From 4e8d2139802ce4f41936a687f06c560b12115247 Mon Sep 17 00:00:00 2001 From: Junho Ryu Date: Tue, 3 Dec 2013 18:10:28 -0500 Subject: ext4: fix use-after-free in ext4_mb_new_blocks ext4_mb_put_pa should hold pa->pa_lock before accessing pa->pa_count. While ext4_mb_use_preallocated checks pa->pa_deleted first and then increments pa->count later, ext4_mb_put_pa decrements pa->pa_count before holding pa->pa_lock and then sets pa->pa_deleted. * Free sequence ext4_mb_put_pa (1): atomic_dec_and_test pa->pa_count ext4_mb_put_pa (2): lock pa->pa_lock ext4_mb_put_pa (3): check pa->pa_deleted ext4_mb_put_pa (4): set pa->pa_deleted=1 ext4_mb_put_pa (5): unlock pa->pa_lock ext4_mb_put_pa (6): remove pa from a list ext4_mb_pa_callback: free pa * Use sequence ext4_mb_use_preallocated (1): iterate over preallocation ext4_mb_use_preallocated (2): lock pa->pa_lock ext4_mb_use_preallocated (3): check pa->pa_deleted ext4_mb_use_preallocated (4): increase pa->pa_count ext4_mb_use_preallocated (5): unlock pa->pa_lock ext4_mb_release_context: access pa * Use-after-free sequence [initial status] pa_deleted = 0, pa_count = 1> ext4_mb_use_preallocated (1): iterate over preallocation ext4_mb_use_preallocated (2): lock pa->pa_lock ext4_mb_use_preallocated (3): check pa->pa_deleted ext4_mb_put_pa (1): atomic_dec_and_test pa->pa_count [pa_count decremented] pa_deleted = 0, pa_count = 0> ext4_mb_use_preallocated (4): increase pa->pa_count [pa_count incremented] pa_deleted = 0, pa_count = 1> ext4_mb_use_preallocated (5): unlock pa->pa_lock ext4_mb_put_pa (2): lock pa->pa_lock ext4_mb_put_pa (3): check pa->pa_deleted ext4_mb_put_pa (4): set pa->pa_deleted=1 [race condition!] pa_deleted = 1, pa_count = 1> ext4_mb_put_pa (5): unlock pa->pa_lock ext4_mb_put_pa (6): remove pa from a list ext4_mb_pa_callback: free pa ext4_mb_release_context: access pa AddressSanitizer has detected use-after-free in ext4_mb_new_blocks Bug report: http://goo.gl/rG1On3 Signed-off-by: Junho Ryu Signed-off-by: "Theodore Ts'o" Cc: stable@vger.kernel.org --- fs/ext4/mballoc.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'fs/ext4/mballoc.c') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 4d113efa024..04766d9a29c 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3442,6 +3442,9 @@ static void ext4_mb_pa_callback(struct rcu_head *head) { struct ext4_prealloc_space *pa; pa = container_of(head, struct ext4_prealloc_space, u.pa_rcu); + + BUG_ON(atomic_read(&pa->pa_count)); + BUG_ON(pa->pa_deleted == 0); kmem_cache_free(ext4_pspace_cachep, pa); } @@ -3455,11 +3458,13 @@ static void ext4_mb_put_pa(struct ext4_allocation_context *ac, ext4_group_t grp; ext4_fsblk_t grp_blk; - if (!atomic_dec_and_test(&pa->pa_count) || pa->pa_free != 0) - return; - /* in this short window concurrent discard can set pa_deleted */ spin_lock(&pa->pa_lock); + if (!atomic_dec_and_test(&pa->pa_count) || pa->pa_free != 0) { + spin_unlock(&pa->pa_lock); + return; + } + if (pa->pa_deleted == 1) { spin_unlock(&pa->pa_lock); return; -- cgit v1.2.3-70-g09d2 From f5a44db5d2d677dfbf12deee461f85e9ec633961 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 20 Dec 2013 09:29:35 -0500 Subject: ext4: add explicit casts when masking cluster sizes The missing casts can cause the high 64-bits of the physical blocks to be lost. Set up new macros which allows us to make sure the right thing happen, even if at some point we end up supporting larger logical block numbers. Thanks to the Emese Revfy and the PaX security team for reporting this issue. Reported-by: PaX Team Reported-by: Emese Revfy Signed-off-by: "Theodore Ts'o" Cc: stable@vger.kernel.org --- fs/ext4/ext4.h | 10 ++++++++++ fs/ext4/extents.c | 26 ++++++++++++-------------- fs/ext4/mballoc.c | 6 +++--- 3 files changed, 25 insertions(+), 17 deletions(-) (limited to 'fs/ext4/mballoc.c') diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index e6185031c1c..ece55565b9c 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -268,6 +268,16 @@ struct ext4_io_submit { /* Translate # of blks to # of clusters */ #define EXT4_NUM_B2C(sbi, blks) (((blks) + (sbi)->s_cluster_ratio - 1) >> \ (sbi)->s_cluster_bits) +/* Mask out the low bits to get the starting block of the cluster */ +#define EXT4_PBLK_CMASK(s, pblk) ((pblk) & \ + ~((ext4_fsblk_t) (s)->s_cluster_ratio - 1)) +#define EXT4_LBLK_CMASK(s, lblk) ((lblk) & \ + ~((ext4_lblk_t) (s)->s_cluster_ratio - 1)) +/* Get the cluster offset */ +#define EXT4_PBLK_COFF(s, pblk) ((pblk) & \ + ((ext4_fsblk_t) (s)->s_cluster_ratio - 1)) +#define EXT4_LBLK_COFF(s, lblk) ((lblk) & \ + ((ext4_lblk_t) (s)->s_cluster_ratio - 1)) /* * Structure of a blocks group descriptor diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 267c9fb53bf..4410cc3d6ee 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1851,8 +1851,7 @@ static unsigned int ext4_ext_check_overlap(struct ext4_sb_info *sbi, depth = ext_depth(inode); if (!path[depth].p_ext) goto out; - b2 = le32_to_cpu(path[depth].p_ext->ee_block); - b2 &= ~(sbi->s_cluster_ratio - 1); + b2 = EXT4_LBLK_CMASK(sbi, le32_to_cpu(path[depth].p_ext->ee_block)); /* * get the next allocated block if the extent in the path @@ -1862,7 +1861,7 @@ static unsigned int ext4_ext_check_overlap(struct ext4_sb_info *sbi, b2 = ext4_ext_next_allocated_block(path); if (b2 == EXT_MAX_BLOCKS) goto out; - b2 &= ~(sbi->s_cluster_ratio - 1); + b2 = EXT4_LBLK_CMASK(sbi, b2); } /* check for wrap through zero on extent logical start block*/ @@ -2521,7 +2520,7 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, * extent, we have to mark the cluster as used (store negative * cluster number in partial_cluster). */ - unaligned = pblk & (sbi->s_cluster_ratio - 1); + unaligned = EXT4_PBLK_COFF(sbi, pblk); if (unaligned && (ee_len == num) && (*partial_cluster != -((long long)EXT4_B2C(sbi, pblk)))) *partial_cluster = EXT4_B2C(sbi, pblk); @@ -2615,7 +2614,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, * accidentally freeing it later on */ pblk = ext4_ext_pblock(ex); - if (pblk & (sbi->s_cluster_ratio - 1)) + if (EXT4_PBLK_COFF(sbi, pblk)) *partial_cluster = -((long long)EXT4_B2C(sbi, pblk)); ex--; @@ -3770,7 +3769,7 @@ int ext4_find_delalloc_cluster(struct inode *inode, ext4_lblk_t lblk) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); ext4_lblk_t lblk_start, lblk_end; - lblk_start = lblk & (~(sbi->s_cluster_ratio - 1)); + lblk_start = EXT4_LBLK_CMASK(sbi, lblk); lblk_end = lblk_start + sbi->s_cluster_ratio - 1; return ext4_find_delalloc_range(inode, lblk_start, lblk_end); @@ -3829,9 +3828,9 @@ get_reserved_cluster_alloc(struct inode *inode, ext4_lblk_t lblk_start, trace_ext4_get_reserved_cluster_alloc(inode, lblk_start, num_blks); /* Check towards left side */ - c_offset = lblk_start & (sbi->s_cluster_ratio - 1); + c_offset = EXT4_LBLK_COFF(sbi, lblk_start); if (c_offset) { - lblk_from = lblk_start & (~(sbi->s_cluster_ratio - 1)); + lblk_from = EXT4_LBLK_CMASK(sbi, lblk_start); lblk_to = lblk_from + c_offset - 1; if (ext4_find_delalloc_range(inode, lblk_from, lblk_to)) @@ -3839,7 +3838,7 @@ get_reserved_cluster_alloc(struct inode *inode, ext4_lblk_t lblk_start, } /* Now check towards right. */ - c_offset = (lblk_start + num_blks) & (sbi->s_cluster_ratio - 1); + c_offset = EXT4_LBLK_COFF(sbi, lblk_start + num_blks); if (allocated_clusters && c_offset) { lblk_from = lblk_start + num_blks; lblk_to = lblk_from + (sbi->s_cluster_ratio - c_offset) - 1; @@ -4047,7 +4046,7 @@ static int get_implied_cluster_alloc(struct super_block *sb, struct ext4_ext_path *path) { struct ext4_sb_info *sbi = EXT4_SB(sb); - ext4_lblk_t c_offset = map->m_lblk & (sbi->s_cluster_ratio-1); + ext4_lblk_t c_offset = EXT4_LBLK_COFF(sbi, map->m_lblk); ext4_lblk_t ex_cluster_start, ex_cluster_end; ext4_lblk_t rr_cluster_start; ext4_lblk_t ee_block = le32_to_cpu(ex->ee_block); @@ -4065,8 +4064,7 @@ static int get_implied_cluster_alloc(struct super_block *sb, (rr_cluster_start == ex_cluster_start)) { if (rr_cluster_start == ex_cluster_end) ee_start += ee_len - 1; - map->m_pblk = (ee_start & ~(sbi->s_cluster_ratio - 1)) + - c_offset; + map->m_pblk = EXT4_PBLK_CMASK(sbi, ee_start) + c_offset; map->m_len = min(map->m_len, (unsigned) sbi->s_cluster_ratio - c_offset); /* @@ -4220,7 +4218,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, */ map->m_flags &= ~EXT4_MAP_FROM_CLUSTER; newex.ee_block = cpu_to_le32(map->m_lblk); - cluster_offset = map->m_lblk & (sbi->s_cluster_ratio-1); + cluster_offset = EXT4_LBLK_CMASK(sbi, map->m_lblk); /* * If we are doing bigalloc, check to see if the extent returned @@ -4288,7 +4286,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, * needed so that future calls to get_implied_cluster_alloc() * work correctly. */ - offset = map->m_lblk & (sbi->s_cluster_ratio - 1); + offset = EXT4_LBLK_COFF(sbi, map->m_lblk); ar.len = EXT4_NUM_B2C(sbi, offset+allocated); ar.goal -= offset; ar.logical -= offset; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 04766d9a29c..04a5c7504be 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -4126,7 +4126,7 @@ ext4_mb_initialize_context(struct ext4_allocation_context *ac, ext4_get_group_no_and_offset(sb, goal, &group, &block); /* set up allocation goals */ - ac->ac_b_ex.fe_logical = ar->logical & ~(sbi->s_cluster_ratio - 1); + ac->ac_b_ex.fe_logical = EXT4_LBLK_CMASK(sbi, ar->logical); ac->ac_status = AC_STATUS_CONTINUE; ac->ac_sb = sb; ac->ac_inode = ar->inode; @@ -4668,7 +4668,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode, * blocks at the beginning or the end unless we are explicitly * requested to avoid doing so. */ - overflow = block & (sbi->s_cluster_ratio - 1); + overflow = EXT4_PBLK_COFF(sbi, block); if (overflow) { if (flags & EXT4_FREE_BLOCKS_NOFREE_FIRST_CLUSTER) { overflow = sbi->s_cluster_ratio - overflow; @@ -4682,7 +4682,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode, count += overflow; } } - overflow = count & (sbi->s_cluster_ratio - 1); + overflow = EXT4_LBLK_COFF(sbi, count); if (overflow) { if (flags & EXT4_FREE_BLOCKS_NOFREE_LAST_CLUSTER) { if (count > overflow) -- cgit v1.2.3-70-g09d2