Skip to contentSkip to conversation

dax: Fix stack overflow when mounting fsdax pmem device

PATCH Superseded Adrian Huang 3 participants 3 messages Sep 16, 2020 – Sep 17, 2020

Subsystems (2)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 dax: Fix stack overflow when mounting fsdax pmem device

From: Adrian Huang <[email protected]>

When mounting fsdax pmem device, commit 6180bb446ab6 ("dax: fix detection of dax support for non-persistent memory block devices") introduces the stack overflow [1][2]. Here is the call path for mounting ext4 file system:
ext4_fill_super
bdev_dax_supported
__bdev_dax_supported
dax_supported
generic_fsdax_supported
__generic_fsdax_supported
bdev_dax_supported

The call path leads to the infinite calling loop, so we cannot call bdev_dax_supported() in __generic_fsdax_supported(). The sanity checking of the variable 'dax_dev' is moved prior to the two bdev_dax_pgoff() checks [3][4].

[1] https://lists.01.org/hyperkitty/list/[email protected]/thread/BULZHRILK7N2WS2JVISNF2QZNRQK6JU4/
[2] https://lists.01.org/hyperkitty/list/[email protected]/thread/OOZGFY3RNQGTGJJCH52YXCSYIDXMOPXO/
[3] https://lists.01.org/hyperkitty/list/[email protected]/message/SMQW2LY3QHPXOAW76RKNSCGG3QJFO7HT/
[4] https://lists.01.org/hyperkitty/list/[email protected]/message/7E2X6UGX5RQ2ISGYNAF66VLY5BKBFI4M/

Fixes 6180bb446ab6 ("dax: fix detection of dax support for non-persistent memory block devices")
Cc Coly Li <[email protected]>
Cc Jan Kara <[email protected]>
Cc Ira Weiny <[email protected]>
Cc John Pittman <[email protected]>
Cc Mikulas Patocka <[email protected]>
Cc Alasdair Kergon <[email protected]>
Cc Mike Snitzer <[email protected]>

---
v1->v2
* Remove the checking for the returned value '-EOPNOTSUPP' of dax_direct_access(). Jan has prepared a patch to address the issue in dm.

---
 drivers/dax/super.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/dax/super.c +6 -6
index e5767c83ea23..11d0541e6f8f 100644
@@ -85,6 +85,12 @@ bool __generic_fsdax_supported(struct dax_device *dax_dev,
 		return false;
 	}
 
+	if (!dax_dev) {
+		pr_debug("%s: error: dax unsupported by block device\n",
+				bdevname(bdev, buf));
+		return false;
+	}
+
 	err = bdev_dax_pgoff(bdev, start, PAGE_SIZE, &pgoff);
 	if (err) {
 		pr_info("%s: error: unaligned partition for dax\n",
@@ -100,12 +106,6 @@ bool __generic_fsdax_supported(struct dax_device *dax_dev,
 		return false;
 	}
 
-	if (!dax_dev || !bdev_dax_supported(bdev, blocksize)) {
-		pr_debug("%s: error: dax unsupported by block device\n",
-				bdevname(bdev, buf));
-		return false;
-	}
-
 	id = dax_read_lock();
 	len = dax_direct_access(dax_dev, pgoff, 1, &kaddr, &pfn);
 	len2 = dax_direct_access(dax_dev, pgoff_end, 1, &end_kaddr, &end_pfn);
Signature · 4 lines
2.17.1
_______________________________________________
Linux-nvdimm mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Reviewed-by on v2 1/1
#1 lore raw
Quoted 24 lines · Adrian Huang

From: Adrian Huang <[email protected]>

When mounting fsdax pmem device, commit 6180bb446ab6 ("dax: fix detection of dax support for non-persistent memory block devices") introduces the stack overflow [1][2]. Here is the call path for mounting ext4 file system:
ext4_fill_super
bdev_dax_supported
__bdev_dax_supported
dax_supported
generic_fsdax_supported
__generic_fsdax_supported
bdev_dax_supported

The call path leads to the infinite calling loop, so we cannot call bdev_dax_supported() in __generic_fsdax_supported(). The sanity checking of the variable 'dax_dev' is moved prior to the two bdev_dax_pgoff() checks [3][4].

[1] https://lists.01.org/hyperkitty/list/[email protected]/thread/BULZHRILK7N2WS2JVISNF2QZNRQK6JU4/
[2] https://lists.01.org/hyperkitty/list/[email protected]/thread/OOZGFY3RNQGTGJJCH52YXCSYIDXMOPXO/
[3] https://lists.01.org/hyperkitty/list/[email protected]/message/SMQW2LY3QHPXOAW76RKNSCGG3QJFO7HT/
[4] https://lists.01.org/hyperkitty/list/[email protected]/message/7E2X6UGX5RQ2ISGYNAF66VLY5BKBFI4M/

Quoted 9 lines
Fixes 6180bb446ab6 ("dax: fix detection of dax support for non-persistent memory block devices")
Cc Coly Li <[email protected]>
Cc Jan Kara <[email protected]>
Cc Ira Weiny <[email protected]>
Cc John Pittman <[email protected]>
Cc Mikulas Patocka <[email protected]>
Cc Alasdair Kergon <[email protected]>
Cc Mike Snitzer <[email protected]>

The patch looks good to me. You can add:

Honza

Quoted 42 lines

---
v1->v2
* Remove the checking for the returned value '-EOPNOTSUPP' of dax_direct_access(). Jan has prepared a patch to address the issue in dm.

---
 drivers/dax/super.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/dax/super.c +6 -6
index e5767c83ea23..11d0541e6f8f 100644
@@ -85,6 +85,12 @@ bool __generic_fsdax_supported(struct dax_device *dax_dev,
 		return false;
 	}
 
+	if (!dax_dev) {
+		pr_debug("%s: error: dax unsupported by block device\n",
+				bdevname(bdev, buf));
+		return false;
+	}
+
 	err = bdev_dax_pgoff(bdev, start, PAGE_SIZE, &pgoff);
 	if (err) {
 		pr_info("%s: error: unaligned partition for dax\n",
@@ -100,12 +106,6 @@ bool __generic_fsdax_supported(struct dax_device *dax_dev,
 		return false;
 	}
 
-	if (!dax_dev || !bdev_dax_supported(bdev, blocksize)) {
-		pr_debug("%s: error: dax unsupported by block device\n",
-				bdevname(bdev, buf));
-		return false;
-	}
-
 	id = dax_read_lock();
 	len = dax_direct_access(dax_dev, pgoff, 1, &kaddr, &pfn);
 	len2 = dax_direct_access(dax_dev, pgoff_end, 1, &end_kaddr, &end_pfn);
Signature · 1 line
2.17.1
Signature · 5 lines
Jan Kara <[email protected]>
SUSE Labs, CR
_______________________________________________
Linux-nvdimm mailing list -- [email protected]
To unsubscribe send an email to [email protected]
#2 lore raw
In reply to Adrian Huang
Quoted 15 lines · Adrian Huang

From: Adrian Huang <[email protected]>

When mounting fsdax pmem device, commit 6180bb446ab6 ("dax: fix detection of dax support for non-persistent memory block devices") introduces the stack overflow [1][2]. Here is the call path for mounting ext4 file system:
ext4_fill_super
bdev_dax_supported
__bdev_dax_supported
dax_supported
generic_fsdax_supported
__generic_fsdax_supported
bdev_dax_supported

Quoted 4 lines

The call path leads to the infinite calling loop, so we cannot call bdev_dax_supported() in __generic_fsdax_supported(). The sanity checking of the variable 'dax_dev' is moved prior to the two bdev_dax_pgoff() checks [3][4].

Looks good, and passes tests.

Quoted 4 lines

[1] https://lists.01.org/hyperkitty/list/[email protected]/thread/BULZHRILK7N2WS2JVISNF2QZNRQK6JU4/
[2] https://lists.01.org/hyperkitty/list/[email protected]/thread/OOZGFY3RNQGTGJJCH52YXCSYIDXMOPXO/
[3] https://lists.01.org/hyperkitty/list/[email protected]/message/SMQW2LY3QHPXOAW76RKNSCGG3QJFO7HT/
[4] https://lists.01.org/hyperkitty/list/[email protected]/message/7E2X6UGX5RQ2ISGYNAF66VLY5BKBFI4M/

lists.01.org has already proven to not be reliable for maintaining permanent list archive links. The most reliable way to reference a message is via lore.kernel.org.

    http://lore.kernel.org/r/$MSG_ID

For example your first link is here:

http://lore.kernel.org/r/[email protected]
_______________________________________________
Linux-nvdimm mailing list -- [email protected] To unsubscribe send an email to [email protected]