On Mon, 2011-11-07 at 03:57 -0500, Christoph Hellwig wrote: > On Sat, Nov 05, 2011 at 12:37:23PM -0700, Nicholas A. Bellinger wrote: > > I'd really like to avoid using tcm_sense_reason_table for function > > return values, or making external fabric code responsible for setting > > ->scsi_sense_reason by propagating up these values as returns. Also, > > tcm_sense_reason_table implies CHECK_CONDITION status, and there are > > cases where we only want to setting ->scsi_status. Although > > TCM_RESERVATION_CONFLICT is currently abusing this a little.. ;) > > > > Anyways, adding new return codes doesn't really seem to add much here > > either. Do you have a strong objecting to sticking with errno as long > > as the fabrics are not response for decoding specific failure status..? > > Yes. Using errno values when you don't care about the actual value > is a bad idea. I'm okay with keeping the TCM_ code separate, but in that > case please switch the routines to simply return bollean arguments. > Fair enough. I'll change the fabric facing functions involved here to return boolean instead of propagating up errno. Thanks, --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
