Date: Sat, 21 Apr 2001 23:27:58 +0200 (MET DST) From: j@uriah.heep.sax.de (J Wunsch) To: freebsd-scsi@freebsd.org Cc: j@uriah.heep.sax.de, mjacob@feral.com Subject: Re: Problem with current sa(4) driver Message-ID: <200104212127.f3LLRwf02467@uriah.heep.sax.de> References: <20010414203925.A63281@uriah.heep.sax.de> <200104150504.f3F544s00932@aslan.scsiguy.com> <20010418225355.U688@uriah.heep.sax.de>
next in thread | previous in thread | raw e-mail | index | archive | help
J Wunsch <j@uriah.heep.sax.de> wrote:
>> While it is true that the sa driver should be filtering out this
>> particular case because there is no error, returning ERESTART for
>> NO_SENSE is also wrong. You should be able to fix that by changing
>> the table entry for that sense code in cam_periph.c.
>
> You mean, like this?
Hmm, no comments?
> I tried to manually patch the return value of cam_periph_error() to 0
> in kgdb, but this just gets me back at the second problem:
>
> % ps axl
> UID PID PPID CPU PRI NI VSZ RSS WCHAN STAT TT TIME COMMAND
> ...
> 107 373 1 0 -8 0 244 33 cbwait DWE p0- 0:00.00 dd if=/dev/sa0 of=/dev/null bs=10
>
> It sits there, and waits indefinately.
OK, i found this one. Obviously, the device queue is still frozen at
this point. The following patch fixes the problem for me, by now only
tested against that ARCHIVE Python DDS drive, but will also test
shortly against my Tandberg TDC4222 drive. What the fix does is
filtering out `no errors', i. e. illegal length indications with a
shorter tape block than requested, and hitting a filemark during read,
and then instead of calling cam_periph_error(), simply making sure
that the device queue is really becoming unfrozen at exit (in case of
an actual error, cam_periph_error() performs the unfreeze).
I'd like to commit the fix to scsi_sa.c (or have Matt commit it, i
don't care).
Also repeated the patch for cam_periph.c that makes a NO_SENSE return
no error, which should IMHO be what Justin was referring to in the
quote above.
Any feedback highly appreciated, since i feel a little far out on thin
ice here in CAM-land.
Index: cam_periph.c
===================================================================
RCS file: /home/ncvs/src/sys/cam/cam_periph.c,v
retrieving revision 1.34
diff -u -r1.34 cam_periph.c
--- cam_periph.c 2001/04/04 18:24:35 1.34
+++ cam_periph.c 2001/04/17 17:46:11
@@ -1369,6 +1369,8 @@
switch (err_action & SS_MASK) {
case SS_NOP:
+ error = 0;
+ break;
case SS_RETRY:
action_string = "Retrying Command";
error = ERESTART;
Index: scsi/scsi_sa.c
===================================================================
RCS file: /home/ncvs/src/sys/cam/scsi/scsi_sa.c,v
retrieving revision 1.67
diff -u -r1.67 scsi_sa.c
--- scsi/scsi_sa.c 2001/03/27 05:45:12 1.67
+++ scsi/scsi_sa.c 2001/04/21 21:05:53
@@ -2286,7 +2286,7 @@
u_int32_t resid = 0;
int32_t info = 0;
int error_code, sense_key, asc, ascq;
- int error, defer_action;
+ int error, defer_action, no_actual_error = FALSE;
periph = xpt_path_periph(ccb->ccb_h.path);
softc = (struct sa_softc *)periph->softc;
@@ -2396,6 +2396,8 @@
if (defer_action) {
error = -1;
softc->flags |= SA_FLAG_EOF_PENDING;
+ } else {
+ no_actual_error = TRUE;
}
/*
* Unconditionally, if we detected a filemark on a read,
@@ -2424,6 +2426,8 @@
softc->flags |= SA_FLAG_EIO_PENDING;
else
error = EIO;
+ } else {
+ no_actual_error = TRUE;
}
/*
* Bump the block number if we hadn't seen a filemark.
@@ -2438,8 +2442,17 @@
}
}
}
- if (error == 0)
+ if (error == 0 && !no_actual_error)
return (cam_periph_error(ccb, cflgs, sflgs, &softc->saved_ccb));
+ if (no_actual_error) {
+ if ((ccb->ccb_h.status & CAM_DEV_QFRZN) != 0)
+ cam_release_devq(ccb->ccb_h.path,
+ /* relsim_flags */0,
+ /* openings */0,
+ /* timeout */0,
+ /* getcount_only */ FALSE);
+ return (0);
+ }
if (error == -1)
return (0);
--
cheers, J"org .-.-. --... ...-- -.. . DL8DTL
http://www.sax.de/~joerg/ NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200104212127.f3LLRwf02467>