On Thu, 23 Oct 2008 10:12:28 +0200 Doron Shoham <dorons@xxxxxxxxxxxx> wrote: > FUJITA Tomonori wrote: > > On Wed, 22 Oct 2008 10:08:20 +0200 > > Doron Shoham <dorons@xxxxxxxxxxxx> wrote: > > > >> return error if tid is not a postive integer > >> > >> Signed-off-by: Doron Shoham <dorons@xxxxxxxxxxxx> > > > > In general, this patch does the right things: > > > > 1. we do more strict checking > > 2. we create a target whose id is 0. > > > > But I'm a bit worry about the latter change since I guess that some > > parts of the code assume that zero tid is an error. > > Exept from the strict checking of tid<=0, > can you remember where in the code we assume such thing? I can't recall so I asked you. Well, I think that there were some reason why I prohibited zero tid. > > If you are confident that all the code can handle zero tid, it's > > fine. But if not, please change this patch to handle zero tid as an > > error. > > Before this patch, when a user gave an invalid tid number the strtol function > returned 0. > This caused the new target to be created with tid=0. > Unless we artificially decide that target's tid cannot be 0, As I wrote above (and in the previous mail) I artificially decided that tid cannot be 0 for some reasons but I can't recall. But I thought that some code can't handle zero tid. > I don't see any problems with that. If the change doesn't break anything, it's fine by me. But seems that you looked at only tgtadm so I don't want to do it for now. > If we do want to use tid 0, we can consider update the tgt-admin script to > start searching the next available tid from 0. -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
