mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: torvalds@linux-foundation.org, akpm@linux-foundation.org,
	alan@lxorguk.ukuu.org.uk,
	Nicholas Bellinger <nab@linux-iscsi.org>,
	Christoph Hellwig <hch@lst.de>
Subject: [ 49/53] target: Fix bug in handling of FILEIO + block_device resize ops
Date: Fri, 18 May 2012 03:33:43 +0100	[thread overview]
Message-ID: <20120518023300.972513351@decadent.org.uk> (raw)
In-Reply-To: <20120518023254.339945758@decadent.org.uk>

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

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

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

commit cd9323fd68aee3c1c6b5b21e5746c9d1b586fb58 upstream.

This patch fixes a bug in the handling of FILEIO w/ underlying block_device
resize operations where the original fd_dev->fd_dev_size was incorrectly being
used in fd_get_blocks() for READ_CAPACITY response payloads.

This patch avoids using fd_dev->fd_dev_size for FILEIO devices with
an underlying block_device, and instead changes fd_get_blocks() to
get the sector count directly from i_size_read() as recommended by hch.

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
 drivers/target/target_core_file.c |   22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
index 7ed58e2..f286955 100644
--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -169,6 +169,7 @@ static struct se_device *fd_create_virtdevice(
 	inode = file->f_mapping->host;
 	if (S_ISBLK(inode->i_mode)) {
 		struct request_queue *q;
+		unsigned long long dev_size;
 		/*
 		 * Setup the local scope queue_limits from struct request_queue->limits
 		 * to pass into transport_add_device_to_core_hba() as struct se_dev_limits.
@@ -183,13 +184,12 @@ static struct se_device *fd_create_virtdevice(
 		 * one (1) logical sector from underlying struct block_device
 		 */
 		fd_dev->fd_block_size = bdev_logical_block_size(inode->i_bdev);
-		fd_dev->fd_dev_size = (i_size_read(file->f_mapping->host) -
+		dev_size = (i_size_read(file->f_mapping->host) -
 				       fd_dev->fd_block_size);
 
 		pr_debug("FILEIO: Using size: %llu bytes from struct"
 			" block_device blocks: %llu logical_block_size: %d\n",
-			fd_dev->fd_dev_size,
-			div_u64(fd_dev->fd_dev_size, fd_dev->fd_block_size),
+			dev_size, div_u64(dev_size, fd_dev->fd_block_size),
 			fd_dev->fd_block_size);
 	} else {
 		if (!(fd_dev->fbd_flags & FBDF_HAS_SIZE)) {
@@ -605,10 +605,20 @@ static u32 fd_get_device_type(struct se_device *dev)
 static sector_t fd_get_blocks(struct se_device *dev)
 {
 	struct fd_dev *fd_dev = dev->dev_ptr;
-	unsigned long long blocks_long = div_u64(fd_dev->fd_dev_size,
-			dev->se_sub_dev->se_dev_attrib.block_size);
+	struct file *f = fd_dev->fd_file;
+	struct inode *i = f->f_mapping->host;
+	unsigned long long dev_size;
+	/*
+	 * When using a file that references an underlying struct block_device,
+	 * ensure dev_size is always based on the current inode size in order
+	 * to handle underlying block_device resize operations.
+	 */
+	if (S_ISBLK(i->i_mode))
+		dev_size = (i_size_read(i) - fd_dev->fd_block_size);
+	else
+		dev_size = fd_dev->fd_dev_size;
 
-	return blocks_long;
+	return div_u64(dev_size, dev->se_sub_dev->se_dev_attrib.block_size);
 }
 
 static struct se_subsystem_api fileio_template = {
-- 
1.7.10




  parent reply	other threads:[~2012-05-18  2:58 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-18  2:32 [ 00/53] 3.2.18-stable review Ben Hutchings
2012-05-18  2:32 ` [ 01/53] regulator: Fix the logic to ensure new voltage setting in valid range Ben Hutchings
2012-05-18  2:32 ` [ 02/53] ARM: OMAP: Revert "ARM: OMAP: ctrl: Fix CONTROL_DSIPHY register fields" Ben Hutchings
2012-05-18  2:32 ` [ 03/53] ALSA: echoaudio: Remove incorrect part of assertion Ben Hutchings
2012-05-18  2:32 ` [ 04/53] ARM: orion5x: Fix GPIO enable bits for MPP9 Ben Hutchings
2012-05-18  2:32 ` [ 05/53] ALSA: HDA: Lessen CPU usage when waiting for chip to respond Ben Hutchings
2012-05-18  2:33 ` [ 06/53] percpu: pcpu_embed_first_chunk() should free unused parts after all allocs are complete Ben Hutchings
2012-05-18  2:33 ` [ 07/53] hugetlb: prevent BUG_ON in hugetlb_fault() -> hugetlb_cow() Ben Hutchings
2012-05-18  2:33 ` [ 08/53] namespaces, pid_ns: fix leakage on fork() failure Ben Hutchings
2012-05-18  2:33 ` [ 09/53] mm: nobootmem: fix sign extend problem in __free_pages_memory() Ben Hutchings
2012-05-18  2:33 ` [ 10/53] asix: Fix tx transfer padding for full-speed USB Ben Hutchings
2012-05-18  2:33 ` [ 11/53] netem: fix possible skb leak Ben Hutchings
2012-05-18  2:33 ` [ 12/53] net: In unregister_netdevice_notifier unregister the netdevices Ben Hutchings
2012-05-18  5:09   ` Herton Ronaldo Krzesinski
2012-05-18  5:41     ` David Miller
2012-05-19  0:31       ` Ben Hutchings
2012-05-19  3:43         ` David Miller
2012-05-18  2:33 ` [ 13/53] net: l2tp: unlock socket lock before returning from l2tp_ip_sendmsg Ben Hutchings
2012-05-18  2:33 ` [ 14/53] sky2: propogate rx hash when packet is copied Ben Hutchings
2012-05-18  2:33 ` [ 15/53] sky2: fix receive length error in mixed non-VLAN/VLAN traffic Ben Hutchings
2012-05-18  2:33 ` [ 16/53] sungem: Fix WakeOnLan Ben Hutchings
2012-05-18  2:33 ` [ 17/53] tg3: Avoid panic from reserved statblk field access Ben Hutchings
2012-05-18  2:33 ` [ 18/53] tcp: fix infinite cwnd in tcp_complete_cwr() Ben Hutchings
2012-05-18  2:33 ` [ 19/53] tcp: change tcp_adv_win_scale and tcp_rmem[2] Ben Hutchings
2012-05-18  2:33 ` [ 20/53] brcm80211: smac: pass missing argument to brcms_b_mute Ben Hutchings
2012-05-18  2:33 ` [ 21/53] phy:icplus:fix Auto Power Saving in ip101a_config_init Ben Hutchings
2012-05-18  2:33 ` [ 22/53] NFSv4: Revalidate uid/gid after open Ben Hutchings
2012-05-18  2:33 ` [ 23/53] target: Drop incorrect se_lun_acl release for dynamic -> explict ACL conversion Ben Hutchings
2012-05-18  2:33 ` [ 24/53] [media] marvell-cam: fix an ARM build error Ben Hutchings
2012-05-18  2:33 ` [ 25/53] [media] rc: Postpone ISR registration Ben Hutchings
2012-05-18  2:33 ` [ 26/53] cdc_ether: Ignore bogus union descriptor for RNDIS devices Ben Hutchings
2012-05-18  2:33 ` [ 27/53] jffs2: Fix lock acquisition order bug in gc path Ben Hutchings
2012-05-18  2:33 ` [ 28/53] [media] s5p-fimc: Fix locking in subdev set_crop op Ben Hutchings
2012-05-18  2:33 ` [ 29/53] dm mpath: check if scsi_dh module already loaded before trying to load Ben Hutchings
2012-05-18  2:33 ` [ 30/53] sparc64: Do not clobber %g2 in xcall_fetch_glob_regs() Ben Hutchings
2012-05-18  2:33 ` [ 31/53] gpio: Add missing spin_lock_init in gpio-ml-ioh driver Ben Hutchings
2012-05-18  2:33 ` [ 32/53] spi-topcliff-pch: Modify pci-bus number dynamically to get DMA device info Ben Hutchings
2012-05-18  2:33 ` [ 33/53] spi-topcliff-pch: Fix issue for transmitting over 4KByte Ben Hutchings
2012-05-18  2:33 ` [ 34/53] spi-topcliff-pch: supports a spi mode setup and bit order setup by IO control Ben Hutchings
2012-05-18  2:33 ` [ 35/53] spi-topcliff-pch: add recovery processing in case wait-event timeout Ben Hutchings
2012-05-18  2:33 ` [ 36/53] ext4: avoid deadlock on sync-mounted FS w/o journal Ben Hutchings
2012-05-18  2:33 ` [ 37/53] ia64: Add accept4() syscall Ben Hutchings
2012-05-18  2:33 ` [ 38/53] brcm80211: smac: fix endless retry of A-MPDU transmissions Ben Hutchings
2012-05-18  2:33 ` [ 39/53] ARM: 7417/1: vfp: ensure preemption is disabled when enabling VFP access Ben Hutchings
2012-05-18  2:33 ` [ 40/53] target: Fix SPC-2 RELEASE bug for multi-session iSCSI client setups Ben Hutchings
2012-05-18  2:33 ` [ 41/53] crypto: mv_cesa requires on CRYPTO_HASH to build Ben Hutchings
2012-05-18  2:33 ` [ 42/53] ALSA: hda/idt - Fix power-map for speaker-pins with some HP laptops Ben Hutchings
2012-05-18  2:33 ` [ 43/53] ASoC: wm8994: Fix AIF2ADC power down Ben Hutchings
2012-05-18  2:33 ` [ 44/53] usbnet: fix skb traversing races during unlink(v2) Ben Hutchings
2012-05-18  2:33 ` [ 45/53] cdc_ether: add Novatel USB551L device IDs for FLAG_WWAN Ben Hutchings
2012-05-18  2:33 ` [ 46/53] ARM: prevent VM_GROWSDOWN mmaps extending below FIRST_USER_ADDRESS Ben Hutchings
2012-05-18  2:33 ` [ 47/53] arch/tile: apply commit 74fca9da0 to the compat signal handling as well Ben Hutchings
2012-05-18  2:33 ` [ 48/53] MD: Add del_timer_sync to mddev_suspend (fix nasty panic) Ben Hutchings
2012-05-18  2:33 ` Ben Hutchings [this message]
2012-05-18  2:33 ` [ 50/53] tcp: do_tcp_sendpages() must try to push data out on oom conditions Ben Hutchings
2012-05-18  2:33 ` [ 51/53] e1000: Prevent reset task killing itself Ben Hutchings
2012-05-18  2:33 ` [ 52/53] mtd: map.h: fix arm cross-build failure Ben Hutchings
2012-05-18  2:33 ` [ 53/53] stmmac: Fix compilation error in mmc_core.c Ben Hutchings
2012-05-18 14:46 ` [ 00/53] 3.2.18-stable review Steven Rostedt
2012-05-18 18:22   ` Greg KH
2012-05-18 22:00     ` Ben Hutchings
2012-05-18 22:04       ` Willy Tarreau
2012-05-19  0:48       ` Steven Rostedt

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=20120518023300.972513351@decadent.org.uk \
    --to=ben@decadent.org.uk \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nab@linux-iscsi.org \
    --cc=stable@vger.kernel.org \
    --cc=torvalds@linux-foundation.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

Powered by JetHome