From patchwork Thu May 12 17:43:06 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jon Derrick X-Patchwork-Id: 9084931 Return-Path: X-Original-To: patchwork-linux-block@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 8BF65BF29F for ; Thu, 12 May 2016 17:45:26 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 93E3B20253 for ; Thu, 12 May 2016 17:45:25 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 88D7B2025A for ; Thu, 12 May 2016 17:45:24 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932065AbcELRpX (ORCPT ); Thu, 12 May 2016 13:45:23 -0400 Received: from mga04.intel.com ([192.55.52.120]:18569 "EHLO mga04.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932076AbcELRpV (ORCPT ); Thu, 12 May 2016 13:45:21 -0400 Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by fmsmga104.fm.intel.com with ESMTP; 12 May 2016 10:45:21 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,610,1455004800"; d="scan'208";a="964653695" Received: from eremita.lm.intel.com ([10.232.112.176]) by fmsmga001.fm.intel.com with ESMTP; 12 May 2016 10:45:20 -0700 From: Jon Derrick To: linux-block@vger.kernel.org Cc: Jon Derrick , "Jens Axboe" , "Alexander Viro" , linux-fsdevel@vger.kernel.org, "Dan Williams" , "Jeff Moyer" , "Stephen Bates" , "Keith Busch" , linux-nvme@lists.infradead.org, "Christoph Hellwig" Subject: [RFC 2/2] block: Introduce S_HIPRI inode flag Date: Thu, 12 May 2016 11:43:06 -0600 Message-Id: <1463074986-3070-3-git-send-email-jonathan.derrick@intel.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1463074986-3070-1-git-send-email-jonathan.derrick@intel.com> References: <1463074986-3070-1-git-send-email-jonathan.derrick@intel.com> Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org X-Spam-Status: No, score=-8.3 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=ham version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP S_HIPRI is a hint that indicates the file (currently only block devices) is a high priority file. This hint allows direct-io to the block device to poll for completions if polling is available to the block device. The motivation for this patch comes from tiered caching solutions. A user may wish to have low-latency block devices act as a cache for higher-latency storage media. With the introduction of block polling, polling could be enabled on a queue of a block device. The preadv2/pwritev2 sets allowed a user to specify per-io polling, but removed the ability to poll per-queue. Instead of having a user modify their software to use preadv2/pwritev2, this patch allows a user to set S_HIPRI on a block device file to request all direct-io for this file to be polled. Signed-off-by: Jon Derrick --- block/ioctl.c | 33 +++++++++++++++++++++++++++++++++ fs/block_dev.c | 3 +++ include/linux/fs.h | 2 ++ include/uapi/linux/fs.h | 2 ++ 4 files changed, 40 insertions(+) diff --git a/block/ioctl.c b/block/ioctl.c index 4ff1f92..5c7f1bd 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -520,6 +520,35 @@ static int blkdev_bszset(struct block_device *bdev, fmode_t mode, return ret; } +/* set the hipri flag */ +static int blkdev_hipriset(struct block_device *bdev, fmode_t mode, + int __user *argp) +{ + int n; + + if (!capable(CAP_SYS_ADMIN)) + return -EACCES; + if (!argp) + return -EINVAL; + if (get_user(n, argp)) + return -EFAULT; + + if (!(mode & FMODE_EXCL)) { + bdgrab(bdev); + if (blkdev_get(bdev, mode | FMODE_EXCL, &bdev) < 0) + return -EBUSY; + } + + if (n) + bdev->bd_inode->i_flags |= S_HIPRI; + else + bdev->bd_inode->i_flags &= ~S_HIPRI; + + if (!(mode & FMODE_EXCL)) + blkdev_put(bdev, mode | FMODE_EXCL); + return 0; +} + /* * always keep this in sync with compat_blkdev_ioctl() */ @@ -601,6 +630,10 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd, case BLKDAXGET: return put_int(arg, !!(bdev->bd_inode->i_flags & S_DAX)); break; + case BLKHIPRISET: + return blkdev_hipriset(bdev, mode, argp); + case BLKHIPRIGET: + return put_int(arg, !!(bdev->bd_inode->i_flags & S_HIPRI)); case IOC_PR_REGISTER: return blkdev_pr_register(bdev, argp); case IOC_PR_RESERVE: diff --git a/fs/block_dev.c b/fs/block_dev.c index d4fa725..6fa81c0 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -170,6 +170,9 @@ blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) if (IS_DAX(inode)) return dax_do_io(iocb, inode, iter, offset, blkdev_get_block, NULL, DIO_SKIP_DIO_COUNT); + + if (IS_HIPRI(inode)) + iocb->ki_flags |= IOCB_HIPRI; return __blockdev_direct_IO(iocb, inode, I_BDEV(inode), iter, offset, blkdev_get_block, NULL, NULL, DIO_SKIP_DIO_COUNT); diff --git a/include/linux/fs.h b/include/linux/fs.h index 70e61b5..8ae39ea 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1788,6 +1788,7 @@ struct super_operations { #else #define S_DAX 0 /* Make all the DAX code disappear */ #endif +#define S_HIPRI 16384 /* IO for this file has high priority */ /* * Note that nosuid etc flags are inode-specific: setting some file-system @@ -1826,6 +1827,7 @@ struct super_operations { #define IS_AUTOMOUNT(inode) ((inode)->i_flags & S_AUTOMOUNT) #define IS_NOSEC(inode) ((inode)->i_flags & S_NOSEC) #define IS_DAX(inode) ((inode)->i_flags & S_DAX) +#define IS_HIPRI(inode) ((inode)->i_flags & S_HIPRI) #define IS_WHITEOUT(inode) (S_ISCHR(inode->i_mode) && \ (inode)->i_rdev == WHITEOUT_DEV) diff --git a/include/uapi/linux/fs.h b/include/uapi/linux/fs.h index a079d50..d6e262c 100644 --- a/include/uapi/linux/fs.h +++ b/include/uapi/linux/fs.h @@ -223,6 +223,8 @@ struct fsxattr { #define BLKROTATIONAL _IO(0x12,126) #define BLKZEROOUT _IO(0x12,127) #define BLKDAXGET _IO(0x12,129) +#define BLKHIPRISET _IOW(0x12,130,int) +#define BLKHIPRIGET _IO(0x12,131) #define BMAP_IOCTL 1 /* obsolete - kept for compatibility */ #define FIBMAP _IO(0x00,1) /* bmap access */