From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: Leif Lindholm <leif.lindholm@linaro.org>
Cc: Charles Garcia-Tobin <charles.garcia-tobin@arm.com>,
"edk2-devel@lists.01.org" <edk2-devel@lists.01.org>
Subject: Re: [PATCH 3/5] ArmPkg/ArmMmuLib AARCH64: implement support for EFI_MEMORY_RP permissions
Date: Mon, 14 Jan 2019 15:59:08 +0100 [thread overview]
Message-ID: <CAKv+Gu-oFsR=XdeWY1NrGnBMw4U=b7MyMpUz=DCnKaL=Uzr8Pg@mail.gmail.com> (raw)
In-Reply-To: <20190114142901.lkwgte6mlsv6bjyk@bivouac.eciton.net>
On Mon, 14 Jan 2019 at 15:29, Leif Lindholm <leif.lindholm@linaro.org> wrote:
>
> On Mon, Jan 07, 2019 at 08:15:02AM +0100, Ard Biesheuvel wrote:
> > Wire up the access flag (AF) page table attribute to the EFI_MEMORY_RP
> > permission attribute, so that attempts to read from such a region will
> > trigger an access flag fault.
> >
> > Note that this is a stronger notion than just read protection, since
> > it now implies that any write or execute attempt is trapped as well.
> > However, this does not really matter in practice since we never assume
> > that a read protected page is writable or executable, and StackGuard
> > and HeapGuard (which are the primary users of this facility) certainly
> > don't care.
>
> So ... I'm cautiously positive to this patch.
> But this use does contradict the UEFI spec (2.7a, 2.3.6.1 Memory
> types), which says EFI_MEMORY_RP is "not used or defined" for AArch64.
>
> Charles?
>
Not defined by the spec means we can use it do whatever we bloody want
with it, at least that is what a typical compiler engineer will tell
you :-)
I think there was a pending ECR to update the AArch64 binding code to
reflect reality, but I don't think we included EFI_MEMORY_RP.
> > Contributed-under: TianoCore Contribution Agreement 1.1
> > Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> > ---
> > ArmPkg/Drivers/CpuDxe/AArch64/Mmu.c | 5 +++--
> > ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c | 14 +++++++++++---
> > 2 files changed, 14 insertions(+), 5 deletions(-)
> >
> > diff --git a/ArmPkg/Drivers/CpuDxe/AArch64/Mmu.c b/ArmPkg/Drivers/CpuDxe/AArch64/Mmu.c
> > index 3e216c7cb235..e62e3fa87112 100644
> > --- a/ArmPkg/Drivers/CpuDxe/AArch64/Mmu.c
> > +++ b/ArmPkg/Drivers/CpuDxe/AArch64/Mmu.c
> > @@ -223,8 +223,9 @@ EfiAttributeToArmAttribute (
> > ArmAttributes = TT_ATTR_INDX_MASK;
> > }
> >
> > - // Set the access flag to match the block attributes
> > - ArmAttributes |= TT_AF;
> > + if ((EfiAttributes & EFI_MEMORY_RP) == 0) {
> > + ArmAttributes |= TT_AF;
> > + }
> >
> > // Determine protection attributes
> > if (EfiAttributes & EFI_MEMORY_RO) {
> > diff --git a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
> > index e1fabfcbea14..b59c081a7e49 100644
> > --- a/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
> > +++ b/ArmPkg/Library/ArmMmuLib/AArch64/ArmMmuLibCore.c
> > @@ -102,6 +102,10 @@ PageAttributeToGcdAttribute (
> > GcdAttributes |= EFI_MEMORY_XP;
> > }
> >
> > + if ((PageAttributes & TT_AF) == 0) {
> > + GcdAttributes |= EFI_MEMORY_RP;
> > + }
> > +
> > return GcdAttributes;
> > }
> >
> > @@ -451,7 +455,11 @@ GcdAttributeToPageAttribute (
> > PageAttributes |= TT_AP_RO_RO;
> > }
> >
> > - return PageAttributes | TT_AF;
> > + if ((GcdAttributes & EFI_MEMORY_RP) == 0) {
> > + PageAttributes |= TT_AF;
> > + }
> > +
> > + return PageAttributes;
> > }
> >
> > EFI_STATUS
> > @@ -474,9 +482,9 @@ ArmSetMemoryAttributes (
> > // No memory type was set in Attributes, so we are going to update the
> > // permissions only.
> > //
> > - PageAttributes &= TT_AP_MASK | TT_UXN_MASK | TT_PXN_MASK;
> > + PageAttributes &= TT_AP_MASK | TT_UXN_MASK | TT_PXN_MASK | TT_AF;
> > PageAttributeMask = ~(TT_ADDRESS_MASK_BLOCK_ENTRY | TT_AP_MASK |
> > - TT_PXN_MASK | TT_XN_MASK);
> > + TT_PXN_MASK | TT_XN_MASK | TT_AF);
> > }
> >
> > TranslationTable = ArmGetTTBR0BaseAddress ();
> > --
> > 2.20.1
> >
next prev parent reply other threads:[~2019-01-14 14:59 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
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 [this message]
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+Gu-oFsR=XdeWY1NrGnBMw4U=b7MyMpUz=DCnKaL=Uzr8Pg@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