linux-kernel - Re: Edited seccomp.2 man page for review [v2]

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:	Mon, 5 Jan 2015 12:25:40 -0800
From:	Kees Cook <keescook@...omium.org>
To:	"Michael Kerrisk (man-pages)" <mtk.manpages@...il.com>
Cc:	Daniel Borkmann <dborkman@...hat.com>,
	Linux API <linux-api@...r.kernel.org>,
	"linux-man@...r.kernel.org" <linux-man@...r.kernel.org>,
	lkml <linux-kernel@...r.kernel.org>,
	Will Drewry <wad@...omium.org>
Subject: Re: Edited seccomp.2 man page for review [v2]

On Tue, Dec 30, 2014 at 11:45 AM, Michael Kerrisk (man-pages)
<mtk.manpages@...il.com> wrote:
>>> The program counter will be as though the system call happened
>>> (i.e., it will not point to the system call instruction).
>>> The return value register will contain an architecture\-dependent value;
>>> if resuming execution, set it to something sensible.
>>> .\" FIXME Regarding the preceding line, can you give an example(s)
>>> .\"       of "something sensible"? (Depending on the answer, maybe it
>>> .\"       might be useful to add some text on this point.)
>>
>> This means sensible in the context of the syscall made, or the desired
>> behavior. For example, setting the return value to ELOOP for something
>> like a "bind" syscall isn't very sensible.
>
> Okay -- I did s/sensible/appropriate for the system call/

Yes, perfect. That captures it nicely.

>>> .\"
>>> .\" FIXME Please check:
>>> .\"     In an attempt to make the text clearer, I changed
>>> .\"     "replacing it with" to "setting the return value register to"
>>> .\"     Okay?
>>> (The architecture dependency is because setting the return value register to
>>> .BR ENOSYS
>>> could overwrite some useful information.)
>>
>> Well, the arch dependency is really because _how_ to change the
>> register, and the register itself, is different between architectures.
>> (i.e. which ptrace call is needed, and which register is being
>> changed.) The overwriting of useful information is certainly true too,
>> though.
>
> So, revert to the previous wording? Or do you have a suggested
> better wording?

I think the previous wording is better. I'm struggling to produce
language that makes more sense here.

> Thanks. We're getting close now.

Excellent! :)

-Kees

-- 
Kees Cook
Chrome OS Security
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/