Re: [Mesa-dev] [PATCH 3/5] intel/fs: Cap dst-aligned region stride to maximum representable hstride value.

2019-02-21 Thread Jason Ekstrand
On Thu, Feb 14, 2019 at 4:53 PM Francisco Jerez wrote: > Jason Ekstrand writes: > > > On Fri, Jan 18, 2019 at 6:09 PM Francisco Jerez > > wrote: > > > >> This is required in combination with the following commit, because > >> otherwise if a source region with an extended 8+ stride is present

Re: [Mesa-dev] [PATCH 3/5] intel/fs: Cap dst-aligned region stride to maximum representable hstride value.

2019-02-15 Thread Francisco Jerez
Jason Ekstrand writes: > On Fri, Jan 18, 2019 at 6:09 PM Francisco Jerez > wrote: > >> This is required in combination with the following commit, because >> otherwise if a source region with an extended 8+ stride is present in >> the instruction (which we're about to declare legal) we'll end up

Re: [Mesa-dev] [PATCH 3/5] intel/fs: Cap dst-aligned region stride to maximum representable hstride value.

2019-02-14 Thread Jason Ekstrand
On Fri, Jan 18, 2019 at 6:09 PM Francisco Jerez wrote: > This is required in combination with the following commit, because > otherwise if a source region with an extended 8+ stride is present in > the instruction (which we're about to declare legal) we'll end up > emitting code that attempts to

[Mesa-dev] [PATCH 3/5] intel/fs: Cap dst-aligned region stride to maximum representable hstride value.

2019-01-18 Thread Francisco Jerez
This is required in combination with the following commit, because otherwise if a source region with an extended 8+ stride is present in the instruction (which we're about to declare legal) we'll end up emitting code that attempts to write to such a region, even though strides greater than four