Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1616873 > unrolled thread

Re: [PATCH 21/26] x86/mm: add support of additional page table level during early boot

Started by"Kirill A. Shutemov" <kirill@shutemov.name>
First post2017-04-05 14:10 +0200
Last post2017-04-05 17:40 +0200
Articles 2 — 1 participant

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH 21/26] x86/mm: add support of additional page table level  during early boot "Kirill A. Shutemov" <kirill@shutemov.name> - 2017-04-05 14:10 +0200
    Re: [PATCH 21/26] x86/mm: add support of additional page table level  during early boot "Kirill A. Shutemov" <kirill@shutemov.name> - 2017-04-05 17:40 +0200

#1616873 — Re: [PATCH 21/26] x86/mm: add support of additional page table level during early boot

From"Kirill A. Shutemov" <kirill@shutemov.name>
Date2017-04-05 14:10 +0200
SubjectRe: [PATCH 21/26] x86/mm: add support of additional page table level during early boot
Message-ID<tsRV7-2sp-3@gated-at.bofh.it>
On Mon, Mar 13, 2017 at 08:18:10AM +0100, Ingo Molnar wrote:
> 
> * Kirill A. Shutemov <kirill.shutemov@linux.intel.com> wrote:
> 
> > This patch adds support for 5-level paging during early boot.
> > It generalizes boot for 4- and 5-level paging on 64-bit systems with
> > compile-time switch between them.
> > 
> > Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> > ---
> >  arch/x86/boot/compressed/head_64.S          | 23 +++++++++--
> >  arch/x86/include/asm/pgtable.h              |  2 +-
> >  arch/x86/include/asm/pgtable_64.h           |  6 ++-
> >  arch/x86/include/uapi/asm/processor-flags.h |  2 +
> >  arch/x86/kernel/espfix_64.c                 |  2 +-
> >  arch/x86/kernel/head64.c                    | 40 +++++++++++++-----
> >  arch/x86/kernel/head_64.S                   | 63 +++++++++++++++++++++--------
> 
> Ok, here I'd like to have a C version instead of further complicating an already 
> complex assembly version...

Just head up: I work on this.

It's great deal of frustration (I can't really read assembly), but I'm
slowly moving forward.

Most of logic in startup_64 in arch/x86/kernel/head_64.S is converted
to C. Dealing with secondary_startup_64 now.

Not sure if it's possible to convert code in
arch/x86/boot/compressed/head_64.S to C.

Assembly code there is in 32-bit mode, but if we move it to C it will
compiled as 64-bit. I've tried to put it in separate translation unit and
compile with -m32, but then link phase breaks as object files have
different types.

Any suggestion how I can get out of the situation?

-- 
 Kirill A. Shutemov

[toc] | [next] | [standalone]


#1617069

From"Kirill A. Shutemov" <kirill@shutemov.name>
Date2017-04-05 17:40 +0200
Message-ID<tsVcm-4qr-21@gated-at.bofh.it>
In reply to#1616873
On Wed, Apr 05, 2017 at 02:36:24PM +0300, Kirill A. Shutemov wrote:
> On Mon, Mar 13, 2017 at 08:18:10AM +0100, Ingo Molnar wrote:
> > 
> > * Kirill A. Shutemov <kirill.shutemov@linux.intel.com> wrote:
> > 
> > > This patch adds support for 5-level paging during early boot.
> > > It generalizes boot for 4- and 5-level paging on 64-bit systems with
> > > compile-time switch between them.
> > > 
> > > Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> > > ---
> > >  arch/x86/boot/compressed/head_64.S          | 23 +++++++++--
> > >  arch/x86/include/asm/pgtable.h              |  2 +-
> > >  arch/x86/include/asm/pgtable_64.h           |  6 ++-
> > >  arch/x86/include/uapi/asm/processor-flags.h |  2 +
> > >  arch/x86/kernel/espfix_64.c                 |  2 +-
> > >  arch/x86/kernel/head64.c                    | 40 +++++++++++++-----
> > >  arch/x86/kernel/head_64.S                   | 63 +++++++++++++++++++++--------
> > 
> > Ok, here I'd like to have a C version instead of further complicating an already 
> > complex assembly version...
> 
> Just head up: I work on this.
> 
> It's great deal of frustration (I can't really read assembly), but I'm
> slowly moving forward.
> 
> Most of logic in startup_64 in arch/x86/kernel/head_64.S is converted
> to C. Dealing with secondary_startup_64 now.

I'm stuck with secondary_startup_64 too. Stack breaks as soon as we switch
to new page tables when onlining secondary CPUs. I don't know how to get
around this.

Would it be sufficient if I only convert startup_64 in
arch/x86/kernel/head_64.S to C?

-- 
 Kirill A. Shutemov

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web