Hi 于 2025年5月14日 GMT+08:00 13:17:47，Christoph Hellwig <hch@xxxxxx> 写道： >On Tue, May 13, 2025 at 05:32:13PM +0800, Yu Kuai wrote: >> I was thinking about record a stack dev depth in mddev to handle the >> weird case inside raid. Is there other stack device have the same >> problem? AFAIK, some dm targets like dm-crypt are using workqueue >> to handle all IO. > >I guess anything that might have to read in metadata to serve a >data I/O. bcache, dm-snapshot and dm-thinkp would be candidates for >that, but I haven't checked the implementation. > >> I'm still interested because this can improve first write latency. >> >>> >>> So instead just write a comment documenting why you switch to a >>> different stack using the workqueue. >> >> Ok, I'll add comment if we keep using the workqueue. > >Maybe do that for getting the new bitmap code in ASAP and then >revisit the above separately? > > Sure, sorry that I am in a business travel suddenly. I will get back to this ASAP I return. Thanks Kuai
