[PATCH 0/3] A few minor fw_devlink fixes

[PATCH 0/3] A few minor fw_devlink fixes

[Saravana Kannan]

Probably easiest for Greg to pull in these changes?

PSA: Do not pull any of these patches into stable kernels. fw_devlink
had a lot of changes that landed in the last year. It's hard to ensure
cherry-picks have picked up all the dependencies correctly. If any of
these really need to get cherry-picked into stable kernels, cc me and
wait for my explicit Ack.

Thanks,
Saravana

Saravana Kannan (3):
  drm: display: Set fwnode for aux bus devices
  phy: tegra: xusb: Set fwnode for xusb port devices
  drivers: core: fw_devlink: Make the error message a bit more useful

 drivers/base/core.c                      | 4 ++--
 drivers/gpu/drm/display/drm_dp_aux_bus.c | 2 +-
 drivers/phy/tegra/xusb.c                 | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

-- 
2.47.0.105.g07ac214952-goog

[PATCH 1/3] drm: display: Set fwnode for aux bus devices

[Saravana Kannan]

fwnode needs to be set for a device for fw_devlink to be able to
track/enforce its dependencies correctly. Without this, you'll see error
messages like this when the supplier has probed and tries to make sure
all its fwnode consumers are linked to it using device links:

mediatek-drm-dp 1c500000.edp-tx: Failed to create device link (0x180) with backlight-lcd0

Reported-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
Closes: https://lore.kernel.org/all/7b995947-4540-4b17-872e-e107adca4598@notapiano/
Tested-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
Signed-off-by: Saravana Kannan <saravanak@google.com>
---
 drivers/gpu/drm/display/drm_dp_aux_bus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/display/drm_dp_aux_bus.c b/drivers/gpu/drm/display/drm_dp_aux_bus.c
index d810529ebfb6..ec7eac6b595f 100644
--- a/drivers/gpu/drm/display/drm_dp_aux_bus.c
+++ b/drivers/gpu/drm/display/drm_dp_aux_bus.c
@@ -292,7 +292,7 @@ int of_dp_aux_populate_bus(struct drm_dp_aux *aux,
 	aux_ep->dev.parent = aux->dev;
 	aux_ep->dev.bus = &dp_aux_bus_type;
 	aux_ep->dev.type = &dp_aux_device_type_type;
-	aux_ep->dev.of_node = of_node_get(np);
+	device_set_node(&aux_ep->dev, of_fwnode_handle(of_node_get(np)));
 	dev_set_name(&aux_ep->dev, "aux-%s", dev_name(aux->dev));
 
 	ret = device_register(&aux_ep->dev);
-- 
2.47.0.105.g07ac214952-goog

[PATCH 2/3] phy: tegra: xusb: Set fwnode for xusb port devices

[Saravana Kannan]

fwnode needs to be set for a device for fw_devlink to be able to
track/enforce its dependencies correctly. Without this, you'll see error
messages like this when the supplier has probed and tries to make sure
all its fwnode consumers are linked to it using device links:

tegra-xusb-padctl 3520000.padctl: Failed to create device link (0x180) with 1-0008

Reported-by: Jon Hunter <jonathanh@nvidia.com>
Closes: https://lore.kernel.org/all/20240910130019.35081-1-jonathanh@nvidia.com/
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Suggested-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
Signed-off-by: Saravana Kannan <saravanak@google.com>
---
 drivers/phy/tegra/xusb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/phy/tegra/xusb.c b/drivers/phy/tegra/xusb.c
index cfdb54b6070a..0a2096085971 100644
--- a/drivers/phy/tegra/xusb.c
+++ b/drivers/phy/tegra/xusb.c
@@ -543,7 +543,7 @@ static int tegra_xusb_port_init(struct tegra_xusb_port *port,
 
 	device_initialize(&port->dev);
 	port->dev.type = &tegra_xusb_port_type;
-	port->dev.of_node = of_node_get(np);
+	device_set_node(&port->dev, of_fwnode_handle(of_node_get(np)));
 	port->dev.parent = padctl->dev;
 
 	err = dev_set_name(&port->dev, "%s-%u", name, index);
-- 
2.47.0.105.g07ac214952-goog

[PATCH 3/3] drivers: core: fw_devlink: Make the error message a bit more useful

[Saravana Kannan]

It would make it easier to debugs issues similar to the ones
reported[1][2] recently where some devices didn't have the fwnode set.

[1] - https://lore.kernel.org/all/7b995947-4540-4b17-872e-e107adca4598@notapiano/
[2] - https://lore.kernel.org/all/20240910130019.35081-1-jonathanh@nvidia.com/

Signed-off-by: Saravana Kannan <saravanak@google.com>
---
 drivers/base/core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index a4c853411a6b..3b13fed1c3e3 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -2181,8 +2181,8 @@ static int fw_devlink_create_devlink(struct device *con,
 		}
 
 		if (con != sup_dev && !device_link_add(con, sup_dev, flags)) {
-			dev_err(con, "Failed to create device link (0x%x) with %s\n",
-				flags, dev_name(sup_dev));
+			dev_err(con, "Failed to create device link (0x%x) with supplier %s for %pfwf\n",
+				flags, dev_name(sup_dev), link->consumer);
 			ret = -EINVAL;
 		}
 
-- 
2.47.0.105.g07ac214952-goog

Re: [PATCH 3/3] drivers: core: fw_devlink: Make the error message a bit more useful

[Saravana Kannan]

Greg,

Can you fix up the commit subject prefix to "driver core: fw_devlink:"
please? Don't want to send v2 just for that.

-Saravana

On Wed, Oct 23, 2024 at 11:14 PM Saravana Kannan <saravanak@google.com> wrote:
>
> It would make it easier to debugs issues similar to the ones
> reported[1][2] recently where some devices didn't have the fwnode set.
>
> [1] - https://lore.kernel.org/all/7b995947-4540-4b17-872e-e107adca4598@notapiano/
> [2] - https://lore.kernel.org/all/20240910130019.35081-1-jonathanh@nvidia.com/
>
> Signed-off-by: Saravana Kannan <saravanak@google.com>
> ---
>  drivers/base/core.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/base/core.c b/drivers/base/core.c
> index a4c853411a6b..3b13fed1c3e3 100644
> --- a/drivers/base/core.c
> +++ b/drivers/base/core.c
> @@ -2181,8 +2181,8 @@ static int fw_devlink_create_devlink(struct device *con,
>                 }
>
>                 if (con != sup_dev && !device_link_add(con, sup_dev, flags)) {
> -                       dev_err(con, "Failed to create device link (0x%x) with %s\n",
> -                               flags, dev_name(sup_dev));
> +                       dev_err(con, "Failed to create device link (0x%x) with supplier %s for %pfwf\n",
> +                               flags, dev_name(sup_dev), link->consumer);
>                         ret = -EINVAL;
>                 }
>
> --
> 2.47.0.105.g07ac214952-goog
>

Re: [PATCH 3/3] drivers: core: fw_devlink: Make the error message a bit more useful

[Greg Kroah-Hartman]

On Wed, Oct 23, 2024 at 11:16:26PM -0700, Saravana Kannan wrote:
> Greg,
> 
> Can you fix up the commit subject prefix to "driver core: fw_devlink:"
> please? Don't want to send v2 just for that.

Will do, thanks!

Re: [PATCH 0/3] A few minor fw_devlink fixes

[Greg Kroah-Hartman]

On Wed, Oct 23, 2024 at 11:13:41PM -0700, Saravana Kannan wrote:
> Probably easiest for Greg to pull in these changes?
> 
> PSA: Do not pull any of these patches into stable kernels. fw_devlink
> had a lot of changes that landed in the last year. It's hard to ensure
> cherry-picks have picked up all the dependencies correctly. If any of
> these really need to get cherry-picked into stable kernels, cc me and
> wait for my explicit Ack.

You can do that with the correct tag in the commit as per the stable
documentation if you really want to :)

But why would these not be able to go backwards?  What changed to
require them now and not be ok for older kernels?

thanks,

greg k-h

Re: [PATCH 0/3] A few minor fw_devlink fixes

[Saravana Kannan]

On Thu, Oct 24, 2024 at 12:08 AM Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
>
> On Wed, Oct 23, 2024 at 11:13:41PM -0700, Saravana Kannan wrote:
> > Probably easiest for Greg to pull in these changes?
> >
> > PSA: Do not pull any of these patches into stable kernels. fw_devlink
> > had a lot of changes that landed in the last year. It's hard to ensure
> > cherry-picks have picked up all the dependencies correctly. If any of
> > these really need to get cherry-picked into stable kernels, cc me and
> > wait for my explicit Ack.
>
> You can do that with the correct tag in the commit as per the stable
> documentation if you really want to :)
>
> But why would these not be able to go backwards?  What changed to
> require them now and not be ok for older kernels?

Depending on how far back we go in terms of kernel version, it'd be a
bunch of cycle detection logic, deferred probe behavior change, etc.
And those patches might have other dependencies themselves, etc.

He's one example of such a breakage:
https://lore.kernel.org/all/20241024-fixup-5-15-v1-1-74d360bd3002@mediatek.com/

-Saravana

Re: [PATCH 1/3] drm: display: Set fwnode for aux bus devices

[Thierry Reding]

[-- Attachment #1: Type: text/plain, Size: 889 bytes --]

On Wed, Oct 23, 2024 at 11:13:42PM -0700, Saravana Kannan wrote:
> fwnode needs to be set for a device for fw_devlink to be able to
> track/enforce its dependencies correctly. Without this, you'll see error
> messages like this when the supplier has probed and tries to make sure
> all its fwnode consumers are linked to it using device links:
> 
> mediatek-drm-dp 1c500000.edp-tx: Failed to create device link (0x180) with backlight-lcd0
> 
> Reported-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
> Closes: https://lore.kernel.org/all/7b995947-4540-4b17-872e-e107adca4598@notapiano/
> Tested-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
> Signed-off-by: Saravana Kannan <saravanak@google.com>
> ---
>  drivers/gpu/drm/display/drm_dp_aux_bus.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Thierry Reding <treding@nvidia.com>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH 2/3] phy: tegra: xusb: Set fwnode for xusb port devices

[Thierry Reding]

[-- Attachment #1: Type: text/plain, Size: 893 bytes --]

On Wed, Oct 23, 2024 at 11:13:43PM -0700, Saravana Kannan wrote:
> fwnode needs to be set for a device for fw_devlink to be able to
> track/enforce its dependencies correctly. Without this, you'll see error
> messages like this when the supplier has probed and tries to make sure
> all its fwnode consumers are linked to it using device links:
> 
> tegra-xusb-padctl 3520000.padctl: Failed to create device link (0x180) with 1-0008
> 
> Reported-by: Jon Hunter <jonathanh@nvidia.com>
> Closes: https://lore.kernel.org/all/20240910130019.35081-1-jonathanh@nvidia.com/
> Tested-by: Jon Hunter <jonathanh@nvidia.com>
> Suggested-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
> Signed-off-by: Saravana Kannan <saravanak@google.com>
> ---
>  drivers/phy/tegra/xusb.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Acked-by: Thierry Reding <treding@nvidia.com>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH 1/3] drm: display: Set fwnode for aux bus devices

[Dmitry Baryshkov]

On Wed, Oct 23, 2024 at 11:13:42PM -0700, Saravana Kannan wrote:
> fwnode needs to be set for a device for fw_devlink to be able to
> track/enforce its dependencies correctly. Without this, you'll see error
> messages like this when the supplier has probed and tries to make sure
> all its fwnode consumers are linked to it using device links:
> 
> mediatek-drm-dp 1c500000.edp-tx: Failed to create device link (0x180) with backlight-lcd0
> 
> Reported-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
> Closes: https://lore.kernel.org/all/7b995947-4540-4b17-872e-e107adca4598@notapiano/
> Tested-by: "Nícolas F. R. A. Prado" <nfraprado@collabora.com>
> Signed-off-by: Saravana Kannan <saravanak@google.com>
> ---
>  drivers/gpu/drm/display/drm_dp_aux_bus.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>

-- 
With best wishes
Dmitry