From: John Hubbard <jhubbard@nvidia.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Jens Axboe <axboe@kernel.dk>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Miklos Szeredi <miklos@szeredi.hu>,
Christoph Hellwig <hch@infradead.org>,
"Darrick J . Wong" <djwong@kernel.org>,
Trond Myklebust <trond.myklebust@hammerspace.com>,
Anna Schumaker <anna@kernel.org>, Jan Kara <jack@suse.cz>,
Logan Gunthorpe <logang@deltatee.com>,
<linux-block@vger.kernel.org>, <linux-fsdevel@vger.kernel.org>,
<linux-xfs@vger.kernel.org>, <linux-nfs@vger.kernel.org>,
<linux-mm@kvack.org>, LKML <linux-kernel@vger.kernel.org>,
John Hubbard <jhubbard@nvidia.com>
Subject: [PATCH 6/6] fuse: convert direct IO paths to use FOLL_PIN
Date: Sat, 27 Aug 2022 01:36:07 -0700 [thread overview]
Message-ID: <20220827083607.2345453-7-jhubbard@nvidia.com> (raw)
In-Reply-To: <20220827083607.2345453-1-jhubbard@nvidia.com>
Convert the fuse filesystem to use pin_user_pages_fast() and
unpin_user_page(), instead of get_user_pages_fast() and put_page().
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
---
fs/fuse/dev.c | 8 ++++++--
fs/fuse/file.c | 31 ++++++++++++++++++++-----------
fs/fuse/fuse_i.h | 1 +
3 files changed, 27 insertions(+), 13 deletions(-)
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 51897427a534..eb841fc82bb9 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -675,7 +675,10 @@ static void fuse_copy_finish(struct fuse_copy_state *cs)
flush_dcache_page(cs->pg);
set_page_dirty_lock(cs->pg);
}
- put_page(cs->pg);
+ if (cs->pipebufs)
+ put_page(cs->pg);
+ else
+ dio_w_unpin_user_page(cs->pg);
}
cs->pg = NULL;
}
@@ -730,7 +733,8 @@ static int fuse_copy_fill(struct fuse_copy_state *cs)
}
} else {
size_t off;
- err = iov_iter_get_pages2(cs->iter, &page, PAGE_SIZE, 1, &off);
+ err = dio_w_iov_iter_pin_pages(cs->iter, &page, PAGE_SIZE, 1,
+ &off);
if (err < 0)
return err;
BUG_ON(!err);
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 1a3afd469e3a..a79aa4fea937 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -625,14 +625,19 @@ void fuse_read_args_fill(struct fuse_io_args *ia, struct file *file, loff_t pos,
}
static void fuse_release_user_pages(struct fuse_args_pages *ap,
- bool should_dirty)
+ bool should_dirty, bool is_kvec)
{
unsigned int i;
- for (i = 0; i < ap->num_pages; i++) {
- if (should_dirty)
- set_page_dirty_lock(ap->pages[i]);
- put_page(ap->pages[i]);
+ if (is_kvec) {
+ for (i = 0; i < ap->num_pages; i++) {
+ if (should_dirty)
+ set_page_dirty_lock(ap->pages[i]);
+ put_page(ap->pages[i]);
+ }
+ } else {
+ dio_w_unpin_user_pages_dirty_lock(ap->pages, ap->num_pages,
+ should_dirty);
}
}
@@ -733,7 +738,7 @@ static void fuse_aio_complete_req(struct fuse_mount *fm, struct fuse_args *args,
struct fuse_io_priv *io = ia->io;
ssize_t pos = -1;
- fuse_release_user_pages(&ia->ap, io->should_dirty);
+ fuse_release_user_pages(&ia->ap, io->should_dirty, io->is_kvec);
if (err) {
/* Nothing */
@@ -1414,10 +1419,10 @@ static int fuse_get_user_pages(struct fuse_args_pages *ap, struct iov_iter *ii,
while (nbytes < *nbytesp && ap->num_pages < max_pages) {
unsigned npages;
size_t start;
- ret = iov_iter_get_pages2(ii, &ap->pages[ap->num_pages],
- *nbytesp - nbytes,
- max_pages - ap->num_pages,
- &start);
+ ret = dio_w_iov_iter_pin_pages(ii, &ap->pages[ap->num_pages],
+ *nbytesp - nbytes,
+ max_pages - ap->num_pages,
+ &start);
if (ret < 0)
break;
@@ -1483,6 +1488,9 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, struct iov_iter *iter,
fl_owner_t owner = current->files;
size_t nbytes = min(count, nmax);
+ /* For use in fuse_release_user_pages(): */
+ io->is_kvec = iov_iter_is_kvec(iter);
+
err = fuse_get_user_pages(&ia->ap, iter, &nbytes, write,
max_pages);
if (err && !nbytes)
@@ -1498,7 +1506,8 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, struct iov_iter *iter,
}
if (!io->async || nres < 0) {
- fuse_release_user_pages(&ia->ap, io->should_dirty);
+ fuse_release_user_pages(&ia->ap, io->should_dirty,
+ io->is_kvec);
fuse_io_free(ia);
}
ia = NULL;
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 488b460e046f..1d927e499395 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -290,6 +290,7 @@ struct fuse_io_priv {
struct kiocb *iocb;
struct completion *done;
bool blocking;
+ bool is_kvec;
};
#define FUSE_IO_PRIV_SYNC(i) \
--
2.37.2
next prev parent reply other threads:[~2022-08-27 8:37 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-27 8:36 [PATCH 0/6] convert most filesystems to pin_user_pages_fast() John Hubbard
2022-08-27 8:36 ` [PATCH 1/6] mm/gup: introduce pin_user_page() John Hubbard
2022-08-29 12:07 ` David Hildenbrand
2022-08-29 19:33 ` John Hubbard
2022-08-30 12:17 ` David Hildenbrand
2022-08-30 21:42 ` John Hubbard
2022-08-31 0:06 ` John Hubbard
2022-08-27 8:36 ` [PATCH 2/6] block: add dio_w_*() wrappers for pin, unpin user pages John Hubbard
2022-08-27 22:27 ` Andrew Morton
2022-08-27 23:59 ` John Hubbard
2022-08-28 0:12 ` Andrew Morton
2022-08-28 0:31 ` John Hubbard
2022-08-28 1:07 ` John Hubbard
2022-08-27 8:36 ` [PATCH 3/6] iov_iter: new iov_iter_pin_pages*() routines John Hubbard
2022-08-27 22:46 ` Al Viro
2022-08-27 22:48 ` John Hubbard
2022-08-27 8:36 ` [PATCH 4/6] block, bio, fs: convert most filesystems to pin_user_pages_fast() John Hubbard
2022-08-27 8:36 ` [PATCH 5/6] NFS: direct-io: convert to FOLL_PIN pages John Hubbard
2022-08-27 22:48 ` Al Viro
2022-08-27 23:55 ` John Hubbard
2022-08-28 0:38 ` Al Viro
2022-08-28 0:39 ` Al Viro
2022-08-28 0:46 ` John Hubbard
2022-08-29 4:59 ` John Hubbard
2022-08-29 16:08 ` Jan Kara
2022-08-29 19:59 ` John Hubbard
2022-08-31 9:43 ` Jan Kara
2022-08-31 18:02 ` John Hubbard
2022-09-01 0:38 ` Al Viro
2022-09-01 9:06 ` Jan Kara
2022-08-27 8:36 ` John Hubbard [this message]
-- strict thread matches above, loose matches on Subject: below --
2022-02-27 9:34 [PATCH 0/6] block, fs: convert most Direct IO cases to FOLL_PIN jhubbard.send.patches
2022-02-27 9:34 ` [PATCH 6/6] fuse: convert direct IO paths to use FOLL_PIN jhubbard.send.patches
2022-02-28 15:59 ` Miklos Szeredi
2022-02-28 21:16 ` John Hubbard
2022-03-01 9:41 ` Miklos Szeredi
2022-03-02 8:07 ` John Hubbard
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=20220827083607.2345453-7-jhubbard@nvidia.com \
--to=jhubbard@nvidia.com \
--cc=akpm@linux-foundation.org \
--cc=anna@kernel.org \
--cc=axboe@kernel.dk \
--cc=djwong@kernel.org \
--cc=hch@infradead.org \
--cc=jack@suse.cz \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=miklos@szeredi.hu \
--cc=trond.myklebust@hammerspace.com \
--cc=viro@zeniv.linux.org.uk \
/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).