On 10/11/2019 12:13 PM, Bart Van Assche wrote: > On 10/11/19 3:12 AM, Bodo Stroesser wrote: >> If I understood correctly, we all would prefer to make TCMU completely >> transparent regarding the CDB. >> >> For pscsi I think we all agree, that the code is wrong or incomplete. >> But for pscsi up to now no one complained. So I'm wondering whether we >> should spend much effort for pscsi to discuss / find out the right >> solution? Especially as the original comment in the code, as posted by >> Mike, as well as the current comment are not very clear. Thus 'fixing' >> pscsi would mean to change or remove something we don't understand >> completely. >> >> So I agree to Hannes: we should simply move that code from >> passthrough_parse_cdb() to pscsi_parse_cdb(), at least as a first step. > > Hannes wrote "[ ... ] delegate it to pscsi if there is a need." Is there > really a need for such code in the SCSI passthrough driver? The upstream > SCSI target code is one of the four Linux SCSI target stacks that I know > of. I haven't found any SCSI-2 LUN number filtering code in the tgt > project (http://stgt.sourceforge.net/). I'm not sure about IET. SCST is > around since 2006 and has SCSI-2 LUN number filtering code in its CDROM > and MODISK passthrough drivers. That LUN number filtering code is there > since the initial commit so it is not the result of a recent request of > a user. SCST does not have LUN filtering code in its disk passthrough > driver nor in its tape passthrough driver. In other words, I do not > expect that anyone's setup will be broken by removing > passthrough_parse_cdb() entirely. I originally thought it was for a really specific case Nick hit a long time ago because the code is so odd. However, the original code's comment sounds like Nick might have read the specs incorrectly or was working off a older iscsi rfc draft. I think this happened in some persistent reservation code that is wrong too. I think if the code it was for a specific initiator then he would have added that in the code comments. For example for a while we had a really specific comment about the mac os initiator or qlogic on windows or something, and there is still a comment about solaris. If on the other hand it was a spec thing then it seems he writes the scsi spec name like he did in the original comment.
