Ian,
I cannot reproduce that bug on my x86_64 CentOS 5.4 system:
dikay@turn29:-/tmp% cat temp.f90
CHARACTER A(2)*2
DATA A/2*'.A'/
CALL CA(A)
STOP 1
END
SUBROUTINE CA(A)
CHARACTER A(2)*2
DO 1 I=1,2
!#### With given data, the condition "A(I)(2:).NE.'A'" is always FALSE,
!#### so we should never execute the 'STOP 1' here:
1 IF(A(I)(2:).NE.'A' .AND. A(I).NE.'..') RETURN
!#### Should always drop through to here:
STOP 2
END
dikay@turn29:-/tmp% ifort temp.f90
dikay@turn29:-/tmp% ./a.out
2
dikay@turn29:-/tmp% file a.out
a.out: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for
GNU/Linux 2.6.9, dynamically linked (uses shared libs), for GNU/Linux
2.6.9, not stripped
dikay@turn29:-/tmp% ifort -m32 temp.f90
dikay@turn29:-/tmp% ./a.out
2
dikay@turn29:-/tmp% file a.out
a.out: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for
GNU/Linux 2.6.9, dynamically linked (uses shared libs), for GNU/Linux
2.6.9, not stripped
dikay@turn29:-/tmp% ifort -v
Version 11.1
dikay@turn29:-/tmp% ifort --vv
ifort: command line warning #10006: ignoring unknown option '-fvv'
ifort: command line error: no files specified; for help type "ifort -help"
dikay@turn29:-/tmp% ifort -V
Intel(R) Fortran Intel(R) 64 Compiler Professional for applications
running on Intel(R) 64, Version 11.1 Build 20091012 Package ID:
l_cprof_p_11.1.059
Copyright (C) 1985-2009 Intel Corporation. All rights reserved.
FOR NON-COMMERCIAL USE ONLY
dikay@turn29:-/tmp% uname -a
Linux turn29.biologie.uni-konstanz.de 2.6.18-164.6.1.el5 #1 SMP Tue Nov
3 16:12:36 EST 2009 x86_64 x86_64 x86_64 GNU/Linux
best,
Kay
Ian Tickle schrieb:
> Before you all go away and install the latest Intel compilers to fix
> this bug - don't bother! We just installed the latest version (ifort
> 11.3) and it has the same bug! We plan to file a bug report with Intel.
>
> In case anyone is interested, this is the code snippet which exhibits
> the buggy behaviour (this was pruned down from s/r CHK_ALT in Refmac):
>
> CHARACTER A(2)*2
> DATA A/2*'.A'/
> CALL CA(A)
> STOP 1
> END
>
> SUBROUTINE CA(A)
> CHARACTER A(2)*2
> DO 1 I=2
> C#### With given data, the condition "A(I)(2:).NE.'A'" is always FALSE,
> C#### so we should never execute the 'STOP 1' here:
> 1 IF(A(I)(2:).NE.'A' .AND. A(I).NE.'..') RETURN
> C#### Should always drop through to here:
> STOP 2
> END
>
> It fails using -O2, -O3 and -Os, i.e. it only gives the right answer
> using -O0 & -O1.
>
> I should say that this is the first time we've ever encountered a bug in
> the Intel compilers - I wish I could say the same for g77 & gfortran!
>
> Cheers
>
> -- Ian
>
>> -----Original Message-----
>> From: [log in to unmask]
>> [mailto:[log in to unmask]] On Behalf Of Ian Tickle
>> Sent: 28 November 2009 13:22
>> To: Garib Murshudov; [log in to unmask]
>> Cc: [log in to unmask]
>> Subject: RE: [ccp4bb] Bad geometry for alt. conformation
>> refined in Refmac5
>>
>> I think I've finally squashed this particular bug which had been
>> annoying me for quite some time! It appears that Intel Fortran v11.0
>> has an optimisation bug in s/r CHK_ALT (i.e. checking alternate atom
>> codes), & possibly also 10.x - I don't have those to test -
>> though v9.1
>> & possibly earlier seem to be bug-free (at least this bug anyway!).
>> Possibly this is fixed in the current version of ifort (11.1), again I
>> don't have it to test, maybe someone who does can try it.
>> This affects
>> compilations using optimisation levels -O2, -Os and -O3, but
>> -O0 and -O1
>> are unaffected.
>>
>> So my solution is to recompile with -O1. The CCP4 6.1.2 distribution
>> uses -O2 so you need either to fix the makefile, or download Garib's
>> source code and make sure the makefile you use has the
>> appropriate flag
>> settings.
>>
>> As was pointed out earlier, having to use the
>> gfortran-compiled code is
>> an issue, because at least for the OS I'm using (Centos 4.6),
>> gfortran-compiled code runs ~ 300% slower than the ifort code!!
>>
>> Hope this helps!
>>
>> -- Ian
>>
>>> -----Original Message-----
>>> From: [log in to unmask]
>>> [mailto:[log in to unmask]] On Behalf Of Garib Murshudov
>>> Sent: 25 November 2009 16:27
>>> To: [log in to unmask]
>>> Cc: [log in to unmask]
>>> Subject: Re: [ccp4bb] Bad geometry for alt. conformation
>>> refined in Refmac5
>>>
>>> Could you please try the version from York:
>>>
>>> www.ysbl.york.ac.uk/refmac/latest_refmac.html
>>>
>>> I think probkem you mention is related with compilation or
>>> something.
>>> At least I cannot repat it on my computer
>>>
>>> regards
>>> Garib
>>>
>>> On 25 Nov 2009, at 16:18, John Pascal wrote:
>>>
>>>> Hello All,
>>>>
>>>> We are trying to refine ARG residues with two conformations in
>>>> Refmac5, and the
>>>> refined atom positions in the output PDB file are all over the
>>>> place, as if the
>>>> geometry restraints are not well defined.
>>>>
>>>> We've tried several different formats for the input file,
>> based on
>>>> previous
>>>> postings to the bulletin board and the PDB standard (two
>> examples
>>>> below), but
>>>> the result is always the same.
>>>>
>>>> We are using Refmac5 in CCP4 Suite 6.1.2, GUI 2.0.5 on Mac OSX.
>>>>
>>>> We'd appreciate any suggestions. Thanks. -John
>>>>
>>>> Examples of ARG format:
>>>>
>>>> 1)
>>>> ATOM 1472 N ARG A 1 -5.737 26.887 38.372 1.00
>>>> 29.53 C N
>>>> ATOM 1473 CA ARG A 1 -5.445 25.560 37.882 1.00
>>>> 30.24 C C
>>>> ATOM 1474 CB ARG A 1 -5.314 24.548 39.036 1.00
>>>> 30.63 C C
>>>> ATOM 1475 CG ARG A 1 -5.426 23.052 38.627 1.00
>>>> 34.81 C C
>>>> ATOM 1476 CD AARG A 1 -4.827 22.075 39.644 0.50
>>>> 37.09 C C
>>>> ATOM 1477 CD BARG A 1 -4.301 22.419 39.279 0.50
>>>> 37.09 C C
>>>> ATOM 1478 NE AARG A 1 -3.430 21.777 39.304 0.50
>>>> 42.71 C N
>>>> ATOM 1479 NE BARG A 1 -4.482 21.902 40.627 0.50
>>>> 42.71 C N
>>>> ATOM 1480 CZ AARG A 1 -2.998 20.868 38.402 0.50
>>>> 44.91 C C
>>>> ATOM 1481 CZ BARG A 1 -3.648 22.142 41.638 0.50
>>>> 44.91 C C
>>>> ATOM 1482 NH1AARG A 1 -3.841 20.117 37.678 0.50
>>>> 45.20 C N
>>>> ATOM 1483 NH1BARG A 1 -2.584 22.912 41.464 0.50
>>>> 45.20 C N
>>>> ATOM 1484 NH2AARG A 1 -1.688 20.715 38.210 0.50
>>>> 44.99 C N
>>>> ATOM 1485 NH2BARG A 1 -3.878 21.619 42.831 0.50
>>>> 44.99 C N
>>>> ATOM 1486 C ARG A 1 -6.518 25.176 36.830 1.00
>>>> 29.97 C C
>>>> ATOM 1487 O ARG A 1 -7.675 25.501 36.971 1.00
>>>> 31.02 C O
>>>>
>>>> 2)
>>>> ATOM 44 N AARG A 1 26.671 62.112 46.990 0.50
>>>> 30.13 A N
>>>> ATOM 45 CA AARG A 1 26.970 63.346 47.667 0.50
>>>> 30.65 A C
>>>> ATOM 46 CB AARG A 1 27.172 64.495 46.676 0.50
>>>> 31.07 A C
>>>> ATOM 47 CG AARG A 1 27.152 65.897 47.322 0.50
>>>> 34.20 A C
>>>> ATOM 48 CD AARG A 1 27.993 66.976 46.599 0.50
>>>> 37.16 A C
>>>> ATOM 49 NE AARG A 1 27.726 67.425 45.342 0.50
>>>> 42.06 A N
>>>> ATOM 50 CZ AARG A 1 28.315 67.639 44.168 0.50
>>>> 44.78 A C
>>>> ATOM 51 NH1AARG A 1 29.525 67.160 43.918 0.50
>>>> 45.37 A N
>>>> ATOM 52 NH2AARG A 1 27.690 68.340 43.240 0.50
>>>> 45.67 A N
>>>> ATOM 53 C AARG A 1 25.839 63.640 48.622 0.50
>>>> 30.37 A C
>>>> ATOM 54 O AARG A 1 24.690 63.377 48.340 0.50
>>>> 31.24 A O
>>>> ATOM 55 N BARG A 1 26.667 62.080 47.010 0.50
>>>> 30.13 A N
>>>> ATOM 56 CA BARG A 1 26.921 63.329 47.640 0.50
>>>> 30.65 A C
>>>> ATOM 57 CB BARG A 1 27.108 64.390 46.581 0.50
>>>> 31.07 A C
>>>> ATOM 58 CG BARG A 1 27.138 65.756 47.103 0.50
>>>> 34.20 A C
>>>> ATOM 59 CD BARG A 1 28.447 66.452 46.933 0.50
>>>> 37.16 A C
>>>> ATOM 60 NE BARG A 1 28.377 67.657 47.707 0.50
>>>> 42.06 A N
>>>> ATOM 61 CZ BARG A 1 29.349 68.269 48.373 0.50
>>>> 44.78 A C
>>>> ATOM 62 NH1BARG A 1 30.572 67.836 48.373 0.50
>>>> 45.37 A N
>>>> ATOM 63 NH2BARG A 1 29.066 69.368 49.034 0.50
>>>> 45.67 A N
>>>> ATOM 64 C BARG A 1 25.822 63.636 48.638 0.50
>>>> 30.37 A C
>>>> ATOM 65 O BARG A 1 24.698 63.372 48.403 0.50
>>>> 31.24 A O
>>>>
>>>>
>>>>
>>>> John Pascal, PhD
>>>> Thomas Jefferson University
>>>> Department of Biochemistry & Molecular Biology
>>>> 233 South 10th Street, BLSB 804
>>>> Philadelphia, Pennsylvania 19107
>>>>
>>>> ph 215.503.4596
>>>> fx 215.923.2117
>>>
>>
>> Disclaimer
>> This communication is confidential and may contain privileged
>> information intended solely for the named addressee(s). It
>> may not be used or disclosed except for the purpose for which
>> it has been sent. If you are not the intended recipient you
>> must not review, use, disclose, copy, distribute or take any
>> action in reliance upon it. If you have received this
>> communication in error, please notify Astex Therapeutics Ltd
>> by emailing [log in to unmask] and destroy all
>> copies of the message and any attached documents.
>> Astex Therapeutics Ltd monitors, controls and protects all
>> its messaging traffic in compliance with its corporate email
>> policy. The Company accepts no liability or responsibility
>> for any onward transmission or use of emails and attachments
>> having left the Astex Therapeutics domain. Unless expressly
>> stated, opinions in this message are those of the individual
>> sender and not of Astex Therapeutics Ltd. The recipient
>> should check this email and any attachments for the presence
>> of computer viruses. Astex Therapeutics Ltd accepts no
>> liability for damage caused by any virus transmitted by this
>> email. E-mail is susceptible to data corruption,
>> interception, unauthorized amendment, and tampering, Astex
>> Therapeutics Ltd only send and receive e-mails on the basis
>> that the Company is not liable for any such alteration or any
>> consequences thereof.
>> Astex Therapeutics Ltd., Registered in England at 436
>> Cambridge Science Park, Cambridge CB4 0QA under number 3751674
>>
>>
>
>
> Disclaimer
> This communication is confidential and may contain privileged information intended solely for the named addressee(s). It may not be used or disclosed except for the purpose for which it has been sent. If you are not the intended recipient you must not review, use, disclose, copy, distribute or take any action in reliance upon it. If you have received this communication in error, please notify Astex Therapeutics Ltd by emailing [log in to unmask] and destroy all copies of the message and any attached documents.
> Astex Therapeutics Ltd monitors, controls and protects all its messaging traffic in compliance with its corporate email policy. The Company accepts no liability or responsibility for any onward transmission or use of emails and attachments having left the Astex Therapeutics domain. Unless expressly stated, opinions in this message are those of the individual sender and not of Astex Therapeutics Ltd. The recipient should check this email and any attachments for the presence of computer viruses. Astex Therapeutics Ltd accepts no liability for damage caused by any virus transmitted by this email. E-mail is susceptible to data corruption, interception, unauthorized amendment, and tampering, Astex Therapeutics Ltd only send and receive e-mails on the basis that the Company is not liable for any such alteration or any consequences thereof.
> Astex Therapeutics Ltd., Registered in England at 436 Cambridge Science Park, Cambridge CB4 0QA under number 3751674
--
Kay Diederichs http://strucbio.biologie.uni-konstanz.de
email: [log in to unmask] Tel +49 7531 88 4049 Fax 3183
Fachbereich Biologie, Universität Konstanz, Box M647, D-78457 Konstanz
This e-mail is digitally signed. If your e-mail client does not have the
necessary capabilities, just ignore the attached signature "smime.p7s".
|
