lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20210614102700.653688505@linuxfoundation.org>
Date:   Mon, 14 Jun 2021 12:26:47 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     linux-kernel@...r.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        stable@...r.kernel.org, Dmitry Fomichev <Dmitry.Fomichev@....com>,
        Johannes Thumshirn <johannes.thumshirn@....com>,
        Naohiro Aota <naohiro.aota@....com>,
        David Sterba <dsterba@...e.com>
Subject: [PATCH 5.12 075/173] btrfs: zoned: fix zone number to sector/physical calculation

From: Naohiro Aota <naohiro.aota@....com>

commit 5b434df8778771d181bc19fb4593bca114d1c4eb upstream.

In btrfs_get_dev_zone_info(), we have "u32 sb_zone" and calculate "sector_t
sector" by shifting it. But, this "sector" is calculated in 32bit, leading
it to be 0 for the 2nd superblock copy.

Since zone number is u32, shifting it to sector (sector_t) or physical
address (u64) can easily trigger a missing cast bug like this.

This commit introduces helpers to convert zone number to sector/LBA, so we
won't fall into the same pitfall again.

Reported-by: Dmitry Fomichev <Dmitry.Fomichev@....com>
Fixes: 12659251ca5d ("btrfs: implement log-structured superblock for ZONED mode")
CC: stable@...r.kernel.org # 5.11+
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@....com>
Signed-off-by: Naohiro Aota <naohiro.aota@....com>
Reviewed-by: David Sterba <dsterba@...e.com>
Signed-off-by: David Sterba <dsterba@...e.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
 fs/btrfs/zoned.c |   23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -150,6 +150,18 @@ static inline u32 sb_zone_number(int shi
 	return (u32)zone;
 }
 
+static inline sector_t zone_start_sector(u32 zone_number,
+					 struct block_device *bdev)
+{
+	return (sector_t)zone_number << ilog2(bdev_zone_sectors(bdev));
+}
+
+static inline u64 zone_start_physical(u32 zone_number,
+				      struct btrfs_zoned_device_info *zone_info)
+{
+	return (u64)zone_number << zone_info->zone_size_shift;
+}
+
 /*
  * Emulate blkdev_report_zones() for a non-zoned device. It slices up the block
  * device into static sized chunks and fake a conventional zone on each of
@@ -405,8 +417,8 @@ int btrfs_get_dev_zone_info(struct btrfs
 		if (sb_zone + 1 >= zone_info->nr_zones)
 			continue;
 
-		sector = sb_zone << (zone_info->zone_size_shift - SECTOR_SHIFT);
-		ret = btrfs_get_dev_zones(device, sector << SECTOR_SHIFT,
+		ret = btrfs_get_dev_zones(device,
+					  zone_start_physical(sb_zone, zone_info),
 					  &zone_info->sb_zones[sb_pos],
 					  &nr_zones);
 		if (ret)
@@ -721,7 +733,7 @@ int btrfs_sb_log_location_bdev(struct bl
 	if (sb_zone + 1 >= nr_zones)
 		return -ENOENT;
 
-	ret = blkdev_report_zones(bdev, sb_zone << zone_sectors_shift,
+	ret = blkdev_report_zones(bdev, zone_start_sector(sb_zone, bdev),
 				  BTRFS_NR_SB_LOG_ZONES, copy_zone_info_cb,
 				  zones);
 	if (ret < 0)
@@ -826,7 +838,7 @@ int btrfs_reset_sb_log_zones(struct bloc
 		return -ENOENT;
 
 	return blkdev_zone_mgmt(bdev, REQ_OP_ZONE_RESET,
-				sb_zone << zone_sectors_shift,
+				zone_start_sector(sb_zone, bdev),
 				zone_sectors * BTRFS_NR_SB_LOG_ZONES, GFP_NOFS);
 }
 
@@ -878,7 +890,8 @@ u64 btrfs_find_allocatable_zones(struct
 			if (!(end <= sb_zone ||
 			      sb_zone + BTRFS_NR_SB_LOG_ZONES <= begin)) {
 				have_sb = true;
-				pos = ((u64)sb_zone + BTRFS_NR_SB_LOG_ZONES) << shift;
+				pos = zone_start_physical(
+					sb_zone + BTRFS_NR_SB_LOG_ZONES, zinfo);
 				break;
 			}
 


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ