echuraev commented on a change in pull request #7980:
URL: https://github.com/apache/tvm/pull/7980#discussion_r627087841
##
File path: src/runtime/metal/metal_module.mm
##
@@ -85,44 +112,45 @@ void SaveToBinary(dmlc::Stream* stream) final {
if (it != e.smap.end()) return i
This is an automated email from the ASF dual-hosted git repository.
jcf94 pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 148b852 Fix post-merge conflict between #7785 and #7945. (#7982)
add bbf7fdb [FIX] Fix autoscheduler tuning on spars
jcf94 merged pull request #7974:
URL: https://github.com/apache/tvm/pull/7974
--
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.
For queries about this service, please conta
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 f85cab2 [AOT] Introducing AOT in TVM (#7785)
add 148b852 Fix post-merge conflict between #7785 and #7945. (#7982)
tqchen merged pull request #7982:
URL: https://github.com/apache/tvm/pull/7982
--
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.
For queries about this service, please cont
huajsj commented on a change in pull request #7892:
URL: https://github.com/apache/tvm/pull/7892#discussion_r627006161
##
File path: tests/python/relay/test_analysis_pipeline.py
##
@@ -0,0 +1,144 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contri
JackYoustra commented on pull request #7985:
URL: https://github.com/apache/tvm/pull/7985#issuecomment-833135460
Hi @tmoreau89 , could you help to review this patch? You reviewed #3843 and
it's just an extension of that.
Thanks
Jack
--
This is an automated message from the Apac
JackYoustra opened a new pull request #7985:
URL: https://github.com/apache/tvm/pull/7985
#3843 fixed the infinite recursive call for the Xilinx boards, but didn't
fix it for the intel boards. This fixes it for the DE10 (same missing symbol
problem with same fix).
--
This is an automate
comaniac commented on a change in pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#discussion_r626970274
##
File path: python/tvm/relay/backend/compile_engine.py
##
@@ -253,7 +255,16 @@ def select_implementation(op, attrs, inputs, out_type,
target, use_autotvm=
tqchen commented on a change in pull request #7983:
URL: https://github.com/apache/tvm/pull/7983#discussion_r626971170
##
File path: include/tvm/runtime/c_backend_api.h
##
@@ -152,6 +152,16 @@ TVM_DLL int TVMBackendParallelBarrier(int task_id,
TVMParallelGroupEnv* penv);
*/
rkimball commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-833097344
Changed to print @tkonolige's message once if any of the kernels are
untuned. If you enable DEBUG logging then the old, detailed message is printed.
--
This is an automated messa
zackcquic commented on pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#issuecomment-833095927
> I will try and comment over the weekend, thanks for the PR!
Hi @jroesch:
Just a gentle ping to receive feedback on this PR.
Thanks,
Zack
--
This is an a
jwfromm closed pull request #7984:
URL: https://github.com/apache/tvm/pull/7984
--
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.
For queries about this service, please con
tqchen commented on pull request #7984:
URL: https://github.com/apache/tvm/pull/7984#issuecomment-833020246
@areusch already rasied https://github.com/apache/tvm/pull/7982
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and u
jwfromm commented on pull request #7984:
URL: https://github.com/apache/tvm/pull/7984#issuecomment-833016297
@tqchen we should probably try to merge this soon to fix builds.
--
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 #7984:
URL: https://github.com/apache/tvm/pull/7984
Somehow a call to `GraphAddCallNode` with a missing argument got merged and
is for some reason only inconsistently breaking builds. This PR adds the
missing argument and should resolve build issues.
--
tkonolige opened a new pull request #7983:
URL: https://github.com/apache/tvm/pull/7983
This PR adds an optional dependency on PAPI
(https://bitbucket.org/icl/papi/) in order to collect hardware performance
counters on CPU and CUDA. These performance counters include data like total
cycle
areusch commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626867573
##
File path: src/relay/backend/utils.cc
##
@@ -0,0 +1,44 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor licens
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626863808
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> st
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626863808
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> st
areusch opened a new pull request #7982:
URL: https://github.com/apache/tvm/pull/7982
@tqchen @giuseros
--
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.
For queries about t
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626849123
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> st
areusch commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626847192
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> stora
echuraev commented on issue #7981:
URL: https://github.com/apache/tvm/issues/7981#issuecomment-832939184
Thank you! I'll check then the PR will be merged.
--
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
comaniac commented on a change in pull request #7892:
URL: https://github.com/apache/tvm/pull/7892#discussion_r626816048
##
File path: tests/python/relay/test_analysis_pipeline.py
##
@@ -0,0 +1,144 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more cont
comaniac commented on a change in pull request #7892:
URL: https://github.com/apache/tvm/pull/7892#discussion_r626814675
##
File path: tests/python/relay/test_analysis_pipeline.py
##
@@ -0,0 +1,144 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more cont
areusch commented on pull request #7785:
URL: https://github.com/apache/tvm/pull/7785#issuecomment-832909339
thanks @giuseros for your hard work on this effort and for bearing with me
through a long review! and thanks @manupa-arm @jroesch @tqchen @tkonolige
@jwfromm for helping to review t
This is an automated email from the ASF dual-hosted git repository.
areusch pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 301ce49 [PROFILER] Add shape, structural hash, and layout information
to profiling (#7894)
add f85cab2 [AOT] Intr
areusch merged pull request #7785:
URL: https://github.com/apache/tvm/pull/7785
--
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.
For queries about this service, please con
comaniac commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832905591
> Another option is to detect when a user provides a tuning log. In that
scenario the intention is for log entries to be applied, and the user likely
should be notified on stdout t
huajsj commented on a change in pull request #7892:
URL: https://github.com/apache/tvm/pull/7892#discussion_r626785383
##
File path: tests/python/relay/test_analysis_pipeline.py
##
@@ -0,0 +1,144 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contri
This is an automated email from the ASF dual-hosted git repository.
junrushao pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 3f788b4 [FIX,VM] Fix get_outputs on the vm with a single output
(#7902)
add 301ce49 [PROFILER] Add shape, struc
junrushao1994 merged pull request #7894:
URL: https://github.com/apache/tvm/pull/7894
--
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.
For queries about this service, plea
csullivan commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832880809
Another option is to detect when a user provides a tuning log. In that
scenario the intention is for log entries to be applied, and the user likely
should be notified on stdout th
comaniac commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832877943
> `One or more operators have not been tuned. Please tune your model for
better performance (link), or use DEBUG logging level to see more details.`
Ah sorry for misunderstan
tkonolige commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832877006
`One or more operators have not been tuned. Please tune your model for
better performance (link), or use DEBUG logging level to see more details.`
--
This is an automated messag
areusch commented on pull request #7785:
URL: https://github.com/apache/tvm/pull/7785#issuecomment-832876837
@manupa-arm please take another look and explicitly approve if you're okay
with this!
https://tvm.apache.org/docs/contribute/code_review.html#approve-and-request-changes-explicit
comaniac commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832876203
> (C) sounds like a good option. I don't think we should say "config" in the
error message. New users will not know what we mean. Maybe say "operators"
instead?
"missing ope
tkonolige commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832872534
(C) sounds like a good option. I don't think we should say "config" in the
error message. New users will not know what we mean. Maybe say "operators"
instead?
--
This is an aut
icemelon9 commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832871378
I vote for option C, which is more comprehensive.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
comaniac edited a comment on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832870770
In summary:
(A) by @masahi: Turn the warning to DEBUG so that only the advance
developers would see it by purpose. We could of course add a paragraph of how
to enable
comaniac edited a comment on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832870770
In summary:
(A) by @masahi: Turn the warning to DEBUG so that only the advance
developers would see it by purpose. We could of course add a paragraph of how
to enable
comaniac commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832870770
In summary:
(A) by @masahi: Turn the warning to DEBUG so that only the advance
developers would see it by purpose. We could of course add a paragraph of how
to enable this m
areusch commented on pull request #7902:
URL: https://github.com/apache/tvm/pull/7902#issuecomment-832854873
thanks @tkonolige , the PR is now merged!
--
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 g
This is an automated email from the ASF dual-hosted git repository.
areusch pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/main by this push:
new 3f788b4 [FIX,VM] Fix get_outputs on the vm with a sin
areusch merged pull request #7902:
URL: https://github.com/apache/tvm/pull/7902
--
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.
For queries about this service, please con
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626736200
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> st
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626736200
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> st
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626736200
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> st
tkonolige commented on pull request #7975:
URL: https://github.com/apache/tvm/pull/7975#issuecomment-832849433
I think we should keep the error message, but maybe we can improve it. Maybe
say something like "tune your model for better performance" and then link to a
page on tuning?
--
T
tqchen commented on pull request #7875:
URL: https://github.com/apache/tvm/pull/7875#issuecomment-832848378
Thanks @apeskov . This is an important step. Would be great if we can think
about de-couple further to reverse the dependency. (e.g. make the macho depend
on libtvm as a plugin).
comaniac merged pull request #7951:
URL: https://github.com/apache/tvm/pull/7951
--
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.
For queries about this service, please co
comaniac commented on pull request #7951:
URL: https://github.com/apache/tvm/pull/7951#issuecomment-832838925
Thanks @zxy844288792 @tkonolige @yongwww
--
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
This is an automated email from the ASF dual-hosted git repository.
comaniac pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 3e35130 [Relay][Parser][Bugfix] Fix parsing hierarchical attibute
names (#7976)
add ae31a33 [Frontend][Tensorflo
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 9b8e994 [DOC] Add how to enable IR debug messages. (#7978)
add 3e35130 [Relay][Parser][Bugfix] Fix parsing hierarch
tqchen merged pull request #7976:
URL: https://github.com/apache/tvm/pull/7976
--
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.
For queries about this service, please cont
tqchen commented on pull request #7976:
URL: https://github.com/apache/tvm/pull/7976#issuecomment-832835440
Thanks @altanh !
--
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.
comaniac commented on issue #7981:
URL: https://github.com/apache/tvm/issues/7981#issuecomment-832833798
https://github.com/apache/tvm/pull/7977 should solve this problem.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and u
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626691300
##
File path: tests/python/unittest/test_micro_model_library_format.py
##
@@ -167,11 +199,68 @@ def @main(%a : Tensor[(1, 2), uint8], %b : Tensor[(1, 2),
manupa-arm commented on pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#issuecomment-832805866
Hi @areusch ,
Thanks for the review.
I've addressed your comments and answered questions. PTAL when you have some
time.
--
This is an automated message from the Apac
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626689505
##
File path: src/relay/backend/utils.cc
##
@@ -0,0 +1,44 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor lic
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626688186
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -551,10 +713,14 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslator> st
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626686759
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -189,9 +192,109 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslatorIsIn
manupa-arm commented on a change in pull request #7938:
URL: https://github.com/apache/tvm/pull/7938#discussion_r626685988
##
File path: src/relay/backend/graph_executor_codegen.cc
##
@@ -189,9 +192,109 @@ class GraphExecutorCodegen : public
backend::MemoizedExprTranslatorIsIn
trevor-m commented on pull request #7949:
URL: https://github.com/apache/tvm/pull/7949#issuecomment-832781097
Thanks for reviewing @mbrookhart @leandron !
Tests are passing
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub a
daniperfer commented on issue #7971:
URL: https://github.com/apache/tvm/issues/7971#issuecomment-832759852
Thanks for answering @masahi.
I see what you said. I have removed my modifications to the tutorial script,
back to the original code, and I've launched again the original script
`t
apeskov commented on pull request #7875:
URL: https://github.com/apache/tvm/pull/7875#issuecomment-832657486
@tqchen Your recommendation was fulfilled.
* The conflict sources was extracted into separate repo
https://github.com/octoml/macho-dyld
* TVM is linked with macho_dyld dynamica
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 26a5e29 [Frontend][Keras] Fix Dense with 3d inputs (#7753)
add 9b8e994 [DOC] Add how to enable IR debug messages. (
tqchen merged pull request #7978:
URL: https://github.com/apache/tvm/pull/7978
--
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.
For queries about this service, please cont
tqchen commented on pull request #7978:
URL: https://github.com/apache/tvm/pull/7978#issuecomment-832648324
Thank you @zackcquic !
--
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
apeskov commented on pull request #7875:
URL: https://github.com/apache/tvm/pull/7875#issuecomment-832645618
Reopen. Was closed by mistake.
--
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
apeskov closed pull request #7875:
URL: https://github.com/apache/tvm/pull/7875
--
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.
For queries about this service, please con
echuraev opened a new issue #7981:
URL: https://github.com/apache/tvm/issues/7981
Hello,
I faced the problem that the TVM export methods doesn’t strip original
collection of constant tensor in case of usage external runtime like BNNS/ACL
and others. First time we store constants in
stoa commented on a change in pull request #7742:
URL: https://github.com/apache/tvm/pull/7742#discussion_r626494235
##
File path: src/runtime/stm32/graph_runtime.cc
##
@@ -0,0 +1,529 @@
+/*
Review comment:
How are you, Andrew @areusch
I have reorganized the code
stoa commented on a change in pull request #7742:
URL: https://github.com/apache/tvm/pull/7742#discussion_r626494235
##
File path: src/runtime/stm32/graph_runtime.cc
##
@@ -0,0 +1,529 @@
+/*
Review comment:
How are you, Andrew @areusch
I have reorganized the code
stoa commented on a change in pull request #7742:
URL: https://github.com/apache/tvm/pull/7742#discussion_r626494235
##
File path: src/runtime/stm32/graph_runtime.cc
##
@@ -0,0 +1,529 @@
+/*
Review comment:
How are you, Andrew @areusch
I have reorganized the code
elvin-nnov commented on a change in pull request #7980:
URL: https://github.com/apache/tvm/pull/7980#discussion_r626489762
##
File path: src/runtime/metal/metal_module.mm
##
@@ -85,44 +112,45 @@ void SaveToBinary(dmlc::Stream* stream) final {
if (it != e.smap.end()) return
echuraev opened a new pull request #7980:
URL: https://github.com/apache/tvm/pull/7980
Refactor Metal module to build each kernel separately. This should help
to avoid potential problem then generated blockSize is more than
maxTotalThreadsPerThreadgroup.
Thanks for contributing
vinceab commented on pull request #7922:
URL: https://github.com/apache/tvm/pull/7922#issuecomment-832582235
> > > The change looks OK to me. Apart from my comments below, is there a
chance to add a test case for this PR?
> >
> >
> > I will do the update of the patch.
> > Rega
Tantalus13A98B5F commented on pull request #7250:
URL: https://github.com/apache/tvm/pull/7250#issuecomment-832537668
Hi @areusch if I'm not mistaking you, I just created a post
[here](https://discuss.tvm.apache.org/t/autotvm-interface-for-fixed-input-data/9893).
--
This is an automated
80 matches
Mail list logo