Re: mm: uninterruptable tasks hanged on mmap_sem

[Jiri Kosina]

On Tue, 2 Feb 2016, Dmitry Vyukov wrote:

> If the following program run in a parallel loop, eventually it leaves
> hanged uninterruptable tasks on mmap_sem.

I am now able to reproduce the issue and will be looking into it.

[ .. snip .. ]
> // autogenerated by syzkaller (http://github.com/google/syzkaller)
> #include <pthread.h>
> #include <stdint.h>
> #include <string.h>
> #include <sys/syscall.h>
> #include <unistd.h>
> 
> #ifndef SYS_mlock2
> #define SYS_mlock2 325
> #endif
> 
> long r[7];
> 
> void* thr(void* arg)
> {
>   switch ((long)arg) {
>   case 0:
>     r[0] = syscall(SYS_mmap, 0x20000000ul, 0x1000ul, 0x3ul, 0x32ul,
>                    0xfffffffffffffffful, 0x0ul);
>     break;
>   case 1:
>     memcpy((void*)0x20000000, "\x2f\x64\x65\x76\x2f\x66\x64\x23", 8);

The memcpy() can be removed and the problem still triggers reliably for 
me.

>  case 3:
>    r[4] = syscall(SYS_mmap, 0x20002000ul, 0x1000ul, 0x3ul, 0x812ul,
>                   r[2], 0x0ul);

0x812 made me wonder (it's not really meaningful flags value), but the bug 
triggers also with 0x12.

-- 
Jiri Kosina
SUSE Labs