May 2003 - openSUSE Programming - openSUSE Mailing Lists

Overriding some C++ class methods
by Filippos Papadopoulos 15 May '03

15 May '03

Hi. I am a bit new to C++ so i would like to ask if someone knows the following: I want to override only one method of a widget in Qt library. To do this i create my own class, which inherits that widget, and i declare just that one method in my new class. The problem is that the compiler (g++) complains about the constructors. Do i need also to override ALL the constructors of that widget? (there are many of them). If so, isn't there another way of implementing this? I mean just to override the method that i am interested in and not any other thing in that widget class. Thanks.

3 6

Troubleshooting help
by Greg Freemyer 14 May '03

14 May '03

I need some advice on how to troubleshoot an app crash. First, I'm used to getting core images in UNIX, I have never seen one in Linux. Do I have to enable those somehow? My issue: 3dmd (3ware's raid daemon) works fine in SuSE 8.0. It is unreliable with SuSE 8.2 3ware distributes it as a binary. I don't think source is available, but I'm hoping I can figure out the problem and fix it by modifying something in the SuSE distro. Per "ldd /usr/sbin/3dmd" it is "not a dynamic executable", so it should not be a library issue. My troubleshooting: I start 3dmd with: strace -ff -o /tmp/3dmd /usr/sbin/3dmd 3dmd then fires off a child processes. The first child in turn forks off at least 3 children of its own. It runs fine until I connect to it with a browser and click "Send Test Message". Then all 5 processes disappear from the process table and the browser times out. Looking at the strace output I see a seg. fault with one of the final 3 child processes. I have included the last 50 lines of the strace below my signature. The seg fault is near the top. The strace is not telling me anything useful, what other steps can I take? Thanks Greg -- Greg Freemyer === strace final output tmp # tail -50 3dmd.1473 read(9, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0pY\1\000"..., 1024) = 1024 fstat64(9, {st_mode=S_IFREG|0755, st_size=1491599, ...}) = 0 old_mmap(NULL, 1268004, PROT_READ|PROT_EXEC, MAP_PRIVATE, 9, 0) = 0x4000d000 mprotect(0x4013c000, 26916, PROT_NONE) = 0 old_mmap(0x4013c000, 16384, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 9, 0x12f000) = 0x4013c000 old_mmap(0x40140000, 10532, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x40140000 close(9) = 0 open("/lib/ld-linux.so.2", O_RDONLY) = 9 read(9, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\200\n\0"..., 1024) = 1024 fstat64(9, {st_mode=S_IFREG|0755, st_size=97518, ...}) = 0 old_mmap(NULL, 83576, PROT_READ|PROT_EXEC, MAP_PRIVATE, 9, 0) = 0x40143000 mprotect(0x40157000, 1656, PROT_NONE) = 0 old_mmap(0x40157000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 9, 0x14000) = 0x40157000 close(9) = 0 --- SIGSEGV (Segmentation fault) @ 0 (0) --- write(4, "\0\374\177\277\2\0\0\0\v\0\0\0\377H\7\10p*\21\10\274\7"..., 148) = 148 rt_sigprocmask(SIG_SETMASK, NULL, [SEGV RTMIN], 8) = 0 rt_sigsuspend([SEGV]--- SIGINT (Interrupt) @ 0 (0) --- ) = -1 EINTR (Interrupted system call) sigreturn() = ? (mask now [SEGV RTMIN]) rt_sigsuspend([SEGV] <unfinished ...> --- SIGINT (Interrupt) @ 0 (0) --- <... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call) sigreturn() = ? (mask now [SEGV RTMIN]) rt_sigsuspend([SEGV] <unfinished ...> --- SIGINT (Interrupt) @ 0 (0) --- <... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call) sigreturn() = ? (mask now [SEGV RTMIN]) rt_sigsuspend([SEGV] <unfinished ...> --- SIGINT (Interrupt) @ 0 (0) --- <... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call) sigreturn() = ? (mask now [SEGV RTMIN]) rt_sigsuspend([SEGV] <unfinished ...> --- SIGINT (Interrupt) @ 0 (0) --- <... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call) sigreturn() = ? (mask now [SEGV RTMIN]) rt_sigsuspend([SEGV] <unfinished ...> --- SIGINT (Interrupt) @ 0 (0) --- <... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call) sigreturn() = ? (mask now [SEGV RTMIN]) rt_sigsuspend([SEGV] <unfinished ...> --- SIGINT (Interrupt) @ 0 (0) --- <... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call) sigreturn() = ? (mask now [SEGV RTMIN]) rt_sigsuspend([SEGV] <unfinished ...> --- SIGTERM (Terminated) @ 0 (0) --- <... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system call) --- SIGCONT (Continued) @ 0 (0) --- sigreturn() = ? (mask now [SEGV TERM]) _exit(15) = ? DualStore2000:/tmp #

1 0

Locked folders
by Linux Mail 14 May '03

14 May '03

Some folders contain a lock over them and I cannot open them unless I am root. How ca I create folders that have a similar lock? regards Leo

2 1

suse 8.2 and kdevelop
by himba 13 May '03

13 May '03

hello i just installed suse 8.2 pro edition and i did full install. i have two problems regarding suse 8.2 and kdevelop, 1. kde documentation doesn't get detected by kdevelop --setup, after manually adding directory, where api docs are stored (in next step), htdig shows up with an error that it can't find htdig.conf! 2. i'm newbie at developing and i need <math.h> header and abs() function from it, it appears that during compile i get error about abs() function not found ?! It all worked fine on preuvisly installed RH8 system. regards, himba

8 14

Determine release of Suse
by Fong Vang 10 May '03

10 May '03

On RedHat systems, the release number is contained in /etc/redhat-release (such as "RedHat Linux 8.0"). Is there something equivalent on Suse? Thanks for your help. This e-mail has been captured and archived by the ZANTAZ Digital Safe(tm) service. For more information, visit us at www.zantaz.com. IMPORTANT: This electronic mail message is intended only for the use of the individual or entity to which it is addressed and may contain information that is privileged, confidential or exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, or the employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify the sender immediately by telephone or directly reply to the original message(s) sent. Thank you.

2 1

g++ from suse 8.2 generates broken code
by Carlos Carrasco Martinez 05 May '03

05 May '03

I have attached the full testcase. basically it boils down to enabling any level of optimizations higher than -O1 with -march=i586 or higher. struct matrix { union { struct { float a, b, c, d; float e, f, g, h; float i, j, k, l; float m, n, o, p; } n; float v[16]; float t[4][4]; } data; [...] void identity_bug() { const float I[16] = { 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 1.0f }; for (int i = 0; i < 16; ++i) data.v[i] = I[i]; } [...] int main() { [...] matrix bug; bug.identity_bug(); [...] } with arch=i486 -O3, the code generated for the indentity_bug() call is: (inlined in main) .LBB36: leal -408(%ebp), %edi cld movl $16, %ecx xorl %eax, %eax rep stosl movl $0x3f800000, -408(%ebp) movl $0x3f800000, -388(%ebp) movl $0x3f800000, -368(%ebp) movl $0x3f800000, -348(%ebp) .loc 1 87 0 .LBB37: but with arch=i586 -O2, the code is: .LBB42: cld leal -392(%ebp), %edi movl $16, %ecx xorl %eax, %eax movl $0x3f800000, -392(%ebp) rep stosl movl $0x3f800000, -372(%ebp) movl $0x3f800000, -352(%ebp) movl $0x3f800000, -332(%ebp) .loc 1 87 0 .LBB43: the second movl gets trashed by the susequent rep stosl. it gets worse with arch=athlon-xp: .LBB38: leal -392(%ebp), %edi movl $16, %ecx xorl %eax, %eax cld movl $0x3f800000, -392(%ebp) movl $0x3f800000, -372(%ebp) movl $0x3f800000, -352(%ebp) rep stosl movl $0x3f800000, -332(%ebp) .loc 1 87 0 .LBB39: now 3 movl get overwritten by the rep stosl. in the test case I also include code for 2x2 and 3x3 matrix. for these the for loop is fully unrolled and is just a sequence of movl, so there is no bug. anybody with a newer gcc knows if this bug is fixed? to test it, untar the attached file, and invoke c.sh. if the result of executing the compiled ./bug is: (last 4 lines) | 0 0 0 0 | | 1 0 0 0 | | 0 1 0 0 | | 0 1 0 0 | | 0 0 1 0 | | 0 0 1 0 | | 0 0 0 1 | | 0 0 0 1 | then your g++ has the bug too. I have tested it with gcc 3.2.1 and it is broken too. a friend tested it on debian testing with gcc 3.2.3 and 3.3 (snapshot 20030415) but they generate the correct code. carlos

1 0

c++ array problem
by Gedi 04 May '03

04 May '03

Hi all, I've just started to learn C++ in the last few days and have some accross a small problem. I am trying to build an array to take a number of inputs and print these inputs to stdout. I have read as much material as I can to try and understand why my small program will not compile to no avail. Can anyone point me in the right direction. Here is a snippet of the code that is causeing me the problem : #include <iostream.h> int index, array[10]; int main() { for(index = 0; index < 10; index++) { cout << "Enter string: "; cin >> array[index]; } cout << array; // a little unsure as to how you print the strings out as a whole and seperatly cout << array[1]; cout << array[3]; } Any help would be much appricieted. Thanks, Gedi

5 14

where to get command's code ?
by Linux Aremania 01 May '03

01 May '03

hello again, I wanna learn the program of linux-command. Where can I get it ?, if it's *.RPM or *.SRPMS, how to extract it so I can see its files one by one. thx a lot ===== Linux Aremania __________________________________________________ Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo http://search.yahoo.com

6 12