Return-Path: Received: from mx2.netapp.com ([216.240.18.37]:56731 "EHLO mx2.netapp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752796Ab0LAPgw (ORCPT ); Wed, 1 Dec 2010 10:36:52 -0500 From: Trond Myklebust To: Linus Torvalds , Nick Bowler Cc: Linux Kernel Mailing List , linux-nfs@vger.kernel.org Subject: [PATCH v2 2/3] NFS: lock the readdir page while it is in use Date: Wed, 1 Dec 2010 10:36:43 -0500 Message-Id: <1291217804-11257-3-git-send-email-Trond.Myklebust@netapp.com> In-Reply-To: <1291217804-11257-2-git-send-email-Trond.Myklebust@netapp.com> References: <1291217804-11257-1-git-send-email-Trond.Myklebust@netapp.com> <1291217804-11257-2-git-send-email-Trond.Myklebust@netapp.com> In-Reply-To: <20101201150428.GA2879@elliptictech.com> References: <20101201150428.GA2879@elliptictech.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: Content-Type: text/plain MIME-Version: 1.0 Before we can fix the memory leak due to not freeing up the contents of the nfs_cache_array when clearing the page cache, we need to ensure that shrink_page_list can't just lock the page and call try_to_release_page() while we're accessing the array. Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 16 +++++++++++++++- 1 files changed, 15 insertions(+), 1 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index e03537f..3ec3f1c 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -639,6 +639,7 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page* page) static void cache_page_release(nfs_readdir_descriptor_t *desc) { + unlock_page(desc->page); page_cache_release(desc->page); desc->page = NULL; } @@ -646,8 +647,20 @@ void cache_page_release(nfs_readdir_descriptor_t *desc) static struct page *get_cache_page(nfs_readdir_descriptor_t *desc) { - return read_cache_page(desc->file->f_path.dentry->d_inode->i_mapping, + struct page *page; + + for (;;) { + page = read_cache_page(desc->file->f_path.dentry->d_inode->i_mapping, desc->page_index, (filler_t *)nfs_readdir_filler, desc); + if (IS_ERR(page)) + break; + lock_page(page); + if (page->mapping != NULL && PageUptodate(page)) + break; + unlock_page(page); + page_cache_release(page); + } + return page; } /* @@ -771,6 +784,7 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, desc->page_index = 0; desc->last_cookie = *desc->dir_cookie; desc->page = page; + lock_page(page); status = nfs_readdir_xdr_to_array(desc, page, inode); if (status < 0) -- 1.7.3.2