On Fri, Sep 12, 2014 at 5:11 PM, Dirk Gouders <d...@gouders.net> wrote: > Yinghai Lu <ying...@kernel.org> writes: > > Yes, I skip them in pci_scan_device() to get the unused bus number space > that 0a.0 and 0b.0 then can use when reconfiguration is triggered.
That is not right. Can not let two bridges use same bus num. > >> Proper simulation could be: >> 1. remove 07:00.0 >> echo 1 > /sys/bus/pci/devices/0000:07:00.0/remove >> 2. change bus range in 00:0a.0 bus >> setpci -s 00:0a.0 0x1c.l=0x00101000 >> 3. remove 00:0a.0 >> echo 1 > /sys/bus/pci/devices/0000:00:0a.0/remove >> 4. rescan pci >> echo 1 > /sys/bus/pci/rescan >> 5. reset pcie link from 00:0a.0 >> setpci -s 00:0a.0 0xc0.b=0x18 >> sleep 1s >> setpci -s 00:0a.0 0xc0.b=0x08 >> 6. rescan pci bus >> echo 1 > /sys/bus/pci/rescan > > OK, I did that (with 3.16) but the reconfiguration code doesn't > get triggered. Here is what dmesg and lspci -tv say: 2 should be setpci -s 00:0a.0 0x18.l=0x00101000 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
