[char-misc-next] mei: gsc_proxy: match component when GSC is on different bus

Message ID 20240211130408.3478-1-tomas.winkler@intel.com
State New
Headers
Series [char-misc-next] mei: gsc_proxy: match component when GSC is on different bus |

Commit Message

Tomas Winkler Feb. 11, 2024, 1:04 p.m. UTC
  From: Alexander Usyskin <alexander.usyskin@intel.com>

On Arrow Lake S systems, MEI is no longer strictly connected to bus 0,
while graphics remain exclusively on bus 0. Adapt the component
matching logic to accommodate this change:

Original behavior: Required both MEI and graphics to be on the same
bus 0.

New behavior: Only enforces graphics to be on bus 0 (integrated),
allowing MEI to reside on any bus.
This ensures compatibility with Arrow Lake S and maintains functionality
for the legacy systems.

Cc: <stable@vger.kernel.org>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
Requires 'mei: me: add arrow lake point S DID'
https://lore.kernel.org/lkml/20240211103912.117105-1-tomas.winkler@intel.com/

 drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
  

Comments

Greg KH Feb. 19, 2024, 10:40 a.m. UTC | #1
On Sun, Feb 11, 2024 at 03:04:08PM +0200, Tomas Winkler wrote:
> From: Alexander Usyskin <alexander.usyskin@intel.com>
> 
> On Arrow Lake S systems, MEI is no longer strictly connected to bus 0,
> while graphics remain exclusively on bus 0. Adapt the component
> matching logic to accommodate this change:
> 
> Original behavior: Required both MEI and graphics to be on the same
> bus 0.
> 
> New behavior: Only enforces graphics to be on bus 0 (integrated),
> allowing MEI to reside on any bus.
> This ensures compatibility with Arrow Lake S and maintains functionality
> for the legacy systems.
> 
> Cc: <stable@vger.kernel.org>
> Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>

What commit id does this fix?  How far back in stable kernels should it
go?

thanks,

greg k-h
  
Tomas Winkler Feb. 20, 2024, 7:52 p.m. UTC | #2
> 
> On Sun, Feb 11, 2024 at 03:04:08PM +0200, Tomas Winkler wrote:
> > From: Alexander Usyskin <alexander.usyskin@intel.com>
> >
> > On Arrow Lake S systems, MEI is no longer strictly connected to bus 0,
> > while graphics remain exclusively on bus 0. Adapt the component
> > matching logic to accommodate this change:
> >
> > Original behavior: Required both MEI and graphics to be on the same
> > bus 0.
> >
> > New behavior: Only enforces graphics to be on bus 0 (integrated),
> > allowing MEI to reside on any bus.
> > This ensures compatibility with Arrow Lake S and maintains
> > functionality for the legacy systems.
> >
> > Cc: <stable@vger.kernel.org>
> > Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
> > Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
> 
> What commit id does this fix?  How far back in stable kernels should it go?
Should be backported together with 'mei: me: add arrow lake point S DID'
And 1dd924f6885b ("mei: gsc_proxy: add gsc proxy driver")
I will send v2

Thanks
Tomas
  

Patch

diff --git a/drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c b/drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c
index be52b113aea937c7c658e06c..89364bdbb1290f5726a34945 100644
--- a/drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c
+++ b/drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c
@@ -96,7 +96,8 @@  static const struct component_master_ops mei_component_master_ops = {
  *
  *    The function checks if the device is pci device and
  *    Intel VGA adapter, the subcomponent is SW Proxy
- *    and the parent of MEI PCI and the parent of VGA are the same PCH device.
+ *    and the VGA is on the bus 0 reserved for built-in devices
+ *    to reject discrete GFX.
  *
  * @dev: master device
  * @subcomponent: subcomponent to match (I915_COMPONENT_SWPROXY)
@@ -123,7 +124,8 @@  static int mei_gsc_proxy_component_match(struct device *dev, int subcomponent,
 	if (subcomponent != I915_COMPONENT_GSC_PROXY)
 		return 0;
 
-	return component_compare_dev(dev->parent, ((struct device *)data)->parent);
+	/* Only built-in GFX */
+	return (pdev->bus->number == 0);
 }
 
 static int mei_gsc_proxy_probe(struct mei_cl_device *cldev,
@@ -146,7 +148,7 @@  static int mei_gsc_proxy_probe(struct mei_cl_device *cldev,
 	}
 
 	component_match_add_typed(&cldev->dev, &master_match,
-				  mei_gsc_proxy_component_match, cldev->dev.parent);
+				  mei_gsc_proxy_component_match, NULL);
 	if (IS_ERR_OR_NULL(master_match)) {
 		ret = -ENOMEM;
 		goto err_exit;