<html><body>
<p><tt><font size="2"><br>
> > Subject:<br>
> > <br>
> > Re: Fedora 22<br>
> > <br>
> > Sent by:<br>
> > <br>
> > ppc-bounces@lists.fedoraproject.org<br>
> > <br>
> > >> Hi<br>
> > >> We briefly discussed priorities for Fedora 22 and I had taken an<br>
> > >> action item to start an email conversation about this. So here is<br>
> > >> what I would like to see for Fedora 22.<br>
> > >><br>
> > >> 1) Get the -mcpu and -mtune flags set properly for the LE build.<br>
> > >> Should be -mcpu=power7 -mtune=power8<br>
> > ><br>
> > > done, all packages that honour the Fedora system wide compiler flags<br>
> > > use them, if they don't it's a packaging bug<br>
> > ><br>
> > >> 2) Have a cloud image available<br>
> > >> 3) For BE I would like another subarch. Same packages as the current<br>
> > >> one but tuned for P8.<br>
> > ><br>
> > > you mean in addition to ppc64p7? can't we just switch ppc64p7 from<br>
> > > -mcpu=power7 -mtune=power7 to -mcpu=power7 -mtune=power8?<br>
> > <br>
> > This makes sense to me as it then mirrors what we have in ppc64le and<br>
> > it saves having more targets.<br>
> <br>
> The disadvantage of this would be to cut off users that have Power 7 systems<br>
> and optimized code. So why would people want to optimize from Fedora 21 to <br>
> Fedora 22. You would be taking a big step back in performance. I don't <br>
> want to suggest keeping a subarch for each type of POWER system out there. I <br>
> was thinking of keeping two. So when the next POWER arch that comes out,<br>
> the Power 7 subarch goes away and you would have Power 8 and the new<br>
> Power arch.<br>
> <br>
> > <br>
> > >> 4) Looking into if we can get Docker enabled for POWER. I will send<br>
> > >> out further emails about this as<br>
> > >> soon as I know more about the current development state<br>
> > >> 5) Start the ground work for a workstation release. But I think we<br>
> > >> should target the release for Fedora-23.<br>
> > ><br>
> > > yeah, the deadlines in F-22 won't give us enough time to work on it<br>
> > <br>
> > Agreed, ultimately all the packages are built and as a result it's<br>
> > possible to install and test all the main components of Workstation<br>
> > for those that actually have desktop capable hardware to do any<br>
> > development/testing/debug that is required to ensure all the<br>
> > underlying workstation components and dependencies are in a good state<br>
> > for Fedora 23.<br>
> > <br>
> > It's likely there will be work needed on the X stack components like<br>
> > mesa etc to ensure all is working.<br>
> > <br>
> > >> I seem to remember hearing that there is a quite a lot of work<br>
> > >> involved to get this build going.<br>
> > ><br>
> > > we should have almost everything built (eg. LibreOffice is known to fail<br>
> > > to build), but it's not tested and if broken, it will need to be fixed<br>
> > ><br>
> > > One thing to not forgot is that also the mainline kernel needs to have<br>
> > > the necessary support for the workstation class hw.<br>
> > <br>
> > I think the biggest thing will be the graphics stack. While all the<br>
> > workstation bits are built I doubt they've had wide and varied testing<br>
> > to ensure they're al sufficiently optimised and robust for a good<br>
> > Workstation experience.<br>
</font></tt><br>
<tt><font size="2">Adding 5) update gcc to version 5. Will that be possible? It is my understanding</font></tt><br>
<tt><font size="2"> that the gccgo changes needed to support Docker are in that release</font></tt></body></html>