intel/2017b toolchain definition (REVIEW)#4618
Conversation
…017.4.196-GCC-7.1.0-2.28.eb, ifort-2017.4.196-GCC-7.1.0-2.28.eb, impi-2017.3.196-iccifort-2017.4.196-GCC-7.1.0-2.28.eb, imkl-2017.3.196-iimpi-2017.05-GCC-7.1.0-2.28.eb, iccifort-2017.4.196-GCC-7.1.0-2.28.eb, iimpi-2017.05-GCC-7.1.0-2.28.eb
|
Test report by @boegel |
|
Test report by @boegel |
|
Reply from Intel support on use of Intel compilers 2017 update 4 on top of GCC 7.1.0: So I guess we'll stick to GCC 6.3.0... Thoughts @wpoely86, @akesandgren, @damianam, @ocaisa? |
|
If that is what Intel says, we stick to it? No reason to go all adventurous, especially as GCC 7 has no killer features (or not as far as I know) |
|
Yeah, better to keep it supported.
Sent with AquaMail for Android
http://www.aqua-mail.com
…On June 21, 2017 18:11:49 Kenneth Hoste ***@***.***> wrote:
Reply from Intel support on use of Intel compilers 2017 update 4 on top of
GCC 7.1.0:
```
In the release note, we specify that we only have support for GCC 6.
https://software.intel.com/en-us/articles/intel-c-compiler-170-for-linux-release-notes-for-intel-parallel-studio-xe-2017
Even the 2018 update will only support GCC 6.3
You might end up having some issues of compatibility.
```
So I guess we'll stick to GCC 6.3.0...
Thoughts @wpoely86, @akesandgren, @damianam, @ocaisa?
--
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub:
#4618 (comment)
|
|
Test report by @boegel |
| toolchain = {'name': 'dummy', 'version': 'dummy'} | ||
|
|
||
| compver = '2017.4.196' | ||
| gccver = '6.3.0' |
There was a problem hiding this comment.
GCC 6.4.0 is imminent: https://gcc.gnu.org/ml/gcc/2017-06/msg00101.html (hat tip @ocaisa)
|
Test report by @boegel |
| toolchainopts = {'usempi': True} | ||
|
|
||
| sources = [SOURCELOWER_TAR_GZ] | ||
| source_urls = ['http://www.netlib.org/benchmark/%(namelower)s'] |
| @@ -0,0 +1,33 @@ | |||
| # This is an easyconfig file for EasyBuild, see http://hpcugent.github.io/easybuild | |||
| @@ -0,0 +1,19 @@ | |||
| # This is an easyconfig file for EasyBuild, see http://hpcugent.github.io/easybuild | |||
| @@ -0,0 +1,36 @@ | |||
| # This is an easyconfig file for EasyBuild, see http://hpcugent.github.io/easybuild | |||
| @@ -0,0 +1,20 @@ | |||
| # This is an easyconfig file for EasyBuild, see http://hpcugent.github.io/easybuild | |||
| @@ -0,0 +1,36 @@ | |||
| # This is an easyconfig file for EasyBuild, see http://hpcugent.github.io/easybuild | |||
| @@ -0,0 +1,33 @@ | |||
| # This is an easyconfig file for EasyBuild, see http://hpcugent.github.io/easybuild | |||
|
|
||
| homepage = 'http://software.intel.com/en-us/intel-cluster-toolkit-compiler/' | ||
| description = """Intel Cluster Toolkit Compiler Edition provides Intel C/C++ and Fortran compilers, Intel MPI & | ||
| Intel MKL.""" |
There was a problem hiding this comment.
Wouldn't it be better to have a proper toolchain description here?
There was a problem hiding this comment.
Well, that just reads like some Intel promotional material.
Compare it with the foss text.
There was a problem hiding this comment.
Something like this then?
Compiler toolchain including Intel compilers, Intel MPI and Intel Math Kernel Library (MKL).
|
@akesandgren You need to make sure a license is available for these... |
|
Yes i know, i was just a bit too tired this morning. New build is already running. |
|
Test report by @akesandgren |
|
@akesandgren Hmm, this one looks trickier, some missing system dependency for Intel MKL? :-/ |
|
No, it's a feature of Debian/Ubuntu that Intel doesn't handle. I have a patch to the icc easyconfig that fixes this. I'll prep a PR for it. |
|
Ignore that failed test... |
|
Test report by @vanzod |
|
Going in, thanks @boegel! |
(created using
eb --new-pr)requires
easybuilders/easybuild-framework#2215since SHA256 checksums are usedrequires
#4794for GCC 6.4.0