Re: [PATCH 2/5] ArmPkg/ArmMmuLib AARCH64: get rid of needless TLB invalidation

[Ard Biesheuvel <ard.biesheuvel@linaro.org>]

On Wed, 23 Jan 2019 at 17:13, Leif Lindholm <leif.lindholm@linaro.org> wrote:
>
> On Wed, Jan 23, 2019 at 04:55:28PM +0100, Ard Biesheuvel wrote:
> > On Wed, 23 Jan 2019 at 16:46, Leif Lindholm <leif.lindholm@linaro.org> wrote:
> > >
> > > On Mon, Jan 07, 2019 at 08:15:01AM +0100, Ard Biesheuvel wrote:
> > > > Currently, we always invalidate the TLBs entirely after making
> > > > any modification to the page tables. Now that we have introduced
> > > > strict memory permissions in quite a number of places, such
> > > > modifications occur much more often, and it is better for performance
> > > > to flush only those TLB entries that are actually affected by
> > > > the changes.
> > > >
> > > > Contributed-under: TianoCore Contribution Agreement 1.1
> > > > Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> > > > ---
> > > >  ArmPkg/Include/Library/ArmMmuLib.h                       |  3 ++-
> > > >  ArmPkg/Library/ArmLib/AArch64/ArmLibSupport.S            |  6 +++---
> > > >  ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c         | 16 +++++++---------
> > > >  ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibReplaceEntry.S | 14 ++++++++------
> > > >  4 files changed, 20 insertions(+), 19 deletions(-)
> > > >
> > > > diff --git a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
> > > > index d66df3e17a02..e1fabfcbea14 100644
> > > > --- a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
> > > > +++ b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
> > > > @@ -129,13 +129,14 @@ STATIC
> > > >  VOID
> > > >  ReplaceLiveEntry (
> > > >    IN  UINT64  *Entry,
> > > > -  IN  UINT64  Value
> > > > +  IN  UINT64  Value,
> > > > +  IN  UINT64  Address
> > > >    )
> > > >  {
> > > >    if (!ArmMmuEnabled ()) {
> > > >      *Entry = Value;
> > > >    } else {
> > > > -    ArmReplaceLiveTranslationEntry (Entry, Value);
> > > > +    ArmReplaceLiveTranslationEntry (Entry, Value, Address);
> > > >    }
> > > >  }
> > > >
> > > > @@ -296,7 +297,8 @@ GetBlockEntryListFromAddress (
> > > >
> > > >          // Fill the BlockEntry with the new TranslationTable
> > > >          ReplaceLiveEntry (BlockEntry,
> > > > -          ((UINTN)TranslationTable & TT_ADDRESS_MASK_DESCRIPTION_TABLE) | TableAttributes | TT_TYPE_TABLE_ENTRY);
> > > > +          (UINTN)TranslationTable | TableAttributes | TT_TYPE_TABLE_ENTRY,
> > > > +          RegionStart);
> > >
> >
> > /me pages in the data ...
> >
> > > OK, this whole patch took a few times around the loop before I think I
> > > caught on what was happening.
> > >
> > > I think I'm down to the only things confusing me being:
> > > - The name "Address" to refer to something that is always the start
> > >   address of a 4KB-aligned translation region.
> > >   Is this because the function will be usable in other contexts in
> > >   later patches?
> >
> > I could change it to VirtualAddress if you prefer.
> > It is only passed
> > for TLB maintenance which is only needed at page granularity, and the
> > low bits are shifted out anyway.
>
> Yeah, exactly. It would just be nice if the name reflected that. Not
> sure VirtualAddress does. VirtualBase? PageBase?
>

Well, the argument passed in is called RegionStart, so let's just
stick with that.

> > > - Why drop the & TT_ADDRESS_MASK_DESCRIPTION_TABLE bit here?
> > >   Was it just always pointless and you decided to drop it while you
> > >   were at it?
> >
> > IIRC yes. It is a newly allocated page, so the masking does not do anything.
>
> Yeah, that's fair enough.
> Just made me wonder if anything unobvious was going on :)
>
> /
>     Leif