[Devel] O_NONBLOCK for directory

[Kir Kolyshkin]

On 04/15/2016 04:48 AM, Dmitry Mishin wrote:
> I am reviewing recent VZ7 libploop commits, in particular, this one:
>
>
>     https://src.openvz.org/projects/OVZ/repos/ploop/commits/36df847b9
>
>     I left a question there, let me repeat it here in a hope someone
>     answers.
>
>>     Igor Sukhihcommitted36df847b99c
>>     <https://src.openvz.org/projects/OVZ/repos/ploop/commits/36df847b99c92557c69255ebfb00d4cc74cb51ac>Yesterday
>>     ploop_copy_init(): open folder with O_DIRECTORY flag
>>
>     ...
>>     -        _h->mntfd = open(mnt, O_RDONLY);
>>     +        _h->mntfd = open(mnt, O_RDONLY|O_NONBLOCK|O_DIRECTORY);
>
>     1. What's the reason for adding O_NONBLOCK here? As far as I can
>     see, it doesn't
>     change anything at all (neither in this open(), nor in subsequent
>     syncfs(), ioctl()
>     and close())? I went as far as the kernel sources to check that
>     O_NONBLOCK
>     doesn't affect syncfs() call, but maybe I'm mistaken?
>
>
> Added accidentally, you are right, it doesn't change anything.
>
>
>
>     2. What's the reason for adding O_DIRECTORY? Ideally, the changelog
>     should say why we're doing it, not what we do (as it's pretty clear
>     from the patch itself).
>
> It does exactly as specified – enforces open file to be a directory.

See, I am not asking what it does -- it's pretty clear from open(2) man 
page.

What I am interested in is the reason _why_ it was added. A bug report
this change is trying to fix, or a train of thought leading to this 
change, etc.
Ideally this is what the changelog message should tell us.

> Any problems with this enforcement?

No, I don't foresee any problems, more to say, this extra check might
actually do some good.

Kir.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openvz.org/pipermail/devel/attachments/20160415/fd3c8a2b/attachment.html>