Received-SPF: pass (google.com: domain of linux-kernel+bounces-155289-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240417160842.76665-1-ryncsn@gmail.com> <20240417160842.76665-7-ryncsn@gmail.com>
 <87mspkx3cy.fsf@yhuang6-desk2.ccr.corp.intel.com>
In-Reply-To: <87mspkx3cy.fsf@yhuang6-desk2.ccr.corp.intel.com>
From: Kairui Song <ryncsn@gmail.com>
Date: Tue, 23 Apr 2024 21:33:54 +0800
Message-ID: <CAMgjq7CH0xCEXF6nwsNZYW7Rcx0YF1+7Sb_ycXe2k10hbZc_tA@mail.gmail.com>
Subject: Re: [PATCH 6/8] mm/swap: get the swap file offset directly
To: "Huang, Ying" <ying.huang@intel.com>
Cc: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>, 
	Matthew Wilcox <willy@infradead.org>, Chris Li <chrisl@kernel.org>, 
	Barry Song <v-songbaohua@oppo.com>, Ryan Roberts <ryan.roberts@arm.com>, Neil Brown <neilb@suse.de>, 
	Minchan Kim <minchan@kernel.org>, Hugh Dickins <hughd@google.com>, 
	David Hildenbrand <david@redhat.com>, Yosry Ahmed <yosryahmed@google.com>, linux-fsdevel@vger.kernel.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Apr 23, 2024 at 9:43=E2=80=AFAM Huang, Ying <ying.huang@intel.com> =
wrote:
>
> Kairui Song <ryncsn@gmail.com> writes:
>
> > From: Kairui Song <kasong@tencent.com>
> >
> > folio_file_pos and page_file_offset are for mixed usage of swap cache
> > and page cache, it can't be page cache here, so introduce a new helper
> > to get the swap offset in swap file directly.
> >
> > Signed-off-by: Kairui Song <kasong@tencent.com>
> > ---
> >  mm/page_io.c | 6 +++---
> >  mm/swap.h    | 5 +++++
> >  2 files changed, 8 insertions(+), 3 deletions(-)
> >
> > diff --git a/mm/page_io.c b/mm/page_io.c
> > index ae2b49055e43..93de5aadb438 100644
> > --- a/mm/page_io.c
> > +++ b/mm/page_io.c
> > @@ -279,7 +279,7 @@ static void sio_write_complete(struct kiocb *iocb, =
long ret)
> >                * be temporary.
> >                */
> >               pr_err_ratelimited("Write error %ld on dio swapfile (%llu=
)\n",
> > -                                ret, page_file_offset(page));
> > +                                ret, swap_file_pos(page_swap_entry(pag=
e)));
> >               for (p =3D 0; p < sio->pages; p++) {
> >                       page =3D sio->bvec[p].bv_page;
> >                       set_page_dirty(page);
> > @@ -298,7 +298,7 @@ static void swap_writepage_fs(struct folio *folio, =
struct writeback_control *wbc
> >       struct swap_iocb *sio =3D NULL;
> >       struct swap_info_struct *sis =3D swp_swap_info(folio->swap);
> >       struct file *swap_file =3D sis->swap_file;
> > -     loff_t pos =3D folio_file_pos(folio);
> > +     loff_t pos =3D swap_file_pos(folio->swap);
> >
> >       count_swpout_vm_event(folio);
> >       folio_start_writeback(folio);
> > @@ -429,7 +429,7 @@ static void swap_read_folio_fs(struct folio *folio,=
 struct swap_iocb **plug)
> >  {
> >       struct swap_info_struct *sis =3D swp_swap_info(folio->swap);
> >       struct swap_iocb *sio =3D NULL;
> > -     loff_t pos =3D folio_file_pos(folio);
> > +     loff_t pos =3D swap_file_pos(folio->swap);
> >
> >       if (plug)
> >               sio =3D *plug;
> > diff --git a/mm/swap.h b/mm/swap.h
> > index fc2f6ade7f80..2de83729aaa8 100644
> > --- a/mm/swap.h
> > +++ b/mm/swap.h
> > @@ -7,6 +7,11 @@ struct mempolicy;
> >  #ifdef CONFIG_SWAP
> >  #include <linux/blk_types.h> /* for bio_end_io_t */
> >
> > +static inline loff_t swap_file_pos(swp_entry_t entry)
> > +{
> > +     return ((loff_t)swp_offset(entry)) << PAGE_SHIFT;
> > +}
> > +
> >  /* linux/mm/page_io.c */
> >  int sio_pool_init(void);
> >  struct swap_iocb;
>
> I feel that the file concept for swap is kind of confusing.  From the
> file cache point of view, one "struct address space" conresponds to one
> file.  If so, we have a simple file system on a swap device (block
> device backed or file backed), where the size of each file is 64M.  The
> swap entry encode the file system (swap_type), the file name
> (swap_offset >> SWAP_ADDRESS_SPACE_SHIFT), and the offset in file (lower
> bits of swap_offset).
>
> If the above definition is good, it's better to rename swap_file_pos()
> to swap_dev_pos(), because it returns the swap device position of the
> swap entry.

Good suggestion! The definition looks good to me, swap_dev_pos also
looks better, "swap_file" looks confusing indeed.

>
> And, when we reaches consensus on the swap file related concept, we may
> document it somewhere and review all naming in swap code to cleanup.
>
> --
> Best Regards,
> Huang, Ying