Quoting Joonas Lahtinen (2018-06-01 11:21:43) > On Thu, 2018-05-31 at 12:35 +0100, Chris Wilson wrote: > > On gen8 and onwards, we can mark GPU accesses through the ppGTT as being > > read-only, that is cause any GPU write onto that page to be discarded > > (not triggering a fault). This is all that we need to finally support > > the read-only flag for userptr! > > > > Testcase: igt/gem_userptr_blits/readonly* > > Signed-off-by: Chris Wilson <ch...@chris-wilson.co.uk> > > Cc: Jon Bloomfield <jon.bloomfi...@intel.com> > > Cc: Joonas Lahtinen <joonas.lahti...@linux.intel.com> > > Please add clarification to the commit message that this expands uAPI > (by supporting RO userptrs), so Rodrigo will notice to highlight for > -next PR.
Ha, I was playing the "this was already in the uAPI" card, just never implemented so hadn't made it in the uABI. :) Being able to import read-only shm segments (or PROT_READ file mmaps) was always on my wishlist, just at the time the only prospect was special case supporting it for byt (not very appealing). PR material it is then. -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
