Bug 972381
| Summary: | kernel panic when attach device to pcie switch | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Suqin Huang <shuang> |
| Component: | qemu-kvm | Assignee: | Radim Krčmář <rkrcmar> |
| Status: | CLOSED WONTFIX | QA Contact: | Virtualization Bugs <virt-bugs> |
| Severity: | low | Docs Contact: | |
| Priority: | medium | ||
| Version: | 7.0 | CC: | acathrow, chayang, juzhang, qiguo, rkrcmar, shuang, sluo, virt-maint, xfu, zhzhang |
| Target Milestone: | rc | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2013-08-26 14:21:18 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Suqin Huang
2013-06-09 03:16:30 UTC
PCIe specification does not allow direct connection of upstream port to the root hub (complex). We have to create root port and connect throught it: -M q35 -device ioh3420,bus=pcie.0,id=root.0 \ -device x3130-upstream,bus=root.0,id=upstream \ -device xio3130-downstream,bus=upstream,id=downstream,chassis=1 Upstream kernel is not happy with a check for misconfigured qemu, so it should be avoided/prevented it in userspace. (Qemu allows even more nonsensical topologies, where downstream port is not connected to upstream port.) Was this command generated by libvirt? (In reply to Radim Krčmář from comment #2) > PCIe specification does not allow direct connection of upstream port to the > root hub (complex). > > We have to create root port and connect throught it: > -M q35 -device ioh3420,bus=pcie.0,id=root.0 \ > -device x3130-upstream,bus=root.0,id=upstream \ > -device xio3130-downstream,bus=upstream,id=downstream,chassis=1 > Re-tested this issue by using the command you provided, hit the same panic ---snip commandline of mine--- /usr/libexec/qemu-kvm -M q35 -device ioh3420,bus=pcie.0,id=root.0 -device x3130-upstream,bus=root.0,addr=0x4,id=upstream -device xio3130-downstream,bus=upstream,id=downstream0,chassis=1 -drive file=/home/rhel7_switch.qcow,if=none,id=drive-system-disk,media=disk,format=qcow2,aio=native,werror=stop,rerror=stop -device virtio-blk-pci,bus=downstream0,drive=drive-system-disk,id=system-disk,bootindex=1 Hi Radim, Would you please have a look again? Any further testing, please let me know. > Upstream kernel is not happy with a check for misconfigured qemu, so it > should be avoided/prevented it in userspace. > (Qemu allows even more nonsensical topologies, where downstream port is not > connected to upstream port.) > > Was this command generated by libvirt? The kernel boots without "addr=0x4", or with "addr=0x0". Also the backtrace now goes through "pci_subsys_init" and not "acpi_init", so the problem is a bit different. How are the addresses chosen? Re-tested this issue without "addr=0x4", or with "addr=0x0". guest boot successfully and no kernel panic. About comment3, Is it a new issue? Do I need open a bug to track it? Upstream kernel decided to drop simple fix for this issue, hoping someone will rewrite aspm support instead. Modeling hardware configurations is a qemu feature, so we won't be fixing this. (I don't have enough information on source of these parameters to open new bugs) |