/sys/class/drm/*/enabled reports “disabled” for enabled monitors

[RMTT]

NVIDIA Open GPU Kernel Modules Version

570.169

Please confirm this issue does not happen with the proprietary driver (of the same version). This issue tracker is only for bugs specific to the open kernel driver.

• I confirm that this does not happen with the proprietary driver package.

Operating System and Version

NixOS

Kernel Release

6.15.2

Please confirm you are running a stable release kernel (e.g. not a -rc). We do not accept bug reports for unreleased kernels.

• I am running on a stable kernel release.

Hardware: GPU

NVIDIA GeForce RTX 5070 Ti Laptop GPU

Describe the bug

That's old problem which had been reported in https://forums.developer.nvidia.com/t/sys-class-drm-enabled-reports-disabled-for-enabled-monitors/51187 , but it doesn't been solved until now.

I found that kernel updates /sys/class/drm/*/enabled by(https://github.com/torvalds/linux/blob/master/drivers/gpu/drm/drm_sysfs.c#L251):

	struct drm_connector *connector = to_drm_connector(device);
	bool enabled;

	enabled = READ_ONCE(connector->encoder);

	return sysfs_emit(buf, enabled ? "enabled\n" : "disabled\n");

Seems just check the existence of field connector->encoder . I skimmed nvidia-drm sources, seems there is no assignment for connector->encoder, and it only uses connector->nv_detected_encoder .

Some softwares which depend on /sys/class/drm/*/enabled to detect monitor's status cannot work incorrectly when use nvidia card, for example, logind in laptop with nvidia card.

To Reproduce

cat /sys/class/drm/*/enabled

Bug Incidence

Once

nvidia-bug-report.log.gz

nvidia-bug-report.log.gz

More Info

No response

[Binary-Eater]

Thanks @RMTT. I was doing my own experiments recently and found the same behavior as well. Looking into resolving this.

[Binary-Eater]

As a FYI, it's a bug if a DRM driver explicitly sets connector->encoder.

/**
 * drm_connector_attach_encoder - attach a connector to an encoder
 * @connector: connector to attach
 * @encoder: encoder to attach @connector to
 *
 * This function links up a connector to an encoder. Note that the routing
 * restrictions between encoders and crtcs are exposed to userspace through the
 * possible_clones and possible_crtcs bitmasks.
 *
 * Returns:
 * Zero on success, negative errno on failure.
 */
int drm_connector_attach_encoder(struct drm_connector *connector,
				 struct drm_encoder *encoder)
{
	/*
	 * In the past, drivers have attempted to model the static association
	 * of connector to encoder in simple connector/encoder devices using a
	 * direct assignment of connector->encoder = encoder. This connection
	 * is a logical one and the responsibility of the core, so drivers are
	 * expected not to mess with this.
	 *
	 * Note that the error return should've been enough here, but a large
	 * majority of drivers ignores the return value, so add in a big WARN
	 * to get people's attention.
	 */
	if (WARN_ON(connector->encoder))
		return -EINVAL;

	connector->possible_encoders |= drm_encoder_mask(encoder);

	return 0;
}
EXPORT_SYMBOL(drm_connector_attach_encoder);

[RMTT]

Looking into resolving this

Thanks! This bug is very crucial for improving linux desktop experience with NVIDIA in laptops.

[Binary-Eater]

Wanted to update we have a fix for this issue internally.

[RMTT]

Wanted to update we have a fix for this issue internally.

Thanks for your work! Will this fix be included in next release?

[abchauhan-nv]

Hi all, the issue was tracked on NVBug #5420364. This is now resolved and the fix will be available in a future driver release. Thanks

[ShadiestGoat]

Hi - is this fix supposed to be in current release? I seem to still be able to reproduce this on latest version

[Binary-Eater]

The fix I wrote has not landed in the current releases. I think @abchauhan-nv will update the thread when it's out.

[Binary-Eater]

We just had a new release

https://www.nvidia.com/en-us/drivers/details/253003/

[alexphanna]

This release did not fix this issue for me. cat /sys/class/drm/*/enabled still reports all disabled. I am using a 3060 and have tested both the 580.82.07 and 580.82.09 drivers, neither fix it for me. I am also on NixOS if that matters.

[RMTT]

This release did not fix this issue for me. cat /sys/class/drm/*/enabled still reports all disabled. I am using a 3060 and have tested both the 580.82.07 and 580.82.09 drivers, neither fix it for me. I am also on NixOS if that matters.

I'm using 5070Ti Laptop with 580.82.09 driver. The problem had been fixed.

[alexphanna]

This release did not fix this issue for me. cat /sys/class/drm/*/enabled still reports all disabled. I am using a 3060 and have tested both the 580.82.07 and 580.82.09 drivers, neither fix it for me. I am also on NixOS if that matters.

I'm using 5070Ti Laptop with 580.82.09 driver. The problem had been fixed.

That's interesting, I am on a desktop and my monitor is plugged into an HDMI port on my graphics card. I feel like this shouldn't matter because the nouveau drivers show it as enabled. I am also on the latest kernel 6.16.7.

[abchauhan-nv]

@alexphanna , Can you please capture a NVIDIA bug report from your system and attach it here? Thanks