[RESEND RFC PATCH v2] arm64: Exposes support for 32-bit syscalls

[David Laight]

From: Catalin Marinas
> Sent: 12 February 2021 13:28
> 
> On Fri, Feb 12, 2021 at 12:35:15PM +0000, Mark Brown wrote:
> > On Fri, Feb 12, 2021 at 11:30:41AM +0000, Steven Price wrote:
> > > On 11/02/2021 20:21, sonicadvance1 at gmail.com wrote:
> > > > Why do we need compatibility layers?
> > > > There are ARMv8 CPUs that only support AArch64 but still need to run
> > > > AArch32 applications.
> > > > Cortex-A34/R82 and other cores are prime examples of this.
> > > > Additionally if a user is needing to run legacy 32-bit x86 software, it
> > > > needs the same compatibility layer.
> >
> > > Unless I'm much mistaken QEMU's user mode already does this - admittedly I
> > > don't tend to run "legacy 32-bit x86 software".
> >
> > Yes, this has been deployed on Debian for a long time - you can install
> > any combination of Debian architectures on a single system and it will
> > use qemu to run binaries that can't be supported natively by the
> > hardware.
> 
> The only downside I think is that for some syscalls it's not that
> efficient. Those using struct iovec come to mind, qemu probably
> duplicates the user structures, having to copy them in both directions
> (well, the kernel compat layer does something similar).
> 
> Anyway, I'm not in favour of this patch. Those binary translation tools
> need to explore the user-only options first and come up with some perf
> numbers to justify the proposal.

I don't think the problem is only the performance.
The difficulty is knowing when structures need changing.
A typical example is driver ioctl requests.
Any user space adaption layer would have to know which actual
driver has been opened and what internal structures it has.
Getting that right is hard and difficult.
The recent changes to move (IIRC) sockopt compatibility down
into the protocol code found quite a few places where it was
previously broken.
It is much easier to get it right in the code that knows about
the actual structures.

For mmap() you certainly want to be able to limit the returned
address to 32 bits (or maybe 31.5 bits).
A MAP_BELOW flag could do that, but the 32bit syscall has to.
(I can't remember what is done for wine - which needs 31bit addresses).

Of course, that only helps for 32bit arm binaries - when the
kernel compat code is written for,
Trying to run x86 binaries adds extra complexity.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)