[gmx-developers] gromacs on macbook pro (intel chip) ... or not?

[David van der Spoel]

Tom Joseph wrote:
> It looks like the Apple-modified version of the GNU assembler (part of 
> binutils) doesn't support the well-known Intel syntax for assembly 
> language.  For some reason the binutils team decided they wanted to use 
> a different source code syntax than the entire rest of the world.  The 
> instruction "mov eax, ebx" in Intel syntax is "mov.l %ebx, %eax" in GNU 
> syntax.  The problem you observe would also arise with any compiler or 
> OS for x86 that doesn't use an assembler that supports this syntax.
> 
> The easiest solutions that I can think of are 1) disable assembly loops 
> and take the speed hit or 2) build your own copies of vanilla binutils 
> and gcc and then use those to build GROMACS.

Erik has already added separate assembly loops for Mac/Intel. It will be 
in 3.3.1.


> 
> --Tom
> 
> On Mar 29, 2006, at 6:46 PM, Arnie Hagler wrote:
> 
>> I have recently attempted to install gromacs on the newly released 
>> Macbook pro (w/ intel chips). I am having a problem w/ binutils. From 
>> the config.log file:
>>
>> configure:26987: checking whether your compiler can handle assembly 
>> files (*.s)
>> configure:27006: cc -c conftestasm.s
>> configure:27009: $? = 0
>> configure:27012: result: yes
>> configure:27031: checking whether as understands 'lock cmpxchg'
>> configure:27051: cc -c conftest.s
>> conftest.s:1:Unknown pseudo-op: .intel_syntax
>> conftest.s:1:Rest of line ignored. 1st junk character valued 110 (n).
>> conftest.s:3:invalid character '[' in first operand
>> configure:27054: $? = 1
>> configure:27059: result: no
>> configure:27061: error: Upgrade to binutils >= 2.12.1 (or disable 
>> assembly loops).
>>
>> corresponding I believe to the code in configure:
>>
>> echo "$as_me:$LINENO: checking whether as understands 'lock cmpxchg'" >&5
>> echo $ECHO_N "checking whether as understands 'lock cmpxchg'... 
>> $ECHO_C" >&6
>>   if test "$enable_x86_64_sse" = "yes"; then
>>     cat > conftest.s << EOF
>> .intel_syntax noprefix
>> checkasm:
>>         lock cmpxchg [rsi], ebx
>>         emms
>>         ret
>> EOF
>>   else
>>     cat > conftest.s << EOF
>> .intel_syntax noprefix
>> checkasm:
>>         lock cmpxchg [esi], ebx
>>         emms
>>         ret
>> EOF
>>   fi
>>
>> I doubt that the problem is an old version of binutils on this new 
>> platform and would prefer not to disable assembly loops.
>>
>> Any advice would be greatly appreciated.
>>
>> Thanks
>>
>>
>> ------------------------------------------------------------------------
>> New Yahoo! Messenger with Voice. 
>> <http://us.rd.yahoo.com/mail_us/taglines/postman4/*http://us.rd.yahoo.com/evt=39666/*http://beta.messenger.yahoo.com> 
>> Call regular phones from your PC for low, low rates.
>> _______________________________________________
>> gmx-developers mailing list
>> gmx-developers at gromacs.org <mailto:gmx-developers at gromacs.org>
>> http://www.gromacs.org/mailman/listinfo/gmx-developers
>> Please don't post (un)subscribe requests to the list. Use the 
>> www interface or send it to gmx-developers-request at gromacs.org 
>> <mailto:gmx-developers-request at gromacs.org>.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-developers-request at gromacs.org.


-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++