linux-kernel - Re: mmotm 2009-04-10-02-21 uploaded

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:	Wed, 15 Apr 2009 01:34:56 -0700
From:	Andrew Morton <akpm@...ux-foundation.org>
To:	Ali Gholami Rudi <ali@...i.ir>
Cc:	Ingo Molnar <mingo@...e.hu>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Valdis.Kletnieks@...edu, Mike Travis <travis@....com>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	mm-commits@...r.kernel.org, Rusty Russell <rusty@...tcorp.com.au>,
	Dave Jones <davej@...hat.com>, Len Brown <lenb@...nel.org>
Subject: Re: mmotm 2009-04-10-02-21 uploaded - forkbombed by work_for_cpu

On Wed, 15 Apr 2009 12:45:34 +0430 Ali Gholami Rudi <ali@...i.ir> wrote:

> Hi,
> 
> In today's tip (v2.6.30-rc2), when my cpu is idle (and the ondemand
> governor correctly uses the lowest frequency) the temperature of my CPU
> rises to above 50^C till the fan turns on (it used to be about 40^C
> before).  Git bisect points to this patch:
> 
>   commit 01599fca6758d2cd133e78f87426fc851c9ea725
>   Author: Andrew Morton <akpm@...ux-foundation.org>
>   Date:   Mon Apr 13 10:27:49 2009 -0700
>   
>       cpufreq: use smp_call_function_[single|many]() in acpi-cpufreq.c
>       
>       Atttempting to rid us of the problematic work_on_cpu().  Just use
>       smp_call_fuction_single() here.
>       
>       This repairs a 10% sysbench(oltp)+mysql regression which Mike reported,
>       due to
>       
>         commit 6b44003e5ca66a3fffeb5bc90f40ada2c4340896
>         Author: Andrew Morton <akpm@...ux-foundation.org>
>         Date:   Thu Apr 9 09:50:37 2009 -0600
>       
>             work_on_cpu(): rewrite it to create a kernel thread on demand
>       
>       It seems that the kernel calls these acpi-cpufreq functions at a quite
>       high frequency.
>       
>       Valdis Kletnieks also reports that this causes 70-90 forks per second on
>       his hardware.
>       
>       Cc: Valdis.Kletnieks@...edu
>       Cc: Rusty Russell <rusty@...tcorp.com.au>
>       Cc: Venkatesh Pallipadi <venkatesh.pallipadi@...el.com>
>       Cc: Len Brown <len.brown@...el.com>
>       Cc: Zhao Yakui <yakui.zhao@...el.com>
>       Acked-by: Dave Jones <davej@...hat.com>
>       Cc: Thomas Gleixner <tglx@...utronix.de>
>       Tested-by: Mike Galbraith <efault@....de>
>       Cc: "Zhang, Yanmin" <yanmin_zhang@...ux.intel.com>
>       Signed-off-by: Andrew Morton <akpm@...ux-foundation.org>
>       Acked-by: Ingo Molnar <mingo@...e.hu>
>       [ Made it use smp_call_function_many() instead of looping over cpu's
>         with smp_call_function_single()    - Linus ]

<stares suspiciously at smp_call_function_many()>

 * smp_call_function_many(): Run a function on a set of other CPUs.

"other".  It refuses to call the function on *this* CPU.  Tricky.

Does this fix it up?

--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c~a
+++ a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -204,7 +204,10 @@ static void drv_read(struct drv_cmd *cmd
 
 static void drv_write(struct drv_cmd *cmd)
 {
-	smp_call_function_many(cmd->mask, do_drv_write, cmd, 1);
+	unsigned cpu;
+
+	for_each_cpu(cpu, cmd->mask)
+		smp_call_function_single(cpu, do_drv_write, cmd, 1);	
 }
 
 static u32 get_cur_val(const struct cpumask *mask)
_

--
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/