diff mbox series

[v4,04/26] cxl/pci: add check for validating capabilities

Message ID 20241017165225.21206-5-alejandro.lucero-palau@amd.com
State New
Headers show
Series cxl: add Type2 device support | expand

Commit Message

Lucero Palau, Alejandro Oct. 17, 2024, 4:52 p.m. UTC
From: Alejandro Lucero <alucerop@amd.com>

During CXL device initialization supported capabilities by the device
are discovered. Type3 and Type2 devices have different mandatory
capabilities and a Type2 expects a specific set including optional
capabilities.

Add a function for checking expected capabilities against those found
during initialization.

Rely on this function for validating capabilities instead of when CXL
regs are probed.

Signed-off-by: Alejandro Lucero <alucerop@amd.com>
---
 drivers/cxl/core/pci.c  | 14 ++++++++++++++
 drivers/cxl/core/regs.c |  9 ---------
 drivers/cxl/pci.c       | 17 +++++++++++++++++
 include/linux/cxl/cxl.h |  3 +++
 4 files changed, 34 insertions(+), 9 deletions(-)

Comments

Alejandro Lucero Palau Oct. 25, 2024, 10:16 a.m. UTC | #1
On 10/17/24 17:52, alejandro.lucero-palau@amd.com wrote:
> From: Alejandro Lucero <alucerop@amd.com>
>
> During CXL device initialization supported capabilities by the device
> are discovered. Type3 and Type2 devices have different mandatory
> capabilities and a Type2 expects a specific set including optional
> capabilities.
>
> Add a function for checking expected capabilities against those found
> during initialization.
>
> Rely on this function for validating capabilities instead of when CXL
> regs are probed.
>
> Signed-off-by: Alejandro Lucero <alucerop@amd.com>
> ---
>   drivers/cxl/core/pci.c  | 14 ++++++++++++++
>   drivers/cxl/core/regs.c |  9 ---------
>   drivers/cxl/pci.c       | 17 +++++++++++++++++
>   include/linux/cxl/cxl.h |  3 +++
>   4 files changed, 34 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c
> index 3d6564dbda57..fa2a5e216dc3 100644
> --- a/drivers/cxl/core/pci.c
> +++ b/drivers/cxl/core/pci.c
> @@ -8,6 +8,7 @@
>   #include <linux/pci-doe.h>
>   #include <linux/aer.h>
>   #include <linux/cxl/pci.h>
> +#include <linux/cxl/cxl.h>
>   #include <cxlpci.h>
>   #include <cxlmem.h>
>   #include <cxl.h>
> @@ -1077,3 +1078,16 @@ bool cxl_endpoint_decoder_reset_detected(struct cxl_port *port)
>   				     __cxl_endpoint_decoder_reset_detected);
>   }
>   EXPORT_SYMBOL_NS_GPL(cxl_endpoint_decoder_reset_detected, CXL);
> +
> +bool cxl_pci_check_caps(struct cxl_dev_state *cxlds, unsigned long *expected_caps,
> +			unsigned long *current_caps)
> +{
> +	if (current_caps)
> +		bitmap_copy(current_caps, cxlds->capabilities, CXL_MAX_CAPS);
> +
> +	dev_dbg(cxlds->dev, "Checking cxlds caps 0x%08lx vs expected caps 0x%08lx\n",
> +		*cxlds->capabilities, *expected_caps);
> +
> +	return bitmap_equal(cxlds->capabilities, expected_caps, CXL_MAX_CAPS);
> +}
> +EXPORT_SYMBOL_NS_GPL(cxl_pci_check_caps, CXL);
> diff --git a/drivers/cxl/core/regs.c b/drivers/cxl/core/regs.c
> index 9d63a2adfd42..6fbc5c57149e 100644
> --- a/drivers/cxl/core/regs.c
> +++ b/drivers/cxl/core/regs.c
> @@ -444,15 +444,6 @@ static int cxl_probe_regs(struct cxl_register_map *map, unsigned long *caps)
>   	case CXL_REGLOC_RBI_MEMDEV:
>   		dev_map = &map->device_map;
>   		cxl_probe_device_regs(host, base, dev_map, caps);
> -		if (!dev_map->status.valid || !dev_map->mbox.valid ||
> -		    !dev_map->memdev.valid) {
> -			dev_err(host, "registers not found: %s%s%s\n",
> -				!dev_map->status.valid ? "status " : "",
> -				!dev_map->mbox.valid ? "mbox " : "",
> -				!dev_map->memdev.valid ? "memdev " : "");
> -			return -ENXIO;
> -		}
> -
>   		dev_dbg(host, "Probing device registers...\n");
>   		break;
>   	default:
> diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
> index 6cd7ab117f80..89c8ac1a61fd 100644
> --- a/drivers/cxl/pci.c
> +++ b/drivers/cxl/pci.c
> @@ -792,6 +792,8 @@ static int cxl_event_config(struct pci_host_bridge *host_bridge,
>   static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>   {
>   	struct pci_host_bridge *host_bridge = pci_find_host_bridge(pdev->bus);
> +	DECLARE_BITMAP(expected, CXL_MAX_CAPS);
> +	DECLARE_BITMAP(found, CXL_MAX_CAPS);
>   	struct cxl_memdev_state *mds;
>   	struct cxl_dev_state *cxlds;
>   	struct cxl_register_map map;
> @@ -853,6 +855,21 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>   	if (rc)
>   		dev_dbg(&pdev->dev, "Failed to map RAS capability.\n");
>   
> +	bitmap_clear(expected, 0, BITS_PER_TYPE(unsigned long));
> +
> +	/* These are the mandatory capabilities for a Type3 device */
> +	bitmap_set(expected, CXL_DEV_CAP_HDM, 1);
> +	bitmap_set(expected, CXL_DEV_CAP_DEV_STATUS, 1);
> +	bitmap_set(expected, CXL_DEV_CAP_MAILBOX_PRIMARY, 1);
> +	bitmap_set(expected, CXL_DEV_CAP_DEV_STATUS, 1);
> +
> +	if (!cxl_pci_check_caps(cxlds, expected, found)) {
> +		dev_err(&pdev->dev,
> +			"Expected capabilities not matching with found capabilities: (%08lx - %08lx)\n",
> +			*expected, *found);
> +		return -ENXIO;
> +	}
> +


This is wrong since a Type3 could have more caps than the mandatory 
ones. I will change the check for at least the mandatory ones being 
there, and do not fail if they are.

I guess a dev_dbg showing always the found versus the expected ones 
would not harm, so adding that as well in v5.


>   	rc = cxl_await_media_ready(cxlds);
>   	if (rc == 0)
>   		cxlds->media_ready = true;
> diff --git a/include/linux/cxl/cxl.h b/include/linux/cxl/cxl.h
> index 4a4f75a86018..78653fa4daa0 100644
> --- a/include/linux/cxl/cxl.h
> +++ b/include/linux/cxl/cxl.h
> @@ -49,4 +49,7 @@ void cxl_set_dvsec(struct cxl_dev_state *cxlds, u16 dvsec);
>   void cxl_set_serial(struct cxl_dev_state *cxlds, u64 serial);
>   int cxl_set_resource(struct cxl_dev_state *cxlds, struct resource res,
>   		     enum cxl_resource);
> +bool cxl_pci_check_caps(struct cxl_dev_state *cxlds,
> +			unsigned long *expected_caps,
> +			unsigned long *current_caps);
>   #endif
Jonathan Cameron Oct. 25, 2024, 2:16 p.m. UTC | #2
On Fri, 25 Oct 2024 11:16:55 +0100
Alejandro Lucero Palau <alucerop@amd.com> wrote:

> On 10/17/24 17:52, alejandro.lucero-palau@amd.com wrote:
> > From: Alejandro Lucero <alucerop@amd.com>
> >
> > During CXL device initialization supported capabilities by the device
> > are discovered. Type3 and Type2 devices have different mandatory
> > capabilities and a Type2 expects a specific set including optional
> > capabilities.
> >
> > Add a function for checking expected capabilities against those found
> > during initialization.
> >
> > Rely on this function for validating capabilities instead of when CXL
> > regs are probed.
> >
> > Signed-off-by: Alejandro Lucero <alucerop@amd.com>
> > ---
> >   drivers/cxl/core/pci.c  | 14 ++++++++++++++
> >   drivers/cxl/core/regs.c |  9 ---------
> >   drivers/cxl/pci.c       | 17 +++++++++++++++++
> >   include/linux/cxl/cxl.h |  3 +++
> >   4 files changed, 34 insertions(+), 9 deletions(-)
> >
> > diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c
> > index 3d6564dbda57..fa2a5e216dc3 100644
> > --- a/drivers/cxl/core/pci.c
> > +++ b/drivers/cxl/core/pci.c
> > @@ -8,6 +8,7 @@
> >   #include <linux/pci-doe.h>
> >   #include <linux/aer.h>
> >   #include <linux/cxl/pci.h>
> > +#include <linux/cxl/cxl.h>
> >   #include <cxlpci.h>
> >   #include <cxlmem.h>
> >   #include <cxl.h>
> > @@ -1077,3 +1078,16 @@ bool cxl_endpoint_decoder_reset_detected(struct cxl_port *port)
> >   				     __cxl_endpoint_decoder_reset_detected);
> >   }
> >   EXPORT_SYMBOL_NS_GPL(cxl_endpoint_decoder_reset_detected, CXL);
> > +
> > +bool cxl_pci_check_caps(struct cxl_dev_state *cxlds, unsigned long *expected_caps,
> > +			unsigned long *current_caps)
> > +{
> > +	if (current_caps)
> > +		bitmap_copy(current_caps, cxlds->capabilities, CXL_MAX_CAPS);
> > +
> > +	dev_dbg(cxlds->dev, "Checking cxlds caps 0x%08lx vs expected caps 0x%08lx\n",
> > +		*cxlds->capabilities, *expected_caps);
> > +
> > +	return bitmap_equal(cxlds->capabilities, expected_caps, CXL_MAX_CAPS);
> > +}
> > +EXPORT_SYMBOL_NS_GPL(cxl_pci_check_caps, CXL);
> > diff --git a/drivers/cxl/core/regs.c b/drivers/cxl/core/regs.c
> > index 9d63a2adfd42..6fbc5c57149e 100644
> > --- a/drivers/cxl/core/regs.c
> > +++ b/drivers/cxl/core/regs.c
> > @@ -444,15 +444,6 @@ static int cxl_probe_regs(struct cxl_register_map *map, unsigned long *caps)
> >   	case CXL_REGLOC_RBI_MEMDEV:
> >   		dev_map = &map->device_map;
> >   		cxl_probe_device_regs(host, base, dev_map, caps);
> > -		if (!dev_map->status.valid || !dev_map->mbox.valid ||
> > -		    !dev_map->memdev.valid) {
> > -			dev_err(host, "registers not found: %s%s%s\n",
> > -				!dev_map->status.valid ? "status " : "",
> > -				!dev_map->mbox.valid ? "mbox " : "",
> > -				!dev_map->memdev.valid ? "memdev " : "");
> > -			return -ENXIO;
> > -		}
> > -
> >   		dev_dbg(host, "Probing device registers...\n");
> >   		break;
> >   	default:
> > diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
> > index 6cd7ab117f80..89c8ac1a61fd 100644
> > --- a/drivers/cxl/pci.c
> > +++ b/drivers/cxl/pci.c
> > @@ -792,6 +792,8 @@ static int cxl_event_config(struct pci_host_bridge *host_bridge,
> >   static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
> >   {
> >   	struct pci_host_bridge *host_bridge = pci_find_host_bridge(pdev->bus);
> > +	DECLARE_BITMAP(expected, CXL_MAX_CAPS);
> > +	DECLARE_BITMAP(found, CXL_MAX_CAPS);
> >   	struct cxl_memdev_state *mds;
> >   	struct cxl_dev_state *cxlds;
> >   	struct cxl_register_map map;
> > @@ -853,6 +855,21 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
> >   	if (rc)
> >   		dev_dbg(&pdev->dev, "Failed to map RAS capability.\n");
> >   
> > +	bitmap_clear(expected, 0, BITS_PER_TYPE(unsigned long));
> > +
> > +	/* These are the mandatory capabilities for a Type3 device */
> > +	bitmap_set(expected, CXL_DEV_CAP_HDM, 1);
> > +	bitmap_set(expected, CXL_DEV_CAP_DEV_STATUS, 1);
> > +	bitmap_set(expected, CXL_DEV_CAP_MAILBOX_PRIMARY, 1);
> > +	bitmap_set(expected, CXL_DEV_CAP_DEV_STATUS, 1);
> > +
> > +	if (!cxl_pci_check_caps(cxlds, expected, found)) {
> > +		dev_err(&pdev->dev,
> > +			"Expected capabilities not matching with found capabilities: (%08lx - %08lx)\n",
> > +			*expected, *found);
> > +		return -ENXIO;
> > +	}
> > +  
> 
> 
> This is wrong since a Type3 could have more caps than the mandatory 
> ones. I will change the check for at least the mandatory ones being 
> there, and do not fail if they are.
> 
> I guess a dev_dbg showing always the found versus the expected ones 
> would not harm, so adding that as well in v5.

I'd also make it clear that we only check for caps that are actually
used by Linux drivers. I don't fancy having to fix emulation up to support
something random that no software cares about.  Will get there
eventually but it's not a priority.

Jonathan

> 
> 
> >   	rc = cxl_await_media_ready(cxlds);
> >   	if (rc == 0)
> >   		cxlds->media_ready = true;
> > diff --git a/include/linux/cxl/cxl.h b/include/linux/cxl/cxl.h
> > index 4a4f75a86018..78653fa4daa0 100644
> > --- a/include/linux/cxl/cxl.h
> > +++ b/include/linux/cxl/cxl.h
> > @@ -49,4 +49,7 @@ void cxl_set_dvsec(struct cxl_dev_state *cxlds, u16 dvsec);
> >   void cxl_set_serial(struct cxl_dev_state *cxlds, u64 serial);
> >   int cxl_set_resource(struct cxl_dev_state *cxlds, struct resource res,
> >   		     enum cxl_resource);
> > +bool cxl_pci_check_caps(struct cxl_dev_state *cxlds,
> > +			unsigned long *expected_caps,
> > +			unsigned long *current_caps);
> >   #endif  
>
diff mbox series

Patch

diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c
index 3d6564dbda57..fa2a5e216dc3 100644
--- a/drivers/cxl/core/pci.c
+++ b/drivers/cxl/core/pci.c
@@ -8,6 +8,7 @@ 
 #include <linux/pci-doe.h>
 #include <linux/aer.h>
 #include <linux/cxl/pci.h>
+#include <linux/cxl/cxl.h>
 #include <cxlpci.h>
 #include <cxlmem.h>
 #include <cxl.h>
@@ -1077,3 +1078,16 @@  bool cxl_endpoint_decoder_reset_detected(struct cxl_port *port)
 				     __cxl_endpoint_decoder_reset_detected);
 }
 EXPORT_SYMBOL_NS_GPL(cxl_endpoint_decoder_reset_detected, CXL);
+
+bool cxl_pci_check_caps(struct cxl_dev_state *cxlds, unsigned long *expected_caps,
+			unsigned long *current_caps)
+{
+	if (current_caps)
+		bitmap_copy(current_caps, cxlds->capabilities, CXL_MAX_CAPS);
+
+	dev_dbg(cxlds->dev, "Checking cxlds caps 0x%08lx vs expected caps 0x%08lx\n",
+		*cxlds->capabilities, *expected_caps);
+
+	return bitmap_equal(cxlds->capabilities, expected_caps, CXL_MAX_CAPS);
+}
+EXPORT_SYMBOL_NS_GPL(cxl_pci_check_caps, CXL);
diff --git a/drivers/cxl/core/regs.c b/drivers/cxl/core/regs.c
index 9d63a2adfd42..6fbc5c57149e 100644
--- a/drivers/cxl/core/regs.c
+++ b/drivers/cxl/core/regs.c
@@ -444,15 +444,6 @@  static int cxl_probe_regs(struct cxl_register_map *map, unsigned long *caps)
 	case CXL_REGLOC_RBI_MEMDEV:
 		dev_map = &map->device_map;
 		cxl_probe_device_regs(host, base, dev_map, caps);
-		if (!dev_map->status.valid || !dev_map->mbox.valid ||
-		    !dev_map->memdev.valid) {
-			dev_err(host, "registers not found: %s%s%s\n",
-				!dev_map->status.valid ? "status " : "",
-				!dev_map->mbox.valid ? "mbox " : "",
-				!dev_map->memdev.valid ? "memdev " : "");
-			return -ENXIO;
-		}
-
 		dev_dbg(host, "Probing device registers...\n");
 		break;
 	default:
diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
index 6cd7ab117f80..89c8ac1a61fd 100644
--- a/drivers/cxl/pci.c
+++ b/drivers/cxl/pci.c
@@ -792,6 +792,8 @@  static int cxl_event_config(struct pci_host_bridge *host_bridge,
 static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	struct pci_host_bridge *host_bridge = pci_find_host_bridge(pdev->bus);
+	DECLARE_BITMAP(expected, CXL_MAX_CAPS);
+	DECLARE_BITMAP(found, CXL_MAX_CAPS);
 	struct cxl_memdev_state *mds;
 	struct cxl_dev_state *cxlds;
 	struct cxl_register_map map;
@@ -853,6 +855,21 @@  static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (rc)
 		dev_dbg(&pdev->dev, "Failed to map RAS capability.\n");
 
+	bitmap_clear(expected, 0, BITS_PER_TYPE(unsigned long));
+
+	/* These are the mandatory capabilities for a Type3 device */
+	bitmap_set(expected, CXL_DEV_CAP_HDM, 1);
+	bitmap_set(expected, CXL_DEV_CAP_DEV_STATUS, 1);
+	bitmap_set(expected, CXL_DEV_CAP_MAILBOX_PRIMARY, 1);
+	bitmap_set(expected, CXL_DEV_CAP_DEV_STATUS, 1);
+
+	if (!cxl_pci_check_caps(cxlds, expected, found)) {
+		dev_err(&pdev->dev,
+			"Expected capabilities not matching with found capabilities: (%08lx - %08lx)\n",
+			*expected, *found);
+		return -ENXIO;
+	}
+
 	rc = cxl_await_media_ready(cxlds);
 	if (rc == 0)
 		cxlds->media_ready = true;
diff --git a/include/linux/cxl/cxl.h b/include/linux/cxl/cxl.h
index 4a4f75a86018..78653fa4daa0 100644
--- a/include/linux/cxl/cxl.h
+++ b/include/linux/cxl/cxl.h
@@ -49,4 +49,7 @@  void cxl_set_dvsec(struct cxl_dev_state *cxlds, u16 dvsec);
 void cxl_set_serial(struct cxl_dev_state *cxlds, u64 serial);
 int cxl_set_resource(struct cxl_dev_state *cxlds, struct resource res,
 		     enum cxl_resource);
+bool cxl_pci_check_caps(struct cxl_dev_state *cxlds,
+			unsigned long *expected_caps,
+			unsigned long *current_caps);
 #endif