| 1 |
Fernando Rodriguez <frodriguez.developer <at> outlook.com> writes: |
| 2 |
|
| 3 |
> Do you know of any plans to enable offloading on the gentoo toolchain? |
| 4 |
|
| 5 |
NO, but I'm sure some devs are keenly aware of this feature in gcc-5. |
| 6 |
|
| 7 |
|
| 8 |
> I was able to build the offloading compiler using crossdev with a few hacks |
| 9 |
> and wrote an ebuild for Intel's simulator[2]. I will work on enabling the |
| 10 |
> host compiler tomorrow and may open a feature request and post patches |
| 11 |
> once I get it working. The changes needed to enable it on the host are |
| 12 |
> pretty trivial. |
| 13 |
|
| 14 |
Sorry Fernando, I just now saw this thread on an old thread. I think that |
| 15 |
'sys-cluster/ceph' is where I'd like to test your spin on the gcc-5. |
| 16 |
Ceph has RDMA (RoCE) in the 0.94 branch (in portage). You are definately |
| 17 |
ahead of me on practical gcc-5 experiments with offloading and other |
| 18 |
new features. |
| 19 |
|
| 20 |
You did not list your second reference. Where I can I get/git your |
| 21 |
compiler and some brief suggestions on taking it for a test drive. |
| 22 |
|
| 23 |
I'm not much interested in the Intel simulator, atm. I like to test on |
| 24 |
old gear running gentoo:: borking is no big deal, if it happens. |
| 25 |
Other codes keen to test gcc-5 (offloading) on are Apache-mesos and |
| 26 |
Apache-spark and mesos-distcc. |
| 27 |
|
| 28 |
Also, per this doc [1] you can get your own gentoo overlay to put |
| 29 |
things up for wider experimentation, if you like. |
| 30 |
|
| 31 |
[1] https://wiki.gentoo.org/wiki/Project:Overlays/Dev_Guide |
| 32 |
|
| 33 |
|
| 34 |
Very cool, what you have done, |
| 35 |
James |
Archives