On Tuesday 21 August 2012, Catalin Marinas wrote:

On Thu, Aug 16, 2012 at 01:37:53PM +0100, Arnd Bergmann wrote:

quoted

On Thursday 16 August 2012, Will Deacon wrote:

quoted

quoted

This looks wrong: PER_LINUX/PER_LINUX32 decides over the output of the
uname system call, while TIF_32BIT decides over the instruction set
when returning to user space. You definitely should not set the personality
to the value you pass from the elf loader. Instead, just do

#define SET_PERSONALITY(ex) clear_thread_flag(TIF_32BIT);
#defined COMPAT_SET_PERSONALITY(ex) set_thread_flag(TIF_32BIT);

In this case, won't uname be incorrect (aarch64l) for aarch32 tasks (which
expect something like armv8l)?

No, the uname output is meant to tell you about the system, not the
instruction set that you are using (you already know that in compiled
code).

OK, so we assumed that compat tasks should get a uname as close as
possible to a 32-bit system, i.e. armv8l, for full compatibility. This
would allow us to run something like 32-bit Debian on an AArch64 kernel
without worrying about any scripts failing.

You can still do that, just boot with init="/sbin/setarch armv7 /sbin/init".

But I can see on x86 that it always reports x86_64 even if the task is
x86_32.

Not just x86, the same behavior is used on powerpc, s390, mips, sparc and
parisc. Not sure about tile though.

	Arnd