On Sep 27, 2012, at 6:43 AM, Davide Viti wrote: > Hi, > >> So its odd that scanning of the second bus didn't report any devices. Do > you have code that implements ppc_md.pci_exclude_device ? > > not that I'm aware > of > >> You might also want to put some code in the indirect PCI ops (indirect.c) > to see what actual values you are getting from various indirect_read_config() > calls. > > To make sure that ppc_md.pci_exclude_device is not implemented, I've > put some printouts inside indirect_read_config(): I print various parameters > when the function is called, and when it returns and note that: > > 1. > indirect_read_config() is called 422 times: > 174 times for > [/pcie@ffe0a000] (controller where the device is not detected) > 248 times > for [/pcie@ffe09000] > 2. ppc_md.pci_exclude_device is always NULL > 3. the > function always returns with PCIBIOS_SUCCESSFUL > 4. the only call to > indirect_read_config() inside which bus_no=0x3, returns with the following > log: > > pci_bus 0001:03: scanning bus > -> ind_r_config - [/pcie@ffe0a000] > devfn=0x0 len=0x4 hose->indirect_type=0x16 hose->first_busno=0x2 bus- >> number=0x3 > -> ind_r_config [/pcie@ffe0a000] - (bus_no=0x3 reg=0x0 > cfg_data=0xffffffff len=0xff7eb004) val=0x4 PCIBIOS_SUCCESSFUL
Can you see what bus_no actually gets set to in the case we scan 0001:03 ? If its set to 03, can you try hack it to 1. - k > > the entire > log is about 116Kb and is available in [1] or [2] (didn't feel like pasting > so > much data on the ML) > thanx alot, > > Davide > > [1] http://pastebin.com/JaPGmmfs > [2] > http://paste2.org/p/2273728 > > > Invita i tuoi amici e Tiscali ti premia! Il consiglio di un amico vale più di > uno spot in TV. Per ogni nuovo abbonato 30 € di premio per te e per lui! Un > amico al mese e parli e navighi sempre gratis: http://freelosophy.tiscali.it/ _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
