svn commit: r332489 - in head: gnu/usr.bin/gdb/kgdb sys/conf sys/dev/dcons sys/dev/hyperv/vmbus/i386 sys/dev/ppc sys/dev/syscons sys/i386/conf sys/i386/i386 sys/i386/include sys/i386/include/pc sys...

[Konstantin Belousov]

On Sun, Apr 22, 2018 at 09:06:56PM +0200, Tijl Coosemans wrote:
> On Fri, 13 Apr 2018 20:30:49 +0000 (UTC) Konstantin Belousov <kib at FreeBSD.org> wrote:
> > Author: kib
> > Date: Fri Apr 13 20:30:49 2018
> > New Revision: 332489
> > URL: https://svnweb.freebsd.org/changeset/base/332489
> > 
> > Log:
> >   i386 4/4G split.
> >   
> >   The change makes the user and kernel address spaces on i386
> >   independent, giving each almost the full 4G of usable virtual addresses
> >   except for one PDE at top used for trampoline and per-CPU trampoline
> >   stacks, and system structures that must be always mapped, namely IDT,
> >   GDT, common TSS and LDT, and process-private TSS and LDT if allocated.
> 
> Could this have broken the linux futex syscall?  I have a linux program
> that gets stuck in linux_sys_futex and becomes unkillable.  Note that the
> routines in sys/i386/linux/linux_support.s try to do atomic operations on
> user space addresses.

Yes, it is quite possible.  I will try to look next week.