From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1757268AbbGGMyr (ORCPT <rfc822;w@1wt.eu>);
	Tue, 7 Jul 2015 08:54:47 -0400
Received: from mail-wi0-f180.google.com ([209.85.212.180]:35189 "EHLO
	mail-wi0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1754408AbbGGMyi (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Tue, 7 Jul 2015 08:54:38 -0400
Date: Tue, 7 Jul 2015 14:54:33 +0200
From: Anton Arapov <arapov@gmail.com>
To: Oleg Nesterov <oleg@redhat.com>
Cc: Ananth Mavinakayanahalli <ananth@in.ibm.com>,
        David Long <dave.long@linaro.org>,
        Denys Vlasenko <dvlasenk@redhat.com>,
        "Frank Ch. Eigler" <fche@redhat.com>, Ingo Molnar <mingo@kernel.org>,
        Jan Willeke <willeke@de.ibm.com>, Jim Keniston <jkenisto@us.ibm.com>,
        Mark Wielaard <mjw@redhat.com>, Pratyush Anand <panand@redhat.com>,
        Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 05/11] uprobes: Change handle_trampoline() to find the
 next chain beforehand
Message-ID: <20150707125433.GE5946@troglodyte.intinfra.com>
References: <20150707012210.GA7466@redhat.com>
 <20150707012250.GA7523@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150707012250.GA7523@redhat.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Tue, Jul 07, 2015 at 03:22:50AM +0200, Oleg Nesterov wrote:
> No functional changes, preparation.
> 
> Add the new helper, find_next_ret_chain(), which finds the first !chained
> entry and returns its ->next. Yes, it is suboptimal. We probably want to
> turn ->chained into ->start_of_this_chain pointer and avoid another loop.
> But this needs the boring changes in dup_utask(), so lets do this later.
> 
> Change the main loop in handle_trampoline() to unwind the stack until ri
> is equal to the pointer returned by this new helper.
> 
> Signed-off-by: Oleg Nesterov <oleg@redhat.com>
> Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Anton Arapov <arapov@gmail.com>


> ---
>  kernel/events/uprobes.c |   27 ++++++++++++++++-----------
>  1 files changed, 16 insertions(+), 11 deletions(-)
> 
> diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
> index 4c941fe..98e4d97 100644
> --- a/kernel/events/uprobes.c
> +++ b/kernel/events/uprobes.c
> @@ -1766,11 +1766,22 @@ handle_uretprobe_chain(struct return_instance *ri, struct pt_regs *regs)
>  	up_read(&uprobe->register_rwsem);
>  }
>  
> +static struct return_instance *find_next_ret_chain(struct return_instance *ri)
> +{
> +	bool chained;
> +
> +	do {
> +		chained = ri->chained;
> +		ri = ri->next;	/* can't be NULL if chained */
> +	} while (chained);
> +
> +	return ri;
> +}
> +
>  static void handle_trampoline(struct pt_regs *regs)
>  {
>  	struct uprobe_task *utask;
> -	struct return_instance *ri;
> -	bool chained;
> +	struct return_instance *ri, *next;
>  
>  	utask = current->utask;
>  	if (!utask)
> @@ -1780,24 +1791,18 @@ static void handle_trampoline(struct pt_regs *regs)
>  	if (!ri)
>  		goto sigill;
>  
> +	next = find_next_ret_chain(ri);
>  	/*
>  	 * TODO: we should throw out return_instance's invalidated by
>  	 * longjmp(), currently we assume that the probed function always
>  	 * returns.
>  	 */
>  	instruction_pointer_set(regs, ri->orig_ret_vaddr);
> -
> -	for (;;) {
> +	do {
>  		handle_uretprobe_chain(ri, regs);
> -
> -		chained = ri->chained;
>  		ri = free_ret_instance(ri);
>  		utask->depth--;
> -
> -		if (!chained)
> -			break;
> -		BUG_ON(!ri);
> -	}
> +	} while (ri != next);
>  
>  	utask->return_instances = ri;
>  	return;
> -- 
> 1.5.5.1
>