Your patch indeed fixed most of the problems I was having but
unfortunately not all of them. Namely a simple "hello.c" program linked
with more than libc now works but complex ones like vim or crash or
cscope still coredump (whilst working very happily on 2.3.24) later on in
a mmap-related areas.
Anyway, now that I can work with 2.3.32-pre2 (+your patch) it is easier
to investigate this further,
Btw, with parenthises as below it silences gcc:
--- mmap.c.0 Sun Dec 12 11:46:52 1999
+++ mmap.c Sun Dec 12 11:54:25 1999
@@ -551,8 +551,7 @@
mpnt->vm_page_prot = area->vm_page_prot;
mpnt->vm_flags = area->vm_flags;
mpnt->vm_ops = area->vm_ops;
- mpnt->vm_pgoff = area->vm_pgoff;
- area->vm_pgoff += (end - area->vm_start) >> PAGE_SHIFT;
+ mpnt->vm_pgoff = area->vm_pgoff + ((end - area->vm_start) >> PAGE_SHIFT);
mpnt->vm_file = area->vm_file;
mpnt->vm_private_data = area->vm_private_data;
if (mpnt->vm_file)
Thanks again guys,
Tigran.
On Sat, 11 Dec 1999, Benjamin C.R. LaHaise wrote:
> On Sat, 11 Dec 1999, Eleonora Autore wrote:
>
> > Hi
> >
> > I have just had a look at the 2.3.24->30 kernels searching for whatever
> > broke the mmap(2) system call and narrowed it down to 24->25 changes.
> > There are quite a lot of mm changes in there so I am still narrowing it
> > down further.
>
> Thanks for narrowing it down. Eyeballing the patch came up with the
> following typo (doh!) introduced in 2.3.25.
>
> -ben
>
> diff -ur clean/2.3.32pre2/mm/mmap.c 2.3.32pre2/mm/mmap.c
> --- clean/2.3.32pre2/mm/mmap.c Mon Dec 6 23:10:56 1999
> +++ 2.3.32pre2/mm/mmap.c Sat Dec 11 16:11:07 1999
> @@ -545,8 +545,7 @@
> mpnt->vm_page_prot = area->vm_page_prot;
> mpnt->vm_flags = area->vm_flags;
> mpnt->vm_ops = area->vm_ops;
> - mpnt->vm_pgoff = area->vm_pgoff;
> - area->vm_pgoff += (end - area->vm_start) >> PAGE_SHIFT;
> + mpnt->vm_pgoff = area->vm_pgoff + (end - area->vm_start) >> PAGE_SHIFT;
> mpnt->vm_file = area->vm_file;
> mpnt->vm_private_data = area->vm_private_data;
> if (mpnt->vm_file)
>
>
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/