EfiCreateProtocolNotifyEvent() takes a (VOID**) for "Registration",
similarly to gBS->RegisterProtocolNotify(). We should pass the address of
an actual pointer-to-VOID, and not the address of an EFI_EVENT. EFI_EVENT
just happens to be specified as (VOID*), and has nothing to do with the
registration.
The same applies to gMmst->MmRegisterProtocolNotify().
"mFtwRegistration", "mFvRegistration", and "mFvbRegistration" are used for
nothing else.
This change is a no-op in practice; it's a semantic improvement.
Cc: Hao A Wu <hao.a.wu@intel.com>
Cc: Jian J Wang <jian.j.wang@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Ray Ni <ray.ni@intel.com>
Cc: Zhichao Gao <zhichao.gao@intel.com>
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Reviewed-by: Philippe Mathieu-Daude <philmd@redhat.com>
GetModuleInfoFromHandle() takes an EFI_HANDLE -- (VOID*) -- as first
parameter, but InsertFpdtRecord() passes (EFI_HANDLE*) -- (VOID**).
(VOID**) converts silently to (VOID*), which is why the wrong cast is
masked.
Note that the *value* that is passed is alright -- therefore this patch
does not change behavior --, it's just semantically wrong to pass an
(EFI_HANDLE*) where an EFI_HANDLE is expected.
Cc: Dandan Bi <dandan.bi@intel.com>
Cc: Eric Dong <eric.dong@intel.com>
Cc: Hao A Wu <hao.a.wu@intel.com>
Cc: Jian J Wang <jian.j.wang@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Ray Ni <ray.ni@intel.com>
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Philippe Mathieu-Daude <philmd@redhat.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
The existing loop is set to call PlatformBootManagerWaitCallback every
second except the last one. We believe this is a mistake as it prevents
the called code from performing timeout expiration tasks such as, for
instance, ensuring that the last segment of a progress bar is displayed
before continuing (which is a current issue for the RPi3 platform).
Signed-off-by: Pete Batard <pete@akeo.ie>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=2218
There is a bug in current driver code that makes
it impossible for it to work with devices that
only work in legacy speed mode since the function
EmmcSwitchToHighSpeed will return with EFI_INVALID_PARAMETER
for such bus mode. Since the logic in that function will
work well for SdMmcMmcLegacy this patch just allows
to call this function with SdMmcMmcLegacy bus mode.
Signed-off-by: Mateusz Albecki <mateusz.albecki@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Tested-by: Hao A Wu <hao.a.wu@intel.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=2218
SD specification recommends switching card bus timing before
switching bus timing in controller. Emmc driver used to do
this switch other way around. This commit adds controller
timing switch in EmmcSwitchBusTiming function to enforce this
order and removes all controller timing programming from
EmmcSwitchToXXX functions.
Signed-off-by: Mateusz Albecki <mateusz.albecki@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Tested-by: Hao A Wu <hao.a.wu@intel.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=2218
Stopping the SD clock before switching host controller
bus timing is only necessary if preset value enable is
set in host controller. Code that performs switch to
HS200 doesn't check this condition right now and we also
don't do the SD clock stop before any other bus timing switch
in host controller. Since the driver is generally not ready to support
host controller with preset value enable set we removed this
code.
Signed-off-by: Mateusz Albecki <mateusz.albecki@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Tested-by: Hao A Wu <hao.a.wu@intel.com>
Tested-by: Marcin Wojtas <mw@semihalf.com>
There is no sourceX offset in case when
Configure->PixelFormat == PixelBlueGreenRedReserved8BitPerColor.
We are copying most left pixels instead of copying required rectangle.
Signed-off-by: Gris87 <Gris87@yandex.ru>
Cc: Ruiyu Ni <ruiyu.ni@intel.com>
Tested-by: Zhichao Gao <zhichao.gao@intel.com>
Reviewed-by: Zhichao Gao <zhichao.gao@intel.com>
For the LoadImage() boot service, with EFI_SECURITY_VIOLATION retval,
the Image was loaded and an ImageHandle was created with a valid
EFI_LOADED_IMAGE_PROTOCOL, but the image can not be started right now.
This follows UEFI Spec.
But if the caller of LoadImage() doesn't have the option to defer
the execution of an image, we can not treat EFI_SECURITY_VIOLATION
like any other LoadImage() error, we should unload image for the
EFI_SECURITY_VIOLATION to avoid resource leak.
This patch is to do error handling for EFI_SECURITY_VIOLATION explicitly
for the caller in PlatformDriOverrideDxe which don't have the policy to
defer the execution of the image.
Cc: Jian J Wang <jian.j.wang@intel.com>
Cc: Hao A Wu <hao.a.wu@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Cc: Philippe Mathieu-Daude <philmd@redhat.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1992
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Reviewed-by: Philippe Mathieu-Daude <philmd@redhat.com>
For the LoadImage() boot service, with EFI_SECURITY_VIOLATION retval,
the Image was loaded and an ImageHandle was created with a valid
EFI_LOADED_IMAGE_PROTOCOL, but the image can not be started right now.
This follows UEFI Spec.
But if the caller of LoadImage() doesn't have the option to defer
the execution of an image, we can not treat EFI_SECURITY_VIOLATION
like any other LoadImage() error, we should unload image for the
EFI_SECURITY_VIOLATION to avoid resource leak.
This patch is to do error handling for EFI_SECURITY_VIOLATION explicitly
for the callers in UefiBootManagerLib which don't have the policy to defer
the execution of the image.
Cc: Jian J Wang <jian.j.wang@intel.com>
Cc: Hao A Wu <hao.a.wu@intel.com>
Cc: Ray Ni <ray.ni@intel.com>
Cc: Zhichao Gao <zhichao.gao@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Cc: Philippe Mathieu-Daude <philmd@redhat.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1992
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Zhichao Gao <zhichao.gao@intel.com>
Reviewed-by: Philippe Mathieu-Daude <philmd@redhat.com>
For the LoadImage() boot service, with EFI_SECURITY_VIOLATION retval,
the Image was loaded and an ImageHandle was created with a valid
EFI_LOADED_IMAGE_PROTOCOL, but the image can not be started right now.
This follows UEFI Spec.
But if the caller of LoadImage() doesn't have the option to defer
the execution of an image, we can not treat EFI_SECURITY_VIOLATION
like any other LoadImage() error, we should unload image for the
EFI_SECURITY_VIOLATION to avoid resource leak.
This patch is to do error handling for EFI_SECURITY_VIOLATION explicitly
for the callers in DxeCapsuleLibFmp which don't have the policy to defer
the execution of the image.
Cc: Jian J Wang <jian.j.wang@intel.com>
Cc: Hao A Wu <hao.a.wu@intel.com>
Cc: Liming Gao <liming.gao@intel.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Cc: Philippe Mathieu-Daude <philmd@redhat.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1992
Signed-off-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Reviewed-by: Philippe Mathieu-Daude <philmd@redhat.com>
The Fpdt driver (FirmwarePerformanceDxe) saves a memory address across
reboots, and then does an AllocatePage for that memory address.
If, on this boot, that memory comes from a Runtime memory bucket,
the MAT table is not updated. This causes Windows to boot into Recovery.
This patch blocks the memory manager from changing the page
from a special bucket to a different memory type. Once the buckets are
allocated, we freeze the memory ranges for the OS, and fragmenting
the special buckets will cause errors resuming from hibernate (S4).
The references to S4 here are the use case that fails. This
failure is root caused to an inconsistent behavior of the
core memory services themselves when type AllocateAddress is used.
The main issue is apparently with the UEFI memory map -- the UEFI memory
map reflects the pre-allocated bins, but the actual allocations at fixed
addresses may go out of sync with that. Everything else, such as:
- EFI_MEMORY_ATTRIBUTES_TABLE (page protections) being out of sync,
- S4 failing
are just symptoms / consequences.
This patch is cherry pick from Project Mu:
a9be767d9b
With the minor change,
1. Update commit message format to keep the message in 80 characters one line.
2. Remove // MU_CHANGE comments in source code.
3. Update comments style to follow edk2 style.
Cc: Jian J Wang <jian.j.wang@intel.com>
Cc: Hao A Wu <hao.a.wu@intel.com>
Cc: Dandan Bi <dandan.bi@intel.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Liming Gao <liming.gao@intel.com>
Acked-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Dandan Bi <dandan.bi@intel.com>
Acked-by: Hao A Wu <hao.a.wu@intel.com>
Initialize PageMapLevel5Entry at the beginning of the function.
This commit will fix a GCC 4.8.5 build failure introduced by commit
b3527dedc3.
OvmfPkg build failure wtih gcc 4.8.5 still exists at latest edk2 version.
The commit 46f8a68916 seems not to fix
the build failure completely.
Cc: Dandan Bi <dandan.bi@intel.com>
Signed-off-by: Shenglei Zhang <shenglei.zhang@intel.com>
Reviewed-by: Liming Gao <liming.gao@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
This commit will fix a GCC 4.8.5 build failure introduced by commit
b3527dedc3.
PageMapLevel5Entry may be uninitialized in original code, which means
uninitialized pointer will be modified at some circumstance.
So relocate the operation of PageMapLevel5Entry++ in order to make sure
the pointer could be modified only when it is initialized.
Cc: Liming Gao <liming.gao@intel.com>
Signed-off-by: Shenglei Zhang <shenglei.zhang@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Reviewed-by: Dandan Bi <dandan.bi@intel.com>
Reviewed-by: Eric Dong <eric.dong@intel.com>
https://bugzilla.tianocore.org/show_bug.cgi?id=1343
This commit adds EDKII_UFS_HC_PLATFORM_PROTOCOL implementation
in UfsPassThruDxe driver in version 1. Driver assumes that at
most one instance of the protocol exists in the system. Presence
of the protocol is not mandatory.
Signed-off-by: Mateusz Albecki <mateusz.albecki@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
https://bugzilla.tianocore.org/show_bug.cgi?id=1343
Private data has been refactored to use EDKII_UFS_HC_INFO structure
to store host controller capabilities and version
information. Getting host controller data has been moved
into single place and is done before host controller enable.
Signed-off-by: Mateusz Albecki <mateusz.albecki@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
https://bugzilla.tianocore.org/show_bug.cgi?id=1343
EDKII_UFS_HC_PLATFORM_PROTOCOL will allow the platform to inject
platform specific logic into standard UFS flows. Right now we
support callbacks pre and post host controller enable and pre
and post link startup. Provided callbacks allow the platform
driver to inject UIC programming after HCE is set which is
a standard initialization step covered by UFS specification as
well as cover some additional use cases during other callbacks.
For instance platform driver may switch to fast mode after link
startup.
We also allow the platform to override host controller capabilities
and version which might be useful to manage silicon bugs or
allow testing experimental features from new versions of the
specification.
Signed-off-by: Mateusz Albecki <mateusz.albecki@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=2008
DxeIpl is responsible to create page table for DXE phase running
either in long mode or in 32bit mode with certain protection
mechanism enabled (refer to ToBuildPageTable()).
The patch updates DxeIpl to create 5-level page table for DXE phase
running in long mode when PcdUse5LevelPageTable is TRUE and CPU
supports 5-level page table.
Signed-off-by: Ray Ni <ray.ni@intel.com>
Reviewed-by: Eric Dong <eric.dong@intel.com>
Regression-tested-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Signed-off-by: Eric Dong <eric.dong@intel.com>
REF: https://bugzilla.tianocore.org/show_bug.cgi?id=1525
The patch is to merge multiple FMP instances into single ESRT entry
when they have the same GUID.
The policy to LastAttemptStatus/LastAttemptVersion of ESRT entry is:
If all the LastAttemptStatus are LAST_ATTEMPT_STATUS_SUCCESS, then
LastAttemptVersion should be the smallest of LastAttemptVersion. If
any of the LastAttemptStatus is not LAST_ATTEMPT_STATUS_SUCCESS,
then the LastAttemptVersion/LastAttemptStatus should be the values
of the first FMP instance whose LastAttemptStatus is not
LAST_ATTEMPT_STATUS_SUCCESS.
To detect possible duplicated GUID/HardwareInstance, a table of
GUID/HardwareInstance pairs from all the EFI_FIRMWARE_IMAGE_DESCRIPTORs
from all FMP instances is built. If a duplicate is found, then generate
a DEBUG_ERROR message, generate an ASSERT(), and ignore the duplicate
EFI_FIRMWARE_IMAGE_DESCRIPTOR.
Add an internal worker function called FmpGetFirmwareImageDescriptor()
that retrieves the list of EFI_FIRMWARE_IMAGE_DESCRIPTORs from a single
FMP instance and returns the descriptors in an allocated buffer. This
function is used to get the descriptors used to build the table of
unique GUID/HardwareInstance pairs. It is then used again to generate
the ESRT Table from all the EFI_FIRMWARE_IMAGE_DESCRIPTORs from all the
FMP instances. 2 passes are performed so the total number of
descriptors is known. This allows the correct sized buffers to always
be allocated.
Cc: Sean Brogan <sean.brogan@microsoft.com>
Cc: Bret Barkelew <Bret.Barkelew@microsoft.com>
Cc: Jian J Wang <jian.j.wang@intel.com>
Signed-off-by: Michael D Kinney <michael.d.kinney@intel.com>
Signed-off-by: Eric Jin <eric.jin@intel.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
Until now, during the USB device enumeration when its PortState
USB_PORT_STAT_CONNECTION bit was not set, the stack was not informed
that the device is not present. Fix that by returning appropriate
error code.
Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Hao A Wu <hao.a.wu@intel.com>
