mirror_edk2.git
5 years agoNetworkPkg: Fix IpsecConfig GCC build failure issue
Jiaxin Wu [Tue, 8 Mar 2016 08:08:40 +0000 (16:08 +0800)]
NetworkPkg: Fix IpsecConfig GCC build failure issue

This issue is caused by the string token ID for help message,
which is defined in the internal head file.
This head file is used for reference more than once. So,
multiple definition for the string token ID error will be
enrolled.

Cc: Fu Siyuan <siyuan.fu@intel.com>
Cc: Zhang Lubo <lubo.zhang@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Zhang, Lubo <lubo.zhang@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: permit access to the full extended config space
Laszlo Ersek [Fri, 4 Mar 2016 00:49:54 +0000 (01:49 +0100)]
OvmfPkg: PciHostBridgeLib: permit access to the full extended config space

By now OVMF makes MdeModulePkg/Bus/Pci/PciHostBridgeDxe go through
MMCONFIG (when running on Q35). Enable the driver to address each B/D/F's
config space up to and including offset 0xFFF.

Cc: Gabriel Somlo <somlo@cmu.edu>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Michał Zegan <webczat_200@poczta.onet.pl>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Gabriel Somlo <somlo@cmu.edu>
Tested-by: Michał Zegan <webczat_200@poczta.onet.pl>
5 years agoOvmfPkg: match PCI config access to machine type (if not USE_OLD_PCI_HOST)
Laszlo Ersek [Fri, 4 Mar 2016 00:33:38 +0000 (01:33 +0100)]
OvmfPkg: match PCI config access to machine type (if not USE_OLD_PCI_HOST)

If USE_OLD_PCI_HOST is FALSE, then we switch all executable module types
supported by DxePciLibI440FxQ35 to the following library instance stack:

  BasePciSegmentLibPci  [class: PciSegmentLib]
    DxePciLibI440FxQ35  [class: PciLib]
      BasePciCf8Lib     [class: PciCf8Lib]
      BasePciExpressLib [class: PciExpressLib]

Every module will select 0xCF8 vs. ECAM based on the OVMF platform type
(i440fx or Q35). Notably, MdeModulePkg/Bus/Pci/PciHostBridgeDxe is among
the affected drivers.

The BasePciExpressLib instance is where the PcdPciExpressBaseAddress PCD
fills its original role.

Cc: Gabriel Somlo <somlo@cmu.edu>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Michał Zegan <webczat_200@poczta.onet.pl>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Gabriel Somlo <somlo@cmu.edu>
Tested-by: Michał Zegan <webczat_200@poczta.onet.pl>
5 years agoOvmfPkg: add DxePciLibI440FxQ35
Laszlo Ersek [Thu, 3 Mar 2016 23:53:12 +0000 (00:53 +0100)]
OvmfPkg: add DxePciLibI440FxQ35

This library is a trivial unification of the following two PciLib
instances (and the result is easily diffable against each):
- MdePkg/Library/BasePciLibCf8
- MdePkg/Library/BasePciLibPciExpress

The PCI config access method is determined in the constructor function,
from the dynamic PCD "PcdOvmfHostBridgePciDevId" that is set by
PlatformPei.

The library instance is usable in DXE phase or later modules: the PciLib
instances being unified have no firmware phase / client module type
restrictions, and here the only PCD access is made in the constructor
function. That is, even before a given client executable's entry point is
invoked.

The library instance depends on PlatformPei both for setting the PCD
mentioned above, and also for enabling MMCONFIG on Q35. PEI and earlier
phase modules are not expected to need extended config access even on Q35.

Cc: Gabriel Somlo <somlo@cmu.edu>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Michał Zegan <webczat_200@poczta.onet.pl>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Gabriel Somlo <somlo@cmu.edu>
Tested-by: Michał Zegan <webczat_200@poczta.onet.pl>
5 years agoOvmfPkg: PlatformPei: enable PCIEXBAR (aka MMCONFIG / ECAM) on Q35
Laszlo Ersek [Thu, 3 Mar 2016 19:46:22 +0000 (20:46 +0100)]
OvmfPkg: PlatformPei: enable PCIEXBAR (aka MMCONFIG / ECAM) on Q35

The comments in the code should speak for themselves; here we note only
two facts:

- The PCI config space writes (to the PCIEXBAR register) are performed
  using the 0xCF8 / 0xCFC IO ports, by virtue of PciLib being resolved to
  BasePciLibCf8. (This library resolution will permanently remain in place
  for the PEI phase.)

- Since PCIEXBAR counts as a chipset register, it is the responsibility of
  the firmware to reprogram it at S3 resume. Therefore
  PciExBarInitialization() is called regardless of the boot path. (Marcel
  recently posted patches for SeaBIOS that implement this.)

This patch suffices to enable PCIEXBAR (and the dependent ACPI table
generation in QEMU), for the sake of "PCIeHotplug" in the Linux guest:

  ACPI: MCFG 0x000000007E17F000 00003C
        (v01 BOCHS  BXPCMCFG 00000001 BXPC 00000001)
  PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem 0x80000000-0x8fffffff]
       (base 0x80000000)
  PCI: MMCONFIG at [mem 0x80000000-0x8fffffff] reserved in E820
  acpi PNP0A08:00: _OSC: OS supports
                   [ExtendedConfig ASPM ClockPM Segments MSI]
  acpi PNP0A08:00: _OSC: OS now controls
                   [PCIeHotplug PME AER PCIeCapability]

In the following patches, we'll equip the core PCI host bridge / root
bridge driver and the rest of DXE as well to utilize ECAM on Q35.

Cc: Gabriel Somlo <somlo@cmu.edu>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Michał Zegan <webczat_200@poczta.onet.pl>
Ref: https://github.com/tianocore/edk2/issues/32
Ref: http://thread.gmane.org/gmane.comp.bios.coreboot.seabios/10548
Suggested-by: Marcel Apfelbaum <marcel@redhat.com>
Reported-by: Michał Zegan <webczat_200@poczta.onet.pl>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Tested-by: Gabriel Somlo <somlo@cmu.edu>
Tested-by: Michał Zegan <webczat_200@poczta.onet.pl>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PlatformPei: lower the 32-bit PCI MMIO base to 2GB on Q35
Laszlo Ersek [Tue, 8 Mar 2016 12:01:03 +0000 (13:01 +0100)]
OvmfPkg: PlatformPei: lower the 32-bit PCI MMIO base to 2GB on Q35

Gerd has advised us that long term support Q35 machine types have no low
RAM above 2GB, hence we should utilize the [2GB, 3GB) gap -- that we
currently leave unused -- for MMIO. (Plus, later in this series, for the
PCIEXBAR too.)

Cc: Gabriel Somlo <somlo@cmu.edu>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Michał Zegan <webczat_200@poczta.onet.pl>
Ref: https://github.com/tianocore/edk2/issues/32
Ref: http://thread.gmane.org/gmane.comp.bios.edk2.devel/8707/focus=8817
Suggested-by: Gerd Hoffmann <kraxel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Gabriel Somlo <somlo@cmu.edu>
Tested-by: Michał Zegan <webczat_200@poczta.onet.pl>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: IndustryStandard/Q35MchIch9.h: add PCIEXBAR macros
Laszlo Ersek [Thu, 3 Mar 2016 18:04:28 +0000 (19:04 +0100)]
OvmfPkg: IndustryStandard/Q35MchIch9.h: add PCIEXBAR macros

Section 5.1.16 ("PCIEXBAR -- PCI Express Register Range Base Address") in
Intel document #316966-002 (already referenced near the top of this header
file) describes the Q35 DRAM Controller register that configures the
memory-mapped PCI config space (also known as MMCONFIG, and ECAM /
Enhanced Configuration Access Method).

In this patch we add the macros we'll need later. We'll only support the
256 MB memory-mapped config space -- enough for buses [0, 255].

Cc: Gabriel Somlo <somlo@cmu.edu>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Michał Zegan <webczat_200@poczta.onet.pl>
Ref: https://github.com/tianocore/edk2/issues/32
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Gabriel Somlo <somlo@cmu.edu>
Tested-by: Michał Zegan <webczat_200@poczta.onet.pl>
5 years agoBaseTools ConvertMasmToNasm: Support Python 3
Jordan Justen [Mon, 7 Mar 2016 07:38:59 +0000 (23:38 -0800)]
BaseTools ConvertMasmToNasm: Support Python 3

The script is updated to support both python 2.7 and python 3.

v2:
 * Use io.open() rather than open() (Jaben)

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: Yonghong Zhu <yonghong.zhu@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Erik Bjorge <erik.c.bjorge@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Erik Bjorge <erik.c.bjorge@intel.com>
5 years agoBaseTools ConvertMasmToNasm: Fix exception when no arguments are given
Jordan Justen [Thu, 3 Mar 2016 08:44:38 +0000 (00:44 -0800)]
BaseTools ConvertMasmToNasm: Fix exception when no arguments are given

Convert to use the argparse library rather than optparse.

As part of the conversion, the script will now give an error message
if no arguments are given. Previously the script would give an
exception when no arguments were given.

Fixes: https://github.com/tianocore/edk2/issues/65
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: Yonghong Zhu <yonghong.zhu@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
5 years agoBaseTools ConvertMasmToNasm: Fix running script outside of a git tree
Jordan Justen [Wed, 10 Feb 2016 18:08:37 +0000 (10:08 -0800)]
BaseTools ConvertMasmToNasm: Fix running script outside of a git tree

The script previously would hit an exception if it was run outside of
a git tree.

The exception looked like:

edk2/BaseTools/Scripts/ConvertMasmToNasm.py Version 0.01
Traceback (most recent call last):
  File "edk2/BaseTools/Scripts/ConvertMasmToNasm.py", line 986, in <module>
    ConvertAsmApp()
  File "edk2/BaseTools/Scripts/ConvertMasmToNasm.py", line 984, in __init__
    ConvertAsmFile(src, dst, self)
  File "edk2/BaseTools/Scripts/ConvertMasmToNasm.py", line 209, in __init__
    CommonUtils.__init__(self, clone)
  File "edk2/BaseTools/Scripts/ConvertMasmToNasm.py", line 69, in __init__
    self.gitemail = clone.gitemail
AttributeError: ConvertAsmApp instance has no attribute 'gitemail'

Fixes: https://github.com/tianocore/edk2/issues/63
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: Yonghong Zhu <yonghong.zhu@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
5 years agoBaseTools: Update ARM/AArch64 GenFv vector processing for encapsulated FVs
Cohen, Eugene [Tue, 1 Mar 2016 23:39:36 +0000 (23:39 +0000)]
BaseTools: Update ARM/AArch64 GenFv vector processing for encapsulated FVs

Instead of only handling SEC Core or PEI Core instances in the outer FV,
the GenFv tool will now recurse into FV image FFS files to look for instances
in encapsulated FVs so the vector area can be updated appropriately.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Eugene Cohen <eugene@hp.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
5 years agoBaseTools: report warning if VOID* PCD with {} value is not 8-byte aligned
Yonghong Zhu [Wed, 9 Mar 2016 06:03:39 +0000 (14:03 +0800)]
BaseTools: report warning if VOID* PCD with {} value is not 8-byte aligned

For VOID* Pcd with {} value, If platform developer wants to put in a
specific hex offset value that is not 8-byte aligned for VOID * then we
allow it with a warning message.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Yonghong Zhu <yonghong.zhu@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
5 years agoBaseTools: Change source files to DOS format.
Liming Gao [Wed, 9 Mar 2016 01:42:06 +0000 (09:42 +0800)]
BaseTools: Change source files to DOS format.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Yonghong Zhu <yonghong.zhu@intel.com>
5 years agoNetworkPkg: Use the New Functions from HttpLib
Ghazi Belaam [Fri, 4 Mar 2016 22:07:50 +0000 (06:07 +0800)]
NetworkPkg: Use the New Functions from HttpLib

After submitting changes for HttpLib, other modules should be able to use
those functions
1 remove the private function and their calls
2 update it with the functions from httpLib

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ghazi Belaam <Ghazi.belaam@hpe.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Samer EL-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Jiaxin Wu <jiaxin.wu@intel.com>
5 years agoModulePkg/DxeHttpLib: Adding Functions to HttpLib
Ghazi Belaam [Fri, 4 Mar 2016 22:07:49 +0000 (06:07 +0800)]
ModulePkg/DxeHttpLib: Adding Functions to HttpLib

There some usefull functions in edk2 private modules that could be used,
so we added them to the httpLib

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ghazi Belaam <Ghazi.belaam@hpe.com>
Reviewed-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Samer EL-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
5 years agoNetworkPkg: Convert UNI file to UTF8 format.
Fu Siyuan [Thu, 10 Mar 2016 01:14:34 +0000 (09:14 +0800)]
NetworkPkg: Convert UNI file to UTF8 format.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Jiaxin Wu <jiaxin.wu@intel.com>
5 years agoUefiCpuPkg/Application/Cpuid: Remove unnecessary code check
Jeff Fan [Thu, 10 Mar 2016 01:33:17 +0000 (09:33 +0800)]
UefiCpuPkg/Application/Cpuid: Remove unnecessary code check

gMaximumBasicFunction is set to CPUID_SIGNATURE as below, so removed the compare code.
UINT32  gMaximumBasicFunction = CPUID_SIGNATURE;

Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
UefiCpuPkg/Application/Cpuid: Remove unnecessary code check

gMaximumBasicFunction is set to CPUID_SIGNATURE as below, so removed the compare code.
UINT32  gMaximumBasicFunction = CPUID_SIGNATURE;

Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
UefiCpuPkg/Application/Cpuid: Remove unnecessary code check

gMaximumBasicFunction is set to CPUID_SIGNATURE as below, so removed the compare code.
UINT32  gMaximumBasicFunction = CPUID_SIGNATURE;

Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
Reviewed-by: Qiu Shumin <shumin.qiu@intel.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
5 years agoUefiCpuPkg/Application/Cpuid: Add check for gMaximumBasicFunction
Jeff Fan [Wed, 9 Mar 2016 02:05:30 +0000 (10:05 +0800)]
UefiCpuPkg/Application/Cpuid: Add check for gMaximumBasicFunction

Add check for gMaximumBasicFunction in CpuidVersionInfo () back.

Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
Reviewed-by: Qiu Shumin <shumin.qiu@intel.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
5 years agoShellPkg: Use DOS format end of line.
Qiu Shumin [Tue, 8 Mar 2016 12:51:43 +0000 (20:51 +0800)]
ShellPkg: Use DOS format end of line.

Use DOS format end of line(CR, LF).

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Qiu Shumin <shumin.qiu@intel.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
5 years agoBaseTools: Add Multiple Workspaces support for custom Makefiles.
Marvin Haeuser [Sun, 21 Feb 2016 00:46:58 +0000 (08:46 +0800)]
BaseTools: Add Multiple Workspaces support for custom Makefiles.

This patch makes sure the MODULE_DIR variable points to the correct
location when multiple workspaces are used. Currently, it is
always prefixed with $(WORKSPACE), which only works as long as the
package is in the Workspace.
Code modules were not effected because the required paths were valid,
but for custom Makefiles, the MODULE_DIR variable is used.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Marvin Haeuser <Marvin.Haeuser@outlook.com>
Reviewed-by: Yonghong Zhu <yonghong.zhu@intel.com>
5 years agoMdeModulePkg: Increase the maximum number of PEI performance log entries
Cinnamon Shia [Tue, 8 Mar 2016 06:00:04 +0000 (14:00 +0800)]
MdeModulePkg: Increase the maximum number of PEI performance log entries

The maximum number of PEI performance log entries is 255.
Add a new PCD, PcdMaxPeiPerformanceLogEntries16, to increase the maximum
number of PEI performance log entries.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Cinnamon Shia <cinnamon.shia@hpe.com>
Reviewed-by: Samer EL-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Joseph Shifflett <joseph.shifflett@hpe.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
Reviewed-by: Star Zeng <star.zeng@intel.com>
5 years agoOvmfPkg: Enable Network2 Shell Commands for IPv6
Gary Lin [Tue, 8 Mar 2016 09:50:44 +0000 (17:50 +0800)]
OvmfPkg: Enable Network2 Shell Commands for IPv6

Enable the network2 commands when NETWORK_IP6_ENABLE is TRUE, so we
would have Ping6 and Ifconfig6.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Gary Ching-Pang Lin <glin@suse.com>
[lersek@redhat.com: added the word "Shell" to the subject]
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
5 years agoShellPkg/UefiDpLib: Support execution break
Cinnamon Shia [Mon, 7 Mar 2016 03:23:08 +0000 (11:23 +0800)]
ShellPkg/UefiDpLib: Support execution break

Support UEFI shell execution break.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Cinnamon Shia <cinnamon.shia@hpe.com>
Reviewed-by: Star Zeng <star.zeng@intel.com>
Reviewed-by: Samer EL-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
5 years agoPerformancePkg/Dp_App: Support execution break
Cinnamon Shia [Mon, 7 Mar 2016 03:23:07 +0000 (11:23 +0800)]
PerformancePkg/Dp_App: Support execution break

Support UEFI shell execution break.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Cinnamon Shia <cinnamon.shia@hpe.com>
Reviewed-by: Star Zeng <star.zeng@intel.com>
Reviewed-by: Samer EL-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
5 years agoMdeModulePkg AcpiTableDxe: Use Rsdt to check against NULL
Star Zeng [Mon, 7 Mar 2016 06:01:46 +0000 (14:01 +0800)]
MdeModulePkg AcpiTableDxe: Use Rsdt to check against NULL

Some static scan tool may regard CurrentRsdtEntry to be potentially
referenced to NULL pointer if CurrentRsdtEntry == NULL is used in
the right above if condition judgment.

        CopyMem (CurrentRsdtEntry, CurrentRsdtEntry + 1, (*NumberOfTableEntries - Index) * sizeof (UINT32));

It is introduced by commit f9bbb8d9c3f065faba9f266cf4e731fe2ca70c4d.
To avoid it and have same style with
"((Xsdt == NULL) || CurrentTablePointer64 == (UINT64) (UINTN) Table->Table)",
use Rsdt instead of CurrentRsdtEntry to check against NULL.

Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Shumin Qiu <shumin.qiu@intel.com>
Cc: Feng Tian <feng.tian@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Star Zeng <star.zeng@intel.com>
Reviewed-by: Shumin Qiu <shumin.qiu@intel.com>
5 years agoOvmfPkg PciHostBridgeDxe: Convert X64/IoFifo.asm to NASM
Jordan Justen [Tue, 8 Mar 2016 03:18:04 +0000 (19:18 -0800)]
OvmfPkg PciHostBridgeDxe: Convert X64/IoFifo.asm to NASM

The BaseTools/Scripts/ConvertMasmToNasm.py script was used to convert
X64/IoFifo.asm to X64/IoFifo.nasm

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
5 years agoOvmfPkg PciHostBridgeDxe: Convert Ia32/IoFifo.asm to NASM
Jordan Justen [Tue, 8 Mar 2016 03:18:03 +0000 (19:18 -0800)]
OvmfPkg PciHostBridgeDxe: Convert Ia32/IoFifo.asm to NASM

The BaseTools/Scripts/ConvertMasmToNasm.py script was used to convert
Ia32/IoFifo.asm to Ia32/IoFifo.nasm

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
5 years agoShellPkg: Merge Ping6 and Ifconfig6 tools to Shell command.
Zhang, Lubo [Wed, 2 Mar 2016 02:13:54 +0000 (10:13 +0800)]
ShellPkg: Merge Ping6 and Ifconfig6 tools to Shell command.

According to the new Shell spec, we add Network2 profile and
merge Ping6 and Ifconfig6 tools to Shell command.

Cc: Carsey Jaben <jaben.carsey@intel.com>
Cc: Wu Jiaxin <jiaxin.wu@intel.com>
Cc: Ye Ting <ting.ye@intel.com>
Cc: Fu Siyuan <siyuan.fu@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Zhang Lubo <lubo.zhang@intel.com>
Reviewed-By: Jaben Carsey <Jaben.carsey@intel.com>
Reviewed-by: Jiaxin Wu <jiaxin.wu@intel.com>
5 years agoNetworkPkg: HttpBootDxe: fix gcc build failure.
Laszlo Ersek [Tue, 8 Mar 2016 03:12:57 +0000 (11:12 +0800)]
NetworkPkg: HttpBootDxe: fix gcc build failure.

Commit fa848a404894 ("NetworkPkg: Add URI configuration form to HTTP boot
driver") causes the build to fail with gcc:

> NetworkPkg/HttpBootDxe/HttpBootConfig.c: In function
>                                          'HttpBootAddBootOption':
> NetworkPkg/HttpBootDxe/HttpBootConfig.c:148:14:
> error: passing argument 3 of 'GetVariable2' from incompatible pointer
>        type [-Werror]
>               );
>               ^
> In file included from NetworkPkg/HttpBootDxe/HttpBootDxe.h:31:0,
>                  from NetworkPkg/HttpBootDxe/HttpBootConfig.c:15:
> MdePkg/Include/Library/UefiLib.h:708:1: note: expected 'void **' but
>                                         argument is of type 'CHAR16 **'
>  GetVariable2 (
>  ^
> cc1: all warnings being treated as errors

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
5 years agoFW: [PATCH 3/9] CorebootPayloadPkg/PlatformBdsLib: Fix spelling error
Leahy, Leroy P [Fri, 4 Mar 2016 17:52:14 +0000 (09:52 -0800)]
FW: [PATCH 3/9] CorebootPayloadPkg/PlatformBdsLib: Fix spelling error

-----Original Message-----
From: Leahy, Leroy P
Sent: Friday, March 4, 2016 8:58 AM
To: Ni, Ruiyu <ruiyu.ni@intel.com>; edk2-devel@lists.01.org; Bjorge, Erik C <erik.c.bjorge@intel.com>
Cc: Leahy, Leroy P <leroy.p.leahy@intel.com>
Subject: [PATCH 3/9] CorebootPayloadPkg/PlatformBdsLib: Fix spelling error

Change vender to vendor

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-by: Prince Agyeman <prince.agyeman@intel.com>
5 years agoMdeModulePkg/Bds: More user-friendly network boot option description
Ruiyu Ni [Fri, 4 Mar 2016 10:12:38 +0000 (18:12 +0800)]
MdeModulePkg/Bds: More user-friendly network boot option description

The patch enhances the UefiBootManagerLib to use more user-friendly
network boot option description.
It builds description like below:
 "PXEv6 (MAC:112233445566 VLAN1)"
 "HTTPv4 (MAC:112233445566)"

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Siyuan Fu <siyuan.fu@intel.com>
5 years agoMdeModulePkg/UefiBootManagerLib: Separate boot description functions.
Ruiyu Ni [Fri, 4 Mar 2016 08:34:43 +0000 (16:34 +0800)]
MdeModulePkg/UefiBootManagerLib: Separate boot description functions.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Siyuan Fu <siyuan.fu@intel.com>
5 years agoMdeModulePkg/Bds: Fix VS2010/VS2012 build failure.
Ruiyu Ni [Mon, 7 Mar 2016 05:29:44 +0000 (13:29 +0800)]
MdeModulePkg/Bds: Fix VS2010/VS2012 build failure.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Shumin Qiu <shumin.qiu@intel.com>
5 years agoMdePkg: Fix ACPI NFIT GUID definitions
Samer El-Haj-Mahmoud [Fri, 4 Mar 2016 02:56:12 +0000 (10:56 +0800)]
MdePkg: Fix ACPI NFIT GUID definitions

Fix ACPI 6.0 and 6.1 NFIT GUID definitions to avoid GCC compiler error
for missing braces.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
5 years agoMdePkg: Fix CPER GUID definitions to match UEFI spec
Samer El-Haj-Mahmoud [Fri, 4 Mar 2016 01:15:01 +0000 (09:15 +0800)]
MdePkg: Fix CPER GUID definitions to match UEFI spec

Add gEfiIa32X64ProcessorErrorSectionGuid and
gEfiArmProcessorErrorSectionGuid to
match the definition in the UEFI 2.6 specification Table 249.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
5 years agoNetworkPkg: Support print help information using -? command.
Zhang, Lubo [Mon, 29 Feb 2016 06:25:50 +0000 (14:25 +0800)]
NetworkPkg: Support print help information using -? command.

v2:
*Modify the logic of show SAD,SPD and PAD help info, include them in -?
instead of follow -p command.

Since Shell supports finding help information from resource section
of application image. We modify the Shell application Under NetworkPkg
to support print help information string using -? command.

Cc: Fu Siyuan <siyuan.fu@intel.com>
Cc: Ye Ting <ting.ye@intel.com>
Cc: Wu Jiaxin <jiaxin.wu@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Zhang Lubo <lubo.zhang@intel.com>
Reviewed-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
5 years agoNetworkPkg: Add URI configuration form to HTTP boot driver.
Fu Siyuan [Mon, 15 Feb 2016 05:55:11 +0000 (13:55 +0800)]
NetworkPkg: Add URI configuration form to HTTP boot driver.

This patch updates the HTTP boot driver to produce a setup page for the boot
file URI configuration. A new boot option will be created for the manual
configured URI address. This change is made to support the HTTP boot usage
in home environment.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Wu Jiaxin <jiaxin.wu@intel.com>
Reviewed-by: Ye Ting <ting.ye@intel.com>
5 years agoCryptoPkg/OpensslLib: Fix CRLF breakage in process_files.sh
David Woodhouse [Sat, 5 Mar 2016 16:44:33 +0000 (16:44 +0000)]
CryptoPkg/OpensslLib: Fix CRLF breakage in process_files.sh

This got broken in committing, due to a catalogue of broken practices.

Firstly, we should *pull* git submissions, never recommit them. You
preserve the correct history then, and don't risk rebasing to result in
a history which *never* worked in the form that gets preserved.

That would have kept the authorship attrbution correct too.

Secondly, we shouldn't be storing CRLF line endings in the objects that
git stores in its database. It is designed to store simple LF line endings,
and then check that out as appropriate for the system (resulting in CRLF
in the working tree for Windows users, as they expect). That would avoid
this problem, and all the other problems we have with patches being
exchanged.

Make it executable too, which also got lost in the commit mess.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
5 years agoCryptoPkg/OpensslLib: Automatically configure OpenSSL and generate file list
Qin Long [Sat, 5 Mar 2016 15:45:59 +0000 (23:45 +0800)]
CryptoPkg/OpensslLib: Automatically configure OpenSSL and generate file list

OpenSSL 1.1 (as well as our backport to 1.0.2) now allows us to run its
standard Configure script and import the result into the EDK II source
repository for others to build natively. The opensslconf.h file and the
list of files in OpensslLib.inf don't need to be managed manually.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Fix OpenSSL link failures on Windows (RT#4310)
Qin Long [Sat, 5 Mar 2016 15:44:40 +0000 (23:44 +0800)]
CryptoPkg/OpensslLib: Fix OpenSSL link failures on Windows (RT#4310)

This is pull request #755 for OpenSSL 1.1, along with a little extra fix
in the RSA_NET code which has been removed from 1.1 so we can't fix it
there.

https://github.com/openssl/openssl/pull/755

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3969
Qin Long [Sat, 5 Mar 2016 15:43:21 +0000 (23:43 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3969

Support for the UEFI target has been added to OpenSSL in commit 4d60c7e10.

Drop our partial implementation and use a backported version of what's
upstream. This includes a couple of fixes which will be needed when we
automatically generate the file list and opensslconf.h instead of
manually maintaining those.

This includes the subsequent fix in commit fb4844bbc.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3992
Qin Long [Sat, 5 Mar 2016 15:42:24 +0000 (23:42 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3992

Instead of commenting out the Signed Certificate Timestamps purely based
on the OPENSSL_SYS_UEFI flag, OpenSSL 1.1 supports a no-sct configuration
option, added in commit 05d7bf6c5. Drop our own hack and use that.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3951
Qin Long [Sat, 5 Mar 2016 15:41:31 +0000 (23:41 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3951

A more complete implementation of the X509_V_FLAG_NO_CHECK_TIME flag was
added to OpenSSL 1.1 as commit d35ff2c0a. Drop our own version and use a
backport of what was committed upstream.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3674
Qin Long [Sat, 5 Mar 2016 15:40:40 +0000 (23:40 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3674

A more complete fix for the no-cms configuration has been added to
OpenSSL 1.1 as commit e968561d5. Drop our own version and use a
backport of what was committed upstream.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3955
Qin Long [Sat, 5 Mar 2016 15:39:47 +0000 (23:39 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3955

A different fix for the excessive stack usage has been merged into
OpenSSL 1.1 as commit 8e704858f. Drop our own version and use a backport
of what was committed upstream.

Note: This requires the free() function to work correctly when passed
      a NULL argument (qv).

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3628
Qin Long [Sat, 5 Mar 2016 15:38:51 +0000 (23:38 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3628

A complete implementation of the no-filenames configuration option was
added to OpenSSL 1.1 in commit 02f7114a7. Drop our own version and use
a backport of what was committed upstream.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3964
Qin Long [Sat, 5 Mar 2016 15:37:42 +0000 (23:37 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#3964

Extensive fixes for the no-stdio configuration have been merged into
OpenSSL 1.1, primarily in commit 984d6c605.

The backport to 1.0.2 is slightly different because we still have a
mixture of no-fp-api and no-stdio in 1.0.2, although they are hopelessly
intertwined. Nevertheless, drop our own original version and switch to
a backported version of what went into 1.1.

This includes subsequent fixes in commit c0cf5b84d for the TS code.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#4175
Qin Long [Sat, 5 Mar 2016 15:36:37 +0000 (23:36 +0800)]
CryptoPkg/OpensslLib: Switch to upstream fix for OpenSSL RT#4175

A different fix for the PKCS7_verify() regression on Authenticode
signatures has landed in the OpenSSL 1.0.2 branch as commit c436c990f
and will be present in the 1.0.2g release.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Regenerate OpenSSL patch
Qin Long [Sat, 5 Mar 2016 15:35:04 +0000 (23:35 +0800)]
CryptoPkg/OpensslLib: Regenerate OpenSSL patch

All the OpenSSL changes we carry in our EDKII_openssl patch for 1.0.2
are now merged into upstream OpenSSL and will be in the upcoming 1.1
release.

As a first step towards switching out our original hacks for backported
versions of the commits which were actually accepted into OpenSSL 1.1,
just regenerate the *existing* patch against the 1.0.2f release using
'git diff'.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg/OpensslLib: Include complete copy of opensslconf.h
Qin Long [Sat, 5 Mar 2016 15:33:59 +0000 (23:33 +0800)]
CryptoPkg/OpensslLib: Include complete copy of opensslconf.h

This can be an auto-generated file, and it *isn't* in the OpenSSL git tree;
it's only in the generated tarballs. So rather than including it in our
OpenSSL patch, just have the user copy it into place.

This makes it easier to manage changes, and is a step towards better
integration.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoCryptoPkg: Use OpenSSL include directory directly
Qin Long [Sat, 5 Mar 2016 15:32:55 +0000 (23:32 +0800)]
CryptoPkg: Use OpenSSL include directory directly

The standard OpenSSL 1.0.2 configuration and build process will already
symlink or copy the necessary header files to the include/openssl/
directory within the OpenSSL source tree.

When we transition to OpenSSL 1.1 it won't even be necessary to link
or copy the files there; they have just been moved outright.

So let's use them from there. Change the include directory specified
in CryptoPkg/CryptoPkg.dec, and modify the Install.cmd and Install.sh
scripts to copy the files to the normal directory within the OpenSSL
source tree, instead of CryptoPkg/Include/openssl/.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qin Long <qin.long@intel.com>
Tested-by: Qin Long <qin.long@intel.com>
5 years agoShellPkg: Fix smbiosview decode of PCIe Extended Capabilities
Samer El-Haj-Mahmoud [Fri, 4 Mar 2016 03:31:45 +0000 (19:31 -0800)]
ShellPkg: Fix smbiosview decode of PCIe Extended Capabilities

Fix SMBIOSVIEW decode of the HeaderLog fields of the PCIe
AER structure . The PCIe 2.1 Base Specification, section 7.10, lists
this as 16 bytes, or 4 DWORDs.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
5 years agoMdeModulePkg AcpiTableDxe: Don't uninstall Acpi Sdt Protocol at ReadyToLock
Star Zeng [Thu, 3 Mar 2016 08:49:54 +0000 (16:49 +0800)]
MdeModulePkg AcpiTableDxe: Don't uninstall Acpi Sdt Protocol at ReadyToLock

1. The consumer of Acpi Sdt Protocol may want to use the API after ReadyToLock.
2. The ACPI system configuration table even could be overwritten,
   we see little issue in leaving Acpi Sdt Protocol installed.

Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Feng Tian <feng.tian@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Star Zeng <star.zeng@intel.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
5 years agoMdeModulePkg/PciHostBridge: Don't assert when setting UC to MMIO fails
Ruiyu Ni [Fri, 4 Mar 2016 05:04:21 +0000 (13:04 +0800)]
MdeModulePkg/PciHostBridge: Don't assert when setting UC to MMIO fails

Failing to set EFI_MEMORY_UC to MMIO aperture is not a fatal error.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Laszlo Ersek <lersek@redhat.com>
5 years agoSecurityPkg/SecureBootConfigDxe: Handle allocation failure gracefully
Dandan Bi [Mon, 29 Feb 2016 06:37:07 +0000 (14:37 +0800)]
SecurityPkg/SecureBootConfigDxe: Handle allocation failure gracefully

The function AllocateCopyPool may return NULL, so need to do check
after calling it. This patch is to enhance the related logic.

Cc: Chao Zhang <chao.b.zhang@intel.com>
Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Eric Dong <eric.dong@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Chao Zhang <chao.b.zhang@intel.com>
Reviewed-by: Eric Dong <eric.dong@intel.com>
5 years agoMdeModulePkg/Bds: Support booting from remote file system.
Ruiyu Ni [Wed, 24 Feb 2016 07:15:02 +0000 (15:15 +0800)]
MdeModulePkg/Bds: Support booting from remote file system.

Enhance BDS to support booting from a remote file system exposed
by a HTTP boot option.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
5 years agoMdeModulePkg/Bds: Support short-form URI boot.
Ruiyu Ni [Wed, 24 Feb 2016 07:04:44 +0000 (15:04 +0800)]
MdeModulePkg/Bds: Support short-form URI boot.

The patch adds short-form URI boot support to follow
UEFI Spec.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
5 years agoMdeModulePkg/Bds: Wide match HTTP boot option.
Ruiyu Ni [Wed, 24 Feb 2016 06:47:34 +0000 (14:47 +0800)]
MdeModulePkg/Bds: Wide match HTTP boot option.

Enhance BDS to wide match the HTTP boot option without matching
the specific device path data in IP device path and URI device
path node.
It's to follow UEFI Spec 2.6.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
5 years agoMdeModulePkg/Bds: Refine the code to load file from FV.
Ruiyu Ni [Mon, 11 Jan 2016 06:14:26 +0000 (14:14 +0800)]
MdeModulePkg/Bds: Refine the code to load file from FV.

Change BmGetFileBufferByMemmapFv to BmGetFileBufferByFvFilePath.
The original function gets the file buffer only from memory mapped
FV device path and leaves GUIDed FV device path to the code below;
The new function gets the file buffer from both formats of FV device
paths.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
5 years agoMdeModulePkg: Fix IPv4 double free
Samer El-Haj-Mahmoud [Fri, 4 Mar 2016 07:18:49 +0000 (15:18 +0800)]
MdeModulePkg: Fix IPv4 double free

Fix a possible ASSERT after NBP finishes loading from a PXE
boot. Ip4Dxe driver calls Ip4Config2SetPolicy which calls
CloseEvent on the Dhcp4Event struct member.  After NBP is
downloaded, it then calls Ip4Config2CleanInstance which calls
CloseEvent again on Dhcp4Event.  This double free can cause an
ASSERT. When the event is closed, set the event pointer to NULL
so the Ip4 code won't call CloseEvent on it again.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Ye Ting <ting.ye@intel.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
5 years agoMdeModulePkg: Make HII configuration settings available to OS runtime
Dandan Bi [Wed, 2 Mar 2016 08:53:50 +0000 (16:53 +0800)]
MdeModulePkg: Make HII configuration settings available to OS runtime

This feature is aimed to allow OS make use of the HII database
during runtime. In this case, the contents of the HII Database
is exported to a buffer. The pointer to the buffer is placed
in the EFI System Configuration Table, where it can be retrieved
by an OS application.

Cc: Liming Gao <liming.gao@intel.com>
Cc: Eric Dong <eric.dong@intel.com>
Cc: Brian J. Johnson <bjohnson@sgi.com>
Cc: Andrew Fish <afish@apple.com>
Cc: El-Haj-Mahmoud Samer <samer.el-haj-mahmoud@hpe.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Eric Dong <eric.dong@intel.com>
5 years agoMdeModulePkg/DisplayEngineDxe: Refine the coding style
Dandan Bi [Thu, 3 Mar 2016 08:51:51 +0000 (16:51 +0800)]
MdeModulePkg/DisplayEngineDxe: Refine the coding style

Remove the empty line to conform with current coding style.

Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Eric Dong <eric.dong@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Eric Dong <eric.dong@intel.com>
5 years agoUefiCpuPkg: Add dynamic type for PcdCpuMaxLogicalProcessorNumber
Jeff Fan [Thu, 3 Mar 2016 07:25:26 +0000 (15:25 +0800)]
UefiCpuPkg: Add dynamic type for PcdCpuMaxLogicalProcessorNumber

Currently, PcdCpuMaxLogicalProcessorNumber only supports fixed type. There are
some requests on this PCD to support dynamic type also. For example, platform
may dynamically set this PCD to actual processor count to save memory or improve
boot performance.

Cc: Feng Tian <feng.tian@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
Reviewed-by: Feng Tian <feng.tian@intel.com>
5 years agoArmPkg: Configure TTBCR register
Evan Lloyd [Wed, 2 Mar 2016 21:08:46 +0000 (21:08 +0000)]
ArmPkg: Configure TTBCR register

Architecturally, the TTBCR register value is undefined at reset for
Non-Secure.
On some platforms the reset value for TTBCR is not zero and
this causes a data abort exception once the MMU is enabled.

This patch configures the TTBCR register to enable translation table
walk using TTBR0.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Evan Lloyd <evan.lloyd@arm.com>
5 years agoMdeModulePkg/PciHostBridgeLib: Correct the function header comments
Ruiyu Ni [Thu, 3 Mar 2016 09:47:52 +0000 (17:47 +0800)]
MdeModulePkg/PciHostBridgeLib: Correct the function header comments

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Shumin Qiu <shumin.qiu@intel.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Qiu Shumin <shumin.qiu@intel.com>
5 years agoOvmfPkg: switch to MdeModulePkg/Bus/Pci/PciHostBridgeDxe
Laszlo Ersek [Tue, 26 Jan 2016 00:18:36 +0000 (01:18 +0100)]
OvmfPkg: switch to MdeModulePkg/Bus/Pci/PciHostBridgeDxe

The old driver is retained for now; it remains available with "-D
USE_OLD_PCI_HOST". This is because I'd like to involve end users and
downstreams in testing the new drier, but also allow them to switch back
to the old driver at the first sight of trouble, while we debug the new
driver in parallel.

In a few weeks the ifdeffery and the "OvmfPkg/PciHostBridgeDxe/" driver
should be removed.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: resolve PciSegmentLib
Laszlo Ersek [Tue, 26 Jan 2016 00:29:24 +0000 (01:29 +0100)]
OvmfPkg: resolve PciSegmentLib

In the next patch we'll build "MdeModulePkg/Bus/Pci/PciHostBridgeDxe".
That driver depends on the PciSegmentLib class. Edk2 offers four
instances:

(1) MdePkg/Library/UefiPciSegmentLibPciRootBridgeIo/

    Inappropriate here because it consumes
    EFI_PCI_ROOT_BRIDGE_IO_PROTOCOL, but
    "MdeModulePkg/Bus/Pci/PciHostBridgeDxe" needs the library class for
    producing that protocol.

(2) MdePkg/Library/PeiPciSegmentLibPciCfg2/

    Restricted to PEIM, SEC, and PEI_CORE client modules.

(3) MdePkg/Library/DxePciSegmentLibEsal/

    "uses ESAL services to perform PCI Configuration cycles"

(4) MdePkg/Library/BasePciSegmentLibPci/

    A simple BASE library instance that sits on top of PciLib. This is our
    choice. We can resolve PciSegmentLib to this instance for all module
    types.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: initialize RootBus->DevicePath
Laszlo Ersek [Mon, 25 Jan 2016 23:52:05 +0000 (00:52 +0100)]
OvmfPkg: PciHostBridgeLib: initialize RootBus->DevicePath

We copy the code from InitRootBridge()
[OvmfPkg/PciHostBridgeDxe/PciHostBridge.c], with a slight change: the
device path is allocated separately now.

This is the final field to initialize in PCI_ROOT_BRIDGE.

The type EFI_PCI_ROOT_BRIDGE_DEVICE_PATH is renamed to
OVMF_PCI_ROOT_BRIDGE_DEVICE_PATH. The original is a misnomer (it is not a
standard UEFI type) that dates back to PcAtChipsetPkg/PciHostBridgeDxe.
Simply removing the EFI_ suffix would result in
PCI_ROOT_BRIDGE_DEVICE_PATH, where PCI_ could incorrectly suggest a
relation with the PCI standards or the PCI-related generic edk2 code.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: set RootBus->NoExtendedConfigSpace
Laszlo Ersek [Fri, 26 Feb 2016 23:06:39 +0000 (00:06 +0100)]
OvmfPkg: PciHostBridgeLib: set RootBus->NoExtendedConfigSpace

In "OvmfPkg/PciHostBridgeDxe/PciRootBridgeIo.c", the
RootBridgeIoCheckParameter() function hard-codes the maximum offset for
the PCI config space as 0xFF (see the MAX_PCI_REG_ADDRESS macro), which
matches OVMF's 0xCF8 / 0xCFC config access method.

The "MdeModulePkg/Bus/Pci/PciHostBridgeDxe" driver abstracts away config
space access via the PciSegmentLib class, so it has to be informed
separately about the config space size.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: set bus, IO and 32-bit MMIO windows in RootBus
Laszlo Ersek [Mon, 25 Jan 2016 23:36:46 +0000 (00:36 +0100)]
OvmfPkg: PciHostBridgeLib: set bus, IO and 32-bit MMIO windows in RootBus

The bus aperture is copied verbatim from InitRootBridge()
[OvmfPkg/PciHostBridgeDxe/PciHostBridge.c].

The IO and 32-bit MMIO apertures are matched to PlatformPei's settings.
PciHostBridgeLibDxe expects PciHostBridgeLib instances to advertize the
exact apertures.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: set RootBus->AllocationAttributes
Laszlo Ersek [Mon, 25 Jan 2016 23:11:43 +0000 (00:11 +0100)]
OvmfPkg: PciHostBridgeLib: set RootBus->AllocationAttributes

InitRootBridge() in "OvmfPkg/PciHostBridgeDxe/PciHostBridge.c" passes the
EFI_PCI_HOST_BRIDGE_COMBINE_MEM_PMEM allocation attribute to
RootBridgeConstructor(); we should do the same here.

From "MdePkg/Include/Protocol/PciHostBridgeResourceAllocation.h":

/// If this bit is set, then the PCI Root Bridge does not support separate
/// windows for Non-prefetchable and Prefetchable memory. A PCI bus driver
/// needs to include requests for Prefetchable memory in the
/// Non-prefetchable memory pool.

Which implies that both the 32-bit and 64-bit prefetchable MMIO apertures
should be marked empty. (The CreateRootBridge() function actually enforces
this in "MdeModulePkg/Bus/Pci/PciHostBridgeDxe/PciRootBridgeIo.c".)

Furthermore, since OvmfPkg/PciHostBridgeDxe does *not* set the
EFI_PCI_HOST_BRIDGE_MEM64_DECODE allocation attribute:

/// If this bit is set, then the PCI Root Bridge supports 64 bit memory
/// windows.  If this bit is not set, the PCI bus driver needs to include
/// requests for 64 bit memory address in the corresponding 32 bit memory
/// pool.

we follow suit in the PciHostBridgeLib instance.

In turn, the 64-bit MMIO apertures (both prefetchable and
non-prefetchable) should be marked empty.
MdeModulePkg/Bus/Pci/PciHostBridgeDxe enforces this too.

(64-bit MMIO aperture support, based on yet more fw_cfg files, is a
planned future improvement.)

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: clear RootBus->DmaAbove4G
Laszlo Ersek [Mon, 25 Jan 2016 22:50:14 +0000 (23:50 +0100)]
OvmfPkg: PciHostBridgeLib: clear RootBus->DmaAbove4G

When this BOOLEAN member is FALSE, and the caller tries to set up a DMA
transfer between a PCI device and a host buffer not entirely under 4GB,
then "MdeModulePkg/Bus/Pci/PciHostBridgeDxe" sets up a bounce buffer under
4GB, in the implementation of EFI_PCI_ROOT_BRIDGE_IO_PROTOCOL.Map().

Since that's exactly what RootBridgeIoMap() does in
"OvmfPkg/PciHostBridgeDxe/PciRootBridgeIo.c", stick with it in this
conversion.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: set supported and initial attributes in RootBus
Laszlo Ersek [Mon, 25 Jan 2016 22:47:19 +0000 (23:47 +0100)]
OvmfPkg: PciHostBridgeLib: set supported and initial attributes in RootBus

These settings are copied from the RootBridgeConstructor() function, file
"OvmfPkg/PciHostBridgeDxe/PciRootBridgeIo.c".

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: set RootBus->Segment
Laszlo Ersek [Mon, 25 Jan 2016 22:43:50 +0000 (23:43 +0100)]
OvmfPkg: PciHostBridgeLib: set RootBus->Segment

This is the first of the patches that set the fields of PCI_ROOT_BRIDGE.
The structure is zero-filled as a precaution for later field additions.

Here we set the Segment member explicitly to zero (so that any later
customization can be easier).

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: implement PciHostBridgeFreeRootBridges()
Laszlo Ersek [Mon, 25 Jan 2016 22:17:22 +0000 (23:17 +0100)]
OvmfPkg: PciHostBridgeLib: implement PciHostBridgeFreeRootBridges()

This function has no counterpart in OvmfPkg/PciHostBridgeDxe/, but the
PciHostBridgeLib class requires it.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: PciHostBridgeLib: convert main loop from PciHostBridgeDxe
Laszlo Ersek [Mon, 25 Jan 2016 21:47:00 +0000 (22:47 +0100)]
OvmfPkg: PciHostBridgeLib: convert main loop from PciHostBridgeDxe

In this patch we import the scan for extra root buses from the
InitializePciHostBridge() function, in file
"OvmfPkg/PciHostBridgeDxe/PciHostBridge.c".

For the time being, the InitRootBridge() and UninitRootBridge() functions
are just placeholders.

The PciHostBridgeGetRootBridges() API expects us to return the
PCI_ROOT_BRIDGE structures in a contiguous array, instead of a linked
list. Therefore the following bits have to be converted manually:

(1) The array is allocated in advance, in a single step.

(2) The calculation of the array size depends on an explicit
    multiplication, which we must check against overflow. Since more than
    255 extra root bridges make no sense anyway, we use (1 + 255) as the
    limit on the main plus all extra root bridges. This also ensures that
    the UINTN multiplication doesn't overflow.

(3) The PciHostBridgeDxe code decrements "ExtraRootBridgesLeft" to
    terminate the scanning early. Here we need track the increasing count
    of used array elements as well, so we employ "ExtraRootBridges" as a
    constant limit, and increment the new local variable "Initialized".

(4) The prototypes of InitRootBridge() and UninitRootBridge() reflect that
    the PCI_ROOT_BRIDGE structure is allocated by the caller; only
    in-place initialization is necessary.

Additionally, macros are employed for standard PCI quantities, from
"MdePkg/Include/IndustryStandard/Pci22.h":

- MAX_PCI_DEVICE_NUMBER (31) is replaced with PCI_MAX_DEVICE (same),
- the constant 255 is replaced with PCI_MAX_BUS,
- the (RootBridgeNumber < 256) condition is replaced with
  (RootBridgeNumber <= PCI_MAX_BUS).

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: clone PciHostBridgeLib from MdeModulePkg's Null instance
Laszlo Ersek [Mon, 25 Jan 2016 19:33:32 +0000 (20:33 +0100)]
OvmfPkg: clone PciHostBridgeLib from MdeModulePkg's Null instance

In this patch we clone "MdeModulePkg/Library/PciHostBridgeLibNull" for
customization under OvmfPkg. Differences relative to a verbatim copy:

- the Null suffix is dropped from file names,
- the UNI file is dropped, together with the corresponding MODULE_UNI_FILE
  reference in the INF file,
- the INF file receives a new FILE_GUID,
- the top comments in the files mention OVMF, not a null instance.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: factor the MMIO aperture shared by all PCI root bridges into PCDs
Laszlo Ersek [Fri, 26 Feb 2016 15:29:19 +0000 (16:29 +0100)]
OvmfPkg: factor the MMIO aperture shared by all PCI root bridges into PCDs

Going forward, two modules will need to know about the aperture:
PlatformPei (as before), and OVMF's upcoming PciHostBridgeLib instance
(because the core PciHostBridgeDxe driver requires the library to state
the exact apertures for all root bridges).

On QEMU, all root bridges share the same MMIO aperture, hence one pair of
PCDs suffices.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: factor the IO aperture shared by all PCI root bridges into PCDs
Laszlo Ersek [Fri, 26 Feb 2016 15:29:19 +0000 (16:29 +0100)]
OvmfPkg: factor the IO aperture shared by all PCI root bridges into PCDs

At the moment we don't intend to customize this aperture at runtime, but
going forward, two modules will need to know about it: PlatformPei (as
before), and OVMF's upcoming PciHostBridgeLib instance (because the core
PciHostBridgeDxe driver requires the library to state the exact apertures
for all root bridges).

On QEMU, all root bridges share the same IO port aperture, hence one pair
of PCDs suffices.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOvmfPkg: remove superfluous [PcdsFixedAtBuild] section header
Laszlo Ersek [Fri, 26 Feb 2016 15:24:35 +0000 (16:24 +0100)]
OvmfPkg: remove superfluous [PcdsFixedAtBuild] section header

At the location of this header an earlier [PcdsFixedAtBuild] section is in
effect already.

Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoMdeModulePkg: PciHostBridgeDxe: don't assume extended config space
Laszlo Ersek [Fri, 26 Feb 2016 22:09:46 +0000 (23:09 +0100)]
MdeModulePkg: PciHostBridgeDxe: don't assume extended config space

The "pc" ("pc-i440fx-*") machine types of QEMU don't support extended
config space. Accordingly, OVMF will use the following library instances
in connection with the core PciHostBridgeDxe driver:

  BasePciSegmentLibPci [class: PciSegmentLib]
    BasePciLibCf8      [class: PciLib]
      BasePciCf8Lib    [class: PciCf8Lib]

Add a new field to the PCI_ROOT_BRIDGE structure so that
RootBridgeIoCheckParameter() can catch config space offsets above 0xFF on
such old (emulated) platforms.

Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
5 years agoMdeModulePkg: PciHostBridgeDxe: insert horizontal whitespace
Laszlo Ersek [Tue, 1 Mar 2016 10:04:59 +0000 (11:04 +0100)]
MdeModulePkg: PciHostBridgeDxe: insert horizontal whitespace

... in preparation for the next patch.

Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoMdeModulePkg/FvSimpleFs: don't open DevicePath with BY_DRIVER mode
Feng Tian [Wed, 2 Mar 2016 01:45:06 +0000 (09:45 +0800)]
MdeModulePkg/FvSimpleFs: don't open DevicePath with BY_DRIVER mode

The device path protocol doesn't get closed at FvSimpleFilesystem
driver stop() when it's opened at start() with BY_DRIVER mode.

We change it to open DevicePath with GET_PROTOCOL mode as FV2 protocol with BY_DRIVER is enough.

Cc: Chao Zhang <chao.b.zhang@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Feng Tian <feng.tian@intel.com>
Reviewed-by: Chao Zhang <chao.b.zhang@intel.com>
5 years agoShellPkg: Update 'ifconfig -r' implementation
Jiaxin Wu [Wed, 2 Mar 2016 07:58:25 +0000 (15:58 +0800)]
ShellPkg: Update 'ifconfig -r' implementation

This patch is used to update ifconfig -r implementation
to sync with UEFI Shell 2.2.

option -r means to reconfigure all or specified interface,
and set DHCP policy. If specified interface is already set
to DHCP, then refresh the IPv4 configuration.

If the interface name is specified
with '-r', DHCP DORA process will be triggered by the policy
transition (static -> dhcp).

Cc: Ye Ting <ting.ye@intel.com>
Cc: Fu Siyuan <siyuan.fu@intel.com>
Cc: Carsey Jaben <jaben.carsey@intel.com>
Cc: El-Haj-Mahmoud Samer <samer.el-haj-mahmoud@hpe.com>
Cc: Subramanian Sriram <sriram-s@hpe.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Sriram Subramanian <sriram-s@hpe.com>
Reviewed-by: Jaben Carsey <Jaben.carsey@intel.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
5 years agoShellPkg: Increase reallocation size for temp memory files
Jim_Dailey@Dell.com [Thu, 18 Feb 2016 15:51:33 +0000 (23:51 +0800)]
ShellPkg: Increase reallocation size for temp memory files

If data of any real size were to be piped from one command to another,
an inordinate amount of time could be taken up by reallocating memory
that is only 10 bytes bigger than what is currently needed. Also, this
could cause unwelcome memory fragmentation.

Added a define to control how much memory is reallocated beyond that
which is currently needed. Set it to 1K vs. the original 10 bytes.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jim Dailey <jim_dailey@dell.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
5 years agoShellPkg: Do not write the UNICODE BOM on ConOut.
Jim_Dailey@Dell.com [Thu, 18 Feb 2016 14:47:07 +0000 (22:47 +0800)]
ShellPkg: Do not write the UNICODE BOM on ConOut.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jim Dailey <jim_dailey@dell.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
5 years agoShellPkg: Add FileSize member to shell memory file structure.
Jim_Dailey@Dell.com [Thu, 18 Feb 2016 14:40:46 +0000 (22:40 +0800)]
ShellPkg: Add FileSize member to shell memory file structure.

The shell uses the memory file structure to manage temporary files in
memory that support piping of output from one command into the the
input of another command.  The BufferSize member is the size of the
internal buffer, not the size of the data that was written to the
file. So, it was possible to read beyond the EOF of these files as
reads used BufferSize. Now FileSize tracks the actual size of these
files (the number of bytes written, not the number of bytes available
in the buffer), and the reads use this member.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jim Dailey <jim_dailey@dell.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
5 years agoBaseTools/LZMA: fix the format issue for last patch
Yonghong Zhu [Tue, 1 Mar 2016 04:41:30 +0000 (12:41 +0800)]
BaseTools/LZMA: fix the format issue for last patch

There are no functional changes in this patch. fixing the format base on
last commit.
The only change is 1) add back the blank line, which can help we better
compare with the original LZMA source code. 2) remove the indent of
#ifndef and #endif.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Yonghong Zhu <yonghong.zhu@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
5 years agoUefiCpuPkg/Application/Cpuid: Change DisplayFamily/DisplayModel type
Jeff Fan [Wed, 2 Mar 2016 02:05:44 +0000 (10:05 +0800)]
UefiCpuPkg/Application/Cpuid: Change DisplayFamily/DisplayModel type

Change DisplayFamily/DisplayModel type to UINT32 to avoid different size in
bitwise operation.

Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
Reviewed-by: Qiu Shumin <shumin.qiu@intel.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
5 years agoUefiCpuPkg/Application/Cpuid: Remove unnecessary code check
Jeff Fan [Wed, 2 Mar 2016 02:00:31 +0000 (10:00 +0800)]
UefiCpuPkg/Application/Cpuid: Remove unnecessary code check

gMaximumBasicFunction is set to CPUID_SIGNATURE as below, so removed the compare
code.
UINT32  gMaximumBasicFunction = CPUID_SIGNATURE;

Cc: Qiu Shumin <shumin.qiu@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
Reviewed-by: Qiu Shumin <shumin.qiu@intel.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
5 years agoUefiCpuPkg/CpuMpPei: Load microcode if found newer revision
Jeff Fan [Wed, 2 Mar 2016 01:45:14 +0000 (09:45 +0800)]
UefiCpuPkg/CpuMpPei: Load microcode if found newer revision

Current implementation only loads the latest revision if there is no microcode
loaded. Per IA32 User Manual, we should load the newer revision by comparing
the existing microcode loaded with the updating revision.

Cc: Feng Tian <feng.tian@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jeff Fan <jeff.fan@intel.com>
Reviewed-by: Feng Tian <feng.tian@intel.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
5 years agoEmulatorPkg: Use UEFI Shell by default
Jordan Justen [Tue, 1 Mar 2016 23:39:43 +0000 (15:39 -0800)]
EmulatorPkg: Use UEFI Shell by default

Now we will build the UEFI shell from source and include it in the
EmulatorPkg build.

USE_OLD_SHELL can be defined on the command line to force using the
old shell.

Fixes: https://github.com/tianocore/edk2/issues/60
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: Andrew Fish <afish@apple.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
5 years agoShellPkg: Revert git 'd6cf1af9' fix
Jiaxin Wu [Thu, 25 Feb 2016 03:09:17 +0000 (11:09 +0800)]
ShellPkg: Revert git 'd6cf1af9' fix

'd6cf1af9' is associated with '3d0a49ad' commit. So, this patch is
used to respond the revert for '3d0a49ad' to adapt the Ipv4 config
policy update.

Cc: Subramanian Sriram <sriram-s@hpe.com>
Cc: El-Haj-Mahmoud Samer <samer.el-haj-mahmoud@hpe.com>
Cc: Ye Ting <ting.ye@intel.com>
Cc: Fu Siyuan <siyuan.fu@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Ye Ting <ting.ye@intel.com>
5 years agoNetworkPkg: Revert git 'eb213f2f' fix
Jiaxin Wu [Thu, 25 Feb 2016 03:06:14 +0000 (11:06 +0800)]
NetworkPkg: Revert git 'eb213f2f' fix

'eb213f2f' is associated with '3d0a49ad' commit. So, this patch is
used to respond the revert for '3d0a49ad' to adapt the Ipv4 config
policy update.

Cc: Subramanian Sriram <sriram-s@hpe.com>
Cc: El-Haj-Mahmoud Samer <samer.el-haj-mahmoud@hpe.com>
Cc: Ye Ting <ting.ye@intel.com>
Cc: Fu Siyuan <siyuan.fu@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Ye Ting <ting.ye@intel.com>
5 years agoNetworkPkg: Change the default IPv6 config policy
Jiaxin Wu [Thu, 25 Feb 2016 02:52:13 +0000 (10:52 +0800)]
NetworkPkg: Change the default IPv6 config policy

The default policy for Ip6Config is Ip6ConfigPolicyAutomatic,
which results in all NIC ports starting SARR process when it
receives RA message with M flag from IPv6 router. So, this
patch is used to changes the the default IPv6 config policy to
Ip6ConfigPolicyManualand also defer the SetData operation after
Ip6ConfigProtocol installed. This update let the other platform
drivers have chance to change the default config data by consume
Ip6ConfigProtocol.

Cc: Subramanian Sriram <sriram-s@hpe.com>
Cc: El-Haj-Mahmoud Samer <samer.el-haj-mahmoud@hpe.com>
Cc: Ye Ting <ting.ye@intel.com>
Cc: Fu Siyuan <siyuan.fu@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Subramanian Sriram <sriram-s@hpe.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Ye Ting <ting.ye@intel.com>
5 years agoMdeModulePkg: Change the default IPv4 config policy
Jiaxin Wu [Thu, 25 Feb 2016 02:48:58 +0000 (10:48 +0800)]
MdeModulePkg: Change the default IPv4 config policy

Git version '3d0a49ad' commit provided a scenario to resolve the
performance issue for IPv4, but it's not workable for IPv6. To
avoid IPv4 and IPv6 inconsistency, we decided to revert that version
fix.

If so, the default policy for Ip4Config2 is Ip4Config2PolicyDhcp,
which results in all NIC ports attempting DHCP. So, this patch is
used to changes the the default IPv4 config policy to
Ip4Config2PolicyStatic and also defer the SetData operation after
Ip4Config2Protocol installed. This update let the other platform
drivers have chance to change the default config data by consume
Ip4Config2Protocol.

Cc: Subramanian Sriram <sriram-s@hpe.com>
Cc: El-Haj-Mahmoud Samer <samer.el-haj-mahmoud@hpe.com>
Cc: Ye Ting <ting.ye@intel.com>
Cc: Fu Siyuan <siyuan.fu@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Jiaxin Wu <jiaxin.wu@intel.com>
Reviewed-by: Subramanian Sriram <sriram-s@hpe.com>
Reviewed-by: Fu Siyuan <siyuan.fu@intel.com>
Reviewed-by: Ye Ting <ting.ye@intel.com>
5 years agoMdeModulePkg: Add ASSERT to make sure pointers are not NULL
Dandan Bi [Mon, 29 Feb 2016 07:08:56 +0000 (15:08 +0800)]
MdeModulePkg: Add ASSERT to make sure pointers are not NULL

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Eric Dong <eric.dong@intel.com>
5 years agoMdeModulePkg: FileExplorerLib: Modify the EFI_FILE_EXPLORE_FORMSET_GUID
Dandan Bi [Tue, 1 Mar 2016 06:36:46 +0000 (14:36 +0800)]
MdeModulePkg: FileExplorerLib: Modify the EFI_FILE_EXPLORE_FORMSET_GUID

Currently the EFI_FILE_EXPLORE_FORMSET_GUID value in BdsDxe in
IntelFrameworkModulePkg is same with the GUID VALUE in FileExplorerLib.
This patch is to modify the EFI_FILE_EXPLORE_FORMSET_GUID in
FileExplorerLib to avoid potential issue.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Eric Dong <eric.dong@intel.com>
5 years agoOvmfPkg: copy log level comments from DebugLib.h
Laszlo Ersek [Tue, 1 Mar 2016 09:39:49 +0000 (10:39 +0100)]
OvmfPkg: copy log level comments from DebugLib.h

Cc: Jordan Justen <jordan.l.justen@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Leif Lindholm <leif.lindholm@linaro.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
5 years agoOmap35xxPkg: sync log level comments to DebugLib.h
Laszlo Ersek [Tue, 1 Mar 2016 09:28:50 +0000 (10:28 +0100)]
Omap35xxPkg: sync log level comments to DebugLib.h

Cc: Leif Lindholm <leif.lindholm@linaro.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Leif Lindholm <leif.lindholm@linaro.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>