On Tue, Feb 04, 2014 at 11:40:20AM +0000, Chris Wilson wrote: > On Tue, Feb 04, 2014 at 12:31:37PM +0100, Daniel Vetter wrote: > > On Fri, Jan 31, 2014 at 03:42:48PM -0600, jeff.mcgee@xxxxxxxxx wrote: > > > From: Jeff McGee <jeff.mcgee@xxxxxxxxx> > > > > > > RPS manual mode disables/ignores load-based inputs and allows render > > > performance state to be controlled externally. The enabling of manual > > > mode and setting of desired frequency is done through debugfs. > > > > > > i915_rps_manual: > > > '0' - RPS controlled normally using load metrics. > > > '1' - RPS controlled manually via i915_cur_freq writes. > > > > > > i915_cur_freq: > > > u64 - Value is the current gpu frequency request in MHz. Writes > > > accepted only if i915_rps_manual = 1. > > > > > > Supports Gen6+ except Valleyview and Broadwell. > > > > > > Signed-off-by: Jeff McGee <jeff.mcgee@xxxxxxxxx> > > > > Hm, can't we fake this by setting min/max to the same values? > > This is one where we can reconfigure the hardware if explicitly set to > disabled. Or maybe we should just put the smarts in to detect min == max > and disable the interrupt generation (as opposed to just masking at the > various points along the path). > -Chris Yes, it is possible to achieve this functionality by coercing the frequency with min and max. That seemed less clean to me and adds extra steps to test code. Also the hardware will still be processing RPS logic unless we add the min==max smarts as Chris mentions. -Jeff _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
