[PATCH] fpga: bridge: return errors in the show() method of the "state" attribute

[PATCH] fpga: bridge: return errors in the show() method of the "state" attribute

[Marco Pagani]

This patch changes the show() method of the "state" sysfs attribute to
propagate errors returned by the enable_show() op. In this way,
userspace can distinguish between when the bridge is actually "enabled"
(i.e., allowing signals) or "disabled" (i.e., gating signals), or when
there is an error.

Currently, enable_show() returns an integer representing the bridge's
state (enabled or disabled) or an error code. However, this integer
value is interpreted in state_show() as a bool, resulting in the method
printing "enabled" (i.e., the bridge allows signals to pass), without
propagating the error, even when enable_show() returns an error code.

Thanks

Signed-off-by: Marco Pagani <marpagan@redhat.com>
---
 drivers/fpga/fpga-bridge.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/fpga/fpga-bridge.c b/drivers/fpga/fpga-bridge.c
index 727704431f61..5cd40acab5bf 100644
--- a/drivers/fpga/fpga-bridge.c
+++ b/drivers/fpga/fpga-bridge.c
@@ -293,12 +293,15 @@ static ssize_t state_show(struct device *dev,
 			  struct device_attribute *attr, char *buf)
 {
 	struct fpga_bridge *bridge = to_fpga_bridge(dev);
-	int enable = 1;
+	int state = 1;
 
-	if (bridge->br_ops && bridge->br_ops->enable_show)
-		enable = bridge->br_ops->enable_show(bridge);
+	if (bridge->br_ops && bridge->br_ops->enable_show) {
+		state = bridge->br_ops->enable_show(bridge);
+		if (state < 0)
+			return state;
+	}
 
-	return sprintf(buf, "%s\n", enable ? "enabled" : "disabled");
+	return sysfs_emit(buf, "%s\n", state ? "enabled" : "disabled");
 }
 
 static DEVICE_ATTR_RO(name);
-- 
2.39.1

Re: [PATCH] fpga: bridge: return errors in the show() method of the "state" attribute

[Xu Yilun]

On 2023-01-25 at 15:06:22 +0100, Marco Pagani wrote:
> This patch changes the show() method of the "state" sysfs attribute to
> propagate errors returned by the enable_show() op. In this way,
> userspace can distinguish between when the bridge is actually "enabled"
> (i.e., allowing signals) or "disabled" (i.e., gating signals), or when
> there is an error.
> 
> Currently, enable_show() returns an integer representing the bridge's
> state (enabled or disabled) or an error code. However, this integer
> value is interpreted in state_show() as a bool, resulting in the method
> printing "enabled" (i.e., the bridge allows signals to pass), without
> propagating the error, even when enable_show() returns an error code.
> 
> Thanks

Removed this line, and applied to for-next.

Acked-by: Xu Yilun <yilun.xu@intel.com>

> 
> Signed-off-by: Marco Pagani <marpagan@redhat.com>
> ---
>  drivers/fpga/fpga-bridge.c | 11 +++++++----
>  1 file changed, 7 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/fpga/fpga-bridge.c b/drivers/fpga/fpga-bridge.c
> index 727704431f61..5cd40acab5bf 100644
> --- a/drivers/fpga/fpga-bridge.c
> +++ b/drivers/fpga/fpga-bridge.c
> @@ -293,12 +293,15 @@ static ssize_t state_show(struct device *dev,
>  			  struct device_attribute *attr, char *buf)
>  {
>  	struct fpga_bridge *bridge = to_fpga_bridge(dev);
> -	int enable = 1;
> +	int state = 1;
>  
> -	if (bridge->br_ops && bridge->br_ops->enable_show)
> -		enable = bridge->br_ops->enable_show(bridge);
> +	if (bridge->br_ops && bridge->br_ops->enable_show) {
> +		state = bridge->br_ops->enable_show(bridge);
> +		if (state < 0)
> +			return state;
> +	}
>  
> -	return sprintf(buf, "%s\n", enable ? "enabled" : "disabled");
> +	return sysfs_emit(buf, "%s\n", state ? "enabled" : "disabled");
>  }
>  
>  static DEVICE_ATTR_RO(name);
> -- 
> 2.39.1
>