[Bug 260049] vtnet modern/1.x panic - data segment exception trap

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=260049

            Bug ID: 260049
           Summary: vtnet modern/1.x panic - data segment exception trap
           Product: Base System
           Version: CURRENT
          Hardware: Any
                OS: Any
            Status: New
          Severity: Affects Many People
          Priority: ---
         Component: kern
          Assignee: bugs@FreeBSD.org
          Reporter: alfredo@freebsd.org

I consistently get fatal kernel trap when sending files using two simultaneous
SCP instances [2] to a powerpc64 QEMU [1] KVM accelerated VM in a Linux host. 

The issue is reproducible with vtnet modern/1.x mode. With vtnet legacy/pre-1.0
mode it works fine:



fatal kernel trap:

   exception       = 0x380 (data segment exception)
   virtual address = 0xdeadc0dedeadc0e8
   dsisr           = 0x0
   srr0            = 0xc0000000006047f8 (0x6047f8)
   srr1            = 0x8000000000009032
   current msr     = 0x8000000000009032
   lr              = 0xc000000000604798 (0x604798)
   frame           = 0xc008000035d57250
   curthread       = 0xc008000035973080
          pid = 12, comm = irq4608:-irtio_pci1

panic: data segment exception trap
cpuid = 4
time = 1637888248
KDB: stack backtrace:
0xc008000035d56e70: at kdb_backtrace+0x60
0xc008000035d56f80: at vpanic+0x1f0
0xc008000035d57030: at panic+0x44
0xc008000035d57060: at trap+0x324
0xc008000035d57190: at powerpc_interrupt+0x1cc
0xc008000035d57220: kernel DSE trap @ 0xdeadc0dedeadc0e8 by
vtnet_rxq_eof+0x174: srr1=0x8000000000009032
            r1=0xc008000035d574d0 cr=0x44c00488 xer=0x20040000 ctr=0x1
r2=0xc0000000013e9000 frame=0xc008000035d57250
0xc008000035d574d0: at vtnet_rxq_eof+0x110
0xc008000035d57600: at vtnet_rx_vq_process+0xe0
0xc008000035d57660: at virtqueue_intr+0x2c
0xc008000035d57690: at vtpci_intx_intr+0x11c
0xc008000035d576d0: at ithread_loop+0x3b8
0xc008000035d57820: at fork_exit+0xc4
0xc008000035d578c0: at fork_trampoline+0x18
0xc008000035d578f0: at -0x4
KDB: enter: panic
[ thread pid 12 tid 100077 ]
Stopped at      kdb_enter+0x78: ori     r0, r0, 0x0
db> 


------

[1] # qemu-system-ppc64 -name guest=alfredo1 -smp 8 -m 8g -drive
file=vdisks/vm1-disk1.qcow2,format=qcow2,if=virtio -device
virtio-net-pci,netdev=br0,disable-legacy=on,disable-modern=off,mac=52:54:a1:f4:ed:01
-netdev bridge,id=br0 -enable-kvm -machine
pseries,accel=kvm,cap-hpt-max-page-size=2M -mem-prealloc -mem-path
/dev/hugepages -vnc :62 -vga none -nographic -snapshot



$ qemu-system-ppc64 --version
QEMU emulator version 6.0.0
Copyright (c) 2003-2021 Fabrice Bellard and the QEMU Project developers

[2] SSH copy command:

sshpass -p "<pass>" scp <2GB_FILE> root@<IP>:d.iso & sshpass -p "pass" scp
<2GB_FILE> root@<IP>:c.iso &

-- 
You are receiving this mail because:
You are the assignee for the bug.