Re: [PATCH RFC] btrfs: implement a new compat RO feature, data_io_size

在 2026/8/1 06:55, Qu Wenruo 写道:
> 
> 
> 在 2026/7/31 21:10, Sun YangKai 写道:
>> On 2026/7/22 18:03, Qu Wenruo wrote:
>>>
>>> Full stripe 16K, then the new RAID5F chunks will require power-of-2
>>> disks, and no more than 16K/4K disks.
>>>
>>> So no 4 disks RAID5F chunk in the first place.
>>
>> I've got your idea and got the following 2 concerns:
>>
>> 1. the device count is limited to 3 or 5 device for RAID5, and 4 or 6
>> devices for RAID6, losing the flexibility currently btrfs have;
> 
> Why you think 16K is the only block size?
> 
> You can go as large as 64K, which means 16 data devices which is more 
> than enough for common usages.

Furthermore, even with 16K full stripe size, you can still go less than 
(2+1) and (4+1), you can also go (1+1).

So I do not think there is any inflexibility.

If you think (3+1) or some other weird combination is a good idea, 
please tell me what this weird combination can provide that 
(1+1)/(2+1)/(4+1) can not do.

> 
> In fact, I do not think you should have overly flex design, which is 
> contributing to the write-hole in the first place.
> 
>>
>> 2. On each device, we will send a lot of 4KiB or 8KiB io, which might be
>> not good for performance;
> 
> It can always be solved usings blk plugs.
> 
>>
>> So I personally don't think this is a good way to fix the write hole 
>> issue.
> 
> I do not think so.
> 
> As I have already said, at least I know XFS is planning doing something 
> similar. Although not to solve write-hole.
>