PowerPC agpmode issues

[Mike]

Put the info forth in an Amiga ppc related thread, agpmode=-1 is also
needed on the Pegasos 2, need do some research/googling but is it safe to
say its broken for all non x86 platforms?

Including those replies here:
----
Hi Christian and Mike,
for my personal experiences on Pegasos2 and MacMini G4 the performance was
 good on 3D and 2D but need the agpmode=-1 or the system freeze
adding the agpmode=-1 make loose performances but make the system working
this because the gpu without that append was over working it was look like
software overclocked.
 About agp macs a guy on ubuntuforums boost his performance adding this to
the kernel append  "quiet splash radeon.modeset=0
video=radeonfb:1280x960-32 at 60 "

About Pegasos2 i had it with 9800pro and the machine was running great.
Alex i know is a geek with this machine that i dont have any moore or
better i have another one died :-
---

Hi Luigi.

Thanks. Remember the modeset=0 flag my self from a while back, had to be
appended to either the kernel or as an option the the module. What a sorry
state.. I think agpmode=-1 masks a bunch of issues, degrading performance
due to some hittage of the cpu unnecessary, thats my suspicion anyway.
likely if ever a fix is found it will reveal a bunch of other issues is my
fear.
----



On 5 February 2016 at 07:49, Denis Kirjanov <kda at linux-powerpc.org> wrote:

> On 2/5/16, Mike <michael.heltne at gmail.com> wrote:
> > Hi.
> > Managed to get the Radeon R300 running on mesa 11.1.1 with an old 2013
> > patch from Michel Dànzer, next problem is of course enabling agpmode,
> > running with pci-mode with radeon.agpmode=-1 works, but is of course
> slow,
> > and seems to load the cpu a lot.
>
> For me nouveau on quad pmac is also broken for years(?) :/
>
> >
> > Upon initial investigation i could not initially believe agp could be
> this
> > this broken for this long, until i found this.
> >  "committed with Ben Skeggs on Feb 26, 2013"
> >
> https://github.com/DespairFactor/bullhead/commit/650e1203c11354ba84d69ba445abc0efcfe3890a
> >
> http://lxr.free-electrons.com/source/drivers/gpu/drm/nouveau/nouveau_agp.c?v=4.2
> > #ifdef __powerpc__
> > /* Disable AGP by default on all PowerPC machines for
> > * now -- At least some UniNorth-2 AGP bridges are
> > * known to be broken: DMA from the host to the card
> > * works just fine, but writeback from the card to the
> > * host goes straight to memory untranslated bypassing
> > * the GATT somehow, making them quite painful to deal
> > * with...
> > */
> > if (nouveau_agpmode == -1)
> > return false;
> > #endif
> >
> >  and now later this:
> >
> https://github.com/torvalds/linux/blob/master/drivers/gpu/drm/nouveau/nvkm/subdev/pci/agp.c
> > #ifdef __powerpc__
> > /* Disable AGP by default on all PowerPC machines for now -- At
> > * least some UniNorth-2 AGP bridges are known to be broken:
> > * DMA from the host to the card works just fine, but writeback
> > * from the card to the host goes straight to memory
> > * untranslated bypassing that GATT somehow, making them quite
> > * painful to deal with...
> > */
> > mode = 0;
> > #endif
> >
> > All seems to point to serious issues had around the time of change to ums
> > to kms and a serious regression hitting the linux kernel? No?
> >
> > Cheers
> > -Mike
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ozlabs.org/pipermail/linuxppc-dev/attachments/20160205/5827238f/attachment-0001.html>