On 06/25/2014 01:32 PM, Sagi Grimberg wrote:

On 6/25/2014 11:48 AM, Sagi Grimberg wrote:

<SNIP>

quoted

quoted

I made the patch below which should fix both bidi
support in iscsi and also WRITE_SAME (and similar commands) support.

I'm a bit confused, for all commands (bidi or not) the iscsi header 
data_length
should describe the scsi_data_buffer length, bidi information will lie 
in AHS header.
(in case bidi will ever co-exist with PI, we might need another helper 
that will look
in req->special + PI, something like scsi_bidi_transfer_length)

So why not keep scsi_transfer_length to work on sdb length (take 
scsi_bufflen(scmnd) or
scsi_out(scmnd)->length as MKP suggested) and that's it - don't touch 
libiscsi.

Let me test that.

So I tested a bidirectional command using:
$ sg_raw --infile=/root/filex --send=1024 --request=1024 
--outfile=/root/filex "/dev/bsg/7:0:0:0" 53 00 00 00 00 00 00 00 02 00

And I see:
kernel: session1: iscsi_prep_scsi_cmd_pdu iscsi prep [bidirectional cid 
0 sc ffff880468ca1e00 cdb 0x53 itt 0x16 len 1024 bidi_len 1024 cmdsn 223 
win 64]

This is a very bad example and tested nothing, since len && bidi_len
are the same. So even if you had a bug and took length from the
wrong side it would come out the same.

You must test with a bidi command that has two different lengths for
each side

If you have a tree that you want me to test I will be glad too.

From this thread I'm confused as to what patches you want me to

test? please point me to a tree you need testing. You can bug me
any time, any tree. I will be happy to test.

Cheers
Boaz

This confirms what I wrote above, so AFAICT no additional fix is 
required for libiscsi wrt bidi commands support.

Mike?

Sagi.
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html