Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <b6d2b1e8-f458-650b-37d1-e73830f1c85e@linux.alibaba.com>
Date:   Fri, 1 Apr 2022 11:23:01 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:91.0)
 Gecko/20100101 Thunderbird/91.6.1
Subject: Re: [PATCH RESEND] uio/uio_pci_generic: Introduce refcnt on
 open/release
To:     "Michael S. Tsirkin" <mst@redhat.com>
Cc:     gregkh@linuxfoundation.org, linux-kernel@vger.kernel.org,
        alikernel-developer@linux.alibaba.com, gang.cao@intel.com,
        shouqu.ssq@alibaba-inc.com, yangxiu.yx@alibaba-inc.com
References: <1648693432-129409-1-git-send-email-yaohongbo@linux.alibaba.com>
 <20220331162952-mutt-send-email-mst@kernel.org>
From:   Yao Hongbo <yaohongbo@linux.alibaba.com>
In-Reply-To: <20220331162952-mutt-send-email-mst@kernel.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

+ cc.

在 2022/4/1 上午4:30, Michael S. Tsirkin 写道:
> On Thu, Mar 31, 2022 at 10:23:52AM +0800, Yao Hongbo wrote:
>> If two userspace programs both open the PCI UIO fd, when one
>> of the program exits uncleanly, the other will cause IO hang
>> due to bus-mastering disabled.
> 
> With two programs poking at the same device, how is this ever
> supposed to work even while both are alive?
> 
>> It's a common usage for spdk/dpdk to use UIO. So, introduce refcnt
>> to avoid such problems.
>>
>> Fixes: 865a11f("uio/uio_pci_generic: Disable bus-mastering on release")
>> Reported-by: Xiu Yang <yangxiu.yx@alibaba-inc.com>
>> Signed-off-by: Yao Hongbo <yaohongbo@linux.alibaba.com>
>> ---
>>  drivers/uio/uio_pci_generic.c | 16 +++++++++++++++-
>>  1 file changed, 15 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/uio/uio_pci_generic.c b/drivers/uio/uio_pci_generic.c
>> index e03f9b5..8add2cf 100644
>> --- a/drivers/uio/uio_pci_generic.c
>> +++ b/drivers/uio/uio_pci_generic.c
>> @@ -31,6 +31,7 @@
>>  struct uio_pci_generic_dev {
>>  	struct uio_info info;
>>  	struct pci_dev *pdev;
>> +	atomic_t refcnt;
>>  };
>>  
>>  static inline struct uio_pci_generic_dev *
>> @@ -39,6 +40,14 @@ struct uio_pci_generic_dev {
>>  	return container_of(info, struct uio_pci_generic_dev, info);
>>  }
>>  
>> +static int open(struct uio_info *info, struct inode *inode)
>> +{
>> +	struct uio_pci_generic_dev *gdev = to_uio_pci_generic_dev(info);
>> +
>> +	atomic_inc(&gdev->refcnt);
>> +	return 0;
>> +}
>> +
>>  static int release(struct uio_info *info, struct inode *inode)
>>  {
>>  	struct uio_pci_generic_dev *gdev = to_uio_pci_generic_dev(info);
>> @@ -51,7 +60,9 @@ static int release(struct uio_info *info, struct inode *inode)
>>  	 * Note that there's a non-zero chance doing this will wedge the device
>>  	 * at least until reset.
>>  	 */
>> -	pci_clear_master(gdev->pdev);
>> +	if (atomic_dec_and_test(&gdev->refcnt))
>> +		pci_clear_master(gdev->pdev);
>> +
>>  	return 0;
>>  }
>>  
>> @@ -92,8 +103,11 @@ static int probe(struct pci_dev *pdev,
>>  
>>  	gdev->info.name = "uio_pci_generic";
>>  	gdev->info.version = DRIVER_VERSION;
>> +	gdev->info.open = open;
>>  	gdev->info.release = release;
>>  	gdev->pdev = pdev;
>> +	atomic_set(&gdev->refcnt, 0);
>> +
>>  	if (pdev->irq && (pdev->irq != IRQ_NOTCONNECTED)) {
>>  		gdev->info.irq = pdev->irq;
>>  		gdev->info.irq_flags = IRQF_SHARED;
>> -- 
>> 1.8.3.1