December 2005 - openSUSE Programming - openSUSE Mailing Lists

Problems Compiling in SUSE 10
by Jorge Almeida 02 Jan '06

02 Jan '06

Hello. I used to work in SuSE 9.2, programming in C++. And i'm working on a program that use a STL template (queue). After i made the upgrade for SuSE 10 the program does not compile anymore! Gives some erros regarding the template declaration and some linking problems with stdc++ library. I've already notice that the gcc version is diferent and the locations of the header and libraries are also different. But the compability should be done!!?? If anyone already have this problem and solved it please report the solution to the list Thanks in advance -- Jorge Almeida j-almeida(a)criticalsoftware.com DISCLAIMER: This message may contain confidential information or privileged material and is intended only for the individual(s) named. If you are not a named addressee and mistakenly received this message you should not copy or otherwise disseminate it: please delete this e-mail from your system and notify the sender immediately. E-mail transmissions are not guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete or contain viruses. Therefore, the sender does not accept liability for any errors or omissions in the contents of this message that arise as a result of e-mail transmissions. Please request a hard copy version if verification is required. Critical Software.

5 15

kdevelop file externally modified
by Karaoulis Marios 02 Jan '06

02 Jan '06

After everytime i try to compile my projects this message appears "kdevelop file externally modified" And trying to save all open files i have at that time. Obviously i haven't changed anything in the files. After i choose to Overwrite the exsisting files kdevelop crashes.... This happens after i updated kdevelop to 3.3.0 (with kde 3.5). Any ideas? __________________________________________________ �� Yahoo!; �� (spam); �� Yahoo! Mail �� http://mail.yahoo.gr

3 3

kprobes
by Frank Beckmann 26 Dec '05

26 Dec '05

Hi :-) i have try to load a kprobe as modul into the kernel my SuSE version is 10 and the runing 2.6.13-15.7-default the kernel is make with make cloneconfig. To enable kprobes i have changed the kernel konfig /usr/src/linux/.config CONFIG_KALLSYMS=y CONFIG_KALLSYMS_ALL=y CONFIG_KALLSYMS_EXTRA_PASS=y # Kernel hacking CONFIG_DEBUG_KERNEL=y CONFIG_MAGIC_SYSRQ=y CONFIG_LOG_BUF_SHIFT=17 CONFIG_DEBUG_BUGVERBOSE=y CONFIG_DEBUG_INFO=y CONFIG_DEBUG_FS=y CONFIG_EARLY_PRINTK=y CONFIG_DEBUG_STACKOVERFLOW=y CONFIG_KPROBES=y I use the exsample from /usr/src/linux/Documentation/kprobes.txt compiling and linking no problem but insmod ./kprobedoc.ko insmod: error inserting './kprobedoc.ko': -1 Operation not permitted messages: Dec 26 19:27:30 mainframe kernel: register_kprobe failed, returned 1 the entry in the messages came from the int init_module function if ((ret = register_kprobe(&kp) < 0)) { printk("register_kprobe failed, returned %d\n", ret); return -1; } i have forgot somthing in the kernel config ? or is it a bug ? Thanks Frank #include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> #include <linux/kprobes.h> #include <linux/kallsyms.h> #include <linux/sched.h> /*For each probe you need to allocate a kprobe structure*/ static struct kprobe kp; /*kprobe pre_handler: called just before the probed instruction is executed*/ int handler_pre(struct kprobe *p, struct pt_regs *regs) { printk("pre_handler: p->addr=0x%p, eip=%lx, eflags=0x%lx\n", p->addr, regs->eip, regs->eflags); dump_stack(); return 0; } /*kprobe post_handler: called after the probed instruction is executed*/ void handler_post(struct kprobe *p, struct pt_regs *regs, unsigned long flags) { printk("post_handler: p->addr=0x%p, eflags=0x%lx\n", p->addr, regs->eflags); } /* fault_handler: this is called if an exception is generated for any * instruction within the pre- or post-handler, or when Kprobes * single-steps the probed instruction. */ int handler_fault(struct kprobe *p, struct pt_regs *regs, int trapnr) { printk("fault_handler: p->addr=0x%p, trap #%dn", p->addr, trapnr); /* Return 0 because we don't handle the fault. */ return 0; } int init_module(void) { int ret; kp.pre_handler = handler_pre; kp.post_handler = handler_post; kp.fault_handler = handler_fault; kp.addr = (kprobe_opcode_t*) kallsyms_lookup_name("do_fork"); /* register the kprobe now */ if (!kp.addr) { printk("Couldn't find %s to plant kprobe\n", "do_fork"); return -1; } printk("%p kp adr \n",&kp); if ((ret = register_kprobe(&kp) < 0)) { printk("register_kprobe failed, returned %d\n", ret); return -1; } printk("kprobe registered\n"); return 0; } void cleanup_module(void) { unregister_kprobe(&kp); printk("kprobe unregistered\n"); } MODULE_LICENSE("GPL"); ####################################################################### cat Makefile TARGET = kprobedoc OBJS = kprobedoc.o MDIR = drivers/misc EXTRA_CFLAGS = -DEXPORT_SYMTAB CURRENT = $(shell uname -r) KDIR = /lib/modules/$(CURRENT)/build PWD = $(shell pwd) DEST = /lib/modules/$(CURRENT)/kernel/$(MDIR) obj-m := $(TARGET).o default: make -C $(KDIR) SUBDIRS=$(PWD) modules $(TARGET).o: $(OBJS) $(LD) $(LD_RFLAG) -r -o $@ $(OBJS) ifneq (,$(findstring 2.4.,$(CURRENT))) install: su -c "cp -v $(TARGET).o $(DEST) && /sbin/depmod -a" else install: su -c "cp -v $(TARGET).ko $(DEST) && /sbin/depmod -a" endif clean: -rm -f *.o *.ko .*.cmd .*.flags *.mod.c -include $(KDIR)/Rules.make ______________________________________________________________ Verschicken Sie romantische, coole und witzige Bilder per SMS! Jetzt bei WEB.DE FreeMail: http://f.web.de/?mc=021193

1 0

Re: [suse-programming-e] Problems Compiling in SUSE 10
by Jorge Almeida 21 Dec '05

21 Dec '05

I was thinking in that. Perhaps i will try that approach. Thanks anyway. Em Wednesday 21 December 2005 16:22, escreveu: > I do not have to tell you that you have linking problems, obviously. :) > > I would carefully compile GCC3 and put it in different hierarchy using > something like .. > > ../gcc-3.4.5/configure --prefix=/usr/local/gcc --enable-threads > --enable-shared --enable-languages=c,c++ --with-libiconv-prefix=/usr > > > then see if it is the compiler. My guess is that it is GCC4. > > I have had too many problems with it thus to the point where I am thinking > about going back to 9.3. > > > > LDB > > Jorge Almeida wrote: > > The errors are the following: > > > > For the lib: > > ############################################### > > /usr/local/lib/libccext2.so: undefined reference to > > `__gnu_cxx::__exchange_and_add(int volatile*, int)' > > /usr/local/lib/libccext2.so: undefined reference to > > `std::_Rb_tree_decrement(std::_Rb_tree_node_base*)' > > /usr/local/lib/libccext2.so: undefined reference to > > `std::basic_string<char, std::char_traits<char>, std::allocator<char> > > >::_Rep::_S_empty_rep_storage' /usr/local/lib/libccext2.so: undefined > > reference to > > `std::basic_streambuf<char, std::char_traits<char> >::seekoff(long long, > > std::_Ios_Seekdir, std::_Ios_Openmode)' > > /usr/local/lib/libccext2.so: undefined reference to > > `std::_Rb_tree_insert_and_rebalance(bool, std::_Rb_tree_node_base*, > > std::_Rb_tree_node_base*, std::_Rb_tree_node_base&)' > > /usr/local/lib/libccext2.so: undefined reference to > > `std::_Rb_tree_rebalance_for_erase(std::_Rb_tree_node_base*, > > std::_Rb_tree_node_base&)' > > /usr/local/lib/libccext2.so: undefined reference to > > `std::_Rb_tree_increment(std::_Rb_tree_node_base*)' > > collect2: ld returned 1 exit status > > ############################################## > > > > and for the source template: > > > > ############################################### > > compiling CMiceReadQueues.cpp (g++) > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Reserve(size_t)’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:205: error: ‘c’ was not > > declared in this scope > > if g++ -DHAVE_CONFIG_H -I. > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/linux/queues -I../../.. > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/ciniparserwrapper > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/ciniparserwrapper/iniparser > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/headers > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/linux/queues > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/linux/systemlog > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/linux/utils > > -I/home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implemen > >tation/smice/./common/message -O0 -Werror -Wall -Wnon-virtual-dtor -g3 > > -I/usr/local/include/cc++2 -D_GNU_SOURCE -MT CMiceWriteQueues.o -M > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Clear()’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:233: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘const > > T& CMiceQueue<T>::Top() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:290: error: there are no > > arguments to ‘top’ that depend on a template parameter, so a declaration > > of ‘top’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:290: error: (if you use > > ‘-fpermissive’, G++ will accept your code, but allowing the use of an > > undeclared name is deprecated > > D -MP -MF ".deps/CMiceWriteQueues.Tpo" -c -o > > CMiceWriteQueues.o > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceWriteQueues.cpp; then mv -f > > ".deps/CMiceWriteQueues.Tpo" ".deps/CMiceWriteQueues.Po"; else rm -f > > ".deps/CMiceWriteQueues.Tpo"; exit 1; fi > > ) > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Pop()’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:318: error: there are no > > arguments to ‘size’ that depend on a template parameter, so a declaration > > of ‘size’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:319: error: there are no > > arguments to ‘pop’ that depend on a template parameter, so a declaration > > of ‘pop’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘bool > > CMiceQueue<T>::Empty() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:348: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘size_t > > CMiceQueue<T>::Size() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:376: error: ‘c’ was not > > declared in this scope > > gmake[4]: *** [CMiceReadQueues.o] Error 1 > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Reserve(size_t)’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:205: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Clear()’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:233: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘const > > T& CMiceQueue<T>::Top() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:290: error: there are no > > arguments to ‘top’ that depend on a template parameter, so a declaration > > of ‘top’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:290: error: (if you use > > ‘-fpermissive’, G++ will accept your code, but allowing the use of an > > undeclared name is deprecated) > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Pop()’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:318: error: there are no > > arguments to ‘size’ that depend on a template parameter, so a declaration > > of ‘size’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:319: error: there are no > > arguments to ‘pop’ that depend on a template parameter, so a declaration > > of ‘pop’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘bool > > CMiceQueue<T>::Empty() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:348: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘size_t > > CMiceQueue<T>::Size() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:376: error: ‘c’ was not > > declared in this scope > > gmake[4]: *** [CMiceWriteQueues.o] Error 1 > > gmake[4]: Target `all' not remade because of errors. > > Making all in cmonitoring > > compiling CManagement.cpp (g++) > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Reserve(size_t)’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:205: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Clear()’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:233: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘const > > T& CMiceQueue<T>::Top() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:290: error: there are no > > arguments to ‘top’ that depend on a template parameter, so a declaration > > of ‘top’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:290: error: (if you use > > ‘-fpermissive’, G++ will accept your code, but allowing the use of an > > undeclared name is deprecated) > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘void > > CMiceQueue<T>::Pop()’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:318: error: there are no > > arguments to ‘size’ that depend on a template parameter, so a declaration > > of ‘size’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:319: error: there are no > > arguments to ‘pop’ that depend on a template parameter, so a declaration > > of ‘pop’ must be available > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘bool > > CMiceQueue<T>::Empty() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:348: error: ‘c’ was not > > declared in this scope > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h: In member function ‘size_t > > CMiceQueue<T>::Size() const’: > > /home/jorge/cvs_critical/marinha/direccao-navios/sistemas-mice/implementa > >tion/smice/./common/linux/queues/CMiceQueue.h:376: error: ‘c’ was not > > declared in this scope > > ############################################################# > > > > > > and more that i do not copy because are similar. > > > > > > I've noticed that SUSE 10 come with version 4.0.X of GCC. Maybe that is > > the problem. > > > > > > Thanks. > > > > Em Wednesday 21 December 2005 15:58, o LDB escreveu: > >> Jorge Almeida wrote: > >>> Hello. > >>> > >>> I used to work in SuSE 9.2, programming in C++. > >>> And i'm working on a program that use a STL template (queue). > >>> > >>> After i made the upgrade for SuSE 10 the program does not compile > >>> anymore! Gives some erros regarding the template declaration and some > >>> linking problems with stdc++ library. > >>> > >>> I've already notice that the gcc version is diferent and the locations > >>> of the header and libraries are also different. But the compability > >>> should be done!!?? > >>> > >>> If anyone already have this problem and solved it please report the > >>> solution to the list > >>> > >>> Thanks in advance > >> > >> What are the errors displayed? > >> > >> I know I had to compile my own version of GCC3 to get some of my > >> development off the ground on 10.0. > >> > >> LDB -- Jorge Almeida j-almeida(a)criticalsoftware.com DISCLAIMER: This message may contain confidential information or privileged material and is intended only for the individual(s) named. If you are not a named addressee and mistakenly received this message you should not copy or otherwise disseminate it: please delete this e-mail from your system and notify the sender immediately. E-mail transmissions are not guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete or contain viruses. Therefore, the sender does not accept liability for any errors or omissions in the contents of this message that arise as a result of e-mail transmissions. Please request a hard copy version if verification is required. Critical Software.

1 0

problem with SUSE 10.0 and gcc, fortran and LAPACK
by bauckhag＠cs.yorku.ca 21 Dec '05

21 Dec '05

Hello everyone I am desperately trying to compile C programs which require LAPACK on SUSE 10.0 (running on an AMD 64 bit architecture). So far, I was using Suse 9.1 and everything worked fine. There, I had the following packages installed: - gcc (3.3.3-41 i586) - gcc-f77 - f2c (0.11-1043 i586) - lapack (3.0-920.1 i586) Using the following line in the makefile, gcc was able to compile C programs involving LAPCK: LDLIBS+= -lgcc -llapck -lg2c -lm Now, under SUSE 10.0, I installed the following packages: - gcc (4.0.2_20050901-3x86_64) - gcc-fortran - compat-g77 (3.3.5-2x86_64) - f2c (0.11-1052x85_64) - lapack (3.0-926x86_64) Alas, using the same linking options as above now results in numerous errors such as: /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../lib64/liblapack.a(dgeev.i): In function 'dgeev_': undefined reference to 'gfortran_copy_string' What can be done to resolve this and to get programs developped under SUSE 9.1 running under SUSE 10.0? Calling SUSE support on this was a nightmare experience! Does anybody here have an idea? Best regards Christian ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

2 3

Profiling problem
by Verdi March 20 Dec '05

20 Dec '05

Hi, I need some advise on profiling. I'm comparing the running time of two simulators, let's call them algo1 and algo2 (a slightly modified version of algo1). My problem, in summary, is that on a particular platform configuration, the runtime of algo2 is almost 3x longer than algo1, yet the profiling shows that algo2 calls the hotspot function only 22% higher than algo1. In addition, a single call to the hotspot function in both algo1 and algo2 should be the same (explained later). So the question is, why is the runtime of algo2 increase 3x? If the runtime increase by 20% to, say 50%, then I can understand, but 300% is a bit difficult to accept. The tests were performed on three platform configurations: 1. P4 3 GHz (HT enabled and 1MB cache), SuSE 9.3, glibc-2.3.4-23.4, and gcc-3.3.5-5. 2. Xeon 2.8 GHz (512KB cache), Fedora Core 3, glibc-2.3.6-0.fc3.1, and gcc-3.4.4-2.fc3. 3. Opteron 2.1 GHz (1MB cache), Fedora Core 4, glibc-2.3.5-10.3, and gcc-4.0.2-8.fc4. Each configuration has >= 1 GB memory, and each run of algo1 or algo2 consume about 256-300 MB. Therefore, overhead because of swapping is not a problem (later confirmed with 100% CPU util as reported by /usr/bin/time). Also, only config-3 is a 64-bit platform. The runtime of each program on each configuration is shown below: - Config-1: * algo1: user=1445.93 s, sys=11.33 s, real=23 m 54.02 s * algo2: user=4311.42 s, sys=17.92 s, real=1 h 11 m 41 s - Config-2: * algo1: user=1926.66 s, sys=18.85 s, real=32 m 14.45 s * algo2: user=1897.31 s, sys=17.94 s, real=31 m 46.54 s - Config-3: * algo1: user=1413.00 s, sys=9.18 s, real=23 m 41.39 s * algo2: user=1214.51 s, sys=6.35 s, real=20 m 21.07 s It can be seen that on config-1 (P4, SuSE), algo2 takes almost 3x longer than algo1. However, on other configurations, both versions takes roughly the same amount of time. Here are some snippets of gprof profiling result of algo1 and algo2, both on config-1. /*************************** Algo 1 ******************************/ FLAT PROFILE: % cumulative self self total time seconds seconds calls s/call s/call name 21.22 2323.95 2323.95 1678059784 0.00 0.00 ineq_compare 9.43 3357.38 1033.43 1538051170 0.00 0.00 biguint_equal 6.52 4071.34 713.96 140185901 0.00 0.00 getNeighbors1 ... 0.00 10954.11 0.00 1 0.00 0.00 initRand 0.00 10954.11 0.00 1 0.00 0.08 initStateDiagrams CALL GRAPH: granularity: each sample hit covers 4 byte(s) for 0.00% of 15130.79 s index % time self children called name <spontaneous> [1] 91.2 143.11 13656.78 main [1] 129.54 12024.06 61206914/61206914 processRequest [2] 94.21 404.78 165336753/165336753 getEvent [21] ... /*****************************************************************/ /*************************** Algo 2 ******************************/ FLAT PROFILE: % cumulative self self total time seconds seconds calls s/call s/call name 21.79 2494.03 2494.03 2057510279 0.00 0.00 ineq_compare 9.35 3564.45 1070.41 2020502220 0.00 0.00 biguint_equal 6.77 4339.43 774.98 148414184 0.00 0.00 getNeighbors1 ... 0.00 11444.75 0.00 1 0.00 0.00 initRand 0.00 11444.75 0.00 1 0.00 0.04 initStateDiagrams CALL GRAPH: granularity: each sample hit covers 4 byte(s) for 0.00% of 14820.76 s index % time self children called name <spontaneous> [1] 90.8 136.76 13327.52 main [1] 129.92 12040.16 44425464/44425464 processRequest [2] 70.57 283.72 123000389/123000389 getEvent [25] /*****************************************************************/ Some observations that I've made are as follows. First, the hotspot is ineq_compare. In algo2, it's called 22% times higher than in algo1. This means, for algo2 to takes almost 3x longer than algo1, there should be a significant increase in the cost of a single call to algo2's ineq_compare. Even though from gprof I couldn't know total_s/call is increased in algo2, but from the source code I'm sure that self_s/call does not change in algo1 and algo2. Function ineq_compare takes two big unsigned integers (biguint) as its parameters and compares them. Each biguint is implemented as an array of long long unsigned int. The function remains the same in both versions and the array in biguint consists of two elements. Hence, the conclusion that each call of ineq_compare costs the same in both algo1 and algo2. Second, gprof shows that main() in both algo1 and algo2 takes roughly the same amount of time. I would expect that the profile shows main() in algo2 is significantly longer than in algo1? Because of the result on the two other platforms, I'm tempted to "blame" either the hardware (P4) or software platform (older glibc and gcc) :D, but somehow I'm still declined to do that. Any ideas on how to better find out the cause of slowdown? Regards, Verdi -- Telefonieren Sie schon oder sparen Sie noch? NEU: GMX Phone_Flat http://www.gmx.net/de/go/telefonie

1 2

Override hotkeys in KDE
by anuradha.vembu＠wipro.com 16 Dec '05

16 Dec '05

Hi, Thanks for the support till now. I have a GUI based application (Qt) in which I have defined some shortcuts like - Ctrl+ Function keys, Ctrl+ Tab, Alt+F1, Alt+F2... etc. They are mapped to some other functionality in SuSe KDE (3.2) by default (For e.g., Ctrl+ Tab is minimizing all the windows). Is there any way to override these default key mappings? In Google search I found the link (http://bdn.borland.com/article/borcon/files/6202/paper/6202.html <http://bdn.borland.com/article/borcon/files/6202/paper/6202.html> chapter 9 -> KDE stealing hot-keys) which said that with Scroll lock enabled, the default hotkeys will not function; but it is not working with my application. The Ctrl+Fx keys combination is working in Gnome environment. Any suggestion to override the default key mappings in KDE, so that the shortcuts I have defined would function would be of great help to me. Thanks in advance /Anuradha The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com

2 1

MIT Scheme on SuSE 10
by Steven T. Hatton 13 Dec '05

13 Dec '05

What do I need in order to work with MIT Scheme on SuSE 10? I can't figure out how to get a chicken nor an egg from the savana site. Theire so-called binary distribution does not reflect the documentation that comes with it. In particular there is no bin directory which is where they claim I will find the executables. If I understand correctly, I need that in order to build any CVS executable of Scheme. Steven

2 3

GDB- doesn't stop in constructors & destructors?
by anuradha.vembu＠wipro.com 10 Dec '05

10 Dec '05

Hi, I am trying to debug my application using gdb. When I try to put a breakpoint in a constructor, it says no such function. When I apply breakpoint in the particular line number, the breakpoint gets set. When I run the application, it should stop in that breakpoint before any execution; but it does not do. So my doubt is, does gdb display when breakpoints are put in the constructor? Please suggest me a way by which I would ensure that all the constructors (objects) are initialized before I run the application. Thanks in advance. Regards, Anuradha Confidentiality Notice The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain confidential or privileged information. If you are not the intended recipient, please notify the sender at Wipro or Mailadmin(a)wipro.com immediately and destroy all copies of this message and any attachments.

3 3

Debugging c++ template classes
by Jerry Feldman 02 Dec '05

02 Dec '05

I'm working on a huge (1,000,000 lines of code + 12 3rd party libraries) project and I'm trying to use gdb to debug some template classes. More specifically, I've got a template within a template. I want to be able to get a breakpoint. Note that my GDB script has a number of dir commands so it can find all the sources and header files. In this case, there is a static class variable I want to set a watchpoint on. But, even when stepping thorough I generally cannot even view the lines of code (templates are generally inlined). In this particular case I am using the Intel C++ compiler on an Itanium processor. What is really strange is that the Intel version of the code fails to obtain some data, but executes and exits cleanly. The g++ code executes cleanly and segfaults on exit. There are other debugging techniques I am using, but the ability to step into or set a breakpoint in a template class function would save me some time. -- Jerry Feldman <gaf(a)blu.org> Boston Linux and Unix user group http://www.blu.org PGP key id:C5061EA9 PGP Key fingerprint:053C 73EC 3AC1 5C44 3E14 9245 FB00 3ED5 C506 1EA9

2 4