[prev in list] [next in list] [prev in thread] [next in thread] 

List:       opensolaris-ufs-discuss
Subject:    Re: [ufs-discuss] Validity of vfs / vnode files
From:       Drew Balfour <andrew.balfour () sun ! com>
Date:       2008-08-18 16:59:33
Message-ID: 32026732.3881219104005676.JavaMail.Twebapp () sf-app1
[Download RAW message or body]

> I'm working on (again) the fix for 6613349 suid not
> allowed message could be better.

> 1. Assuming I do a 'refstr_hold(vfs_t->vfs_mntpnt);
> /* use mntpnt value */
> refstr_rele(vfs_t->vfs_mntpnt);', are there any
> situations where the value might be invalid (and
> could cause a panic or other nasty behavior)?

You should be alright; Although, I'd suggest using 

mntpoint = vfs_getmntpoint(vfs_t)
[...]
refstr_rele(mntpoint); 

as this will lock the vfs_list while it gets the mntpoint.

> 2. Similarly, I believe that vnode_t->v_path is not
> always guaranteed to be valid (i.e. it may be NULL).
> In this situation, I am thinking to merely omit this
> it from the output, but again, is there any special
>  conditions I need to be aware of before using it?

Not as such; You might also want to consider vnodetopath() if the v_path is NULL. 

-Drew
 
 
This message posted from opensolaris.org
_______________________________________________
ufs-discuss mailing list
ufs-discuss@opensolaris.org
[prev in list] [next in list] [prev in thread] [next in thread]