Re: "BUG: using smp_processor_id() in preemptible" with KPTI on 4.14.11

[Thomas Gleixner]

On Thu, 4 Jan 2018, Thomas Zeitlhofer wrote:
> On Thu, Jan 04, 2018 at 01:55:28PM +0100, Greg Kroah-Hartman wrote:
> > > > > > on an Ivybridge CPU, I get with 4.14.11:
> > > > > > 
> > > > > >    BUG: using smp_processor_id() in preemptible [00000000] code: ovsdb-server/4510
> > > > > >    caller is native_flush_tlb_single+0x57/0xc0
> > > > > >    CPU: 3 PID: 4510 Comm: ovsdb-server Not tainted 4.14.11-kvm-00434-gcd0b8eb84f5c #3
> > > > > >    Hardware name: MSI MS-7798/B75MA-P45 (MS-7798), BIOS V1.9 09/30/2013
> > > > > >    Call Trace:
> > > > > >     dump_stack+0x5c/0x86
> > > > > >     check_preemption_disabled+0xdd/0xe0
> > > > > >     native_flush_tlb_single+0x57/0xc0
> > > > > >     ? __set_pte_vaddr+0x2d/0x40
> > > > > >     __set_pte_vaddr+0x2d/0x40
> > > > > >     set_pte_vaddr+0x2f/0x40
> > > > > >     cea_set_pte+0x30/0x40
> > > > > >     ds_update_cea.constprop.4+0x4d/0x70
> > > > > >     reserve_ds_buffers+0x159/0x410
> > > > > >     ? wp_page_copy+0x36d/0x6a0
> > > > > >     x86_reserve_hardware+0x150/0x160
> > > > > >     x86_pmu_event_init+0x3e/0x1f0
> > > > > >     perf_try_init_event+0x69/0x80
> > > > > >     perf_event_alloc+0x652/0x740
> > > > > >     SyS_perf_event_open+0x3f6/0xd60
> > > > > >     do_syscall_64+0x5c/0x190
> > > > > >     entry_SYSCALL64_slow_path+0x25/0x25
> > > > > >    RIP: 0033:0x74a1d94580b9
> > > > > >    RSP: 002b:00007fff0c01d5d8 EFLAGS: 00000206 ORIG_RAX: 000000000000012a
> > > > > >    RAX: ffffffffffffffda RBX: 00007fff0c01d7b0 RCX: 000074a1d94580b9
> > > > > >    RDX: 00000000ffffffff RSI: 0000000000000000 RDI: 00007fff0c01d5e0
> > > > > >    RBP: 0000000000000000 R08: 0000000000000000 R09: 0000007000000000
> > > > > >    R10: 00000000ffffffff R11: 0000000000000206 R12: 0000000000000008
> > > > > >    R13: 0000000000000000 R14: 00007fff0c01d790 R15: 00005df43a799600
> > > > > > 
> > > > > > This does not show up when booting with pti=off.

Right, because the code path is not invoked ....

> > Odd, does 4.15-rc6 also trigger the same error? 
> 
> Yes:
> 
>    BUG: using smp_processor_id() in preemptible [00000000] code: ovsdb-server/4498
>    caller is native_flush_tlb_single+0x57/0xc0
>    CPU: 2 PID: 4498 Comm: ovsdb-server Not tainted 4.15.0-rc6-kvm-00423-gea1908c252eb #3
>    Hardware name: MSI MS-7798/B75MA-P45 (MS-7798), BIOS V1.9 09/30/2013
>    Call Trace:
>     dump_stack+0x5c/0x86
>     check_preemption_disabled+0xdd/0xe0
>     native_flush_tlb_single+0x57/0xc0
>     ? __set_pte_vaddr+0x2d/0x40
>     __set_pte_vaddr+0x2d/0x40
>     set_pte_vaddr+0x2f/0x40
>     cea_set_pte+0x30/0x40
>     ds_update_cea.constprop.4+0x4d/0x70
>     reserve_ds_buffers+0x159/0x410
>     ? wp_page_copy+0x370/0x6c0
>     x86_reserve_hardware+0x150/0x160
>     x86_pmu_event_init+0x3e/0x1f0
>     perf_try_init_event+0x69/0x80
>     perf_event_alloc+0x652/0x740
>     SyS_perf_event_open+0x3f6/0xd60
>     do_syscall_64+0x5c/0x190
>     entry_SYSCALL64_slow_path+0x25/0x25
>    RIP: 0033:0x72bff0a3c0b9
>    RSP: 002b:00007ffed11c2f18 EFLAGS: 00000206 ORIG_RAX: 000000000000012a
>    RAX: ffffffffffffffda RBX: 00007ffed11c30f0 RCX: 000072bff0a3c0b9
>    RDX: 00000000ffffffff RSI: 0000000000000000 RDI: 00007ffed11c2f20
>    RBP: 0000000000000000 R08: 0000000000000000 R09: 0000007000000000
>    R10: 00000000ffffffff R11: 0000000000000206 R12: 0000000000000008
>    R13: 0000000000000000 R14: 00007ffed11c30d0 R15: 000060986ecfb600
>    device ovs-system entered promiscuous mode
>    netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
> 
> In addition, with v4.15-rc6, netlink messages like in the last line show
> up, but I guess this is a different openvswitch related issue.
> 
> > Thomas is working on an
> > issue with KALSR (see lkml with:
> > 	Subject: Re: "bad pmd" errors + oops with KPTI on 4.14.11 after loading X.509 certs
> > )
> 
> Yes, I have also seen that thread, but I did not see any similarities to
> my issue. Anyway, I also tried out the patch proposed in
> https://lkml.org/lkml/2018/1/4/313 but it does not change anything here.

Correct. I'm looking into a fix. Stay tuned.

Thanks,

	tglx