public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	alan@lxorguk.ukuu.org.uk, Ferry <iscsitmp@bananateam.nl>,
	Christoph Hellwig <hch@lst.de>,
	Nicholas Bellinger <nab@linux-iscsi.org>
Subject: [ 32/76] target/file: Re-enable optional fd_buffered_io=1 operation
Date: Thu, 18 Oct 2012 19:46:56 -0700	[thread overview]
Message-ID: <20121019024355.294213734@linuxfoundation.org> (raw)
In-Reply-To: <20121019024350.087156547@linuxfoundation.org>

3.6-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Nicholas Bellinger <nab@linux-iscsi.org>

commit b32f4c7ed85c5cee2a21a55c9f59ebc9d57a2463 upstream.

This patch re-adds the ability to optionally run in buffered FILEIO mode
(eg: w/o O_DSYNC) for device backends in order to once again use the
Linux buffered cache as a write-back storage mechanism.

This logic was originally dropped with mainline v3.5-rc commit:

commit a4dff3043c231d57f982af635c9d2192ee40e5ae
Author: Nicholas Bellinger <nab@linux-iscsi.org>
Date:   Wed May 30 16:25:41 2012 -0700

    target/file: Use O_DSYNC by default for FILEIO backends

This difference with this patch is that fd_create_virtdevice() now
forces the explicit setting of emulate_write_cache=1 when buffered FILEIO
operation has been enabled.

(v2: Switch to FDBD_HAS_BUFFERED_IO_WCE + add more detailed
     comment as requested by hch)

Reported-by: Ferry <iscsitmp@bananateam.nl>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/target/target_core_file.c |   41 +++++++++++++++++++++++++++++++++++---
 drivers/target/target_core_file.h |    1 
 2 files changed, 39 insertions(+), 3 deletions(-)

--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -125,6 +125,19 @@ static struct se_device *fd_create_virtd
 	 * of pure timestamp updates.
 	 */
 	flags = O_RDWR | O_CREAT | O_LARGEFILE | O_DSYNC;
+	/*
+	 * Optionally allow fd_buffered_io=1 to be enabled for people
+	 * who want use the fs buffer cache as an WriteCache mechanism.
+	 *
+	 * This means that in event of a hard failure, there is a risk
+	 * of silent data-loss if the SCSI client has *not* performed a
+	 * forced unit access (FUA) write, or issued SYNCHRONIZE_CACHE
+	 * to write-out the entire device cache.
+	 */
+	if (fd_dev->fbd_flags & FDBD_HAS_BUFFERED_IO_WCE) {
+		pr_debug("FILEIO: Disabling O_DSYNC, using buffered FILEIO\n");
+		flags &= ~O_DSYNC;
+	}
 
 	file = filp_open(fd_dev->fd_dev_name, flags, 0600);
 	if (IS_ERR(file)) {
@@ -188,6 +201,12 @@ static struct se_device *fd_create_virtd
 	if (!dev)
 		goto fail;
 
+	if (fd_dev->fbd_flags & FDBD_HAS_BUFFERED_IO_WCE) {
+		pr_debug("FILEIO: Forcing setting of emulate_write_cache=1"
+			" with FDBD_HAS_BUFFERED_IO_WCE\n");
+		dev->se_sub_dev->se_dev_attrib.emulate_write_cache = 1;
+	}
+
 	fd_dev->fd_dev_id = fd_host->fd_host_dev_id_count++;
 	fd_dev->fd_queue_depth = dev->queue_depth;
 
@@ -407,6 +426,7 @@ enum {
 static match_table_t tokens = {
 	{Opt_fd_dev_name, "fd_dev_name=%s"},
 	{Opt_fd_dev_size, "fd_dev_size=%s"},
+	{Opt_fd_buffered_io, "fd_buffered_io=%d"},
 	{Opt_err, NULL}
 };
 
@@ -418,7 +438,7 @@ static ssize_t fd_set_configfs_dev_param
 	struct fd_dev *fd_dev = se_dev->se_dev_su_ptr;
 	char *orig, *ptr, *arg_p, *opts;
 	substring_t args[MAX_OPT_ARGS];
-	int ret = 0, token;
+	int ret = 0, arg, token;
 
 	opts = kstrdup(page, GFP_KERNEL);
 	if (!opts)
@@ -459,6 +479,19 @@ static ssize_t fd_set_configfs_dev_param
 					" bytes\n", fd_dev->fd_dev_size);
 			fd_dev->fbd_flags |= FBDF_HAS_SIZE;
 			break;
+		case Opt_fd_buffered_io:
+			match_int(args, &arg);
+			if (arg != 1) {
+				pr_err("bogus fd_buffered_io=%d value\n", arg);
+				ret = -EINVAL;
+				goto out;
+			}
+
+			pr_debug("FILEIO: Using buffered I/O"
+				" operations for struct fd_dev\n");
+
+			fd_dev->fbd_flags |= FDBD_HAS_BUFFERED_IO_WCE;
+			break;
 		default:
 			break;
 		}
@@ -490,8 +523,10 @@ static ssize_t fd_show_configfs_dev_para
 	ssize_t bl = 0;
 
 	bl = sprintf(b + bl, "TCM FILEIO ID: %u", fd_dev->fd_dev_id);
-	bl += sprintf(b + bl, "        File: %s  Size: %llu  Mode: O_DSYNC\n",
-		fd_dev->fd_dev_name, fd_dev->fd_dev_size);
+	bl += sprintf(b + bl, "        File: %s  Size: %llu  Mode: %s\n",
+		fd_dev->fd_dev_name, fd_dev->fd_dev_size,
+		(fd_dev->fbd_flags & FDBD_HAS_BUFFERED_IO_WCE) ?
+		"Buffered-WCE" : "O_DSYNC");
 	return bl;
 }
 
--- a/drivers/target/target_core_file.h
+++ b/drivers/target/target_core_file.h
@@ -14,6 +14,7 @@
 
 #define FBDF_HAS_PATH		0x01
 #define FBDF_HAS_SIZE		0x02
+#define FDBD_HAS_BUFFERED_IO_WCE 0x04
 
 struct fd_dev {
 	u32		fbd_flags;



  parent reply	other threads:[~2012-10-19  3:00 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-19  2:46 [ 00/76] 3.6.3-stable review Greg Kroah-Hartman
2012-10-19  2:46 ` [ 01/76] ARM: vfp: fix saving d16-d31 vfp registers on v6+ kernels Greg Kroah-Hartman
2012-10-19  2:46 ` [ 02/76] pnfsblock: fix partial page buffer wirte Greg Kroah-Hartman
2012-10-19  2:46 ` [ 03/76] NFS41: fix error of setting blocklayoutdriver Greg Kroah-Hartman
2012-10-19  2:46 ` [ 04/76] NFS: Remove bad delegations during open recovery Greg Kroah-Hartman
2012-10-19  2:46 ` [ 05/76] nfsd4: dont pin clientids to pseudoflavors Greg Kroah-Hartman
2012-10-19  2:46 ` [ 06/76] nfsd4: fix nfs4 stateid leak Greg Kroah-Hartman
2012-10-19  2:46 ` [ 07/76] NFSD: pass null terminated buf to kstrtouint() Greg Kroah-Hartman
2012-10-19  2:46 ` [ 08/76] lockd: per-net NSM client creation and destruction helpers introduced Greg Kroah-Hartman
2012-10-19  2:46 ` [ 09/76] lockd: use rpc clients cl_nodename for id encoding Greg Kroah-Hartman
2012-10-19 22:10   ` Ben Hutchings
2012-10-19 22:13     ` Ben Hutchings
2012-10-19  2:46 ` [ 10/76] lockd: create and use per-net NSM RPC clients on MON/UNMON requests Greg Kroah-Hartman
2012-10-19  2:46 ` [ 11/76] ACPI: EC: Make the GPE storm threshold a module parameter Greg Kroah-Hartman
2012-10-19  2:46 ` [ 12/76] ACPI: EC: Add a quirk for CLEVO M720T/M730T laptop Greg Kroah-Hartman
2012-10-19  2:46 ` [ 13/76] ALSA: hda - Add missing hda_gen_spec to struct via_spec Greg Kroah-Hartman
2012-10-19  2:46 ` [ 14/76] ALSA: hda - do not detect jack on internal speakers for Realtek Greg Kroah-Hartman
2012-10-19  2:46 ` [ 15/76] ALSA: hda - Fix memory leaks at error path in patch_cirrus.c Greg Kroah-Hartman
2012-10-19  2:46 ` [ 16/76] mips,kgdb: fix recursive page fault with CONFIG_KPROBES Greg Kroah-Hartman
2012-10-19  2:46 ` [ 17/76] tmpfs,ceph,gfs2,isofs,reiserfs,xfs: fix fh_len checking Greg Kroah-Hartman
2012-10-19  2:46 ` [ 18/76] iscsi-target: Correctly set 0xffffffff field within ISCSI_OP_REJECT PDU Greg Kroah-Hartman
2012-10-19  2:46 ` [ 19/76] iscsit: remove incorrect unlock in iscsit_build_sendtargets_resp Greg Kroah-Hartman
2012-10-19  2:46 ` [ 20/76] iscsi-target: Add explicit set of cache_dynamic_acls=1 for TPG demo-mode Greg Kroah-Hartman
2012-10-19  2:46 ` [ 21/76] iscsi-target: Bump defaults for nopin_timeout + nopin_response_timeout values Greg Kroah-Hartman
2012-10-19  2:46 ` [ 22/76] SCSI: storvsc: Account for in-transit packets in the RESET path Greg Kroah-Hartman
2012-10-19  2:46 ` [ 23/76] SCSI: scsi_debug: Fix off-by-one bug when unmapping region Greg Kroah-Hartman
2012-10-19  2:46 ` [ 24/76] SCSI: virtio-scsi: initialize scatterlist structure Greg Kroah-Hartman
2012-10-19  2:46 ` [ 25/76] ARM: 7541/1: Add ARM ERRATA 775420 workaround Greg Kroah-Hartman
2012-10-19  2:46 ` [ 26/76] ARM: OMAP: counter: add locking to read_persistent_clock Greg Kroah-Hartman
2012-10-19  2:46 ` [ 27/76] firewire: cdev: fix user memory corruption (i386 userland on amd64 kernel) Greg Kroah-Hartman
2012-10-19  2:46 ` [ 28/76] SUNRPC: Ensure that the TCP socket is closed when in CLOSE_WAIT Greg Kroah-Hartman
2012-10-19  2:46 ` [ 29/76] target: support zero allocation length in INQUIRY Greg Kroah-Hartman
2012-10-19  2:46 ` [ 30/76] target: fix truncation of mode data, support zero allocation length Greg Kroah-Hartman
2012-10-19  2:46 ` [ 31/76] target: fix return code in target_core_init_configfs error path Greg Kroah-Hartman
2012-10-19  2:46 ` Greg Kroah-Hartman [this message]
2012-10-19  2:46 ` [ 33/76] qla2xxx: Fix endianness of task management response code Greg Kroah-Hartman
2012-10-19  2:46 ` [ 34/76] vfio: Move PCI INTx eventfd setting earlier Greg Kroah-Hartman
2012-10-19  2:46 ` [ 35/76] vfio: Fix PCI INTx disable consistency Greg Kroah-Hartman
2012-10-19  2:47 ` [ 36/76] xen/pv-on-hvm kexec: add quirk for Xen 3.4 and shutdown watches Greg Kroah-Hartman
2012-10-19  2:47 ` [ 37/76] xen/bootup: allow {read|write}_cr8 pvops call Greg Kroah-Hartman
2012-10-19  2:47 ` [ 38/76] xen/bootup: allow read_tscp call for Xen PV guests Greg Kroah-Hartman
2012-10-19  2:47 ` [ 39/76] block: fix request_queue->flags initialization Greg Kroah-Hartman
2012-10-19  2:47 ` [ 40/76] autofs4 - fix reset pending flag on mount fail Greg Kroah-Hartman
2012-10-19  2:47 ` [ 41/76] module: taint kernel when lve module is loaded Greg Kroah-Hartman
2012-10-19  2:47 ` [ 42/76] video/udlfb: fix line counting in fb_write Greg Kroah-Hartman
2012-10-19  2:47 ` [ 43/76] viafb: dont touch clock state on OLPC XO-1.5 Greg Kroah-Hartman
2012-10-19  2:47 ` [ 44/76] timekeeping: Cast raw_interval to u64 to avoid shift overflow Greg Kroah-Hartman
2012-10-19  2:47 ` [ 45/76] timers: Fix endless looping between cascade() and internal_add_timer() Greg Kroah-Hartman
2012-10-19  2:47 ` [ 46/76] nohz: Fix one jiffy count too far in idle cputime Greg Kroah-Hartman
2012-10-19  2:47 ` [ 47/76] ath9k: use ieee80211_free_txskb Greg Kroah-Hartman
2012-10-19  2:47 ` [ 48/76] mac80211: use ieee80211_free_txskb to fix possible skb leaks Greg Kroah-Hartman
2012-10-19  2:47 ` [ 49/76] md/raid10: use correct limit variable Greg Kroah-Hartman
2012-10-19  2:47 ` [ 50/76] kdb,vt_console: Fix missed data due to pager overruns Greg Kroah-Hartman
2012-10-19  2:47 ` [ 51/76] pktgen: fix crash when generating IPv6 packets Greg Kroah-Hartman
2012-10-19  2:47 ` [ 52/76] MIPS: ath79: Fix CPU/DDR frequency calculation for SRIF PLLs Greg Kroah-Hartman
2012-10-19  2:47 ` [ 53/76] kbuild: Fix accidental revert in commit fe04ddf Greg Kroah-Hartman
2012-10-19  2:47 ` [ 54/76] Add CDC-ACM support for the CX93010-2x UCMxx USB Modem Greg Kroah-Hartman
2012-10-19  2:47 ` [ 55/76] fs: handle failed audit_log_start properly Greg Kroah-Hartman
2012-10-19  2:47 ` [ 56/76] fs: prevent use after free in auditing when symlink following was denied Greg Kroah-Hartman
2012-10-19  2:47 ` [ 57/76] drm/radeon: Dont destroy I2C Bus Rec in radeon_ext_tmds_enc_destroy() Greg Kroah-Hartman
2012-10-19  2:47 ` [ 58/76] drm/i915: remove useless BUG_ON which caused a regression in 3.5 Greg Kroah-Hartman
2012-10-19  2:47 ` [ 59/76] drm/i915: Set guardband clipping workaround bit in the right register Greg Kroah-Hartman
2012-10-19  2:47 ` [ 60/76] drm/nouveau/bios: fix shadowing of ACPI ROMs larger than 64KiB Greg Kroah-Hartman
2012-10-19 16:01   ` Heinz Diehl
2012-10-19 17:48     ` Greg Kroah-Hartman
2012-10-19 19:11       ` Heinz Diehl
2012-10-21 16:31         ` Greg Kroah-Hartman
2012-10-19  2:47 ` [ 61/76] drm/i915: use adjusted_mode instead of mode for checking the 6bpc force flag Greg Kroah-Hartman
2012-10-19  2:47 ` [ 62/76] mcs7830: Fix link state detection Greg Kroah-Hartman
2012-10-19  2:47 ` [ 63/76] jbd: Fix assertion failure in commit code due to lacking transaction credits Greg Kroah-Hartman
2012-10-19  2:47 ` [ 64/76] mtd: nand: allow NAND_NO_SUBPAGE_WRITE to be set from driver Greg Kroah-Hartman
2012-10-19  2:47 ` [ 65/76] e1000e: Change wthresh to 1 to avoid possible Tx stalls Greg Kroah-Hartman
2012-10-19  2:47 ` [ 66/76] tpm: Propagate error from tpm_transmit to fix a timeout hang Greg Kroah-Hartman
2012-10-19  2:47 ` [ 67/76] usb: gadget: at91_udc: fix dt support Greg Kroah-Hartman
2012-10-19  2:47 ` [ 68/76] ALSA: hda - Fix registration race of VGA switcheroo Greg Kroah-Hartman
2012-10-19  2:47 ` [ 69/76] ALSA: hda - Stop LPIB delay counting on broken hardware Greg Kroah-Hartman
2012-10-19  2:47 ` [ 70/76] ALSA: hda - Always check array bounds in alc_get_line_out_pfx Greg Kroah-Hartman
2012-10-19  2:47 ` [ 71/76] ASoC: fsi: dont reschedule DMA from an atomic context Greg Kroah-Hartman
2012-10-19  2:47 ` [ 72/76] ASoC: wm2200: Use rev A register patches on rev B Greg Kroah-Hartman
2012-10-19  2:47 ` [ 73/76] ASoC: wm2200: Fix non-inverted OUT2 mute control Greg Kroah-Hartman
2012-10-19  2:47 ` [ 74/76] ASoC: omap-abe-twl6040: Fix typo of Vibrator Greg Kroah-Hartman
2012-10-19  2:47 ` [ 75/76] ALSA: ac97 - Fix missing NULL check in snd_ac97_cvol_new() Greg Kroah-Hartman
2012-10-19  2:47 ` [ 76/76] ALSA: emu10k1: add chip details for E-mu 1010 PCIe card Greg Kroah-Hartman

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=20121019024355.294213734@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=hch@lst.de \
    --cc=iscsitmp@bananateam.nl \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nab@linux-iscsi.org \
    --cc=stable@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