[PATCH] slab: fix /proc/slabinfo alignment

[PATCH] slab: fix /proc/slabinfo alignment

[? ?]

Signed-off-by: mordor <mordorw@hotmail.com>
/proc/slabinfo is not aligned, it is difficult to read, so correct it

---
 mm/slab_common.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/mm/slab_common.c b/mm/slab_common.c
index 10f127b..7111549 100644
--- a/mm/slab_common.c
+++ b/mm/slab_common.c
@@ -1232,7 +1232,6 @@ void cache_random_seq_destroy(struct kmem_cache *cachep)
 #else
 #define SLABINFO_RIGHTS S_IRUSR
 #endif
-
 static void print_slabinfo_header(struct seq_file *m)
 {
 	/*
@@ -1244,7 +1243,7 @@ static void print_slabinfo_header(struct seq_file *m)
 #else
 	seq_puts(m, "slabinfo - version: 2.1\n");
 #endif
-	seq_puts(m, "# name            <active_objs> <num_objs> <objsize> <objperslab> <pagesperslab>");
+	seq_puts(m, "# name                         <active_objs> <num_objs> <objsize> <objperslab> <pagesperslab>");
 	seq_puts(m, " : tunables <limit> <batchcount> <sharedfactor>");
 	seq_puts(m, " : slabdata <active_slabs> <num_slabs> <sharedavail>");
 #ifdef CONFIG_DEBUG_SLAB
@@ -1291,6 +1290,7 @@ memcg_accumulate_slabinfo(struct kmem_cache *s, struct slabinfo *info)
 	}
 }
 
+
 static void cache_show(struct kmem_cache *s, struct seq_file *m)
 {
 	struct slabinfo sinfo;
@@ -1300,13 +1300,13 @@ static void cache_show(struct kmem_cache *s, struct seq_file *m)
 
 	memcg_accumulate_slabinfo(s, &sinfo);
 
-	seq_printf(m, "%-17s %6lu %6lu %6u %4u %4d",
+	seq_printf(m, "%-30s %13lu %10lu %9u %12u %14d",
 		   cache_name(s), sinfo.active_objs, sinfo.num_objs, s->size,
 		   sinfo.objects_per_slab, (1 << sinfo.cache_order));
 
-	seq_printf(m, " : tunables %4u %4u %4u",
+	seq_printf(m, " : tunables %7u %12u %14u",
 		   sinfo.limit, sinfo.batchcount, sinfo.shared);
-	seq_printf(m, " : slabdata %6lu %6lu %6lu",
+	seq_printf(m, " : slabdata %14lu %11lu %13lu",
 		   sinfo.active_slabs, sinfo.num_slabs, sinfo.shared_avail);
 	slabinfo_show_stats(m, s);
 	seq_putc(m, '\n');
-- 
2.7.4

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Michal Hocko]

On Tue 20-02-18 02:29:13, ? ? wrote:
> Signed-off-by: mordor <mordorw@hotmail.com>
> /proc/slabinfo is not aligned, it is difficult to read, so correct it

I do not see this as an improvement, to be honest. Moreover you risk a
regression when some dumb parsing tool relies on the current layout
format. I find the later rather unlikely but there would have to be a
very good reason to take the risk.

> ---
>  mm/slab_common.c | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/mm/slab_common.c b/mm/slab_common.c
> index 10f127b..7111549 100644
> --- a/mm/slab_common.c
> +++ b/mm/slab_common.c
> @@ -1232,7 +1232,6 @@ void cache_random_seq_destroy(struct kmem_cache *cachep)
>  #else
>  #define SLABINFO_RIGHTS S_IRUSR
>  #endif
> -
>  static void print_slabinfo_header(struct seq_file *m)
>  {
>  	/*
> @@ -1244,7 +1243,7 @@ static void print_slabinfo_header(struct seq_file *m)
>  #else
>  	seq_puts(m, "slabinfo - version: 2.1\n");
>  #endif
> -	seq_puts(m, "# name            <active_objs> <num_objs> <objsize> <objperslab> <pagesperslab>");
> +	seq_puts(m, "# name                         <active_objs> <num_objs> <objsize> <objperslab> <pagesperslab>");
>  	seq_puts(m, " : tunables <limit> <batchcount> <sharedfactor>");
>  	seq_puts(m, " : slabdata <active_slabs> <num_slabs> <sharedavail>");
>  #ifdef CONFIG_DEBUG_SLAB
> @@ -1291,6 +1290,7 @@ memcg_accumulate_slabinfo(struct kmem_cache *s, struct slabinfo *info)
>  	}
>  }
>  
> +
>  static void cache_show(struct kmem_cache *s, struct seq_file *m)
>  {
>  	struct slabinfo sinfo;
> @@ -1300,13 +1300,13 @@ static void cache_show(struct kmem_cache *s, struct seq_file *m)
>  
>  	memcg_accumulate_slabinfo(s, &sinfo);
>  
> -	seq_printf(m, "%-17s %6lu %6lu %6u %4u %4d",
> +	seq_printf(m, "%-30s %13lu %10lu %9u %12u %14d",
>  		   cache_name(s), sinfo.active_objs, sinfo.num_objs, s->size,
>  		   sinfo.objects_per_slab, (1 << sinfo.cache_order));
>  
> -	seq_printf(m, " : tunables %4u %4u %4u",
> +	seq_printf(m, " : tunables %7u %12u %14u",
>  		   sinfo.limit, sinfo.batchcount, sinfo.shared);
> -	seq_printf(m, " : slabdata %6lu %6lu %6lu",
> +	seq_printf(m, " : slabdata %14lu %11lu %13lu",
>  		   sinfo.active_slabs, sinfo.num_slabs, sinfo.shared_avail);
>  	slabinfo_show_stats(m, s);
>  	seq_putc(m, '\n');
> -- 
> 2.7.4
> 
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Christopher Lameter]

On Tue, 20 Feb 2018, ? ? wrote:

> /proc/slabinfo is not aligned, it is difficult to read, so correct it

How does it look on a terminal with 80 characters per line?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Matthew Wilcox]

On Tue, Feb 20, 2018 at 08:56:11AM -0600, Christopher Lameter wrote:
> On Tue, 20 Feb 2018, ? ? wrote:
> 
> > /proc/slabinfo is not aligned, it is difficult to read, so correct it
> 
> How does it look on a terminal with 80 characters per line?

That ship sailed long ago ...

kmalloc-8192         433    435   8192    1    2 : tunables    8    4    0 : sla
bdata    433    435      0

(I put in a manual carriage return at 80 columns for those not reading on
an 80 column terminal).

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Christopher Lameter]

On Tue, 20 Feb 2018, Matthew Wilcox wrote:

> On Tue, Feb 20, 2018 at 08:56:11AM -0600, Christopher Lameter wrote:
> > On Tue, 20 Feb 2018, ? ? wrote:
> >
> > > /proc/slabinfo is not aligned, it is difficult to read, so correct it
> >
> > How does it look on a terminal with 80 characters per line?
>
> That ship sailed long ago ...
>
> kmalloc-8192         433    435   8192    1    2 : tunables    8    4    0 : sla
> bdata    433    435      0
>
> (I put in a manual carriage return at 80 columns for those not reading on
> an 80 column terminal).

Well yes but if someone is fixing things then the 80 character issue also
should be fixed.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Matthew Wilcox]

On Tue, Feb 20, 2018 at 10:05:23AM -0600, Christopher Lameter wrote:
> On Tue, 20 Feb 2018, Matthew Wilcox wrote:
> 
> > On Tue, Feb 20, 2018 at 08:56:11AM -0600, Christopher Lameter wrote:
> > > On Tue, 20 Feb 2018, ? ? wrote:
> > >
> > > > /proc/slabinfo is not aligned, it is difficult to read, so correct it
> > >
> > > How does it look on a terminal with 80 characters per line?
> >
> > That ship sailed long ago ...
> >
> > kmalloc-8192         433    435   8192    1    2 : tunables    8    4    0 : sla
> > bdata    433    435      0
> >
> > (I put in a manual carriage return at 80 columns for those not reading on
> > an 80 column terminal).
> 
> Well yes but if someone is fixing things then the 80 character issue also
> should be fixed.

I don't think it's fixable; there's just too much information per slab.
Anyway, I preferred the solution you & I were working on to limit the
length of names to 16 bytes, except for the cgroup slabs.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Christopher Lameter]

On Tue, 20 Feb 2018, Matthew Wilcox wrote:

> I don't think it's fixable; there's just too much information per slab.
> Anyway, I preferred the solution you & I were working on to limit the
> length of names to 16 bytes, except for the cgroup slabs.

So what do we do with the cgroup slab names? have a slabinfo per cgroup?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Matthew Wilcox]

On Tue, Feb 20, 2018 at 10:23:41AM -0600, Christopher Lameter wrote:
> On Tue, 20 Feb 2018, Matthew Wilcox wrote:
> 
> > I don't think it's fixable; there's just too much information per slab.
> > Anyway, I preferred the solution you & I were working on to limit the
> > length of names to 16 bytes, except for the cgroup slabs.
> 
> So what do we do with the cgroup slab names? have a slabinfo per cgroup?

What I had in mind ...

struct kmem_cache_attr {
        const char name[16];
        unsigned int size;
        unsigned int align;
        unsigned int useroffset;
        unsigned int usersize;
        slab_flags_t flags;
        kmem_cache_ctor ctor;
}

struct kmem_cache {
        const struct kmem_cache_attr *a;
        const char *name;
	...
};

In kmem_cache_create_usercopy:

	s->name = a->name;

In memcg_create_kmem_cache:

	s->name = kasprintf(GFP_KERNEL, "%s(%llu:%s)", a->name,
				css->serial_nr, memcg_name_buf);

In slab_kmem_cache_release:

	if (s->name != s->a->name)
		kfree_const(s->name);

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Christopher Lameter]

On Tue, 20 Feb 2018, Matthew Wilcox wrote:

> In memcg_create_kmem_cache:
>
> 	s->name = kasprintf(GFP_KERNEL, "%s(%llu:%s)", a->name,
> 				css->serial_nr, memcg_name_buf);
>

But that creates the long name that shows up in /proc/slabinfo.


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] slab: fix /proc/slabinfo alignment

[Matthew Wilcox]

On Tue, Feb 20, 2018 at 12:43:52PM -0600, Christopher Lameter wrote:
> On Tue, 20 Feb 2018, Matthew Wilcox wrote:
> 
> > In memcg_create_kmem_cache:
> >
> > 	s->name = kasprintf(GFP_KERNEL, "%s(%llu:%s)", a->name,
> > 				css->serial_nr, memcg_name_buf);
> >
> 
> But that creates the long name that shows up in /proc/slabinfo.

Yes.  People who enable cgroups get an ugly /proc/slabinfo.  I don't have
a solution to that, but at least those of us who don't enable cgroups will
no longer have:

drm_i915_gem_request    243    357    576    7    1 : tunables   54   27    8 : slabdata     51     51      0
fat_cache              0      0     32  124    1 : tunables  120   60    8 : slabdata      0      0      0

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>