[PATCH v4] ext4: initialize multi-block allocator before checking block descriptors

From: Azat Khuzhin
Date: Sat Apr 05 2014 - 11:49:27 EST


With EXT4FS_DEBUG ext4_count_free_clusters() will call
ext4_read_block_bitmap() without s_group_info initialized, so we need to
initialize multi-block allocator before.

And dependencies that must be solved, to allow this:
- multi-block allocator needs in group descriptors
- need to install s_op before initializing multi-block allocator,
because in ext4_mb_init_backend() new inode is created.
- initialize number of group desc blocks (s_gdb_count) otherwise
number of clusters returned by ext4_free_clusters_after_init() is not correct.
(see ext4_bg_num_gdb_nometa())

Here is bt:
(gdb) bt
#0 ext4_get_group_info (group=0, sb=0xffff880079a10000) at ext4.h:2430
#1 ext4_validate_block_bitmap (sb=sb@entry=0xffff880079a10000, desc=desc@entry=0xffff880056510000, block_group=block_group@entry=0,
bh=bh@entry=0xffff88007bf2b2d8) at balloc.c:358
#2 0xffffffff81232202 in ext4_wait_block_bitmap (sb=sb@entry=0xffff880079a10000, block_group=block_group@entry=0,
bh=bh@entry=0xffff88007bf2b2d8) at balloc.c:476
#3 0xffffffff81232eaf in ext4_read_block_bitmap (sb=sb@entry=0xffff880079a10000, block_group=block_group@entry=0) at balloc.c:489
#4 0xffffffff81232fc0 in ext4_count_free_clusters (sb=sb@entry=0xffff880079a10000) at balloc.c:665
#5 0xffffffff81259ffa in ext4_check_descriptors (first_not_zeroed=<synthetic pointer>, sb=0xffff880079a10000) at super.c:2143
#6 ext4_fill_super (sb=sb@entry=0xffff880079a10000, data=<optimized out>, data@entry=0x0 <irq_stack_union>, silent=silent@entry=0)
at super.c:3851
#7 0xffffffff811b8340 in mount_bdev (fs_type=<optimized out>, flags=0, dev_name=<optimized out>, data=0x0 <irq_stack_union>,
fill_super=fill_super@entry=0xffffffff812589c0 <ext4_fill_super>) at super.c:987
#8 0xffffffff8124ec35 in ext4_mount (fs_type=<optimized out>, flags=<optimized out>, dev_name=<optimized out>, data=<optimized out>)
at super.c:5365
#9 0xffffffff811b8cf9 in mount_fs (type=type@entry=0xffffffff81c71840 <ext4_fs_type>, flags=flags@entry=0,
name=name@entry=0xffff880077a80c70 "/dev/loop4", data=data@entry=0x0 <irq_stack_union>) at super.c:1090
#10 0xffffffff811d2ff3 in vfs_kern_mount (type=type@entry=0xffffffff81c71840 <ext4_fs_type>, flags=0,
name=name@entry=0xffff880077a80c70 "/dev/loop4", data=data@entry=0x0 <irq_stack_union>) at namespace.c:813
#11 0xffffffff811d55de in do_new_mount (data=0x0 <irq_stack_union>, name=0xffff880077a80c70 "/dev/loop4", mnt_flags=32,
flags=<optimized out>, fstype=0xffff880077a80ca0 "ext4-insane", path=0xffff88007a5b1ed0) at namespace.c:2068
#12 do_mount (dev_name=0xffff880077a80c70 "/dev/loop4", dir_name=<optimized out>, type_page=0xffff880077a80ca0 "ext4-insane",
flags=<optimized out>, flags@entry=3236757504, data_page=0x0 <irq_stack_union>) at namespace.c:2392
#13 0xffffffff811d6183 in SYSC_mount (data=0x0 <irq_stack_union>, flags=3236757504, type=<optimized out>, dir_name=<optimized out>,
dev_name=0x7ffad9649c20 "/dev/loop4") at namespace.c:2586
#14 SyS_mount (dev_name=140715365800992, dir_name=<optimized out>, type=<optimized out>, flags=3236757504, data=0) at namespace.c:2559

Signed-off-by: Azat Khuzhin <a3at.mail@xxxxxxxxx>
---
Link: http://marc.info/?l=linux-ext4&m=139493754530149&w=2
v5: initialize number of group desc blocks (s_gdb_count) otherwise
number of clusters returned by ext4_free_clusters_after_init() is not correct.

fs/ext4/super.c | 51 +++++++++++++++++++++++++++------------------------
1 file changed, 27 insertions(+), 24 deletions(-)

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index f3c6670..6f9e6fa 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -3869,19 +3869,38 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
goto failed_mount2;
}
}
+
+ /*
+ * set up enough so that it can read an inode,
+ * and create new inode for buddy allocator
+ */
+ sbi->s_gdb_count = db_count;
+ if (!test_opt(sb, NOLOAD) &&
+ EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL))
+ sb->s_op = &ext4_sops;
+ else
+ sb->s_op = &ext4_nojournal_sops;
+
+ ext4_ext_init(sb);
+ err = ext4_mb_init(sb);
+ if (err) {
+ ext4_msg(sb, KERN_ERR, "failed to initialize mballoc (%d)",
+ err);
+ goto failed_mount2;
+ }
+
if (!ext4_check_descriptors(sb, &first_not_zeroed)) {
ext4_msg(sb, KERN_ERR, "group descriptors corrupted!");
- goto failed_mount2;
+ goto failed_mount2a;
}
if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG))
if (!ext4_fill_flex_info(sb)) {
ext4_msg(sb, KERN_ERR,
"unable to initialize "
"flex_bg meta info!");
- goto failed_mount2;
+ goto failed_mount2a;
}

- sbi->s_gdb_count = db_count;
get_random_bytes(&sbi->s_next_generation, sizeof(u32));
spin_lock_init(&sbi->s_next_gen_lock);

@@ -3916,14 +3935,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
sbi->s_stripe = ext4_get_stripe_size(sbi);
sbi->s_extent_max_zeroout_kb = 32;

- /*
- * set up enough so that it can read an inode
- */
- if (!test_opt(sb, NOLOAD) &&
- EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL))
- sb->s_op = &ext4_sops;
- else
- sb->s_op = &ext4_nojournal_sops;
sb->s_export_op = &ext4_export_ops;
sb->s_xattr = ext4_xattr_handlers;
#ifdef CONFIG_QUOTA
@@ -4113,21 +4124,13 @@ no_journal:
if (err) {
ext4_msg(sb, KERN_ERR, "failed to reserve %llu clusters for "
"reserved pool", ext4_calculate_resv_clusters(sb));
- goto failed_mount4a;
+ goto failed_mount5;
}

err = ext4_setup_system_zone(sb);
if (err) {
ext4_msg(sb, KERN_ERR, "failed to initialize system "
"zone (%d)", err);
- goto failed_mount4a;
- }
-
- ext4_ext_init(sb);
- err = ext4_mb_init(sb);
- if (err) {
- ext4_msg(sb, KERN_ERR, "failed to initialize mballoc (%d)",
- err);
goto failed_mount5;
}

@@ -4204,11 +4207,8 @@ failed_mount8:
failed_mount7:
ext4_unregister_li_request(sb);
failed_mount6:
- ext4_mb_release(sb);
-failed_mount5:
- ext4_ext_release(sb);
ext4_release_system_zone(sb);
-failed_mount4a:
+failed_mount5:
dput(sb->s_root);
sb->s_root = NULL;
failed_mount4:
@@ -4232,11 +4232,14 @@ failed_mount3:
percpu_counter_destroy(&sbi->s_extent_cache_cnt);
if (sbi->s_mmp_tsk)
kthread_stop(sbi->s_mmp_tsk);
+failed_mount2a:
+ ext4_mb_release(sb);
failed_mount2:
for (i = 0; i < db_count; i++)
brelse(sbi->s_group_desc[i]);
ext4_kvfree(sbi->s_group_desc);
failed_mount:
+ ext4_ext_release(sb);
if (sbi->s_chksum_driver)
crypto_free_shash(sbi->s_chksum_driver);
if (sbi->s_proc) {
--
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/