[dpdk-dev] [dpdk-stable] [PATCH 1/2] vfio: fix check for vfio_group_fd

Kevin Traynor ktraynor at redhat.com
Wed May 27 12:45:04 CEST 2020


On 21/05/2020 13:53, Burakov, Anatoly wrote:
> On 19-May-20 8:43 AM, David Marchand wrote:
>> On Tue, May 19, 2020 at 5:42 AM wangyunjian <wangyunjian at huawei.com> wrote:
>>>
>>> From: Yunjian Wang <wangyunjian at huawei.com>
>>>
>>> The issue is that a file descriptor at 0 is a valid one. Currently
>>> the file not found, the return value will be set to 0. As a result,
>>> it is impossible to distinguish between a correct descriptor and a
>>> failed return value. Fix it to return -ENOENT instead of 0.
>>>
>>> Fixes: b758423bc4fe ("vfio: fix race condition with sysfs")
>>> Fixes: ff0b67d1c868 ("vfio: DMA mappinge")
>>> Cc: stable at dpdk.org
>>
>> This patch reverts the (marked for stable) fix b758423bc4fe ("vfio:
>> fix race condition with sysfs") and comes with a different fix.
>>
>> - Kevin, Luca, I would put this b758423bc4fe backport on hold.
>>
>> - Anatoly, I don't want to put 20.05 at risk.
>> My simple question for 20.05 is, should we revert b758423bc4fe?
> 
> No, let's not revert anything.
> 
> As far as i can tell, this patch is a more complete fix, but it 
> essentially does the same thing, just in a different (and better) way. 
> Still, i haven't reviewed it in detail.
> 

b758423bc4fe was not reverted in master - should that be backported now?
or should we wait until this new fix is ready for backport too?

>>
>> - This patchset will go to 20.08 as I don't feel confident in taking it now.
>>
> 
> Yes, let's leave it for 20.08.
> 



More information about the dev mailing list