Skip to contentSkip to conversation

Kernel doesn't create core for threads

DISCUSSION Zdenek Kabelac 2 participants 2 messages May 2, 2001

lore
Share & tools
lore
#0 lore raw

Hello

Looks to me like the latest 2.4.5-pre1 is not creating coredump for multithreaded aplications:

----

$ ulimit  -a
core file size (blocks)     unlimited
data seg size (kbytes)      unlimited
file size (blocks)          unlimited
max locked memory (kbytes)  unlimited
max memory size (kbytes)    unlimited
open files                  1024
pipe size (512 bytes)       8
stack size (kbytes)         8192
cpu time (seconds)          unlimited
max user processes          6144
virtual memory (kbytes)     unlimited
$ ./testcore 
Segmentation fault

--- test program --

#include <stdio.h>
#include <pthread.h>

#define TASKS 10

void* thread(void* arg)

{
    printf("yuk %s\n", 0xffff0000);
    return 0;
}

int main(int argc, char *argv[])

{
    pthread_t task[TASKS];
    void* ret;
    int i = 0;
    for(i = 0; i < TASKS; i++)
	pthread_create(&task[i], NULL, thread, 0);
    for(i = 0; i < TASKS; i++)
	pthread_join(task[i], &ret);

    return 0;
}

---

for single threadad apps there is no problem:
(just put printf after int i declaration)

$ ./testcore 
Segmentation fault (core dumped)

---

Am I doing something wrong ???

Also for quite a long time Alan's AC patches were efictively locking my machine when threaded application was crashing - resolvin almost original behaviour solved this problem - but I like per-thread coredump - is there any special reason why this is still not present in the current 2.4.4 kernel as I can see this as quite usefull.

bye

[email protected]

#1 lore raw

Zdenek-

Yes, this is a known problem. I sent out a patch last month that fixed this problem. Unfortunately, the patch caused some kernel hangs that I just fixed today. I'm testing the fix now so expect a new patch either this afternoon or tomorrow that will fix this.

Quoted 78 lines · Zdenek Kabelac

Hello

Looks to me like the latest 2.4.5-pre1 is not creating coredump for multithreaded aplications:

----

$ ulimit  -a
core file size (blocks)     unlimited
data seg size (kbytes)      unlimited
file size (blocks)          unlimited
max locked memory (kbytes)  unlimited
max memory size (kbytes)    unlimited
open files                  1024
pipe size (512 bytes)       8
stack size (kbytes)         8192
cpu time (seconds)          unlimited
max user processes          6144
virtual memory (kbytes)     unlimited
$ ./testcore 
Segmentation fault

--- test program --

#include <stdio.h>
#include <pthread.h>

#define TASKS 10

void* thread(void* arg)

{
    printf("yuk %s\n", 0xffff0000);
    return 0;
}

int main(int argc, char *argv[])

{
    pthread_t task[TASKS];
    void* ret;
    int i = 0;
    for(i = 0; i < TASKS; i++)
	pthread_create(&task[i], NULL, thread, 0);
    for(i = 0; i < TASKS; i++)
	pthread_join(task[i], &ret);

    return 0;
}

---

for single threadad apps there is no problem:
(just put printf after int i declaration)

$ ./testcore 
Segmentation fault (core dumped)

---

Am I doing something wrong ???

Also for quite a long time Alan's AC patches were efictively locking my machine when threaded application was crashing - resolvin almost original behaviour solved this problem - but I like per-thread coredump - is there any special reason why this is still not present in the current 2.4.4 kernel as I can see this as quite usefull.

bye

[email protected]

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Signature · 4 lines
Don Dugger
"Censeo Toto nos in Kansa esse decisse." - D. Gale
[email protected]
Ph: 303/938-9838