From: Dan Streetman <ddstreet@ieee.org>
To: Weijie Yang <weijie.yang.kh@gmail.com>
Cc: Seth Jennings <sjennings@variantweb.net>,
Minchan Kim <minchan@kernel.org>, Nitin Gupta <ngupta@vflare.org>,
Andrew Morton <akpm@linux-foundation.org>,
Bob Liu <bob.liu@oracle.com>, Hugh Dickins <hughd@google.com>,
Mel Gorman <mgorman@suse.de>, Rik van Riel <riel@redhat.com>,
Weijie Yang <weijie.yang@samsung.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
Linux-MM <linux-mm@kvack.org>,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 2/4] mm: zpool: implement zsmalloc shrinking
Date: Sun, 27 Apr 2014 00:13:30 -0400 [thread overview]
Message-ID: <CALZtONB4j=yd=cGBnkHy0+H0nyUCwG3PGb4K6XYCyRHA=mqt-g@mail.gmail.com> (raw)
In-Reply-To: <CAL1ERfMPcfyUeACnmZ2QF5WxJUQ2PaKbtRzis8sPbQsjnvf_GQ@mail.gmail.com>
On Sat, Apr 26, 2014 at 4:37 AM, Weijie Yang <weijie.yang.kh@gmail.com> wrote:
> On Sat, Apr 19, 2014 at 11:52 PM, Dan Streetman <ddstreet@ieee.org> wrote:
>> Add zs_shrink() and helper functions to zsmalloc. Update zsmalloc
>> zs_create_pool() creation function to include ops param that provides
>> an evict() function for use during shrinking. Update helper function
>> fix_fullness_group() to always reinsert changed zspages even if the
>> fullness group did not change, so they are updated in the fullness
>> group lru. Also update zram to use the new zsmalloc pool creation
>> function but pass NULL as the ops param, since zram does not use
>> pool shrinking.
>>
>
> I only review the code without test, however, I think this patch is
> not acceptable.
>
> The biggest problem is it will call zswap_writeback_entry() under lock,
> zswap_writeback_entry() may sleep, so it is a bug. see below
thanks for catching that!
>
> The 3/4 patch has a lot of #ifdef, I don't think it's a good kind of
> abstract way.
it has the #ifdef's because there's no point in compiling in code to
use zbud/zsmalloc if zbud/zsmalloc isn't compiled...what alternative
to #ifdef's would you suggest? Or are there just specific #ifdefs you
suggest to remove?
>
> What about just disable zswap reclaim when using zsmalloc?
> There is a long way to optimize writeback reclaim(both zswap and zram) ,
> Maybe a small and simple step forward is better.
I think it's possible to just remove the zspage from the class while
under lock, then unlock and reclaim it. As long as there's a
guarantee that zswap (or any zpool/zsmalloc reclaim user) doesn't
map/access the handle after evict() completes successfully, that
should work. There does need to be some synchronization between
zs_free() and each handle's eviction though, similar to zbud's
under_reclaim flag. I'll work on a v2 patch.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2014-04-27 4:13 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-19 15:52 [PATCH 0/4] mm: zpool: add common api for zswap to use zbud/zsmalloc Dan Streetman
2014-04-19 15:52 ` [PATCH 1/4] mm: zpool: zbud_alloc() minor param change Dan Streetman
2014-04-19 15:52 ` [PATCH 2/4] mm: zpool: implement zsmalloc shrinking Dan Streetman
2014-04-26 8:37 ` Weijie Yang
2014-04-27 4:13 ` Dan Streetman [this message]
2014-05-02 20:01 ` Seth Jennings
2014-05-04 20:38 ` Dan Streetman
2014-04-19 15:52 ` [PATCH 3/4] mm: zpool: implement common zpool api to zbud/zsmalloc Dan Streetman
2014-04-22 10:05 ` Sergey Senozhatsky
2014-04-22 13:43 ` Dan Streetman
2014-04-19 15:52 ` [PATCH 4/4] mm: zpool: update zswap to use zpool Dan Streetman
2014-04-21 2:47 ` [PATCH 0/4] mm: zpool: add common api for zswap to use zbud/zsmalloc Weijie Yang
2014-05-07 21:51 ` [PATCHv2 0/4] mm/zpool: " Dan Streetman
2014-05-07 21:51 ` [PATCHv2 1/4] mm/zbud: zbud_alloc() minor param change Dan Streetman
2014-05-09 3:33 ` Seth Jennings
2014-05-07 21:51 ` [PATCH 2/4] mm/zbud: change zbud_alloc size type to size_t Dan Streetman
2014-05-09 3:33 ` Seth Jennings
2014-05-07 21:51 ` [PATCHv2 3/4] mm/zpool: implement common zpool api to zbud/zsmalloc Dan Streetman
2014-05-09 4:13 ` Seth Jennings
2014-05-10 16:06 ` Dan Streetman
2014-05-07 21:51 ` [PATCHv2 4/4] mm/zswap: update zswap to use zpool Dan Streetman
2014-05-24 19:06 ` [PATCHv3 0/6] mm/zpool: add common api for zswap to use zbud/zsmalloc Dan Streetman
2014-05-24 19:06 ` [PATCHv2 1/6] mm/zbud: zbud_alloc() minor param change Dan Streetman
2014-05-24 19:06 ` [PATCH 2/6] mm/zbud: change zbud_alloc size type to size_t Dan Streetman
2014-05-24 19:06 ` [PATCHv3 3/6] mm/zpool: implement common zpool api to zbud/zsmalloc Dan Streetman
2014-05-27 22:06 ` Seth Jennings
2014-05-27 22:48 ` Seth Jennings
2014-05-28 0:06 ` Dan Streetman
2014-05-29 3:48 ` Seth Jennings
2014-05-24 19:06 ` [PATCH 4/6] mm/zpool: zbud/zsmalloc implement zpool Dan Streetman
2014-05-24 19:06 ` [PATCHv3 5/6] mm/zpool: update zswap to use zpool Dan Streetman
2014-05-24 19:06 ` [PATCH 6/6] mm/zpool: prevent zbud/zsmalloc from unloading when used Dan Streetman
2014-05-27 22:40 ` Seth Jennings
2014-05-28 0:40 ` Dan Streetman
2014-05-27 22:44 ` [PATCHv3 0/6] mm/zpool: add common api for zswap to use zbud/zsmalloc Seth Jennings
2014-06-02 22:19 ` [PATCHv4 " Dan Streetman
2014-06-02 22:19 ` [PATCHv2 1/6] mm/zbud: zbud_alloc() minor param change Dan Streetman
2014-06-23 21:19 ` Andrew Morton
2014-06-24 15:24 ` Dan Streetman
2014-06-02 22:19 ` [PATCH 2/6] mm/zbud: change zbud_alloc size type to size_t Dan Streetman
2014-06-02 22:19 ` [PATCHv4 3/6] mm/zpool: implement common zpool api to zbud/zsmalloc Dan Streetman
2014-06-23 21:46 ` Andrew Morton
2014-06-24 15:39 ` Dan Streetman
2014-06-24 23:08 ` Andrew Morton
2014-06-27 17:11 ` Dan Streetman
2014-06-27 19:17 ` Andrew Morton
2014-06-02 22:19 ` [PATCHv2 4/6] mm/zpool: zbud/zsmalloc implement zpool Dan Streetman
2014-06-02 22:19 ` [PATCHv4 5/6] mm/zpool: update zswap to use zpool Dan Streetman
2014-06-02 22:19 ` [PATCHv2 6/6] mm/zpool: prevent zbud/zsmalloc from unloading when used Dan Streetman
2014-06-23 21:48 ` Andrew Morton
2014-06-24 15:41 ` Dan Streetman
2014-06-04 1:38 ` [PATCHv4 0/6] mm/zpool: add common api for zswap to use zbud/zsmalloc Bob Liu
2014-06-06 21:01 ` Seth Jennings
2014-07-02 21:43 ` [PATCHv5 0/4] " Dan Streetman
2014-07-02 21:45 ` Dan Streetman
2014-07-02 21:45 ` [PATCHv2 1/4] mm/zbud: change zbud_alloc size type to size_t Dan Streetman
2014-07-02 21:45 ` [PATCHv5 2/4] mm/zpool: implement common zpool api to zbud/zsmalloc Dan Streetman
2014-07-02 21:45 ` [PATCHv3 3/4] mm/zpool: zbud/zsmalloc implement zpool Dan Streetman
2014-07-02 21:45 ` [PATCHv5 4/4] mm/zpool: update zswap to use zpool Dan Streetman
2014-07-14 18:10 ` [PATCHv5 0/4] mm/zpool: add common api for zswap to use zbud/zsmalloc Dan Streetman
2014-07-16 20:59 ` Seth Jennings
2014-07-16 21:05 ` Dan Streetman
2014-07-16 22:00 ` Seth Jennings
2014-07-25 16:59 ` Dan Streetman
2014-07-28 20:40 ` Seth Jennings
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CALZtONB4j=yd=cGBnkHy0+H0nyUCwG3PGb4K6XYCyRHA=mqt-g@mail.gmail.com' \
--to=ddstreet@ieee.org \
--cc=akpm@linux-foundation.org \
--cc=bob.liu@oracle.com \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=minchan@kernel.org \
--cc=ngupta@vflare.org \
--cc=riel@redhat.com \
--cc=sergey.senozhatsky@gmail.com \
--cc=sjennings@variantweb.net \
--cc=weijie.yang.kh@gmail.com \
--cc=weijie.yang@samsung.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox