Hello.
I found the following code fails on arm processor,
while it works on x86 processor.
Kernel version is 2.4.31-a9-3 (for Armadillo-9).
--- Code start ---
#include <stdio.h>
#include <sys/types.h>
#include <sys/wait.h>
#include <unistd.h>
#include <sched.h>
#include <errno.h>
#include <stdlib.h>
static int child(void *arg) {
_exit(1);
}
int main(int argc, char *argv[]) {
//char *stack = malloc(1024);
//const pid_t pid = clone(child, stack + (1024 / 2), CLONE_NEWNS, (void *) argv);
static char stack[1024];
const pid_t pid = clone(child, stack + (sizeof(stack) / 2), CLONE_NEWNS, (void *) argv);
while (waitpid(pid, NULL, __WALL) == EOF && errno == EINTR);
return 0;
}
--- Code end ---
The __do_user_fault() in arch/arm/mm/fault-common.c is called by clone(2).
--- dmesg start ---
newns: unhandled page fault at pc=0x40007ad4, lr=0x40007aa0 (bad address=0x8c4001ca, code 5)
pc : [<40007ad4>] lr : [<40007aa0>] Not tainted
sp : 000107f9 ip : 00000000 fp : 00010885
r10: 4001c6b8 r9 : 0065cf04 r8 : 0001089d
r7 : 00008164 r6 : 00008260 r5 : 00010825 r4 : fffffc44
r3 : 0000006f r2 : 4001c2fc r1 : 8c4001ca r0 : 00000000
Flags: Nzcv IRQs on FIQs on Mode USER_32 Segment user
Control: C000317F Table: C56EC000 DAC: 00000015
--- dmesg end ---
The above code works if I use malloc(3).
Is passing 'statically' allocated stack memory to clone(2) illegal?
Regards.
--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive: http://mail.nl.linux.org/kernelnewbies/
FAQ: http://kernelnewbies.org/faq/
