From mboxrd@z Thu Jan  1 00:00:00 1970
From: "H. Peter Anvin" <hpa@zytor.com>
Subject: Re: gcc inlining heuristics was Re: [PATCH -v7][RFC]: mutex:	implement
 adaptive spinning
Date: Tue, 20 Jan 2009 17:22:20 -0800
Message-ID: <497678CC.5060703@zytor.com>
References: <alpine.LFD.2.00.0901121128080.6528@localhost.localdomain> <496BBE27.2020206@t-online.de> <20090119001345.GA9880@elte.hu> <20090119062212.GC22584@wotan.suse.de> <alpine.LFD.2.00.0901200754100.2974@localhost.localdomain> <20090120005124.GD16304@wotan.suse.de> <20090120123824.GD7790@elte.hu> <1232480940.22233.1435.camel@macbook.infradead.org> <20090120210515.GC19710@elte.hu> <alpine.LFD.2.00.0901210814130.2892@localhost.localdomain> <20090120220516.GA10483@elte.hu>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	David Woodhouse <dwmw2@infradead.org>,
	Nick Piggin <npiggin@suse.de>,
	Bernd Schmidt <bernds_cb1@t-online.de>,
	Andi Kleen <andi@firstfloor.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	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>,
	Peter Morreale <pmorreale@novell.com>,
	Sven Dietrich <SDietrich@novell.com>, jh@suse.cz
To: Ingo Molnar <mingo@elte.hu>
Return-path: <linux-fsdevel-owner@vger.kernel.org>
In-Reply-To: <20090120220516.GA10483@elte.hu>
List-ID: <linux-btrfs.vger.kernel.org>

Ingo Molnar wrote:
> 
> Hm, GCC uses __restrict__, right?
> 
> I'm wondering whether there's any internal tie-up between alias analysis 
> and the __restrict__ keyword - so if we turn off aliasing optimizations 
> the __restrict__ keyword's optimizations are turned off as well.
> 

Actually I suspect that "restrict" makes little difference for inlines 
or even statics, since gcc generally can do alias analysis fine there. 
However, in the presence of an intermodule function call, all alias 
analysis is off.  This is presumably why type-based analysis is used at 
all ... to at least be able to a modicum of, say, loop invariant removal 
in the presence of a library call.  This is also where "restrict" comes 
into play.

	-hpa