[PATCH] ata: ahci: Make ahci_ignore_port() handle empty mask_port_map

[PATCH] ata: ahci: Make ahci_ignore_port() handle empty mask_port_map

[Niklas Cassel]

Commit 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port
numbers") added a skip to ahci_platform_enable_phys() for ports that are
not in mask_port_map.

The code in ahci_platform_get_resources(), will currently set mask_port_map
for each child "port" node it finds in the device tree.

However, device trees that do not have any child "port" nodes will not have
mask_port_map set, and for non-device tree platforms mask_port_map will
only exist as a quirk for specific PCI device + vendor IDs, or as a kernel
module parameter, but will not be set by default.

Therefore, the common thing is that mask_port_map is only set if you do not
want to use all ports (as defined by Offset 0Ch: PI – Ports Implemented
register), but instead only want to use the ports in mask_port_map. If
mask_port_map is not set, all ports are available.

Thus, ahci_ignore_port() must be able to handle an empty mask_port_map.

Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Closes: https://lore.kernel.org/linux-ide/10b31dd0-d0bb-4f76-9305-2195c3e17670@samsung.com/
Co-developed-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Fixes: 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port numbers")
Fixes: 2c202e6c4f4d ("ata: libahci_platform: Do not set mask_port_map when not needed")
Signed-off-by: Niklas Cassel <cassel@kernel.org>
---
 drivers/ata/ahci.h    | 8 ++++++--
 drivers/ata/libahci.c | 1 +
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
index aea30df50c58..b2e0ef4efbdc 100644
--- a/drivers/ata/ahci.h
+++ b/drivers/ata/ahci.h
@@ -386,8 +386,12 @@ struct ahci_host_priv {
 static inline bool ahci_ignore_port(struct ahci_host_priv *hpriv,
 				    unsigned int portid)
 {
-	return portid >= hpriv->nports ||
-		!(hpriv->mask_port_map & (1 << portid));
+	if (portid >= hpriv->nports)
+		return true;
+	/* mask_port_map not set means that all ports are available */
+	if (!hpriv->mask_port_map)
+		return false;
+	return !(hpriv->mask_port_map & (1 << portid));
 }
 
 extern int ahci_ignore_sss;
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index fdfa7b266218..e7ace4b10f15 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -541,6 +541,7 @@ void ahci_save_initial_config(struct device *dev, struct ahci_host_priv *hpriv)
 		hpriv->saved_port_map = port_map;
 	}
 
+	/* mask_port_map not set means that all ports are available */
 	if (hpriv->mask_port_map) {
 		dev_warn(dev, "masking port_map 0x%lx -> 0x%lx\n",
 			port_map,
-- 
2.48.1

Re: [PATCH] ata: ahci: Make ahci_ignore_port() handle empty mask_port_map

[Marek Szyprowski]

On 25.02.2025 15:16, Niklas Cassel wrote:
> Commit 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port
> numbers") added a skip to ahci_platform_enable_phys() for ports that are
> not in mask_port_map.
>
> The code in ahci_platform_get_resources(), will currently set mask_port_map
> for each child "port" node it finds in the device tree.
>
> However, device trees that do not have any child "port" nodes will not have
> mask_port_map set, and for non-device tree platforms mask_port_map will
> only exist as a quirk for specific PCI device + vendor IDs, or as a kernel
> module parameter, but will not be set by default.
>
> Therefore, the common thing is that mask_port_map is only set if you do not
> want to use all ports (as defined by Offset 0Ch: PI – Ports Implemented
> register), but instead only want to use the ports in mask_port_map. If
> mask_port_map is not set, all ports are available.
>
> Thus, ahci_ignore_port() must be able to handle an empty mask_port_map.
>
> Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
> Closes: https://lore.kernel.org/linux-ide/10b31dd0-d0bb-4f76-9305-2195c3e17670@samsung.com/
> Co-developed-by: Damien Le Moal <dlemoal@kernel.org>
> Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
> Fixes: 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port numbers")
> Fixes: 2c202e6c4f4d ("ata: libahci_platform: Do not set mask_port_map when not needed")
> Signed-off-by: Niklas Cassel <cassel@kernel.org>

This fixes the issue I've observed.

Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>

> ---
>   drivers/ata/ahci.h    | 8 ++++++--
>   drivers/ata/libahci.c | 1 +
>   2 files changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
> index aea30df50c58..b2e0ef4efbdc 100644
> --- a/drivers/ata/ahci.h
> +++ b/drivers/ata/ahci.h
> @@ -386,8 +386,12 @@ struct ahci_host_priv {
>   static inline bool ahci_ignore_port(struct ahci_host_priv *hpriv,
>   				    unsigned int portid)
>   {
> -	return portid >= hpriv->nports ||
> -		!(hpriv->mask_port_map & (1 << portid));
> +	if (portid >= hpriv->nports)
> +		return true;
> +	/* mask_port_map not set means that all ports are available */
> +	if (!hpriv->mask_port_map)
> +		return false;
> +	return !(hpriv->mask_port_map & (1 << portid));
>   }
>   
>   extern int ahci_ignore_sss;
> diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
> index fdfa7b266218..e7ace4b10f15 100644
> --- a/drivers/ata/libahci.c
> +++ b/drivers/ata/libahci.c
> @@ -541,6 +541,7 @@ void ahci_save_initial_config(struct device *dev, struct ahci_host_priv *hpriv)
>   		hpriv->saved_port_map = port_map;
>   	}
>   
> +	/* mask_port_map not set means that all ports are available */
>   	if (hpriv->mask_port_map) {
>   		dev_warn(dev, "masking port_map 0x%lx -> 0x%lx\n",
>   			port_map,

Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland

Re: [PATCH] ata: ahci: Make ahci_ignore_port() handle empty mask_port_map

[Damien Le Moal]

On 2/25/25 11:16 PM, Niklas Cassel wrote:
> Commit 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port
> numbers") added a skip to ahci_platform_enable_phys() for ports that are
> not in mask_port_map.
> 
> The code in ahci_platform_get_resources(), will currently set mask_port_map
> for each child "port" node it finds in the device tree.
> 
> However, device trees that do not have any child "port" nodes will not have
> mask_port_map set, and for non-device tree platforms mask_port_map will
> only exist as a quirk for specific PCI device + vendor IDs, or as a kernel
> module parameter, but will not be set by default.
> 
> Therefore, the common thing is that mask_port_map is only set if you do not
> want to use all ports (as defined by Offset 0Ch: PI – Ports Implemented
> register), but instead only want to use the ports in mask_port_map. If
> mask_port_map is not set, all ports are available.
> 
> Thus, ahci_ignore_port() must be able to handle an empty mask_port_map.
> 
> Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
> Closes: https://lore.kernel.org/linux-ide/10b31dd0-d0bb-4f76-9305-2195c3e17670@samsung.com/
> Co-developed-by: Damien Le Moal <dlemoal@kernel.org>
> Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
> Fixes: 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port numbers")
> Fixes: 2c202e6c4f4d ("ata: libahci_platform: Do not set mask_port_map when not needed")

I think that it also would be a good idea to add:

Fixes: c9b5be909e65 ("ahci: Introduce ahci_ignore_port() helper")

So that this patch also gets picked up for backporting.

> Signed-off-by: Niklas Cassel <cassel@kernel.org>
> ---
>  drivers/ata/ahci.h    | 8 ++++++--
>  drivers/ata/libahci.c | 1 +
>  2 files changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
> index aea30df50c58..b2e0ef4efbdc 100644
> --- a/drivers/ata/ahci.h
> +++ b/drivers/ata/ahci.h
> @@ -386,8 +386,12 @@ struct ahci_host_priv {
>  static inline bool ahci_ignore_port(struct ahci_host_priv *hpriv,
>  				    unsigned int portid)
>  {
> -	return portid >= hpriv->nports ||
> -		!(hpriv->mask_port_map & (1 << portid));
> +	if (portid >= hpriv->nports)
> +		return true;
> +	/* mask_port_map not set means that all ports are available */
> +	if (!hpriv->mask_port_map)
> +		return false;
> +	return !(hpriv->mask_port_map & (1 << portid));
>  }
>  
>  extern int ahci_ignore_sss;
> diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
> index fdfa7b266218..e7ace4b10f15 100644
> --- a/drivers/ata/libahci.c
> +++ b/drivers/ata/libahci.c
> @@ -541,6 +541,7 @@ void ahci_save_initial_config(struct device *dev, struct ahci_host_priv *hpriv)
>  		hpriv->saved_port_map = port_map;
>  	}
>  
> +	/* mask_port_map not set means that all ports are available */
>  	if (hpriv->mask_port_map) {
>  		dev_warn(dev, "masking port_map 0x%lx -> 0x%lx\n",
>  			port_map,


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH] ata: ahci: Make ahci_ignore_port() handle empty mask_port_map

[Niklas Cassel]

On Tue, 25 Feb 2025 15:16:12 +0100, Niklas Cassel wrote:
> Commit 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port
> numbers") added a skip to ahci_platform_enable_phys() for ports that are
> not in mask_port_map.
> 
> The code in ahci_platform_get_resources(), will currently set mask_port_map
> for each child "port" node it finds in the device tree.
> 
> [...]

Applied to libata/linux.git (for-6.14), thanks!

[1/1] ata: ahci: Make ahci_ignore_port() handle empty mask_port_map
      https://git.kernel.org/libata/linux/c/130ff5c8

Kind regards,
Niklas