linux-kernel - Re: [PATCH 3/3] tools/vm/slabinfo: update struct slabinfo members' types

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:	Thu, 12 Nov 2015 15:17:01 +0900
From:	Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
To:	David Rientjes <rientjes@...gle.com>
Cc:	Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>,
	Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Christoph Lameter <cl@...ux.com>,
	Pekka Enberg <penberg@...nel.org>,
	Joonsoo Kim <iamjoonsoo.kim@....com>, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/3] tools/vm/slabinfo: update struct slabinfo members'
 types

On (11/11/15 21:07), David Rientjes wrote:
[..]
> > > >  	/* Object size */
> > > > -	unsigned long long min_objsize = max, max_objsize = 0, avg_objsize;
> > > > +	unsigned int min_objsize = UINT_MAX, max_objsize = 0, avg_objsize;
> > > >  
> > > >  	/* Number of partial slabs in a slabcache */
> > > >  	unsigned long long min_partial = max, max_partial = 0,
> > > 
> > > avg_objsize should not be unsigned int.
> > 
> > Hm. the assumption is that `avg_objsize' cannot be larger
> > than `max_objsize', which is
> > 	`int object_size;' in `struct kmem_cache' from slab_def.h
> > and
> > 	`unsigned int object_size;' in `struct kmem_cache' from slab.h.
> > 
> > 
> >  avg_objsize = total_used / total_objects;
> > 
> 

I'm not sure I clearly understand the problems you're pointing
me to.

> This has nothing to do with object_size in the kernel.

what we have in slabinfo as slab_size(), ->object_size, etc.
comming from slub's sysfs attrs:

	chdir("/sys/kernel/slab")
	while readdir
		...
		slab->object_size = get_obj("object_size");
		slab->slab_size = get_obj("slab_size");
		...

and attr show handlers are:

...
 static ssize_t slab_size_show(struct kmem_cache *s, char *buf)
 {
 	return sprintf(buf, "%d\n", s->size);
 }
 SLAB_ATTR_RO(slab_size);

 static ssize_t object_size_show(struct kmem_cache *s, char *buf)
 {
 	return sprintf(buf, "%d\n", s->object_size);
 }
 SLAB_ATTR_RO(object_size);
...

so those are sprintf("%d") of `struct kmem_cache'-s `int'
values.


> total_used and total_objects are unsigned long long.

yes, that's correct.
but `total_used / total_objects' cannot be larger that the size
of the largest object, which is represented in the kernel and
returned to user space as `int'. it must fit into `unsigned int'.


> If you need to convert max_objsize to be unsigned long long as
> well, that would be better.

... in case if someday `struct kmem_cache' will be updated to keep
`unsigned long' sized objects and sysfs attrs will do sprintf("%lu")?
IOW, if slabs will keep objects bigger that 4gig?

	-ss
--
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/