From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx115.postini.com [74.125.245.115]) by kanga.kvack.org (Postfix) with SMTP id 120C26B0062 for ; Wed, 31 Oct 2012 05:10:20 -0400 (EDT) Date: 31 Oct 2012 05:10:18 -0400 Message-ID: <20121031091018.24875.qmail@science.horizon.com> From: "George Spelvin" Subject: Re: [dm-devel] [PATCH v8 01/16] hashtable: introduce a small and naive hashtable Sender: owner-linux-mm@kvack.org List-ID: To: tj@kernel.org Cc: dm-devel@redhat.com, levinsasha928@gmail.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-nfs@vger.kernel.org, netdev@vger.kernel.org Tejun Heo wrote: >> +#define hash_min(val, bits) \ >> +({ \ >> + sizeof(val) <= 4 ? \ >> + hash_32(val, bits) : \ >> + hash_long(val, bits); \ >> +}) > Also, you probably want () around at least @val. In general, > it's a good idea to add () around any macro argument to avoid nasty > surprises. Er... not in this case, you don't. If a macro argument is passed verbatim as an argument to a function, it doesn't need additional parens. That's because the one guarantee you have about a macro argument is that it can't contain any (unquoted) commas, and there's nothing lower precedence than the comma. So it's safe to delimit a macro argument with *either* parens *or* a comma. So you can go ahead and write: #define hash_min(val, bits) \ (sizeof(val) <= 4 ? hash_32(val, bits) : hash_long(val, bits)) ... which is easier to read, anyway. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org