Quoting Chris Wilson (2020-04-21 17:41:30) > Normally when we create a new context, and a new ppGTT to go with it, we > point all the unused pages in the ppGTT to a 'safe' scratch page. Any > inadvertent access outside of the declared user's area will result in a > read/write to scratch instead. However, sometimes it is preferrable to > that to cause a fault instead. This does not trap execution of the > faulting batch, but it does record the error: > > FAULT_TLB_DATA: 0x00000000 0x00000004 > Address 0x0000000000004000 PPGTT > > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > Cc: Jason Ekstrand <jason@xxxxxxxxxxxxxx> > --- > The name and value semantics are horrendous. The non-trapping behaviour > is also less than ideal. Worth it? Note that we can ask for an interrupt on a page access error, however it is a 'validation' error and we get an interrupt for every single access and NOOP fixup (the single bit for all classes of validation errors). They were quite frequent. -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
