Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 3 Nov 2001 19:46:08 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 3 Nov 2001 19:45:59 -0500 Received: from neon-gw-l3.transmeta.com ([63.209.4.196]:3081 "EHLO neon-gw.transmeta.com") by vger.kernel.org with ESMTP id ; Sat, 3 Nov 2001 19:45:52 -0500 Date: Sat, 3 Nov 2001 16:43:07 -0800 (PST) From: Linus Torvalds To: Richard Henderson cc: Alan Cox , Juergen Doelle , Subject: Re: Pls apply this spinlock patch to the kernel In-Reply-To: <20011103130156.D5984@twiddle.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, 3 Nov 2001, Richard Henderson wrote: > > On Sat, Nov 03, 2001 at 12:20:53PM -0800, Linus Torvalds wrote: > > If you have a 4-byte entry that is aligned to 128 bytes, you have 124 > > bytes of stuff that the linker _will_ fill up with other things. > > If you put the alignment on the type, not the variable, e.g. That doesn't work. The whole _point_ here is to make the thing variable-specific, so that we can say _this_ spinlock needs a cache-line of its own, without blowing up all spinlocks to 128 bytes. There's no way we want 128-byte spinlocks in general. We want to mark 4-5 spinlocks as being so critical that they can have a cacheline of their own. But I do _not_ want to have special operations for those spinlocks, so I don't want those spinlocks to have any special types (ie the actual _user_ should not need to care, so that you can play around with testing the different spinlocks one by one without having to edit all the users of any specific spinlock). Linus - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/