[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-343655159
 
 
   Cool, this should be tried. Weird it does not raise issues for all the other 
setups. Well, c++ compilers are not all the same.
   
   The exp function is deep in a nested namespace, though


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-343414021
 
 
   @asmushetzel Hi Asmus, some people continue to have problems building on 
some CentOS setup that I cannot work on.
   Maybe we can go over my changes briefly, and you flag things that may be 
non-CUDA?
   However, I worked on this with @piiswrong 
   The files in questions are src/operator/
   - mshadow_op.h
   - math_functions-inl.h (new)
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-343413652
 
 
   Hello,
   I'd like to help, but I just cannot reproduce the error (given it is still 
the same as the one described above). Here some points:
   
   - I'd need the full error output. If some template is expanded, would have 
to know for which
 ops in mshadow_op.h. An obvious thing to try is to revert things back to 
before my changes
 for these ops only
   - Why was the original issue closed? Something must have solved the problem, 
and now maybe
  this is just a different problem? At least, we need error outputs for the 
new problem, and maybe
  this should be a different issue?
   - What is your hunch what is happening here? What cc or cu code is actually 
built here?
   
   My hunches of what to try:
   - The new math_functions-inl.h introduces a lot of functions exp, log, ... 
in the namespace
  mxnet::op::math. Maybe this clashes with something else you are doing? 
One thing to try
  would be to rename the namespace math => math_debug. This also needs 
mshadow_op.h
  changes, but nowhere else (I think)
   - Figure out exactly which ops in mshadow_op.h are implicated, and then 
change their code
  back to older versions
   
   - To explain what my changes did: The mshadow_op.h ops now consistently cast 
inputs to
  float for all DType != double, but leave them double for DType = double. 
The computation
  is then done in float or double, and if DType != double, we cast back to 
DType at the end.
  The input is cast with static_cast(a), the result is cast back 
with DType(result).
   - The code before my changes did things differently:
  The forward ops were always computed in float, also if DType = double. 
Also, they would
   do ::expf(a) instead of ::expf(static_cast(a)), which should be 
the same.
   More serious, the gradient (backward) ops were often doing computations 
in DType if
   only arithmetic was involved (*, +, even /), but in float if math.h 
functions were involved.
   This is plain wrong. In my changes, also all gradient ops cast to float, 
whether math.h
   functions are involved or not
   
   @zhreshold How would this fail in CI in windows, etc., if it does not fail 
for all other PRs that passed CI since my changes? This makes no sense, right? 
If CI fails for you, please provide some full error outputs. BTW: CI fails 
randomly for many reasons all the time, unfortunately. Make sure it fails with 
the errors you are reporting here (and please, re-report them).
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-343263971
 
 
   Please contact the person who closed this issue, because it was solved for 
him. 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-343066287
 
 
   Hello,
   The issue was closed, because the person who raised it figured it out.
   Please contact this person.
   
   On Nov 8, 2017 11:42 PM, "Joshua Z. Zhang"  wrote:
   
   > @mseeger  Have you figured out what is the
   > cause of this problem?
   > I encountered exactly same error during CI compilation in #8582
   > 
   >
   > /workspace/mshadow/mshadow/././././cuda/tensor_gpu-inl.cuh(75): error: 
expression preceding parentheses of apparent call must have (pointer-to-) 
function type
   >
   > I can't reproduce it on ubuntu, and cannot locate where's the origin of
   > this error in source code.
   >
   > ?
   > You are receiving this because you were mentioned.
   > Reply to this email directly, view it on GitHub
   > 
,
   > or mute the thread
   > 

   > .
   >
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-341126321
 
 
   Hello, does your problem persist?
   We have just pushed a new PR, changing mshadow_op.h once more, and also 
smooth_l1_*.
   Please try again.
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-338326612
 
 
   Any news here? Did you try with the latest master?
   
   You could also try replacing smooth_l1_* with a trivial function, returning 
DType(0). Does this compile?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-337862092
 
 
   The error messages seem to depend on mshadow_op.h only through 
smooth_l1_gradient. And that code is really independent of the changes I did.
   
   Are you sure this worked before my changes?
   
   Another thing: The code for smooth_l1_gradient does not do anything useful 
for integral DType. The error messages are about using it for DType = uint8_t. 
Seems to make very little sense to me.
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-337678983
 
 
   The line in question in tensor_gpu-inl.cuh is:
   
   Saver::Save(dst.REval(y, x), exp.Eval(y, x));
   
   Is this the complete error you are getting? Why can't I see what DstPlan or 
Plan are? Or for what arguments dst or exp this is called when it fails?
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on issue #8338: master branch cannot build on centos 7 with cuda-8.0

[GitBox]

mseeger commented on issue #8338: master branch cannot build on centos 7 with 
cuda-8.0
URL: 
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-337676796
 
 
   The warnings relate to smooth_l1_loss in mshadow_op.h. What I did there, is 
to replace constants like 0.5f by DType(0.5f), to make it look similar to 
elsewhere.
   
   This can be reverted.
   
   I am not enough an expert on CUDA or mshadow to understand what problems 
this is causing, and I also don't have a clue about "centos 7 with cuda-8.0". 
After all, lots of GPU unit tests work without any issues.
   
   But the first attempt would be to replace smooth_l1_loss and 
smooth_l1_gradient in mshadow_op.h by the old code, and see whether that helps.
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services