From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin Herrenschmidt To: Raymond Waldrop , Bill Fink , Subject: Re: aic7xxx breakage Date: Sun, 11 Aug 2002 10:26:15 +0200 Message-Id: <20020811082616.1423@192.168.4.1> In-Reply-To: <20020810084648.A1510@adsl-208-190-252-230.dsl.wchtks.swbell.net> References: <20020810084648.A1510@adsl-208-190-252-230.dsl.wchtks.swbell.net> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Sender: owner-linuxppc-dev@lists.linuxppc.org List-Id: >Ben, > >I also encountered this problem on a dual G4 450 (AGP graphics). On the >theory that more information is often useful (or at least easily deleted >from the email queue), I am sending the results of "lspci -vv" as well as >a couple of dmesg logs. The last kernel I made that works with the card >is a 2.4.19-rc1 kernel from the linuxppc_2_4 tree. This is the kernel >I was running for the lspci output. I am also sending dmesg logs from >this kernel, and from a 2.4.19 kernel built from the linuxppc_2_4_devel >tree. If they would be useful, I can also send diffs of any of the PCI >driver source code files between the two trees. (I don't know how many >versions you might have laying around easily fed into diff.) Ok, can you both try editing arch/ppc/kernel/pci.c, and in pcibios_init(), replace the call to pcibios_assign_resource() with a call to pci_assign_unassigned_resources() and tell me what happens ? If the resulting kernel boots, please send me the dmesg log and the lspci -vv output. Thanks Ben. ** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/