On Wed, Nov 20, 2013 at 7:00 AM, S, Deepak <deepak.s@xxxxxxxxx> wrote: >>- have you done measurements on this? given how infrequently we > ought to be waking the wells when they're idle, and how long we > generally keep them awake, is this a real power win? > [Deepak] By Individually controlling the wells we observed around 100mW - 200mW saving in different scenarios (GL Beanchmark & Media playback). This kind of information is really important and should be part of the commit message. This rule holds generally for performance/power tuning work - the commit message should at least mention the order of magnitude of the improvements seen and which workloads have been tested. If you're afraid to disclose confidential information (e.g. for power savings) just make the language fuzzy enough, e.g. here "We've seen power savings in the lower sub-1W range on workloads that only neeed on of the power wells, e.g. glbenchmark, media playback." Thanks, Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
