[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5a556476-05b2-483e-8875-eeb3e2a51151@linux.dev>
Date: Wed, 28 Feb 2024 13:14:48 +0800
From: Chengming Zhou <chengming.zhou@...ux.dev>
To: Sergey Senozhatsky <senozhatsky@...omium.org>,
Chengming Zhou <zhouchengming@...edance.com>
Cc: yosryahmed@...gle.com, hannes@...xchg.org, nphamcs@...il.com,
Andrew Morton <akpm@...ux-foundation.org>, Minchan Kim <minchan@...nel.org>,
linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH 1/2] mm/zsmalloc: don't hold locks of all pages when
free_zspage()
On 2024/2/28 12:33, Sergey Senozhatsky wrote:
> On (24/02/27 03:02), Chengming Zhou wrote:
> [..]
>> @@ -978,10 +974,11 @@ static struct zspage *alloc_zspage(struct zs_pool *pool,
>> pages[i] = page;
>> }
>>
>> - create_page_chain(class, zspage, pages);
>> init_zspage(class, zspage);
>> zspage->pool = pool;
>> zspage->class = class->index;
>> + /* RCU set_zspage() after zspage initialized. */
>> + create_page_chain(class, zspage, pages);
>
> So this hasn't been tested, has it?
I have tested it in my test vm, but it hasn't KASAN enabled. I tested the
kernel build in tmpfs with zswap enabled using zsmalloc pool, not sure
why the kernel didn't crash then...
>
> init_zspage() does not like to be invoked before create_page_chain(),
> because we haven't setup required pointers yet.
You're right, I can reproduce the problem with KASAN enabled this time,
create_page_chain() should be put before init_zspage(), which will iterate
over the pages to create free objects list.
>
> So when init_zspage() calls get_first_page() it gets NULL zspage->first_page
> which we then use in is_first_page(first_page)->PagePrivate(page). As far as
> I can tell.
Thanks! I will fix it and test throughly before send an update.
Powered by blists - more mailing lists