similar to: [PATCH AUTOSEL 5.7 49/61] drm/nouveau: Fix reference count leak in nouveau_connector_detect

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 990a1162986e8eff7ca18cc5a0e03b4304392ae2 ] nouveau_connector_detect() calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 990a1162986e8eff7ca18cc5a0e03b4304392ae2 ] nouveau_connector_detect() calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 990a1162986e8eff7ca18cc5a0e03b4304392ae2 ] nouveau_connector_detect() calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 990a1162986e8eff7ca18cc5a0e03b4304392ae2 ] nouveau_connector_detect() calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 990a1162986e8eff7ca18cc5a0e03b4304392ae2 ] nouveau_connector_detect() calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 990a1162986e8eff7ca18cc5a0e03b4304392ae2 ] nouveau_connector_detect() calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

nouveau_connector_detect() calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> --- drivers/gpu/drm/nouveau/nouveau_connector.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c

It's true we can't resume the device from poll workers in nouveau_connector_detect(). We can however, prevent the autosuspend timer from elapsing immediately if it hasn't already without risking any sort of deadlock with the runtime suspend/resume operations. So do that instead of entirely avoiding grabbing a power reference. Signed-off-by: Lyude Paul <lyude at redhat.com> Cc:

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit bfad51c7633325b5d4b32444efe04329d53297b2 ] nouveau_fbcon_open() calls calls pm_runtime_get_sync() that increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit a2cdf39536b0d21fb06113f5e16692513d7bcb9c ] nv50_disp_atomic_commit() calls calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

nouveau's ->runtime_suspend hook calls drm_kms_helper_poll_disable(), which waits for the output poll worker to finish if it's running. The output poll worker meanwhile calls pm_runtime_get_sync() in nouveau_connector_detect() which waits for the ongoing suspend to finish, causing a deadlock. Fix by not acquiring a runtime PM ref if nouveau_connector_detect() is called in the output

When we disable hotplugging on the GPU, we need to be able to synchronize with each connector's hotplug interrupt handler before the interrupt is finally disabled. This can be a problem however, since nouveau_connector_detect() currently grabs a runtime power reference when handling connector probing. This will deadlock the runtime suspend handler like so: [ 861.480896] INFO: task

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit a2cdf39536b0d21fb06113f5e16692513d7bcb9c ] nv50_disp_atomic_commit() calls calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 8f29432417b11039ef960ab18987c7d61b2b5396 ] nouveau_debugfs_strap_peek() calls pm_runtime_get_sync() that increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 659fb5f154c3434c90a34586f3b7aa1c39cf6062 ] On calling pm_runtime_get_sync() the reference count of the device is incremented. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com> Signed-off-by: Sasha

No functional changes Signed-off-by: Lyude Paul <lyude at redhat.com> --- drivers/gpu/drm/nouveau/nouveau_connector.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 9a9a7f5003d3..0d42a7e5deff 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++

On Wed, Aug 01, 2018 at 05:14:57PM -0400, Lyude Paul wrote: > When we disable hotplugging on the GPU, we need to be able to > synchronize with each connector's hotplug interrupt handler before the > interrupt is finally disabled. This can be a problem however, since > nouveau_connector_detect() currently grabs a runtime power reference > when handling connector probing. This

Thanks, I've grabbed this, and the others of the same sort you sent out at the same time. Ben. On Mon, 15 Jun 2020 at 17:29, Aditya Pakki <pakki001 at umn.edu> wrote: > > nouveau_debugfs_strap_peek() calls pm_runtime_get_sync() that > increments the reference count. In case of failure, decrement the > ref count before returning the error. > > Signed-off-by: Aditya

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 8f29432417b11039ef960ab18987c7d61b2b5396 ] nouveau_debugfs_strap_peek() calls pm_runtime_get_sync() that increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com>

From: Aditya Pakki <pakki001 at umn.edu> [ Upstream commit 659fb5f154c3434c90a34586f3b7aa1c39cf6062 ] On calling pm_runtime_get_sync() the reference count of the device is incremented. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> Signed-off-by: Ben Skeggs <bskeggs at redhat.com> Signed-off-by: Sasha