On Thu, May 8, 2014 at 2:14 PM, shiv prakash Agarwal <chhotu.shiv@xxxxxxxxx> wrote: > Below is dmesg: > > [ 0.763264] PCI host bridge to bus 0000:00 > [ 0.763380] pci_bus 0000:00: root bus resource [mem 0x32100000-0x3fffffff] > [ 0.763694] pci_bus 0000:00: root bus resource [mem > 0x12100000-0x320fffff pref] > [ 0.764030] pci_bus 0000:00: root bus resource [io 0x1000-0xffff] > [ 0.764315] pci_bus 0000:00: No busn resource found for root bus, > will use [bus 00-ff] > [ 0.765085] PCI: bus0: Fast back to back transfers disabled > [ 0.765278] pci 0000:00:00.0: bridge configuration invalid ([bus > 00-00]), reconfiguring > [ 0.766292] PCI: bus1: Fast back to back transfers disabled > [ 0.766658] PCIE: arch_setup_msi_irq(2735) > [ 0.767168] pcieport 0000:00:00.0: Signaling PME through PCIe PME interrupt > [ 0.767413] pci 0000:01:00.0: Signaling PME through PCIe PME interrupt > [ 0.768101] pcieport 0000:00:00.0: BAR 9: assigned [mem > 0x12100000-0x121fffff 64bit pref] > [ 0.768405] pcieport 0000:00:00.0: PCI bridge to [bus 01] > [ 0.768683] pcieport 0000:00:00.0: bridge window [mem > 0x12100000-0x121fffff 64bit pref] I asked for the *complete* dmesg because it gives more useful information, such as the architecture (of course, you can elide proprietary things). But I try to avoid extra question/answer exchanges by requesting the easiest-to-collect, most complete information up front. Typically that's the complete dmesg log and complete "lspci -vv" output. It's also convention on these mailing lists to avoid top-posting. Bjorn > On Thu, May 8, 2014 at 9:37 PM, Bjorn Helgaas <bhelgaas@xxxxxxxxxx> wrote: >> [Your original mail was multipart and therefore didn't appear on the >> list; see http://vger.kernel.org/majordomo-info.html. Please use >> plain text for the mailing list.] >> >> On Thu, May 8, 2014 at 6:18 AM, shiv prakash Agarwal >> <chhotu.shiv@xxxxxxxxx> wrote: >>> Why some of bars are unassigned as below? >>> >>> I have one pcie card which needs only 16K pref memory but it remains >>> unassigned although host can provide around 512M pref memory. >>> >>> >>> Region 1: Memory at <unassigned> (32-bit, prefetchable) >> >> The PCI doesn't assign resources by itself. Most architectures call >> pci_assign_unassigned_resources() somewhere, which does. >> >> But I can't give any more clues without seeing a complete dmesg log >> and "lspci -vv" output. -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
