Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Doug Ledford]

On Thu, Feb 06, 2003 at 06:25:02PM -0800, Patrick Mansfield wrote:
> On Thu, Feb 06, 2003 at 06:01:06PM -0800, Martin J. Bligh wrote:
> > 
> > Curious. I've no idea why the changes brought this out then ... I've done
> > hundreds and hundreds of reboots on 2.5 on all sorts of different kernels,
> > and never, ever seen this. Yet in 2.5.59-bk I see it every single time.
> > Very odd.
> > 
> > M.
> 
> Okay:
> 
> There were some bk scsi changes that ignored the queue depth (qlogicisp
> sets them all to one). 
> 
> Current bk (I just pulled and checked) has a fix, the cleaner shinier 
> better scsi_lib.c scsi_request_fn now has this code:
> 
> 	if (sdev->device_busy >= sdev->queue_depth)
> 		break;
> 
> So the oops has to do with the isp handling multiple requests in a row or
> in quick succession.
> 
> Hopefully going to the latest bk will fix your oops.

It might, but please understand this.  The qlogicisp driver does things to
the scsi mid layer that the scsi mid layer does not protect itself against
and as a result is the biggest pile of steaming, unsupportable, crap code
in the universe!  The scsi mid layer was designed from day one to think
that the host->can_queue, sdev->queue_depth, and host->sg_tablesize items
were *static* on a given host/device unless specifically changed by
calling into the adjustment routines (scsi_adjust_queue_depth).  The
qlogicisp driver violates those principles and I make no warranty of any
kind that said driver will continue to operate properly unless someone
takes the time to actually audit the qlogicisp_queuecommand() and
qlogicisp_irq() routine to make sure it is actually doing the right thing
when making those changes!

If I understand correctly, Matthew Jacob's latest isp driver set drives
*all* qlogic hardware (or at least all the older stuff like the qlogicisp
driver drives).  I would much prefer that people simply test out Matthew's
driver and use it instead.  In fact, if it's ready for 2.5 kernel use, I
would strongly recommend that it be considered as a possible replacement
in the linux kernel for the default driver on all qlogic cards not handled
by the new qla2x00 driver version 6 (DaveM may have objections to that 
related to sparc if Matthew's driver isn't sparc friendly, but I don't 
know of any other reason not to switch over).

-- 
  Doug Ledford <dledford@redhat.com>     919-754-3700 x44233
         Red Hat, Inc. 
         1801 Varsity Dr.
         Raleigh, NC 27606
  

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Anton Blanchard]

 
Hi,

> If I understand correctly, Matthew Jacob's latest isp driver set drives
> *all* qlogic hardware (or at least all the older stuff like the qlogicisp
> driver drives).  I would much prefer that people simply test out Matthew's
> driver and use it instead.  In fact, if it's ready for 2.5 kernel use, I
> would strongly recommend that it be considered as a possible replacement
> in the linux kernel for the default driver on all qlogic cards not handled
> by the new qla2x00 driver version 6 (DaveM may have objections to that 
> related to sparc if Matthew's driver isn't sparc friendly, but I don't 
> know of any other reason not to switch over).

I had a bunch of problems with the in kernel and vendor qlogic drivers
on my ppc64 box. Matt Jacob's driver worked out of the box. Davem
sounded positive last time I asked him about it.

I did a quick forward port to 2.5 a month or two ago, sounds like we
should work to get it in the kernel. There are some rough edges but
Mike kindly offered to lend a hand here.

Anton

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Mike Anderson]

I removed the other Mike Anderson (mikeand@us.ibm.com) from the cc
list, we seem to have many Mike Anderson's around hear :-).

Anton Blanchard [anton@samba.org] wrote:
>  
> Hi,
> 
> > If I understand correctly, Matthew Jacob's latest isp driver set drives
> > *all* qlogic hardware (or at least all the older stuff like the qlogicisp
> > driver drives).  I would much prefer that people simply test out Matthew's
> > driver and use it instead.  In fact, if it's ready for 2.5 kernel use, I
> > would strongly recommend that it be considered as a possible replacement
> > in the linux kernel for the default driver on all qlogic cards not handled
> > by the new qla2x00 driver version 6 (DaveM may have objections to that 
> > related to sparc if Matthew's driver isn't sparc friendly, but I don't 
> > know of any other reason not to switch over).
> 
> I had a bunch of problems with the in kernel and vendor qlogic drivers
> on my ppc64 box. Matt Jacob's driver worked out of the box. Davem
> sounded positive last time I asked him about it.
> 
> I did a quick forward port to 2.5 a month or two ago, sounds like we
> should work to get it in the kernel. There are some rough edges but
> Mike kindly offered to lend a hand here.

I have been buried lately so I have only taken the patch you sent me
and updated it so it will compile with the most recent SCSI
changes. I also made a few tweaks to the make files.

Currently it is running on my 2202 card system.

When I ran it my other system that has a Qlogic ISP1020 and two Qlogic
2300's it would hang post initing the driver. The driver seemed to be
responsive to external events like port downs, but appeared to not
complete its init. When I use the driver disable defines to turn off
detection of the ISP1020 the driver loaded ok (I only ran it up to 60MB
on a few spindles so not a very good test).

I will look at this more and see if I can understand the cause.

-andmike
--
Michael Anderson
andmike@us.ibm.com

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Andrew Morton]

Doug Ledford <dledford@redhat.com> wrote:
>
> I would much prefer that people simply test out Matthew's
> driver and use it instead.

Where is it?

http://www.feral.com/isp.html seems to be 2.4.x-only.

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Doug Ledford]

On Thu, Feb 06, 2003 at 08:19:39PM -0800, Andrew Morton wrote:
> Doug Ledford <dledford@redhat.com> wrote:
> >
> > I would much prefer that people simply test out Matthew's
> > driver and use it instead.
> 
> Where is it?
> 
> http://www.feral.com/isp.html seems to be 2.4.x-only.

As answered elsewhere, the 2.5 port isn't done yet.  That's why I said in 
my email "if it's ready for 2.5" because I was afraid Matthew hadn't 
gotten around to doing the 2.5 update yet.  However, if no one else can do 
it, I can make a 2.5 update of this driver happen (I don't suspect it 
would be that hard actually, not *that* much has to change).

-- 
  Doug Ledford <dledford@redhat.com>     919-754-3700 x44233
         Red Hat, Inc. 
         1801 Varsity Dr.
         Raleigh, NC 27606
  

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[William Lee Irwin III]

On Thu, Feb 06, 2003 at 08:19:39PM -0800, Andrew Morton wrote:
>> http://www.feral.com/isp.html seems to be 2.4.x-only.

On Thu, Feb 06, 2003 at 11:24:40PM -0500, Doug Ledford wrote:
> As answered elsewhere, the 2.5 port isn't done yet.  That's why I said in 
> my email "if it's ready for 2.5" because I was afraid Matthew hadn't 
> gotten around to doing the 2.5 update yet.  However, if no one else can do 
> it, I can make a 2.5 update of this driver happen (I don't suspect it 
> would be that hard actually, not *that* much has to change).

This driver's bugginess is a _major_ nuisance to me and I don't have
the SCSI know-how to fix it myself. I'd _love_ to test a driver with a
prayer of working anytime this century.

Thanks.

-- wli

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Matthew Jacob]

The other thing to note is that I'm not really very happy with my driver
at present. It may be working well for some people, but *I* think it
needs some rework before it's really ready for primetime again. I need
to split out the SCSI && FC dependencies. I need to move the name server
code out of the main body and make it more policy driven.

The trouble is also that it's just a hobby for me right now (no clients
with direct Linux support requirements) , and as a recent parent I've
had a lot less hobby time.


On Thu, 6 Feb 2003, William Lee Irwin III wrote:

> On Thu, Feb 06, 2003 at 08:19:39PM -0800, Andrew Morton wrote:
> >> http://www.feral.com/isp.html seems to be 2.4.x-only.
>
> On Thu, Feb 06, 2003 at 11:24:40PM -0500, Doug Ledford wrote:
> > As answered elsewhere, the 2.5 port isn't done yet.  That's why I said in
> > my email "if it's ready for 2.5" because I was afraid Matthew hadn't
> > gotten around to doing the 2.5 update yet.  However, if no one else can do
> > it, I can make a 2.5 update of this driver happen (I don't suspect it
> > would be that hard actually, not *that* much has to change).
>
> This driver's bugginess is a _major_ nuisance to me and I don't have
> the SCSI know-how to fix it myself. I'd _love_ to test a driver with a
> prayer of working anytime this century.
>
> Thanks.
>
> -- wli
> -
> 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
>

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Matthew Jacob]

I haven't integrated Anton's patches yet. Bad me.


On Thu, 6 Feb 2003, Andrew Morton wrote:

> Doug Ledford <dledford@redhat.com> wrote:
> >
> > I would much prefer that people simply test out Matthew's
> > driver and use it instead.
>
> Where is it?
>
> http://www.feral.com/isp.html seems to be 2.4.x-only.
> -
> 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
>

Re: Broken SCSI code in the BK tree (was: 2.5.59-mm8)

[Martin J. Bligh]

>> There were some bk scsi changes that ignored the queue depth (qlogicisp
>> sets them all to one). 
>> 
>> Current bk (I just pulled and checked) has a fix, the cleaner shinier 
>> better scsi_lib.c scsi_request_fn now has this code:
>> 
>> 	if (sdev->device_busy >= sdev->queue_depth)
>> 		break;
>> 
>> So the oops has to do with the isp handling multiple requests in a row or
>> in quick succession.
>> 
>> Hopefully going to the latest bk will fix your oops.
> 
> It might, but please understand this.  The qlogicisp driver does things to
> the scsi mid layer that the scsi mid layer does not protect itself against
> and as a result is the biggest pile of steaming, unsupportable, crap code
> in the universe!  The scsi mid layer was designed from day one to think
> that the host->can_queue, sdev->queue_depth, and host->sg_tablesize items
> were *static* on a given host/device unless specifically changed by
> calling into the adjustment routines (scsi_adjust_queue_depth).  The
> qlogicisp driver violates those principles and I make no warranty of any
> kind that said driver will continue to operate properly unless someone
> takes the time to actually audit the qlogicisp_queuecommand() and
> qlogicisp_irq() routine to make sure it is actually doing the right thing
> when making those changes!
> 
> If I understand correctly, Matthew Jacob's latest isp driver set drives
> *all* qlogic hardware (or at least all the older stuff like the qlogicisp
> driver drives).  I would much prefer that people simply test out Matthew's
> driver and use it instead.  In fact, if it's ready for 2.5 kernel use, I
> would strongly recommend that it be considered as a possible replacement
> in the linux kernel for the default driver on all qlogic cards not handled
> by the new qla2x00 driver version 6 (DaveM may have objections to that 
> related to sparc if Matthew's driver isn't sparc friendly, but I don't 
> know of any other reason not to switch over).

If you can send me a patch, I'll willingly test it .... I have plenty of
these cards on very racy machines ;-)

M.