<html> <head> <base href="https://bugs.freedesktop.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - [SKL/KBL] Init time is high" href="https://bugs.freedesktop.org/show_bug.cgi?id=94722#c21">Comment # 21</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - [SKL/KBL] Init time is high" href="https://bugs.freedesktop.org/show_bug.cgi?id=94722">bug 94722</a> from <a class="email" href="mailto:david.weinehall@intel.com" title="David Weinehall <david.weinehall@intel.com>"> David Weinehall</a> <pre>(In reply to Chris Wilson from <a href="show_bug.cgi?id=94722#c20">comment #20</a>) > (In reply to David Weinehall from <a href="show_bug.cgi?id=94722#c19">comment #19</a>) > > Measurements using SuspendResume on a KBL-Y yields roughly these numbers: > > > > suspend: 320ms -- out of this 300ms is spent turning off panel + backlight; > > the delays here cannot be optimised away, so at most we can shave off a few > > milliseconds. This values will vary greatly depending on the panel used. > > > > resume: 470ms -- out of this 30ms is spent in restore_gtt, 230ms is spent > > turning on panel + backlight, 170ms on probing for new displays, and 30ms in > > one of the opregion-related calls. > > Could you test <a href="https://patchwork.freedesktop.org/patch/82422/">https://patchwork.freedesktop.org/patch/82422/</a> that should > help the restore_gtt slightly. Not just slightly. On my Broadwell it cuts restore_gtt from 15ms to less than 5ms. The Kabylake-Y I've been testing on is busy today -- will run similar tests on it tomorrow to see whether the savings are similar there.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> <li>You are on the CC list for the bug.</li> </ul> </body> </html>