mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH v2] cxl/acpi: Check ACPI companion before use
@ 2026-09-26  7:16 Jiale Yao
  2026-09-27 22:50 ` Jonathan Cameron
  2026-09-28 23:50 ` Dave Jiang
  0 siblings, 2 replies; 3+ messages in thread
From: Jiale Yao @ 2026-09-26  7:16 UTC (permalink / raw)
  To: Davidlohr Bueso, Jonathan Cameron, Dave Jiang, Alison Schofield,
	Vishal Verma, Dan Williams, Ira Weiny, Li Ming, linux-cxl,
	linux-kernel
  Cc: Jiale Yao, stable

Platform drivers can be forced to match devices outside their ID tables
through driver_override.  cxl_acpi_probe() assumes that every bound device
has an ACPI companion and dereferences adev->dev.bus without checking the
result of ACPI_COMPANION().  Force-binding cxl_acpi to a platform device
without a companion therefore causes a NULL pointer dereference.

This was reproduced by setting the driver override for the pcspkr platform
device to cxl_acpi and binding it through sysfs:

  BUG: kernel NULL pointer dereference, address: 0000000000000280
  #PF: supervisor read access in kernel mode
  RIP: cxl_acpi_probe+0xf4/0x220
  Call Trace:
   platform_probe+0x4d/0x80
   really_probe+0x106/0x370
   device_driver_attach+0x4c/0xa0
   bind_store+0xd0/0x100

Commit 2b3a5dabe89e ("platform/surface: acpi-notify: Check ACPI
companion before use") fixed the same force-binding issue in another
platform driver.  Check the companion before setting up the CXL root and
return -ENODEV when it is absent.

Fixes: 7d4b5ca2e2cb ("cxl/acpi: Add downstream port data to cxl_port instances")
Cc: stable@vger.kernel.org
Signed-off-by: Jiale Yao <yaojiale02@163.com>
---

Notes:
    Changes in v2:
    - Move the ACPI companion assignment immediately before its NULL check.
    - Reorder local declarations in reverse Christmas tree order.

 drivers/cxl/acpi.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/cxl/acpi.c b/drivers/cxl/acpi.c
index 3b818adbd38b..fb09a5ff48c1 100644
--- a/drivers/cxl/acpi.c
+++ b/drivers/cxl/acpi.c
@@ -885,13 +885,17 @@ static int pair_cxl_resource(struct device *dev, void *data)
 
 static int cxl_acpi_probe(struct platform_device *pdev)
 {
-	int rc;
+	struct cxl_cfmws_context ctx;
+	struct acpi_device *adev;
 	struct resource *cxl_res;
 	struct cxl_root *cxl_root;
 	struct cxl_port *root_port;
 	struct device *host = &pdev->dev;
-	struct acpi_device *adev = ACPI_COMPANION(host);
-	struct cxl_cfmws_context ctx;
+	int rc;
+
+	adev = ACPI_COMPANION(host);
+	if (!adev)
+		return -ENODEV;
 
 	device_lock_set_class(&pdev->dev, &cxl_root_key);
 	rc = devm_add_action_or_reset(&pdev->dev, cxl_acpi_lock_reset_class,
-- 
2.34.1


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v2] cxl/acpi: Check ACPI companion before use
  2026-09-26  7:16 [PATCH v2] cxl/acpi: Check ACPI companion before use Jiale Yao
@ 2026-09-27 22:50 ` Jonathan Cameron
  2026-09-28 23:50 ` Dave Jiang
  1 sibling, 0 replies; 3+ messages in thread
From: Jonathan Cameron @ 2026-09-27 22:50 UTC (permalink / raw)
  To: Jiale Yao
  Cc: Davidlohr Bueso, Dave Jiang, Alison Schofield, Vishal Verma,
	Dan Williams, Ira Weiny, Li Ming, linux-cxl, linux-kernel,
	stable

On Sat, 26 Sep 2026 15:16:05 +0800
Jiale Yao <yaojiale02@163.com> wrote:

> Platform drivers can be forced to match devices outside their ID tables
> through driver_override.  cxl_acpi_probe() assumes that every bound device
> has an ACPI companion and dereferences adev->dev.bus without checking the
> result of ACPI_COMPANION().  Force-binding cxl_acpi to a platform device
> without a companion therefore causes a NULL pointer dereference.
> 
Given I had this drawn to my attention earlier (thanks to Uwe)
let me draw attention to an attempt to add a flag to remove the need
for this sort of defense by blocking driver_override.

https://lore.kernel.org/lkml/20260922-driver-override-opt-out-v1-0-58c35ded3b83@nvidia.com/

So maybe we can replace what we do here once that lands.
In the meantime it's a valid bug so 

Reviewed-by: Jonathan Cameron <jonathan.cameron@oss.qualcomm.com>

> This was reproduced by setting the driver override for the pcspkr platform
> device to cxl_acpi and binding it through sysfs:
> 
>   BUG: kernel NULL pointer dereference, address: 0000000000000280
>   #PF: supervisor read access in kernel mode
>   RIP: cxl_acpi_probe+0xf4/0x220
>   Call Trace:
>    platform_probe+0x4d/0x80
>    really_probe+0x106/0x370
>    device_driver_attach+0x4c/0xa0
>    bind_store+0xd0/0x100
> 
> Commit 2b3a5dabe89e ("platform/surface: acpi-notify: Check ACPI
> companion before use") fixed the same force-binding issue in another
> platform driver.  Check the companion before setting up the CXL root and
> return -ENODEV when it is absent.
> 
> Fixes: 7d4b5ca2e2cb ("cxl/acpi: Add downstream port data to cxl_port instances")
> Cc: stable@vger.kernel.org
> Signed-off-by: Jiale Yao <yaojiale02@163.com>
> ---
> 
> Notes:
>     Changes in v2:
>     - Move the ACPI companion assignment immediately before its NULL check.
>     - Reorder local declarations in reverse Christmas tree order.
> 
>  drivers/cxl/acpi.c | 10 +++++++---
>  1 file changed, 7 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/cxl/acpi.c b/drivers/cxl/acpi.c
> index 3b818adbd38b..fb09a5ff48c1 100644
> --- a/drivers/cxl/acpi.c
> +++ b/drivers/cxl/acpi.c
> @@ -885,13 +885,17 @@ static int pair_cxl_resource(struct device *dev, void *data)
>  
>  static int cxl_acpi_probe(struct platform_device *pdev)
>  {
> -	int rc;
> +	struct cxl_cfmws_context ctx;
> +	struct acpi_device *adev;
>  	struct resource *cxl_res;
>  	struct cxl_root *cxl_root;
>  	struct cxl_port *root_port;
>  	struct device *host = &pdev->dev;
> -	struct acpi_device *adev = ACPI_COMPANION(host);
> -	struct cxl_cfmws_context ctx;
> +	int rc;
> +
> +	adev = ACPI_COMPANION(host);
> +	if (!adev)
> +		return -ENODEV;
>  
>  	device_lock_set_class(&pdev->dev, &cxl_root_key);
>  	rc = devm_add_action_or_reset(&pdev->dev, cxl_acpi_lock_reset_class,


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v2] cxl/acpi: Check ACPI companion before use
  2026-09-26  7:16 [PATCH v2] cxl/acpi: Check ACPI companion before use Jiale Yao
  2026-09-27 22:50 ` Jonathan Cameron
@ 2026-09-28 23:50 ` Dave Jiang
  1 sibling, 0 replies; 3+ messages in thread
From: Dave Jiang @ 2026-09-28 23:50 UTC (permalink / raw)
  To: Jiale Yao, Davidlohr Bueso, Jonathan Cameron, Alison Schofield,
	Vishal Verma, Dan Williams, Ira Weiny, Li Ming, linux-cxl,
	linux-kernel
  Cc: stable



On 9/26/26 12:16 AM, Jiale Yao wrote:
> Platform drivers can be forced to match devices outside their ID tables
> through driver_override.  cxl_acpi_probe() assumes that every bound device
> has an ACPI companion and dereferences adev->dev.bus without checking the
> result of ACPI_COMPANION().  Force-binding cxl_acpi to a platform device
> without a companion therefore causes a NULL pointer dereference.
> 
> This was reproduced by setting the driver override for the pcspkr platform
> device to cxl_acpi and binding it through sysfs:
> 
>   BUG: kernel NULL pointer dereference, address: 0000000000000280
>   #PF: supervisor read access in kernel mode
>   RIP: cxl_acpi_probe+0xf4/0x220
>   Call Trace:
>    platform_probe+0x4d/0x80
>    really_probe+0x106/0x370
>    device_driver_attach+0x4c/0xa0
>    bind_store+0xd0/0x100
> 
> Commit 2b3a5dabe89e ("platform/surface: acpi-notify: Check ACPI
> companion before use") fixed the same force-binding issue in another
> platform driver.  Check the companion before setting up the CXL root and
> return -ENODEV when it is absent.
> 
> Fixes: 7d4b5ca2e2cb ("cxl/acpi: Add downstream port data to cxl_port instances")
> Cc: stable@vger.kernel.org
> Signed-off-by: Jiale Yao <yaojiale02@163.com>

Applied to cxl/next:
371c2ad16242

> ---
> 
> Notes:
>     Changes in v2:
>     - Move the ACPI companion assignment immediately before its NULL check.
>     - Reorder local declarations in reverse Christmas tree order.
> 
>  drivers/cxl/acpi.c | 10 +++++++---
>  1 file changed, 7 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/cxl/acpi.c b/drivers/cxl/acpi.c
> index 3b818adbd38b..fb09a5ff48c1 100644
> --- a/drivers/cxl/acpi.c
> +++ b/drivers/cxl/acpi.c
> @@ -885,13 +885,17 @@ static int pair_cxl_resource(struct device *dev, void *data)
>  
>  static int cxl_acpi_probe(struct platform_device *pdev)
>  {
> -	int rc;
> +	struct cxl_cfmws_context ctx;
> +	struct acpi_device *adev;
>  	struct resource *cxl_res;
>  	struct cxl_root *cxl_root;
>  	struct cxl_port *root_port;
>  	struct device *host = &pdev->dev;
> -	struct acpi_device *adev = ACPI_COMPANION(host);
> -	struct cxl_cfmws_context ctx;
> +	int rc;
> +
> +	adev = ACPI_COMPANION(host);
> +	if (!adev)
> +		return -ENODEV;
>  
>  	device_lock_set_class(&pdev->dev, &cxl_root_key);
>  	rc = devm_add_action_or_reset(&pdev->dev, cxl_acpi_lock_reset_class,


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2026-09-28 23:50 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-09-26  7:16 [PATCH v2] cxl/acpi: Check ACPI companion before use Jiale Yao
2026-09-27 22:50 ` Jonathan Cameron
2026-09-28 23:50 ` Dave Jiang

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®