Date: Fri, 25 Jan 2008 19:08:28 +0100
From: Ingo Molnar <mingo@elte.hu>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Harvey Harrison <harvey.harrison@gmail.com>,
       Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
       Andi Kleen <ak@suse.de>
Subject: Re: [PATCH UPDATE] x86: ignore spurious faults
Message-ID: <20080125180828.GB10993@elte.hu>
References: <4797D64D.1060105@goop.org> <1201133916.16972.124.camel@brick> <4797DBA0.5020909@goop.org> <20080125153032.GE11846@elte.hu> <479A0632.9090205@goop.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <479A0632.9090205@goop.org>
User-Agent: Mutt/1.5.17 (2007-11-01)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1463
Lines: 38


* Jeremy Fitzhardinge <jeremy@goop.org> wrote:

> Ingo Molnar wrote:
>> thanks, applied.
>>
>> it would be nice to expose this ability of the architecture to the core 
>> Linux kernel mprotect code as well, and let it skip on a TLB flush when 
>> doing a RO->RW transition.
>
> The usermode fault handler already effectively does this; this patch 
> just does it for kernel mode as well.  I don't know if mprotect takes 
> advantage of this.

spurious faults happen all the time on SMP, in the native kernel.

And what i mean is that Linux mprotect currently does not take advantage 
of x86's ability to just change the ptes, because there's no structured 
way to tell mm/mprotect.c that "it's safe to skip the TLB flush here". 

The flush happens in mm/mprotect.c's change_protection() function:

        flush_tlb_range(vma, start, end);

and that is unnecessary when we increase the protection rights, such as 
in a RO->RW change. (all that is needed is an smp_wmb() instead, to make 
sure all the pte modifications are visible when the syscall returns.)

and it's a really rare case these days that you can find an area where 
Linux does not make use of a hardware MMU feature - so we should fix 
this ;-)

	Ingo
--
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/