Messages in this thread | | | Date | Thu, 27 Sep 2012 20:38:19 +0400 | From | Glauber Costa <> | Subject | Re: [PATCH 4/4] sl[au]b: process slabinfo_show in common code |
| |
On 09/27/2012 07:07 PM, Christoph Lameter wrote: > On Thu, 27 Sep 2012, Glauber Costa wrote: > >> --- a/mm/slab_common.c >> +++ b/mm/slab_common.c >> @@ -239,7 +239,23 @@ static void s_stop(struct seq_file *m, void *p) >> >> static int s_show(struct seq_file *m, void *p) >> { >> - return slabinfo_show(m, p); >> + struct kmem_cache *s = list_entry(p, struct kmem_cache, list); >> + struct slabinfo sinfo; >> + >> + memset(&sinfo, 0, sizeof(sinfo)); >> + get_slabinfo(s, &sinfo); > > Could get_slabinfo() also set the objects per slab etc in some additional > field in struct slabinfo? Then we can avoid the exporting of the oo_ > functions and we do not need the cache_order() etc functions. > Yes. As a matter of fact, I first implemented it this way, and later switched. I was anticipating that people would be likely to point out that those properties are directly derivable from the caches, and it would be better to just get them from there.
I am more than happy to stick them in the slabinfo struct.
| |