similar to: [PATCH v2] drm/nouveau/i2c: Disable i2c bus access after ->fini()

On Thu, 2019-04-11 at 08:48 +1000, Ben Skeggs wrote: > On Wed, 10 Apr 2019 at 06:23, Lyude Paul <lyude at redhat.com> wrote: > > For a while, we've had the problem of i2c bus access not grabbing > > a runtime PM ref when it's being used in userspace by i2c-dev, resulting > > in nouveau spamming the kernel log with errors if anything attempts to > > access

On Wed, 10 Apr 2019 at 06:23, Lyude Paul <lyude at redhat.com> wrote: > > For a while, we've had the problem of i2c bus access not grabbing > a runtime PM ref when it's being used in userspace by i2c-dev, resulting > in nouveau spamming the kernel log with errors if anything attempts to > access the i2c bus while the GPU is in runtime suspend. An example: > > [

mhh, we shouldn't call to Linux APIs from within of nvkm. Rather gaurd the Linux glue code to the i2c stuff instead, but this is all done from inside of nvkm. I think we should move it out into drm/nouveau/nouveau_i2c.c and do the handling there. On Tue, Jul 17, 2018 at 1:59 AM, Lyude Paul <lyude at redhat.com> wrote: > The i2c bus can be both accessed by DRM itself, along with any

For a while, we've had the problem of i2c bus access not grabbing a runtime PM ref when it's being used in userspace by i2c-dev, resulting in nouveau spamming the kernel log with errors if anything attempts to access the i2c bus while the GPU is in runtime suspend. An example: [ 130.078386] nouveau 0000:01:00.0: i2c: aux 000d: begin idle timeout ffffffff Since the GPU is in runtime

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

The i2c bus can be both accessed by DRM itself, along with any of it's devnodes (/sys/class/i2c). So, we need to make sure that all codepaths using the i2c bus keep the GPU resumed. Signed-off-by: Lyude Paul <lyude at redhat.com> Cc: Karol Herbst <karolherbst at gmail.com> Cc: stable at vger.kernel.org --- drivers/gpu/drm/nouveau/nvkm/subdev/i2c/bus.c | 12 +++++++++++- 1 file

On Tue, 17 Jul 2018 at 20:18, Karol Herbst <kherbst at redhat.com> wrote: > > mhh, we shouldn't call to Linux APIs from within of nvkm. Rather gaurd > the Linux glue code to the i2c stuff instead, but this is all done > from inside of nvkm. I think we should move it out into > drm/nouveau/nouveau_i2c.c and do the handling there. Huh? No, this is completely fine. >

It turns out that while disabling i2c bus access from software when the GPU is suspended was a step in the right direction with: commit 342406e4fbba ("drm/nouveau/i2c: Disable i2c bus access after ->fini()") We also ended up accidentally breaking the vbios init scripts on some older Tesla GPUs, as apparently said scripts can actually use the i2c bus. Since these scripts are executed

The power sensors can be configured to sample the readout values over time. Nvidia does this too, so nouveau should probably do that too. Karol Herbst (4): iccsense: remove read function iccsense: convert to linked list iccsense: split sensor into own struct iccsense: configure sensors like nvidia does drm/nouveau/include/nvkm/subdev/iccsense.h | 6 +- drm/nouveau/nouveau_hwmon.c

This is another attempt at fixing an issue with yes | sensors-detect Causing some machines with nouveau loaded to hang if certain kinds of displays are attached. I've also included one minor fix that I found along the way of troubleshooting this issue. Lyude Paul (2): drm/nouveau: Fix missing elses in g94_i2c_aux_xfer drm/nouveau: Don't retry infinitely when receiving no data on i2c

The power sensors can be configured to sample the readout values over time. Nvidia does this too, so nouveau should probably do that too. v2: use list_del and rework an error message Karol Herbst (4): iccsense: remove read function iccsense: convert to linked list iccsense: split sensor into own struct iccsense: configure sensors like nvidia does

Right now, we make the mistake of allowing interlacing on all connectors. Nvidia hardware does not always support interlacing with DP though, so we need to make sure that we don't allow interlaced modes to be set in such situations as otherwise we'll end up accidentally hanging the display HW. This fixes some hangs with Turing, which would be caused by attempting to set an interlaced mode

Currently, nouveau doesn't actually bother to try probing whether or not it can actually handle interlaced modes over DisplayPort. As a result, on volta and later we'll end up trying to set an interlaced mode even when it's not supported and cause the front end for the display engine to hang. So, let's teach nouveau to reject interlaced modes on hardware that can't actually

Signed-off-by: Karol Herbst <nouveau at karolherbst.de> --- drm/nouveau/include/nvkm/subdev/iccsense.h | 1 + drm/nouveau/nvkm/subdev/iccsense/base.c | 141 ++++++++++++++++++++--------- drm/nouveau/nvkm/subdev/iccsense/priv.h | 15 ++- 3 files changed, 112 insertions(+), 45 deletions(-) diff --git a/drm/nouveau/include/nvkm/subdev/iccsense.h

The INA3221 is usually found on mid and high end kepler+ gpus Marins Patch implements the new iccsense subdev and all needed bits for the INA3221 power sensor. My Patch implements the hwmon power1 interface to expose the current power consumption through hwmon (and can be read out via sysfs or the sensors tool) Please test these patches for Fermi+ GPUs, that nothing gets messed up and works as

We are deprecating calls which return NULL in favor of new variants which return an ERR_PTR. Only build tested. Wolfram Sang (6): drm/amdgpu: convert to use i2c_new_client_device() drm/gma500: convert to use i2c_new_client_device() drm/i2c/sil164: convert to use i2c_new_client_device() drm/i2c/tda998x: convert to use i2c_new_client_device() drm/nouveau/therm: convert to use

From: Oleg Vasilev <oleg.vasilev at intel.com> Since DP-specific information is stored in driver's structures, every driver needs to implement subconnector property by itself. v2: rebase Cc: Ben Skeggs <bskeggs at redhat.com> Cc: nouveau at lists.freedesktop.org Signed-off-by: Jeevan B <jeevan.b at intel.com> Signed-off-by: Oleg Vasilev <oleg.vasilev at intel.com>

Since DP-specific information is stored in driver's structures, every driver needs to implement subconnector property by itself. Signed-off-by: Oleg Vasilev <oleg.vasilev at intel.com> Cc: nouveau at lists.freedesktop.org --- drivers/gpu/drm/nouveau/nouveau_connector.c | 13 +++++++++++++ drivers/gpu/drm/nouveau/nouveau_dp.c | 9 +++++++++

This is a complete rework from the first version I sent out. Now the implementation is more centered around the power_rails we find in the SENSE table instead of extdev centered. This makes the implementation a lot easier and straightforward. I've added support for the INA219, INA209 and INA3221 sensors found on multiple Fermi and Kepler cards. The power consumption is also exported via

From: Ville Syrjälä <ville.syrjala at linux.intel.com> We now have per-device driver_features, so let's use that to disable atomic only for pre-nv50. Cc: Ben Skeggs <bskeggs at redhat.com> Cc: Lyude Paul <lyude at redhat.com> Cc: nouveau at lists.freedesktop.org Cc: Daniel Vetter <daniel.vetter at ffwll.ch> Reviewed-by: Daniel Vetter <daniel.vetter at ffwll.ch>