Ard Biesheuvel [Tue, 4 Aug 2015 18:41:19 +0000 (18:41 +0000)]
ArmVirtPkg/ArmVirtQemu: drop ARM BDS and make Intel BDS the default
ARM BDS support in ArmVirtQemu has been broken since SVN r17969
("ArmPkg/BdsLib: Remove Linux loader from BdsLib") dated July 14th.
Instead of fixing this, let's get rid of the ARM BDS and LinuxLoader
altogether: they violate both the UEFI spec and the arm64 Linux boot
protocol, and lack the level of integration with the QEMU command
line that the Intel BDS has when running under ArmVirtPkg or OvmfPkg.
Dandan Bi [Tue, 4 Aug 2015 09:09:10 +0000 (09:09 +0000)]
MdeModulePkg:Fix the issue that refresh the question fail in DriverSample
Once the question is refreshed,the processing should happen in the callback function
of EFI_BROWSER_ACTION_RETRIEVE case,rather than EFI_BROWSER_ACTION_CHANGING in DriverSample.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Dandan Bi <dandan.bi@intel.com> Reviewed-by: Eric Dong <eric.dong@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18150 6f19259b-4bc3-4df7-8a09-765794883524
Why?
Some platforms maybe report both below 4G and above 4G memory resource hob as tested,
then CoreInitializeMemoryServices() will find and add the tested above 4G memory resource hob to memory descriptor for early memory allocation services,
then if ProcessLibraryConstructorList() that has library constructor tries to allocate below 4G memory and will fail.
In fact, the following CoreInitializeGcdServices() will add all the memory resource hob to GCD map,
and add the tested below 4G memory resource hob to memory descriptor,
but it has been too late for ProcessLibraryConstructorList().
Also move below two lines as they are needed to be after
the constructor of DxeCorePerfomanceLib.
PERF_END (NULL,"PEI", NULL, 0) ;
PERF_START (NULL,"DXE", NULL, 0) ;
Tapan Shah [Tue, 4 Aug 2015 01:09:49 +0000 (01:09 +0000)]
MdePkg/UefiDevicePathLib: Fix RAM Disk Device Path To Text Issue
Perform Left Shift 32 bits of a 32-bit StartAddr[1] and EndingAddr[1]
instead of Right Shift when displaying 64-bit Start and End Address Value.
StartAddr[1] and EndingAddr[1] are already a 32-bit value and it should
perform left shift 32-bit to generate a complete 64-bit value along with
StartAddr[0] and EndingAddr[0] respectively.
Dandan Bi [Mon, 3 Aug 2015 09:44:04 +0000 (09:44 +0000)]
MdeModulePkg:Fix the issue FindQuestionFromProgress in SetupBrowserDxe is broken
If the storage of the question is EFI_HII_VARSTORE_BUFFER/EFI_HII_VARSTORE_EFI_VARIABLE_BUFFER,
in SetupBrowserDxe the configuration stings contain uppercase,but HiiDataBaseDxe generates the
ConfigResp string in lowercase,they mismatch,so FindQuestionFromProgress function is broken.
Now convert the configuration string in SetupBrowserDxe to lowercase to fix this issue.
Ard Biesheuvel [Mon, 3 Aug 2015 08:23:59 +0000 (08:23 +0000)]
BaseTools IA32/X64: Use GccBase.lds instead of gcc*-ld-script
These scripts all now have the same contents, so we only need to use
GccBase.lds. Therefore we can delete gcc-4K-align-ld-script,
gcc4.4-ld-script and gcc4.9-ld-script.
Ard Biesheuvel [Mon, 3 Aug 2015 08:23:26 +0000 (08:23 +0000)]
ArmVirtPkg: move to unified GCC linker script
Move to the parametrised generic GCC linker script and set 64 KB
alignment, instead of using the AARCH64 specific incremental linker
script for 64 KB alignment which is about to be removed.
Ard Biesheuvel [Mon, 3 Aug 2015 08:23:16 +0000 (08:23 +0000)]
ArmPlatformPkg/ArmVExpressPkg: move to unified GCC linker script
Move to the parametrised generic GCC linker script and set 64 KB
alignment, instead of using the AARCH64 specific incremental linker
script for 64 KB alignment which is about to be removed.
Ard Biesheuvel [Mon, 3 Aug 2015 08:22:50 +0000 (08:22 +0000)]
BaseTools GCC: move AutoGen.obj contents to .text section
All AutoGen.obj files consist of global GUID definitions, fixed
and patchable PCDs and other data that is essentially read-only at
runtime but has not been declared as such for various reasons.
By moving these contents to .text we achieve two things:
- global GUIDs and other data items which must be constant for correct
program operation can no longer be modified, for instance, when
running a DXE_RUNTIME_MODULE binary under the OS with the Properties
Table feature for memory protection enabled;
- the .data section becomes smaller, and may be dropped completely for
many XIP modules, which reduces wasted FV space if the PE/COFF section
alignment is large.
Ard Biesheuvel [Mon, 3 Aug 2015 08:22:39 +0000 (08:22 +0000)]
BaseTools GCC: align start of .data to .text alignment
Now that GenFw honors the ELF section alignment when placing the
PE/COFF sections in the output, the start of the PE/COFF version of
.data will be aligned to the alignment of .text if its alignment is
higher than the default. So duplicate this behavior in the ELF output,
this will make the memory layout of the PE/COFF binary match the
layout of the ELF version more closely.
Ard Biesheuvel [Mon, 3 Aug 2015 08:22:28 +0000 (08:22 +0000)]
BaseTools GCC: add unified GCC linker script for all archs and versions
This unifies all GCC linker scripts into a single parametrised GCC
linker script that can be used for all GCC versions and architectures.
The two parameters that can be set on the linker command line are:
- PECOFF_HEADER_SIZE, this is a build time property of GenFw, but
its value is different between 32-bit and 64-bit;
- common-page-size, this can be set using -z on the ld command line,
and controls the value of the COMMONPAGESIZE constant when used in
a linker script. This value is used for the minimum section alignment.
Ard Biesheuvel [Mon, 3 Aug 2015 08:22:16 +0000 (08:22 +0000)]
BaseTools IA32/X64: get header size and alignment from ld commandline
Instead of hardcoding the values for the PE/COFF header size and the
section alignment, set them on the linker command line. This factors
out these values from the various linker scripts, which will allow us
to unify them in a subsequent patch.
Ard Biesheuvel [Mon, 3 Aug 2015 08:22:00 +0000 (08:22 +0000)]
BaseTools IA32/X64: move .got contents to the PE/COFF .text section
Move the .got contents to the PE/COFF .text section. This should be
a no-op, since we typically don't generate position independent code
(i.e., using -fPIC). But since the GOT contains variable addresses that
are updated at relocation time only, its contents are best kept in .text
to prevent them from being overwritten inadvertently.
Ard Biesheuvel [Mon, 3 Aug 2015 08:21:49 +0000 (08:21 +0000)]
BaseTools IA32/X64: drop redundant alignment from linker script
There is no need to pad out the end of a section of the start of
the following section is aligned to the same value. So drop the
redundant ALIGN() statements.
Ard Biesheuvel [Mon, 3 Aug 2015 08:21:39 +0000 (08:21 +0000)]
BaseTools IA32/X64: move .rodata to PE/COFF .text section
The .rodata ELF section contains constant non-executable data that
should never be modified by the program itself. Since the risk of
inadvertent modification is typically higher than the risk of
inadvertent execution, it makes sense to put this data in the
R-X .text section rather than in the RW- .data section.
So move it there.
Ard Biesheuvel [Mon, 3 Aug 2015 08:21:28 +0000 (08:21 +0000)]
BaseTools IA32/X64: remove NOP padding from X86/IA32 GCC linker scripts
The NOP padding in the GCC linker scripts ensures that all empty
regions in the ELF binary are filled with x86 NOP instructions.
There is no upside to doing this: if the CPU ends up executing these
instructions, we have little hope of resuming normal execution of the
program anyway. And having NOP slides in memory only makes it easier
for attackers to launch exploits. So remove them.
Liming Gao [Mon, 3 Aug 2015 03:18:37 +0000 (03:18 +0000)]
MdeModulePkg PeiCore: Add PCD to specify PEIM Shadow
v2 changelog:
Check CurrentPeimHandle to check the matched PeimHandle.
Add check point to ShadowPeiCore based on PCD.
v1 changelog:
PeiCore LoadImage always shadow itself and PEIM on normal boot after
the physical memory is installed. On the emulator platform, the shadow
may be not necessary. To support such usage, new PCD PcdShadowPeimOnBoot
is introduced to specify whether loads PEIM in memory by default.
Yao, Jiewen [Sun, 2 Aug 2015 04:02:37 +0000 (04:02 +0000)]
Add Dual-FSP support (MemoryInitUpd/SiliconInitUpd)
Add FspUpdSignatureCheck() API in FspSecPlatformLib, so that FspSecCore can check if UPD data is valid in FSP API.
Add Set/GetFspMemoryInitUpdDataPointer() and Set/GetFspSiliconInitUpdDataPointer() API in FspCommonLib,
so that core can set this UdpDataPointer and platform code may get UpdDataPointer easily.
Add UpdateMemSiUpdInitOffsetValue function in GenCfgOpt.py tool, so that the MemoryInitUpdOffset and SiUpdInitOffset is recorded.
Add missing EMBED comment in GenCfgOptUserManual.docx
ArmVirtPkg: use 'auto' alignment and FIXED placement for XIP modules
Now that GenFw correctly propagates the minimum alignment of the ELF
input sections to the PE/COFF binary, we can simply select 'auto'
alignment in the FDF Rule section instead of tweaking it by hand.
Also add the FIXED FFS attribute to the module types that may execute
in place. This enables a newly added optimization in GenFfs that strips
redundant padding, preventing excessive waste of FV space if the section
alignment is considerable (i.e., 2 KB or 4 KB)
Harry Liebel [Thu, 30 Jul 2015 09:51:04 +0000 (09:51 +0000)]
StdLib: Add support for AArch64
- Use some files from ARM version.
- Use NetBSD software floating point library to provide floating point
operations not handled directly by hardware floating point enabled
GCC compiler.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Harry Liebel <Harry.Liebel@arm.com> Reviewed-by: Olivier Martin <Olivier.Martin@arm.com> Reviewed-by: Daryl McDaniel <edk2-lists@mc2research.org>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18118 6f19259b-4bc3-4df7-8a09-765794883524
Harry Liebel [Thu, 30 Jul 2015 09:50:58 +0000 (09:50 +0000)]
StdLib/LibC: Provide missing ARM symbols
Provide missing functionality by using files from LLVM.
Changes made:
- Formatting changes (tabs to spaces, DOS line endings etc).
- Simplified 'int_endianness.h' to work for our case.
- Added LLVM licence to the individual files.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Harry Liebel <Harry.Liebel@arm.com> Reviewed-by: Olivier Martin <Olivier.Martin@arm.com> Reviewed-by: Daryl McDaniel <edk2-lists@mc2research.org>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18117 6f19259b-4bc3-4df7-8a09-765794883524
StdLib/LibC: Add software floating point library from NetBSD
Floating point processing is not supported on ARM for UEFI. In order to
support UEFI applications in AppPkg we use this library to provide the
required functionality.
Changes as compared to the NetBSD version:
- Formatting changes (tabs to spaces, DOS line endings etc).
- Disable exceptions as described in the float_raise() function.
- Disable definition of 'Symbolic Boolean literals' in milieu.h.
Source originally from: NetBSD project
- Source: http://cvsweb.netbsd.org/bsdweb.cgi/?only_with_tag=MAIN
- Licensing and Copyright: http://www.netbsd.org/about/redistribution.html
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Harry Liebel <Harry.Liebel@arm.com> Reviewed-by: Olivier Martin <Olivier.Martin@arm.com> Reviewed-by: Daryl McDaniel <edk2-lists@mc2research.org>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18116 6f19259b-4bc3-4df7-8a09-765794883524
Ruiyu Ni [Thu, 30 Jul 2015 06:40:36 +0000 (06:40 +0000)]
MdeModulePkg: Enhance PciBusDxe to handle high 32bit of MEM64 BAR returns 0
According to the PCI spec, when software writes all-one to BAR for size probing,
the value read back should be 0b1...10...0 after masking the BAR type bits.
But in real world, it's possible that certain device returns 0b0...01...10...0
for MEM64 BAR size probing: some bits in the high 32bit may be 0.
PciBus driver has the code to handle such case. However, it doesn't handle the
case that the high 32bit is totally 0. The patch is to handle the special case.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com> Reviewed-by: Eric Dong <eric.dong@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18114 6f19259b-4bc3-4df7-8a09-765794883524
ArmPlatformPkg: remove obsolete ARM and AARCH64 platforms
Remove obsolete ARM and AARCH64 platforms so the maintainers can
focus on the ones that are still supported, which are:
- TC2 (ArmVExpress-CTA15-A7.dsc)
- Foundation model and Fast model emulators (ArmVExpress-FVP-AArch64.dsc)
- Juno (ArmJunoPkg/ArmJuno.dsc)
- Cortex-A15 MPcore RTSM (ArmVExpress-RTSM-A15_MPCore)
MdeModulePkg: Fix the issue cannot boot to UEFI Network after reset
DHCP4 service allows only one of its children to be configured
in the active state,If the DHCP4 D.O.R.A started by IP4 auto
configuration and has not been completed, the Dhcp4 state machine
will not be in the right state for the PXE to start a new round
D.O.R.A., so we need to switch it's policy to static.
NetworkPkg: Fix the issue cannot boot to UEFI Network after reset
DHCP4 service allows only one of its children to be configured
in the active state,If the DHCP4 D.O.R.A started by IP4 auto
configuration and has not been completed, the Dhcp4 state machine
will not be in the right state for the PXE to start a new round D.O.R.A.
so we need to switch it's policy to static.
Since it is arguably incorrect to infer the GIC revision from CPU
ID and GIC feature registers on platforms that describe the GIC in
the device tree, this implements the library class ArmGicArchLib
tailored for such platforms.
The supported GIC revision is retrieved from the dynamic PCD that
is set based on the GIC DT node.
This means this library can only execute post DXE core, but this is
not a problem for any of the virt platforms.
In order to allow a ArmGicArchLib to be implemented that returns
the supported GIC revision based on the device tree, add handling
to VirtFdtDxe to record the GIC revision at DT parsing time.
Instead of inferring the GIC revision from the CPU id registers
and the presence/availability of the system register interface
upon each invocation, move the logic to a constructor and cache
the result.
Clone ArmGicArchLib into a SEC phase specific ArmGicArchSecLib
so that we can modify the former in a subsequent patch to cache
the GIC revision in a global variable.
The current implementation of ArmGicGetSupportedArchRevision ()
that is used by all ARM platforms is entirely stateless (in order
to support being executed from flash) so it needs to interrogate
the hardware for the supported GIC revision upon each invocation.
However, this statelessness is only needed for SEC type modules;
in all other cases, we could easily determine the GIC revision once,
and store the result in a global variable.
In preparation of having separate early and normal versions, this patch
introduces the ArmGicArchLib library class and default implementation,
and moves the existing ArmGicGetSupportedArchRevision () into it.
Before splitting off ArmGicArchLib and moving it out of
ArmPkg/Drivers/ArmGic into ArmPkg/Library, make sure that the
GIC specific declarations it depends on are not hidden away in
local headers "GicV2/GicV2Lib.h" and "GicV3/GicV3Lib.h".
So merge them with <Library/ArmGicLib.h>. This is entirely
appropriate, since this is not a header that declares a public
interface into ArmGicLib, but defines implementation internals.
Add a new module MemoryOverwriteRequestControlLock to register VarCheck handler to enforce MorLock Policy.
Only SMM version is added because MOR is only supported in SMM variable case.
Downgrade one debug message level in DxeTpm2MeasureBootHandler from EFI_D_ERROR to EFI_D_INFO.
No TPM2 is considered as valid case. For example, a platform may only have TPM1.2, without TPM2.0 So this is NOT an ERROR message, but more an INFO message.
GLOBAL_REMOVE_IF_UNREFERENCED means __declspec(selectany) for MS tool
chain IA32/X64 build.
According to MSDN, "selectany" would tell the compiler that the declared
global data item (variable or object) is a pick-any COMDAT (a packaged
function). At link time, if multiple definitions of a COMDAT are seen,
the linker picks one and discards the rest.
So we have to remove GLOBAL_REMOVE_IF_UNREFERENCED key word from two global
variable's extern definitions of .h file as we have used this keyword in
.c file, otherwise the MS tool chain would pick the definitions in .h file
and treat it as uninitialized data to generate full-0 content for these two
global variables.
IntelFrameworkModulePkg: Add missing PCD usage information in UNI and DEC files.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Qiu Shumin <shumin.qiu@intel.com> Signed-off-by: Jeff Fan <jeff.fan@intel.com> Reviewed-by: Jeff Fan <jeff.fan@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18084 6f19259b-4bc3-4df7-8a09-765794883524
To prevent double padding of XIP modules leading to excessive
waste of FV space, try to adjust existing padding rather than
adding more.
Instead of adding a pad file to the FV to line up an FFS file that
itself may contain padding to line up the payload, try to find a
dedicated padding section inside the FFS, and reduce its size to
place all subsequent aligned FFS section at their respective minimum
alignments.
When using 4 KB section alignment (which is required on AARCH64 in
some cases), this will save 4 KB for each XIP module.
BaseTools: use GUID identifiable section for FFS alignment padding
Instead of using an anonymous section of type EFI_SECTION_RAW to pad
out the first aligned FFS section to its required alignment, use a
section with a dedicated GUID if the size of the padding permits it.
This allows for more flexibility when placing such FFS images in a
firmware volume, because we will now be able to remove padding rather
than add more, by shrinking the size of this section instead of
padding out the start of the FFS image to file alignment.
BaseTools/GenFw: move PE/COFF header closer to payload
The secondary header (not the DOS header) of a PE/COFF binary
does not reside at a fixed offset. Instead, its offset into the
file is recorded in the DOS header.
This gives us the flexibility to move it, along with the section
headers, to right before the first section if there is considerable
space before it, i.e., when the PE/COFF file alignment is substantially
larger than the size of the header.
Since the PE/COFF to TE conversion replaces everything before the
section headers with a simple TE header, this change removes all
the header padding from such images, leading to smaller files.
BaseTools/GenFw: move .debug contents to .data to save space
In order to reduce the memory footprint of PE/COFF images when
using large values for the PE/COFF section alignment, move the
contents of the .debug section to data, and point the debug data
directory entry to it. This allows us to drop the .debug section
entirely, as well as any associated rounding. Since our .debug
section only contains the filename of the ELF input image, the
penalty of keeping this data in a non-discardable section is
negligible.
Note that the PE/COFF spec v6.3 explicitly mentions that this is
allowed.
Dandan Bi [Mon, 27 Jul 2015 09:49:01 +0000 (09:49 +0000)]
MdeModulePkg:Add ESC key support in setup browser pop-up message box
When load default value or save changes will pop up message box to let user
confirm,Y means confirm and N means ignore,now add Esc key with the
same function of N key.And change the content of pop up message in .uni file,
now will display "Press 'Y' to confirm, 'N'/'ESC' to ignore."
Dandan Bi [Mon, 27 Jul 2015 09:44:39 +0000 (09:44 +0000)]
MdeModulePkg:SetupBrowser output debug message when DisplayEngineDxe is not installed
Enhance the setupbrowserdxe to output debug message when DisplayEngineDxe is
not installed and this will be easy for user to find the reason why can not
enter Setup page.
Reduce reserved memory consumption by page table buffer,
then OS can have more available memory to use.
Take PhysicalAddressBits = 48 and 2MB page granularity as example,
1:1 Virtual to Physical identity mapping page table buffer needs to be
((512 + 1) * 512 + 1) * 4096 = 1075843072 bytes = 0x40201000 bytes.
The code is updated to only allocate 2 pages (1G page enabled) or
6 pages for 4G page table, and 8 extra pages to handles > 4G request
by page fault.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Star Zeng <star.zeng@intel.com> Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18070 6f19259b-4bc3-4df7-8a09-765794883524
We are going to reduce reserved memory consumption by page table buffer,
then OS can have more available memory to use.
Take PhysicalAddressBits = 48 and 2MB page granularity as example,
1:1 Virtual to Physical identity mapping page table buffer needs to be
((512 + 1) * 512 + 1) * 4096 = 1075843072 bytes = 0x40201000 bytes.
The code is updated to build 4G page table by default and only use 8 extra
pages to handles > 4G request by page fault.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Star Zeng <star.zeng@intel.com> Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18069 6f19259b-4bc3-4df7-8a09-765794883524
Reduce reserved memory consumption by page table buffer,
then OS can have more available memory to use.
Take PhysicalAddressBits = 48 and 2MB page granularity as example,
1:1 Virtual to Physical identity mapping page table buffer needs to be
((512 + 1) * 512 + 1) * 4096 = 1075843072 bytes = 0x40201000 bytes.
When BIOS does not support long mode waking vector, only allocate
2 pages (1G page enabled) or 6 pages for 4G page table, and 8 extra
pages to handles > 4G request by page fault.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Star Zeng <star.zeng@intel.com> Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18068 6f19259b-4bc3-4df7-8a09-765794883524
We are going to reduce reserved memory consumption by page table buffer,
then OS can have more available memory to use.
Take PhysicalAddressBits = 48 and 2MB page granularity as example,
1:1 Virtual to Physical identity mapping page table buffer needs to be
((512 + 1) * 512 + 1) * 4096 = 1075843072 bytes = 0x40201000 bytes.
Current BootScriptExecutorDxe handles > 4G request by page fault because
S3ResumePeim only builds 4G page table when long mode waking vector is
not needed, but BootScriptExecutorDxe still assume the page table buffer for
page table is at 1:1 Virtual to Physical identity mapping.
To reduce reserved memory consumption, the code is updated to only use
8 extra pages to handles > 4G request by page fault.
Another, when both BIOS and OS wants long mode waking vector,
S3ResumePei should have established 1:1 Virtual to Physical identity mapping
page table for ACPI spec requirement, so no need to hook page fault handler.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Star Zeng <star.zeng@intel.com> Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18067 6f19259b-4bc3-4df7-8a09-765794883524
Star Zeng [Mon, 27 Jul 2015 00:49:00 +0000 (00:49 +0000)]
MdeModulePkg Variable: Read MonotonicCount by ReadUnaligned64()
As variable HEADER_ALIGNMENT = 4, the MonotonicCount in
AUTHENTICATED_VARIABLE_HEADER may be not UINT64 aligned,
so go to use ReadUnaligned64() to ensure read data correctly.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Star Zeng <star.zeng@intel.com> Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18064 6f19259b-4bc3-4df7-8a09-765794883524
Ruiyu Ni [Sun, 26 Jul 2015 08:07:15 +0000 (08:07 +0000)]
MdeModulePkg: Make boot option description unique
When there are multiple network boot options, user will see multiple
"UEFI Network" boot options. It's hard to distinguish them using the
description.
The patch enhances the boot option generation logic to append " 2"
/" 3"/" 4"/... number suffix to the non-first network boot options.
So the 2nd one becomes "UEFI Network 2".
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com> Cc: Eric Dong <eric.dong@intel.com> Reviewed-by: Eric Dong <eric.dong@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18062 6f19259b-4bc3-4df7-8a09-765794883524
if (ImageHandle == NULL) {
/* ... */
}
if ((ImageHandle == NULL) || (EFI_ERROR(Status))) {
/* ... */
*/
If BdsExpandUsbShortFormDevicePath() returns a non-NULL value, then the
second "if" statement will check Status without the function having
initialized or assigned it.
When BdsExpandUsbShortFormDevicePath() returns non-NULL, Status should be
EFI_SUCCESS; so let us assign it that value up-front.
That is, the bug was not introduced, only exposed, by the patch -- in the
pre-patch state, although the Status variable was set early and
unconditionally, the error code that it may have carried from the failed
gEfiAcpiS3SaveProtocolGuid lookup had nothing to do with the second "if"
statement above.
Cc: Jiewen Yao <jiewen.yao@intel.com> Cc: Jeff Fan <jeff.fan@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Laszlo Ersek <lersek@redhat.com> Reviewed-by: Jeff Fan <jeff.fan@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18061 6f19259b-4bc3-4df7-8a09-765794883524
ShellPkg: Fix the ASSERT issue in Shell 'for' loop
The Length parameter of 'GetNextParameter' is the buffer size in bytes.
While StrnCpys requires user to pass the max number of dest unicode char,
we should convert size in bytes to the number of char.
Scott Duplichan [Sun, 26 Jul 2015 08:04:10 +0000 (08:04 +0000)]
BaseTools: Make AutoGen.h array declaration match AutoGen.c definition
When a quoted string is used as initialization data in a DEC file PCD
entry, the PCD data type in that entry must be VOID*. The created
AutoGen.c defines the PCD data as UINT8[] or UINT16[], depending on
the string type. The created AutoGen.h, however, declares the PCD data
as VOID*. For a standard compile/link, this works because AutoGen.c
doesn't include AutoGen.h. But when GCC LTO is used, the link time
code generation detects the mismatch and the build fails. This
change makes the AutoGen.h PCD data declaration match the AutoGen.c
definition.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Scott Duplichan <scott@notabs.org> Reviewed-by: Yingke Liu <yingke.d.liu@intel.com> Signed-off-by: Laszlo Ersek <lersek@redhat.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18058 6f19259b-4bc3-4df7-8a09-765794883524
MdeModulePkg: Add old IPv4_DEVICE_PATH support for new IScsiDxe
GatewayIpAddress and SubnetMask do not exist in old IPv4_DEVICE_PATH, this
will lead new IScsiDxe to error if IPv4_DEVICE_PATH in system is not
updated.
Following UEFI2.5 spec of IPv4_DEVICE_PATH do a check before accessing
fields only defined in new version, add a judgement here to make old
IPv4_DEVICE_PATH and new IScsiDxe can cowork.
NetworkPkg: Add old IPv4_DEVICE_PATH and IPv6_DEVICE_PATH support
GatewayIpAddress and SubnetMask do not exist in old IPv4_DEVICE_PATH,
IPAddressOrigin, PrefixLength and GatewayIPAddress do not exist in old
IPv6_DEVICE_PATH. This will lead new IScsiDxe to error without updating
IPv4_DEVICE_PATH and IPv6_DEVICE_PATH in system.
Following UEFI2.5 spec of IPv4_DEVICE_PATH do a check before accessing
fields only defined in new version's IPv4_DEVICE_PATH, and revise the same
issue for IPv6_DEVICE_PATH in Iscsi driver.
Dandan Bi [Sun, 26 Jul 2015 08:03:44 +0000 (08:03 +0000)]
MdeModulePkg: Check the case caused by mismatch
When mismatch happens,there exists one case that exit current
form and display last form.Assert code don't cover this case.
Now add check to handle this situation.
Andrew Fish [Sun, 26 Jul 2015 08:03:35 +0000 (08:03 +0000)]
ShellPkg: Fix bad TimeZone (TZ) conversion.
EFI_UNSPECIFIED_TIMEZONE means display local time. TZ of 0 is UTC.
Thus EFI_UNSPECIFIED_TIMEZONE means ignore TZ, 0 means UTC. When this code is
fixed to adust file TZ to local TZ you need to preserve
EFI_UNSPECIFIED_TIMEZONE.
FAT always return EFI_UNSPECIFIED_TIMEZONE.
Modern filesystems, HFS+, NTFS, ext3, etc store time in UTC.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Andrew Fish <afish@apple.com> Reviewed-by: Jaben Carsey <jaben.carsey@intel.com> Signed-off-by: Laszlo Ersek <lersek@redhat.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18051 6f19259b-4bc3-4df7-8a09-765794883524
MdeModulePkg: Fix the issue EfiPxeBcDhcp() may return wrong status.
if the instance of the DHCP4 protocol driver is in the Dhcp4Bound
status that is DHCP configuration has completed, so the Dhcp4->Start
FUNC in the EfiPxcBcDhcp() will return EFI_ALREADY_STARTED status
which lead to EfiPxeBcDhcp FUNC not in correspondence with UEFI spec.
Contributed-under: TianoCore Contribution Agreement 1.0 Signed-off-by: Zhang Lubo <lubo.zhang@intel.com> Reviewed-by: Fu Siyuan <siyuan.fu@intel.com> Reviewed-by: Ye Ting <ting.ye@intel.com>
[lersek@redhat.com: updated copyright year as Siyuan asked] Signed-off-by: Laszlo Ersek <lersek@redhat.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18050 6f19259b-4bc3-4df7-8a09-765794883524
NetworkPkg: Fix the issue EfiPxeBcDhcp() may return wrong status.
if the instance of the EFI DHCP4 protocol driver is in the Dhcp4Bound status
that is DHCP configuration has completed, so the Dhcp4->Start FUNC in
the PxeBcDhcpDora() will return EFI_ALREADY_STARTED status which lead to
EfiPxeBcDhcp FUNC not in correspondence with UEFI spec.
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> Reviewed-by: Ye Ting <ting.ye@intel.com>
[lersek@redhat.com: updated copyright year as requested by Siyuan] Signed-off-by: Laszlo Ersek <lersek@redhat.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18049 6f19259b-4bc3-4df7-8a09-765794883524
MdeModulePkg: Remove TransmitReceive() and ActiveChild dependency
Fix git 59a8cfd4 (SVN r17869) removes DHCP4.TransmitReceive()and DORA
process dependency, but it updated TransmitReceive() to take the ownership
of DhcpSb->ActiveChild but never release it. This will break the retransmit
and lease time out counter of DORA.
To fix that, TransmitReceive() doesn't need to be the ActiveChild, and the
timer routine should be updated to handle the TransmitReceive specially.
Cc: Fu Siyuan <siyuan.fu@intel.com> Cc: Ye Ting <ting.ye@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> Signed-off-by: Laszlo Ersek <lersek@redhat.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18048 6f19259b-4bc3-4df7-8a09-765794883524
projects / mirror_edk2.git / log