public inbox for linux-btrfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Trollkarlen Marklund <robbelibobban@gmail.com>
To: dsterba@suse.cz
Cc: linux-btrfs@vger.kernel.org
Subject: Re: [PATCH] check: check so offset is not bigger then the leaf
Date: Mon, 29 Jun 2015 14:16:28 +0300	[thread overview]
Message-ID: <3FC07E8E-F789-4C2A-9099-F649EFADCD1F@gmail.com> (raw)
In-Reply-To: <20150618164443.GH6761@twin.jikos.cz>


> On 18 Jun 2015, at 19:44, David Sterba <dsterba@suse.cz> wrote:
> 
> On Thu, Jun 18, 2015 at 01:59:13AM +0200, Robert Marklund wrote:
>> This could crash before because of dangerous dangling
>> offset of pointer.
> 
> That's right, this can happen. There are more btrfs_item_ptr that would
> be good to validate that way, namely in the checker as it's most likely
> to see corrupted data.
> 
> I think it's worth to add a wrapper macro for that, that would be like
> 
> (int) btrfs_item_ptr_validate(ei, leaf, slot, struct ..., *optional_key)
> 
> and return 0 if it's ok, 1 if there's a problem and prints the details.
> 
>> Signed-off-by: Robert Marklund <robbelibobban@gmail.com>
>> ---
>> cmds-check.c | 10 ++++++++++
>> 1 file changed, 10 insertions(+)
>> 
>> diff --git a/cmds-check.c b/cmds-check.c
>> index 778f141..da36758 100644
>> --- a/cmds-check.c
>> +++ b/cmds-check.c
>> @@ -8906,6 +8906,16 @@ static int build_roots_info_cache(struct btrfs_fs_info *info)
>> 			goto next;
>> 
>> 		ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
>> +
>> +		if ((long long)ei > info->extent_root->leafsize) {
>> +			 fprintf(stderr, "Bad leaf = %p, slot = %d\n", leaf, slot);
>> +			 fprintf(stderr, "item ptr = %p\n", ei);
>> +			 fprintf(stderr, "objectid = %llx\n", found_key.objectid);
>> +			 fprintf(stderr, "type     = %x\n", found_key.type);
>> +			 fprintf(stderr, "offset   = %llx\n", found_key.offset);
> 
> Hm, I'm not sure whether to continue or fail at this point.
> 

Im not either :)
But for me its better to keep trying until you hot the wall for real.


> Do you have a crafted filesystem image that can reproduce that or was
> that found by code inspection?

I have a failed filesystem caused by a failing disk that I tried to fix/recover.
Then i stumbled on this, and later on on some more places other then this.
Ill submit that also and in a nicer way when my filesystem is rescued. 
 

> 
>> +			 goto next;
>> +		}
>> +
>> 		flags = btrfs_extent_flags(leaf, ei);
>> 
>> 		if (found_key.type == BTRFS_EXTENT_ITEM_KEY &&


  parent reply	other threads:[~2015-06-29 11:16 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-17 23:59 [PATCH] check: check so offset is not bigger then the leaf Robert Marklund
2015-06-18 16:44 ` David Sterba
2015-06-18 17:16   ` Josef Bacik
2015-06-25 16:06     ` David Sterba
2015-06-25 16:24       ` Josef Bacik
2015-06-25 16:49         ` David Sterba
2015-06-29 11:16   ` Trollkarlen Marklund [this message]
2015-07-01 13:26     ` David Sterba

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3FC07E8E-F789-4C2A-9099-F649EFADCD1F@gmail.com \
    --to=robbelibobban@gmail.com \
    --cc=dsterba@suse.cz \
    --cc=linux-btrfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox