linux-kernel - RE: Linux 2.6.38.4 : regression due to x86 AMD ARAT feature

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: <EE335F95F28A664DB4A21289D2AA053BD1807651@SAUSEXMBP01.amd.com>
Date:	Sun, 24 Apr 2011 14:44:26 -0500
From:	"Ostrovsky, Boris" <Boris.Ostrovsky@....com>
To:	Jörg-Volker Peetz <peetz@...ato-kyma.net>,
	Borislav Petkov <bp@...en8.de>,
	Jörg-Volker Peetz <jvpeetz@....de>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"Herrmann3, Andreas" <Andreas.Herrmann3@....com>,
	"Petkov, Borislav" <Borislav.Petkov@....com>
Subject: RE: Linux 2.6.38.4 : regression due to x86 AMD ARAT feature

It's an old family fh processor (CPUID 00020F42) that cpu_has_amd_erratum(amd_erratum_400) apparently hasn't picked up. Since all fh processors are affected by this erratum I'll replace "c->x86 >= 0xf" with "c->x86 > 0xf" (and probably fix erratum formula anyway).

-boris
________________________________________
From: Jörg-Volker Peetz [peetz@...ato-kyma.net]
Sent: Sunday, April 24, 2011 2:55 PM
To: Borislav Petkov; Jörg-Volker Peetz; linux-kernel@...r.kernel.org; Ostrovsky, Boris; Herrmann3, Andreas; Petkov, Borislav
Subject: Re: Linux 2.6.38.4 : regression due to x86 AMD ARAT feature

Borislav Petkov wrote, on 04/24/11 20:20:
> On Sun, Apr 24, 2011 at 10:06:36AM +0200, Jörg-Volker Peetz wrote:
>> With 2.6.38.4 the clock slows down and the affected laptop sometimes, e.g. for
>> the output of "ps axuw" to appear, needs an input from the keyboard or the
>> touchpad. Please find attached the kernel config file and the output of dmesg.
>>
>> Commenting out the the lines
>>
>>      if (c->x86 >= 0xf && !cpu_has_amd_erratum(amd_erratum_400))
>>              set_cpu_cap(c, X86_FEATURE_ARAT);
>>
>> in arch/x86/kernel/cpu/amd.c makes this laptop work as before with kernel
>> 2.6.38.3. I think these lines came with commit
>> d7b7f0c401303dd7eafcff11752fe74d3b75c766.
>
> Hmm, interesting.
>
> Can you send also /proc/cpuinfo please?
>
> Looks like your CPU is a single core K8 revE or something but we'd need
> the exact family info.
>
> Thanks.
>

Here it comes:


processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 36
model name      : AMD Turion(tm) 64 Mobile Technology MT-40
stepping        : 2
cpu MHz         : 800.000
cache size      : 1024 KB
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36
clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt lm 3dnowext 3dnow rep_good
nopl pni lahf_lm
bogomips        : 1592.98
TLB size        : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp tm stc

This is a nice mobile processor and i thought it was sold quite a lot.
--
Best regards,
Jörg-Volker.


--
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/