Description of problem: Boot guest with >=21 xhci controler, seabios will give info "No bootable device". Then sendkey ctrl-alt-delete to reboot qemu-kvm. But qemu-kvm will hang at booting seabios. Version-Release number of selected component (if applicable): seabios-bin-1.7.2.2-10.el7.x86_64 qemu-kvm-1.5.3-41.el7.x86_64 3.10.0-79.el7.x86_64 How reproducible: 100% Steps to Reproduce: 1.Boot guest with >=21 xhci controler via the following shell script. # sh 4230878-capability-of-xhci-attached-to-switch.sh 21 # cat 4230878-capability-of-xhci-attached-to-switch.sh #! /bin/sh CLI="/usr/libexec/qemu-kvm -S -M q35 -m 4G -smp 4,socket=2,core=2,thread=1 \ -device virtio-scsi-pci,bus=pcie.0,id=scsi,addr=0x6 \ -drive file=/home/juli/rhel7.0.qcow2,format=qcow2,if=none,media=disk,id=aa \ -device scsi-hd,drive=aa,id=sys-image,bootindex=0 \ -boot menu=on,strict=on \ -monitor stdio \ -serial unix:/tmp/ttyS0,server,nowait \ -spice port=5931,disable-ticketing \ -vga qxl -global qxl-vga.revision=3 \ -chardev stdio,id=log \ -device isa-debugcon,iobase=0x402,chardev=log" CLI="$CLI -device ioh3420,bus=pcie.0,id=root.0 " CLI="$CLI -device x3130-upstream,bus=root.0,id=upstream1,multifunction=on " CLI="$CLI -device xio3130-downstream,bus=upstream1,id=downstream1,chassis=1 " NUM_DOWNSTREAM=${1:-255} for ((i=1,i_downstream=1;i<=$NUM_DOWNSTREAM;i++,i_downstream++)); do #m=$((i/7)) #n=$((i%7+1)) #m=`printf "%02x" $m` #n=`printf "%02x" $n` #echo $i_downstream CLI="$CLI -device nec-usb-xhci,bus=downstream1,id=usb_controller$i" CLI="$CLI -drive file=/home/juli/usb-storages/usb$i,if=none,format=qcow2,id=aa-usb$i,media=disk " CLI="$CLI -device usb-storage,drive=aa-usb$i,bus=usb_controller$i.0,id=device-usb$i" done $CLI 2.sendkey ctrl-alt-delete via HMP when seabios give info "No bootable device". (qemu) sendkey ctrl-alt-delete 3. Actual results: (qemu) sendkey ctrl-alt-delete (qemu) In resume (status=0) In 32bit resume Attempting a hard reboot xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! Start bios (version seabios-1.7.2.2-10.el7) Ram Size=0xb0000000 (0x0000000050000000 high) Relocating low data from 0x000e4140 to 0x000ef780 (size 2161) Relocating init from 0x000e49b1 to 0xaffe2e30 (size 53415) Expected results: No seabios hang. Can see the seabios info "No bootable device" again. Additional info:
please restest with seabios 1.7.5 builds: http://people.redhat.com/ghoffman/bz1101500/
> please restest with seabios 1.7.5 builds: > http://people.redhat.com/ghoffman/bz1101500/ Hi Gerd, I just do following retestings. --- 1> Retest with # rpm -qa|grep seabios seabios-bin-1.7.5-1.el7_0.bz1101500.3.x86_64 seabios-1.7.5-1.el7_0.bz1101500.3.x86_64 # rpm -qa|grep qemu qemu-kvm-1.5.3-65.el7.x86_64 Result: This bz has gone. After comment #0 step 2, qemu-kvm can be reboot again. --- 2> Retest with # rpm -qa|grep seabios seabios-1.7.2.2-12.el7.x86_64 seabios-bin-1.7.2.2-12.el7.x86_64 # rpm -qa|grep qemu qemu-kvm-1.5.3-65.el7.x86_64 Result: This bz has gone. After comment #0 step 2, qemu-kvm can be reboot again. --- 3> Retest with # rpm -qa|grep seabios seabios-1.7.2.2-10.el7.x86_64 seabios-bin-1.7.2.2-10.el7.x86_64 # rpm -qa|grep qemu qemu-kvm-1.5.3-65.el7.x86_64 Result: This bz can be reproduced. ====== FYI, I just do the above retestings. Using the newest qemu-kvm this bz has gone. Any further testing needed, free to update it in the bz. Best Regards, Jun Li
reproduce this bug with seabios-1.7.2.2-10.el7.x86_64 & qemu-kvm-1.5.3-60.el7.x86_64. according to comment0's steps. result: 1.HMP output No bootable device. (qemu) sendkey alt-ctrl-delete (qemu) In resume (status=0) In 32bit resume Attempting a hard reboot xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! xhci: reset while running! Start bios (version seabios-1.7.2.2-10.el7) Ram Size=0x80000000 (0x0000000080000000 high) Relocating low data from 0x000e4140 to 0x000ef780 (size 2161) Relocating init from 0x000e49b1 to 0x7ffe2e30 (size 53415) 2. guest show black screen verify this bug with seabios-1.7.5-1.el7.x86_64. according to comment 0. result: 1. Guest boot successfully. 2. sendkey alt-ctrl-delete works well so. issue has gone.
Based on C#6, this issue has been resolved in seabios-1.7.5-1.el7.x86_64 Moving to VERIFIED.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-0345.html