From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-pci-owner@vger.kernel.org>
Received: from szxga02-in.huawei.com ([119.145.14.65]:29001 "EHLO
	szxga02-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1754616Ab2IUGBo (ORCPT
	<rfc822;linux-pci@vger.kernel.org>); Fri, 21 Sep 2012 02:01:44 -0400
Message-ID: <505C02A1.50601@huawei.com>
Date: Fri, 21 Sep 2012 14:01:05 +0800
From: Yijing Wang <wangyijing@huawei.com>
MIME-Version: 1.0
To: Huang Ying <ying.huang@intel.com>
CC: Bjorn Helgaas <bhelgaas@google.com>,
	Chen Gong <gong.chen@linux.intel.com>, <jiang.liu@huawei.com>,
	Hanjun Guo <guohanjun@huawei.com>, <linux-pci@vger.kernel.org>
Subject: Re: [PATCH v4 2/5] PCI/AER: introduce pci_bus_ops_get() to avoid
 a small race condition window
References: <1348199056-7696-1-git-send-email-wangyijing@huawei.com>  <1348199056-7696-3-git-send-email-wangyijing@huawei.com> <1348204119.23395.21.camel@yhuang-dev>
In-Reply-To: <1348204119.23395.21.camel@yhuang-dev>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-pci-owner@vger.kernel.org
List-ID: <linux-pci.vger.kernel.org>

On 2012/9/21 13:08, Huang Ying wrote:
> On Fri, 2012-09-21 at 11:44 +0800, Yijing Wang wrote:
>> When we rmmod aer_inject module, there is a small race condition window between pci_bus_ops_pop()
>> and pci_bus_set_ops() in aer_inject_exit, eg. pci_read_aer/pci_write_aer was called between
>> them. So introduce pci_bus_ops_get() to avoid this.
>>
>> Signed-off-by: Yijing Wang <wangyijing@huawei.com>
>> Signed-off-by: Huang Ying <ying.huang@intel.com>
> 
> Please don't add my signed-off.
> 
>> ---
>>  drivers/pci/pcie/aer/aer_inject.c |   21 ++++++++++++++++++---
>>  1 files changed, 18 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/pci/pcie/aer/aer_inject.c b/drivers/pci/pcie/aer/aer_inject.c
>> index fdab3bb..0123120 100644
>> --- a/drivers/pci/pcie/aer/aer_inject.c
>> +++ b/drivers/pci/pcie/aer/aer_inject.c
>> @@ -67,6 +67,8 @@ struct pci_bus_ops {
>>  	struct pci_ops *ops;
>>  };
>>  
>> +#define to_pci_bus_ops(n) container_of(n, struct pci_bus_ops, list)
>> +
>>  static LIST_HEAD(einjected);
>>  
>>  static LIST_HEAD(pci_bus_ops_list);
>> @@ -160,6 +162,18 @@ static struct pci_bus_ops *pci_bus_ops_pop(void)
>>  	return bus_ops;
>>  }
>>  
>> +static struct pci_bus_ops *pci_bus_ops_get(struct pci_bus_ops *from)
>> +{
>> +	struct pci_bus_ops *bus_ops = NULL;
>> +	struct list_head *n;
>> +
>> +	n = from ? from->list.next : pci_bus_ops_list.next;
>> +	if (n != &pci_bus_ops_list)
>> +		bus_ops = to_pci_bus_ops(n);
>> +
>> +	return bus_ops;
>> +}
>> +
>>  static u32 *find_pci_config_dword(struct aer_error *err, int where,
>>  				  int *prw1cs)
>>  {
>> @@ -539,14 +553,15 @@ static void __exit aer_inject_exit(void)
>>  {
>>  	struct aer_error *err, *err_next;
>>  	unsigned long flags;
>> -	struct pci_bus_ops *bus_ops;
>> +	struct pci_bus_ops *bus_ops = NULL;
>>  
>>  	misc_deregister(&aer_inject_device);
>>  
>> -	while ((bus_ops = pci_bus_ops_pop())) {
>> +	while ((bus_ops = pci_bus_ops_get(bus_ops)))
>>  		pci_bus_set_ops(bus_ops->bus, bus_ops->ops);
> 
> Why not just use
> 
> 	list_for_each_entry(&pci_bus_ops_list) {
> 		pci_bus_set_ops();
> 	}
> 

Yes, you are right, use list_for_each_entry(&pci_bus_ops_list) is cleaner.

Thanks
Yijing

> Best Regards,
> Huang Ying
> 
>> +
>> +	while ((bus_ops = pci_bus_ops_pop()))
>>  		kfree(bus_ops);
>> -	}
>>  
>>  	spin_lock_irqsave(&inject_lock, flags);
>>  	list_for_each_entry_safe(err, err_next, &einjected, list) {
> 
> 
> 
> .
> 


-- 
Thanks!
Yijing