From owner-freebsd-hackers  Fri Dec 15 02:59:40 1995
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id CAA13270
          for hackers-outgoing; Fri, 15 Dec 1995 02:59:40 -0800 (PST)
Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id CAA13259
          for <freebsd-hackers@freebsd.org>; Fri, 15 Dec 1995 02:59:32 -0800 (PST)
Received: by Sysiphos id AA17408
  (5.67b/IDA-1.5 for freebsd-hackers@freebsd.org); Fri, 15 Dec 1995 11:57:34 +0100
Message-Id: <199512151057.AA17408@Sysiphos>
From: se@zpr.uni-koeln.de (Stefan Esser)
Date: Fri, 15 Dec 1995 11:57:33 +0100
In-Reply-To: Mihoko Tanaka <m_tanaka@pa.yokogawa.co.jp>
       "SMC EtherPower doesn't work." (Dec 15, 18:45)
X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95)
To: Mihoko Tanaka <m_tanaka@pa.yokogawa.co.jp>
Subject: Re: SMC EtherPower doesn't work.
Cc: freebsd-hackers@freebsd.org
Sender: owner-hackers@freebsd.org
Precedence: bulk

On Dec 15, 18:45, Mihoko Tanaka wrote:
} Subject: SMC EtherPower doesn't work.
} 
} Hello all,
} 
} I use FreeBSD-2.1.0R on HP Vectra VL590. and I try to install the network
} card 'SMC EtherPowerPCI BT'.
} but it could not be recognized. The error message is:
} 
}         pci_map_port failed: device's iorange 0xff80-0xffff
}   	     is incompatible with its bridge's range 0x0-0xffff
} 
} But the bridge's iorange '0x0-0xffff' includes the device's iorange 
} '0xff80-0xffff'. I think the device's iorange is right.
} Is it a bug of pci driver ?

Hmm, it has up to now worked just fine for lots of I/O port mappings.
But I guess none extended up to port address 0xffff ...

} --- pci.c  Mon Oct  9 15:35:59 1995
} +++ pci.c.new       Thu Dec 14 12:09:19 1995
} @@ -922,7 +922,7 @@
}         };
}         iosize = -(data &  PCI_MAP_IO_ADDRESS_MASK);
}         if (ioaddr < pcicb->pcicb_iobase
} -               || ioaddr + iosize > pcicb->pcicb_iolimit) {
} +               || (ioaddr + iosize -1) > pcicb->pcicb_iolimit) {
}                 printf ("pci_map_port failed: device's iorange 0x%x-0x%x "
}                         "is incompatible with its bridge's range 0x%x-0x%x\n",
}                         (unsigned) ioaddr, (unsigned) ioaddr + iosize - 1,

Thanks for sending the bug report !
I'll probably apply a slightly different fix, though ...

Regards, STefan

-- 
 Stefan Esser, Zentrum fuer Paralleles Rechnen		Tel:	+49 221 4706021
 Universitaet zu Koeln, Weyertal 80, 50931 Koeln	FAX:	+49 221 4705160
 ==============================================================================
 http://www.zpr.uni-koeln.de/~se			  <se@ZPR.Uni-Koeln.DE>