From: Yosry Ahmed <yosryahmed@google.com>
To: Chris Li <chrisl@kernel.org>
Cc: lsf-pc@lists.linux-foundation.org,
Johannes Weiner <hannes@cmpxchg.org>,
Linux-MM <linux-mm@kvack.org>, Michal Hocko <mhocko@kernel.org>,
Shakeel Butt <shakeelb@google.com>,
David Rientjes <rientjes@google.com>,
Hugh Dickins <hughd@google.com>,
Seth Jennings <sjenning@redhat.com>,
Dan Streetman <ddstreet@ieee.org>,
Vitaly Wool <vitaly.wool@konsulko.com>,
Yang Shi <shy828301@gmail.com>, Peter Xu <peterx@redhat.com>,
Minchan Kim <minchan@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [LSF/MM/BPF TOPIC] Swap Abstraction / Native Zswap
Date: Wed, 1 Mar 2023 17:00:23 -0800 [thread overview]
Message-ID: <CAJD7tka2sr=9KbHRKLso2SxHBsjOX1otoadm_VMQXEGuqBMBgw@mail.gmail.com> (raw)
In-Reply-To: <CAJD7tkbvGvhTKCOqRpcht797Uw41fWgNd3r2kpN3ObfnUuaUxw@mail.gmail.com>
On Wed, Mar 1, 2023 at 4:30 PM Yosry Ahmed <yosryahmed@google.com> wrote:
>
> On Tue, Feb 28, 2023 at 3:11 PM Chris Li <chrisl@kernel.org> wrote:
> >
> > Hi Yosry,
> >
> > On Sat, Feb 18, 2023 at 02:38:40PM -0800, Yosry Ahmed wrote:
> > > Hello everyone,
> > >
> > > I would like to propose a topic for the upcoming LSF/MM/BPF in May
> > > 2023 about swap & zswap (hope I am not too late).
> >
> > I am very interested in participating in this discussion as well.
>
> That's great to hear!
>
> >
> > > ==================== Objective ====================
> > > Enabling the use of zswap without a backing swapfile, which makes
> > > zswap useful for a wider variety of use cases. Also, when zswap is
> > > used with a swapfile, the pages in zswap do not use up space in the
> > > swapfile, so the overall swapping capacity increases.
> >
> > Agree.
> >
> > >
> > > ==================== Idea ====================
> > > Introduce a data structure, which I currently call a swap_desc, as an
> > > abstraction layer between swapping implementation and the rest of MM
> > > code. Page tables & page caches would store a swap id (encoded as a
> > > swp_entry_t) instead of directly storing the swap entry associated
> > > with the swapfile. This swap id maps to a struct swap_desc, which acts
> >
> > Can you provide a bit more detail? I am curious how this swap id
> > maps into the swap_desc? Is the swp_entry_t cast into "struct
> > swap_desc*" or going through some lookup table/tree?
>
> swap id would be an index in a radix tree (aka xarray), which contains
> a pointer to the swap_desc struct. This lookup should be free with
> this design as we also use swap_desc to directly store the swap cache
> pointer, so this lookup essentially replaces the swap cache lookup.
>
> >
> > > as our abstraction layer. All MM code not concerned with swapping
> > > details would operate in terms of swap descs. The swap_desc can point
> > > to either a normal swap entry (associated with a swapfile) or a zswap
> > > entry. It can also include all non-backend specific operations, such
> > > as the swapcache (which would be a simple pointer in swap_desc), swap
> >
> > Does the zswap entry still use the swap slot cache and swap_info_struct?
>
> In this design no, it shouldn't.
>
> >
> > > This work enables using zswap without a backing swapfile and increases
> > > the swap capacity when zswap is used with a swapfile. It also creates
> > > a separation that allows us to skip code paths that don't make sense
> > > in the zswap path (e.g. readahead). We get to drop zswap's rbtree
> > > which might result in better performance (less lookups, less lock
> > > contention).
> > >
> > > The abstraction layer also opens the door for multiple cleanups (e.g.
> > > removing swapper address spaces, removing swap count continuation
> > > code, etc). Another nice cleanup that this work enables would be
> > > separating the overloaded swp_entry_t into two distinct types: one for
> > > things that are stored in page tables / caches, and for actual swap
> > > entries. In the future, we can potentially further optimize how we use
> > > the bits in the page tables instead of sticking everything into the
> > > current type/offset format.
> >
> > Looking forward to seeing more details in the upcoming discussion.
> > >
> > > ==================== Cost ====================
> > > The obvious downside of this is added memory overhead, specifically
> > > for users that use swapfiles without zswap. Instead of paying one byte
> > > (swap_map) for every potential page in the swapfile (+ swap count
> > > continuation), we pay the size of the swap_desc for every page that is
> > > actually in the swapfile, which I am estimating can be roughly around
> > > 24 bytes or so, so maybe 0.6% of swapped out memory. The overhead only
> > > scales with pages actually swapped out. For zswap users, it should be
> >
> > Is there a way to avoid turning 1 byte into 24 byte per swapped
> > pages? For the users that use swap but no zswap, this is pure overhead.
>
> That's what I could think of at this point. My idea was something like this:
>
> struct swap_desc {
> union { /* Use one bit to distinguish them */
> swp_entry_t swap_entry;
> struct zswap_entry *zswap_entry;
> };
> struct folio *swapcache;
> atomic_t swap_count;
> u32 id;
> }
>
> Having the id in the swap_desc is convenient as we can directly map
> the swap_desc to a swp_entry_t to place in the page tables, but I
> don't think it's necessary. Without it, the struct size is 20 bytes,
> so I think the extra 4 bytes are okay to use anyway if the slab
> allocator only allocates multiples of 8 bytes.
>
> The idea here is to unify the swapcache and swap_count implementation
> between different swap backends (swapfiles, zswap, etc), which would
> create a better abstraction and reduce reinventing the wheel.
>
> We can reduce to only 8 bytes and only store the swap/zswap entry, but
> we still need the swap cache anyway so might as well just store the
> pointer in the struct and have a unified lookup-free swapcache, so
> really 16 bytes is the minimum.
>
> If we stop at 16 bytes, then we need to handle swap count separately
> in swapfiles and zswap. This is not the end of the world, but are the
> 8 bytes worth this?
>
> Keep in mind that the current overhead is 1 byte O(max swap pages) not
> O(swapped). Also, 1 byte is assuming we do not use the swap
> continuation pages. If we do, it may end up being more. We also
> allocate continuation in full 4k pages, so even if one swap_map
> element in a page requires continuation, we will allocate an entire
> page. What I am trying to say is that to get an actual comparison you
> need to also factor in the swap utilization and the rate of usage of
> swap continuation. I don't know how to come up with a formula for this
> tbh.
>
> Also, like Johannes said, the worst case overhead (32 bytes if you
> count the reverse mapping) is 0.8% of swapped memory, aka 8M for every
> 1G swapped. It doesn't sound *very* bad. I understand that it is pure
> overhead for people not using zswap, but it is not very awful.
Oh I forgot. I think the 24 bytes *might* actually be reduced to 16
bytes if we free the underlying swap entry / zswap entry once we add
the page to the swapcache. I did not post anything about it yet as I
am still thinking if there might be any synchronization problems with
this approach, but I will try it out.
>
> >
> > It seems what you really need is one bit of information to indicate
> > this page is backed by zswap. Then you can have a seperate pointer
> > for the zswap entry.
>
> If you use one bit in swp_entry_t (or one of the available swap types)
> to indicate whether the page is backed with a swapfile or zswap it
> doesn't really work. We lose the indirection layer. How do we move the
> page from zswap to swapfile? We need to go update the page tables and
> the shmem page cache, similar to swapoff.
>
> Instead, if we store a key else in swp_entry_t and use this to lookup
> the swp_entry_t or zswap_entry pointer then that's essentially what
> the swap_desc does. It just goes the extra mile of unifying the
> swapcache as well and storing it directly in the swap_desc instead of
> storing it in another lookup structure.
>
> >
> > Depending on how much you are going to reuse the swap cache, you might
> > need to have something like a swap_info_struct to keep the locks happy.
>
> My current intention is to reimplement the swapcache completely as a
> pointer in struct swap_desc. This would eliminate this need and a lot
> of the locking we do today if I get things right.
>
> >
> > > Another potential concern is readahead. With this design, we have no
> >
> > Readahead is for spinning disk :-) Even a normal swap file with an SSD can
> > use some modernization.
>
> Yeah, I initially thought we would only need the swp_entry_t ->
> swap_desc reverse mapping for readahead, and that we can only store
> that for spinning disks, but I was wrong. We need for other things as
> well today: swapoff, when trying to find an empty swap slot and we
> start trying to free swap slots used only by the swapcache. However, I
> think both of these cases can be fixed (I can share more details if
> you want). If everything goes well we should only need to maintain the
> reverse mapping (extra overhead above 24 bytes) for swap files on
> spinning disks for readahead.
>
> >
> > Looking forward to your discussion.
> >
> > Chris
> >
next prev parent reply other threads:[~2023-03-02 1:01 UTC|newest]
Thread overview: 105+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-18 22:38 [LSF/MM/BPF TOPIC] Swap Abstraction / Native Zswap Yosry Ahmed
2023-02-19 4:31 ` Matthew Wilcox
2023-02-19 9:34 ` Yosry Ahmed
2023-02-28 23:22 ` Chris Li
2023-03-01 0:08 ` Matthew Wilcox
2023-03-01 23:22 ` Chris Li
2023-02-21 18:39 ` Yang Shi
2023-02-21 18:56 ` Yosry Ahmed
2023-02-21 19:26 ` Yang Shi
2023-02-21 19:46 ` Yosry Ahmed
2023-02-21 23:34 ` Yang Shi
2023-02-21 23:38 ` Yosry Ahmed
2023-02-22 16:57 ` Johannes Weiner
2023-02-22 22:46 ` Yosry Ahmed
2023-02-28 4:29 ` Kalesh Singh
2023-02-28 8:09 ` Yosry Ahmed
2023-02-28 4:54 ` Sergey Senozhatsky
2023-02-28 8:12 ` Yosry Ahmed
2023-02-28 23:29 ` Minchan Kim
2023-03-02 0:58 ` Yosry Ahmed
2023-03-02 1:25 ` Yosry Ahmed
2023-03-02 17:05 ` Chris Li
2023-03-02 17:47 ` Chris Li
2023-03-02 18:15 ` Johannes Weiner
2023-03-02 18:56 ` Chris Li
2023-03-02 18:23 ` Rik van Riel
2023-03-02 21:42 ` Chris Li
2023-03-02 22:36 ` Rik van Riel
2023-03-02 22:55 ` Yosry Ahmed
2023-03-03 4:05 ` Chris Li
2023-03-03 0:01 ` Chris Li
2023-03-02 16:58 ` Chris Li
2023-03-01 10:44 ` Sergey Senozhatsky
2023-03-02 1:01 ` Yosry Ahmed
2023-02-28 23:11 ` Chris Li
2023-03-02 0:30 ` Yosry Ahmed
2023-03-02 1:00 ` Yosry Ahmed [this message]
2023-03-02 16:51 ` Chris Li
2023-03-03 0:33 ` Minchan Kim
2023-03-03 0:49 ` Yosry Ahmed
2023-03-03 1:25 ` Minchan Kim
2023-03-03 17:15 ` Yosry Ahmed
2023-03-09 12:48 ` Huang, Ying
2023-03-09 19:58 ` Chris Li
2023-03-09 20:19 ` Yosry Ahmed
2023-03-10 3:06 ` Huang, Ying
2023-03-10 23:14 ` Chris Li
2023-03-13 1:10 ` Huang, Ying
2023-03-15 7:41 ` Yosry Ahmed
2023-03-16 1:42 ` Huang, Ying
2023-03-11 1:06 ` Yosry Ahmed
2023-03-13 2:12 ` Huang, Ying
2023-03-15 8:01 ` Yosry Ahmed
2023-03-16 7:50 ` Huang, Ying
2023-03-17 10:19 ` Yosry Ahmed
2023-03-17 18:19 ` Chris Li
2023-03-17 18:23 ` Yosry Ahmed
2023-03-20 2:55 ` Huang, Ying
2023-03-20 6:25 ` Chris Li
2023-03-23 0:56 ` Huang, Ying
2023-03-23 6:46 ` Chris Li
2023-03-23 6:56 ` Huang, Ying
2023-03-23 18:28 ` Chris Li
2023-03-23 18:40 ` Yosry Ahmed
2023-03-23 19:49 ` Chris Li
2023-03-23 19:54 ` Yosry Ahmed
2023-03-23 21:10 ` Chris Li
2023-03-24 17:28 ` Chris Li
2023-03-22 5:56 ` Yosry Ahmed
2023-03-23 1:48 ` Huang, Ying
2023-03-23 2:21 ` Yosry Ahmed
2023-03-23 3:16 ` Huang, Ying
2023-03-23 3:27 ` Yosry Ahmed
2023-03-23 5:37 ` Huang, Ying
2023-03-23 15:18 ` Yosry Ahmed
2023-03-24 2:37 ` Huang, Ying
2023-03-24 7:28 ` Yosry Ahmed
2023-03-24 17:23 ` Chris Li
2023-03-27 1:23 ` Huang, Ying
2023-03-28 5:54 ` Yosry Ahmed
2023-03-28 6:20 ` Huang, Ying
2023-03-28 6:29 ` Yosry Ahmed
2023-03-28 6:59 ` Huang, Ying
2023-03-28 7:59 ` Yosry Ahmed
2023-03-28 14:14 ` Johannes Weiner
2023-03-28 19:59 ` Yosry Ahmed
2023-03-28 21:22 ` Chris Li
2023-03-28 21:30 ` Yosry Ahmed
2023-03-28 20:50 ` Chris Li
2023-03-28 21:01 ` Yosry Ahmed
2023-03-28 21:32 ` Chris Li
2023-03-28 21:44 ` Yosry Ahmed
2023-03-28 22:01 ` Chris Li
2023-03-28 22:02 ` Yosry Ahmed
2023-03-29 1:31 ` Huang, Ying
2023-03-29 1:41 ` Yosry Ahmed
2023-03-29 16:04 ` Chris Li
2023-04-04 8:24 ` Huang, Ying
2023-04-04 8:10 ` Huang, Ying
2023-04-04 8:47 ` Yosry Ahmed
2023-04-06 1:40 ` Huang, Ying
2023-03-29 15:22 ` Chris Li
2023-03-10 2:07 ` Luis Chamberlain
2023-03-10 2:15 ` Yosry Ahmed
2023-05-12 3:07 ` Yosry Ahmed
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='CAJD7tka2sr=9KbHRKLso2SxHBsjOX1otoadm_VMQXEGuqBMBgw@mail.gmail.com' \
--to=yosryahmed@google.com \
--cc=akpm@linux-foundation.org \
--cc=chrisl@kernel.org \
--cc=ddstreet@ieee.org \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=linux-mm@kvack.org \
--cc=lsf-pc@lists.linux-foundation.org \
--cc=mhocko@kernel.org \
--cc=minchan@kernel.org \
--cc=peterx@redhat.com \
--cc=rientjes@google.com \
--cc=shakeelb@google.com \
--cc=shy828301@gmail.com \
--cc=sjenning@redhat.com \
--cc=vitaly.wool@konsulko.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;
as well as URLs for NNTP newsgroup(s).