On 6/18/22 04:30, Sergey Shtylyov wrote:
> While ata_ioc32() returns *int*, its result gets assigned to and compared
> with the *unsigned long* variable 'val' in ata_sas_scsi_ioctl(), its only
> caller, which implies a problematic implicit cast -- fix that by returning
> *unsigned long* instead.
>
> Found by Linux Verification Center (linuxtesting.org) with the SVACE static
> analysis tool.
>
> Signed-off-by: Sergey Shtylyov <s.shtylyov@xxxxxx>
>
> ---
> This patch is against the 'for-next' branch of Damien's 'libata.git' repo.
>
> drivers/ata/libata-scsi.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> Index: libata/drivers/ata/libata-scsi.c
> ===================================================================
> --- libata.orig/drivers/ata/libata-scsi.c
> +++ libata/drivers/ata/libata-scsi.c
> @@ -539,7 +539,7 @@ int ata_task_ioctl(struct scsi_device *s
> return rc;
> }
>
> -static int ata_ioc32(struct ata_port *ap)
> +static unsigned long ata_ioc32(struct ata_port *ap)
> {
> if (ap->flags & ATA_FLAG_PIO_DMA)
> return 1;
Actually, this should be a bool I think and the ioctl code cleaned to use
that type since the val argument of the ioctl is also used as a bool.
--
Damien Le Moal
Western Digital Research
