linux-kernel - Re: [PATCH] add i386 idle notifier (take 3)

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] [day] [month] [year] [list]

Message-ID: <20070109104907.GE27034@frankl.hpl.hp.com>
Date:	Tue, 9 Jan 2007 02:49:07 -0800
From:	Stephane Eranian <eranian@....hp.com>
To:	Adrian Bunk <bunk@...sta.de>
Cc:	Andrew Morton <akpm@...l.org>, linux-kernel@...r.kernel.org,
	ak@...e.de, Stephane Eranian <eranian@....hp.com>
Subject: Re: [PATCH] add i386 idle notifier (take 3)

Hello,

On Fri, Jan 05, 2007 at 02:36:53PM +0100, Adrian Bunk wrote:
> > > > > Where does the perfmon code use the EXPORT_SYMBOL's?
> > > > 
> > > > The perfmon patch includes several kernel modules which make use of
> > > > the exported entry points. The following symbols are exported:
> > > > 
> > > > pfm_pmu_register/pfm_pmu_unregister:
> > > > 	* PMU description module registration.
> > > > 	* Used to describe PMU model.
> > > > 	* Used by perfmon_p4.c, perfmon_core.c, perfmon_mckinley.c, and others
> > > > 
> > > > pfm_fmt_register/pfm_fmt_unregister:
> > > > 	* Sampling format module registration
> > > > 	* Used by perfmon_dfl_smpl.c, perfmon_pebs_smpl.c
> > > > 
> > > > pfm_interrupt_handler:
> > > > 	* PMU interrupt handler
> > > > 	* Used by MIPS-specific perfmon code
> > > > 
> > > > pfm_pmu_conf/pfm_controls:
> > > > 	* global state/control variable
> > > > 
> > > > All exported symbols are currently used. Why are you saying this adds bloat?
> > > 
> > > Which module uses idle_notifier_register/idle_notifier_unregister?
> > > 
> > None.
> > 
> > I have no issue with removing the EXPORT_SYMBOL on i386 and x86_64 if you
> > think that would help.
> 

As a follow-up to this discussion, here is a patch that remove unused
EXPORT_SYMBOL by the i386 idle notifier. It also make it more explicit
that the enter_idle() store must be atomic.

changelog:
	- do not export i386 idle notification registration entry points
	  because there is currently no user for this
	- make it more explicit that the store to idle_state must be atomic

signed-off-by: stephane eranian <eranian@....hp.com>

--- linux-2.6.20-rc3-mm1.orig/arch/i386/kernel/process.c	2007-01-08 01:58:08.000000000 -0800
+++ linux-2.6.20-rc3-mm1/arch/i386/kernel/process.c	2007-01-08 02:01:59.000000000 -0800
@@ -87,19 +87,18 @@ void idle_notifier_register(struct notif
 {
 	atomic_notifier_chain_register(&idle_notifier, n);
 }
-EXPORT_SYMBOL_GPL(idle_notifier_register);
 
 void idle_notifier_unregister(struct notifier_block *n)
 {
 	atomic_notifier_chain_unregister(&idle_notifier, n);
 }
-EXPORT_SYMBOL(idle_notifier_unregister);
 
 static DEFINE_PER_CPU(volatile unsigned long, idle_state);
 
 void enter_idle(void)
 {
-	__get_cpu_var(idle_state) = 1;
+	/* needs to be atomic w.r.t. interrupts, not against other CPUs */
+	__set_bit(0, &__get_cpu_var(idle_state));
 	atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
 }
 
-
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/