Groups > comp.os.linux.development.system > #707

Re: SCSI generic - asynchronous execution

Newsgroups	comp.os.linux.development.system
Date	2014-05-28 10:32 -0700
References	<ihum92$tiu$1@nntp.fujitsu-siemens.com> <ii2mi1$kku$3@news.eternal-september.org> <ii60fi$aob$1@nntp.fujitsu-siemens.com>
Message-ID	<41c5fc41-4854-4192-8fa9-fb0e6c4a4ec6@googlegroups.com> (permalink)
Subject	Re: SCSI generic - asynchronous execution
From	lucasrangit@gmail.com

Show all headers | View raw

On Monday, January 31, 2011 1:47:29 AM UTC-8, Josef Moellers wrote:
> > On Fri 1/28/2011 7:10 AM, Josef Moellers wrote:
> >> Is anyone using the read()/write() interface to SCSI generic rather than
> >> the ioctl(SG_IO)?

Did you get synchronous working using ioctl(SG_IO)? I tried testing this but my SCSI SAT command is getting rejected by the LLD. It does not go out the wire. I'm still trying to figure out where the command goes after scsi_dispatch_cmnd(). 

sd 13:0:0:0: [sda] CDB: ATA command pass through(16): 85 0f 0d 00 01 00 00 00 00 00 00 00 00 40 60 00
sd 13:0:0:0: [sda]  Sense Key : Illegal Request [current] 
sd 13:0:0:0: [sda]  Add. Sense: Invalid field in cdb

> >>          cdb[ 5] = (tag>>  8)&  0xff;
> >>          cdb[ 6] = (tag     )&  0xff;
> >> Any ideas what is wrong?

Table 67 (READ FPDMA QUEUED command inputs) in ACS-3 shows the tag is 5 bits in the low-order byte only.

cdb[5] = 0; /* RSVD, PRIO */
cdb[6] = (tag & 0x1F) << 3; /* 7:3 NCQ TAG, RSVD */

Back to comp.os.linux.development.system | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

Re: SCSI generic - asynchronous execution Cat22 <oops@invlid.org> - 2011-01-29 19:39 -0800
  Re: SCSI generic - asynchronous execution Josef Moellers <josef.moellers@ts.fujitsu.com> - 2011-01-31 10:47 +0100
    Re: SCSI generic - asynchronous execution lucasrangit@gmail.com - 2014-05-28 10:32 -0700
      Re: SCSI generic - asynchronous execution shanecowman@gmail.com - 2015-07-22 15:01 -0700

csiph-web