Re: Undefined SysCall
- From: Paul Pluzhnikov <ppluzhnikov-nsp@xxxxxxxxxxx>
- Date: Sun, 15 Jul 2007 06:53:06 -0700
"Scott J. McCaughrin" <sjmccaug@xxxxxxxxxxxxxxx> writes:
I have a RedHat Linux V Enterprise box with GNU 4.1: while inspecting
the glib, I found that: pthreadlib.a contains a `pthread_create.o'
with the lines:
mov eax,0x137 ;set EAX = 311 (decimal)
int 0x80 ;syscall with undefined code
My x86_64 box only has syscall numbers up to 0x117 (279 decimal),
On Fedora Core 6 (kernel 2.6.19), /usr/include/asm-i386/unistd.h:
....
#define __NR_set_robust_list 311
#define __NR_get_robust_list 312
#define __NR_splice 313
#define __NR_sync_file_range 314
#define __NR_tee 315
#define __NR_vmsplice 316
#define __NR_move_pages 317
#define __NR_getcpu 318
#define __NR_epoll_pwait 319
#endif /* _ASM_I386_UNISTD_H_ */
It is a "standard" glibc technique to issue e.g.
syscall(__NR_set_robust_list), and fall back to other mechanism if
it returns ENOSYS.
Cheers,
--
In order to understand recursion you must first understand recursion.
Remove /-nsp/ for email.
.
- References:
- Undefined SysCall
- From: Scott J. McCaughrin
- Undefined SysCall
- Prev by Date: Re: Undefined SysCall
- Next by Date: Re: modularized kernels
- Previous by thread: Re: Undefined SysCall
- Next by thread: char device driver and bus ?
- Index(es):
Relevant Pages
|