<html> <head> <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type"> </head> <body> <div class="moz-cite-prefix">On 05/03/2017 08:52 AM, Mika Kuoppala wrote: </div> <blockquote cite="mid:87shkmb9uw.fsf@gaia.fi.intel.com" type="cite"> <pre wrap="">Oscar Mateo <a class="moz-txt-link-rfc2396E" href="mailto:oscar.mateo@intel.com"><oscar.mateo@intel.com></a> writes: </pre> <blockquote type="cite"> <pre wrap="">On 05/02/2017 09:17 AM, Mika Kuoppala wrote: </pre> <blockquote type="cite"> <pre wrap="">Chris Wilson <a class="moz-txt-link-rfc2396E" href="mailto:chris@chris-wilson.co.uk"><chris@chris-wilson.co.uk></a> writes: </pre> <blockquote type="cite"> <pre wrap="">On Fri, Apr 28, 2017 at 09:11:06AM +0000, Oscar Mateo wrote: </pre> <blockquote type="cite"> <pre wrap="">The new batchbuffer for CNL surpasses the 4096 byte mark. Cc: Mika Kuoppala <a class="moz-txt-link-rfc2396E" href="mailto:mika.kuoppala@intel.com"><mika.kuoppala@intel.com></a> Cc: Ben Widawsky <a class="moz-txt-link-rfc2396E" href="mailto:ben@bwidawsk.net"><ben@bwidawsk.net></a> Signed-off-by: Oscar Mateo <a class="moz-txt-link-rfc2396E" href="mailto:oscar.mateo@intel.com"><oscar.mateo@intel.com></a> </pre> </blockquote> <pre wrap="">Evil, 4k+ of nothing-ness that userspace then has to configure for itself for correctness anyway. Patch looks ok, but still question the sanity. </pre> </blockquote> <pre wrap="">Is there a requirement for CNL to init the renderstate? I would like to drop the render state init from CNL if we can't find evidence that it needs it. Bspec indicates that it doesnt. -Mika </pre> </blockquote> <pre wrap=""> Hi Mika, I can double-check with the hardware architects, but word around here is that render state init has never stopped being a requirement. Where did you see in the BSpec that it is not required for CNL? </pre> </blockquote> <pre wrap=""> It would be great if you could refresh the answer and perhaps even get some answers to the 'why' parts. In the "Context Descriptor Format" section, it says: "Render CS Only: Render state need not be initialized; the Render Context Restore Inhibit bit in the Context/Save image in memory should be set to prevent restoring garbage render context." -Mika </pre> </blockquote> :_( The same section also says:<o:p></o:p> <o:p> </o:p> “See the Logical Ring Context Format section for details.”<o:p></o:p> <o:p> </o:p> And then “Logical Ring Context Format” section goes on to say:<o:p></o:p> <o:p> </o:p> “It is tedious for software to populate the engine context as per the requirements, it is recommended to implicitly use engine to populate this portion of the context. […] Software must program all the state required to initialize the engine in the ring buffer which would initialize the hardware state.”<o:p></o:p> <o:p> </o:p> I’ll try to clarify it... </body> </html>