linux-kernel - Re: [PATCH 1/2] fs/efivarfs: restrict inode permissions

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

Date:   Tue, 20 Feb 2018 23:39:27 +0000
From:   Matthew Garrett <mjg59@...gle.com>
To:     tony.luck@...el.com
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        joe.konno@...ux.intel.com, linux-efi <linux-efi@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Ard Biesheuvel <ard.biesheuvel@...aro.org>,
        matthew.garrett@...ula.com, Jeremy Kerr <jk@...abs.org>,
        ak@...ux.intel.com, pjones@...hat.com, luto@...nel.org,
        James Bottomley <james.bottomley@...senpartnership.com>
Subject: Re: [PATCH 1/2] fs/efivarfs: restrict inode permissions

On Tue, Feb 20, 2018 at 3:30 PM Luck, Tony <tony.luck@...el.com> wrote:
> [1] I didn't dig through the Linux code to check whether we manage to
> get those four SMIs from a single EFI call, or whether we make multiple
> EFI calls to open/read/close one file. It is possible that we stink a
> bit too if we are doing more EFI calls than required.

read() will make two calls - one to obtain the size of the variable, the
other to read it. It looks like cat will also trigger an fstat(), so we're
probably also making a call for that. There's presumably some optimisation
that could be made there if we trust the firmware not to change the size
behind our back.