linux-kernel - Re: [PATCH] x86/cpuid: Deal with broken firmware once more

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 13 Nov 2016 01:05:11 +0300
From:   "M. Vefa Bicakci" <m.v.b@...box.com>
To:     "Charles (Chas) Williams" <ciwillia@...cade.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        "x86@...nel.org" <x86@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Borislav Petkov <bp@...en8.de>
Subject: Re: [PATCH] x86/cpuid: Deal with broken firmware once more

On 11/10/2016 01:50 PM, Charles (Chas) Williams wrote:
> 
> 
> On 11/09/2016 10:57 PM, M. Vefa Bicakci wrote:
>> [    0.002000] mvb: CPU: Physical Processor ID: 0
>> [    0.002000] mvb: CPU: Processor Core ID: 0
>> [    0.002000] mvb: identify_cpu:1112: c: ffff880013b0a040,
>> c->logical_proc_id: 65535
>> [    0.002000] mvb: __default_cpu_present_to_apicid:612: Returning
>> 65535! mps_cpu: 1, nr_cpu_ids: 2, cpu_present(mps_cpu): 1
>> [    0.002000] smpboot: mvb: topology_update_package_map:270: cpu: 1,
>> pkg: 4095
>> [    0.002000] smpboot: APIC(ffff) Converting physical 4095 to logical
>> package 0
>> [    0.002000] smpboot: mvb: topology_update_package_map:305: cpu: 1,
>> cpu_data(cpu).logical_proc_id: 0
> 
> This seems strange.  0xffff is BAD_APICID.  Why didn't this fail here:
> 
>         for_each_present_cpu(cpu) {
>                 unsigned int apicid = apic->cpu_present_to_apicid(cpu);
> 
>                 if (apicid == BAD_APICID ||
> !apic->apic_id_valid(apicid))  <<<<<<<<<<
>                         continue;
>                 if (!topology_update_package_map(apicid, cpu))
>                         continue;
> 
> topology_update_package_map() should never have been called?

(Sorry for the delay!)

It appears that this is a different call path than the one in
smp_init_package_map function. I *think* the following call path is the
one shown in the dmesg, but I am not sure:

cpu_bringup_and_idle
  cpu_bringup (arch/x86/xen/smp.c)
    smp_store_cpu_info (this call path branch is included for context)
      identify_secondary_cpu
        identify_cpu
          detect_ht
    topology_update_package_map

Sorry about the potentially misleading dmesg excerpt I posted.

Vefa