Re: [PATCH] PCI: Fix incorrect retval type in remove_id_store()

[Bjorn Helgaas <helgaas@kernel.org>]

On Tue, Feb 24, 2026 at 10:14:29AM -0800, Alok Tiwari wrote:
> remove_id_store() initializes retval with -ENODEV, but retval was
> declared as size_t. Since size_t is unsigned, the negative value is
> converted to a large positive number, breaking error handling.

Does this actually cause a problem?  size_t and ssize_t are the same
size, so I would think the conversions would end up with the same
bits:

  (gdb) p (long)((unsigned long) -19)
  $1 = -19
  (gdb) p (int)((unsigned int) -19)
  $2 = -19

We have lots of sysfs show/store functions in pci-sysfs.c that return
size_t values.

> Use ssize_t to match the sysfs store() return type.
> 
> Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
> ---
>  drivers/pci/pci-driver.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
> index dd9075403987..b2d3dffcc4f3 100644
> --- a/drivers/pci/pci-driver.c
> +++ b/drivers/pci/pci-driver.c
> @@ -261,7 +261,7 @@ static ssize_t remove_id_store(struct device_driver *driver, const char *buf,
>  	u32 vendor, device, subvendor = PCI_ANY_ID,
>  		subdevice = PCI_ANY_ID, class = 0, class_mask = 0;
>  	int fields;
> -	size_t retval = -ENODEV;
> +	ssize_t retval = -ENODEV;
>  
>  	fields = sscanf(buf, "%x %x %x %x %x %x",
>  			&vendor, &device, &subvendor, &subdevice,
> -- 
> 2.50.1
>