linux-kernel - Re: [PATCH] staging: lustre: ko2iblbd: handle ib_dereg

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20161003073104.GA28412@infradead.org>
Date:   Mon, 3 Oct 2016 00:31:04 -0700
From:   Christoph Hellwig <hch@...radead.org>
To:     James Simmons <jsimmons@...radead.org>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        devel@...verdev.osuosl.org,
        Andreas Dilger <andreas.dilger@...el.com>,
        Oleg Drokin <oleg.drokin@...el.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Lustre Development List <lustre-devel@...ts.lustre.org>,
        linux-rdma@...r.kernel.org
Subject: Re: [PATCH] staging: lustre: ko2iblbd: handle ib_dereg_mr removal

Hell no.  If you had actually read the changelog you'd see that there
is a reason the functions went away, and simply calling the underlying
methods (which also will go away) will not fix it.

Please go back reading the changelogs, and while we're at it read some
basic introductions into the RDMA programming model.  As a little
shortcut: relying on any form of a global registration for the rkey
is inherently unsafe and only allowed as an option variant for
very trusted environments (thing single box appliances).  You'll need
to do a real memory registration here and not drag that old crap along.