Re: [Qemu-devel] [PATCH 1/6] softfloat: remove HPPA specific code

[Andreas Färber]

Am 05.01.2011 um 11:21 schrieb Aurelien Jarno:

> On Wed, Jan 05, 2011 at 09:15:43AM +0100, Andreas Färber wrote:
> > Am 05.01.2011 um 00:56 schrieb Aurelien Jarno:
> >
> > > HPPA code has still to be ported
> > > to TCG.
> > > Yes it still uses dyngen.
> >
> > Oh. I was pretty sure I saw TCG host support patches...
>
> Yes, there is TCG *host* code in upstream, and this code has been  
> updated
> recently.
>
> But that's not the subject we are talking about HPPA guest support.

[Thanks, I do know the difference. Obviously, by inference I assumed  
that the target would then use TCG as well. Repo doesn't make it  
obvious what the changes to qemu.git are, since it was merged, not  
rebased.]

> We have forks still using dyngen. Rebasing them against HEAD will  
> break
> them (including this HPPA target). Still you didn't complain about
> dyngen removal.

For the record, I did. Replacing dyngen with TCG was a good step, but  
it broke Darwin/ppc host for MONTHS.

I also did for kqemu. Its removal effectively means NO acceleration on  
(Open)Solaris and Haiku.

And I'm curious how Windows support will work out. We have seen some  
patches lately (one yet to be reviewed comes to mind) but no official  
maintainer yet, which was demanded by Anthony to not drop Win32. On  
the other hand, AdaCore and CodeSourcery have spoken up as commercial  
Win32 users.

Anyway, for target-ppc, me and others helped in a concerted effort to  
get things converted to TCG and reviewed. Might work for other  
projects, too.

> And this is only an example. QEMU evolves quite fast
> recently, if you don't rebase regularly, your code becomes  
> incompatible,
> and useless quite soon.

Which is why I've volunteered to step up as PReP maintainer. Can't do  
that for all projects, but we should encourage people to push their  
favorite features into upstream before things diverge more making it  
even more complicated.

This was sort-of evoked by the "1st QEMU Users Forum" thread, which  
expressed a sentiment that the qemu-devel community had shifted  
towards KVM/virtualization and were neglecting the emulation part of  
QEMU.

> I'll commit it in 6 months if
> nothing has changed with regards to HPPA.

Just pointing out that there were 5 months without activity from  
yourself last year. It happens, Real Life can be gripping.

> That will make 2 years without
> activity.

My haiku-user project is still in its infancy for 1 year now, with the  
ELF runtime_loader not yet fully ported, since I dedicated more time  
to getting ppc system emulation working. While I try to emulate ppc,  
the Haiku/ppc kernel gets neglected, etc. Two years go by really  
quickly when you're busy!

Cheers,

Andreas