linux-kernel - Re: [PATCH] omit POSIX timer stuff from task

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]

Message-ID: <CALAqxLV1WVkHS4D87zmrZz541CJ110xOwJLmf9BL-tCZw7RMQA@mail.gmail.com>
Date:   Fri, 20 Jan 2017 14:48:23 -0800
From:   John Stultz <john.stultz@...aro.org>
To:     Nicolas Pitre <nicolas.pitre@...aro.org>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        lkml <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] omit POSIX timer stuff from task_struct when disabled

On Fri, Jan 20, 2017 at 2:46 PM, Nicolas Pitre <nicolas.pitre@...aro.org> wrote:
> On Fri, 20 Jan 2017, John Stultz wrote:
>
>> On Mon, Jan 9, 2017 at 10:49 AM, Nicolas Pitre <nicolas.pitre@...aro.org> wrote:
>> > When CONFIG_POSIX_TIMERS is disabled, it is preferable to remove related
>> > structures from struct task_struct and struct signal_struct as they
>> > won't contain anything useful and shouldn't be relied upon by mistake.
>> > Code still referencing those structures is also disabled here.
>> >
>> > Signed-off-by: Nicolas Pitre <nico@...aro.org>
>> >
>> [snip]
>> > diff --git a/kernel/fork.c b/kernel/fork.c
>> > index 11c5c8ab82..8e333e55a9 100644
>> > --- a/kernel/fork.c
>> > +++ b/kernel/fork.c
>> > @@ -1309,6 +1309,7 @@ void __cleanup_sighand(struct sighand_struct *sighand)
>> >   */
>> >  static void posix_cpu_timers_init_group(struct signal_struct *sig)
>> >  {
>> > +#ifdef CONFIG_POSIX_TIMERS
>> >         unsigned long cpu_limit;
>> >
>> >         cpu_limit = READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur);
>> > @@ -1321,6 +1322,7 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig)
>> >         INIT_LIST_HEAD(&sig->cpu_timers[0]);
>> >         INIT_LIST_HEAD(&sig->cpu_timers[1]);
>> >         INIT_LIST_HEAD(&sig->cpu_timers[2]);
>> > +#endif
>> >  }
>>
>> So apologies for not catching this earlier. I was just queuing this
>> up, when I noticed the style issue here.
>>
>> Aren't in-function ifdefs frowned upon? Wouldn't it be better to do:
>> #ifdef CONFIG_POSIX_TIMERS
>> static void posix_cpu_timers_init_group(struct signal_struct *sig)
>> {
>> ...
>> }
>> #else
>> static void posix_cpu_timers_init_group(struct signal_struct *sig) {}
>> #endif
>>
>> And similar for most of the ifdef'ed out functions in this patch?
>
> Well... I don't mind either ways.  In this case those functions are
> rather small and doing it the way you suggest doubles the number of
> added lines in this hunk. That's why I opted for the current style.
>
> Just tell me if you prefer that I respin the patch and I'll do it.

Yea. I'm not so finicky but I'm sure I'll probably get yelled at if I
pass it on, so if you don't mind respinning it, I'll get it applied
for testing here quickly :)

thanks
-john