[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMzpN2hvPHWYOeyzfpRmk39XYwCrSJx0UyqxE48F1TjTNyoKAg@mail.gmail.com>
Date: Fri, 24 Feb 2023 16:38:41 -0500
From: Brian Gerst <brgerst@...il.com>
To: Usama Arif <usama.arif@...edance.com>
Cc: linux-kernel@...r.kernel.org, x86@...nel.org,
David Woodhouse <dwmw2@...radead.org>,
Thomas Gleixner <tglx@...utronix.de>,
Borislav Petkov <bp@...en8.de>,
"H . Peter Anvin" <hpa@...or.com>,
Peter Zijlstra <peterz@...radead.org>,
Andy Lutomirski <luto@...nel.org>,
Ingo Molnar <mingo@...nel.org>
Subject: Re: [External] [PATCH v2 0/5] x86-64: Remove global variables from boot
On Fri, Feb 24, 2023 at 3:40 PM Usama Arif <usama.arif@...edance.com> wrote:
>
>
>
> On 24/02/2023 15:42, Brian Gerst wrote:
> > This is based on the parallel boot v11 series.
> >
> > Remove the global variables initial_gs, initial_stack, and
> > early_gdt_descr from the 64-bit boot code. The stack, GDT, and GSBASE
> > can be determined from the CPU number.
> >
> > v2: - Dropped first two patches which were merged into another patch in
> > the parallel boot series.
> > - Fixed a compile error in patch 1 found by the kernel test robot.
> > - Split out the removal of STARTUP_SECONDARY into a separate patch
> > and renumber the remaining flags.
> >
> > Brian Gerst (5):
> > x86/smpboot: Remove initial_stack on 64-bit
> > x86/smpboot: Remove early_gdt_descr on 64-bit
> > x86/smpboot: Remove initial_gs
> > x86/smpboot: Simplify boot CPU setup
> > x86/smpboot: Remove STARTUP_SECONDARY
> >
> > arch/x86/include/asm/processor.h | 6 +-
> > arch/x86/include/asm/realmode.h | 1 -
> > arch/x86/include/asm/smp.h | 5 +-
> > arch/x86/kernel/acpi/sleep.c | 5 +-
> > arch/x86/kernel/head_64.S | 99 ++++++++++++--------------------
> > arch/x86/kernel/smpboot.c | 6 +-
> > arch/x86/xen/xen-head.S | 2 +-
> > 7 files changed, 49 insertions(+), 75 deletions(-)
> >
>
> Its weird putting something in earlier patches like
> STARTUP_APICID_CPUID_*,STARTUP_SECONDARY.. and removing/changing it
> later on in *the same series*. Maybe better to keep this as a separate
> series from parallel smp? i.e. not include this in v12 and review it
> separately? Happy with whatever you and David decide..
>
> Thanks,
> Usama
Removing the globals before the parallel boot series, would be the
best option IMO. That would make the transition simpler.
--
Brian Gerst.
Powered by blists - more mailing lists