[PATCH v7 0/2] cxl: Export cxl1.1 device link status to sysfs

[PATCH v7 0/2] cxl: Export cxl1.1 device link status to sysfs

[Kobayashi,Daisuke]

Export cxl1.1 device link status register value to pci device sysfs.

CXL devices are extensions of PCIe. Therefore, from CXL2.0 onwards,
the link status can be output in the same way as traditional PCIe.
However, unlike devices from CXL2.0 onwards, CXL1.1 requires a
different method to obtain the link status from traditional PCIe.
This is because the link status of the CXL1.1 device is not mapped
in the configuration space (as per cxl3.0 specification 8.1).
Instead, the configuration space containing the link status is mapped
to the memory mapped register region (as per cxl3.0 specification 8.2,
Table 8-18). Therefore, the current lspci has a problem where it does
not display the link status of the CXL1.1 device. 
Solve these issues with sysfs attributes to export the status
registers hidden in the RCRB.

The procedure is as follows:
First, obtain the RCRB address within the cxl driver, then access 
the configuration space. Next, output the link status information from
the configuration space to sysfs. Ultimately, the expectation is that
this sysfs file will be consumed by PCI user tools to utilize link status
information.


Changes
v1[1] -> v2:
- Modified to perform rcrb access within the CXL driver.
- Added new attributes to the sysfs of the PCI device.
- Output the link status information to the sysfs of the PCI device.
- Retrieve information from sysfs as the source when displaying information in lspci.

v2[2] -> v3:
- Fix unnecessary initialization and wrong types (Bjohn).
- Create a helper function for getting a PCIe capability offset (Bjohn).
- Move platform-specific implementation to the lib directory in pciutils (Martin).

v3[3] -> v4:
- RCRB register values are read once and cached.
- Added a new attribute to the sysfs of the PCI device.
- Separate lspci implementation from this patch.

v4[4] -> v5:
- Use macros for bitwise operations
- Fix RCRB access to use cxl_memdev

v5[5] -> v6:
- Add and use masks for RCRB register values

v6[6] -> v7:
- Fix comments on white space inline

[1]
https://lore.kernel.org/linux-cxl/20231220050738.178481-1-kobayashi.da-06@fujitsu.com/
[2]
https://lore.kernel.org/linux-cxl/20240227083313.87699-1-kobayashi.da-06@fujitsu.com/
[3]
https://lore.kernel.org/linux-cxl/20240312080559.14904-1-kobayashi.da-06@fujitsu.com/
[4]
https://lore.kernel.org/linux-cxl/20240409073528.13214-1-kobayashi.da-06@fujitsu.com/
[5]
https://lore.kernel.org/linux-cxl/20240412070715.16160-1-kobayashi.da-06@fujitsu.com/
[6]
https://lore.kernel.org/linux-cxl/20240424050102.26788-1-kobayashi.da-06@fujitsu.com/

Kobayashi,Daisuke (2):
  cxl: Add rcd_regs to cxl_rcrb_info
  cxl/pci: Add sysfs attribute for CXL 1.1 device link statu

 drivers/cxl/core/core.h |   4 ++
 drivers/cxl/core/regs.c |  16 +++++++
 drivers/cxl/cxl.h       |   3 ++
 drivers/cxl/pci.c       | 101 ++++++++++++++++++++++++++++++++++++++++
 4 files changed, 124 insertions(+)

-- 
2.44.0

[PATCH v7 1/2] cxl/core/regs: Add rcd_regs initialization at __rcrb_to_component()

[Kobayashi,Daisuke]

Add rcd_regs and its initialization at __rcrb_to_component() to cache
the cxl1.1 device link status information. Reduce access to the memory
map area where the RCRB is located by caching the cxl1.1 device
link status information.

Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com>
---
 drivers/cxl/core/core.h |  4 ++++
 drivers/cxl/core/regs.c | 16 ++++++++++++++++
 drivers/cxl/cxl.h       |  3 +++
 3 files changed, 23 insertions(+)

diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
index 3b64fb1b9ed0..42e3483b4a14 100644
--- a/drivers/cxl/core/core.h
+++ b/drivers/cxl/core/core.h
@@ -75,6 +75,10 @@ resource_size_t __rcrb_to_component(struct device *dev,
 				    enum cxl_rcrb which);
 u16 cxl_rcrb_to_aer(struct device *dev, resource_size_t rcrb);
 
+#define PCI_RCRB_CAP_LIST_ID_MASK	GENMASK(7, 0)
+#define PCI_RCRB_CAP_HDR_ID_MASK	GENMASK(7, 0)
+#define PCI_RCRB_CAP_HDR_NEXT_MASK	GENMASK(15, 8)
+
 extern struct rw_semaphore cxl_dpa_rwsem;
 extern struct rw_semaphore cxl_region_rwsem;
 
diff --git a/drivers/cxl/core/regs.c b/drivers/cxl/core/regs.c
index 372786f80955..1ad58c464488 100644
--- a/drivers/cxl/core/regs.c
+++ b/drivers/cxl/core/regs.c
@@ -514,6 +514,8 @@ resource_size_t __rcrb_to_component(struct device *dev, struct cxl_rcrb_info *ri
 	u32 bar0, bar1;
 	u16 cmd;
 	u32 id;
+	u16 offset;
+	u32 cap_hdr;
 
 	if (which == CXL_RCRB_UPSTREAM)
 		rcrb += SZ_4K;
@@ -537,6 +539,20 @@ resource_size_t __rcrb_to_component(struct device *dev, struct cxl_rcrb_info *ri
 	cmd = readw(addr + PCI_COMMAND);
 	bar0 = readl(addr + PCI_BASE_ADDRESS_0);
 	bar1 = readl(addr + PCI_BASE_ADDRESS_1);
+	offset = FIELD_GET(PCI_RCRB_CAP_LIST_ID_MASK, readw(addr + PCI_CAPABILITY_LIST));
+	cap_hdr = readl(addr + offset);
+	while ((FIELD_GET(PCI_RCRB_CAP_HDR_ID_MASK, cap_hdr)) != PCI_CAP_ID_EXP) {
+		offset = FIELD_GET(PCI_RCRB_CAP_HDR_NEXT_MASK, cap_hdr);
+		if (offset == 0 || offset > SZ_4K)
+			break;
+		cap_hdr = readl(addr + offset);
+	}
+	if (offset) {
+		ri->rcd_lnkcap = readl(addr + offset + PCI_EXP_LNKCAP);
+		ri->rcd_lnkctrl = readl(addr + offset + PCI_EXP_LNKCTL);
+		ri->rcd_lnkstatus = readl(addr + offset + PCI_EXP_LNKSTA);
+	}
+
 	iounmap(addr);
 	release_mem_region(rcrb, SZ_4K);
 
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index 003feebab79b..808818ccc255 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -646,6 +646,9 @@ cxl_find_dport_by_dev(struct cxl_port *port, const struct device *dport_dev)
 
 struct cxl_rcrb_info {
 	resource_size_t base;
+	u16 rcd_lnkstatus;
+	u16 rcd_lnkctrl;
+	u32 rcd_lnkcap;
 	u16 aer_cap;
 };
 
-- 
2.44.0

[PATCH v7 2/2] cxl/pci: Add sysfs attribute for CXL 1.1 device link status

[Kobayashi,Daisuke]

Add sysfs attribute for CXL 1.1 device link status to the cxl pci device.

In CXL1.1, the link status of the device is included in the RCRB mapped to
the memory mapped register area. Critically, that arrangement makes the
link status and control registers invisible to existing PCI user tooling.

Export those registers via sysfs with the expectation that PCI user
tooling will alternatively look for these sysfs files when attempting to
access to these CXL 1.1 endpoints registers.

Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com>
---
 drivers/cxl/pci.c | 101 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 101 insertions(+)

diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
index 2ff361e756d6..c10797adde2c 100644
--- a/drivers/cxl/pci.c
+++ b/drivers/cxl/pci.c
@@ -786,6 +786,106 @@ static int cxl_event_config(struct pci_host_bridge *host_bridge,
 	return 0;
 }
 
+static ssize_t rcd_link_cap_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
+	struct cxl_memdev *cxlmd = cxlds->cxlmd;
+	struct device *endpoint_parent;
+	struct cxl_dport *dport;
+	struct cxl_port *port;
+
+	port = cxl_mem_find_port(cxlmd, &dport);
+	if (!port)
+		return -EINVAL;
+
+	endpoint_parent = port->uport_dev;
+	if (!endpoint_parent)
+		return -ENXIO;
+
+	guard(device)(endpoint_parent);
+	if (!endpoint_parent->driver)
+		return -ENXIO;
+
+	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkcap);
+}
+static DEVICE_ATTR_RO(rcd_link_cap);
+
+static ssize_t rcd_link_ctrl_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
+	struct cxl_memdev *cxlmd = cxlds->cxlmd;
+	struct device *endpoint_parent;
+	struct cxl_dport *dport;
+	struct cxl_port *port;
+
+	port = cxl_mem_find_port(cxlmd, &dport);
+	if (!port)
+		return -EINVAL;
+
+	endpoint_parent = port->uport_dev;
+	if (!endpoint_parent)
+		return -ENXIO;
+
+	guard(device)(endpoint_parent);
+	if (!endpoint_parent->driver)
+		return -ENXIO;
+
+	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkctrl);
+}
+static DEVICE_ATTR_RO(rcd_link_ctrl);
+
+static ssize_t rcd_link_status_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
+	struct cxl_memdev *cxlmd = cxlds->cxlmd;
+	struct device *endpoint_parent;
+	struct cxl_dport *dport;
+	struct cxl_port *port;
+
+	port = cxl_mem_find_port(cxlmd, &dport);
+	if (!port)
+		return -EINVAL;
+
+	endpoint_parent = port->uport_dev;
+	if (!endpoint_parent)
+		return -ENXIO;
+
+	guard(device)(endpoint_parent);
+	if (!endpoint_parent->driver)
+		return -ENXIO;
+
+	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkstatus);
+}
+static DEVICE_ATTR_RO(rcd_link_status);
+
+static struct attribute *cxl_rcd_attrs[] = {
+		&dev_attr_rcd_link_cap.attr,
+		&dev_attr_rcd_link_ctrl.attr,
+		&dev_attr_rcd_link_status.attr,
+		NULL
+};
+
+static umode_t cxl_rcd_visible(struct kobject *kobj,
+					  struct attribute *a, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct pci_dev *pdev = to_pci_dev(dev);
+
+	if (is_cxl_restricted(pdev))
+		return a->mode;
+
+	return 0;
+}
+
+static struct attribute_group cxl_rcd_group = {
+		.attrs = cxl_rcd_attrs,
+		.is_visible = cxl_rcd_visible,
+};
+__ATTRIBUTE_GROUPS(cxl_rcd);
+
 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);
@@ -969,6 +1069,7 @@ static struct pci_driver cxl_pci_driver = {
 	.id_table		= cxl_mem_pci_tbl,
 	.probe			= cxl_pci_probe,
 	.err_handler		= &cxl_error_handlers,
+	.dev_groups		= cxl_rcd_groups,
 	.driver	= {
 		.probe_type	= PROBE_PREFER_ASYNCHRONOUS,
 	},
-- 
2.44.0

RE: [PATCH v7 0/2] cxl: Export cxl1.1 device link status to sysfs

[Daisuke Kobayashi (Fujitsu)]

Kobayashi Daisuke wrote:
> 
> Export cxl1.1 device link status register value to pci device sysfs.
> 
> CXL devices are extensions of PCIe. Therefore, from CXL2.0 onwards,
> the link status can be output in the same way as traditional PCIe.
> However, unlike devices from CXL2.0 onwards, CXL1.1 requires a
> different method to obtain the link status from traditional PCIe.
> This is because the link status of the CXL1.1 device is not mapped
> in the configuration space (as per cxl3.0 specification 8.1).
> Instead, the configuration space containing the link status is mapped
> to the memory mapped register region (as per cxl3.0 specification 8.2,
> Table 8-18). Therefore, the current lspci has a problem where it does
> not display the link status of the CXL1.1 device.
> Solve these issues with sysfs attributes to export the status
> registers hidden in the RCRB.
> 
> The procedure is as follows:
> First, obtain the RCRB address within the cxl driver, then access
> the configuration space. Next, output the link status information from
> the configuration space to sysfs. Ultimately, the expectation is that
> this sysfs file will be consumed by PCI user tools to utilize link status
> information.
> 
> 
> Changes
> v1[1] -> v2:
> - Modified to perform rcrb access within the CXL driver.
> - Added new attributes to the sysfs of the PCI device.
> - Output the link status information to the sysfs of the PCI device.
> - Retrieve information from sysfs as the source when displaying information in
> lspci.
> 
> v2[2] -> v3:
> - Fix unnecessary initialization and wrong types (Bjohn).
> - Create a helper function for getting a PCIe capability offset (Bjohn).
> - Move platform-specific implementation to the lib directory in pciutils
> (Martin).
> 
> v3[3] -> v4:
> - RCRB register values are read once and cached.
> - Added a new attribute to the sysfs of the PCI device.
> - Separate lspci implementation from this patch.
> 
> v4[4] -> v5:
> - Use macros for bitwise operations
> - Fix RCRB access to use cxl_memdev
> 
> v5[5] -> v6:
> - Add and use masks for RCRB register values
> 
> v6[6] -> v7:
> - Fix comments on white space inline
> 
> [1]
> https://lore.kernel.org/linux-cxl/20231220050738.178481-1-kobayashi.da-06
> @fujitsu.com/
> [2]
> https://lore.kernel.org/linux-cxl/20240227083313.87699-1-kobayashi.da-06@
> fujitsu.com/
> [3]
> https://lore.kernel.org/linux-cxl/20240312080559.14904-1-kobayashi.da-06@
> fujitsu.com/
> [4]
> https://lore.kernel.org/linux-cxl/20240409073528.13214-1-kobayashi.da-06@
> fujitsu.com/
> [5]
> https://lore.kernel.org/linux-cxl/20240412070715.16160-1-kobayashi.da-06@
> fujitsu.com/
> [6]
> https://lore.kernel.org/linux-cxl/20240424050102.26788-1-kobayashi.da-06@
> fujitsu.com/
> 
> Kobayashi,Daisuke (2):
>   cxl: Add rcd_regs to cxl_rcrb_info
>   cxl/pci: Add sysfs attribute for CXL 1.1 device link statu
> 
>  drivers/cxl/core/core.h |   4 ++
>  drivers/cxl/core/regs.c |  16 +++++++
>  drivers/cxl/cxl.h       |   3 ++
>  drivers/cxl/pci.c       | 101
> ++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 124 insertions(+)
> 
> --
> 2.44.0

Hi all.
Gentle ping.
Is there anything I can do to help with merging the patch? 
I believe I have addressed all of the points raised in the review.

Re: [PATCH v7 1/2] cxl/core/regs: Add rcd_regs initialization at __rcrb_to_component()

[Jonathan Cameron]

On Fri, 10 May 2024 16:37:09 +0900
"Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com> wrote:

> Add rcd_regs and its initialization at __rcrb_to_component() to cache

This rcd_regs doesn't immediately align with what is in the code. I'd just
call it out as

Add caching of Link Status, Link Control and Link Capability registers for
a Restricted CXL Device to struct cxl_rcrb_info.

> the cxl1.1 device link status information. Reduce access to the memory
> map area where the RCRB is located by caching the cxl1.1 device
> link status information.

Why do we care about accessing that memory mapped area?
Avoiding the walk to find the offset might be an alternative as then
we could directly read these registers when needed.
So handle these similarly to aer_cap.

That will allow for them changing at runtime without carefully needing
to update the cached values.

> 
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

New day, new comments :( 

> Signed-off-by: "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com>
> ---
>  drivers/cxl/core/core.h |  4 ++++
>  drivers/cxl/core/regs.c | 16 ++++++++++++++++
>  drivers/cxl/cxl.h       |  3 +++
>  3 files changed, 23 insertions(+)
> 
> diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
> index 3b64fb1b9ed0..42e3483b4a14 100644
> --- a/drivers/cxl/core/core.h
> +++ b/drivers/cxl/core/core.h
> @@ -75,6 +75,10 @@ resource_size_t __rcrb_to_component(struct device *dev,
>  				    enum cxl_rcrb which);
>  u16 cxl_rcrb_to_aer(struct device *dev, resource_size_t rcrb);
>  
> +#define PCI_RCRB_CAP_LIST_ID_MASK	GENMASK(7, 0)
> +#define PCI_RCRB_CAP_HDR_ID_MASK	GENMASK(7, 0)
> +#define PCI_RCRB_CAP_HDR_NEXT_MASK	GENMASK(15, 8)
> +
>  extern struct rw_semaphore cxl_dpa_rwsem;
>  extern struct rw_semaphore cxl_region_rwsem;
>  
> diff --git a/drivers/cxl/core/regs.c b/drivers/cxl/core/regs.c
> index 372786f80955..1ad58c464488 100644
> --- a/drivers/cxl/core/regs.c
> +++ b/drivers/cxl/core/regs.c
> @@ -514,6 +514,8 @@ resource_size_t __rcrb_to_component(struct device *dev, struct cxl_rcrb_info *ri
>  	u32 bar0, bar1;
>  	u16 cmd;
>  	u32 id;
> +	u16 offset;
> +	u32 cap_hdr;
>  
>  	if (which == CXL_RCRB_UPSTREAM)
>  		rcrb += SZ_4K;
> @@ -537,6 +539,20 @@ resource_size_t __rcrb_to_component(struct device *dev, struct cxl_rcrb_info *ri
>  	cmd = readw(addr + PCI_COMMAND);
>  	bar0 = readl(addr + PCI_BASE_ADDRESS_0);
>  	bar1 = readl(addr + PCI_BASE_ADDRESS_1);
> +	offset = FIELD_GET(PCI_RCRB_CAP_LIST_ID_MASK, readw(addr + PCI_CAPABILITY_LIST));
> +	cap_hdr = readl(addr + offset);
> +	while ((FIELD_GET(PCI_RCRB_CAP_HDR_ID_MASK, cap_hdr)) != PCI_CAP_ID_EXP) {
> +		offset = FIELD_GET(PCI_RCRB_CAP_HDR_NEXT_MASK, cap_hdr);
> +		if (offset == 0 || offset > SZ_4K)
> +			break;
> +		cap_hdr = readl(addr + offset);
> +	}
> +	if (offset) {
> +		ri->rcd_lnkcap = readl(addr + offset + PCI_EXP_LNKCAP);
> +		ri->rcd_lnkctrl = readl(addr + offset + PCI_EXP_LNKCTL);
> +		ri->rcd_lnkstatus = readl(addr + offset + PCI_EXP_LNKSTA);
> +	}
> +
>  	iounmap(addr);
>  	release_mem_region(rcrb, SZ_4K);
>  
> diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
> index 003feebab79b..808818ccc255 100644
> --- a/drivers/cxl/cxl.h
> +++ b/drivers/cxl/cxl.h
> @@ -646,6 +646,9 @@ cxl_find_dport_by_dev(struct cxl_port *port, const struct device *dport_dev)
>  
>  struct cxl_rcrb_info {
>  	resource_size_t base;
> +	u16 rcd_lnkstatus;
> +	u16 rcd_lnkctrl;
> +	u32 rcd_lnkcap;
>  	u16 aer_cap;
>  };
>  

Re: [PATCH v7 2/2] cxl/pci: Add sysfs attribute for CXL 1.1 device link status

[Jonathan Cameron]

On Fri, 10 May 2024 16:37:10 +0900
"Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com> wrote:

> Add sysfs attribute for CXL 1.1 device link status to the cxl pci device.
> 
> In CXL1.1, the link status of the device is included in the RCRB mapped to
> the memory mapped register area. Critically, that arrangement makes the
> link status and control registers invisible to existing PCI user tooling.
> 
> Export those registers via sysfs with the expectation that PCI user
> tooling will alternatively look for these sysfs files when attempting to
> access to these CXL 1.1 endpoints registers.

The lspci support raced with this series and has landed.

https://github.com/pciutils/pciutils/commit/49efa87fcce4f7d5b351238668ae1d4491802b88

A few follow up comments on suggestion to cache the offset, not the register value
(which might change over time)
> 
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Signed-off-by: "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com>
> ---
>  drivers/cxl/pci.c | 101 ++++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 101 insertions(+)
> 
> diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
> index 2ff361e756d6..c10797adde2c 100644
> --- a/drivers/cxl/pci.c
> +++ b/drivers/cxl/pci.c
> @@ -786,6 +786,106 @@ static int cxl_event_config(struct pci_host_bridge *host_bridge,
>  	return 0;
>  }
>  
> +static ssize_t rcd_link_cap_show(struct device *dev,
> +				   struct device_attribute *attr, char *buf)
> +{
> +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> +	struct device *endpoint_parent;
> +	struct cxl_dport *dport;
> +	struct cxl_port *port;
> +
> +	port = cxl_mem_find_port(cxlmd, &dport);
> +	if (!port)
> +		return -EINVAL;
> +
> +	endpoint_parent = port->uport_dev;
> +	if (!endpoint_parent)
> +		return -ENXIO;
> +
> +	guard(device)(endpoint_parent);
> +	if (!endpoint_parent->driver)
> +		return -ENXIO;
> +
> +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkcap);

If you follow the suggestion of caching the offset, not the register
content then, these are all very similar. Define one common
function which takes a u8 offset and call that from each of the
functions.

If you stick to separate caches (and ensure they are up to date),
then you can add a utility function to take the dev here and
get you to the rcrb structure.  Then you can call that and only
have the final line in each instances of this code.

> +}
> +static DEVICE_ATTR_RO(rcd_link_cap);
> +
> +static ssize_t rcd_link_ctrl_show(struct device *dev,
> +				   struct device_attribute *attr, char *buf)
> +{
> +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> +	struct device *endpoint_parent;
> +	struct cxl_dport *dport;
> +	struct cxl_port *port;
> +
> +	port = cxl_mem_find_port(cxlmd, &dport);
> +	if (!port)
> +		return -EINVAL;
> +
> +	endpoint_parent = port->uport_dev;
> +	if (!endpoint_parent)
> +		return -ENXIO;
> +
> +	guard(device)(endpoint_parent);
> +	if (!endpoint_parent->driver)
> +		return -ENXIO;
> +
> +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkctrl);
> +}
> +static DEVICE_ATTR_RO(rcd_link_ctrl);
> +
> +static ssize_t rcd_link_status_show(struct device *dev,
> +				   struct device_attribute *attr, char *buf)
> +{
> +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> +	struct device *endpoint_parent;
> +	struct cxl_dport *dport;
> +	struct cxl_port *port;
> +
> +	port = cxl_mem_find_port(cxlmd, &dport);
> +	if (!port)
> +		return -EINVAL;
> +
> +	endpoint_parent = port->uport_dev;
> +	if (!endpoint_parent)
> +		return -ENXIO;
> +
> +	guard(device)(endpoint_parent);
> +	if (!endpoint_parent->driver)
> +		return -ENXIO;
> +
> +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkstatus);
> +}
> +static DEVICE_ATTR_RO(rcd_link_status);
> +
> +static struct attribute *cxl_rcd_attrs[] = {
> +		&dev_attr_rcd_link_cap.attr,
> +		&dev_attr_rcd_link_ctrl.attr,
> +		&dev_attr_rcd_link_status.attr,
> +		NULL
> +};
> +
> +static umode_t cxl_rcd_visible(struct kobject *kobj,
> +					  struct attribute *a, int n)
> +{
> +	struct device *dev = kobj_to_dev(kobj);
> +	struct pci_dev *pdev = to_pci_dev(dev);
> +
> +	if (is_cxl_restricted(pdev))
> +		return a->mode;
> +
> +	return 0;
> +}
> +
> +static struct attribute_group cxl_rcd_group = {
> +		.attrs = cxl_rcd_attrs,
> +		.is_visible = cxl_rcd_visible,
> +};
> +__ATTRIBUTE_GROUPS(cxl_rcd);
> +
>  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);
> @@ -969,6 +1069,7 @@ static struct pci_driver cxl_pci_driver = {
>  	.id_table		= cxl_mem_pci_tbl,
>  	.probe			= cxl_pci_probe,
>  	.err_handler		= &cxl_error_handlers,
> +	.dev_groups		= cxl_rcd_groups,
>  	.driver	= {
>  		.probe_type	= PROBE_PREFER_ASYNCHRONOUS,
>  	},

RE: [PATCH v7 2/2] cxl/pci: Add sysfs attribute for CXL 1.1 device link status

[Daisuke Kobayashi (Fujitsu)]

Jonathan Cameron wrote:
> On Fri, 10 May 2024 16:37:10 +0900
> "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com> wrote:
> 
> > Add sysfs attribute for CXL 1.1 device link status to the cxl pci device.
> >
> > In CXL1.1, the link status of the device is included in the RCRB mapped to
> > the memory mapped register area. Critically, that arrangement makes the
> > link status and control registers invisible to existing PCI user tooling.
> >
> > Export those registers via sysfs with the expectation that PCI user
> > tooling will alternatively look for these sysfs files when attempting to
> > access to these CXL 1.1 endpoints registers.
> 
> The lspci support raced with this series and has landed.
> 
> https://github.com/pciutils/pciutils/commit/49efa87fcce4f7d5b351238668ae
> 1d4491802b88
> 
> A few follow up comments on suggestion to cache the offset, not the register
> value
> (which might change over time)
> >
> > Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> > Signed-off-by: "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com>
> > ---
> >  drivers/cxl/pci.c | 101
> ++++++++++++++++++++++++++++++++++++++++++++++
> >  1 file changed, 101 insertions(+)
> >
> > diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
> > index 2ff361e756d6..c10797adde2c 100644
> > --- a/drivers/cxl/pci.c
> > +++ b/drivers/cxl/pci.c
> > @@ -786,6 +786,106 @@ static int cxl_event_config(struct pci_host_bridge
> *host_bridge,
> >  	return 0;
> >  }
> >
> > +static ssize_t rcd_link_cap_show(struct device *dev,
> > +				   struct device_attribute *attr, char *buf)
> > +{
> > +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> > +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> > +	struct device *endpoint_parent;
> > +	struct cxl_dport *dport;
> > +	struct cxl_port *port;
> > +
> > +	port = cxl_mem_find_port(cxlmd, &dport);
> > +	if (!port)
> > +		return -EINVAL;
> > +
> > +	endpoint_parent = port->uport_dev;
> > +	if (!endpoint_parent)
> > +		return -ENXIO;
> > +
> > +	guard(device)(endpoint_parent);
> > +	if (!endpoint_parent->driver)
> > +		return -ENXIO;
> > +
> > +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkcap);
> 
> If you follow the suggestion of caching the offset, not the register
> content then, these are all very similar. Define one common
> function which takes a u8 offset and call that from each of the
> functions.
> 
> If you stick to separate caches (and ensure they are up to date),
> then you can add a utility function to take the dev here and
> get you to the rcrb structure.  Then you can call that and only
> have the final line in each instances of this code.
> 

Thank you for your comment.

To confirm my understanding of the proposed implementation approach, 
I would like to share my interpretation:

- The values stored in `cxl_rcrb_info` should be the offsets of the registers, 
rather than the register values themselves.
- Similar to aer, the values should be read from the memory-mapped region 
using the offset when they are needed.
- To achieve this, the offsets (like `cxl_rcrb_pcie_caps`) will be stored in `cxl_rcrb_info` 
during initialization, and the memory-mapped region will be accessed within the `rcd_*_show()`
functions where the values are actually used.
- Accessing the memory-mapped region only requires the `struct device *dev`, 
so a common utility function will be created and used for this purpose.

If the cost of accessing the memory-mapped region at runtime is not high, especially considering 
the avoidance of the walk to find the offset, then I believe it is more reasonable to access 
the memory-mapped region at runtime.

Could you let me know if there are any errors in my understanding.

If there are no issues, I will update the implementation with this approach 
and submit a v8 patch (hopefully tomorrow).

> > +}
> > +static DEVICE_ATTR_RO(rcd_link_cap);
> > +
> > +static ssize_t rcd_link_ctrl_show(struct device *dev,
> > +				   struct device_attribute *attr, char *buf)
> > +{
> > +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> > +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> > +	struct device *endpoint_parent;
> > +	struct cxl_dport *dport;
> > +	struct cxl_port *port;
> > +
> > +	port = cxl_mem_find_port(cxlmd, &dport);
> > +	if (!port)
> > +		return -EINVAL;
> > +
> > +	endpoint_parent = port->uport_dev;
> > +	if (!endpoint_parent)
> > +		return -ENXIO;
> > +
> > +	guard(device)(endpoint_parent);
> > +	if (!endpoint_parent->driver)
> > +		return -ENXIO;
> > +
> > +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkctrl);
> > +}
> > +static DEVICE_ATTR_RO(rcd_link_ctrl);
> > +
> > +static ssize_t rcd_link_status_show(struct device *dev,
> > +				   struct device_attribute *attr, char *buf)
> > +{
> > +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> > +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> > +	struct device *endpoint_parent;
> > +	struct cxl_dport *dport;
> > +	struct cxl_port *port;
> > +
> > +	port = cxl_mem_find_port(cxlmd, &dport);
> > +	if (!port)
> > +		return -EINVAL;
> > +
> > +	endpoint_parent = port->uport_dev;
> > +	if (!endpoint_parent)
> > +		return -ENXIO;
> > +
> > +	guard(device)(endpoint_parent);
> > +	if (!endpoint_parent->driver)
> > +		return -ENXIO;
> > +
> > +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkstatus);
> > +}
> > +static DEVICE_ATTR_RO(rcd_link_status);
> > +
> > +static struct attribute *cxl_rcd_attrs[] = {
> > +		&dev_attr_rcd_link_cap.attr,
> > +		&dev_attr_rcd_link_ctrl.attr,
> > +		&dev_attr_rcd_link_status.attr,
> > +		NULL
> > +};
> > +
> > +static umode_t cxl_rcd_visible(struct kobject *kobj,
> > +					  struct attribute *a, int n)
> > +{
> > +	struct device *dev = kobj_to_dev(kobj);
> > +	struct pci_dev *pdev = to_pci_dev(dev);
> > +
> > +	if (is_cxl_restricted(pdev))
> > +		return a->mode;
> > +
> > +	return 0;
> > +}
> > +
> > +static struct attribute_group cxl_rcd_group = {
> > +		.attrs = cxl_rcd_attrs,
> > +		.is_visible = cxl_rcd_visible,
> > +};
> > +__ATTRIBUTE_GROUPS(cxl_rcd);
> > +
> >  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);
> > @@ -969,6 +1069,7 @@ static struct pci_driver cxl_pci_driver = {
> >  	.id_table		= cxl_mem_pci_tbl,
> >  	.probe			= cxl_pci_probe,
> >  	.err_handler		= &cxl_error_handlers,
> > +	.dev_groups		= cxl_rcd_groups,
> >  	.driver	= {
> >  		.probe_type	= PROBE_PREFER_ASYNCHRONOUS,
> >  	},

Re: [PATCH v7 2/2] cxl/pci: Add sysfs attribute for CXL 1.1 device link status

[Jonathan Cameron]

On Wed, 5 Jun 2024 07:21:44 +0000
"Daisuke Kobayashi (Fujitsu)" <kobayashi.da-06@fujitsu.com> wrote:

> Jonathan Cameron wrote:
> > On Fri, 10 May 2024 16:37:10 +0900
> > "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com> wrote:
> >   
> > > Add sysfs attribute for CXL 1.1 device link status to the cxl pci device.
> > >
> > > In CXL1.1, the link status of the device is included in the RCRB mapped to
> > > the memory mapped register area. Critically, that arrangement makes the
> > > link status and control registers invisible to existing PCI user tooling.
> > >
> > > Export those registers via sysfs with the expectation that PCI user
> > > tooling will alternatively look for these sysfs files when attempting to
> > > access to these CXL 1.1 endpoints registers.  
> > 
> > The lspci support raced with this series and has landed.
> > 
> > https://github.com/pciutils/pciutils/commit/49efa87fcce4f7d5b351238668ae
> > 1d4491802b88
> > 
> > A few follow up comments on suggestion to cache the offset, not the register
> > value
> > (which might change over time)  
> > >
> > > Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> > > Signed-off-by: "Kobayashi,Daisuke" <kobayashi.da-06@fujitsu.com>
> > > ---
> > >  drivers/cxl/pci.c | 101  
> > ++++++++++++++++++++++++++++++++++++++++++++++  
> > >  1 file changed, 101 insertions(+)
> > >
> > > diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
> > > index 2ff361e756d6..c10797adde2c 100644
> > > --- a/drivers/cxl/pci.c
> > > +++ b/drivers/cxl/pci.c
> > > @@ -786,6 +786,106 @@ static int cxl_event_config(struct pci_host_bridge  
> > *host_bridge,  
> > >  	return 0;
> > >  }
> > >
> > > +static ssize_t rcd_link_cap_show(struct device *dev,
> > > +				   struct device_attribute *attr, char *buf)
> > > +{
> > > +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> > > +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> > > +	struct device *endpoint_parent;
> > > +	struct cxl_dport *dport;
> > > +	struct cxl_port *port;
> > > +
> > > +	port = cxl_mem_find_port(cxlmd, &dport);
> > > +	if (!port)
> > > +		return -EINVAL;
> > > +
> > > +	endpoint_parent = port->uport_dev;
> > > +	if (!endpoint_parent)
> > > +		return -ENXIO;
> > > +
> > > +	guard(device)(endpoint_parent);
> > > +	if (!endpoint_parent->driver)
> > > +		return -ENXIO;
> > > +
> > > +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkcap);  
> > 
> > If you follow the suggestion of caching the offset, not the register
> > content then, these are all very similar. Define one common
> > function which takes a u8 offset and call that from each of the
> > functions.
> > 
> > If you stick to separate caches (and ensure they are up to date),
> > then you can add a utility function to take the dev here and
> > get you to the rcrb structure.  Then you can call that and only
> > have the final line in each instances of this code.
> >   
> 
> Thank you for your comment.
> 
> To confirm my understanding of the proposed implementation approach, 
> I would like to share my interpretation:
> 
> - The values stored in `cxl_rcrb_info` should be the offsets of the registers, 
> rather than the register values themselves.
> - Similar to aer, the values should be read from the memory-mapped region 
> using the offset when they are needed.
> - To achieve this, the offsets (like `cxl_rcrb_pcie_caps`) will be stored in `cxl_rcrb_info` 
> during initialization, and the memory-mapped region will be accessed within the `rcd_*_show()`
> functions where the values are actually used.
> - Accessing the memory-mapped region only requires the `struct device *dev`, 
> so a common utility function will be created and used for this purpose.
> 
> If the cost of accessing the memory-mapped region at runtime is not high, especially considering 
> the avoidance of the walk to find the offset, then I believe it is more reasonable to access 
> the memory-mapped region at runtime.
> 
> Could you let me know if there are any errors in my understanding.
> 
> If there are no issues, I will update the implementation with this approach 
> and submit a v8 patch (hopefully tomorrow).

Your description matches my intent. Looking forwards to v8.

Jonathan

> 
> > > +}
> > > +static DEVICE_ATTR_RO(rcd_link_cap);
> > > +
> > > +static ssize_t rcd_link_ctrl_show(struct device *dev,
> > > +				   struct device_attribute *attr, char *buf)
> > > +{
> > > +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> > > +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> > > +	struct device *endpoint_parent;
> > > +	struct cxl_dport *dport;
> > > +	struct cxl_port *port;
> > > +
> > > +	port = cxl_mem_find_port(cxlmd, &dport);
> > > +	if (!port)
> > > +		return -EINVAL;
> > > +
> > > +	endpoint_parent = port->uport_dev;
> > > +	if (!endpoint_parent)
> > > +		return -ENXIO;
> > > +
> > > +	guard(device)(endpoint_parent);
> > > +	if (!endpoint_parent->driver)
> > > +		return -ENXIO;
> > > +
> > > +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkctrl);
> > > +}
> > > +static DEVICE_ATTR_RO(rcd_link_ctrl);
> > > +
> > > +static ssize_t rcd_link_status_show(struct device *dev,
> > > +				   struct device_attribute *attr, char *buf)
> > > +{
> > > +	struct cxl_dev_state *cxlds = dev_get_drvdata(dev);
> > > +	struct cxl_memdev *cxlmd = cxlds->cxlmd;
> > > +	struct device *endpoint_parent;
> > > +	struct cxl_dport *dport;
> > > +	struct cxl_port *port;
> > > +
> > > +	port = cxl_mem_find_port(cxlmd, &dport);
> > > +	if (!port)
> > > +		return -EINVAL;
> > > +
> > > +	endpoint_parent = port->uport_dev;
> > > +	if (!endpoint_parent)
> > > +		return -ENXIO;
> > > +
> > > +	guard(device)(endpoint_parent);
> > > +	if (!endpoint_parent->driver)
> > > +		return -ENXIO;
> > > +
> > > +	return sysfs_emit(buf, "%x\n", dport->rcrb.rcd_lnkstatus);
> > > +}
> > > +static DEVICE_ATTR_RO(rcd_link_status);
> > > +
> > > +static struct attribute *cxl_rcd_attrs[] = {
> > > +		&dev_attr_rcd_link_cap.attr,
> > > +		&dev_attr_rcd_link_ctrl.attr,
> > > +		&dev_attr_rcd_link_status.attr,
> > > +		NULL
> > > +};
> > > +
> > > +static umode_t cxl_rcd_visible(struct kobject *kobj,
> > > +					  struct attribute *a, int n)
> > > +{
> > > +	struct device *dev = kobj_to_dev(kobj);
> > > +	struct pci_dev *pdev = to_pci_dev(dev);
> > > +
> > > +	if (is_cxl_restricted(pdev))
> > > +		return a->mode;
> > > +
> > > +	return 0;
> > > +}
> > > +
> > > +static struct attribute_group cxl_rcd_group = {
> > > +		.attrs = cxl_rcd_attrs,
> > > +		.is_visible = cxl_rcd_visible,
> > > +};
> > > +__ATTRIBUTE_GROUPS(cxl_rcd);
> > > +
> > >  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);  
> > > @@ -969,6 +1069,7 @@ static struct pci_driver cxl_pci_driver = {
> > >  	.id_table		= cxl_mem_pci_tbl,
> > >  	.probe			= cxl_pci_probe,
> > >  	.err_handler		= &cxl_error_handlers,
> > > +	.dev_groups		= cxl_rcd_groups,
> > >  	.driver	= {
> > >  		.probe_type	= PROBE_PREFER_ASYNCHRONOUS,
> > >  	},  
>