From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-yk0-x230.google.com (mail-yk0-x230.google.com [IPv6:2607:f8b0:4002:c07::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 3FEB41A0026 for ; Sat, 6 Feb 2016 06:44:28 +1100 (AEDT) Received: by mail-yk0-x230.google.com with SMTP id r207so64877947ykd.2 for ; Fri, 05 Feb 2016 11:44:28 -0800 (PST) MIME-Version: 1.0 In-Reply-To: References: <969BB53B-CB30-49DC-B6C5-4D2381ECE6EE@gmail.com> Date: Fri, 5 Feb 2016 19:44:25 +0000 Message-ID: Subject: Re: PowerPC agpmode issues From: Mike To: "Herminio Hernandez, Jr." Cc: Pranith Kumar , "open list:LINUX FOR POWERPC..." , "Aneesh Kumar K.V" Content-Type: multipart/alternative; boundary=94eb2c0b163020393b052b0b14cd List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , --94eb2c0b163020393b052b0b14cd Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Thanks, found this now - https://www.mail-archive.com/search?l=3Dmesa-dev@lists.freedesktop.org&q=3D= subject:%22%5C%5BMesa%5C-dev%5C%5D+gallium+r300+driver+for+PowerPC%22&o=3Dn= ewest&f=3D1 On 5 February 2016 at 18:08, Herminio Hernandez, Jr. < herminio.hernandezjr@gmail.com> wrote: > Yeah the issue is with the driOpenDriver function at least that what I > think from what I saw in gdb. > > On Fri, Feb 5, 2016 at 12:06 PM, Mike wrote: > >> True, a permanent patch was needed, but at least we are up and running >> and can identify other issues.. In the thread with the patches it is >> highlighted that the drivers are broken in the first place >> >> On 5 February 2016 at 17:51, Herminio Hernandez, Jr. < >> herminio.hernandezjr@gmail.com> wrote: >> >>> I have already applied his patches on my PowerBook running Jessie to ge= t >>> accelerated graphics. However they are only a work around and not a rea= l >>> fix so they will be committed to mesa. >>> >>> On Fri, Feb 5, 2016 at 8:44 AM, Mike wrote: >>> >>>> Herminio, >>>> >>>> Ok, take a look at this. >>>> https://github.com/threader/mesa-11.1.1-ppc-r300-debian or if on >>>> stable maybe this is more fitting >>>> https://github.com/threader/mesa-10.6.3-ppc-r300-debian - this only >>>> fixes the issue for r300 on powerpc though. >>>> >>>> On 5 February 2016 at 14:32, Herminio Hernandez Jr. < >>>> herminio.hernandezjr@gmail.com> wrote: >>>> >>>>> I have been experiencing the same thing with my iBook and PowerBook. >>>>> >>>>> Sent from my iPhone >>>>> >>>>> On Feb 4, 2016, at 8:47 PM, Mike wrote: >>>>> >>>>> Hi. >>>>> Managed to get the Radeon R300 running on mesa 11.1.1 with an old 201= 3 >>>>> patch from Michel D=C3=A0nzer, next problem is of course enabling agp= mode, >>>>> running with pci-mode with radeon.agpmode=3D-1 works, but is of cours= e slow, >>>>> and seems to load the cpu a lot. >>>>> >>>>> 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/650e1203c11354ba84d6= 9ba445abc0efcfe3890a >>>>> >>>>> http://lxr.free-electrons.com/source/drivers/gpu/drm/nouveau/nouveau_= agp.c?v=3D4.2 >>>>> #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 =3D=3D -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 =3D 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 >>>>> >>>>> _______________________________________________ >>>>> Linuxppc-dev mailing list >>>>> Linuxppc-dev@lists.ozlabs.org >>>>> https://lists.ozlabs.org/listinfo/linuxppc-dev >>>>> >>>>> >>>> >>> >> > --94eb2c0b163020393b052b0b14cd Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
On 5 February 2016 at 18:08, Herminio Hernande= z, Jr. <herminio.hernandezjr@gmail.com> wrote:<= br>
Yeah the issue is with t= he driOpenDriver function at least that what I think from what I saw in gdb= .
<= br>
On Fri, Feb 5, 2016 at 12:06 PM, Mike <michael.heltne@gmail.com> wrote:
True, a permanent patch was needed, but at least w= e are up and running and can identify other issues.. In the thread with the= patches it is highlighted that the drivers are broken in the first place

On = 5 February 2016 at 17:51, Herminio Hernandez, Jr. <herminio.h= ernandezjr@gmail.com> wrote:
I have already applied his patches on my PowerBook runni= ng Jessie to get accelerated graphics. However they are only a work around = and not a real fix so they will be committed to mesa.

On Fri, Feb 5, 2016 at = 8:44 AM, Mike <michael.heltne@gmail.com> wrote:
Herminio,=C2=A0=

On 5 Fe= bruary 2016 at 14:32, Herminio Hernandez Jr. <herminio.herna= ndezjr@gmail.com> wrote:
I have been experiencing the same thing with my iBook = and PowerBook.=C2=A0

Sent from my iPhone

On = Feb 4, 2016, at 8:47 PM, Mike <michael.heltne@gmail.com> wrote:

<= blockquote type=3D"cite">
<= div class=3D"gmail_extra">Hi.=C2=A0
Managed= to get the Radeon R300 running on mesa 11.1.1 with an old 2013 patch from = Michel D=C3=A0nzer, next problem is of course enabling agpmode, running wit= h pci-mode with radeon.agpmode=3D-1 works, but is of course slow, and seems= to load the cpu a lot.

Upon initial investigation i could not initially believe = agp could be this this broken for this long, until i found this.
=C2=A0"committed with= Ben Skeggs on Feb 26, 2013"
https://github.com/DespairFactor/bull= head/commit/650e1203c11354ba84d69ba445abc0efcfe3890a
=
#ifdef __powerpc__
/* Disable AGP by default = on all PowerPC machines for
* now -- At least some UniNorth-2 AGP bridg= es 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 m= emory untranslated bypassing
* the GATT somehow, making them quite pain= ful to deal
* with...
*/
if (nouveau_agpmode =3D=3D -1)
retu= rn false;
#endif
=C2=A0
=C2=A0and now later this:=C2= =A0
https://github.com/torvalds/linux/blob/master/drivers/gpu/drm/nouve= au/nvkm/subdev/pci/agp.c
#ifdef __power= pc__
= /* Disable AGP by default on all PowerPC machines for now -- At
*= least some UniNorth-2 AGP bridges are known to be broken:
* DMA from t= he host to the card works just fine, but writeback
* from the card to th= e host goes straight to memory
* untranslated bypassing that GATT someh= ow, making them quite
* painful to deal with...
*/
mode =3D 0;
#endif

All seems to point to serious issues had aroun= d the time of change to ums to kms and a serious regression hitting the lin= ux kernel? No?

Cheers
-Mike
_______= ________________________________________
Linuxppc-dev maili= ng list
Linuxppc-dev@lists.ozlabs.org
https= ://lists.ozlabs.org/listinfo/linuxppc-dev





--94eb2c0b163020393b052b0b14cd--