From: Laszlo Ersek <lersek@redhat.com>
To: Ard Biesheuvel <ard.biesheuvel@linaro.org>, edk2-devel@lists.01.org
Cc: sigmaepsilon92@gmail.com
Subject: Re: [PATCH 1/2] ArmVirtPkg/HighMemDxe: use CPU arch protocol to apply memprotect policy
Date: Tue, 21 Mar 2017 11:36:08 +0100 [thread overview]
Message-ID: <19fedc88-4567-72c6-fca3-8c81e95c6d5e@redhat.com> (raw)
In-Reply-To: <1490088209-8564-2-git-send-email-ard.biesheuvel@linaro.org>
On 03/21/17 10:23, Ard Biesheuvel wrote:
> Instead of invoking gDS->SetMemorySpaceAttributes to set the EFI_MEMORY_XP
> attribute on newly added regions, which is guaranteed to fail if the same
> attribute was not declared as a capability of the region when it as added,
> invoke the CPU arch protocol directly to set the EFI_MEMORY_XP attribute
> if our memory protection policy demands it.
>
> Reported-by: Michael Zimmermann <sigmaepsilon92@gmail.com>
> Contributed-under: TianoCore Contribution Agreement 1.0
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> ---
> ArmVirtPkg/HighMemDxe/HighMemDxe.c | 31 +++++++++++++++-----
> ArmVirtPkg/HighMemDxe/HighMemDxe.inf | 1 +
> 2 files changed, 25 insertions(+), 7 deletions(-)
>
> diff --git a/ArmVirtPkg/HighMemDxe/HighMemDxe.c b/ArmVirtPkg/HighMemDxe/HighMemDxe.c
> index f70978f6414f..4e41120deff3 100644
> --- a/ArmVirtPkg/HighMemDxe/HighMemDxe.c
> +++ b/ArmVirtPkg/HighMemDxe/HighMemDxe.c
> @@ -20,6 +20,7 @@
> #include <Library/PcdLib.h>
> #include <Library/UefiBootServicesTableLib.h>
>
> +#include <Protocol/Cpu.h>
> #include <Protocol/FdtClient.h>
>
> EFI_STATUS
> @@ -30,6 +31,7 @@ InitializeHighMemDxe (
> )
> {
> FDT_CLIENT_PROTOCOL *FdtClient;
> + EFI_CPU_ARCH_PROTOCOL *Cpu;
> EFI_STATUS Status, FindNodeStatus;
> INT32 Node;
> CONST UINT32 *Reg;
> @@ -43,6 +45,10 @@ InitializeHighMemDxe (
> (VOID **)&FdtClient);
> ASSERT_EFI_ERROR (Status);
>
> + Status = gBS->LocateProtocol (&gEfiCpuArchProtocolGuid, NULL,
> + (VOID **)&Cpu);
> + ASSERT_EFI_ERROR (Status);
> +
> //
> // Check for memory node and add the memory spaces except the lowest one
> //
> @@ -78,13 +84,24 @@ InitializeHighMemDxe (
> continue;
> }
>
> + Status = gDS->SetMemorySpaceAttributes (CurBase, CurSize,
> + EFI_MEMORY_WB);
> + if (EFI_ERROR (Status)) {
> + DEBUG ((DEBUG_WARN,
> + "%a: gDS->SetMemorySpaceAttributes() failed on region 0x%lx - 0x%lx (%r)\n",
> + __FUNCTION__, CurBase, CurBase + CurSize - 1, Status));
> + }
> +
> + //
> + // Due to the ambiguous nature of the RO/XP GCD memory space attributes,
> + // it is impossible to add a memory space with the XP attribute in a way
> + // that does not result in the XP attribute being set on *all* UEFI
> + // memory map entries that are carved from it, including code regions
> + // that require executable permissions.
> //
> - // Take care not to strip any permission attributes that will have been
> - // set by DxeCore on the region we just added if a strict permission
> - // policy is in effect for EfiConventionalMemory regions.
> - // Unfortunately, we cannot interrogate the GCD memory space map for
> - // those permissions, since they are not recorded there (for historical
> - // reasons), so check the policy directly.
> + // So instead, we never set the RO/XP attributes in the GCD memory space
> + // capabilities or attribute fields, and apply any protections directly
> + // on the page table mappings by going through the cpu arch protocol.
> //
> Attributes = EFI_MEMORY_WB;
> if ((PcdGet64 (PcdDxeNxMemoryProtectionPolicy) &
> @@ -92,7 +109,7 @@ InitializeHighMemDxe (
> Attributes |= EFI_MEMORY_XP;
> }
>
> - Status = gDS->SetMemorySpaceAttributes (CurBase, CurSize, Attributes);
> + Status = Cpu->SetMemoryAttributes (Cpu, CurBase, CurSize, Attributes);
>
> if (EFI_ERROR (Status)) {
> DEBUG ((EFI_D_ERROR,
> diff --git a/ArmVirtPkg/HighMemDxe/HighMemDxe.inf b/ArmVirtPkg/HighMemDxe/HighMemDxe.inf
> index 89c743ebe058..ac1761974f52 100644
> --- a/ArmVirtPkg/HighMemDxe/HighMemDxe.inf
> +++ b/ArmVirtPkg/HighMemDxe/HighMemDxe.inf
> @@ -41,6 +41,7 @@ [LibraryClasses]
> UefiDriverEntryPoint
>
> [Protocols]
> + gEfiCpuArchProtocolGuid ## CONSUMES
> gFdtClientProtocolGuid ## CONSUMES
>
> [Pcd]
>
gEfiCpuArchProtocolGuid is already part of the DEPEX, so:
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
next prev parent reply other threads:[~2017-03-21 10:36 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-21 9:23 [PATCH 0/2] ArmVirtPkg/HighMemDxe: fix issues reported by Michael Ard Biesheuvel
2017-03-21 9:23 ` [PATCH 1/2] ArmVirtPkg/HighMemDxe: use CPU arch protocol to apply memprotect policy Ard Biesheuvel
2017-03-21 10:36 ` Laszlo Ersek [this message]
2017-03-21 9:23 ` [PATCH 2/2] ArmVirtPkg/HighMemDxe: check new regions against GCD memory space map Ard Biesheuvel
2017-03-21 10:40 ` Laszlo Ersek
2017-03-21 10:42 ` [PATCH 0/2] ArmVirtPkg/HighMemDxe: fix issues reported by Michael 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=19fedc88-4567-72c6-fca3-8c81e95c6d5e@redhat.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