Mike,
I have filed https://bugzilla.tianocore.org/show_bug.cgi?id=1428 and
https://bugzilla.tianocore.org/show_bug.cgi?id=1429 to address this. I have
assigned these to me as I already have fix for these.
Also, how do we ensure all components which may have this issue (may not have
exposed it
On Fri, 2019-01-04 at 12:03 +0100, Ard Biesheuvel wrote:
> Fix a couple of occurrences of typo Standlone -> Standalone. Since
> _PiMmStandloneArmTfCpuDriverEntry() is never referenced, drop it
> altogether.
>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Ard Biesheuvel
On Fri, 2019-01-04 at 12:03 +0100, Ard Biesheuvel wrote:
> StandaloneMmCoreEntryPoint calls SerialPortInitialize() explicitly,
> so add SerialPortLib to its list of LibraryClasses.
>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Ard Biesheuvel
Reviewed-by: Supreeth Ve
On Fri, 2019-01-04 at 12:03 +0100, Ard Biesheuvel wrote:
> PE/COFF section names are ASCII strings so use %a not %s.
>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Ard Biesheuvel
Reviewed-by: Supreeth Venkatesh
> ---
> StandaloneMmPkg/Library/StandaloneMmCoreEntry
We introduced support for non-shareable cached mappings to the AArch64
version of ArmMmuLib a while ago, but the ARM version was left behind,
so fix it.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
ArmPkg/Library/ArmMmuLib/Arm/ArmMmuLibCore.c | 13 +++
PopulateLevel2PageTable () is invoked for [parts of] mappings that
start or end on a non-1 MB aligned address (or both). The size of
the mapping depends on both the start address modulo 1 MB and the
length of the mapping, but the logic that calculates this size is
flawed: subtracting 'start address
On Fri, 2019-01-04 at 12:03 +0100, Ard Biesheuvel wrote:
> ASSERT_EFI_ERROR (x) is a shorthand for ASSERT(!EFI_ERROR(x)), and so
> it should only be used with EFI_STATUS type expressions.
>
> So drop two instances that operate on other types, since neither
> looks
> particularly useful.
>
> Contr
During cleanup in case of initialization failure, some driver
bindings are not installed. Using abstractions in UEFILib takes
care of it.
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1428
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ashish Singhal
---
NetworkPkg
An issue was seen in IScsiDxe in NetworkPkg where driver cleanup after
initialization failure was not done right. Bug 1428 was filed in this regard.
As per discussions with Mike, it was also discussed that having UEFILib
provide protocol uninstallation abstraction would help to avoid these
issues i
Provided functions in UEFILib that abstract driver model protocol
uninstallation. This helps drivers to install and uninstall protocols
using a library to keep things seemless.
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1429
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-
Update interfaces as exposed by UEFILib for protocol
installation and uninstallation abstraction.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ashish Singhal
---
NetworkPkg/IScsiDxe/IScsiDriver.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/NetworkPkg/IScsi
Add a helper function to operate upon protocol installation and
uninstallation instead of every function doing it by itself.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ashish Singhal
---
MdePkg/Include/Library/UefiLib.h | 26 +-
MdePkg/Library/UefiLib/UefiDr
Mike,
I have addressed the issue along with some optimizations and have submitted
patches for review.
Thanks
Ashish
From: Ashish Singhal
Sent: Friday, January 4, 2019 10:33 AM
To: 'Kinney, Michael D' ; edk2-devel@lists.01.org
Cc: Gao, Liming ; Fu, Siyuan ; Wu,
Jiaxin
Subject: RE: Uninstalling
An issue was seen in IScsiDxe in NetworkPkg where driver cleanup after
initialization failure was not done right. Bug 1428 was filed in this regard.
As per discussions with Mike, it was also discussed that having UEFILib
provide protocol uninstallation abstraction would help to avoid these
issues i
Provided functions in UEFILib that abstract driver model protocol
uninstallation. This helps drivers to install and uninstall protocols
using a library to keep things seemless.
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1429
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-
Add a helper function to operate upon protocol installation and
uninstallation instead of every function doing it by itself.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ashish Singhal
---
MdePkg/Include/Library/UefiLib.h | 26 +-
MdePkg/Library/UefiLib/UefiDr
During cleanup in case of initialization failure, some driver
bindings are not installed. Using abstractions in UEFILib takes
care of it.
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1428
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ashish Singhal
---
NetworkPkg
Update interfaces as exposed by UEFILib for protocol
installation and uninstallation abstraction.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ashish Singhal
---
NetworkPkg/IScsiDxe/IScsiDriver.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/NetworkPkg/IScsi
On Thu, 3 Jan 2019 at 17:05, Carsey, Jaben wrote:
>
> Reviewed-by: Jaben Carsey
>
>
Thanks all
Pushed as 672601cfcc6f..8ef653aa5aad
> > -Original Message-
> > From: edk2-devel [mailto:edk2-devel-boun...@lists.01.org] On Behalf Of
> > Ard Biesheuvel
> > Sent: Thursday, January 03, 2019
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1426
In function PeCoffLoaderRelocateImageForRuntime, it doesn't
do much check when applies relocation fixups. For API level
consideration, it's not safe enough.
This patch is to replace the same code logic with calling
function PeCoffLoaderImage
I think any solution like that should be publicly available for an
unlimited number of users. As it stands now, mailing lists and IRC are
open to all - if software like Microsoft Teams is to be used, it's
pretty likely most communication would quickly become invitation-only
to avoid paying $8 per u
(add the CryptoPkg maintainer)
On Wed, 2 Jan 2019 at 14:14, Jagadeesh Ujja wrote:
>
> “SmmCryptLib” library can be used by MM_STANDALONE drivers as well.
> So add MM_STANDALONE as the module type this library supports.
>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: J
On Wed, 2 Jan 2019 at 14:14, Jagadeesh Ujja wrote:
>
> “VarCheckLib” library can be used by MM_STANDALONE drivers as well.
> So add MM_STANDALONE as the module type this library supports.
>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Jagadeesh Ujja
Reviewed-by: Ard
On Thu, 3 Jan 2019 at 07:15, Jagadeesh Ujja wrote:
>
> On Thu, Jan 3, 2019 at 6:45 AM Zhang, Chao B wrote:
> >
> > Reviewed-by : Chao Zhang
>
> Hi Chao Zhang,
>
> Thanks for the review, I will not be having any new changes with this
> specific patch.
> Can you please merge this patch, so that I
ASSERT_EFI_ERROR (x) is a shorthand for ASSERT(!EFI_ERROR(x)), and so
it should only be used with EFI_STATUS type expressions.
So drop two instances that operate on other types, since neither looks
particularly useful.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Bie
This series addresses a number of issues I ran into while bringing up
the standalone MM based authenticated variable store on the SynQuacer
(AArch64) platform.
The first six patches are obvious fixes/improvements.
Patch #7 adds support for TE formatted MM_CORE_STANDALONE binaries.
This is useful
TE images take up less space when using 4 KB section alignment, since
the FFS/FV generation code optimizes away the redundant, nested padding.
This saves 4 KB of space, which is a worthwhile improvement for code
that executes in place in secure context.
Contributed-under: TianoCore Contribution Ag
Standalone MM requires 4 KB section alignment for all images, so that
strict permissions can be applied. Unfortunately, this results in a
lot of wasted space, which is usually costly in the secure world
environment that standalone MM is expected to operate in.
So let's permit the standalone MM dri
The special handling of the EFI_IMAGE_SUBSYSTEM_EFI_RUNTIME_DRIVER
attribute is only necessary for images that are relocated twice, i.e.,
in the context of SetVirtualAddressMap (). This does not apply to
standalone MM modules, so drop the check.
Drop some redundant DEBUG output while at it.
Contr
StandaloneMmCoreEntryPoint calls SerialPortInitialize() explicitly,
so add SerialPortLib to its list of LibraryClasses.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
StandaloneMmPkg/Library/StandaloneMmCoreEntryPoint/StandaloneMmCoreEntryPoint.inf
|
The dispatcher uses the PE/COFF loader to load images into the heap,
but only does so after copying the entire image first, leading to
two copies being made for no good reason.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
StandaloneMmPkg/Core/Dispatch
Fix a couple of occurrences of typo Standlone -> Standalone. Since
_PiMmStandloneArmTfCpuDriverEntry() is never referenced, drop it
altogether.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
StandaloneMmPkg/Drivers/StandaloneMmCpu/AArch64/EventHandle.c
PE/COFF section names are ASCII strings so use %a not %s.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
StandaloneMmPkg/Library/StandaloneMmCoreEntryPoint/AArch64/SetPermissions.c |
2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
a/St
On Thu, 3 Jan 2019 at 17:14, Laszlo Ersek wrote:
>
> On 01/03/19 12:03, Ard Biesheuvel wrote:
> > On Wed, 2 Jan 2019 at 14:14, Jagadeesh Ujja wrote:
> >>
> >> Some of the existing DXE drivers can be refactored to execute within
> >> the Standalone MM execution environment as well. Allow such driv
G'Day!
I'm trying to get networking under edk2 up and running. I tried
AppPkg/Applications/Sockets/RawIp4Tx
under OVMF. The raw packet is sent out on the network, but the
application never returns from the
socket close routine.
I'm currently using UDK2017 with the latest security patches (dow
Implement a variant of the FIP006 NOR flash driver that can execute
in standalone MM context. This is the foundation for hosting the
EFI authenticated variable store in the secure world.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
Silicon/Socionext/S
Wire up the various pieces so that the authenticated variable store
runs entirely in standalone MM context residing in a secure partition.
This primarily involves refactoring the platform's NOR flash driver so
we can build a version that can work in the standalone MM context.
Beyond that, it is ju
The FIP006 NOR flash driver contains implementations of the block I/O
and disk I/O protocols, but never exposes them to other drivers (i.e.,
it never installs the protocol interfaces). So let's drop this code
altogether: the NOR flash is for code and variables, not for arbitrary
files.
Contributed
This code may execute in SMM context, where unaligned accesses are
not permitted. So use ReadUnaligned32() instead of performing a
direct UINT32* cast.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
Silicon/Socionext/SynQuacer/Drivers/Fip006Dxe/NorFlash
In preparation of creating a SMM version of the FIP006 NOR flash
driver, refactor the existing pieces into a core driver, the FVB
methods and the DXE instantiation code.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
Silicon/Socionext/SynQuacer/Drivers/
This implements support for UEFI secure boot on DeveloperBox using
the standalone MM framework. This moves all of the software handling
of the UEFI authenticated variable store into the standalone MM
context residing in a secure partition.
Note that SynQuacer as configured today is not a truly sec
Create a pair of .DSC/.FDF files that describe the components and
the firmware volumes and flash device that will be dispatched into
a secure partition in the secure world to control the UEFI secure
variable store.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuv
We are going to add a separate .DSC/.FDF combo for the standalone
MM components. So put all the pieces we will share in an include
file that both .DSC files can include.
Contributed-under: TianoCore Contribution Agreement 1.1
Signed-off-by: Ard Biesheuvel
---
Platform/Socionext/DeveloperBox/Deve
Ashish,
Thanks for the pointer. I agree there is an issue here.
Please enter a Bugzilla against the IScsiDxe module for this issue so we can
fix this failure.
You are also welcome to enter a Bugzilla for a feature request to add UefiLib
APIs that can be used to safely uninstall all the driver
44 matches
Mail list logo