Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 12 Nov 1996 14:20:03 -0700
From:      "Mike Durian" <durian@plutotech.com>
To:        se@zpr.uni-koeln.de (Stefan Esser)
Cc:        freebsd-hackers@freebsd.org
Subject:   Re: small bugs in pci code 
Message-ID:  <199611122120.OAA04654@pluto.plutotech.com>
In-Reply-To: Your message of "Tue, 12 Nov 1996 21:52:32 %2B0100."

next in thread | raw e-mail | index | archive | help
On Tue, 12 Nov 1996 21:52:32 +0100, se@zpr.uni-koeln.de (Stefan Esser) wrote:
>
>Well, I'm sure it never was set to anything but 0x28 as the 
>address of the first register beyond the map registers.
>(I even checked the CVS file ...)

  I've tracked it down.  We did change it because it limited the
registers that could be mapped with pci_map_mem.  We need to be
able to read and write (reprogram) our adaptecs' seeproms.  To
do this, we map in the eeprom memory and thus need to work around
the pci_map_mem restriction.  We never realized (or even noticed until
now) the ramifications of changing the PCI_MAP_REG_END value.  Thanks
for showing us the error of our ways.

>But it was a false alarm anyway: The code will restore the old
>value of each map register at the end of the probe, and it will
>disbale the ROM at that point again.

  And it looks like we go and re-enable it later.  Sigh.  I need to
talk to some people about this.

mike



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611122120.OAA04654>