linux-kernel - Re: [PATCH] x86,sched: Fix sched_smt_power

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:	Tue, 10 Jan 2012 22:14:55 +0530
From:	Vaidyanathan Srinivasan <svaidy@...ux.vnet.ibm.com>
To:	Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc:	Youquan Song <youquan.song@...el.com>,
	linux-kernel@...r.kernel.org, mingo@...e.hu, tglx@...utronix.de,
	hpa@...or.com, akpm@...ux-foundation.org, stable@...r.kernel.org,
	suresh.b.siddha@...el.com, arjan@...ux.intel.com,
	len.brown@...el.com, anhua.xu@...el.com, chaohong.guo@...el.com,
	Youquan Song <youquan.song@...ux.intel.com>
Subject: Re: [PATCH] x86,sched: Fix sched_smt_power_savings totally broken

* Peter Zijlstra <a.p.zijlstra@...llo.nl> [2012-01-09 12:05:56]:

> On Mon, 2012-01-09 at 19:14 -0500, Youquan Song wrote:
> > Fine, I will base your suggestion to develop another patch soon.
> > 
> 
> > 
> > @@ -3923,6 +3923,10 @@ static inline void update_sg_lb_stats(struct
> > sched_domain *sd,
> >                                                 SCHED_POWER_SCALE);
> >         if (!sgs->group_capacity)
> >                 sgs->group_capacity = fix_small_capacity(sd, group);
> > +
> > +       if (sched_smt_power_savings)
> > +               sgs->group_capacity *= 2; 
> 
> Note, this has the hard-coded assumption you only have 2 threads per
> core, which while true for intel, isn't true in general. I think you
> meant to write *= group->group_weight or somesuch.
> 
> Also, you forgot to limit this to the SD_SHARE_CPUPOWER domain, you're
> now doubling the capacity for all domains.
> 
> Furthermore, have a look at the SD_PREFER_SIBLING logic and make sure
> you're not fighting that.

Hi Peter,

I think I had proposed the following fix earlier.  Can we revisit this
now?  This works, but it is not the best solution.

    sched: Fix group_capacity for sched_smt_powersavings
    
    sched_smt_powersavings for threaded systems need this fix for
    consolidation to sibling threads to work.  Since threads have
    fractional capacity, group_capacity will turn out to be one
    always and not accommodate another task in the sibling thread.
    
    This fix makes group_capacity a function of cpumask_weight that
    will enable the power saving load balancer to pack tasks among
    sibling threads and keep more cores idle.
    
    Signed-off-by: Vaidyanathan Srinivasan <svaidy@...ux.vnet.ibm.com>

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 8e42de9..77ac4ca 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4010,6 +4010,21 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu,
 		 */
 		if (prefer_sibling && !local_group && sds->this_has_capacity)
 			sgs.group_capacity = min(sgs.group_capacity, 1UL);
+		/*
+		 * If power savings balance is set at this domain, then
+		 * make capacity equal to number of hardware threads to
+		 * accommodate more tasks until capacity is reached.
+		 */
+		else if (sd->flags & SD_POWERSAVINGS_BALANCE)
+			sgs.group_capacity =
+				cpumask_weight(sched_group_cpus(sg));
+
+			/*
+			 * The default group_capacity is rounded from sum of
+			 * fractional cpu_powers of sibling hardware threads
+			 * in order to enable fair use of available hardware
+			 * resources.
+			 */
 
 		if (local_group) {
 			sds->this_load = sgs.avg_load;
@@ -4432,7 +4447,8 @@ static int need_active_balance(struct sched_domain *sd, int idle,
 		 * move_tasks() will succeed.  ld_moved will be true and this
 		 * active balance code will not be triggered.
 		 */
-		if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
+		if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP &&
+		    sched_smt_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
 			return 0;
 	}
 

--
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/