NetBSD-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: bin/55815: tar opens device files



The following reply was made to PR bin/55815; it has been noted by GNATS.

From: Joerg Sonnenberger <joerg%bec.de@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost, martin%NetBSD.org@localhost
Subject: Re: bin/55815: tar opens device files
Date: Wed, 25 Nov 2020 19:06:26 +0100

 On Wed, Nov 25, 2020 at 03:55:01PM +0000, Robert Elz wrote:
 >  I am not sure I understand the point here though, why is tar opening
 >  files at all if it doesn't plan on reading from them?   Is this some
 >  kind of optimisation attempt, where since most files will be regular
 >  filesystem files, and will need to be opened, open() then fstat() has
 >  one less pathname lookup (namei) than stat() followed when needed by open()
 >  (the normal case)?    If so, we should just stop doing that, as it is broken,
 >  and revert to the stat()+open() model (where the open only happens if we
 >  want the data contained in the file, rather than just its name and attributes).
 >  We need that for symnlinks too (ie: the stat() would actually be lstat()),
 >  surely.
 
 No, it is about ensuring that the extattr and ACL actually belong to the
 same object as the rest of the permissions.
 
 Joerg
 


Home | Main Index | Thread Index | Old Index