[bug] btrfs check --subvol-extents segfault

[bug] btrfs check --subvol-extents segfault

[Chris Murphy]

https://bugzilla.redhat.com/show_bug.cgi?id=1142421

kernel-3.17.0-0.rc5.git0.1.fc22.x86_64
btrfs-progs-3.16-1.fc21.x86_64

Steps to Reproduce:
1. mkfs.btrfs /dev/sdc
2. btrfs check --subvol-extents /dev/sdc

Results:
Segmentation fault (core dumped)

[  632.749774] btrfs[692]: segfault at 0 ip 00007fc9963fdd77 sp 00007fff93450880 error 4 in libc-2.19.90.so[7fc9963c2000+1bd000]

Eric Sandeed writes in the bugzilla that this was broken by commit d34cbe76c723a05a14bbcfdc46e8ede789c3f1f9 which drops the requirement to specify a subvolid, but still does arg_strtou64(optarg) -> arg_strtou64(NULL) so it crashes.


Chris Murphy

Re: [bug] btrfs check --subvol-extents segfault

[Eric Sandeen]

On 9/16/14 4:33 PM, Chris Murphy wrote:
> https://bugzilla.redhat.com/show_bug.cgi?id=1142421
> 
> kernel-3.17.0-0.rc5.git0.1.fc22.x86_64
> btrfs-progs-3.16-1.fc21.x86_64
> 
> Steps to Reproduce:
> 1. mkfs.btrfs /dev/sdc
> 2. btrfs check --subvol-extents /dev/sdc
> 
> Results:
> Segmentation fault (core dumped)
> 
> [  632.749774] btrfs[692]: segfault at 0 ip 00007fc9963fdd77 sp 00007fff93450880 error 4 in libc-2.19.90.so[7fc9963c2000+1bd000]
> 
> Eric Sandeed writes in the bugzilla that this was broken by commit
> d34cbe76c723a05a14bbcfdc46e8ede789c3f1f9 which drops the requirement
> to specify a subvolid, but still does arg_strtou64(optarg) ->
> arg_strtou64(NULL) so it crashes.

Yeah, AFAICT (although it's not documented anywhere) the functionality is
*supposed* to take a subvolid as an argument.

The commit lets it pass w/ no args, but still does arg_strtou64(optarg) to get
the subvolid, and boom.

I think the commit just needs to be reverted, and then manpages updated &
testcases written...

-Eric

Re: [bug] btrfs check --subvol-extents segfault

[Qu Wenruo]

Ping.

No response?

Thanks,
Qu
-------- Original Message --------
Subject: Re: [bug] btrfs check --subvol-extents segfault
From: Eric Sandeen <sandeen@redhat.com>
To: Chris Murphy <lists@colorremedies.com>, Btrfs BTRFS 
<linux-btrfs@vger.kernel.org>
Date: 2014年09月17日 05:34
> On 9/16/14 4:33 PM, Chris Murphy wrote:
>> https://bugzilla.redhat.com/show_bug.cgi?id=1142421
>>
>> kernel-3.17.0-0.rc5.git0.1.fc22.x86_64
>> btrfs-progs-3.16-1.fc21.x86_64
>>
>> Steps to Reproduce:
>> 1. mkfs.btrfs /dev/sdc
>> 2. btrfs check --subvol-extents /dev/sdc
>>
>> Results:
>> Segmentation fault (core dumped)
>>
>> [  632.749774] btrfs[692]: segfault at 0 ip 00007fc9963fdd77 sp 00007fff93450880 error 4 in libc-2.19.90.so[7fc9963c2000+1bd000]
>>
>> Eric Sandeed writes in the bugzilla that this was broken by commit
>> d34cbe76c723a05a14bbcfdc46e8ede789c3f1f9 which drops the requirement
>> to specify a subvolid, but still does arg_strtou64(optarg) ->
>> arg_strtou64(NULL) so it crashes.
> Yeah, AFAICT (although it's not documented anywhere) the functionality is
> *supposed* to take a subvolid as an argument.
>
> The commit lets it pass w/ no args, but still does arg_strtou64(optarg) to get
> the subvolid, and boom.
>
> I think the commit just needs to be reverted, and then manpages updated &
> testcases written...
>
> -Eric
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [bug] btrfs check --subvol-extents segfault

[Zygo Blaxell]

[-- Attachment #1: Type: text/plain, Size: 1998 bytes --]

I just stumbled across this bug a few hours ago.
It's still in btrfs-progs 3.17.

On Mon, Sep 29, 2014 at 11:20:06AM +0800, Qu Wenruo wrote:
> Ping.
> 
> No response?
> 
> Thanks,
> Qu
> -------- Original Message --------
> Subject: Re: [bug] btrfs check --subvol-extents segfault
> From: Eric Sandeen <sandeen@redhat.com>
> To: Chris Murphy <lists@colorremedies.com>, Btrfs BTRFS
> <linux-btrfs@vger.kernel.org>
> Date: 2014年09月17日 05:34
> >On 9/16/14 4:33 PM, Chris Murphy wrote:
> >>https://bugzilla.redhat.com/show_bug.cgi?id=1142421
> >>
> >>kernel-3.17.0-0.rc5.git0.1.fc22.x86_64
> >>btrfs-progs-3.16-1.fc21.x86_64
> >>
> >>Steps to Reproduce:
> >>1. mkfs.btrfs /dev/sdc
> >>2. btrfs check --subvol-extents /dev/sdc
> >>
> >>Results:
> >>Segmentation fault (core dumped)
> >>
> >>[  632.749774] btrfs[692]: segfault at 0 ip 00007fc9963fdd77 sp 00007fff93450880 error 4 in libc-2.19.90.so[7fc9963c2000+1bd000]
> >>
> >>Eric Sandeed writes in the bugzilla that this was broken by commit
> >>d34cbe76c723a05a14bbcfdc46e8ede789c3f1f9 which drops the requirement
> >>to specify a subvolid, but still does arg_strtou64(optarg) ->
> >>arg_strtou64(NULL) so it crashes.
> >Yeah, AFAICT (although it's not documented anywhere) the functionality is
> >*supposed* to take a subvolid as an argument.
> >
> >The commit lets it pass w/ no args, but still does arg_strtou64(optarg) to get
> >the subvolid, and boom.
> >
> >I think the commit just needs to be reverted, and then manpages updated &
> >testcases written...
> >
> >-Eric
> >--
> >To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> >the body of a message to majordomo@vger.kernel.org
> >More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]