similar to: [PATCH v2] drm: don't continue with anything after the GPU couldn't be woken up

On Tue, Nov 21, 2017 at 6:46 PM, Thierry Reding <thierry.reding at gmail.com> wrote: > On Tue, Nov 21, 2017 at 04:01:16PM +0100, Karol Herbst wrote: >> This should make systems more stable where resuming the GPU fails. This >> can happen due to bad firmware or due to a bug within the kernel. The >> last thing which should happen in either case is an unusable system.

On Tue, Nov 21, 2017 at 04:01:16PM +0100, Karol Herbst wrote: > This should make systems more stable where resuming the GPU fails. This > can happen due to bad firmware or due to a bug within the kernel. The > last thing which should happen in either case is an unusable system. > > v2: do the same in nouveau_pmops_resume > > Tested-by: Karl Hastings <kazen at

On Tue, Nov 21, 2017 at 08:03:20PM +0100, Karol Herbst wrote: > On Tue, Nov 21, 2017 at 6:46 PM, Thierry Reding > <thierry.reding at gmail.com> wrote: > > On Tue, Nov 21, 2017 at 04:01:16PM +0100, Karol Herbst wrote: > >> This should make systems more stable where resuming the GPU fails. This > >> can happen due to bad firmware or due to a bug within the kernel.

Hi all, if a device doesn''t support power management (pm_cap == 0) but it is acpi_pci_power_manageable() because there is a _PS0 method declared for it and _EJ0 is also declared for the slot then nobody is going to set current_state = PCI_D0 for this device. This is what I think it is happening: pci_enable_device | __pci_enable_device_flags /* here we do not set current_state

why the nouveau_is_rpm_worker stuff? On Sat, Nov 17, 2018 at 2:50 AM Lyude Paul <lyude at redhat.com> wrote: > > Now that we have ->pre_transfer() and ->post_transfer() for DP AUX > channel devices, we can implement these hooks in order to ensure that > the GPU is actually woken up before AUX transactions happen. This fixes > /dev/drm_dp_aux* not working while the GPU is

Here's some fixes for the less important DP AUX issues I mentioned a while back. Lyude Paul (2): drm/dp: Add ->pre/post_transfer() hooks for drm_dp_aux drm/nouveau: Grab an rpm reference before/after DP AUX transactions drivers/gpu/drm/drm_dp_helper.c | 5 ++ drivers/gpu/drm/nouveau/nouveau_connector.c | 36 ++++++++ drivers/gpu/drm/nouveau/nouveau_drm.c | 12 ++-

On Sat, 2019-05-04 at 18:32 +0200, Karol Herbst wrote: > Signed-off-by: Karol Herbst <kherbst at redhat.com> > --- > drm/nouveau/nouveau_acpi.c | 6 +++--- > drm/nouveau/nouveau_acpi.h | 4 ++-- > drm/nouveau/nouveau_drm.c | 15 +++++++++++---- > drm/nouveau/nouveau_drv.h | 2 ++ > 4 files changed, 18 insertions(+), 9 deletions(-) > > diff --git

Now that we have ->pre_transfer() and ->post_transfer() for DP AUX channel devices, we can implement these hooks in order to ensure that the GPU is actually woken up before AUX transactions happen. This fixes /dev/drm_dp_aux* not working while the GPU is suspended, along with some more rare issues where the GPU might runtime-suspend if the time between two DP AUX channel transactions ends up

This should make systems more stable where resuming the GPU fails. This can happen due to bad firmware or due to a bug within the kernel. The last thing which should happen in either case is an unusable system. Signed-off-by: Karol Herbst <kherbst at redhat.com> --- drm/nouveau/nouveau_drm.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git

While investigating the runpm issues on my GP107 I noticed that something inside devinit makes runpm break. If Nouveau loads up to the point right before doing devinit, runpm works without any issues, if devinit is ran, not anymore. Out of curiousity I even tried to "bisect" devinit by not running it on vbios provided signed PMU image, but on the devinit parser we have inside Nouveau.

CCing linux-pci and Bjorn Helgaas. Maybe we could get better insights on how a reasonable fix would look like. Anyway, to me this entire issue looks like something which has to be fixed on a PCI level instead of inside a driver, so it makes sense to ask the pci folks if they have any better suggestions. Original cover letter: While investigating the runpm issues on my GP107 I noticed that

Hi Ben, On Tue, Nov 08, 2016 at 09:29:38PM +0100, Peter Wu wrote: > On Tue, Nov 08, 2016 at 12:57:00PM +0100, Lukas Wunner wrote: > > nouveau's ->suspend and ->resume callbacks are currently skipped if the > > device's status is either DRM_SWITCH_POWER_OFF (powered off by > > vga_switcheroo manual power control) or DRM_SWITCH_POWER_DYNAMIC_OFF > > (runtime

nouveau's ->suspend and ->resume callbacks are currently skipped if the device's status is either DRM_SWITCH_POWER_OFF (powered off by vga_switcheroo manual power control) or DRM_SWITCH_POWER_DYNAMIC_OFF (runtime suspended). In the former case this makes sense since the device is powered off behind the PM core's back: It will try to execute the ->suspend and ->resume

Let host know of our state (partial yet) so that the host PCI device is up to where we were when we were suspended. This is still an RFC as this doesn't completely work: an unused device at the time of suspend will work fine after resume, but a device that has seen some activity doesn't behave well after resume. Especially, host->guest communication doesn't go through. This could

Let host know of our state (partial yet) so that the host PCI device is up to where we were when we were suspended. This is still an RFC as this doesn't completely work: an unused device at the time of suspend will work fine after resume, but a device that has seen some activity doesn't behave well after resume. Especially, host->guest communication doesn't go through. This could

Hello, Turns out S3 is not different from S4 for virtio devices: the device is assumed to be reset, so the host and guest state are to be assumed to be out of sync upon resume. We handle the S4 case with exactly the same scenario, so just point the suspend/resume routines to the freeze/restore ones. Once that is done, we also use the PM API's macro to initialise the sleep functions. A

Hello, Turns out S3 is not different from S4 for virtio devices: the device is assumed to be reset, so the host and guest state are to be assumed to be out of sync upon resume. We handle the S4 case with exactly the same scenario, so just point the suspend/resume routines to the freeze/restore ones. Once that is done, we also use the PM API's macro to initialise the sleep functions. A

This fixes quite a number of runtime PM bugs I found that have been causing some pretty nasty issues such as: - Deadlocking on boot - Connector probing potentially not working while the GPU is in runtime suspend - i2c char dev not working while the GPU is in runtime suspend - aux char dev not working while the GPU is in runtime suspend There's definitely more parts of nouveau that need

Signed-off-by: Karol Herbst <kherbst at redhat.com> --- drm/nouveau/nouveau_acpi.c | 6 +++--- drm/nouveau/nouveau_acpi.h | 4 ++-- drm/nouveau/nouveau_drm.c | 15 +++++++++++---- drm/nouveau/nouveau_drv.h | 2 ++ 4 files changed, 18 insertions(+), 9 deletions(-) diff --git a/drm/nouveau/nouveau_acpi.c b/drm/nouveau/nouveau_acpi.c index ffb19585..49b11d22 100644 ---

v2: rework detection of if Nouveau calling a DSM method or not Signed-off-by: Karol Herbst <kherbst at redhat.com> --- drm/nouveau/nouveau_acpi.c | 7 ++++++- drm/nouveau/nouveau_acpi.h | 2 ++ drm/nouveau/nouveau_drm.c | 14 +++++++++++--- drm/nouveau/nouveau_drv.h | 2 ++ 4 files changed, 21 insertions(+), 4 deletions(-) diff --git a/drm/nouveau/nouveau_acpi.c