On 12/5/18 7:41 AM, Christoph Hellwig wrote: > On Tue, Dec 04, 2018 at 03:47:46PM -0700, Jens Axboe wrote: >> If we attempt a direct issue to a SCSI device, and it returns BUSY, then >> we queue the request up normally. However, the SCSI layer may have >> already setup SG tables etc for this particular command. If we later >> merge with this request, then the old tables are no longer valid. Once >> we issue the IO, we only read/write the original part of the request, >> not the new state of it. >> >> This causes data corruption, and is most often noticed with the file >> system complaining about the just read data being invalid: > > Looks good as the first quick fix: > > Reviewed-by: Christoph Hellwig <hch@xxxxxx> > > But as we discussed I think we need to be even more careful. including > the only read and write check below in the thread or some higher level > approach. I agree, I'm reviewing the patches from Jianchao now and will do something cleaner on top of that. -- Jens Axboe
