On Fri, 31 May 2013 01:19:44 +0300 nezhinsky@xxxxxxxxx wrote: > From: Alexander Nezhinsky <nezhinsky@xxxxxxxxx> > > Windows iscsi initiator failed to format devices because in read > scsi commands, mainly READ(10), the last (but not the first) Data-IN > pdu contained zero buffer offset. Obviously this may happen only > with relatively small MaxRecvDSL, and relatively large cmd sizes, > so that multiple Data-IN pdus are sent and the last one carrying > the GOOD status. Linux sets much larger values of MaxRecvDSL than > Windows, and employes relaxed checks of pdu correctness. > It seems that it relies on DataSN and disregards buffer offset. > > The fix was to handle data-in with status and scsi resp separately. > In addition a rectification is made for the tx pdu generation. > > Alexander Nezhinsky (2): > Fix buffer_offset set to 0 in the last Data-IN carrying status > Create tx pdu by explicit scsi cmd type (rd,wr,bidir) > > usr/iscsi/iscsid.c | 43 +++++++++++++++++++++++++++++++++++-------- > 1 file changed, 35 insertions(+), 8 deletions(-) You confirmed that Windows format works with this patch? -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
