[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180117193114.GB25862@bombadil.infradead.org>
Date: Wed, 17 Jan 2018 11:31:14 -0800
From: Matthew Wilcox <willy@...radead.org>
To: Christopher Lameter <cl@...ux.com>
Cc: Kees Cook <keescook@...omium.org>, linux-mm@...ck.org,
Rik van Riel <riel@...hat.com>, Matthew Garrett <mjg59@...gle.com>,
kernel-hardening@...ts.openwall.com
Subject: Re: kmem_cache_attr (was Re: [PATCH 04/36] usercopy: Prepare for
usercopy whitelisting)
On Wed, Jan 17, 2018 at 11:42:34AM -0600, Christopher Lameter wrote:
> On Tue, 16 Jan 2018, Matthew Wilcox wrote:
>
> > struct kmem_cache_attr {
> > char name[16];
>
> That doesnt work. memcg needs long slab names. Sigh.
Oof. That's ugly. Particularly ugly in that it could actually share the
attr with the root cache, except for the name.
We could put a char * in the kmem_cache which (if not NULL) overrides
the attr->name? Probably want a helper to replace the endearingly short
's->name'. Something like:
#define slab_name(s) s->name ?: s->attr->name
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
