write barrier in RAID5

I can see how FUA is handled, but REQ_FLUSH seems problematic. My understanding of REQ_FLUSH semantics: all requests before the flush request must *complete* before the flush request complete. For IO remap layer like MD, we need to guarantee all previous requests are *submitted* to component devices before submitting the flush request. Is that correct?

If so why I don't see the code which make sure the pending write requests in stripe cache are processed & submitted before submit_flushes() get called? Thanks

Eric

No.

See Documentation/block/writeback_cache_control.txt

Only "previously completed" requests are affect by REQ_FLUSH.

So any code using REQ_FLUSH must first wait for previous requests to complete, and then submit REQ_FLUSH. The REQ_FLUSH will not return until the data written by those requests is actually in stable storage, rather than write-behind cache.

NeilBrown