On 18/12/2019 09:35, Chris Wilson wrote:
For very light workloads that frequently park, acquiring the display power well (required to prevent the dmc from trashing the system) takes longer than the execution. A good example is the igt_coherency selftest, which is slowed down by an order of magnitude in the worst case with powerwell cycling. To prevent frequent cycling, while keeping our fast soft-rc6, use a timer to delay release of the display powerwell. Fixes: 311770173fac ("drm/i915/gt: Schedule request retirement when timeline idles") References: https://gitlab.freedesktop.org/drm/intel/issues/848 Signed-off-by: Chris Wilson <[email protected]> Cc: Tvrtko Ursulin <[email protected]> --- drivers/gpu/drm/i915/gt/intel_gt_pm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c index bb57e3443a50..f36ce36dabeb 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c @@ -88,8 +88,9 @@ static int __gt_park(struct intel_wakeref *wf) /* Everything switched off, flush any residual interrupt just in case */ intel_synchronize_irq(i915);+ /* Defer dropping the display power well for 100ms, it's slow! */GEM_BUG_ON(!wakeref); - intel_display_power_put(i915, POWER_DOMAIN_GT_IRQ, wakeref); + intel_display_power_put_async(i915, POWER_DOMAIN_GT_IRQ, wakeref);i915_globals_park();
Checked that intel_display_power_get cancels the delayed put since I wasn't familiar with the API and it looks good.
Reviewed-by: Tvrtko Ursulin <[email protected]> Regards, Tvrtko _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
