Ooops in filemap_write_page in test8

Ooops in filemap_write_page in test8

[Juan J. Quintela]

Hi
        I was running mmap001 over NFS when I got one Oops, with the
        following backtrace.  The problem is that the page->mapping is
        NULL, and it causes a NULL access at filemap_write_page.
        If you need any more info, let me know.

Later, Juan.



0xc29ff574 0xc012229f filemap_write_page+0xb (0xc760fec0, 0xc11a9540, 0x0)      
                               kernel .text 0xc0100000 0xc0122294 0xc01222ac    
           0xc01222be filemap_swapout+0x12 (0xc11a9540, 0xc760fec0)             
                               kernel .text 0xc0100000 0xc01222ac 0xc01222d0    
           0xc0126ca0 try_to_swap_out+0x150 (0xc1298120, 0xc2522c20, 0x4015d000,
 0xc29ff574, 0x4)                                                               
                               kernel .text 0xc0100000 0xc0126b50 0xc0126d70    
           0xc0126e83 swap_out_vma+0x113 (0xc1298120, 0xc2522c20, 0x4015d000, 0x
4)                                                                              
                               kernel .text 0xc0100000 0xc0126d70 0xc0126f08    
           0xc0126f4c swap_out_mm+0x44 (0xc1298120, 0x4)                        
                               kernel .text 0xc0100000 0xc0126f08 0xc0126f74    
           0xc012705f swap_out+0xeb (0x3c, 0x4)                                 
                               kernel .text 0xc0100000 0xc0126f74 0xc01270e4    
           0xc0127276 do_try_to_free_pages+0x192 (0x4, 0x10f00, 0xc1255fb8, 0x0)
                               kernel .text 0xc0100000 0xc01270e4 0xc012730c    
           0xc01273a9 kswapd+0x9d                                               
                               kernel .text 0xc0100000 0xc012730c 0xc01273b0    
           0xc0107488 kernel_thread+0x28                                        
                               kernel .text 0xc0100000 0xc0107460 0xc0107498    

-- 
In theory, practice and theory are the same, but in practice they 
are different -- Larry McVoy
--
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.eu.org/Linux-MM/

Re: Ooops in filemap_write_page in test8

[Petr Vandrovec]

On 11 Sep 00 at 17:00, Juan J. Quintela wrote:
>         I was running mmap001 over NFS when I got one Oops, with the
>         following backtrace.  The problem is that the page->mapping is
>         NULL, and it causes a NULL access at filemap_write_page.
>         If you need any more info, let me know.

Hi Juan,
  is your machine near to VMware or not? I reported same oopses
last week on linux-kernel - they happened after heavy swapped VMware session
on VMware exit (when exit_mmap was cleaning up address space). If you
have idea where mapping gets set to NULL (and why is such page passed
to filemap_write_page), I'd like to know it.
                                          Thanks,
                                            Petr Vandrovec
                                            vandrove@vc.cvut.cz
                                            
--
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.eu.org/Linux-MM/

Re: Ooops in filemap_write_page in test8

[Juan J. Quintela]

>>>>> "petr" == Petr Vandrovec <VANDROVE@vc.cvut.cz> writes:

petr> On 11 Sep 00 at 17:00, Juan J. Quintela wrote:
>> I was running mmap001 over NFS when I got one Oops, with the
>> following backtrace.  The problem is that the page->mapping is
>> NULL, and it causes a NULL access at filemap_write_page.
>> If you need any more info, let me know.

petr> Hi Juan,
petr>   is your machine near to VMware or not? I reported same oopses
petr> last week on linux-kernel - they happened after heavy swapped VMware session
petr> on VMware exit (when exit_mmap was cleaning up address space). If you
petr> have idea where mapping gets set to NULL (and why is such page passed
petr> to filemap_write_page), I'd like to know it.
petr>                                           Thanks,
petr>                                             Petr Vandrovec
petr>                                             vandrove@vc.cvut.cz

I don't have vmware here, I am using NFSv3 (kernel NFS).  I am
investigating where the page puts ->mapping to NULL.  The Oops happend
because a page is in a vma with address operations, but the page
hasn't a mapping :((((

I continue working on that ....

Later, Juan.

-- 
In theory, practice and theory are the same, but in practice they 
are different -- Larry McVoy
--
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.eu.org/Linux-MM/