From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jamie Lokier <jamie@shareable.org>
Subject: Re: gcc inlining heuristics was Re: [PATCH -v7][RFC]: mutex: implement adaptive spinning
Date: Mon, 12 Jan 2009 23:01:17 +0000
Message-ID: <20090112230117.GC26430@shareable.org>
References: <alpine.LFD.2.00.0901111248060.6528@localhost.localdomain> <alpine.LFD.2.00.0901111458290.6528@localhost.localdomain> <20090112001255.GR26290@one.firstfloor.org> <alpine.LFD.2.00.0901111619400.6528@localhost.localdomain> <20090112005228.GS26290@one.firstfloor.org> <496B86B5.3090707@t-online.de> <alpine.LFD.2.00.0901121034170.6528@localhost.localdomain> <20090112193201.GA23848@one.firstfloor.org> <496B9890.1090002@zytor.com> <alpine.LFD.2.00.0901121143070.6528@localhost.localdomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: "H. Peter Anvin" <hpa@zytor.com>, Andi Kleen <andi@firstfloor.org>,
	Bernd Schmidt <bernds_cb1@t-online.de>,
	David Woodhouse <dwmw2@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Ingo Molnar <mingo@elte.hu>,
	Harvey Harrison <harvey.harrison@gmail.com>,
	Chris Mason <chris.mason@oracle.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Steven Rostedt <rostedt@goodmis.org>,
	paulmck@linux.vnet.ibm.com, Gregory Haskins <ghaskins@novell.com>,
	Matthew Wilcox <matthew@wil.cx>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	linux-btrfs <linux-btrfs@vger.kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Nick Piggin <npiggin@suse.de>,
	Peter Morreale <pmorreale@novell.com>,
	Sven Dietrich <SDietrich@novell.com>, jh@suse.cz
To: Linus Torvalds <torvalds@linux-foundation.org>
Return-path: <linux-btrfs-owner@vger.kernel.org>
In-Reply-To: <alpine.LFD.2.00.0901121143070.6528@localhost.localdomain>
List-ID: <linux-btrfs.vger.kernel.org>

Linus Torvalds wrote:
> > This is about storage allocation, not aliases.  Storage allocation only
> > depends on lifetime.
> 
> Well, the thing is, code motion does extend life-times, and if you think 
> you can move stores across each other (even when you can see that they 
> alias statically) due to type-based alias decisions, that does essentially 
> end up making what _used_ to be disjoint lifetimes now be potentially 
> overlapping.

Sometimes code motion makes code faster and/or smaller but use more
stack space.  If you want to keep the stack use down, it blocks some
other optimisations.

Register allocation is similar: code motion optimisations may use more
registers due to overlapping lifetimes, which causes more register
spills and changes the code.  The two interact; it's not trivial to
optimise fully.

-- Jamie