[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160617205526.5cjfm56prpstvowc@treble>
Date: Fri, 17 Jun 2016 15:55:26 -0500
From: Josh Poimboeuf <jpoimboe@...hat.com>
To: Andy Lutomirski <luto@...nel.org>
Cc: x86@...nel.org, linux-kernel@...r.kernel.org, linux-arch@...r.kernel.org,
Borislav Petkov <bp@...en8.de>, Nadav Amit <nadav.amit@...il.com>,
Kees Cook <keescook@...omium.org>, Brian Gerst <brgerst@...il.com>,
"kernel-hardening@...ts.openwall.com" <kernel-hardening@...ts.openwall.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Jann Horn <jann@...jh.net>, Heiko Carstens <heiko.carstens@...ibm.com>,
Vladimir Davydov <vdavydov@...tuozzo.com>,
Johannes Weiner <hannes@...xchg.org>, Michal Hocko <mhocko@...nel.org>,
linux-mm@...ck.org
Subject: Re: [PATCH v2 05/13] mm: Move memcg stack accounting to
account_kernel_stack
On Fri, Jun 17, 2016 at 01:00:41PM -0700, Andy Lutomirski wrote:
> We should account for stacks regardless of stack size. Move it into
> account_kernel_stack.
>
> Fixes: 12580e4b54ba8 ("mm: memcontrol: report kernel stack usage in cgroup2 memory.stat")
> Cc: Vladimir Davydov <vdavydov@...tuozzo.com>
> Cc: Johannes Weiner <hannes@...xchg.org>
> Cc: Michal Hocko <mhocko@...nel.org>
> Cc: linux-mm@...ck.org
> Signed-off-by: Andy Lutomirski <luto@...nel.org>
> ---
> kernel/fork.c | 15 ++++++---------
> 1 file changed, 6 insertions(+), 9 deletions(-)
>
> diff --git a/kernel/fork.c b/kernel/fork.c
> index be7f006af727..cd2abe6e4e41 100644
> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -165,20 +165,12 @@ static struct thread_info *alloc_thread_info_node(struct task_struct *tsk,
> struct page *page = alloc_kmem_pages_node(node, THREADINFO_GFP,
> THREAD_SIZE_ORDER);
>
> - if (page)
> - memcg_kmem_update_page_stat(page, MEMCG_KERNEL_STACK,
> - 1 << THREAD_SIZE_ORDER);
> -
> return page ? page_address(page) : NULL;
> }
>
> static inline void free_thread_info(struct thread_info *ti)
> {
> - struct page *page = virt_to_page(ti);
> -
> - memcg_kmem_update_page_stat(page, MEMCG_KERNEL_STACK,
> - -(1 << THREAD_SIZE_ORDER));
> - __free_kmem_pages(page, THREAD_SIZE_ORDER);
> + free_kmem_pages((unsigned long)ti, THREAD_SIZE_ORDER);
> }
> # else
> static struct kmem_cache *thread_info_cache;
> @@ -227,6 +219,11 @@ static void account_kernel_stack(struct thread_info *ti, int account)
>
> mod_zone_page_state(zone, NR_KERNEL_STACK_KB,
> THREAD_SIZE / 1024 * account);
> +
> + /* All stack pages belong to the same memcg. */
> + memcg_kmem_update_page_stat(
> + virt_to_page(ti), MEMCG_KERNEL_STACK,
> + account * (THREAD_SIZE / PAGE_SIZE));
> }
Won't this be broken in the case where THREAD_SIZE < PAGE_SIZE?
--
Josh
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.
