On Fri, Aug 30, 2019 at 01:50:53PM +0300, Joonas Lahtinen wrote: > To ensure cross-driver locking compatibility, document the expected > guidelines for implementing the GEM locking in i915. Note that this > is a description of how things should end up after being reworked, > and does not reflect the current state of things. > > Signed-off-by: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx> > Signed-off-by: Daniel Vetter <daniel.vetter@xxxxxxxxx> > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > Cc: Dave Airlie <airlied@xxxxxxxxxx> > Cc: Matthew Auld <matthew.auld@xxxxxxxxx> > Cc: Abdiel Janulgue <abdiel.janulgue@xxxxxxxxxxxxxxx> > Cc: CQ Tang <cq.tang@xxxxxxxxx> > --- > Documentation/gpu/i915.rst | 45 ++++++++++++++++++++++++++++++++++++++ > 1 file changed, 45 insertions(+) > > diff --git a/Documentation/gpu/i915.rst b/Documentation/gpu/i915.rst > index e249ea7b0ec7..63a72d10f2c7 100644 > --- a/Documentation/gpu/i915.rst > +++ b/Documentation/gpu/i915.rst > @@ -320,6 +320,51 @@ for execution also include a list of all locations within buffers that > refer to GPU-addresses so that the kernel can edit the buffer correctly. > This process is dubbed relocation. > > +Locking Guidelines > +------------------ > + > +**NOTE:** This is a description of how the locking should be after > +refactoring is done. Does not necessarily reflect what the locking > +looks like while WIP. Maybe use rst note block for this? .. note:: > + > +#. All locking rules and interface contracts with cross-driver interfaces > + (dma-buf, dma_fence) need to be followed. > + > +#. No struct_mutex anywhere in the code > + > +#. dma_resv will be the outermost lock (when needed) and ww_acquire_ctx > + is to be hoisted at highest level and passed down within i915_gem_ctx > + in the call chain > + > +#. While holding lru/memory manager (buddy, drm_mm, whatever) locks > + system memory allocations are not allowed > + > + * Enforce this by priming lockdep (with fs_reclaim). If we > + allocate memory while holding these looks we get a rehash > + of the shrinker vs. struct_mutex saga, and that would be > + real bad. > + > +#. Do not nest different lru/memory manager locks within each other. > + Take them in turn to update memory allocations, relying on the object’s > + dma_resv ww_mutex to serialize against other operations. > + > +#. The suggestion for lru/memory managers locks is that they are small > + enough to be spinlocks. > + > +#. All features need to come with exhaustive kernel selftests and/or > + IGT tests when appropriate > + > +#. All LMEM uAPI paths need to be fully restartable (_interruptible() > + for all locks/waits/sleeps) > + > + * Error handling validation through signal injection. > + Still the best strategy we have for validating GEM uAPI > + corner cases. > + Must be excessively used in the IGT, and we need to check > + that we really have full path coverage of all error cases. > + > + * -EDEADLK handling with ww_mutex > + It seems clear and clean to me. So from my point of view: Reviewed-by: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx> > GEM BO Management Implementation Details > ---------------------------------------- > > -- > 2.20.1 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@xxxxxxxxxxxxxxxxxxxxx > https://lists.freedesktop.org/mailman/listinfo/intel-gfx _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
