linux-kernel - Re: This kernel requires the following features not present on the CPU... (on a VIA C7 CPU)

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] [day] [month] [year] [list]

Date:	Tue, 22 May 2007 17:53:11 +0200
From:	Christian Volkmann <haveaniceday@...sv.de>
To:	Claas Langbehn <claas@...tdir.de>
CC:	Artur Kedzierski <artur@...zierski.org>,
	linux-kernel@...r.kernel.org
Subject: Re: This kernel requires the following features not present on the
 CPU... (on a VIA C7 CPU)

Claas Langbehn wrote:
> Could anyone explain to me what CMPXCHG64 / cx8 is and what happens if
> the kernel has been compiled to use it but the CPU does not have it?
> 
> Regards,
> claas
> 

Hi Claas,

the bug is at another place. But it's hidden if X86_CMPXCHG64 is
switched of.

There are 3 errors.

First: Via is capable of CMPXCHG64, but switch the related CPU bit off
to work with old Win NT versions. ( So the NT error is the fourth... but who cares
about that error. ;-) )

Second:  verify_cpu.S
This checks during the boot the capability of the CPU. It expected
the related bit, but... see first.
This error is switched of if X86_CMPXCHG64 is disabled.

Third: setup.S "PANIC: CPU too old for this kernel." is not printed
cause of a missing set register in setup.S.

As I have learned the NX-bit of the C7 is a BIOS issue.

The only real required patch is below. It enables the right bits for the C3/C7.

The first error is out of scope, and the fix for the "CPU too old.." is for
me just a cosmetic print out in case of boot errors.

I suppose .rc3 should contain a fix for this bug.

Best regards,

Christian

--- linux.orig/arch/i386/kernel/verify_cpu.S
+++ linux/arch/i386/kernel/verify_cpu.S
@@ -2,6 +2,7 @@
    This runs in 16bit mode so that the caller can still use the BIOS
    to output errors on the screen */
 #include <asm/cpufeature.h>
+#include <asm/msr.h>

 verify_cpu:
        pushfl                          # Save caller passed flags
@@ -45,6 +46,28 @@
        cmpl    $0x1,%eax
        jb      bad                     # no cpuid 1

+#if REQUIRED_MASK1 & NEED_CMPXCHG64
+       /* Some VIA C3s need magic MSRs to enable CX64. Do this here */
+       cmpl    $0x746e6543,%ebx        # Cent
+       jne     1f
+       cmpl    $0x48727561,%edx        # aurH
+       jne     1f
+       cmpl    $0x736c7561,%ecx        # auls
+       jne     1f
+       movl    $1,%eax                 # check model
+       cpuid
+       shr     $4,%eax
+       andl    $0xf,%eax               # get model
+       cmpl    $6,%eax
+       jb      1f
+       cmpl    $10,%eax                 # newer vias hopefully don't require
+       ja      1f                      # this anymore
+       movl    $MSR_VIA_FCR,%ecx
+       rdmsr
+       orl     $((1<<1)|(1<<7)),%eax   # enable CMPXCHG64 and PGE
+       wrmsr
+1:
+#endif
        movl    $0x1,%eax               # Does the cpu have what it takes
        cpuid
-
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/