-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Difficulty compiling on NVidia tegra #7
Comments
Hi Sam, |
Thank you very much for the prompt reply!
I am out of the office until Monday: I will try to get you a response before then, but I may be unable. I will definitely reply by then!
Sam
…
|
Good news, Marek: the updated files from commit 05addaf build on the
NVidia TX1 with Tegra Linux. I have not yet run the code, I can only report
that they successfully build.
However, I did have to turn off the python and swig bindings, since the
configuration there is still a bit off. I think the issue stems from the
tegra-ubuntu's use of cmake 3.5.1. First, cmake couldn't find the python3
libraries (even though they are there) and then cmake had trouble with the
swig_add_library command. Poking around, it appears "swig_add_library" was
added in cmake 3.8, so I believe this has to do with the fact that I am
running an old configuration.
I'll start working with the examples and see how it runs, then get back to
you with more if I have any. Thanks!
…------------------------------------
Sam Waldman
On Wed, Jul 3, 2019 at 5:48 PM Sam Waldman <[email protected]>
wrote:
Thank you very much for the prompt reply!
I am out of the office until Monday: I will try to get you a response
before then, but I may be unable. I will definitely reply by then!
Sam
On Jul 3, 2019, at 5:25 PM, Marek Moeckel ***@***.***>
wrote:
The minor CMake issues you saw should be fixed with cb486dc
<cb486dc>
5b2097e
<5b2097e>
and 794a8f1
<794a8f1>
.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#7>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AMQUD2ELV323IPLXTKW54JTP5U7QBANCNFSM4H5L5H6Q>
.
|
Hi Sam, thanks for reporting back! Re SWIG, the command swig_add_library was added to replace an older command. If you dig around in CMakeLists.txt's history (1f7cc4b#diff-af3b638bc2a3e6c650974192a53c7291) you'll find it. Maybe it should be added back in as an option for older versions of CMake. |
Fixed in 0802510. |
Goodmorning OPI Team -
I am hoping to use OPI for some basic work -- I have a generic interest in GPU based computation, as well as a specific interest in orbital propagation and radiation tolerant computing. OPI seems like a very promising framework, but I am unable to get the code compiled. Any help would be greatly appreciated, as I suspect I am doing something basic incorrectly.
System:
NVidia TX1 running tegra-ubuntu aarch64
CUDA release 9.0
cmake version 3.5.1
g++ (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609
After
mkdir build; cd build; cmake ..
I receive the following error:This is fixed by replacing with "NOT ${CMAKE_VERSION} VERSION_LESS "3.12 " based on the insight gleaned here: [https://github.com/Dav1dde/glad/issues/134]. This modification allows cmake to finish.
Make proceeds to the 64% level and crashes out when building propagator_properties_cpp.cpp with an error about "virtual functions are pure". This is something I tried to "fix" by changing the function definition so the virtual function is properly overridden, but this seems indicative of a compiler configuration error. As far as I can tell, the compiler is running with C++11 enabled based on CMAKE messages. However, the virtual message and later messages regarding "nullptr" definition suggest that a compiler configuration is wrong still.
Any help greatly appreciated.
Thanks!
sam
The text was updated successfully, but these errors were encountered: