On Tue, 2015-05-26 at 09:30 -0400, James Smart wrote: > Sebastian, > > Re: more than 1 space between a type declaration and a variable name - I > do not believe that's a hard requirement. It fully passes checkpatch. > Yes, consistent style use (aligning all variable names at same offset, > or always 1) would be good - but code has been there so long with > althernate styles it doesn't really matter at this point. I did clean > up those in your last review as I needed to do a mod for the LS_RJT > behavior. But... this seems like a nit. I did promise Christoph that I > would pick a good point and retrofit the sources for all sparse warnings > - and still owe him. Checkpatch is a guideline rather than absolute. It picks up a lot of useful stuff, but also whines about a lot of irrelevant things. In general, errors have to be fixed but a lot of warnings are ignorable (some warnings, like space instead of tabs aren't, but a lot are). In the case of warnings, it's up to the maintainer to judge if they match the current style of the driver. > Re: Checkpatch and string splitting. I understand we aren't passing > checkpatch for that rule, but joining them would have checkpatch > flagging us for beyond 80 character lines. I'd much rather have the > splits and keep the indenting for readability. We have also had this > error quite a bit in the past and believe we have been grandfathered as > there's a lot of this already. I consider the line over 80 characters one of the most bogus checkpatch warnings, so I would prefer not splitting strings, but it's only a warning, so well within the bounds of the maintainer to decide based on the internal style of the driver. James > James B - any comments on the above ? -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
