Boost user list

OvermindDL1 | 1 Aug 01:22

Re: boost::lambda::bind temporary function objects.. and raw C

On Fri, Jul 31, 2009 at 9:20 AM, Steven Watanabe<watanabesj <at> gmail.com> wrote:
> AMDG
>
> Tomasz Mloduchowski wrote:
>>
>> Hello everyone!
>>
>> I have a rather challenging question, that perhaps can be resolved without
>> too much dark magic.
>>
>> A little background: I'm patching an executable (x86 platform, but could
>> be easily extended to other architectures).
>>
>> players:
>> void TargetFunc(some parameters);
>> void Hook(some parameters);
>>
>> method:
>> Grab first 5 bytes of TargetFunc. Back them up. Replace them with a jmp
>> Hook.
>>
>> Make Hook finish as follows:
>>    Restore the backup of 5 bytes.
>>    Call TargetFunc again, this time the real one, not the jmp.
>>    Replace the 5 bytes with jmp Hook, getting ready for the next
>> invocation.
>>
>> In a nutshell - typical trampoline.
>>
>> Now, I want to write a framework for this. I would prefer to be able to

(Continue reading)

headers

Caligula | 1 Aug 09:16

copy_graph - accessing property maps

If you use copy_graph(g, g2) how can you access the corresponding property maps in g2?

Thanks
-Caligula

_______________________________________________
Boost-users mailing list
Boost-users <at> lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/boost-users

headers

er | 1 Aug 21:58

reference_wrapper

Hi All,

1) I don't understand this part of reference_wrapper:

template<class T> class reference_wrapper
{
	operator T& () const { return *t_; }
};

Could someone please explain it?

2) Is there already a reference wrapper for n-ary functors which exposes 
the same result_type (if it exists) as the n-ary functor, but whose 
arguments types are arbitrary (deduced from the arguments)?

headers

Gennadiy Rozental | 1 Aug 23:49

Re: [Boost.Test] [1.38] Automated test suite registration with muti-file-test-modules

Greg Christopher wrote:
> #include <boost/test/execution_monitor.hpp>
> #include <boost/test/unit_test.hpp>
> 
> #define BOOST_TEST_MODULE Yeah

You obviously need to define the symbol *before* include, not after.

Gennadiy

headers

Gennadiy Rozental | 1 Aug 23:50

Re: Boost.test applicability

Joel Falcou wrote:
> Other point, I saw that there is check in Boost.test which check for an 
> arbitrary precision result. However it uses raw value (like 0.001) where 
> having a check based on ULP or RMS maybe better. Is there anyway to 
> integrate those test using existing test framework ?

So, how did you mean to use RMS for FP value comparison?

Gennadiy

headers

Selçuk Cevher | 1 Aug 10:03

non-POD C++ objects and Boost.Interprocess

Hi,

Is it possible to pass user-defined "non-POD" C++ class objects to message queues and retrieve them back on the receiver side using Boost.Interprocess ?

Thanks.

_______________________________________________
Boost-users mailing list
Boost-users <at> lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/boost-users

headers

Seth Fowler | 2 Aug 01:37

Valgrind's DRD tool produces false positives when used with Boost.Thread?

Hi all,

I've been trying to debug some data races in an application that uses  
Boost.Thread. I tried running valgrind's DRD tool against my binary to  
see if it could locate the source of the problem. It produced more  
errors than I expected, so I progressively removed portions of my  
application unless I was left with almost nothing. I found that even  
with the most minimal program I could produce, the DRD tool still  
reported errors - in fact, it even reported errors with the examples  
included with the Boost distribution! I was able to reproduce this  
behavior on Mac OS X 10.5.7 with Boost 1.39, and on Ubuntu 9.04 with  
Boost 1.37.

Here are steps to reproduce the problem:

1. Build the sample program thread.cpp included in the Boost  
distribution in libs/thread/example. For Boost 1.39, I used the  
following command:

g++ -L<Boost lib directory> -lboost_thread-xgcc40-mt-1_39 thread.cpp - 
o thread

2. Run it using valgrind's DRD tool:

valgrind --tool=drd --var-info=yes ./thread

On Mac OS X 10.5.7, I get the following output:

==20729== drd, a thread error detector.
==20729== Copyright (C) 2006-2009, and GNU GPL'd, by Bart Van Assche.
==20729== Using Valgrind-3.5.0.SVN and LibVEX; rerun with -h for  
copyright info
==20729== Command: ./thread
==20729==
--20729-- ./thread:
--20729-- dSYM directory is missing; consider using --auto-run- 
dsymutil=yes
setting alarm for 5 seconds...
==20729== Conflicting load by thread 1 at 0x00012018 size 4
==20729==    at 0x58B1: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x390C: boost::thread::join() (in ./thread)
==20729==    by 0x9A13: main (in ./thread)
==20729== Allocation context: Data section of ./thread
==20729== Other segment start (thread 2)
==20729==    at 0x396B3: pthread_mutex_lock (drd_pthread_intercepts.c: 
580)
==20729==    by 0x5993: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment end (thread 2)
==20729==    at 0x39BC2: pthread_mutex_unlock  
(drd_pthread_intercepts.c:633)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment start (thread 2)
==20729==    at 0x396B3: pthread_mutex_lock (drd_pthread_intercepts.c: 
580)
==20729==    by 0x58D2: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment end (thread 2)
==20729==    at 0x39BC2: pthread_mutex_unlock  
(drd_pthread_intercepts.c:633)
==20729==    by 0x5988: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729==
==20729== Conflicting load by thread 1 at 0x00012014 size 4
==20729==    at 0x58B4: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x390C: boost::thread::join() (in ./thread)
==20729==    by 0x9A13: main (in ./thread)
==20729== Allocation context: Data section of ./thread
==20729== Other segment start (thread 2)
==20729==    at 0x396B3: pthread_mutex_lock (drd_pthread_intercepts.c: 
580)
==20729==    by 0x5993: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment end (thread 2)
==20729==    at 0x39BC2: pthread_mutex_unlock  
(drd_pthread_intercepts.c:633)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment start (thread 2)
==20729==    at 0x396B3: pthread_mutex_lock (drd_pthread_intercepts.c: 
580)
==20729==    by 0x58D2: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment end (thread 2)
==20729==    at 0x39BC2: pthread_mutex_unlock  
(drd_pthread_intercepts.c:633)
==20729==    by 0x5988: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729==
==20729== Conflicting load by thread 1 at 0x00013b5c size 4
==20729==    at 0x993A: boost::detail::get_once_per_thread_epoch()  
(in ./thread)
==20729==    by 0x58BA: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x390C: boost::thread::join() (in ./thread)
==20729==    by 0x9A13: main (in ./thread)
==20729== Allocation context: Data section of ./thread
==20729== Other segment start (thread 2)
==20729==    at 0x227FF0: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment end (thread 2)
==20729==    at 0x396B3: pthread_mutex_lock (drd_pthread_intercepts.c: 
580)
==20729==    by 0x58D2: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729==
==20729== Conflicting load by thread 1 at 0x00013b5c size 4
==20729==    at 0x9972: boost::detail::get_once_per_thread_epoch()  
(in ./thread)
==20729==    by 0x58BA: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x390C: boost::thread::join() (in ./thread)
==20729==    by 0x9A13: main (in ./thread)
==20729== Allocation context: Data section of ./thread
==20729== Other segment start (thread 2)
==20729==    at 0x227FF0: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment end (thread 2)
==20729==    at 0x396B3: pthread_mutex_lock (drd_pthread_intercepts.c: 
580)
==20729==    by 0x58D2: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
==20729==
==20729== Conflicting load by thread 1 at 0x00397680 size 4
==20729==    at 0x20038D: pthread_setspecific (in /usr/lib/ 
libSystem.B.dylib)
==20729==    by 0x997F: boost::detail::get_once_per_thread_epoch()  
(in ./thread)
==20729==    by 0x58BA: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x390C: boost::thread::join() (in ./thread)
==20729==    by 0x9A13: main (in ./thread)
==20729== Allocation context: Data section of /usr/lib/libSystem.B.dylib
==20729== Other segment start (thread 2)
==20729==    at 0x227FF0: thread_start (in /usr/lib/libSystem.B.dylib)
==20729== Other segment end (thread 2)
==20729==    at 0x396B3: pthread_mutex_lock (drd_pthread_intercepts.c: 
580)
==20729==    by 0x58D2: void boost::call_once<void (*) 
()>(boost::once_flag&, void (*)()) (in ./thread)
==20729==    by 0x30CC: thread_proxy (in ./thread)
==20729==    by 0x3707B: vgDrd_thread_wrapper  
(drd_pthread_intercepts.c:272)
==20729==    by 0x228154: _pthread_start (in /usr/lib/libSystem.B.dylib)
==20729==    by 0x228011: thread_start (in /usr/lib/libSystem.B.dylib)
alarm sounded...
==20729==
==20729== ERROR SUMMARY: 5 errors from 5 contexts (suppressed: 725  
from 8)

The output on linux is similar.

Does this indicate a problem with valgrind's DRD tool, Boost.Thread,  
or both? Does any workaround exist?

Thanks in advance,
- Seth Fowler

headers

Ion Gaztañaga | 2 Aug 07:05

Re: non-POD C++ objects and Boost.Interprocess

Selçuk Cevher escribió:
> Hi,
> 
> Is it possible to pass user-defined "non-POD" C++ class objects to 
> message queues and retrieve them back on the receiver side using 
> Boost.Interprocess ?
> 
> Thanks.

Yes, provided both processes are compiled with the same ABI, but 
remember that non-POD is limited to "non-virtual, non mutable static 
classes with non-raw pointers", the same classes you are allowed to put 
in managed_shared_memory. See:

"Limitations When Constructing Objects In Mapped Regions"

http://www.boost.org/doc/libs/1_39_0/doc/html/interprocess/sharedmemorybetweenprocesses.html

Best,

Ion

headers

Archie14 | 2 Aug 14:24

Remove elements from accumulator

I am trying to use accumulator to calculate statistics in "rolling window", 
traversing the data. I searched accumulators library for a method to remove 
first N elements from the accumulator and did not find one.

Question: is is possible to remove first N values from the accumulator?

headers

Eric Niebler | 2 Aug 17:45

Re: Remove elements from accumulator

Archie14 wrote:
> I am trying to use accumulator to calculate statistics in "rolling window", 
> traversing the data. I searched accumulators library for a method to remove 
> first N elements from the accumulator and did not find one.
> 
> Question: is is possible to remove first N values from the accumulator?

No, but there are some rolling window statistical accumulators: 
rolling_sum and rolling_mean.

--

-- 
Eric Niebler
BoostPro Computing
http://www.boostpro.com

140	February 2012
426	January 2012
402	December 2011
627	November 2011
501	October 2011
554	September 2011
636	August 2011
600	July 2011
546	June 2011
604	May 2011
641	April 2011
794	March 2011
569	February 2011
731	January 2011
593	December 2010
812	November 2010
774	October 2010
854	September 2010
857	August 2010
849	July 2010
916	June 2010
857	May 2010
727	April 2010
1079	March 2010
1019	February 2010
725	January 2010
907	December 2009
782	November 2009
850	October 2009
1094	September 2009
952	August 2009
1051	July 2009
791	June 2009
896	May 2009
1014	April 2009
1083	March 2009
1013	February 2009
991	January 2009
889	December 2008
921	November 2008
834	October 2008
987	September 2008
1209	August 2008
1142	July 2008
810	June 2008
1015	May 2008
806	April 2008
732	March 2008
617	February 2008
754	January 2008
531	December 2007
600	November 2007
662	October 2007
505	September 2007
722	August 2007
813	July 2007
767	June 2007
964	May 2007
694	April 2007
822	March 2007
896	February 2007
675	January 2007
484	December 2006
628	November 2006
582	October 2006
706	September 2006
758	August 2006
686	July 2006
629	June 2006
667	May 2006
783	April 2006
761	March 2006
891	February 2006
663	January 2006
571	December 2005
643	November 2005
524	October 2005
645	September 2005
773	August 2005
489	July 2005
395	June 2005
517	May 2005
593	April 2005
728	March 2005
390	February 2005
526	January 2005
586	December 2004
449	November 2004
241	October 2004
342	September 2004
347	August 2004
332	July 2004
223	June 2004
191	May 2004
170	April 2004
280	March 2004
277	February 2004
179	January 2004
117	December 2003
210	November 2003
176	October 2003
163	September 2003
315	August 2003
301	July 2003
297	June 2003
333	May 2003
330	April 2003
341	March 2003
238	February 2003
166	January 2003
201	December 2002
329	November 2002
213	October 2002
360	September 2002
255	August 2002
266	July 2002
176	June 2002
193	May 2002
112	April 2002
1	February 2002

Mon	Tue	Wed	Thu	Fri	Sat	Sun

					1	2
3	4	5	6	7	8	9
10	11	12	13	14	15	16
17	18	19	20	21	22	23
24	25	26	27	28	29	30
31