Re: Kernel BUG linux-2.5.67

["dada1" <dada1@cosmosbay.com>]

Yes... the futex was placed un a hugetlb page...

Thanks Andrew, I will test your patch very soon.

From: "Andrew Morton" <akpm@digeo.com>
> "dada1" <dada1@cosmosbay.com> wrote:
> >
> > Hello
> > 
> > I tried linux-2.5.67 this morning...
> > 
> > instant oops with a small multi-threaded program using futex()
> 
> Was the futex placed inside a hugetlb page?  Please say yes.
> 
> There is a stunning bug.
> 
> --- 25/include/linux/mm.h~a 2003-04-08 02:03:19.000000000 -0700
> +++ 25-akpm/include/linux/mm.h 2003-04-08 02:03:24.000000000 -0700
> @@ -231,8 +231,8 @@ static inline void get_page(struct page 
>  static inline void put_page(struct page *page)
>  {
>   if (PageCompound(page)) {
> + page = (struct page *)page->lru.next;
>   if (put_page_testzero(page)) {
> - page = (struct page *)page->lru.next;
>   if (page->lru.prev) { /* destructor? */
>   (*(void (*)(struct page *))page->lru.prev)(page);
>   } else {
> 
> _
>