From patchwork Wed Aug 8 02:03:11 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Peng Tao X-Patchwork-Id: 1291841 Return-Path: X-Original-To: patchwork-linux-nfs@patchwork.kernel.org Delivered-To: patchwork-process-083081@patchwork2.kernel.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by patchwork2.kernel.org (Postfix) with ESMTP id E5EA2DF280 for ; Wed, 8 Aug 2012 02:03:34 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755430Ab2HHCDe (ORCPT ); Tue, 7 Aug 2012 22:03:34 -0400 Received: from mail-pb0-f46.google.com ([209.85.160.46]:65416 "EHLO mail-pb0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754699Ab2HHCDd (ORCPT ); Tue, 7 Aug 2012 22:03:33 -0400 Received: by mail-pb0-f46.google.com with SMTP id rr13so617318pbb.19 for ; Tue, 07 Aug 2012 19:03:33 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; bh=B1keC74NJ8hW2f95onmc6krefwBTsWlyI01XnhkoO/c=; b=eqUc6jTqFKtq8bkmGN8VZwmmZHG3RX5HedJiM+ONf7uF+kWJ/w0VM/0yKlcXaRs+nl yzA47e5C0d64SrxoV6p9rmadeJ0hEDoBbCflNMRBlGFHN6AoUtKBsX6EaWZ1u7jWPQHM XrqHuztosV91vuahrSc3mIq6EcuLkfk+GbrZHaTRvk4RALRCVPvSLAIlbp8ebTd9yzhk 2TcAwDK2NL45Zbf8g+lnC424cK/Gyr/4hdNrlLpUmE7DX5RD+uF4T4pzX/FzZShJTWNq w4noHVaBC9psV/Kf1puVXW7644Y6yWGEI+0Hh0le2hBr32GskWEcmwKn9SMSq3trY41q Nz4Q== Received: by 10.68.196.193 with SMTP id io1mr32700071pbc.17.1344391413347; Tue, 07 Aug 2012 19:03:33 -0700 (PDT) Received: from debian-sid.localdomain ([123.115.212.137]) by mx.google.com with ESMTPS id pj10sm12281300pbb.46.2012.08.07.19.03.31 (version=SSLv3 cipher=OTHER); Tue, 07 Aug 2012 19:03:33 -0700 (PDT) From: Peng Tao To: bharrosh@panasas.com Cc: linux-nfs@vger.kernel.org, Peng Tao Subject: [PATCH RFC 2/3] NFS41: send real write size in layoutget Date: Wed, 8 Aug 2012 10:03:11 +0800 Message-Id: <1344391392-1948-3-git-send-email-bergwolf@gmail.com> X-Mailer: git-send-email 1.7.1.262.g5ef3d In-Reply-To: <1344391392-1948-1-git-send-email-bergwolf@gmail.com> References: <1344391392-1948-1-git-send-email-bergwolf@gmail.com> Sender: linux-nfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org From: Peng Tao For bufferred write, scan dirty pages to find out longest continuous dirty pages. In this case, also allow layout driver to specify a maximum layoutget size which is useful to avoid busy scanning dirty pages for block layout client. For direct write, just use dreq->bytes_left. Signed-off-by: Peng Tao --- fs/nfs/direct.c | 7 ++++++ fs/nfs/internal.h | 1 + fs/nfs/pnfs.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 64 insertions(+), 2 deletions(-) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index c39f775..c1899dd 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include @@ -191,6 +192,12 @@ static void nfs_direct_req_release(struct nfs_direct_req *dreq) kref_put(&dreq->kref, nfs_direct_req_free); } +ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq) +{ + return dreq->bytes_left; +} +EXPORT_SYMBOL_GPL(nfs_dreq_bytes_left); + /* * Collects and returns the final error value/byte-count. */ diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 31fdb03..e68d329 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -464,6 +464,7 @@ static inline void nfs_inode_dio_wait(struct inode *inode) { inode_dio_wait(inode); } +extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq); /* nfs4proc.c */ extern void __nfs4_read_done_cb(struct nfs_read_data *); diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 2e00fea..e61a373 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -29,6 +29,7 @@ #include #include +#include #include #include "internal.h" #include "pnfs.h" @@ -1172,19 +1173,72 @@ pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *r } EXPORT_SYMBOL_GPL(pnfs_generic_pg_init_read); +/* + * Return the number of contiguous bytes in dirty pages for a given inode + * starting at page frame idx. + */ +static u64 pnfs_num_dirty_bytes(struct inode *inode, pgoff_t idx) +{ + struct address_space *mapping = inode->i_mapping; + pgoff_t index; + struct pagevec pvec; + pgoff_t num = 1; /* self */ + int i, done = 0; + + pagevec_init(&pvec, 0); + idx++; /* self */ + while (!done) { + index = idx; + pagevec_lookup_tag(&pvec, mapping, &index, + PAGECACHE_TAG_DIRTY, (pgoff_t)PAGEVEC_SIZE); + if (pagevec_count(&pvec) == 0) + break; + + for (i = 0; i < pagevec_count(&pvec); i++) { + struct page *page = pvec.pages[i]; + + lock_page(page); + if (unlikely(page->mapping != mapping) || + !PageDirty(page) || + PageWriteback(page) || + page->index != idx) { + done = 1; + unlock_page(page); + break; + } + unlock_page(page); + if (done) + break; + idx++; + num++; + } + pagevec_release(&pvec); + } + return num << PAGE_CACHE_SHIFT; +} + void -pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) +pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) { + u64 wb_size; + BUG_ON(pgio->pg_lseg != NULL); if (req->wb_offset != req->wb_pgbase) { nfs_pageio_reset_write_mds(pgio); return; } + + if (pgio->pg_dreq == NULL) + wb_size = pnfs_num_dirty_bytes(pgio->pg_inode, req->wb_index); + else + wb_size = nfs_dreq_bytes_left(pgio->pg_dreq); + pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, req_offset(req), - req->wb_bytes, + wb_size?:req->wb_bytes, IOMODE_RW, GFP_NOFS); /* If no lseg, fall back to write through mds */