public inbox for linux-rdma@vger.kernel.org
 help / color / mirror / Atom feed
From: sagi grimberg <sagig@mellanox.com>
To: Mike Christie <michaelc@cs.wisc.edu>,
	martin.petersen@oracle.com, Christoph Hellwig <hch@infradead.org>
Cc: nab@linux-iscsi.org, roland@kernel.org,
	linux-scsi@vger.kernel.org, target-devel@vger.kernel.org,
	linux-rdma@vger.kernel.org
Subject: Re: [PATCH v2 1/3] scsi_cmnd: Introduce scsi_transfer_length helper
Date: Tue, 24 Jun 2014 16:01:10 +0300	[thread overview]
Message-ID: <53A97696.6040302@mellanox.com> (raw)
In-Reply-To: <53A920B2.9060503@cs.wisc.edu>

On 6/24/2014 9:54 AM, Mike Christie wrote:
> On 06/11/2014 04:09 AM, Sagi Grimberg wrote:
>> In case protection information exists on the wire
>> scsi transports should include it in the transfer
>> byte count (even if protection information does not
>> exist in the host memory space). This helper will
>> compute the total transfer length from the scsi
>> command data length and protection attributes.
>>
>> Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
>> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
>> ---
>>   include/scsi/scsi_cmnd.h |   17 +++++++++++++++++
>>   1 files changed, 17 insertions(+), 0 deletions(-)
>>
>> diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
>> index dd7c998..a100c6e 100644
>> --- a/include/scsi/scsi_cmnd.h
>> +++ b/include/scsi/scsi_cmnd.h
>> @@ -7,6 +7,7 @@
>>   #include <linux/types.h>
>>   #include <linux/timer.h>
>>   #include <linux/scatterlist.h>
>> +#include <scsi/scsi_device.h>
>>   
>>   struct Scsi_Host;
>>   struct scsi_device;
>> @@ -306,4 +307,20 @@ static inline void set_driver_byte(struct scsi_cmnd *cmd, char status)
>>   	cmd->result = (cmd->result & 0x00ffffff) | (status << 24);
>>   }
>>   
>> +static inline unsigned scsi_transfer_length(struct scsi_cmnd *scmd)
>> +{
>> +	unsigned int xfer_len = blk_rq_bytes(scmd->request);
>> +	unsigned int prot_op = scsi_get_prot_op(scmd);
>> +	unsigned int sector_size = scmd->device->sector_size;
>> +
>> +	switch (prot_op) {
>> +	case SCSI_PROT_NORMAL:
>> +	case SCSI_PROT_WRITE_STRIP:
>> +	case SCSI_PROT_READ_INSERT:
>> +		return xfer_len;
>> +	}
>> +
>> +	return xfer_len + (xfer_len >> ilog2(sector_size)) * 8;
>> +}
>> +
>>   #endif /* _SCSI_SCSI_CMND_H */
>>
> I found the issue Christoph is hitting in the other thread.
>
> The problem is WRITE_SAME requests are setup so that req->__data_len is
> the value of the entire request when the setup is completed but during
> the setup process it's value changes
>
> So __data_len could be thousands of bytes but
> scsi_out(scsi_cmnd)->length for this case was only returning 512 which
> is the sector size. This is because sd_setup_-write_same_cmnd does:
>
>
> rq->__data_len = sdp->sector_size;
> ....
> scsi_setup_blk_pc_cmnd()
> ....
> rq->__data_len = nr_bytes;
>
> and scsi_setup_blk_pc_cmnd does scsi_init_io() -> scsi_init_sgtable()
> and that does
>
> sdb->length = blk_rq_bytes(req);
>
> and at this time because before we called scsi_setup_blk_pc_cmnd we set
> the __data_len to sector size, the sdb length is going to be only 512
> but the final request->__data_len is the total size of the operation.


Hey Christoph, Mike &MKP,

So just got first look in this thread, didn't have time to reproduce it yet.
Mike's analysis makes sense to me, I think the below change should 
resolve this one.

diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index a100c6e..2afd9c2 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -309,7 +309,7 @@ static inline void set_driver_byte(struct scsi_cmnd 
*cmd, char status)

  static inline unsigned scsi_transfer_length(struct scsi_cmnd *scmd)
  {
-       unsigned int xfer_len = blk_rq_bytes(scmd->request);
+       unsigned int xfer_len = scsi_bufflen(scmd);
         unsigned int prot_op = scsi_get_prot_op(scmd);
         unsigned int sector_size = scmd->device->sector_size;


Moreover, since bidi and dif are adjacent, this will also needed:

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 3f46234..abf0c3e 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -386,12 +386,14 @@ static int iscsi_prep_scsi_cmd_pdu(struct 
iscsi_task *task)
                 rc = iscsi_prep_bidi_ahs(task);
                 if (rc)
                         return rc;
+               transfer_length = scsi_in(sc)->length;
+       } else {
+               transfer_length = scsi_transfer_length(sc);
         }

         if (scsi_get_prot_op(sc) != SCSI_PROT_NORMAL)
                 task->protected = true;

-       transfer_length = scsi_transfer_length(sc);
         hdr->data_length = cpu_to_be32(transfer_length);
         if (sc->sc_data_direction == DMA_TO_DEVICE) {
                 struct iscsi_r2t_info *r2t = &task->unsol_r2t;

Let me test and queue it up.

Sagi.

  parent reply	other threads:[~2014-06-24 13:01 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-06-11  9:09 [PATCH v2 0/3] Include protection information in transport header Sagi Grimberg
2014-06-11  9:09 ` [PATCH v2 1/3] scsi_cmnd: Introduce scsi_transfer_length helper Sagi Grimberg
2014-06-11 23:39   ` Martin K. Petersen
2014-06-23 21:24   ` Mike Christie
     [not found]     ` <53A89B0F.4040300-hcNo3dDEHLuVc3sceRu5cw@public.gmane.org>
2014-06-24  1:58       ` Martin K. Petersen
2014-06-25  1:17         ` Vladislav Bolkhovitin
2014-07-27  8:45           ` Boaz Harrosh
     [not found]   ` <1402477799-24610-2-git-send-email-sagig-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2014-06-24  6:54     ` Mike Christie
2014-06-24 12:53       ` Martin K. Petersen
     [not found]         ` <yq1mwd2h3ju.fsf-+q57XtR/GgMb6DWv4sQWN6xOck334EZe@public.gmane.org>
2014-06-24 13:08           ` Sagi Grimberg
2014-06-24 14:55             ` Christoph Hellwig
2014-06-24 15:29           ` sagi grimberg
2014-06-24 14:03         ` Christoph Hellwig
2014-06-24 16:08         ` Michael Christie
2014-06-24 16:27           ` Christoph Hellwig
2014-06-24 16:27           ` Sagi Grimberg
2014-06-24 16:30             ` Christoph Hellwig
     [not found]               ` <20140624163040.GA11499-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-06-24 17:00                 ` Mike Christie
2014-06-24 17:04                   ` Martin K. Petersen
2014-06-24 17:08                   ` Mike Christie
     [not found]                     ` <53A9B0A0.6000103-hcNo3dDEHLuVc3sceRu5cw@public.gmane.org>
2014-06-25  3:32                       ` Mike Christie
2014-06-25  8:48                         ` Sagi Grimberg
2014-06-25  9:17                           ` Christoph Hellwig
2014-06-25 10:32                           ` Sagi Grimberg
     [not found]                             ` <53AAA547.40300-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2014-06-25 11:35                               ` Christoph Hellwig
     [not found]                                 ` <20140625113536.GA30312-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-06-25 15:59                                   ` Michael Christie
2014-07-27  9:11                             ` Boaz Harrosh
     [not found]                               ` <53D4C22F.8050904-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-07-27 13:52                                 ` Christoph Hellwig
2014-08-06 12:15                               ` Sagi Grimberg
2014-06-25  9:14                         ` Christoph Hellwig
2014-06-25 11:29                           ` Martin K. Petersen
2014-06-24 16:31           ` Martin K. Petersen
     [not found]             ` <yq1fviugtgq.fsf-+q57XtR/GgMb6DWv4sQWN6xOck334EZe@public.gmane.org>
2014-06-24 17:05               ` Mike Christie
2014-06-24 13:01       ` sagi grimberg [this message]
2014-06-26 14:53   ` Bart Van Assche
     [not found]     ` <53AC3402.2080302-HInyCGIudOg@public.gmane.org>
2014-06-26 14:55       ` James Bottomley
2014-06-26 15:41         ` Atchley, Scott
2014-06-26 16:38           ` James Bottomley
     [not found]             ` <fbbc6688-5a52-4437-93b1-71e8ff84c36c-2ueSQiBKiTY7tOexoI0I+QC/G2K4zDHf@public.gmane.org>
2014-06-26 21:17               ` Atchley, Scott
2014-07-13 11:37   ` Christoph Hellwig
2014-07-13 11:40     ` Martin K. Petersen
2014-06-11  9:09 ` [PATCH v2 2/3] libiscsi, iser: Adjust data_length to include protection information Sagi Grimberg
2014-06-23 20:59   ` Christoph Hellwig
     [not found]     ` <20140623205948.GA15165-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-06-24  6:31       ` Mike Christie
     [not found]   ` <1402477799-24610-3-git-send-email-sagig-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2014-07-27 10:08     ` Boaz Harrosh
     [not found]       ` <53D4CFAB.3040804-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-08-06 12:43         ` Sagi Grimberg
2014-08-06 13:25           ` Boaz Harrosh
2014-08-13 13:09             ` Sagi Grimberg
     [not found]               ` <53EB639C.3080307-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2014-08-14  7:17                 ` Boaz Harrosh
     [not found]           ` <53E22300.3090907-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2014-08-06 15:54             ` Martin K. Petersen
2014-06-11  9:09 ` [PATCH v2 3/3] TARGET/sbc,loopback: Adjust command data length in case pi exists on the wire Sagi Grimberg
2014-06-11 21:36 ` [PATCH v2 0/3] Include protection information in transport header Nicholas A. Bellinger

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=53A97696.6040302@mellanox.com \
    --to=sagig@mellanox.com \
    --cc=hch@infradead.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=michaelc@cs.wisc.edu \
    --cc=nab@linux-iscsi.org \
    --cc=roland@kernel.org \
    --cc=target-devel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox