linux-kernel - Re: [PATCH 2/2] sched/deadline: Don't return invalid cpu in cpudl_maximum

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: <20170607001454.GE3623@X58A-UD3R>
Date:   Wed, 7 Jun 2017 09:14:54 +0900
From:   Byungchul Park <byungchul.park@....com>
To:     Juri Lelli <juri.lelli@....com>
Cc:     peterz@...radead.org, mingo@...nel.org,
        linux-kernel@...r.kernel.org, juri.lelli@...il.com,
        rostedt@...dmis.org, kernel-team@....com
Subject: Re: [PATCH 2/2] sched/deadline: Don't return invalid cpu in
 cpudl_maximum_cpu()

On Wed, Jun 07, 2017 at 08:42:24AM +0900, Byungchul Park wrote:
> On Tue, Jun 06, 2017 at 04:12:25PM +0100, Juri Lelli wrote:
> > Hi,
> > 
> > On 02/06/17 16:31, Byungchul Park wrote:
> > > When the heap tree is empty, cp->elements[0].cpu has meaningless value.
> 
> Hi,
> 
> The meaningless value is 0.
> 
> > > We need to consider the case.
> > > 
> > > Signed-off-by: Byungchul Park <byungchul.park@....com>
> > > ---
> > >  kernel/sched/cpudeadline.c | 3 ++-
> > >  1 file changed, 2 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
> > > index d4a6963..9b314a9 100644
> > > --- a/kernel/sched/cpudeadline.c
> > > +++ b/kernel/sched/cpudeadline.c
> > > @@ -110,7 +110,8 @@ static void cpudl_heapify(struct cpudl *cp, int idx)
> > >  
> > >  static inline int cpudl_maximum_cpu(struct cpudl *cp)
> > >  {
> > > -	return cp->elements[0].cpu;
> > > +	int cpu = cp->elements[0].cpu;
> > > +	return cp->elements[cpu].idx == IDX_INVALID ? -1 : cpu;
> > 
> > Mmm, don't we get a WARN from cpumask_check() if we return -1 here?
> 
> The function does not return -1 without my patch.
> 
> Right?

Or the following patch would be needed, instead.

----->8-----

>From cada1345bf0ff8e6b5743999509d2abcacd79a9e Mon Sep 17 00:00:00 2001
From: Byungchul Park <byungchul.park@....com>
Date: Wed, 7 Jun 2017 09:05:34 +0900
Subject: [PATCH 2/2] sched/deadline: Initialize cp->elements[].cpu to an
 invalid value

Currently, when the heap tree is empty, cpudl_maximum_cpu() returns 0,
which causes unnecessary migration. It has to return an invalid value
e.g. -1 to prevent that.

Signed-off-by: Byungchul Park <byungchul.park@....com>
---
 kernel/sched/cpudeadline.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
index d4a6963..0f67cea 100644
--- a/kernel/sched/cpudeadline.c
+++ b/kernel/sched/cpudeadline.c
@@ -266,8 +266,10 @@ int cpudl_init(struct cpudl *cp)
 		return -ENOMEM;
 	}
 
-	for_each_possible_cpu(i)
+	for_each_possible_cpu(i) {
+		cp->elements[i].cpu = -1;
 		cp->elements[i].idx = IDX_INVALID;
+	}
 
 	return 0;
 }
-- 
1.9.1