| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <57992d1f-434e-3900-1958-542cd830aad1@de.ibm.com>
Date: Mon, 4 May 2020 10:59:19 +0200
From: Christian Borntraeger <borntraeger@...ibm.com>
To: Qian Cai <cailca@...oud.com>, Pierre Morel <pmorel@...ux.ibm.com>,
Tony Krowiak <akrowiak@...ux.ibm.com>
Cc: Vasily Gorbik <gor@...ux.ibm.com>,
David Hildenbrand <david@...hat.com>,
Janosch Frank <frankja@...ux.ibm.com>, kvm@...r.kernel.org,
linux-s390@...r.kernel.org, LKML <linux-kernel@...r.kernel.org>
Subject: Re: s390 KVM warning in handle_pqap()
On 02.05.20 04:38, Qian Cai wrote:
> This line,
>
> if (WARN_ON_ONCE(fc != 0x03))
>
> qemu-kvm-2.12.0-99.module+el8.2.0+5827+8c39933c with this kernel config,
>
> https://raw.githubusercontent.com/cailca/linux-mm/master/s390.config
>
> # /usr/libexec/qemu-kvm -name ubuntu-18.04-server-cloudimg -cpu host -smp 2 -m 2G -hda ubuntu-18.04-server-cloudimg.qcow2 -cdrom ubuntu-18.04-server-cloudimg.iso -nic user,hostfwd=tcp::2222-:22 -nographic
>
> 00: [ 424.578896] WARNING: CPU: 0 PID: 1533 at arch/s390/kvm/priv.c:632 handle_
> 00: pqap+0x2b6/0x468 [kvm]
> 00: [ 424.578934] Modules linked in: kvm ip_tables x_tables xfs dasd_fba_mod da
> 00: sd_eckd_mod dm_mirror dm_region_hash dm_log dm_mod
> 00: [ 424.579026] CPU: 0 PID: 1533 Comm: qemu-kvm Not tainted 5.7.0-rc3-next-20
> 00: 200501 #2
> 00: [ 424.579064] Hardware name: IBM 2964 N96 400 (z/VM 6.4.0)
You run nested unter z/VM (under LPAR). So it looks like z/VM behaves different to
LPAR regarding the interception of the PQAP instruction.
Tony, can you talk to your z/VM colleagues about this? I guess we might need to
remove the WARN_ON_ONCE(fc != 0x03) and simply return EOPNOTSUPP if our assumption
is not right. I guess z/VM has its ECA field set to 0 so the effective ECA field
is also 0.
> 00: [ 424.579101] Krnl PSW : 0704d00180000000 000003ff80440dc2 (handle_pqap+0x2
> 00: ba/0x468 [kvm])
> 00: [ 424.579239] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:1 PM:0
> 00: RI:0 EA:3
> 00: [ 424.579282] Krnl GPRS: 0000000000000000 0000030000000000 0000030000000000
> 00: 00000000e1ca6148
> 00: [ 424.579320] 0000030000000000 000003ff80440c14 0000000000000000
> 00: 00000000822e8520
> 00: [ 424.579359] 00000000e1ca6000 000000009c79a000 00000000822e8008
> 00: 0000007c00877e70
> 00: [ 424.579399] 000003ff803f5000 000003ff80467528 000003ff80440c14
> 00: 000003e0043bf2c8
> 00: [ 424.579461] Krnl Code: 000003ff80440db6: a774ff5a brc 7,00
> 00: 0003ff80440c6a
> 00: [ 424.579461] 000003ff80440dba: a7f4ff54 brc 15,0
> 00: 00003ff80440c62
> 00: [ 424.579461] #000003ff80440dbe: af000000 mc 0,0
> 00: [ 424.579461] >000003ff80440dc2: a798ffa1 lhi %r9,
> 00: -95
> 00: [ 424.579461] 000003ff80440dc6: a51d0300 llihl %r1,
> 00: 768
> 00: [ 424.579461] 000003ff80440dca: b90800b1 agr %r11
> 00: ,%r1
> 00: [ 424.579461] 000003ff80440dce: d70bb000b000 xc 0(12
> 00: ,%r11),0(%r11)
> 00: [ 424.579461] 000003ff80440dd4: b9140029 lgfr %r2,
> 00: %r9
> 00: [ 424.586765] Call Trace:
> 00: [ 424.586894] [<000003ff80440dc2>] handle_pqap+0x2ba/0x468 [kvm]
> 00: [ 424.587026] [<000003ff80446fa6>] kvm_s390_handle_b2+0x2f6/0x950 [kvm]
> 00: [ 424.587156] [<000003ff8042d74c>] kvm_handle_sie_intercept+0x154/0x1db0 [
> 00: kvm]
> 00: [ 424.587287] [<000003ff80426950>] __vcpu_run+0x1040/0x2150 [kvm]
> 00: [ 424.587414] [<000003ff8042941a>] kvm_arch_vcpu_ioctl_run+0x5fa/0x1338 [k
> 00: vm]
> 00: [ 424.587540] [<000003ff8040195e>] kvm_vcpu_ioctl+0x346/0xa10 [kvm]
> 00: [ 424.587590] [<00000001433fbd16>] ksys_ioctl+0x276/0xbb8
> 00: [ 424.587630] [<00000001433fc682>] __s390x_sys_ioctl+0x2a/0x38
> 00: [ 424.587674] [<000000014393c880>] system_call+0xd8/0x2b4
> 00: [ 424.587715] 2 locks held by qemu-kvm/1533:
> 00: [ 424.587748] #0: 00000000822e80d0 (&vcpu->mutex){+.+.}-{3:3}, at: kvm_vcp
> 00: u_ioctl+0x170/0xa10 [kvm]
> 00: [ 424.587898] #1: 0000000081fe3980 (&kvm->srcu){....}-{0:0}, at: __vcpu_ru
> 00: n+0x60a/0x2150 [kvm]
> 00: [ 424.588045] Last Breaking-Event-Address:
> 00: [ 424.588169] [<000003ff80440c1e>] handle_pqap+0x116/0x468 [kvm]
> 00: [ 424.588204] irq event stamp: 23141
> 00: [ 424.588246] hardirqs last enabled at (23149): [<000000014308f3de>] conso
> 00: le_unlock+0x766/0xa20
> 00: [ 424.588287] hardirqs last disabled at (23156): [<000000014308ee40>] conso
> 00: le_unlock+0x1c8/0xa20
> 00: [ 424.588536] softirqs last enabled at (22998): [<000000014393e162>] __do_
> 00: softirq+0x6e2/0xa48
> 00: [ 424.588583] softirqs last disabled at (22983): [<0000000142f652dc>] do_so
> 00: ftirq_own_stack+0xe4/0x100
> 00: [ 424.588625] ---[ end trace e420441aa7c001ac ]---
>
Powered by blists - more mailing lists