Quoting Chris Wilson (2018-06-18 11:10:23) > We should we have all the kinks worked out and full-ppgtt now works > reliably on gen7 (Ivybridge, Valleyview/Baytrail and Haswell). If we can > let userspace have full control over their own ppgtt, it makes softpinning > far more effective, in turn making GPU dispatch far more efficient and > more secure (due to better mm segregation). On the other hand, switching > over to a different GTT for every client does incur noticeable overhead. > > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > Cc: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx> > Cc: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx> > Cc: Matthew Auld <matthew.william.auld@xxxxxxxxx> > Reviewed-by: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx> > Cc: Jason Ekstrand <jason.ekstrand@xxxxxxxxx> > Cc: Kenneth Graunke <kenneth@xxxxxxxxxxxxx> > --- > > This has been run through piglit for ivb/vlv/hsw locally and hsw on > kernel's CI, but we would like at least one ack from Mesa as well. If > it's possible to run it through the full gamut of your testing, that > would be great. The patches applies to drm-tip, and requires fixes in drm-tip for at least one GPU hang full-ppgtt caused in piglit. -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
