On Thu, 2017-08-24 at 11:11 +0200, Christoph Hellwig wrote: > On Wed, Aug 23, 2017 at 02:40:07PM -0700, Bart Van Assche wrote: > > Avoid that the following compiler warning is reported when building > > with W=1: > > > > drivers/scsi/scsi_transport_srp.c:92:19: warning: comparison is always false due to limited range of data type [-Wtype-limits] > > > > Signed-off-by: Bart Van Assche <bart.vanassche@xxxxxxx> > > Cc: Christoph Hellwig <hch@xxxxxx> > > Cc: Hannes Reinecke <hare@xxxxxxx> > > Cc: Johannes Thumshirn <jthumshirn@xxxxxxx> > > --- > > drivers/scsi/scsi_transport_srp.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/drivers/scsi/scsi_transport_srp.c b/drivers/scsi/scsi_transport_srp.c > > index 698cc4681706..b8f5e4c47579 100644 > > --- a/drivers/scsi/scsi_transport_srp.c > > +++ b/drivers/scsi/scsi_transport_srp.c > > @@ -89,7 +89,7 @@ int srp_tmo_valid(int reconnect_delay, int fast_io_fail_tmo, int dev_loss_tmo) > > if (fast_io_fail_tmo < 0 && > > dev_loss_tmo > SCSI_DEVICE_BLOCK_MAX_TIMEOUT) > > return -EINVAL; > > - if (dev_loss_tmo >= LONG_MAX / HZ) > > + if (dev_loss_tmo + 0UL >= LONG_MAX / HZ) > > return -EINVAL; > > That's a weird change.. If we can to promote dev_loss_tmo to > long we should just cast it. And of course we should always ask > us why we got this warning. If long is 64-bit and int is 32-bit > the warning makes sense, as for every reasonable comparism > the 32-bit timeout can't be larger than LONG_MAX divided by 100 or 100. > > But for 32-bit longs it can, so we should keep it, maybe with a comment. Hello Christoph, The purpose of that check is to avoid that dev_loss_tmo * HZ can overflow. That check is only needed on 32-bit systems since only on these systems sizeof(long) == sizeof(int). How about changing the type of the dev_loss_tmo argument from int to long such that no explicit cast is needed? Thanks, Bart.
