[Intel-gfx] [PATCH] drm/i915: Extend Haswell GT1 PSMI workaround to all

Chris Wilson chris at chris-wilson.co.uk
Tue Sep 17 20:26:21 UTC 2019


Quoting Chris Wilson (2019-09-17 21:23:01)
> Quoting Chris Wilson (2019-09-17 20:47:46)
> > A few times in CI, we have detected a GPU hang on our Haswell GT2
> > systems with the characteristic IPEHR of 0x780c0000. When the PSMI w/a
> > was first introducted, it was applied to all Haswell, but later on we
> > found an erratum that supposedly restricted the issue to GT1 and so
> > constrained it only be applied on GT1. That may have been a mistake...
> 
> Something else to bear in mind about why this is showing up now, is that
> the enabling of iommu on these machines. It's the last instruction in
> the context image... Could we need to expand the context?

Fwiw, we say the maximum size for the haswell context is 70270, which
even expanding for prefetch is well inside the next page boundary of
73728. Furthermore, no DMAR faults. The coincidence may be a timing
artifact of the iommu indirection? Or just a mere coincidence.
-Chris


More information about the Intel-gfx mailing list