It was found by an actual buffer overflow occurring.
-Shumin
-Original Message-
From: Carsey, Jaben
Sent: Friday, May 08, 2015 11:36 PM
To: Qiu, Shumin; edk2-devel@lists.sourceforge.net
Cc: Yao, Jiewen; Carsey, Jaben
Subject: RE: [edk2] [PATCH] ShellPkg: Fix buffer overflow issue in 'map
This patch fixes a crash in PEI on the AArch64 FVP platform when using SEC for
EL3. The current SEC code is not performing initialization of the TZC-400
TrustZone controller such that DRAM is not usable by PEI in EL2. This is a
minimal fix - long term a library abstraction for the TZC-400 woul
Hi, Jordan,
Thank you for your feedback.
Ideally I don't want CorebootPayloadPkg to depend on DuetPkg.But on the
other side I hesitate to duplicate drivers into the CorebootPayloadPkg from
other packages.
If I have to make a choice, I'd like the former.
I just tried to use the standard Pc
Can you use git send-email to send your patches? (Note the email
double-space strangeness below)
I don't think CorebootPayloadPkg should depend on DuetPkg.
Regarding the PciEnumeration issue, in OVMF we set
gEfiMdeModulePkgTokenSpaceGuid.PcdPciDisableBusEnumeration to TRUE at
runtime for Xen to d
On Fri, May 08, 2015 at 08:51:53PM +0200, Laszlo Ersek wrote:
> The PMBA_RTE and ACPI_TIMER_OFFSET macros apply equally to both boards,
> plus they are triplicated between the various AcpiTimerLib instances.
> Define them centrally in "OvmfPlatforms.h".
Patches 01..04 of the series
Acked-by: Gabri
Are you using 'git format-patch --subject-prefix=Patch'? If you leave
off --subject-prefix, then the default is PATCH. You should only need
subject prefix for special cases, like "RFC" or "PATCH v2".
Why not just migrate to the new BDS entirely? Or, why not make it the
default? What feature gaps a
Leandro Becker [mailto:lbec...@positivo.com.br] wrote:
]Sent: Friday, May 08, 2015 02:11 PM
]To: edk2-devel@lists.sourceforge.net
]Subject: [edk2] Fresh checkout from EDKII and VfrCompile failed when compiling
Nt32Pkg
]
]Hello,
]
]I made a checkout of EDKII and started compiling Nt32Pkg, but it f
Hello,
I made a checkout of EDKII and started compiling Nt32Pkg, but it failed
on Visual Studio 2012. After some minutes I got the following error:
"VfrCompile" -l -n --string-db
d:\positivo\edkii\Build\NT32IA32\DEBUG_VS
2012x86\IA32\MdeModulePkg\Universal\DriverSampleDxe\DriverSampleDxe
Move platform specific macros to their own include files. In addition,
define some new macros for register addresses (both PCI and IO) and
register values (bits).
Cc: Gabriel Somlo
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek
---
OvmfPkg/Include/IndustrySt
All POWER_MGMT_REGISTER_Q35() macro invocations in OvmfPkg should use the
macros in "Q35MchIch9.h" as arguments.
Cc: Gabriel Somlo
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek
---
OvmfPkg/Library/AcpiTimerLib/BaseAcpiTimerLib.c| 7 +++
OvmfPkg/Libr
All POWER_MGMT_REGISTER_PIIX4() macro invocations in OvmfPkg should use
the macros in "I440FxPiix4.h" as arguments.
Cc: Gabriel Somlo
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek
---
OvmfPkg/Library/AcpiTimerLib/BaseAcpiTimerLib.c| 7 +++
OvmfPkg/L
The
Quark_EDKII_v1.1.0/IA32FamilyCpuBasePkg/CpuArchDxe
driver provides the following capability in its implementation of
EFI_CPU_ARCH_PROTOCOL: whenever the SetMemoryAttributes() member is used
(directly, or indirectly via gDS->SetMemorySpaceAttributes()) to change
MTRR settings, the complete s
The PMBA_RTE and ACPI_TIMER_OFFSET macros apply equally to both boards,
plus they are triplicated between the various AcpiTimerLib instances.
Define them centrally in "OvmfPlatforms.h".
Cc: Gabriel Somlo
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek
---
Ovm
The logic visible in the previous patch should be done at startup as well,
not only when MTRR settings are changed. The inspiration is taken from
Quark_EDKII_v1.1.0/IA32FamilyCpuBasePkg/CpuMpDxe/
(see the EarlyMpInit() function and its call sites in
"ProcessorConfig.c").
Cc: Jeff Fan
Cc: Chen
EFI_MP_SERVICES_PROTOCOL can be provided even when the system has only one
logical processor (the BSP); the behavior of EFI_MP_SERVICES_PROTOCOL
member functions is well defined. Some drivers (will) depend on the
presence of EFI_MP_SERVICES_PROTOCOL even if there's just a BSP, so let's
install thi
The
Quark_EDKII_v1.1.0/IA32FamilyCpuBasePkg/CpuArchDxe
driver applies any MTRR changes to APs, if the EFI_MP_SERVICES_PROTOCOL is
available. We should do the same; port it to UefiCpuPkg.
The broadcast should occur even if PcdCpuSyncMtrrToAcpiNvs is set to
FALSE; in that case we just use a loca
When the ExecuteSmmCoreFromSmram() function fails, SmmIplEntry() restores
the SMRAM range to EFI_MEMORY_UC. However, it saves the return value of
gDS->SetMemorySpaceAttributes() in the same Status variable that contains
the return value of ExecuteSmmCoreFromSmram(). Therefore, if
gDS->SetMemorySpac
SVN r15305 (git 5a217a06), "OvmfPkg: S3 Suspend: save boot script after
ACPI context", made this driver install gEfiDxeSmmReadyToLockProtocolGuid
in SaveS3BootScript() -- for valid reasons --, however in the INF file the
protocol was marked as "ALWAYS_CONSUMED". Fix it.
Contributed-under: TianoCor
BaseExtractGuidedSectionLib uses a table at the static physical address
PcdGuidedExtractHandlerTableAddress, and modules that are linked against
BaseExtractGuidedSectionLib are expected to work together on that table.
Namely, some modules can register handlers for GUIDed sections, some other
module
My SMM-for-OVMF branch has 52 patches now, and I'm not seeing the light
at the end of the tunnel. It's probably prudent to flush some
preparatory patches now that should be fine / non-controversial in their
own right.
Public branch: https://github.com/lersek/edk2/commits/bits_and_pieces
Thanks
La
Reviewed-by: Jordan Justen
On 2015-05-05 07:21:34, Laszlo Ersek wrote:
> When the Q35 machine type(s) of QEMU are used with libvirt, libvirt tends
> to place some devices behind PCI bridges. This is then reflected in the
> "bootorder" fw_cfg file. For example:
>
> /pci@i0cf8/pci-bridge@1e/pci-
This updates the SecureBootConfigDxe component to allow the enrollment
of the digests of EFI executable that are built for ARM or AARCH64.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ard Biesheuvel
---
.../SecureBootConfigDxe/SecureBootConfigImpl.c|
Hello All,
I was trying SW Vlan implementation today. There is a problem i am facing.
There are some issues with respect to
protocol type field in the ethernet packet.
following are the obervations:
1. When the client receives a packet from the server, the Vlan id is
stripped by hardware.
Felix,
I like the more detailed comments.
Mike
-Original Message-
From: Felix Poludov [mailto:fel...@ami.com]
Sent: Friday, May 08, 2015 6:43 AM
To: Ni, Ruiyu; Kinney, Michael D; edk2-devel@lists.sourceforge.net; Andrew Fish
Subject: RE: [edk2] [Patch] MdePkg: Add EFI_LOAD_OPTION defini
Current CorebootPayloadPkg uses PciHostBridge and PciBusNoEnumeration
driver. It will cause the PCI bus resource incorrectly set in root
bridge instance. As a result all PCI devices behind a PCI bridge will
not show up in Shell 'PCI' command.
To resolve it use PciRootBridgeNoEnumeration driver ins
Reviewed-by: Jaben Carsey
Was this found by an actual buffer overflow occurring or by an automated test?
> -Original Message-
> From: Qiu, Shumin
> Sent: Friday, May 08, 2015 6:03 AM
> To: edk2-devel@lists.sourceforge.net; Carsey, Jaben
> Cc: Yao, Jiewen
> Subject: RE: [edk2] [PATCH] Sh
Ray, Mike,
I'd like to propose a more detailed comment block. How about this?
+//
+// EFI Load Option. This data structure describes format of UEFI boot option
variables.
+// NOTE: EFI Load Option is a byte packed buffer of variable length fields.
+// The first two fields have fixed length. The
Looks good!
Series Reviewed-by: Samer El-Haj-Mahmoud
-Original Message-
From: Ruiyu Ni [mailto:ruiyu...@intel.com]
Sent: Friday, May 08, 2015 2:23 AM
To: edk2-devel@lists.sourceforge.net
Subject: [edk2] [Patch V2 0/2] MdePkg: Add URI_DEVICE_PATH.
The patch adds URI device path node de
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Feng Tian
Reviewed-by: Star Zeng
---
MdePkg/Include/Protocol/DevicePath.h | 11
.../Library/UefiDevicePathLib/DevicePathFromText.c | 29
.../Library/UefiDevicePathLib/DevicePathToTe
IsochronousTransfer/AsyncIsochronousTransfer() interface
descriptions get updated to follow UEFI2.5 spec.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Feng Tian
---
MdePkg/Include/Protocol/Usb2HostController.h | 69 ++--
1 file changed, 66 insert
Hi Jiewen,
You are right. Patch updated.
-Shumin
-Original Message-
From: Yao, Jiewen [mailto:jiewen@intel.com]
Sent: Friday, May 08, 2015 4:44 PM
To: edk2-devel@lists.sourceforge.net; Carsey, Jaben
Subject: Re: [edk2] [PATCH] ShellPkg: Fix buffer overflow issue in 'map'
command.
H
On Fri, May 08, 2015 at 10:39:43AM +0100, Ian Campbell wrote:
> On Fri, 2015-05-08 at 07:31 +0100, Wei Liu wrote:
> > On Thu, May 07, 2015 at 10:29:55AM +0100, Ian Campbell wrote:
> > > On Thu, 2015-05-07 at 09:02 +0200, Laszlo Ersek wrote:
> > > > (Plus, you are cloning a git repo that may or may
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni
CC: Eric Dong
---
.../PlatformBootManagerLib/PlatformBootManager.c | 220 +
.../PlatformBootManagerLib/PlatformBootManager.h | 78
.../PlatformBootManagerLib.inf
On 05/08/15 04:26, Roy Franz wrote:
> On Thu, May 7, 2015 at 7:03 PM, Kinney, Michael D
> wrote:
>> Roy,
>>
>> Did you try the command line at the UEFI Shell prompt through a
>> serial terminal? There is line editing available, where backspace
>> and delete keys perform different actions. I thin
Amy:
I have two comments.
1. PosixLink for EDKII BaseTools, not for third party tool. I don't
think it is good to add Mtoc here.
2. Why remove nasm build rules from build_rule.template?
3. Who uses *_XCODE5_*_PLATFORM_FLAGS and *_XCODE5_X64_PLINK_FLAGS?
Thanks
Liming
From:
Hello
The 2nd parameter should be: The maximum number of Destination Unicode char,
including terminating null char.
Should we use "(StrSize(Specific) + sizeof(CHAR16))/sizeof(CHAR16)" for 2nd
parameter?
Thank you
Yao Jiewen
-Original Message-
From: Qiu Shumin [mailto:shumin@intel.
This patch replace 'StrnCat' with 'StrnCatS' to avoid the buffer overflow in
'map.c'.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Qiu Shumin
---
ShellPkg/Library/UefiShellLevel2CommandsLib/Map.c | 24 +++
1 file changed, 20 insertions(+), 4 deletio
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Star Zeng
Reviewed-by: Liming Gao
---
MdeModulePkg/Core/Dxe/Gcd/Gcd.c | 78
MdeModulePkg/Core/Dxe/Mem/Page.c | 6 +++-
2 files changed, 53 insertions(+), 31 deletions(-)
diff --git
The patches follow UEFI 2.5 and PI 1.4 Spec to add memory more reliable support.
Star Zeng (2):
MdePkg: Add memory more reliable definitions.
MdeModulePkg DxeCore: Add memory more reliable support.
MdeModulePkg/Core/Dxe/Gcd/Gcd.c | 78
MdeModulePkg/C
PiHob.h: EFI_RESOURCE_ATTRIBUTE_MORE_RELIABLE resource attribute.
PiDxeCis: EfiGcdMemoryTypeMoreReliable GCD memory type.
UefiSpec.h: EFI_MEMORY_MORE_RELIABLE memory attribute.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Star Zeng
Reviewed-by: Liming Gao
---
MdePkg/In
Hi, Chao
The patch looks good.
-Original Message-
From: edk2-devel@lists.sourceforge.net
[mailto:edk2-devel@lists.sourceforge.net] On Behalf Of Zhang
Sent: Friday, May 08, 2015 3:56 PM
To: Zhang; Zhang, Chao B
Cc: Fu, Siyuan
Subject: [Patch] SecurityPkg Variable : PK can be cleared in ru
This looks good.
Reviewed-by: Long, Qin
Best Regards & Thanks,
LONG, Qin
-Original Message-
From: Tian, Feng
Sent: Friday, May 08, 2015 3:37 PM
To: edk2-devel@lists.sourceforge.net
Cc: Long, Qin; Tian, Feng
Subject: [patch] MdePkg: Add missing GUID variables in SmartCardEdge.h
Declar
Reviewed-by: Star Zeng
-Original Message-
From: Tian, Feng
Sent: Friday, May 8, 2015 3:36 PM
To: edk2-devel@lists.sourceforge.net
Cc: Zeng, Star; Tian, Feng
Subject: [patch] MdePkg: Add missing GUID variables in BlockIoCrypto.h
Declares some extern EFI_GUID variables which correspond to
Declares some extern EFI_GUID variables which correspond to
guid macros in BlockIoCrypto.h for C code use.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Feng Tian
---
MdePkg/Include/Protocol/BlockIoCrypto.h | 4
MdePkg/MdePkg.dec | 4
2 fi
Declares some extern EFI_GUID variables which correspond to
guid macros in SmartCardEdge.h for C code use.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Feng Tian
---
MdePkg/Include/Protocol/SmartCardEdge.h | 13 +
MdePkg/MdePkg.dec | 7
Good to me.
Reviewed-by: Feng.Tian
-Original Message-
From: Ruiyu Ni [mailto:ruiyu...@intel.com]
Sent: Friday, May 08, 2015 15:23
To: edk2-devel@lists.sourceforge.net
Subject: [edk2] [Patch V2 0/2] MdePkg: Add URI_DEVICE_PATH.
The patch adds URI device path node definition and FromText
The patch adds URI device path node definition and FromText/ToText support.
V2 uses zero-sized array instead of 1-sized array for the definition of
URI_DEVICE_PATH.
Ruiyu Ni (2):
MdePkg: Add URI_DEVICE_PATH definition.
MdePkg: Add ToText/FromText support for URI device path node
MdePkg/Inclu
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni
CC: Feng Tian
---
MdePkg/Include/Protocol/DevicePath.h | 14 ++
1 file changed, 14 insertions(+)
diff --git a/MdePkg/Include/Protocol/DevicePath.h
b/MdePkg/Include/Protocol/DevicePath.h
index 4d2f340..d
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni
CC: Feng Tian
---
.../Library/UefiDevicePathLib/DevicePathFromText.c | 33 ++
.../Library/UefiDevicePathLib/DevicePathToText.c | 40 ++
2 files changed, 73 insertions(+)
diff --
Mike,
How about the below comments?
+//
+// EFI Load Option
+// NOTE: Some of the variable length structure fields are commented out on
purpose.
+//
+typedef struct _EFI_LOAD_OPTION {
Thanks,
Ray
-Original Message-
From: Kinney, Michael D
Sent: Friday, May 8, 2015 3:15 AM
To: Felix Po
50 matches
Mail list logo
