From: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
To: linux-scsi <linux-scsi@vger.kernel.org>,
James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Nicholas Bellinger <nab@linux-iscsi.org>,
Joel Becker <jlbec@evilplan.org>, Christoph Hellwig <hch@lst.de>
Subject: [PATCH 20/31] target: Move subdev release logic into ->release() callback
Date: Wed, 9 Feb 2011 15:34:55 -0800 [thread overview]
Message-ID: <1297294506-23579-21-git-send-email-nab@linux-iscsi.org> (raw)
In-Reply-To: <1297294506-23579-1-git-send-email-nab@linux-iscsi.org>
From: Nicholas Bellinger <nab@linux-iscsi.org>
This patch moves the se_free_virtual_device() / se_subsystem_api->free_device()
and subsequent release of struct se_subsystem_dev memory to inside of the
configfs callback target_core_dev_item_ops->release() called from within
fs/configfs/item.c:config_item_cleanup() context. This patch resolves the
following SLUB 'Poison overwritten' warning when calling target_core_drop_subdev()
-> kfree(se_dev) directly after config_item_put():
[ 5147.638639] =============================================================================
[ 5147.638959] BUG kmalloc-4096: Poison overwritten
[ 5147.639124] -----------------------------------------------------------------------------
[ 5147.639124]
[ 5147.639294] INFO: 0xffff88000d2f887c-0xffff88000d2f887c. First byte 0x6a instead of 0x6b
[ 5147.639294] INFO: Allocated in kzalloc+0xf/0x11 [target_core_mod] age=2602 cpu=0 pid=14639
[ 5147.639294] INFO: Freed in target_core_drop_subdev+0x18d/0x199 [target_core_mod] age=25 cpu=0 pid=14654
[ 5147.639294] INFO: Slab 0xffffea00002e2640 objects=7 used=1 fp=0xffff88000d2f8000 flags=0x1000000000040c1
[ 5147.639294] INFO: Object 0xffff88000d2f8000 @offset=0 fp=0xffff88000d2fb0d8
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas A. Bellinger <nab@linux-iscsi.org>
---
drivers/target/target_core_configfs.c | 70 +++++++++++++++------------------
1 files changed, 32 insertions(+), 38 deletions(-)
diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c
index 3438e73..812eb8b 100644
--- a/drivers/target/target_core_configfs.c
+++ b/drivers/target/target_core_configfs.c
@@ -2003,13 +2003,35 @@ static void target_core_dev_release(struct config_item *item)
{
struct se_subsystem_dev *se_dev = container_of(to_config_group(item),
struct se_subsystem_dev, se_dev_group);
- struct config_group *dev_cg;
-
- if (!(se_dev))
- return;
+ struct se_hba *hba = item_to_hba(&se_dev->se_dev_hba->hba_group.cg_item);
+ struct se_subsystem_api *t = hba->transport;
+ struct config_group *dev_cg = &se_dev->se_dev_group;
- dev_cg = &se_dev->se_dev_group;
kfree(dev_cg->default_groups);
+ /*
+ * This pointer will set when the storage is enabled with:
+ *`echo 1 > $CONFIGFS/core/$HBA/$DEV/dev_enable`
+ */
+ if (se_dev->se_dev_ptr) {
+ printk(KERN_INFO "Target_Core_ConfigFS: Calling se_free_"
+ "virtual_device() for se_dev_ptr: %p\n",
+ se_dev->se_dev_ptr);
+
+ se_free_virtual_device(se_dev->se_dev_ptr, hba);
+ } else {
+ /*
+ * Release struct se_subsystem_dev->se_dev_su_ptr..
+ */
+ printk(KERN_INFO "Target_Core_ConfigFS: Calling t->free_"
+ "device() for se_dev_su_ptr: %p\n",
+ se_dev->se_dev_su_ptr);
+
+ t->free_device(se_dev->se_dev_su_ptr);
+ }
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Deallocating se_subsystem"
+ "_dev_t: %p\n", se_dev);
+ kfree(se_dev);
}
static ssize_t target_core_dev_show(struct config_item *item,
@@ -2803,13 +2825,11 @@ static void target_core_drop_subdev(
struct se_subsystem_api *t;
struct config_item *df_item;
struct config_group *dev_cg, *tg_pt_gp_cg;
- int i, ret;
+ int i;
hba = item_to_hba(&se_dev->se_dev_hba->hba_group.cg_item);
- if (mutex_lock_interruptible(&hba->hba_access_mutex))
- goto out;
-
+ mutex_lock(&hba->hba_access_mutex);
t = hba->transport;
spin_lock(&se_global->g_device_lock);
@@ -2832,38 +2852,12 @@ static void target_core_drop_subdev(
dev_cg->default_groups[i] = NULL;
config_item_put(df_item);
}
-
- config_item_put(item);
/*
- * This pointer will set when the storage is enabled with:
- * `echo 1 > $CONFIGFS/core/$HBA/$DEV/dev_enable`
+ * The releasing of se_dev and associated se_dev->se_dev_ptr is done
+ * from target_core_dev_item_ops->release() ->target_core_dev_release().
*/
- if (se_dev->se_dev_ptr) {
- printk(KERN_INFO "Target_Core_ConfigFS: Calling se_free_"
- "virtual_device() for se_dev_ptr: %p\n",
- se_dev->se_dev_ptr);
-
- ret = se_free_virtual_device(se_dev->se_dev_ptr, hba);
- if (ret < 0)
- goto hba_out;
- } else {
- /*
- * Release struct se_subsystem_dev->se_dev_su_ptr..
- */
- printk(KERN_INFO "Target_Core_ConfigFS: Calling t->free_"
- "device() for se_dev_su_ptr: %p\n",
- se_dev->se_dev_su_ptr);
-
- t->free_device(se_dev->se_dev_su_ptr);
- }
-
- printk(KERN_INFO "Target_Core_ConfigFS: Deallocating se_subsystem"
- "_dev_t: %p\n", se_dev);
-
-hba_out:
+ config_item_put(item);
mutex_unlock(&hba->hba_access_mutex);
-out:
- kfree(se_dev);
}
static struct configfs_group_operations target_core_hba_group_ops = {
--
1.7.4
next prev parent reply other threads:[~2011-02-09 23:35 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-02-09 23:34 [PATCH 00/31] target: mainline updates for .38-rc5 Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 01/31] target: iblock/pscsi claim checking for NULL instead of IS_ERR Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 02/31] target: fix dubious one-bit signed bitfield Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 03/31] target/iblock: Fix failed bd claim NULL pointer dereference Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 04/31] target: Fix memory leak on error path Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 05/31] target/file: Fix memory leak in fd_set_configfs_dev_params() Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 06/31] target/iblock: Fix memory leak in iblock_set_configfs_dev_params Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 07/31] target: Fix memory leaks in target_core_dev_pr_store_attr_res_aptpl_metadata Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 08/31] target: Fix demo-mode MappedLUN shutdown UA/PR breakage Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 09/31] target: Release left-over demo-mode NodeACLs w/ tfo->tpg_check_demo_mode_cache()=1 Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 10/31] target: tcm_mod_builder.py generated Makefile cleanups Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 11/31] target: do not include target_core_mib.h under include/target Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 12/31] target: Convert backend ->create_virtdevice() call to return ERR_PTR Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 13/31] target: Drop nacl->device_list_lock on core_update_device_list_for_node failure Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 14/31] target: Convert rd_build_device_space() to use errno Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 15/31] target: Convert TMR REQ/RSP definitions to target namespace Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 16/31] target core v4.0.0-rc7 Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 17/31] target: Avoid mem leak and needless work in transport_generic_get_mem Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 18/31] target: Fix top-level configfs_subsystem default_group shutdown breakage Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 19/31] target: Move core_delete_hba() into ->release() callback Nicholas A. Bellinger
2011-02-09 23:34 ` Nicholas A. Bellinger [this message]
2011-02-09 23:34 ` [PATCH 21/31] target: Move core_alua_free_lu_gp() " Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 22/31] target: Move core_alua_free_tg_pt_gp() " Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 23/31] target: Move fabric dependent struct se_wwn free " Nicholas A. Bellinger
2011-02-09 23:34 ` [PATCH 24/31] target: Move fabric dependent se_portal_group " Nicholas A. Bellinger
2011-02-09 23:35 ` [PATCH 25/31] target: Move fabric dependent se_node_acl free into ->release callback() Nicholas A. Bellinger
2011-02-09 23:35 ` [PATCH 26/31] target: Move fabric dependent struct se_tpg_np free into ->release() callback Nicholas A. Bellinger
2011-02-09 23:35 ` [PATCH 27/31] target: Move fabric independent se_lun_acl " Nicholas A. Bellinger
2011-02-09 23:35 ` [PATCH 28/31] target: Remove procfs based target_core_mib.c code Nicholas A. Bellinger
2011-02-09 23:35 ` [PATCH 29/31] target: Fix SCF_SCSI_CONTROL_SG_IO_CDB breakage Nicholas A. Bellinger
2011-02-09 23:35 ` [PATCH 30/31] target: Fix bogus return in transport_add_device_to_core_hba failure path Nicholas A. Bellinger
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=1297294506-23579-21-git-send-email-nab@linux-iscsi.org \
--to=nab@linux-iscsi.org \
--cc=James.Bottomley@HansenPartnership.com \
--cc=hch@lst.de \
--cc=jlbec@evilplan.org \
--cc=linux-scsi@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;
as well as URLs for NNTP newsgroup(s).