From patchwork Fri Jul 8 15:37:35 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: drm/amdgpu: Disable RPM helpers while reprobing connectors on resume From: cpaul@redhat.com X-Patchwork-Id: 97837 Message-Id: <1467992256-23832-1-git-send-email-cpaul@redhat.com> To: amd-gfx@lists.freedesktop.org Cc: Tom St Denis , Jammy Zhou , open list , stable@vger.kernel.org, "open list:RADEON and AMDGPU DRM DRIVERS" , Alex Deucher , Lyude , Flora Cui , =?UTF-8?q?Christian=20K=C3=B6nig?= , Monk Liu Date: Fri, 8 Jul 2016 11:37:35 -0400 Just about all of amdgpu's connector probing functions try to acquire runtime PM refs. If we try to do this in the context of amdgpu_resume_kms by calling drm_helper_hpd_irq_event(), we end up deadlocking the system. Since we're guaranteed to be holding the spinlock for RPM in amdgpu_resume_kms, and we already know the GPU is in working order, we need to prevent the RPM helpers from trying to run during the initial connector reprobe on resume. There's a couple of solutions I've explored for fixing this, but this one by far seems to be the simplest and most reliable (plus I'm pretty sure that's what disable_depth is there for anyway). Reproduction recipe: - Get any laptop dual GPUs using PRIME - Make sure runtime PM is enabled for amdgpu - Boot the machine - If the machine managed to boot without hanging, switch out of X to another VT. This should definitely cause X to hang infinitely. Cc: stable@vger.kernel.org Signed-off-by: Lyude --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 6e92008..46c1fee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1841,7 +1841,19 @@ int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon) } drm_kms_helper_poll_enable(dev); + + /* + * Most of the connector probing functions try to acquire runtime pm + * refs to ensure that the GPU is powered on when connector polling is + * performed. Since we're calling this from a runtime PM callback, + * trying to acquire rpm refs will cause us to deadlock. + * + * Since we're guaranteed to be holding the rpm lock, it's safe to + * temporarily disable the rpm helpers so this doesn't deadlock us. + */ + dev->dev->power.disable_depth++; drm_helper_hpd_irq_event(dev); + dev->dev->power.disable_depth--; if (fbcon) { amdgpu_fbdev_set_suspend(adev, 0);