linux-kernel - Re: [patch 4/4] x86: apic_is_clustered_box to indicate unsynched TSC's on multiboard vSMP systems

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]

Message-ID: <20080320190240.GB6085@localdomain>
Date:	Thu, 20 Mar 2008 12:02:40 -0700
From:	Ravikiran G Thirumalai <kiran@...lex86.org>
To:	Yinghai Lu <yhlu.kernel@...il.com>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Ingo Molnar <mingo@...e.hu>, linux-kernel@...r.kernel.org,
	Glauber de Oliveira Costa <gcosta@...hat.com>,
	Andi Kleen <ak@...e.de>, shai@...lex86.org
Subject: Re: [patch 4/4] x86: apic_is_clustered_box to indicate unsynched TSC's on multiboard vSMP systems

On Thu, Mar 20, 2008 at 12:53:25AM -0700, Yinghai Lu wrote:
>On Thu, Mar 20, 2008 at 12:45 AM, Ravikiran G Thirumalai
><kiran@...lex86.org> wrote:
>> Indicate TSCs are unreliable as time sources if the platform is
>>  a multi chassi ScaleMP vSMPowered machine.
>>
>>  Signed-off-by: Ravikiran Thirumalai <kiran@...lex86.org>
>>
>>  Index: linux.git.trees/arch/x86/kernel/apic_64.c
>>  ===================================================================
>>  --- linux.git.trees.orig/arch/x86/kernel/apic_64.c      2008-03-19 20:42:28.187659497 -0700
>>  +++ linux.git.trees/arch/x86/kernel/apic_64.c   2008-03-19 21:24:17.705515003 -0700
>>  @@ -1208,7 +1208,7 @@ __cpuinit int apic_is_clustered_box(void
>>          * will be [4, 0x23] or [8, 0x27] could be thought to
>>          * vsmp box still need checking...
>>          */
>>  -       if (!is_vsmp_box() && (boot_cpu_data.x86_vendor == X86_VENDOR_AMD))
>>  +       if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box())
>>                 return 0;
>
>why?
>
>>
>>         bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr;
>>  @@ -1248,6 +1248,12 @@ __cpuinit int apic_is_clustered_box(void
>>                         ++zeros;
>>         }
>>
>>  +       /* ScaleMP vSMPowered boxes have one cluster per board and TSCs are
>>  +        * not guaranteed to be synced between boards
>>  +        */
>>  +       if (is_vsmp_box() && clusters > 1)
>>  +               return 1;
>
>you still call that for intel box.

Yes.  The first call is to state that TSC's are synced if it is an  AMD box
and if it is _not_ a vSMPowered box (for the apic id lifting case on Sun
boxes), the second call is for _any_ vSMPowered system with more than one
board -- TSC's are not guaranteed to be synched in that case.  Both these
calls are needed.

Thanks,
Kiran
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/