Old implementation only finds first matched full device path for a
given short-form device path.
The patch adds internal function BmGetNextLoadOptionBuffer() to finds
all matched full device path for a given short-form device path.
There are 6 kinds of device paths. Some of them match to multiple
load options, some of them don't.
1. Media device path:
Returns multiple load options: The media device path may point
to a physical BlockIo which contains multiple logic partitions,
each logic partitions contains \EFI\BOOT\BOOT${ARCH}.EFI.
2. Short-form hard-drive device path:
Returns one load option because the partition signature is unique.
3. Short-form file-path device path:
Returns multiple load options: There are multiple SimpleFileSystem
instances and each contains the same file.
4. Short-form URI device path:
Returns multiple load options: There are multiple LoadFile
instances and each can boot.
5. Short-form USB device path:
Returns multiple load options: There are multiple UsbIo instances
and each contains the boot-able file.
6. FV device path, device path pointing to SimpleFileSystem, device
path pointing to LoadFile
Returns one load option.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Feng Tian <feng.tian@intel.com>
Cc: Eric Dong <eric.dong@intel.com>
Cc: Jeff Fan <jeff.fan@intel.com>
There are cases that the operands of an expression are all with rank less
than UINT64/INT64 and the result of the expression is explicitly cast to
UINT64/INT64 to fit the target size.
An example will be:
UINT32 a,b;
// a and b can be any unsigned int type with rank less than UINT64, like
// UINT8, UINT16, etc.
UINT64 c;
c = (UINT64) (a + b);
Some static code checkers may warn that the expression result might
overflow within the rank of "int" (integer promotions) and the result is
then cast to a bigger size.
The commit refines codes by the following rules:
1). When the expression is possible to overflow the range of unsigned int/
int:
c = (UINT64)a + b;
2). When the expression will not overflow within the rank of "int", remove
the explicit type casts:
c = a + b;
3). When the expression will be cast to pointer of possible greater size:
UINT32 a,b;
VOID *c;
c = (VOID *)(UINTN)(a + b); --> c = (VOID *)((UINTN)a + b);
4). When one side of a comparison expression contains only operands with
rank less than UINT32:
UINT8 a;
UINT16 b;
UINTN c;
if ((UINTN)(a + b) > c) {...} --> if (((UINT32)a + b) > c) {...}
For rule 4), if we remove the 'UINTN' type cast like:
if (a + b > c) {...}
The VS compiler will complain with warning C4018 (signed/unsigned
mismatch, level 3 warning) due to promoting 'a + b' to type 'int'.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Hao Wu <hao.a.wu@intel.com>
Reviewed-by: Feng Tian <feng.tian@intel.com>
For pointer subtraction, the result is of type "ptrdiff_t". According to
the C11 standard (Committee Draft - April 12, 2011):
"When two pointers are subtracted, both shall point to elements of the
same array object, or one past the last element of the array object; the
result is the difference of the subscripts of the two array elements. The
size of the result is implementation-defined, and its type (a signed
integer type) is ptrdiff_t defined in the <stddef.h> header. If the result
is not representable in an object of that type, the behavior is
undefined."
In our codes, there are cases that the pointer subtraction is not
performed by pointers to elements of the same array object. This might
lead to potential issues, since the behavior is undefined according to C11
standard.
Also, since the size of type "ptrdiff_t" is implementation-defined. Some
static code checkers may warn that the pointer subtraction might underflow
first and then being cast to a bigger size. For example:
UINT8 *Ptr1, *Ptr2;
UINTN PtrDiff;
...
PtrDiff = (UINTN) (Ptr1 - Ptr2);
The commit will refine the pointer subtraction expressions by casting each
pointer to UINTN first and then perform the subtraction:
PtrDiff = (UINTN) Ptr1 - (UINTN) Ptr2;
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Hao Wu <hao.a.wu@intel.com>
Acked-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Feng Tian <feng.tian@intel.com>
BmExpandMediaDevicePath contains a bug that it uses the
uninitialized Handle.
Since the function is called when the Handle supports BlockIo
or SimpleFileSystem, when there is no SimpleFileSystem installed
on the Handle, BlockIo is *guaranteed* to be installed on the Handle.
The fix initializes the Handle by locating the BlockIo protocol
from the device path.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Cc: Jeff Fan <jeff.fan@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
The type of return value for function EfiBootManagerFindLoadOption() is
INTN. When checking its return value, it is unnecessary to type cast -1 to
type UINTN.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Hao Wu <hao.a.wu@intel.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
When firmware boots to UiApp, the memory type information settings
are saved to NV storage and the settings in HOB are changed as well.
Because UiApp is an APPLICATION type of boot option, system doesn't
reset when settings change.
But when user selects OS to boot in UiApp, because the settings in HOB
was updated when booting to UiApp, the BDS doesn't think the settings
change, expected reset doesn't happen.
The patch fixes this issue to not update the settings in HOB.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Jiewen Yao <jiewen.yao@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
The API dispatches the deferred images that are returned from all
DeferredImageLoad instances.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Chao B Zhang <chao.b.zhang@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
BootManagerMenu boot option is handled by EfiBootManagerGetBootManagerMenu.
Don't need to handle it again when parse LoadFile protocol.
In V2, use "BootManagerMenu" instead of "BootMenuApp".
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
Rename local function name BootMenuApp to BootManagerMenu to align to
other public function name.
In V2, use "BootManagerMenu" instead of "BootMenuApp".
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
Instead of comparing a GUID with gZeroGuid via the CompareGuid API, the
commit uses the IsZeroGuid API to check if the given GUID is a zero GUID.
Cc: Feng Tian <feng.tian@intel.com>
Cc: Star Zeng <star.zeng@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Hao Wu <hao.a.wu@intel.com>
Reviewed-by: Feng Tian <feng.tian@Intel.com>
Update boot description to support LoadFile protocol based on FV file.
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
New LoadFileOnFv2 driver will install LoadFile protocol based on FV file.
Update UefiBootManagerLib to find BootMenuApp with LoadFile protocol.
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
This function abstracts the common logic to find BootMenuApp file.
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
The patch re-orders the sequences by putting updating memory type
information before loading the boot option so that the reserved
memory usage by HTTP RAM disk boot can be excluded by the memory
type information updating.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
Reviewed-by: Star Zeng <star.zeng@intel.com>
This is a enhancement to support the case when platform firmware
doesn't support Boot Manager Menu. For now, if BootManagerMenu FFS
can not be retrieved from FV, BDS core code will still register a
boot option for it. Then, this non-functional boot option will
still be booted by user's request (like HotKey or Exit from shell)
to cause additional boot time and error status code reported.
Therefore, it would be good to skip BootManagerMenu boot option
registration and then return error status and Invalid BootOption
data for this case so that the BootManagerBoot() or other consumers
can directly return without doing anything.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Sunny Wang <sunnywang@hpe.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
After booting a large-size ISO RAM disk (HTTP boot option pointing to
a ISO file) and reboot system, system will possibly run into the
following ASSERT because the BDS core code doesn't consider the
case that Memory page management (Page.c) would possibly NOT update
current memory usage statistics(CurrentMemoryTypeInformation) if
system allocates a memory buffer with a large number of pages.
ASSERT [DxeCore] u:\MdeModulePkg\Core\Dxe\Gcd\Gcd.c(2273):
Length >= MinimalMemorySizeNeeded
The BDS code block for skipping counting reserved memory occupied
by RAM Disk didn't consider the Memory page management's behavior
mentioned above, which caused that the
CurrentMemoryTypeInformation[Index1].NumberOfPages will be updated
to a "very big value" because RamDiskSizeInPages is bigger than
CurrentMemoryTypeInformation[Index1].NumberOfPages. For example,
NumberOfPages is 0x9000 (current use) and RamDiskSizeInPages is
0xC0000 (ISO image size). The result will become a very big value
0xFFF49000.
Therefore, we need to add a check to prevent BDS core code updating
wrong data (very big value) to MemoryTypeInformation variable. This
code change is a improvement for fixing this issue for most cases.
There is still a corner case even when the memory bins don't include
the RAM disk memory, the memory used by all other modules exceeds
RamDiskSizeInPages. Ray will send the other patch to fix this corner
case.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Sunny Wang <sunnywang@hpe.com>
Reviewed-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
Reviewed-by: Ruiyu Ni <Ruiyu.ni@intel.com>
Reviewed-by: Star Zeng <star.zeng@intel.com>
It is the follow up of 3ab41b7a32
to replace UnicodeStrToAsciiStr/AsciiStrToUnicodeStr with
UnicodeStrToAsciiStrS/AsciiStrToUnicodeStrS.
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Eric Dong <eric.dong@intel.com>
Cc: Feng Tian <feng.tian@intel.com>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Star Zeng <star.zeng@intel.com>
Reviewed-by: Jaben Carsey <jaben.carsey@intel.com>
Add gPerformanceProtocolGuid in BdsDxe.inf explicitly.
Currently, BdsDxe could build pass as it inherits
gPerformanceProtocolGuid from UefiBootManagerLib.inf.
Also update the usage of gPerformanceProtocolGuid in UefiBootManagerLib.inf.
Cc: Liming Gao <liming.gao@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Star Zeng <star.zeng@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Prevent the BmRepairAllControllers routine in an infinite loop
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Gary Li <garyli@hpe.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Redfine the BmIsValidLoadOptionVariableName function to allow public use. Change name to EfiBootManagerIsValidLoadOptionVariableName to match naming scheme. Check that VariableName is never NULL and allow OptionType and OptionNumber to be optional.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Thomas Palmer <thomas.palmer@hpe.com>
Reviewed-by: Sunny Wang <sunnywang@hpe.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
MemoryTypeInformation don't count the reserved memory used by RAM Disk,
but it still check all types of memory and do reset when any type
of memory size changes.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Siyuan Fu <siyuan.fu@intel.com>
Reviewed-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
The resource free includes to un-register the ram disk device and
free the memory occupied by the ram disk.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Siyuan Fu <siyuan.fu@intel.com>
Reviewed-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
Use reserved memory to hold the buffer for the RAM disk to
follow the ACPI spec requirement.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Siyuan Fu <siyuan.fu@intel.com>
Reviewed-by: Samer El-Haj-Mahmoud <elhaj@hpe.com>
Check for NULL from AllocateCopyPool before setting Count to 1. Also
change sizeof (EFI_HANDLE*) to sizeof (EFI_HANDLE). Handles is a
EFI_HANDLE pointer, so the allocated memory must be the size of
EFI_HANDLE.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Thomas Palmer <thomas.palmer@hpe.com>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
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>
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>
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>
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>
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>
Change EfiBootManagerDeleteLoadOptionVariable() to not just
remove #### from BootOrder but also remove Boot#### variable.
The old behavior tries to do less for performance but it leaves
unreferenced Boot#### which cannot be reclaimed in variable
reclaim operation though the Boot#### will be eventually be overwritten
by EfiBootManagerAddLoadOptionVariable().
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@19626 6f19259b-4bc3-4df7-8a09-765794883524
BDS puts a special GUID in boot option optional data for
auto-discovered boot option. But when launching that boot
option, the BDS core unconditionally pass the special GUID
to the executable.
A good written application/OS loader can ignore the unexpected
parameters, but BDS core should still avoid passing the
unnecessary GUID.
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@19007 6f19259b-4bc3-4df7-8a09-765794883524
Align to old BDS behavior (IntelFrameworkModulePkg/BDS) to always create
MemoryTypeInfo variable regardless of the PcdResetOnMemoryTypeInformationChange
value.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Michael Kinney <michael.d.kinney@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18926 6f19259b-4bc3-4df7-8a09-765794883524
Ruiyu Ni
Star Zeng
Hao Wu
Liming Gao
Laszlo Ersek
Gary Lin
Sunny Wang
Gary Li
Thomas Palmer
Jordan Justen