gcl-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Gcl-devel] Re: ACL2 with x86-64 Debian Lenny GCL


From: Camm Maguire
Subject: [Gcl-devel] Re: ACL2 with x86-64 Debian Lenny GCL
Date: Thu, 06 Aug 2009 11:23:48 -0400
User-agent: Gnus/5.11 (Gnus v5.11) Emacs/22.2 (gnu/linux)

Greetings!  It appears that you have a flavor of security protection
on your system which is mandating that the sbrk system call is random,
despite GCL's attempt to instruct the kernel to make it non-random.
GCL attemtpts to accomplish this via the ADDR_NO_RANDOMIZE argument to
the personality() system call.  Your kernel may not be honoring this.
>From the GCL configure log at the failure point you indicate:

        AC_MSG_ERROR([Cannot build with randomized sbrk. Your options:
         - upgrade to a kernel/libc that knows about 
personality(ADDR_NO_RANDOMIZE)"
         - recompile your kernel with CONFIG_COMPAT_BRK (if it has that option)"
         - run sysctl kernel.randomize_va_space=0 before using gcl])

I'd suggest trying the sysctl at this point.  All GCL apps have the
same attempted workaround code built in.  

Please let me know if problems persist here.

Take care,

Peter Dillinger <address@hidden> writes:

> I recently updated my work machine from Debian Etch to Lenny and I
> think that may have introduced a GCL problem with ACL2.  I built ACL2
> 3.5 using the GCL 2.6.7 build that comes with the AMD64 Lenny, and I
> get a
>
> Error: Caught fatal error [memory may be damaged]
> Fast links are on: do (si::use-fast-links nil) for debugging
> Error signalled by ACL2_*1*_ACL2::THM-FN.
> Backtrace: funcall > system:top-level > lisp:lambda-closure > lp >
> acl2_*1*_acl2::ld-fn > ld-fn > acl2_*1*_acl2::thm-fn > thm-fn >
> system:universal-error-handler > system::break-level-for-acl2 > let* >
> UNLESS
>
> if I do this:
>
> - at the first prompt, type Ctrl+C
> - :q
> - :mini-proveall
>
> I can also get it (sometimes) by interrupting a :mini-proveall,
> doing a :q, :ubt 1, and another :mini-proveall.
>
> I seem to recall this was more reliable under Etch, but maybe i'm
> wrong.  What's up?
>
> Also, I haven't figured out how to build GCL myself due to
>
>  checking for randomized sbrk... yes
>  checking for randomized brk remedy... no
>  Cannot build with randomized sbrk
>
> Thanks for any info.
>
> -- 
> Peter Dillinger
> address@hidden
>
>
>
>

-- 
Camm Maguire                                        address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah




reply via email to

[Prev in Thread] Current Thread [Next in Thread]