Linux NFS development
 help / color / mirror / Atom feed
From: trondmy@kernel.org
To: linux-nfs@vger.kernel.org
Subject: [PATCH v7 10/21] NFS: Reduce use of uncached readdir
Date: Wed, 23 Feb 2022 16:12:54 -0500	[thread overview]
Message-ID: <20220223211305.296816-11-trondmy@kernel.org> (raw)
In-Reply-To: <20220223211305.296816-10-trondmy@kernel.org>

From: Trond Myklebust <trond.myklebust@hammerspace.com>

When reading a very large directory, we want to try to keep the page
cache up to date if doing so is inexpensive. With the change to allow
readdir to continue reading even when the cache is incomplete, we no
longer need to fall back to uncached readdir in order to scale to large
directories.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/dir.c | 23 +++--------------------
 1 file changed, 3 insertions(+), 20 deletions(-)

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 9b0f13b52dbf..982b5dbe30d7 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -986,28 +986,11 @@ 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;
-}
-
 /* 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;
@@ -1262,10 +1245,10 @@ static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int whence)
 	}
 	if (offset != filp->f_pos) {
 		filp->f_pos = offset;
-		if (!nfs_readdir_use_cookie(filp)) {
+		dir_ctx->page_index = 0;
+		if (!nfs_readdir_use_cookie(filp))
 			dir_ctx->dir_cookie = 0;
-			dir_ctx->page_index = 0;
-		} else
+		else
 			dir_ctx->dir_cookie = offset;
 		if (offset == 0)
 			memset(dir_ctx->verf, 0, sizeof(dir_ctx->verf));
-- 
2.35.1


  reply	other threads:[~2022-02-23 21:21 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-23 21:12 [PATCH v7 00/21] Readdir improvements trondmy
2022-02-23 21:12 ` [PATCH v7 01/21] NFS: constify nfs_server_capable() and nfs_have_writebacks() trondmy
2022-02-23 21:12   ` [PATCH v7 02/21] NFS: Trace lookup revalidation failure trondmy
2022-02-23 21:12     ` [PATCH v7 03/21] NFS: Use kzalloc() to avoid initialising the nfs_open_dir_context trondmy
2022-02-23 21:12       ` [PATCH v7 04/21] NFS: Calculate page offsets algorithmically trondmy
2022-02-23 21:12         ` [PATCH v7 05/21] NFS: Store the change attribute in the directory page cache trondmy
2022-02-23 21:12           ` [PATCH v7 06/21] NFS: If the cookie verifier changes, we must invalidate the " trondmy
2022-02-23 21:12             ` [PATCH v7 07/21] NFS: Don't re-read the entire page cache to find the next cookie trondmy
2022-02-23 21:12               ` [PATCH v7 08/21] NFS: Adjust the amount of readahead performed by NFS readdir trondmy
2022-02-23 21:12                 ` [PATCH v7 09/21] NFS: Simplify nfs_readdir_xdr_to_array() trondmy
2022-02-23 21:12                   ` trondmy [this message]
2022-02-23 21:12                     ` [PATCH v7 11/21] NFS: Improve heuristic for readdirplus trondmy
2022-02-23 21:12                       ` [PATCH v7 12/21] NFS: Don't ask for readdirplus unless it can help nfs_getattr() trondmy
2022-02-23 21:12                         ` [PATCH v7 13/21] NFSv4: Ask for a full XDR buffer of readdir goodness trondmy
2022-02-23 21:12                           ` [PATCH v7 14/21] NFS: Readdirplus can't help lookup for case insensitive filesystems trondmy
2022-02-23 21:12                             ` [PATCH v7 15/21] NFS: Don't request readdirplus when revalidation was forced trondmy
2022-02-23 21:13                               ` [PATCH v7 16/21] NFS: Add basic readdir tracing trondmy
2022-02-23 21:13                                 ` [PATCH v7 17/21] NFS: Trace effects of readdirplus on the dcache trondmy
2022-02-23 21:13                                   ` [PATCH v7 18/21] NFS: Trace effects of the readdirplus heuristic trondmy
2022-02-23 21:13                                     ` [PATCH v7 19/21] NFS: Convert readdir page cache to use a cookie based index trondmy
2022-02-23 21:13                                       ` [PATCH v7 20/21] NFS: Fix up forced readdirplus trondmy
2022-02-23 21:13                                         ` [PATCH v7 21/21] NFS: Remove unnecessary cache invalidations for directories trondmy
2022-02-24 17:31                                       ` [PATCH v7 19/21] NFS: Convert readdir page cache to use a cookie based index Benjamin Coddington
2022-02-25  2:33                                         ` Trond Myklebust
2022-02-25  3:17                                           ` NeilBrown
2022-02-25  4:25                                             ` Trond Myklebust
2022-02-25 12:33                                               ` Benjamin Coddington
2022-02-25 13:11                                                 ` Trond Myklebust
2022-02-24 15:53                                 ` [PATCH v7 16/21] NFS: Add basic readdir tracing Benjamin Coddington
2022-02-25  2:35                                   ` Trond Myklebust
2022-02-24 16:55                     ` [PATCH v7 10/21] NFS: Reduce use of uncached readdir Anna Schumaker
2022-02-25  4:07                       ` Trond Myklebust
2022-02-24 16:30                 ` [PATCH v7 08/21] NFS: Adjust the amount of readahead performed by NFS readdir Anna Schumaker
2022-02-24 16:18             ` [PATCH v7 06/21] NFS: If the cookie verifier changes, we must invalidate the page cache Anna Schumaker
2022-02-24 14:53           ` [PATCH v7 05/21] NFS: Store the change attribute in the directory " Benjamin Coddington
2022-02-25  2:26             ` Trond Myklebust
2022-02-25  3:51               ` Trond Myklebust
2022-02-25 11:38                 ` Benjamin Coddington
2022-02-25 13:10                   ` Trond Myklebust
2022-02-25 13:26                     ` Trond Myklebust
2022-02-25 14:44                     ` Benjamin Coddington
2022-02-25 15:18                       ` Trond Myklebust
2022-02-25 15:34                         ` Benjamin Coddington
2022-02-25 20:23                           ` Benjamin Coddington
2022-02-25 20:28                             ` Benjamin Coddington
2022-02-25 20:41                             ` Trond Myklebust
2022-02-25 22:04                               ` Benjamin Coddington
2022-02-25 22:29                               ` Trond Myklebust
2022-02-24 14:15         ` [PATCH v7 04/21] NFS: Calculate page offsets algorithmically Benjamin Coddington
2022-02-25  2:11           ` Trond Myklebust
2022-02-25 11:28             ` Benjamin Coddington
2022-02-25 12:44               ` Trond Myklebust
2022-02-24 14:14     ` [PATCH v7 02/21] NFS: Trace lookup revalidation failure Benjamin Coddington
2022-02-25  2:09       ` Trond Myklebust
2022-02-24 12:25 ` [PATCH v7 00/21] Readdir improvements David Wysochanski
2022-02-25  4:00   ` Trond Myklebust
2022-02-24 15:07 ` 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=20220223211305.296816-11-trondmy@kernel.org \
    --to=trondmy@kernel.org \
    --cc=linux-nfs@vger.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