linux-kernel - Re: [PATCH v4 5/7] mm: rework non-root kmem

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: <20190520175640.GA24204@tower.DHCP.thefacebook.com>
Date:   Mon, 20 May 2019 17:56:46 +0000
From:   Roman Gushchin <guro@...com>
To:     Waiman Long <longman9394@...il.com>
CC:     Shakeel Butt <shakeelb@...gle.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Linux MM <linux-mm@...ck.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Kernel Team <Kernel-team@...com>,
        "Johannes Weiner" <hannes@...xchg.org>,
        Michal Hocko <mhocko@...nel.org>,
        Rik van Riel <riel@...riel.com>,
        Christoph Lameter <cl@...ux.com>,
        Vladimir Davydov <vdavydov.dev@...il.com>,
        Cgroups <cgroups@...r.kernel.org>
Subject: Re: [PATCH v4 5/7] mm: rework non-root kmem_cache lifecycle
 management

On Mon, May 20, 2019 at 10:54:24AM -0400, Waiman Long wrote:
> On 5/14/19 8:06 PM, Shakeel Butt wrote:
> > diff --git a/mm/slab_common.c b/mm/slab_common.c
> > index 4e5b4292a763..1ee967b4805e 100644
> > --- a/mm/slab_common.c
> > +++ b/mm/slab_common.c
> > @@ -45,6 +45,8 @@ static void slab_caches_to_rcu_destroy_workfn(struct work_struct *work);
> >  static DECLARE_WORK(slab_caches_to_rcu_destroy_work,
> >                     slab_caches_to_rcu_destroy_workfn);
> >
> > +static void kmemcg_queue_cache_shutdown(struct percpu_ref *percpu_ref);
> > +
> 
> kmemcg_queue_cache_shutdown is only defined if CONFIG_MEMCG_KMEM is
> defined. If it is not defined, a compilation warning can be produced.
> Maybe putting the declaration inside a CONFIG_MEMCG_KMEM block:

Hi Waiman!

Yes, that makes total sense to me. Thank you for letting me know!
How about this one?

--

>From 0fa19369adc240cc93281911a59713822a4f3e07 Mon Sep 17 00:00:00 2001
From: Roman Gushchin <guro@...com>
Date: Mon, 20 May 2019 10:52:07 -0700
Subject: [PATCH] mm: guard kmemcg_queue_cache_shutdown() with
 CONFIG_MEMCG_KMEM

Currently kmemcg_queue_cache_shutdown() is defined only
if CONFIG_MEMCG_KMEM is set, however the declaration is not guarded
with corresponding ifdefs. So a compilation warning might be produced.

Let's move the declaration to the section of slab_common.c, where all
kmemcg-specific stuff is defined.

Reported-by: Waiman Long <longman9394@...il.com>
Signed-off-by: Roman Gushchin <guro@...com>
---
 mm/slab_common.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/slab_common.c b/mm/slab_common.c
index 9d2a3d6245dc..e818609c8209 100644
--- a/mm/slab_common.c
+++ b/mm/slab_common.c
@@ -45,8 +45,6 @@ static void slab_caches_to_rcu_destroy_workfn(struct work_struct *work);
 static DECLARE_WORK(slab_caches_to_rcu_destroy_work,
 		    slab_caches_to_rcu_destroy_workfn);
 
-static void kmemcg_queue_cache_shutdown(struct percpu_ref *percpu_ref);
-
 /*
  * Set of flags that will prevent slab merging
  */
@@ -134,6 +132,8 @@ int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t nr,
 LIST_HEAD(slab_root_caches);
 static DEFINE_SPINLOCK(memcg_kmem_wq_lock);
 
+static void kmemcg_queue_cache_shutdown(struct percpu_ref *percpu_ref);
+
 void slab_init_memcg_params(struct kmem_cache *s)
 {
 	s->memcg_params.root_cache = NULL;
-- 
2.20.1