Re: breakage: flex mmap patch for x86-64

[Ricky Beam]

On Sat, 6 Nov 2004, Andi Kleen wrote:
>On Sat, Nov 06, 2004 at 10:50:27AM +0100, Rafael J. Wysocki wrote:
>> On Saturday 06 of November 2004 10:12, Andi Kleen wrote:
>> > >  static inline int mmap_is_legacy(void)
>> > >  {
>> > > +       if (test_thread_flag(TIF_IA32))
>> > > +               return 1;
>> >
>> > That's definitely not the right fix because for 32bit you need flexmmap
>> > more than for 64bit because it gives you more address space.
>>
>> So let's call it temporary, but I like 32-bit apps having less address space
>> rather than segfaulting.
>
>If you want a temporary fix use the appended one.  But I think Linus pulled it anyways.

Right, wrong, whatever.  Using the legacy mmap for IA32 works.  And it works
exactly as before flexmmap was added.  Setting "vm.legacy_va_layout" will
work, but disables flexmmap across the board.

What is it with flexmmap that's causing 32bit apps to fail?

--Ricky


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/