BTRFS warning (device dm-0): unhandled fiemap cache detected

BTRFS warning (device dm-0): unhandled fiemap cache detected

[Christoph Anton Mitterer]

[-- Attachment #1: Type: text/plain, Size: 846 bytes --]

Hey.

Just got the following with 4.12.6:

Aug 21 03:29:51 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 flags=0x0
Aug 21 03:29:56 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 flags=0x0
Aug 21 03:30:58 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=813760614400 len=32768 flags=0x0
Aug 21 03:31:15 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 flags=0x0

Is it what should be fixed with:
https://patchwork.kernel.org/patch/9803291/

?

Is this harmless or must I assume that some part of my data/fs is now
corrupt and should I recover from backup?

Thanks,
Chris.

[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 5930 bytes --]

Re: BTRFS warning (device dm-0): unhandled fiemap cache detected

[Qu Wenruo]

On 2017年08月21日 10:29, Christoph Anton Mitterer wrote:
> Hey.
> 
> Just got the following with 4.12.6:
> 
> Aug 21 03:29:51 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 flags=0x0
> Aug 21 03:29:56 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 flags=0x0
> Aug 21 03:30:58 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=813760614400 len=32768 flags=0x0
> Aug 21 03:31:15 heisenberg kernel: BTRFS warning (device dm-0): unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 flags=0x0

This is just warning about there is some fiemap extent cached but not 
submitted.
So in fact harmless, only mean to info developers.
> 
> Is it what should be fixed with:
> https://patchwork.kernel.org/patch/9803291/
> 
> ?

But it is already in v4.12, so there is some other case I missed.

Any reproducer?

> 
> Is this harmless or must I assume that some part of my data/fs is now
> corrupt and should I recover from backup?

Harmless, it is only designed to merge fiemap output.
No data is touched at all.

Sorry for the trouble.

Thanks,
Qu

> 
> Thanks,
> Chris.
> 

Re: BTRFS warning (device dm-0): unhandled fiemap cache detected

[Qu Wenruo]

On 2017年08月21日 10:43, Qu Wenruo wrote:
> 
> 
> On 2017年08月21日 10:29, Christoph Anton Mitterer wrote:
>> Hey.
>>
>> Just got the following with 4.12.6:
>>
>> Aug 21 03:29:51 heisenberg kernel: BTRFS warning (device dm-0): 
>> unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 
>> flags=0x0
>> Aug 21 03:29:56 heisenberg kernel: BTRFS warning (device dm-0): 
>> unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 
>> flags=0x0
>> Aug 21 03:30:58 heisenberg kernel: BTRFS warning (device dm-0): 
>> unhandled fiemap cache detected: offset=0 phys=813760614400 len=32768 
>> flags=0x0
>> Aug 21 03:31:15 heisenberg kernel: BTRFS warning (device dm-0): 
>> unhandled fiemap cache detected: offset=0 phys=812641906688 len=12288 
>> flags=0x0
> 
> This is just warning about there is some fiemap extent cached but not 
> submitted.
> So in fact harmless, only mean to info developers.
>>
>> Is it what should be fixed with:
>> https://patchwork.kernel.org/patch/9803291/
>>
>> ?
> 
> But it is already in v4.12, so there is some other case I missed.
> 
> Any reproducer?

OK, I checked the v4.12.6 code, and it doesn't have the fix.

While v4.12.8 (current Archlinux official repo linux package) has the fix.
If you're using Arch, updating linux will solve the problem.

Quite strange, according to upstream git log, that commit is merged 
between v4.12-rc7 and v4.12.
Maybe I misunderstand the stable kernel release cycle.

Thanks,
Qu

> 
>>
>> Is this harmless or must I assume that some part of my data/fs is now
>> corrupt and should I recover from backup?
> 
> Harmless, it is only designed to merge fiemap output.
> No data is touched at all.
> 
> Sorry for the trouble.
> 
> Thanks,
> Qu
> 
>>
>> Thanks,
>> Chris.
>>
> -- 
> 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: BTRFS warning (device dm-0): unhandled fiemap cache detected

[Christoph Anton Mitterer]

[-- Attachment #1: Type: text/plain, Size: 423 bytes --]

On Mon, 2017-08-21 at 10:43 +0800, Qu Wenruo wrote:
> Harmless, it is only designed to merge fiemap output.
Thanks for the info :)


On Mon, 2017-08-21 at 10:57 +0800, Qu Wenruo wrote:
> Quite strange, according to upstream git log, that commit is merged 
> between v4.12-rc7 and v4.12.
> Maybe I misunderstand the stable kernel release cycle.

Seems it was only added in 4.12.7? Maybe a typo?

Cheers,
Chris.

[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 5930 bytes --]

Re: BTRFS warning (device dm-0): unhandled fiemap cache detected

[Qu Wenruo]

On 2017年08月21日 11:27, Christoph Anton Mitterer wrote:
> On Mon, 2017-08-21 at 10:43 +0800, Qu Wenruo wrote:
>> Harmless, it is only designed to merge fiemap output.
> Thanks for the info :)
> 
> 
> On Mon, 2017-08-21 at 10:57 +0800, Qu Wenruo wrote:
>> Quite strange, according to upstream git log, that commit is merged
>> between v4.12-rc7 and v4.12.
>> Maybe I misunderstand the stable kernel release cycle.
> 
> Seems it was only added in 4.12.7? Maybe a typo?

I'm talking about upstream, which is managed by Linus.
The 4.12.x is managed by maintainers of stable releases.

In Linus repo, the commit is between v4.12 and v4.12-rc7 tag.
Maybe v4.12.x is started before v4.12 release.

Thanks,
Qu

> 
> Cheers,
> Chris.
> 

Re: BTRFS warning (device dm-0): unhandled fiemap cache detected

[Omar Sandoval]

On Mon, Aug 21, 2017 at 11:31:34AM +0800, Qu Wenruo wrote:
> 
> 
> On 2017年08月21日 11:27, Christoph Anton Mitterer wrote:
> > On Mon, 2017-08-21 at 10:43 +0800, Qu Wenruo wrote:
> > > Harmless, it is only designed to merge fiemap output.
> > Thanks for the info :)
> > 
> > 
> > On Mon, 2017-08-21 at 10:57 +0800, Qu Wenruo wrote:
> > > Quite strange, according to upstream git log, that commit is merged
> > > between v4.12-rc7 and v4.12.
> > > Maybe I misunderstand the stable kernel release cycle.
> > 
> > Seems it was only added in 4.12.7? Maybe a typo?
> 
> I'm talking about upstream, which is managed by Linus.
> The 4.12.x is managed by maintainers of stable releases.
> 
> In Linus repo, the commit is between v4.12 and v4.12-rc7 tag.
> Maybe v4.12.x is started before v4.12 release.

It's not in 4.12.

$ git tag --contains 848c23b78fafdcd3270b06a30737f8dbd70c347f
for-4.13-part2-tag
v4.13-rc1
v4.13-rc2
v4.13-rc3
v4.13-rc4
v4.13-rc5