linux-kernel - Re: [PATCH 09/18] frv: switch do_timer() to xtime

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: <alpine.LFD.2.00.1101221051570.2971@localhost6.localdomain6>
Date:	Sat, 22 Jan 2011 11:01:49 +0100 (CET)
From:	Thomas Gleixner <tglx@...utronix.de>
To:	Torben Hohn <torbenh@....de>
cc:	linux-kernel@...r.kernel.org, johnstul@...ibm.com,
	hch@...radead.org, yong.zhang0@...il.com
Subject: Re: [PATCH 09/18] frv: switch do_timer() to xtime_update()

On Sat, 22 Jan 2011, Torben Hohn wrote:

> this code looks like its protecting __set_LEDS() with this lock also.
> i dont think thats necessary.

This changelog is horrible. 

This code does not look like protecting the __set_LEDS() call with the
xtime_lock. The call happens to be inside the xtime_lock held region.

Now instead of bringing up a weak argument "I dont think ..." you
should provide a proper analysis why it's safe to move that code out.

It's pretty simple:

No other call site of __set_LEDS() is protected by xtime_lock, so
xtime_lock does not protect anything related to __set_LEDS(). It's
just inside the xtime_lock region for no good reason at all.

Please be more careful when writing change logs, so a reviewer can
understand the reasoning behind your change easily.

Also all arch/* patches are missing a "Cc: arch-maintainer@...ewhere.xxx".

>  {
> @@ -61,10 +61,11 @@ static irqreturn_t timer_interrupt(int irq, void *dummy)
>  	 * CPU. We need to avoid to SMP race with it. NOTE: we don't need
>  	 * the irq version of write_lock because as just said we have irq
>  	 * locally disabled. -arca
> +	 *
> +	 * xtime_update takes the writelock.

  Errm. xtime_update write locks xtime_lock. Please be careful with comments.

Thanks,

	tglx
--
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/