From: "Benjamin Coddington" <bcodding@redhat.com>
To: trondmy@kernel.org
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH v4 21/21] NFS: Do uncached readdir when we're seeking a cookie in an empty page cache
Date: Mon, 09 Nov 2020 16:41:12 -0500 [thread overview]
Message-ID: <86F25343-0860-44A2-BA40-CFB640147D50@redhat.com> (raw)
In-Reply-To: <20201107140325.281678-22-trondmy@kernel.org>
On 7 Nov 2020, at 9:03, trondmy@kernel.org wrote:
> From: Trond Myklebust <trond.myklebust@hammerspace.com>
>
> If the directory is changing, causing the page cache to get
> invalidated
> while we are listing the contents, then the NFS client is currently
> forced
> to read in the entire directory contents from scratch, because it
> needs
> to perform a linear search for the readdir cookie. While this is not
> an issue for small directories, it does not scale to directories with
> millions of entries.
> In order to be able to deal with large directories that are changing,
> add a heuristic to ensure that if the page cache is empty, and we are
> searching for a cookie that is not the zero cookie, we just default to
> performing uncached readdir.
>
> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
> ---
> fs/nfs/dir.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
>
> diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
> index 238872d116f7..d7a9efd31ecd 100644
> --- a/fs/nfs/dir.c
> +++ b/fs/nfs/dir.c
> @@ -917,11 +917,28 @@ static int find_and_lock_cache_page(struct
> nfs_readdir_descriptor *desc)
> return res;
> }
>
> +static bool nfs_readdir_dont_search_cache(struct
> nfs_readdir_descriptor *desc)
> +{
> + struct address_space *mapping = desc->file->f_mapping;
> + struct inode *dir = file_inode(desc->file);
> + unsigned int dtsize = NFS_SERVER(dir)->dtsize;
> + loff_t size = i_size_read(dir);
> +
> + /*
> + * Default to uncached readdir if the page cache is empty, and
> + * we're looking for a non-zero cookie in a large directory.
> + */
> + return desc->dir_cookie != 0 && mapping->nrpages == 0 && size >
> dtsize;
inode size > dtsize is a little hand-wavy. We have a lot of customers
trying to
reverse-engineer nfs_readdir() behavior instead of reading the code,
this
is sure to drive them crazy.
That said, in the absence of an easy way to make it tunable, I don't
have
anything better to suggest.
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Ben
> +}
> +
> /* Search for desc->dir_cookie from the beginning of the page cache
> */
> static int readdir_search_pagecache(struct nfs_readdir_descriptor
> *desc)
> {
> int res;
>
> + if (nfs_readdir_dont_search_cache(desc))
> + return -EBADCOOKIE;
> +
> do {
> if (desc->page_index == 0) {
> desc->current_index = 0;
> --
> 2.28.0
next prev parent reply other threads:[~2020-11-09 21:41 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-07 14:03 [PATCH v4 00/21] Readdir enhancements trondmy
2020-11-07 14:03 ` [PATCH v4 01/21] NFS: Remove unnecessary inode locking in nfs_llseek_dir() trondmy
2020-11-07 14:03 ` [PATCH v4 02/21] NFS: Remove unnecessary inode lock in nfs_fsync_dir() trondmy
2020-11-07 14:03 ` [PATCH v4 03/21] NFS: Ensure contents of struct nfs_open_dir_context are consistent trondmy
2020-11-07 14:03 ` [PATCH v4 04/21] NFS: Clean up readdir struct nfs_cache_array trondmy
2020-11-07 14:03 ` [PATCH v4 05/21] NFS: Clean up nfs_readdir_page_filler() trondmy
2020-11-07 14:03 ` [PATCH v4 06/21] NFS: Clean up directory array handling trondmy
2020-11-07 14:03 ` [PATCH v4 07/21] NFS: Don't discard readdir results trondmy
2020-11-07 14:03 ` [PATCH v4 08/21] NFS: Remove unnecessary kmap in nfs_readdir_xdr_to_array() trondmy
2020-11-07 14:03 ` [PATCH v4 09/21] NFS: Replace kmap() with kmap_atomic() in nfs_readdir_search_array() trondmy
2020-11-07 14:03 ` [PATCH v4 10/21] NFS: Simplify struct nfs_cache_array_entry trondmy
2020-11-07 14:03 ` [PATCH v4 11/21] NFS: Support larger readdir buffers trondmy
2020-11-07 14:03 ` [PATCH v4 12/21] NFS: More readdir cleanups trondmy
2020-11-07 14:03 ` [PATCH v4 13/21] NFS: nfs_do_filldir() does not return a value trondmy
2020-11-07 14:03 ` [PATCH v4 14/21] NFS: Reduce readdir stack usage trondmy
2020-11-07 14:03 ` [PATCH v4 15/21] NFS: Cleanup to remove nfs_readdir_descriptor_t typedef trondmy
2020-11-07 14:03 ` [PATCH v4 16/21] NFS: Allow the NFS generic code to pass in a verifier to readdir trondmy
2020-11-07 14:03 ` [PATCH v4 17/21] NFS: Handle NFS4ERR_NOT_SAME and NFSERR_BADCOOKIE from readdir calls trondmy
2020-11-07 14:03 ` [PATCH v4 18/21] NFS: Improve handling of directory verifiers trondmy
2020-11-07 14:03 ` [PATCH v4 19/21] NFS: Optimisations for monotonically increasing readdir cookies trondmy
2020-11-07 14:03 ` [PATCH v4 20/21] NFS: Reduce number of RPC calls when doing uncached readdir trondmy
2020-11-07 14:03 ` [PATCH v4 21/21] NFS: Do uncached readdir when we're seeking a cookie in an empty page cache trondmy
2020-11-09 21:41 ` Benjamin Coddington [this message]
2020-11-09 21:46 ` Trond Myklebust
2020-11-11 16:43 ` Benjamin Coddington
2020-11-11 17:34 ` Trond Myklebust
2020-11-11 19:53 ` Benjamin Coddington
2020-11-11 20:11 ` Trond Myklebust
2020-11-10 14:48 ` David Wysochanski
2020-11-10 20:55 ` Trond Myklebust
2020-11-09 20:59 ` [PATCH v4 20/21] NFS: Reduce number of RPC calls when doing uncached readdir Benjamin Coddington
2020-11-09 13:15 ` [PATCH v4 00/21] Readdir enhancements David Wysochanski
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=86F25343-0860-44A2-BA40-CFB640147D50@redhat.com \
--to=bcodding@redhat.com \
--cc=linux-nfs@vger.kernel.org \
--cc=trondmy@kernel.org \
/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).