linux-kernel - Re: [GIT PULL] RCU changes for v2.6.38

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 for Android: free password hash cracker in your pocket

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

Date:	Fri, 7 Jan 2011 00:54:57 +0100
From:	Ingo Molnar <mingo@...e.hu>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	linux-kernel@...r.kernel.org,
	"Paul E. McKenney" <paulmck@...ibm.com>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Thomas Gleixner <tglx@...utronix.de>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [GIT PULL] RCU changes for v2.6.38


* Linus Torvalds <torvalds@...ux-foundation.org> wrote:

> On Thu, Jan 6, 2011 at 1:08 AM, Ingo Molnar <mingo@...e.hu> wrote:
> >
> > Please pull the latest core-rcu-for-linus git tree from:
> 
> So I pulled, but I'm not convinced about the crazy busy-looping
> SRCU_SYNCHRONIZE_DELAY thing.
> 
> Why does it do a silly udelay(), instead of just looping over the
> srcu_readers_active_idx() for a few times? You're wasting CPU time
> anyway, why ask the user how many usecs to waste?
> 
> IOW, why isn't that "wait for no active readers" a nice helper
> function, and why doesn't it do
> 
>    for (i = 0; i < CONFIG_SRCU_SYNCHRONIZE_DELAY; i++) {
>       if (!srcu_readers_active_idx(sp, idx))
>          return;
>       udelay(1);
>    }
>    while (srcu_readers_active_idx(sp, idx))
>       schedule_timeout_interruptible(1);
> 
> instead? And is it really sane to ask the kernel configurator to come
> up with a random value (ie that "CONFIG_SRCU_SYNCHRONIZE_DELAY" is
> just stupid and wrong)?
> 
> Please fix this. And don't make people answer unanswerable questions.
> If YOU and Paul don't know the answer, why the hell do you expect
> somebody who does a "make config" to know the answer?
> 
> Either pick a number, or pick an algorithm that self-tunes.
> 
> Don't use the Kconfig system as a way to tell people that it's their
> fault when you made a bad decision. Really.

Fully agreed - we'll fix this.

Thanks,

	Ingo
--
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/