On Thu, Apr 16, 2009 at 2:16 AM, Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote: > On Sat, Apr 11, 2009 at 02:55:32PM +0200, walter harms wrote: >> >> >> Al Viro schrieb: >> > On Wed, Apr 08, 2009 at 01:31:43PM +0200, Bert Wesarg wrote: >> >> If typeof(c) == CommandList_struct*, that you could replace >> >> sizeof(CommandList_struct) with sizeof(*c) to make the expression >> >> shorter and fit into one line. >> > >> > This is not commonly accepted as a good thing. In particular, I would >> > seriously ask you to avoid doing anything of that kind around fs/* >> >> just for the curious ... >> What is the dark side of replacing ? >> foo=malloc(sizeof(struct fo)) -> foo=malloc(sizeof( *foo) )) > > The latter is ungreppable. For the curious, round 2: I was under the impression (perhaps incorrectly) that sizeof(*foo) was useful because if the type of foo changes (say from struct fo to struct new_fo), you do not need to go fishing around the source changing all the "struct fo" to "struct new_fo." This probably doesn't matter though if all the other code is already doing sizeof(struct fo), because you'll have to go fishing anyway. Joe -- To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
