Re: SCSI QLA not working on latest *-mm SN2

[Jeremy Higdon <jeremy@sgi.com>]

On Tue, Sep 21, 2004 at 02:22:48PM -0700, Andrew Vasquez wrote:
> On Tuesday, September 21, 2004 2:06 PM, Jeremy Higdon wrote:
> > On Tue, Sep 21, 2004 at 01:50:02PM -0700, Andrew Vasquez wrote:
> > > The only requirement after reception of a soft-reset request (by PIO
> > > or MMIO) by the RISC is for the driver to wait 16 PCI clocks before
> > > issuing another request.  The problem of course is determining when
> > > to start timing within the driver.
> > 
> > So I think that we just wait for some reasonable worst case time for
> > the write to complete.  We can't really do anything else.
> >
> 
> That seems to be the case.
>  
> > Are these resets done as part of error recovery?  I.e., do we have
> > to be concerned about long the write will take on a busy system?
> > 
> 
> Just during initialization (qla2x00_chip_diag() and qla2x00_setup_chip())
> and after a firmware dump (qla_dbg.c -- which requires a full 
> re-initialization to recover).  If the driver is in the process of
> performing a firmware-dump, the RISC has already been paused and nothing
> is executing within the firmware.
> 
> --
> Andrew

Well then 10us should be plenty of time.  Maybe 20 just to be extra sure.

jeremy