Re: [PATCH 11/23] Update ARM AArch64 VM parameter definitions for bsd-user

[Warner Losh <imp@bsdimp.com>]

[-- Attachment #1: Type: text/plain, Size: 1658 bytes --]

On Tue, Jun 18, 2024 at 4:16 PM Richard Henderson <
richard.henderson@linaro.org> wrote:

> On 6/17/24 11:57, Ajeet Singh wrote:
> > From: Stacey Son <sson@FreeBSD.org>
> >
> > Defined address spaces for FreeBSD/arm64 and added function for
> > getting stack pointer from CPU and setting a return value.
> >
> > Signed-off-by: Stacey Son <sson@FreeBSD.org>
> > Signed-off-by: Warner Losh <imp@bsdimp.com>
> > Signed-off-by: Ajeet Singh <itachis@FreeBSD.org>
> > Co-authored-by: Sean Bruno <sbruno@freebsd.org>
> > Co-authored-by: Warner Losh <imp@bsdimp.com>
> > ---
> >   bsd-user/aarch64/target_arch_vmparam.h | 68 ++++++++++++++++++++++++++
> >   1 file changed, 68 insertions(+)
> >   create mode 100644 bsd-user/aarch64/target_arch_vmparam.h
>
> Acked-by: Richard Henderson <richard.henderson@linaro.org>
>
> > +                /* KERNBASE - 512 MB */
> > +#define TARGET_VM_MAXUSER_ADDRESS   (0x00007fffff000000ULL - (512 *
> MiB))
> > +#define TARGET_USRSTACK             TARGET_VM_MAXUSER_ADDRESS
>
> I will note that this may conflict with -R reserved_size,
> and is an existing issue with the x86_64 port as well.
>

There are indeed existing issues with address space management. We're
working through
them right now in the blitz branch. We have finally found where the atomic
issues were
coming from and it is <blush> not setting the flag saying we want atomic
ops when creating
the CPU structures (that's a quick summary, I'll post more on this later
when we review it).
So I'd suggest, for the moment, allowing this in and fixing it when we get
those details
ironed out. Does that sound OK?

Warner

[-- Attachment #2: Type: text/html, Size: 2462 bytes --]