junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668476702
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668438352
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
This is an automated email from the ASF dual-hosted git repository.
masahi pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from d043cb9 [BugFix][TOPI] Fix the integer overflow problem of the
scatter_nd op. (#8415)
add 807373c Add qnn batch_ma
masahi commented on pull request #8401:
URL: https://github.com/apache/tvm/pull/8401#issuecomment-878787487
Thanks @elvin-n @jwfromm
--
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
masahi merged pull request #8401:
URL: https://github.com/apache/tvm/pull/8401
--
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...
chiwwang commented on pull request #8448:
URL: https://github.com/apache/tvm/pull/8448#issuecomment-878777359
Thanks @echuraev @comaniac !
Great suggestions! I will add information on the node rectangle.
This PR will keep as draft until some of following TODOs done.
1. Display suita
rijulg commented on issue #7886:
URL: https://github.com/apache/tvm/issues/7886#issuecomment-878765235
> Unfortunately this gotcha is common and not well documented. The error
means you need to use VM compiler and runtime for this model. I'll talk to
people to discuss doc improvement.
ya0guang commented on issue #8439:
URL: https://github.com/apache/tvm/issues/8439#issuecomment-878745630
> @ya0guang Sorry for the late reply, this app has not been maintained since
#6862, if you want to make the demo work, it's highly suggested to downgrade
the version of TVM to `44ff1f3b
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668395730
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668395363
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
mbrookhart commented on pull request #8415:
URL: https://github.com/apache/tvm/pull/8415#issuecomment-878738326
Thanks @zhuwenxi @tkonolige
--
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 s
wyc-ruiker commented on a change in pull request #8443:
URL: https://github.com/apache/tvm/pull/8443#discussion_r668389145
##
File path: tests/python/frontend/pytorch/test_forward.py
##
@@ -3912,6 +3912,109 @@ def forward(self, x):
verify_model(Flip(axis=-1), input_data=in
This is an automated email from the ASF dual-hosted git repository.
mbrookhart pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 62adc77 [MyPy] Extend type checking and annotation for TIR (#8429)
add d043cb9 [BugFix][TOPI] Fix the integer o
mbrookhart merged pull request #8415:
URL: https://github.com/apache/tvm/pull/8415
--
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-unsubsc
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668390088
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
wyc-ruiker commented on a change in pull request #8443:
URL: https://github.com/apache/tvm/pull/8443#discussion_r668389145
##
File path: tests/python/frontend/pytorch/test_forward.py
##
@@ -3912,6 +3912,109 @@ def forward(self, x):
verify_model(Flip(axis=-1), input_data=in
wyc-ruiker commented on a change in pull request #8443:
URL: https://github.com/apache/tvm/pull/8443#discussion_r668389145
##
File path: tests/python/frontend/pytorch/test_forward.py
##
@@ -3912,6 +3912,109 @@ def forward(self, x):
verify_model(Flip(axis=-1), input_data=in
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668387903
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668387313
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668386464
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668383535
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
zhuwenxi commented on pull request #8415:
URL: https://github.com/apache/tvm/pull/8415#issuecomment-878724817
@mbrookhart I've fixed the UT failure.
--
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
masahi commented on pull request #8447:
URL: https://github.com/apache/tvm/pull/8447#issuecomment-878694020
thanks @vvchernov this is great. Please fix the lint issue.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use t
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from cd5a20e [COMMUNITY] @junrushao1994 -> PMC (#8450)
add 62adc77 [MyPy] Extend type checking and annotation for TIR (#
tqchen merged pull request #8429:
URL: https://github.com/apache/tvm/pull/8429
--
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...
comaniac commented on pull request #8069:
URL: https://github.com/apache/tvm/pull/8069#issuecomment-878686422
Yes. Please go to the discuss forum. You can refer to this PR and tag
relevant people in the post.
--
This is an automated message from the Apache Git Service.
To respond to the
comaniac commented on a change in pull request #8457:
URL: https://github.com/apache/tvm/pull/8457#discussion_r668341809
##
File path: python/tvm/topi/cuda/injective.py
##
@@ -54,6 +64,26 @@ def schedule_injective_from_existing(sch, out):
try:
const_size = utils
vinx13 commented on a change in pull request #8457:
URL: https://github.com/apache/tvm/pull/8457#discussion_r668341059
##
File path: python/tvm/topi/cuda/injective.py
##
@@ -54,6 +64,26 @@ def schedule_injective_from_existing(sch, out):
try:
const_size = utils.g
CoinCheung commented on pull request #8069:
URL: https://github.com/apache/tvm/pull/8069#issuecomment-878683239
Hi,
I am not sure whether it is a usage question or the code can be refined, I
am using a quite new commit pulled from github:
![图片](https://user-images.githubuser
YuchenJin commented on a change in pull request #8418:
URL: https://github.com/apache/tvm/pull/8418#discussion_r668337617
##
File path: python/tvm/relay/build_module.py
##
@@ -555,6 +559,9 @@ def create_executor(kind="debug", mod=None, device=None,
target="llvm"):
else:
YuchenJin commented on a change in pull request #8418:
URL: https://github.com/apache/tvm/pull/8418#discussion_r668337617
##
File path: python/tvm/relay/build_module.py
##
@@ -555,6 +559,9 @@ def create_executor(kind="debug", mod=None, device=None,
target="llvm"):
else:
comaniac opened a new pull request #8457:
URL: https://github.com/apache/tvm/pull/8457
Per discussion in
https://discuss.tvm.apache.org/t/cuda-enable-half2-in-cuda-injective-schedule/10441,
this PR improves the CUDA injective schedule to benefit more from `half2` when
working on `float16`
comaniac commented on pull request #8069:
URL: https://github.com/apache/tvm/pull/8069#issuecomment-878678403
- You might need to rebuild TVM.
- Please do not ask usage questions under feature PRs. Please post the
question to the discuss forum instead.
--
This is an automated message
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668335454
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
CoinCheung commented on pull request #8069:
URL: https://github.com/apache/tvm/pull/8069#issuecomment-878677560
HI,
I tried this:
```python
def compile_model(mod, params, target, logfile, save_path):
tvm.relay.backend.compile_engine.get().clear()
mod = tvm.relay.tr
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from c81d533 [Bugfix] Fix broadcast type func with incomplete type (#8438)
add cd5a20e [COMMUNITY] @junrushao1994 -> PMC
tqchen merged pull request #8450:
URL: https://github.com/apache/tvm/pull/8450
--
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...
trevor-m commented on a change in pull request #8455:
URL: https://github.com/apache/tvm/pull/8455#discussion_r668329230
##
File path: tests/python/contrib/test_onnx.py
##
@@ -655,6 +655,26 @@ def verify_cast(dshape, dtype):
verify_cast(i, o_dtype)
+def test_re
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668328703
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
tkonolige commented on a change in pull request #7983:
URL: https://github.com/apache/tvm/pull/7983#discussion_r668327260
##
File path: include/tvm/runtime/contrib/papi.h
##
@@ -0,0 +1,99 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contribu
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668326765
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668326423
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668325916
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668319732
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668323104
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668322103
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668319732
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
jwfromm commented on pull request #8456:
URL: https://github.com/apache/tvm/pull/8456#issuecomment-878658050
@mbrookhart @AndrewZhaoLuo can you guys take a look at this PR?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
jwfromm opened a new pull request #8456:
URL: https://github.com/apache/tvm/pull/8456
While playing with [onnxruntime's quantization
tools](https://www.onnxruntime.ai/docs/how-to/quantization.html) I found that
`ConvInteger` nodes were heavily used and pretty easy to map to `qnn.conv2d`.
mbrookhart commented on issue #8177:
URL: https://github.com/apache/tvm/issues/8177#issuecomment-878656650
I don't see any excessive memory use when I put the test on a loop locally
or run the full tensorflow test suite locally with this version of TF...
--
This is an automated messa
csullivan commented on a change in pull request #8366:
URL: https://github.com/apache/tvm/pull/8366#discussion_r668312348
##
File path: src/tir/transforms/update_pointer_storage_scope.cc
##
@@ -0,0 +1,91 @@
+/*
Review comment:
Thanks @masahi. I tracked this issue down t
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668311883
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
ymwangg commented on a change in pull request #8440:
URL: https://github.com/apache/tvm/pull/8440#discussion_r668311147
##
File path: python/tvm/relay/frontend/onnx.py
##
@@ -678,25 +678,32 @@ def _impl_v1(cls, inputs, attr, params):
# When performing a batch matmul, w
tqchen commented on a change in pull request #7983:
URL: https://github.com/apache/tvm/pull/7983#discussion_r668303761
##
File path: include/tvm/runtime/profiling.h
##
@@ -150,6 +151,26 @@ class Timer : public ObjectRef {
Timer DefaultTimer(Device dev);
namespace profiling
tqchen commented on a change in pull request #7983:
URL: https://github.com/apache/tvm/pull/7983#discussion_r668303499
##
File path: include/tvm/runtime/contrib/papi.h
##
@@ -0,0 +1,99 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor
mbrookhart commented on issue #8177:
URL: https://github.com/apache/tvm/issues/8177#issuecomment-878644292
Still hitting the Tensorflow OOM issue on the T4. Any ideas?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use t
schilkunda-amba opened a new pull request #8455:
URL: https://github.com/apache/tvm/pull/8455
* Added support for resize2d op
* Added unit 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
This is an automated email from the ASF dual-hosted git repository.
mbrookhart pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 15bdf28 Fix address and port reported by android_rpc to tracker
(#8405)
add c81d533 [Bugfix] Fix broadcast typ
mbrookhart commented on pull request #8438:
URL: https://github.com/apache/tvm/pull/8438#issuecomment-878641974
Thanks @comaniac !
--
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
mbrookhart merged pull request #8438:
URL: https://github.com/apache/tvm/pull/8438
--
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-unsubsc
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668299195
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
mbrookhart commented on pull request #8390:
URL: https://github.com/apache/tvm/pull/8390#issuecomment-878640463
@electriclilies It looks like the CPU image doesn't have the necessary
depencencies for ONNX. Could you make a PR to run install onnx in the CPU
docker file?
--
This is an aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668296259
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668282356
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668281045
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668275733
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668275556
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
masahi commented on a change in pull request #8366:
URL: https://github.com/apache/tvm/pull/8366#discussion_r668267879
##
File path: src/tir/transforms/update_pointer_storage_scope.cc
##
@@ -0,0 +1,91 @@
+/*
Review comment:
Hmm interesting. `attr::storage_scope` and the
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668259724
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
Lunderberg commented on pull request #8446:
URL: https://github.com/apache/tvm/pull/8446#issuecomment-878597423
@mvermeulen Thank you for the fix! I had initially thought that the CI
would catch a failing compile, but [it looks
like](https://github.com/apache/tvm/blob/main/conda/recipe/bu
comaniac commented on a change in pull request #8440:
URL: https://github.com/apache/tvm/pull/8440#discussion_r668243600
##
File path: python/tvm/relay/frontend/onnx.py
##
@@ -678,25 +678,32 @@ def _impl_v1(cls, inputs, attr, params):
# When performing a batch matmul,
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668238062
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
junrushao1994 commented on a change in pull request #5:
URL: https://github.com/apache/tvm-rfcs/pull/5#discussion_r668227109
##
File path: rfcs/0001-meta-schedule-autotensorir.md
##
@@ -0,0 +1,444 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+* Feature Name: Meta Schedule (Formerly Aut
emijiayw commented on a change in pull request #8426:
URL: https://github.com/apache/tvm/pull/8426#discussion_r668211743
##
File path: tests/python/frontend/onnx/test_forward.py
##
@@ -4872,6 +4872,66 @@ def test_qlinearadd():
verify_qlinearadd([5, 1, 7], [2, 7], [5, 2, 7]
Mousius commented on pull request #8453:
URL: https://github.com/apache/tvm/pull/8453#issuecomment-878534891
Did you try adding `--unpacked-api=1` to the target options? That should
reduce the stack usage
--
This is an automated message from the Apache Git Service.
To respond to the mess
tkonolige commented on a change in pull request #8437:
URL: https://github.com/apache/tvm/pull/8437#discussion_r668068298
##
File path: tests/python/topi/python/test_topi_sparse.py
##
@@ -212,14 +210,15 @@ def check_device(device):
def test_csrmv():
-verify_dynamic_csr
manupa-arm commented on a change in pull request #9:
URL: https://github.com/apache/tvm-rfcs/pull/9#discussion_r667922010
##
File path: rfcs/0009_Unified_Static_Memory_Planning.md
##
@@ -0,0 +1,467 @@
+Feature Name: Unified Static Memory Planner
+Start Date: 2021 June 1
ymwangg commented on a change in pull request #8440:
URL: https://github.com/apache/tvm/pull/8440#discussion_r668122649
##
File path: python/tvm/relay/frontend/onnx.py
##
@@ -670,68 +670,67 @@ class MatMul(OnnxOpConverter):
@classmethod
def _impl_v1(cls, inputs, attr,
jwfromm commented on pull request #8447:
URL: https://github.com/apache/tvm/pull/8447#issuecomment-878407254
@masahi can you help review and shepherd this PR?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL ab
tqchen commented on pull request #8366:
URL: https://github.com/apache/tvm/pull/8366#issuecomment-878485708
will let @vinx13 manage this PR
--
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 sp
Lunderberg opened a new pull request #8451:
URL: https://github.com/apache/tvm/pull/8451
Previously, any class that doesn't raise a TypeError in copy.deepcopy could
be used as a return value in a `@tvm.testing.fixture`. This has the
possibility of incorrectly copying classes inherit the d
emijiayw commented on a change in pull request #8451:
URL: https://github.com/apache/tvm/pull/8451#discussion_r668178028
##
File path: python/tvm/testing.py
##
@@ -1199,18 +1238,15 @@ def wrapper(*args, **kwargs):
except KeyError:
cached_value = ca
echuraev commented on pull request #8448:
URL: https://github.com/apache/tvm/pull/8448#issuecomment-878266242
I like this feature! It is very useful. I quickly took a look on your PR and
I have one question.
I think it will be useful to display some information on the nodes of the
grap
mbrookhart commented on pull request #8434:
URL: https://github.com/apache/tvm/pull/8434#issuecomment-878431672
This is interesting, I tested locally with a debug build and don't see these
errors, but if I recompile with a release build, they show up. Will debug.
--
This is an automated
Mousius opened a new pull request #8452:
URL: https://github.com/apache/tvm/pull/8452
This patch passes the resource_handle variable through the AOT executor
down to the backend operators. In order to model this on the `PrimFunc`
the resource_handle was added as a property, the resourc
emijiayw commented on a change in pull request #8447:
URL: https://github.com/apache/tvm/pull/8447#discussion_r668195327
##
File path: tests/python/frontend/pytorch/test_lstms.py
##
@@ -0,0 +1,382 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contr
rohanmukh closed pull request #8238:
URL: https://github.com/apache/tvm/pull/8238
--
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
Lunderberg commented on pull request #8383:
URL: https://github.com/apache/tvm/pull/8383#issuecomment-878410739
Current failures.
* `test_conv2d_nchw` doesn't support asymmetric padding on cudnn. Can add
an explicit `pytest.xfail` on that test case once #8433 is merged.
* `test_b
mehrdadh opened a new pull request #8453:
URL: https://github.com/apache/tvm/pull/8453
Due to recent changes in AOT, we need to increase the stack size to pass
test_zephyr_aot test.
cc @gromero @areusch
--
This is an automated message from the Apache Git Service.
To respond t
mehrdadh commented on pull request #8453:
URL: https://github.com/apache/tvm/pull/8453#issuecomment-878545044
--
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
mikepapadim commented on pull request #8429:
URL: https://github.com/apache/tvm/pull/8429#issuecomment-878125080
@jroesch @tqchen @YuchenJin
--
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
tkonolige commented on a change in pull request #7983:
URL: https://github.com/apache/tvm/pull/7983#discussion_r668177949
##
File path: include/tvm/runtime/contrib/papi.h
##
@@ -0,0 +1,99 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contribu
ANSHUMAN87 commented on a change in pull request #8437:
URL: https://github.com/apache/tvm/pull/8437#discussion_r668159979
##
File path: tests/python/topi/python/test_topi_sparse.py
##
@@ -212,14 +210,15 @@ def check_device(device):
def test_csrmv():
-verify_dynamic_cs
csullivan commented on a change in pull request #8366:
URL: https://github.com/apache/tvm/pull/8366#discussion_r668144726
##
File path: src/tir/transforms/update_pointer_storage_scope.cc
##
@@ -0,0 +1,91 @@
+/*
Review comment:
I think we need to also update `lower_devic
jwfromm merged pull request #8405:
URL: https://github.com/apache/tvm/pull/8405
--
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..
vvchernov commented on pull request #8447:
URL: https://github.com/apache/tvm/pull/8447#issuecomment-878478472
Hello @masahi, I have finished my work and waiting for your review and
response
--
This is an automated message from the Apache Git Service.
To respond to the message, please lo
zxy844288792 opened a new pull request #8454:
URL: https://github.com/apache/tvm/pull/8454
This PR adds support for TensorList ops in TF2 frontend parser. Ops like
TensorListFromTensor, TensorListGetItem etc are common in graphdef of TF2 OD
models like Faster-RCNN etc.
This PR is in
Tomakko commented on issue #6801:
URL: https://github.com/apache/tvm/issues/6801#issuecomment-878247737
Have you found a workaround 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
tqchen commented on a change in pull request #7983:
URL: https://github.com/apache/tvm/pull/7983#discussion_r668121949
##
File path: include/tvm/runtime/profiling.h
##
@@ -150,6 +151,26 @@ class Timer : public ObjectRef {
Timer DefaultTimer(Device dev);
namespace profiling
tkonolige commented on a change in pull request #9:
URL: https://github.com/apache/tvm-rfcs/pull/9#discussion_r668065713
##
File path: rfcs/0009_Unified_Static_Memory_Planning.md
##
@@ -0,0 +1,473 @@
+Feature Name: Unified Static Memory Planner
+Start Date: 2021 June 1
1 - 100 of 121 matches
Mail list logo