On Sun, Feb 23, 2020 at 09:03:05AM -0700, Allison Collins wrote: > On 2/23/20 2:34 AM, Amir Goldstein wrote: > > On Sun, Feb 23, 2020 at 4:07 AM Allison Collins > > A struct inititializer macro would have been nice, so code like this: > > > > + struct xfs_name xname; > > + > > + xfs_name_init(&xname, name); > > > > Would become: > > + struct xfs_name xname = XFS_NAME_STR_INIT(name); > > > > As a matter of fact, in most of the cases a named local variable is > > not needed at > > all and the code could be written with an anonymous local struct variable macro: > > > > + error = xfs_attr_remove(XFS_I(inode), XFS_NAME_STR(name), flags); > > > The macro does look nice. I would be the third iteration of initializers > that this patch has been through though. Can I get a consensus of how many > people like the macro? Not me. All it means is that every time I look at this code I have to go look at what that shouty macro does, then have to understand it's being "smart and clever" to save a couple of lines of clear, obviously correct code. That's not an improvement. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx
