Re: [PATCH v4] btrfs: trim: fix underflow in trim length to prevent access beyond device boundary

[Qu Wenruo <quwenruo.btrfs@gmx.com>]

On 2020/8/12 下午2:10, David Sterba wrote:
> On Tue, Aug 11, 2020 at 03:42:31PM +0800, Qu Wenruo wrote:
>>>> CHUNK_TRIMMED and CHUNK_ALLOCATED.
>>>>
>>>> Thus what we're doing is to clear all utilized bits.
>>>
>>> Which is true for now, adding a new bit would change that.
>>
>> Yep, that's also why I had the comment here.
>
> The comment is not enough to make the code future proof, as we really
> want do clear all bits in the range beyond device->total_size and that
> works only if we specify them all (clear_state_bit).
>
> Either we can define a bitmask of all the bits next to their definition
> (slightly more error prone if we forget to add that manually), or pass a
> all ones to the clear_extent_bit. Though this is a bit uncommon, with a
> comment I'd prefer this option.

OK, I'd go that (u64)-1 as clear_bits direction, with updated comment.

Thanks,
Qu

>
>>>>> But I still have doubts about just clearing the range, why are there any
>>>>> device->alloc_state entries at all after device is shrunk?
>>>>
>>>> Because the alloc_state is mostly only utilized by trim facility, thus
>>>> existing functions won't bother clearing/setting it.
>>>>
>>>> In this particular case, previous fstrim run would set the CHUNK_TRIMMED
>>>> bit for all unallocated range (except the super reserve).
>>>> Then shrink doesn't clear the exceed range, and cause problem.
>>>
>>> So the unallocated range on a device is also represented in the
>>> alloc_state tree?
>>
>> If the unallocated range has been trimmed, then it has an state, with
>> CHUNK_TRIMMED bit set.
>
> To clarify my previous doubts: I was talking about the slack space (ie.
> space beyond device->total_size up to block device size), 'unallocated'
> is ambiguous and confusing here.
>
> The missing bit about clearing the whole range lies inside
> clear_state_bit that deletes all state tracking once all bits are
> dropped.
>