<div dir="ltr">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</div><div class="gmail_extra"><br><div class="gmail_quote">On 5 February 2016 at 17:51, Herminio Hernandez, Jr. <span dir="ltr"><<a href="mailto:herminio.hernandezjr@gmail.com" target="_blank">herminio.hernandezjr@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">I have already applied his patches on my PowerBook running Jessie to get accelerated graphics. However they are only a work around and not a real fix so they will be committed to mesa.</div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Feb 5, 2016 at 8:44 AM, Mike <span dir="ltr"><<a href="mailto:michael.heltne@gmail.com" target="_blank">michael.heltne@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Herminio, <div><br></div><div>Ok, take a look at this. <a href="https://github.com/threader/mesa-11.1.1-ppc-r300-debian" target="_blank">https://github.com/threader/mesa-11.1.1-ppc-r300-debian</a> or if on stable maybe this is more fitting <a href="https://github.com/threader/mesa-10.6.3-ppc-r300-debian" target="_blank">https://github.com/threader/mesa-10.6.3-ppc-r300-debian</a> - this only fixes the issue for r300 on powerpc though. </div></div><div><div><div class="gmail_extra"><br><div class="gmail_quote">On 5 February 2016 at 14:32, Herminio Hernandez Jr. <span dir="ltr"><<a href="mailto:herminio.hernandezjr@gmail.com" target="_blank">herminio.hernandezjr@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><div>I have been experiencing the same thing with my iBook and PowerBook. <br><br>Sent from my iPhone</div><div><div><div><br>On Feb 4, 2016, at 8:47 PM, Mike <<a href="mailto:michael.heltne@gmail.com" target="_blank">michael.heltne@gmail.com</a>> wrote:<br><br></div><blockquote type="cite"><div><div dir="ltr"><div class="gmail_extra"><div class="gmail_extra">Hi. </div><div class="gmail_extra">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.</div><div class="gmail_extra"><br></div><div class="gmail_extra">Upon initial investigation i could not initially believe agp could be this this broken for this long, until i found this.</div><div class="gmail_extra"><div class="gmail_extra"> "committed with Ben Skeggs on Feb 26, 2013"</div><div class="gmail_extra"><a href="https://github.com/DespairFactor/bullhead/commit/650e1203c11354ba84d69ba445abc0efcfe3890a" target="_blank">https://github.com/DespairFactor/bullhead/commit/650e1203c11354ba84d69ba445abc0efcfe3890a</a></div></div><div class="gmail_extra"><a href="http://lxr.free-electrons.com/source/drivers/gpu/drm/nouveau/nouveau_agp.c?v=4.2" target="_blank">http://lxr.free-electrons.com/source/drivers/gpu/drm/nouveau/nouveau_agp.c?v=4.2</a></div><div class="gmail_extra">#ifdef __powerpc__</div><div class="gmail_extra"><span style="white-space:pre-wrap">                </span>/* Disable AGP by default on all PowerPC machines for</div><div class="gmail_extra"><span style="white-space:pre-wrap">              </span> * now -- At least some UniNorth-2 AGP bridges are</div><div class="gmail_extra"><span style="white-space:pre-wrap">         </span> * known to be broken: DMA from the host to the card</div><div class="gmail_extra"><span style="white-space:pre-wrap">               </span> * works just fine, but writeback from the card to the</div><div class="gmail_extra"><span style="white-space:pre-wrap">             </span> * host goes straight to memory untranslated bypassing</div><div class="gmail_extra"><span style="white-space:pre-wrap">             </span> * the GATT somehow, making them quite painful to deal</div><div class="gmail_extra"><span style="white-space:pre-wrap">             </span> * with...</div><div class="gmail_extra"><span style="white-space:pre-wrap">         </span> */</div><div class="gmail_extra"><span style="white-space:pre-wrap">                </span>if (nouveau_agpmode == -1)</div><div class="gmail_extra"><span style="white-space:pre-wrap">                 </span>return false;</div><div class="gmail_extra">#endif</div><div class="gmail_extra"> </div><div class="gmail_extra"> and now later this: </div><div class="gmail_extra"><a href="https://github.com/torvalds/linux/blob/master/drivers/gpu/drm/nouveau/nvkm/subdev/pci/agp.c" target="_blank">https://github.com/torvalds/linux/blob/master/drivers/gpu/drm/nouveau/nvkm/subdev/pci/agp.c</a></div><div class="gmail_extra">#ifdef __powerpc__</div><div class="gmail_extra"><span style="white-space:pre-wrap">       </span>/* Disable AGP by default on all PowerPC machines for now -- At</div><div class="gmail_extra"><span style="white-space:pre-wrap">    </span> * least some UniNorth-2 AGP bridges are known to be broken:</div><div class="gmail_extra"><span style="white-space:pre-wrap">       </span> * DMA from the host to the card works just fine, but writeback</div><div class="gmail_extra"><span style="white-space:pre-wrap">    </span> * from the card to the host goes straight to memory</div><div class="gmail_extra"><span style="white-space:pre-wrap">       </span> * untranslated bypassing that GATT somehow, making them quite</div><div class="gmail_extra"><span style="white-space:pre-wrap">     </span> * painful to deal with...</div><div class="gmail_extra"><span style="white-space:pre-wrap"> </span> */</div><div class="gmail_extra"><span style="white-space:pre-wrap">        </span>mode = 0;</div><div class="gmail_extra">#endif</div><div class="gmail_extra"><br></div><div class="gmail_extra">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?</div><div class="gmail_extra"><br></div><div class="gmail_extra">Cheers</div><div class="gmail_extra">-Mike</div></div></div>
</div></blockquote></div></div><blockquote type="cite"><div><span>_______________________________________________</span><br><span>Linuxppc-dev mailing list</span><br><span><a href="mailto:Linuxppc-dev@lists.ozlabs.org" target="_blank">Linuxppc-dev@lists.ozlabs.org</a></span><br><span><a href="https://lists.ozlabs.org/listinfo/linuxppc-dev" target="_blank">https://lists.ozlabs.org/listinfo/linuxppc-dev</a></span></div></blockquote></div></blockquote></div><br></div>
</div></div></blockquote></div><br></div>
</div></div></blockquote></div><br></div>