diff mbox series

[5/7] ext4: fix slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists()

Message ID 20240126085716.1363019-6-libaokun1@huawei.com
State Superseded
Headers show
Series ext4: avoid sysfs variables overflow causing BUG_ON/SOOB | expand

Commit Message

Baokun Li Jan. 26, 2024, 8:57 a.m. UTC 
We can trigger a slab-out-of-bounds with the following commands:

    mkfs.ext4 -F /dev/$disk 10G
    mount /dev/$disk /tmp/test
    echo 2147483647 > /sys/fs/ext4/$disk/mb_group_prealloc
    echo test > /tmp/test/file && sync

==================================================================
BUG: KASAN: slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
Read of size 8 at addr ffff888121b9d0f0 by task kworker/u2:0/11
CPU: 0 PID: 11 Comm: kworker/u2:0 Tainted: GL 6.7.0-next-20240118 #521
Call Trace:
 dump_stack_lvl+0x2c/0x50
 kasan_report+0xb6/0xf0
 ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
 ext4_mb_regular_allocator+0x19e9/0x2370 [ext4]
 ext4_mb_new_blocks+0x88a/0x1370 [ext4]
 ext4_ext_map_blocks+0x14f7/0x2390 [ext4]
 ext4_map_blocks+0x569/0xea0 [ext4]
 ext4_do_writepages+0x10f6/0x1bc0 [ext4]
[...]
==================================================================

The flow of issue triggering is as follows:

// Set s_mb_group_prealloc to 2147483647 via sysfs
ext4_mb_new_blocks
  ext4_mb_normalize_request
    ext4_mb_normalize_group_request
      ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc
  ext4_mb_regular_allocator
    ext4_mb_choose_next_group
      ext4_mb_choose_next_group_best_avail
        mb_avg_fragment_size_order
          order = fls(len) - 2 = 29
        ext4_mb_find_good_group_avg_frag_lists
          frag_list = &sbi->s_mb_avg_fragment_size[order]
          if (list_empty(frag_list)) // Trigger SOOB!

At 4k block size, the length of the s_mb_avg_fragment_size list is 14, but
an oversized s_mb_group_prealloc is set, causing slab-out-of-bounds to be
triggered by an attempt to access an element at index 29.

Therefore it is not allowed to set s_mb_group_prealloc to a value greater
than s_clusters_per_group via sysfs, and to avoid returning an order from
mb_avg_fragment_size_order() that is greater than MB_NUM_ORDERS(sb).

Fixes: 7e170922f06b ("ext4: Add allocation criteria 1.5 (CR1_5)")
CC: stable@vger.kernel.org
Signed-off-by: Baokun Li <libaokun1@huawei.com>
---
 fs/ext4/mballoc.c | 2 ++
 fs/ext4/sysfs.c   | 9 ++++++++-
 2 files changed, 10 insertions(+), 1 deletion(-)

Comments

Zhang Yi Jan. 27, 2024, 2:09 a.m. UTC | #1 
On 2024/1/26 16:57, Baokun Li wrote:
> We can trigger a slab-out-of-bounds with the following commands:
> 
>     mkfs.ext4 -F /dev/$disk 10G
>     mount /dev/$disk /tmp/test
>     echo 2147483647 > /sys/fs/ext4/$disk/mb_group_prealloc
>     echo test > /tmp/test/file && sync
> 
> ==================================================================
> BUG: KASAN: slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
> Read of size 8 at addr ffff888121b9d0f0 by task kworker/u2:0/11
> CPU: 0 PID: 11 Comm: kworker/u2:0 Tainted: GL 6.7.0-next-20240118 #521
> Call Trace:
>  dump_stack_lvl+0x2c/0x50
>  kasan_report+0xb6/0xf0
>  ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
>  ext4_mb_regular_allocator+0x19e9/0x2370 [ext4]
>  ext4_mb_new_blocks+0x88a/0x1370 [ext4]
>  ext4_ext_map_blocks+0x14f7/0x2390 [ext4]
>  ext4_map_blocks+0x569/0xea0 [ext4]
>  ext4_do_writepages+0x10f6/0x1bc0 [ext4]
> [...]
> ==================================================================
> 
> The flow of issue triggering is as follows:
> 
> // Set s_mb_group_prealloc to 2147483647 via sysfs
> ext4_mb_new_blocks
>   ext4_mb_normalize_request
>     ext4_mb_normalize_group_request
>       ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc
>   ext4_mb_regular_allocator
>     ext4_mb_choose_next_group
>       ext4_mb_choose_next_group_best_avail
>         mb_avg_fragment_size_order
>           order = fls(len) - 2 = 29
>         ext4_mb_find_good_group_avg_frag_lists
>           frag_list = &sbi->s_mb_avg_fragment_size[order]
>           if (list_empty(frag_list)) // Trigger SOOB!
> 
> At 4k block size, the length of the s_mb_avg_fragment_size list is 14, but
> an oversized s_mb_group_prealloc is set, causing slab-out-of-bounds to be
> triggered by an attempt to access an element at index 29.
> 
> Therefore it is not allowed to set s_mb_group_prealloc to a value greater
> than s_clusters_per_group via sysfs, and to avoid returning an order from
> mb_avg_fragment_size_order() that is greater than MB_NUM_ORDERS(sb).
> 
> Fixes: 7e170922f06b ("ext4: Add allocation criteria 1.5 (CR1_5)")
> CC: stable@vger.kernel.org
> Signed-off-by: Baokun Li <libaokun1@huawei.com>

Looks good to me.

Reviewed-by: Zhang Yi <yi.zhang@huawei.com>

> ---
>  fs/ext4/mballoc.c | 2 ++
>  fs/ext4/sysfs.c   | 9 ++++++++-
>  2 files changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index f44f668e407f..1ea6491b6b00 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -832,6 +832,8 @@ static int mb_avg_fragment_size_order(struct super_block *sb, ext4_grpblk_t len)
>  		return 0;
>  	if (order == MB_NUM_ORDERS(sb))
>  		order--;
> +	if (WARN_ON_ONCE(order > MB_NUM_ORDERS(sb)))
> +		order = MB_NUM_ORDERS(sb) - 1;
>  	return order;
>  }
>  
> diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c
> index 6f9f96e00f2f..60ca7b2797b2 100644
> --- a/fs/ext4/sysfs.c
> +++ b/fs/ext4/sysfs.c
> @@ -29,6 +29,7 @@ typedef enum {
>  	attr_trigger_test_error,
>  	attr_first_error_time,
>  	attr_last_error_time,
> +	attr_group_prealloc,
>  	attr_feature,
>  	attr_pointer_pi,
>  	attr_pointer_ui,
> @@ -211,13 +212,14 @@ EXT4_ATTR_FUNC(sra_exceeded_retry_limit, 0444);
>  
>  EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, inode_readahead,
>  		 ext4_sb_info, s_inode_readahead_blks);
> +EXT4_ATTR_OFFSET(mb_group_prealloc, 0644, group_prealloc,
> +		 ext4_sb_info, s_mb_group_prealloc);
>  EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal);
>  EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats);
>  EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan);
>  EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan);
>  EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs);
>  EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request);
> -EXT4_RW_ATTR_SBI_PI(mb_group_prealloc, s_mb_group_prealloc);
>  EXT4_RW_ATTR_SBI_UI(mb_max_linear_groups, s_mb_max_linear_groups);
>  EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb);
>  EXT4_ATTR(trigger_fs_error, 0200, trigger_test_error);
> @@ -380,6 +382,7 @@ static ssize_t ext4_generic_attr_show(struct ext4_attr *a,
>  
>  	switch (a->attr_id) {
>  	case attr_inode_readahead:
> +	case attr_group_prealloc:
>  	case attr_pointer_pi:
>  	case attr_pointer_ui:
>  		if (a->attr_ptr == ptr_ext4_super_block_offset)
> @@ -453,6 +456,10 @@ static ssize_t ext4_generic_attr_store(struct ext4_attr *a,
>  		return ret;
>  
>  	switch (a->attr_id) {
> +	case attr_group_prealloc:
> +		if (t > sbi->s_clusters_per_group)
> +			return -EINVAL;
> +		fallthrough;
>  	case attr_pointer_pi:
>  		if ((int)t < 0)
>  			return -EINVAL;
>
Jan Kara Feb. 13, 2024, 4:14 p.m. UTC | #2 
On Fri 26-01-24 16:57:14, Baokun Li wrote:
> We can trigger a slab-out-of-bounds with the following commands:
> 
>     mkfs.ext4 -F /dev/$disk 10G
>     mount /dev/$disk /tmp/test
>     echo 2147483647 > /sys/fs/ext4/$disk/mb_group_prealloc
>     echo test > /tmp/test/file && sync
> 
> ==================================================================
> BUG: KASAN: slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
> Read of size 8 at addr ffff888121b9d0f0 by task kworker/u2:0/11
> CPU: 0 PID: 11 Comm: kworker/u2:0 Tainted: GL 6.7.0-next-20240118 #521
> Call Trace:
>  dump_stack_lvl+0x2c/0x50
>  kasan_report+0xb6/0xf0
>  ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
>  ext4_mb_regular_allocator+0x19e9/0x2370 [ext4]
>  ext4_mb_new_blocks+0x88a/0x1370 [ext4]
>  ext4_ext_map_blocks+0x14f7/0x2390 [ext4]
>  ext4_map_blocks+0x569/0xea0 [ext4]
>  ext4_do_writepages+0x10f6/0x1bc0 [ext4]
> [...]
> ==================================================================
> 
> The flow of issue triggering is as follows:
> 
> // Set s_mb_group_prealloc to 2147483647 via sysfs
> ext4_mb_new_blocks
>   ext4_mb_normalize_request
>     ext4_mb_normalize_group_request
>       ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc
>   ext4_mb_regular_allocator
>     ext4_mb_choose_next_group
>       ext4_mb_choose_next_group_best_avail
>         mb_avg_fragment_size_order
>           order = fls(len) - 2 = 29
>         ext4_mb_find_good_group_avg_frag_lists
>           frag_list = &sbi->s_mb_avg_fragment_size[order]
>           if (list_empty(frag_list)) // Trigger SOOB!
> 
> At 4k block size, the length of the s_mb_avg_fragment_size list is 14, but
> an oversized s_mb_group_prealloc is set, causing slab-out-of-bounds to be
> triggered by an attempt to access an element at index 29.
> 
> Therefore it is not allowed to set s_mb_group_prealloc to a value greater
> than s_clusters_per_group via sysfs, and to avoid returning an order from
> mb_avg_fragment_size_order() that is greater than MB_NUM_ORDERS(sb).
> 
> Fixes: 7e170922f06b ("ext4: Add allocation criteria 1.5 (CR1_5)")
> CC: stable@vger.kernel.org
> Signed-off-by: Baokun Li <libaokun1@huawei.com>

Looks good. Feel free to add:

Reviewed-by: Jan Kara <jack@suse.cz>

								Honza


> ---
>  fs/ext4/mballoc.c | 2 ++
>  fs/ext4/sysfs.c   | 9 ++++++++-
>  2 files changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index f44f668e407f..1ea6491b6b00 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -832,6 +832,8 @@ static int mb_avg_fragment_size_order(struct super_block *sb, ext4_grpblk_t len)
>  		return 0;
>  	if (order == MB_NUM_ORDERS(sb))
>  		order--;
> +	if (WARN_ON_ONCE(order > MB_NUM_ORDERS(sb)))
> +		order = MB_NUM_ORDERS(sb) - 1;
>  	return order;
>  }
>  
> diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c
> index 6f9f96e00f2f..60ca7b2797b2 100644
> --- a/fs/ext4/sysfs.c
> +++ b/fs/ext4/sysfs.c
> @@ -29,6 +29,7 @@ typedef enum {
>  	attr_trigger_test_error,
>  	attr_first_error_time,
>  	attr_last_error_time,
> +	attr_group_prealloc,
>  	attr_feature,
>  	attr_pointer_pi,
>  	attr_pointer_ui,
> @@ -211,13 +212,14 @@ EXT4_ATTR_FUNC(sra_exceeded_retry_limit, 0444);
>  
>  EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, inode_readahead,
>  		 ext4_sb_info, s_inode_readahead_blks);
> +EXT4_ATTR_OFFSET(mb_group_prealloc, 0644, group_prealloc,
> +		 ext4_sb_info, s_mb_group_prealloc);
>  EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal);
>  EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats);
>  EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan);
>  EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan);
>  EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs);
>  EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request);
> -EXT4_RW_ATTR_SBI_PI(mb_group_prealloc, s_mb_group_prealloc);
>  EXT4_RW_ATTR_SBI_UI(mb_max_linear_groups, s_mb_max_linear_groups);
>  EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb);
>  EXT4_ATTR(trigger_fs_error, 0200, trigger_test_error);
> @@ -380,6 +382,7 @@ static ssize_t ext4_generic_attr_show(struct ext4_attr *a,
>  
>  	switch (a->attr_id) {
>  	case attr_inode_readahead:
> +	case attr_group_prealloc:
>  	case attr_pointer_pi:
>  	case attr_pointer_ui:
>  		if (a->attr_ptr == ptr_ext4_super_block_offset)
> @@ -453,6 +456,10 @@ static ssize_t ext4_generic_attr_store(struct ext4_attr *a,
>  		return ret;
>  
>  	switch (a->attr_id) {
> +	case attr_group_prealloc:
> +		if (t > sbi->s_clusters_per_group)
> +			return -EINVAL;
> +		fallthrough;
>  	case attr_pointer_pi:
>  		if ((int)t < 0)
>  			return -EINVAL;
> -- 
> 2.31.1
>
Ojaswin Mujoo Feb. 20, 2024, 5:39 a.m. UTC | #3 
On Fri, Jan 26, 2024 at 04:57:14PM +0800, Baokun Li wrote:

Hey Baokun, 

Good catch! I've added some minor comments below. Other than that feel
free to add 

Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>

> We can trigger a slab-out-of-bounds with the following commands:
> 
>     mkfs.ext4 -F /dev/$disk 10G
>     mount /dev/$disk /tmp/test
>     echo 2147483647 > /sys/fs/ext4/$disk/mb_group_prealloc
>     echo test > /tmp/test/file && sync
> 
> ==================================================================
> BUG: KASAN: slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
> Read of size 8 at addr ffff888121b9d0f0 by task kworker/u2:0/11
> CPU: 0 PID: 11 Comm: kworker/u2:0 Tainted: GL 6.7.0-next-20240118 #521
> Call Trace:
>  dump_stack_lvl+0x2c/0x50
>  kasan_report+0xb6/0xf0
>  ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
>  ext4_mb_regular_allocator+0x19e9/0x2370 [ext4]
>  ext4_mb_new_blocks+0x88a/0x1370 [ext4]
>  ext4_ext_map_blocks+0x14f7/0x2390 [ext4]
>  ext4_map_blocks+0x569/0xea0 [ext4]
>  ext4_do_writepages+0x10f6/0x1bc0 [ext4]
> [...]
> ==================================================================
> 
> The flow of issue triggering is as follows:
> 
> // Set s_mb_group_prealloc to 2147483647 via sysfs
> ext4_mb_new_blocks
>   ext4_mb_normalize_request
>     ext4_mb_normalize_group_request
>       ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc
>   ext4_mb_regular_allocator
>     ext4_mb_choose_next_group
>       ext4_mb_choose_next_group_best_avail
>         mb_avg_fragment_size_order
>           order = fls(len) - 2 = 29
>         ext4_mb_find_good_group_avg_frag_lists
>           frag_list = &sbi->s_mb_avg_fragment_size[order]
>           if (list_empty(frag_list)) // Trigger SOOB!
> 
> At 4k block size, the length of the s_mb_avg_fragment_size list is 14, but
> an oversized s_mb_group_prealloc is set, causing slab-out-of-bounds to be
> triggered by an attempt to access an element at index 29.
> 
> Therefore it is not allowed to set s_mb_group_prealloc to a value greater
> than s_clusters_per_group via sysfs, and to avoid returning an order from
> mb_avg_fragment_size_order() that is greater than MB_NUM_ORDERS(sb).
> 
> Fixes: 7e170922f06b ("ext4: Add allocation criteria 1.5 (CR1_5)")
> CC: stable@vger.kernel.org
> Signed-off-by: Baokun Li <libaokun1@huawei.com>
> ---
>  fs/ext4/mballoc.c | 2 ++
>  fs/ext4/sysfs.c   | 9 ++++++++-
>  2 files changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
> index f44f668e407f..1ea6491b6b00 100644
> --- a/fs/ext4/mballoc.c
> +++ b/fs/ext4/mballoc.c
> @@ -832,6 +832,8 @@ static int mb_avg_fragment_size_order(struct super_block *sb, ext4_grpblk_t len)
>     return 0;
>   if (order == MB_NUM_ORDERS(sb))
>     order--;
> + if (WARN_ON_ONCE(order > MB_NUM_ORDERS(sb)))
> +   order = MB_NUM_ORDERS(sb) - 1;
>   return order;
>  }

So along with this change, I think it'll also be good to add an extra 
check in ext4_mb_choose_next_group_best_avail() as:

  if (1 << min_order < ac->ac_o_ex.fe_len)
    min_order = fls(ac->ac_o_ex.fe_len);
 
+ if (order >= MB_NUM_ORDERS(ac->ac_sb))
+   order = MB_NUM_ORDERS(ac->ac_sb) - 1;
+
  for (i = order; i >= min_order; i--) {
    int frag_order;
    /*


The reason for this is that otherwise when order is large eg 29,
we would unnecessarily loop from i=29 to i=13 while always
looking at the same avg_fragment_list[13].

Regards,
ojaswin
Baokun Li Feb. 20, 2024, 6:31 a.m. UTC | #4 
On 2024/2/20 13:39, Ojaswin Mujoo wrote:
> On Fri, Jan 26, 2024 at 04:57:14PM +0800, Baokun Li wrote:
>
> Hey Baokun,
>
> Good catch! I've added some minor comments below. Other than that feel
> free to add
>
> Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
>
>> We can trigger a slab-out-of-bounds with the following commands:
>>
>>      mkfs.ext4 -F /dev/$disk 10G
>>      mount /dev/$disk /tmp/test
>>      echo 2147483647 > /sys/fs/ext4/$disk/mb_group_prealloc
>>      echo test > /tmp/test/file && sync
>>
>> ==================================================================
>> BUG: KASAN: slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
>> Read of size 8 at addr ffff888121b9d0f0 by task kworker/u2:0/11
>> CPU: 0 PID: 11 Comm: kworker/u2:0 Tainted: GL 6.7.0-next-20240118 #521
>> Call Trace:
>>   dump_stack_lvl+0x2c/0x50
>>   kasan_report+0xb6/0xf0
>>   ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4]
>>   ext4_mb_regular_allocator+0x19e9/0x2370 [ext4]
>>   ext4_mb_new_blocks+0x88a/0x1370 [ext4]
>>   ext4_ext_map_blocks+0x14f7/0x2390 [ext4]
>>   ext4_map_blocks+0x569/0xea0 [ext4]
>>   ext4_do_writepages+0x10f6/0x1bc0 [ext4]
>> [...]
>> ==================================================================
>>
>> The flow of issue triggering is as follows:
>>
>> // Set s_mb_group_prealloc to 2147483647 via sysfs
>> ext4_mb_new_blocks
>>    ext4_mb_normalize_request
>>      ext4_mb_normalize_group_request
>>        ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc
>>    ext4_mb_regular_allocator
>>      ext4_mb_choose_next_group
>>        ext4_mb_choose_next_group_best_avail
>>          mb_avg_fragment_size_order
>>            order = fls(len) - 2 = 29
>>          ext4_mb_find_good_group_avg_frag_lists
>>            frag_list = &sbi->s_mb_avg_fragment_size[order]
>>            if (list_empty(frag_list)) // Trigger SOOB!
>>
>> At 4k block size, the length of the s_mb_avg_fragment_size list is 14, but
>> an oversized s_mb_group_prealloc is set, causing slab-out-of-bounds to be
>> triggered by an attempt to access an element at index 29.
>>
>> Therefore it is not allowed to set s_mb_group_prealloc to a value greater
>> than s_clusters_per_group via sysfs, and to avoid returning an order from
>> mb_avg_fragment_size_order() that is greater than MB_NUM_ORDERS(sb).
>>
>> Fixes: 7e170922f06b ("ext4: Add allocation criteria 1.5 (CR1_5)")
>> CC: stable@vger.kernel.org
>> Signed-off-by: Baokun Li <libaokun1@huawei.com>
>> ---
>>   fs/ext4/mballoc.c | 2 ++
>>   fs/ext4/sysfs.c   | 9 ++++++++-
>>   2 files changed, 10 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
>> index f44f668e407f..1ea6491b6b00 100644
>> --- a/fs/ext4/mballoc.c
>> +++ b/fs/ext4/mballoc.c
>> @@ -832,6 +832,8 @@ static int mb_avg_fragment_size_order(struct super_block *sb, ext4_grpblk_t len)
>>      return 0;
>>    if (order == MB_NUM_ORDERS(sb))
>>      order--;
>> + if (WARN_ON_ONCE(order > MB_NUM_ORDERS(sb)))
>> +   order = MB_NUM_ORDERS(sb) - 1;
>>    return order;
>>   }
> So along with this change, I think it'll also be good to add an extra
> check in ext4_mb_choose_next_group_best_avail() as:
>
>    if (1 << min_order < ac->ac_o_ex.fe_len)
>      min_order = fls(ac->ac_o_ex.fe_len);
>   
> + if (order >= MB_NUM_ORDERS(ac->ac_sb))
> +   order = MB_NUM_ORDERS(ac->ac_sb) - 1;
> +
>    for (i = order; i >= min_order; i--) {
>      int frag_order;
>      /*
>
>
> The reason for this is that otherwise when order is large eg 29,
> we would unnecessarily loop from i=29 to i=13 while always
> looking at the same avg_fragment_list[13].
>
> Regards,
> ojaswin


Yeah, good point! This will cut down on some unnecessary loops.

I'll add this extra check in the next version.

Thanks for having a look!
diff mbox series

Patch

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index f44f668e407f..1ea6491b6b00 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -832,6 +832,8 @@  static int mb_avg_fragment_size_order(struct super_block *sb, ext4_grpblk_t len)
 		return 0;
 	if (order == MB_NUM_ORDERS(sb))
 		order--;
+	if (WARN_ON_ONCE(order > MB_NUM_ORDERS(sb)))
+		order = MB_NUM_ORDERS(sb) - 1;
 	return order;
 }
 
diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c
index 6f9f96e00f2f..60ca7b2797b2 100644
--- a/fs/ext4/sysfs.c
+++ b/fs/ext4/sysfs.c
@@ -29,6 +29,7 @@  typedef enum {
 	attr_trigger_test_error,
 	attr_first_error_time,
 	attr_last_error_time,
+	attr_group_prealloc,
 	attr_feature,
 	attr_pointer_pi,
 	attr_pointer_ui,
@@ -211,13 +212,14 @@  EXT4_ATTR_FUNC(sra_exceeded_retry_limit, 0444);
 
 EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, inode_readahead,
 		 ext4_sb_info, s_inode_readahead_blks);
+EXT4_ATTR_OFFSET(mb_group_prealloc, 0644, group_prealloc,
+		 ext4_sb_info, s_mb_group_prealloc);
 EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal);
 EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats);
 EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan);
 EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan);
 EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs);
 EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request);
-EXT4_RW_ATTR_SBI_PI(mb_group_prealloc, s_mb_group_prealloc);
 EXT4_RW_ATTR_SBI_UI(mb_max_linear_groups, s_mb_max_linear_groups);
 EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb);
 EXT4_ATTR(trigger_fs_error, 0200, trigger_test_error);
@@ -380,6 +382,7 @@  static ssize_t ext4_generic_attr_show(struct ext4_attr *a,
 
 	switch (a->attr_id) {
 	case attr_inode_readahead:
+	case attr_group_prealloc:
 	case attr_pointer_pi:
 	case attr_pointer_ui:
 		if (a->attr_ptr == ptr_ext4_super_block_offset)
@@ -453,6 +456,10 @@  static ssize_t ext4_generic_attr_store(struct ext4_attr *a,
 		return ret;
 
 	switch (a->attr_id) {
+	case attr_group_prealloc:
+		if (t > sbi->s_clusters_per_group)
+			return -EINVAL;
+		fallthrough;
 	case attr_pointer_pi:
 		if ((int)t < 0)
 			return -EINVAL;