Re: [PATCH] mm: fix possible cause of a page_mapped BUG

["Robert Święcki" <robert@swiecki.net>]

On Thu, Apr 7, 2011 at 4:24 PM, Hugh Dickins <hughd@google.com> wrote:
> On Thu, 7 Apr 2011, Robert Swiecki wrote:
>> >
>> > Testing with Linus' patch. Will let you know in a few hours.
>>
>> Ok, nothing happened after ~20h. The bug, usually, was triggered within 5-10h.
>>
>> I can add some printk in this condition, and let it run for a few days
>> (I will not have access to my testing machine throughout that time),
>> if you think this will confirm your hypothesis.
>
> That's great, thanks Robert.  If the machine has nothing better to do,
> then it would be nice to let it run a little longer (a few days if that's
> what suits you), but it does look good so far.  Though I'm afraid you'll
> now discover something else entirely ;)

Ok, I added printk here:

        if (new_len > old_len) {
                unsigned long pgoff;

                if (vma->vm_flags & (VM_DONTEXPAND | VM_PFNMAP))
                        goto Efault;
                pgoff = (addr - vma->vm_start) >> PAGE_SHIFT;
                pgoff += vma->vm_pgoff;
                if (pgoff + (new_len >> PAGE_SHIFT) < pgoff) {
                        printk("VMA_TO_RESIZE: ADDR:%lx OLD_LEN:%lx
NEW_LEN:%lx PGOFF: %lx VMA->VM_START:%lx VMA->VM_FLAGS:%lx",
                                addr, old_len, new_len, pgoff,
vma->vm_start, vma->vm_flags);

                        goto Einval;
                }
        }


and after a few mins of fuzzing I get:

[  584.224028] VMA_TO_RESIZE: ADDR:f751f000 OLD_LEN:6000 NEW_LEN:c000
PGOFF: fffffffffffffffa VMA->VM_START:f751f000 VMA->VM_FLAGS:2321fa
[  639.777561] VMA_TO_RESIZE: ADDR:f751f000 OLD_LEN:6000 NEW_LEN:b000
PGOFF: fffffffffffffffa VMA->VM_START:f751f000 VMA->VM_FLAGS:2301f8

So, if this case is not caught later on in the code, I guess it solves
the problem. During the fuzzing I didn't experience any panic's, but
some other problems arose, i.e. cannot read /proc/<pid>/maps for some
processes (sys_read hangs, and such process cannot be killed or
stopped with any signal, still it's running (R state) and using CPU -
I'll submit another report for that).

-- 
Robert Święcki

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>