Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1290229 > unrolled thread

Re: [PATCH 2/4] mm: memcontrol: reign in the CONFIG space madness

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH 2/4] mm: memcontrol: reign in the CONFIG space madness Vladimir Davydov <vdavydov@virtuozzo.com> - 2015-12-12 17:40 +0100
    Re: [PATCH 2/4] mm: memcontrol: reign in the CONFIG space madness Johannes Weiner <hannes@cmpxchg.org> - 2015-12-12 18:30 +0100
      Re: [PATCH 2/4] mm: memcontrol: reign in the CONFIG space madness Andrew Morton <akpm@linux-foundation.org> - 2015-12-23 00:20 +0100
        Re: [PATCH 2/4] mm: memcontrol: reign in the CONFIG space madness Andrew Morton <akpm@linux-foundation.org> - 2015-12-23 00:20 +0100
          Re: [PATCH 2/4] mm: memcontrol: reign in the CONFIG space madness Johannes Weiner <hannes@cmpxchg.org> - 2015-12-23 00:40 +0100

#1290229 — Re: [PATCH 2/4] mm: memcontrol: reign in the CONFIG space madness

On Fri, Dec 11, 2015 at 02:54:11PM -0500, Johannes Weiner wrote:
> What CONFIG_INET and CONFIG_LEGACY_KMEM guard inside the memory
> controller code is insignificant, having these conditionals is not
> worth the complication and fragility that comes with them.
> 
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>

Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>

> @@ -4374,17 +4342,11 @@ static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
>  {
>  	struct mem_cgroup *memcg = mem_cgroup_from_css(css);
>  
> -#ifdef CONFIG_INET
>  	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
>  		static_branch_dec(&memcg_sockets_enabled_key);
> -#endif
> -
> -	memcg_free_kmem(memcg);

I wonder where the second call to memcg_free_kmem comes from. Luckily,
it couldn't result in a breakage. And now it's removed.

>  
> -#if defined(CONFIG_MEMCG_LEGACY_KMEM) && defined(CONFIG_INET)
>  	if (memcg->tcp_mem.active)
>  		static_branch_dec(&memcg_sockets_enabled_key);
> -#endif
>  
>  	memcg_free_kmem(memcg);
>  
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1290247

On Sat, Dec 12, 2015 at 07:33:32PM +0300, Vladimir Davydov wrote:
> On Fri, Dec 11, 2015 at 02:54:11PM -0500, Johannes Weiner wrote:
> > What CONFIG_INET and CONFIG_LEGACY_KMEM guard inside the memory
> > controller code is insignificant, having these conditionals is not
> > worth the complication and fragility that comes with them.
> > 
> > Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> 
> Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
> 
> > @@ -4374,17 +4342,11 @@ static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
> >  {
> >  	struct mem_cgroup *memcg = mem_cgroup_from_css(css);
> >  
> > -#ifdef CONFIG_INET
> >  	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
> >  		static_branch_dec(&memcg_sockets_enabled_key);
> > -#endif
> > -
> > -	memcg_free_kmem(memcg);
> 
> I wonder where the second call to memcg_free_kmem comes from. Luckily,
> it couldn't result in a breakage. And now it's removed.

Lol, I had to double check my trees to see what's going on as I don't
remember this being part of the patch. But it looks like the double
free came from the "net: drop tcp_memcontrol.c" patch and I must have
removed it again during conflict resolution when rebasing this patch
on top of yours. I must have thought git's auto-merge added it.

However, this causes an underflow of the kmem static branch, so we
will have to fix this directly in "net: drop tcp_memcontrol.c".

Andrew, could you please pick this up? However, it's important to also
then remove the hunk above from THIS patch, the one that deletes the
excessive memcg_free_kmem(). We need exactly one memcg_free_kmem() in
mem_cgroup_css_free(). :-)

From 94a14b7b0f7ed5b5ac88ca285a7e8ec3215ea59c Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@cmpxchg.org>
Date: Sat, 12 Dec 2015 12:14:31 -0500
Subject: [PATCH] net: drop tcp_memcontrol.c fix

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
---
 mm/memcontrol.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 5cf7fd2..a4ce8d3 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -4386,8 +4386,6 @@ static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
 		static_branch_dec(&memcg_sockets_enabled_key);
 #endif
 
-	memcg_free_kmem(memcg);
-
 	__mem_cgroup_free(memcg);
 }
 
-- 
2.6.4
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1297141

On Sat, 12 Dec 2015 12:20:57 -0500 Johannes Weiner <hannes@cmpxchg.org> wrote:

> On Sat, Dec 12, 2015 at 07:33:32PM +0300, Vladimir Davydov wrote:
> > On Fri, Dec 11, 2015 at 02:54:11PM -0500, Johannes Weiner wrote:
> > > What CONFIG_INET and CONFIG_LEGACY_KMEM guard inside the memory
> > > controller code is insignificant, having these conditionals is not
> > > worth the complication and fragility that comes with them.
> > > 
> > > Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> > 
> > Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
> > 
> > > @@ -4374,17 +4342,11 @@ static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
> > >  {
> > >  	struct mem_cgroup *memcg = mem_cgroup_from_css(css);
> > >  
> > > -#ifdef CONFIG_INET
> > >  	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
> > >  		static_branch_dec(&memcg_sockets_enabled_key);
> > > -#endif
> > > -
> > > -	memcg_free_kmem(memcg);
> > 
> > I wonder where the second call to memcg_free_kmem comes from. Luckily,
> > it couldn't result in a breakage. And now it's removed.
> 
> Lol, I had to double check my trees to see what's going on as I don't
> remember this being part of the patch. But it looks like the double
> free came from the "net: drop tcp_memcontrol.c" patch and I must have
> removed it again during conflict resolution when rebasing this patch
> on top of yours. I must have thought git's auto-merge added it.
> 
> However, this causes an underflow of the kmem static branch, so we
> will have to fix this directly in "net: drop tcp_memcontrol.c".
> 
> Andrew, could you please pick this up? However, it's important to also
> then remove the hunk above from THIS patch, the one that deletes the
> excessive memcg_free_kmem(). We need exactly one memcg_free_kmem() in
> mem_cgroup_css_free(). :-)

So you want to retain
mm-memcontrol-reign-in-the-config-space-madness.patch's removal of the
ifdef CONFIG_INET?


What I have is

Against net-drop-tcp_memcontrolc.patch:

--- a/mm/memcontrol.c~net-drop-tcp_memcontrolc-fix
+++ a/mm/memcontrol.c
@@ -4421,8 +4421,6 @@ static void mem_cgroup_css_free(struct c
 		static_branch_dec(&memcg_sockets_enabled_key);
 #endif
 
-	memcg_free_kmem(memcg);
-
 	__mem_cgroup_free(memcg);
 }
 

and against mm-memcontrol-reign-in-the-config-space-madness.patch:

--- a/mm/memcontrol.c~mm-memcontrol-reign-in-the-config-space-madness-fix
+++ a/mm/memcontrol.c
@@ -4380,6 +4380,8 @@ static void mem_cgroup_css_free(struct c
 	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
 		static_branch_dec(&memcg_sockets_enabled_key);
 
+	memcg_free_kmem(memcg);
+
 	if (memcg->tcp_mem.active)
 		static_branch_dec(&memcg_sockets_enabled_key);
 

Producing

static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
{
	struct mem_cgroup *memcg = mem_cgroup_from_css(css);

	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
		static_branch_dec(&memcg_sockets_enabled_key);

	memcg_free_kmem(memcg);

	if (memcg->tcp_mem.active)
		static_branch_dec(&memcg_sockets_enabled_key);

	__mem_cgroup_free(memcg);
}

And I did s/reign/rein/;)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1297142

On Tue, 22 Dec 2015 15:11:38 -0800 Andrew Morton <akpm@linux-foundation.org> wrote:

> What I have is

And after a bit of reject resolution in
mm-memcontrol-clean-up-alloc-online-offline-free-functions.patch we
have


static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
{
	struct mem_cgroup *memcg = mem_cgroup_from_css(css);

	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
		static_branch_dec(&memcg_sockets_enabled_key);

	vmpressure_cleanup(&memcg->vmpressure);
	cancel_work_sync(&memcg->high_work);
	mem_cgroup_remove_from_trees(memcg);
	memcg_free_kmem(memcg);

	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && memcg->tcpmem_active)
		static_branch_dec(&memcg_sockets_enabled_key);

	mem_cgroup_free(memcg);
}

code looks a bit strange.  Can we move the static_branch_dec's together
and run cgroup_subsys_on_dfl just once?

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1297145

On Tue, Dec 22, 2015 at 03:15:27PM -0800, Andrew Morton wrote:
> On Tue, 22 Dec 2015 15:11:38 -0800 Andrew Morton <akpm@linux-foundation.org> wrote:
> 
> > What I have is
> 
> And after a bit of reject resolution in
> mm-memcontrol-clean-up-alloc-online-offline-free-functions.patch we
> have
> 
> 
> static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
> {
> 	struct mem_cgroup *memcg = mem_cgroup_from_css(css);
> 
> 	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
> 		static_branch_dec(&memcg_sockets_enabled_key);
> 
> 	vmpressure_cleanup(&memcg->vmpressure);
> 	cancel_work_sync(&memcg->high_work);
> 	mem_cgroup_remove_from_trees(memcg);
> 	memcg_free_kmem(memcg);
> 
> 	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && memcg->tcpmem_active)
> 		static_branch_dec(&memcg_sockets_enabled_key);
> 
> 	mem_cgroup_free(memcg);
> }
> 
> code looks a bit strange.  Can we move the static_branch_dec's together
> and run cgroup_subsys_on_dfl just once?

Thanks for fixing it up. I think we can at least put the branches next
to each other. Here is what I have in my local tree:

static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
{
	struct mem_cgroup *memcg = mem_cgroup_from_css(css);

	if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
		static_branch_dec(&memcg_sockets_enabled_key);

	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && memcg->tcpmem_active)
		static_branch_dec(&memcg_sockets_enabled_key);

	vmpressure_cleanup(&memcg->vmpressure);
	cancel_work_sync(&memcg->high_work);
	mem_cgroup_remove_from_trees(memcg);
	memcg_free_kmem(memcg);
	mem_cgroup_free(memcg);
}

However, I don't think turning it into this would be an improvement:

	if (cgroup_subsys_on_dfl(memory_cgrp_subsys)) {
		if (!cgroup_memory_nosocket)
			static_branch_dec(&memcg_sockets_enabled_key);
	} else if (memcg->tcpmem_active) {
		static_branch_dec(&memcg_sockets_enabled_key);
	}

Plus, I'm a little worried that conflating cgroup and cgroup2 blocks
will get us into trouble. Yeah, that code looks a little unusual, but
I can't help but think it's easier to follow the code flow for one
particular mode when the jump labels are always explicit. Then the
brain can easily pattern-match and ignore blocks of the other mode.
It doesn't work the same when we hide keywords in implicit else ifs.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web