print after the pass
siju-samuel commented on pull request #5507:
URL: https://github.com/apache/incubator-tvm/pull/5507#issuecomment-625811064
@tqchen Thanks a lot. The 40kb ram reduced for me was due to the free of
graph_json I freed immediately after saving to runtime.
The other fix on setupstorage
dhruvaray commented on pull request #5329:
URL: https://github.com/apache/incubator-tvm/pull/5329#issuecomment-625768588
@siju-samuel - rebased and incorporated comments. Kindly review.
This is an automated message from the
dhruvaray commented on a change in pull request #5329:
URL: https://github.com/apache/incubator-tvm/pull/5329#discussion_r422081874
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -579,6 +582,63 @@ def convert_tanh(self, op):
return out
+def
dhruvaray commented on a change in pull request #5329:
URL: https://github.com/apache/incubator-tvm/pull/5329#discussion_r422081644
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -650,6 +693,82 @@ def test_all_resize():
dhruvaray commented on a change in pull request #5329:
URL: https://github.com/apache/incubator-tvm/pull/5329#discussion_r422081555
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -579,6 +582,63 @@ def convert_tanh(self, op):
return out
+def
dhruvaray commented on a change in pull request #5329:
URL: https://github.com/apache/incubator-tvm/pull/5329#discussion_r422081372
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -579,6 +582,63 @@ def convert_tanh(self, op):
return out
+def
dhruvaray commented on pull request #5447:
URL: https://github.com/apache/incubator-tvm/pull/5447#issuecomment-625745563
@siju-samuel - rebased and used new method
This is an automated message from the Apache Git Service.
To
xqdan commented on pull request #5483:
URL: https://github.com/apache/incubator-tvm/pull/5483#issuecomment-625717683
Binds in some pass function is not clean for round trip dump ir, how do we
deal with it?
This is an
wangzhao123456 opened a new issue #5543:
URL: https://github.com/apache/incubator-tvm/issues/5543
Thanks for participating in the TVM community! We use https://discuss.tvm.ai
for any general usage questions and discussions. The issue tracker is used for
actionable items such as feature
kazum commented on a change in pull request #5502:
URL: https://github.com/apache/incubator-tvm/pull/5502#discussion_r421972844
##
File path: tests/python/relay/test_op_grad_level1.py
##
@@ -69,7 +69,12 @@ def check_single_op(opfunc, ref):
kazum commented on a change in pull request #5502:
URL: https://github.com/apache/incubator-tvm/pull/5502#discussion_r421965191
##
File path: python/tvm/relay/op/_tensor_grad.py
##
@@ -116,12 +117,60 @@ def sinh_grad(orig, grad):
x = orig.args[0]
return [grad *
masahi commented on pull request #5539:
URL: https://github.com/apache/incubator-tvm/pull/5539#issuecomment-625660908
Thanks @mbrookhart
This is an automated message from the Apache Git Service.
To respond to the message,
This is an automated email from the ASF dual-hosted git repository.
masahi pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 1b17b73 Changes to cpp_rpc to make it work on Android (+ Hexagon
offloading) (#5535)
add 36d7fd9 Add
tobegit3hub opened a new pull request #5542:
URL: https://github.com/apache/incubator-tvm/pull/5542
Since different operating systems build the dynamic libraries with different
postfix, we will check the operating system to load the platform-specific file
in Python API.
67 matches
Mail list logo