From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: Leif Lindholm <leif.lindholm@linaro.org>
Cc: "edk2-devel@lists.01.org" <edk2-devel@lists.01.org>
Subject: Re: [PATCH 2/5] ArmPkg/ArmMmuLib AARCH64: get rid of needless TLB invalidation
Date: Wed, 23 Jan 2019 16:55:28 +0100 [thread overview]
Message-ID: <CAKv+Gu8T4BEM52SakL00OVAAi1zRLkMpQF1WZjeknSAZLr6J-g@mail.gmail.com> (raw)
In-Reply-To: <20190123154622.pibw6mi5gh6ywb26@bivouac.eciton.net>
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/Include/Library/ArmMmuLib.h b/ArmPkg/Include/Library/ArmMmuLib.h
> > index fb7fd006417c..d2725810f1c6 100644
> > --- a/ArmPkg/Include/Library/ArmMmuLib.h
> > +++ b/ArmPkg/Include/Library/ArmMmuLib.h
> > @@ -59,7 +59,8 @@ VOID
> > EFIAPI
> > ArmReplaceLiveTranslationEntry (
> > IN UINT64 *Entry,
> > - IN UINT64 Value
> > + IN UINT64 Value,
> > + IN UINT64 Address
> > );
> >
> > EFI_STATUS
> > diff --git a/ArmPkg/Library/ArmLib/AArch64/ArmLibSupport.S b/ArmPkg/Library/ArmLib/AArch64/ArmLibSupport.S
> > index b7173e00b039..175fb58206b6 100644
> > --- a/ArmPkg/Library/ArmLib/AArch64/ArmLibSupport.S
> > +++ b/ArmPkg/Library/ArmLib/AArch64/ArmLibSupport.S
> > @@ -124,15 +124,15 @@ ASM_FUNC(ArmSetMAIR)
> > // IN VOID *MVA // X1
> > // );
> > ASM_FUNC(ArmUpdateTranslationTableEntry)
> > - dc civac, x0 // Clean and invalidate data line
> > - dsb sy
> > + dsb nshst
> > + lsr x1, x1, #12
> > EL1_OR_EL2_OR_EL3(x0)
> > 1: tlbi vaae1, x1 // TLB Invalidate VA , EL1
> > b 4f
> > 2: tlbi vae2, x1 // TLB Invalidate VA , EL2
> > b 4f
> > 3: tlbi vae3, x1 // TLB Invalidate VA , EL3
> > -4: dsb sy
> > +4: dsb nsh
> > isb
> > ret
> >
> > 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.
> - 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.
> /
> Leif
>
> > }
> > } else {
> > if (IndexLevel != PageLevel) {
> > @@ -375,6 +377,8 @@ UpdateRegionMapping (
> > *BlockEntry &= BlockEntryMask;
> > *BlockEntry |= (RegionStart & TT_ADDRESS_MASK_BLOCK_ENTRY) | Attributes | Type;
> >
> > + ArmUpdateTranslationTableEntry (BlockEntry, (VOID *)RegionStart);
> > +
> > // Go to the next BlockEntry
> > RegionStart += BlockEntrySize;
> > RegionLength -= BlockEntrySize;
> > @@ -487,9 +491,6 @@ ArmSetMemoryAttributes (
> > return Status;
> > }
> >
> > - // Invalidate all TLB entries so changes are synced
> > - ArmInvalidateTlb ();
> > -
> > return EFI_SUCCESS;
> > }
> >
> > @@ -512,9 +513,6 @@ SetMemoryRegionAttribute (
> > return Status;
> > }
> >
> > - // Invalidate all TLB entries so changes are synced
> > - ArmInvalidateTlb ();
> > -
> > return EFI_SUCCESS;
> > }
> >
> > diff --git a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibReplaceEntry.S b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibReplaceEntry.S
> > index 90192df24f55..d40c19b2e3e5 100644
> > --- a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibReplaceEntry.S
> > +++ b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibReplaceEntry.S
> > @@ -32,11 +32,12 @@
> > dmb sy
> > dc ivac, x0
> >
> > - // flush the TLBs
> > + // flush translations for the target address from the TLBs
> > + lsr x2, x2, #12
> > .if \el == 1
> > - tlbi vmalle1
> > + tlbi vaae1, x2
> > .else
> > - tlbi alle\el
> > + tlbi vae\el, x2
> > .endif
> > dsb sy
> >
> > @@ -48,12 +49,13 @@
> > //VOID
> > //ArmReplaceLiveTranslationEntry (
> > // IN UINT64 *Entry,
> > -// IN UINT64 Value
> > +// IN UINT64 Value,
> > +// IN UINT64 Address
> > // )
> > ASM_FUNC(ArmReplaceLiveTranslationEntry)
> >
> > // disable interrupts
> > - mrs x2, daif
> > + mrs x4, daif
> > msr daifset, #0xf
> > isb
> >
> > @@ -69,7 +71,7 @@ ASM_FUNC(ArmReplaceLiveTranslationEntry)
> > b 4f
> > 3:__replace_entry 3
> >
> > -4:msr daif, x2
> > +4:msr daif, x4
> > ret
> >
> > ASM_GLOBAL ASM_PFX(ArmReplaceLiveTranslationEntrySize)
> > --
> > 2.20.1
> >
next prev parent reply other threads:[~2019-01-23 15:55 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-07 7:14 [PATCH 0/5] memory/MMU hardening for AArch64 Ard Biesheuvel
2019-01-07 7:15 ` [PATCH 1/5] ArmPkg/ArmMmuLib AARCH64: fix out of bounds access Ard Biesheuvel
2019-01-14 12:00 ` Leif Lindholm
2019-01-14 18:48 ` Ard Biesheuvel
2019-01-07 7:15 ` [PATCH 2/5] ArmPkg/ArmMmuLib AARCH64: get rid of needless TLB invalidation Ard Biesheuvel
2019-01-23 15:46 ` Leif Lindholm
2019-01-23 15:55 ` Ard Biesheuvel [this message]
2019-01-23 16:12 ` Leif Lindholm
2019-01-23 16:16 ` Ard Biesheuvel
2019-01-23 16:20 ` Leif Lindholm
2019-01-28 12:29 ` Ard Biesheuvel
2019-01-28 18:01 ` Leif Lindholm
2019-01-29 10:32 ` Ard Biesheuvel
2019-01-07 7:15 ` [PATCH 3/5] ArmPkg/ArmMmuLib AARCH64: implement support for EFI_MEMORY_RP permissions Ard Biesheuvel
2019-01-14 14:29 ` Leif Lindholm
2019-01-14 14:59 ` Ard Biesheuvel
2019-01-14 15:06 ` Leif Lindholm
2019-01-07 7:15 ` [PATCH 4/5] ArmPkg/ArmMmuLib AARCH64: add support for read-only page tables Ard Biesheuvel
2019-01-07 7:15 ` [PATCH 5/5] ArmPkg/CpuDxe: switch to read-only page tables at EndOfDxe Ard Biesheuvel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-list from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAKv+Gu8T4BEM52SakL00OVAAi1zRLkMpQF1WZjeknSAZLr6J-g@mail.gmail.com \
--to=devel@edk2.groups.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox