In UnloadImage() we need to know if an image is already started.
Add a field to the handle structure identifying loaded and started images.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
If LocateHandle() does not find an entry EFI_NOT_FOUND has to be returned
even if BufferSize is NULL.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
CreateEventEx() does not allow the following event types:
* EVT_SIGNAL_EXIT_BOOT_SERVICES
* EVT_SIGNAL_VIRTUAL_ADDRESS_CHANGE
This check is needed to pass the UEFI SCT conformance test.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Alexander Graf <agraf@csgraf.de>
In case of a failure exit data may be passed to Exit() which in turn is
returned by StartImage().
Let the `bootefi` command print the exit data string in case of an error.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
CreateEvent() and CreateEventEx() should check that a notify function is
provided for either of EVT_NOTIFY_SIGNAL or EVT_NOTIFY_WAIT.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
In the current implementation, bootefi command and EFI boot manager
don't use load_image API, instead, use more primitive and internal
functions. This will introduce duplicated code and potentially
unknown bugs as well as inconsistent behaviours.
With this patch, do_efibootmgr() and do_boot_efi() are completely
overhauled and re-implemented using load_image API.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Use efi_root as parent handle for the loaded image.
LoadImage() should be called with BootPolicy = true by the boot manager.
Avoid duplicate free_pool().
Eliminate variable memdp which is not needed after anymore due to
"efi_loader: correctly split device path of loaded image".
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
When the LoadImage() service is called for an image that is already loaded
to memory the file path may be NULL or it will contain both a device path
as well as a media path. We should not assume that there is no media path.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Export function efi_install_multiple_protocol_interfaces() so that we can
call it in others parts of the UEFI subsystem.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
We should not install the HII protocols on every loaded image. It is
sufficient to install them once on the root node.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Those two functions will be used later to re-implement do_bootefi_exec().
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
To meet UEFI spec v2.7a section 9.2, we should add
EFI_LOADED_IMAGE_DEVICE_PATH_PROTOCOL to image handle,
instead of EFI_DEVICE_PATH_PROTOCOL.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Add parameter checks in the StartImage() and Exit() boottime services:
- check that the image handle is valid and has the loaded image protocol
installed
- in StartImage() record the current image
- in Exit() check that the image is the current image
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Commit 0e18f584de ("efi_loader: LoadImage: always allocate new pages")
ensured that whether we load an image from file or from memory we end up
with the same number of newly allocated buffers. But essentially we ended
up with one buffer too many in both cases:
efi_load_pe() copies and rebases the UEFI image.
We do not need the buffer with the file contents afterwards.
Fixes: 0e18f584de ("efi_loader: LoadImage: always allocate new pages")
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
It does not make any sense to check if a pointer is NULL if we have
dereferenced it before.
Reported-by: Coverity (CID 185827)
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
`bootefi selftest` fails on qemu-x86_defconfig if efi_selftest() is not
invoked using EFI_CALL().
Likewise we call the entry point of EFI payloads with
EFI_CALL(efi_start_image()).
entry_count indicates if we are in U-Boot (1) or in EFI payload code (0).
As we start in U-Boot code the initial value has to be 1.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Remove the duplicate code in efi_do_enter() and use efi_start_image() to
start the image invoked by the bootefi command.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Up to now efi_load_pe() returns the entry point or NULL in case of an
error. This does not allow to return correct error codes from LoadImage().
Let efi_load_pe() return a status code and fill in the entry point in the
corresponding field of the image object.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
If we want to properly unload images in Exit() the memory should always be
allocated in the same way. As we allocate memory when reading from file we
should do the same when the original image is in memory.
A further patch will be needed to free the memory when Exit() is called.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
In case of an error we should set the returned pointers to NULL. This
ensures that an illegal free does not occur even if the caller calls
free() for the handles.
If protocols cannot be installed, release all resources.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Heinrich ran into issues with HII and iPXE which lead to #SErrors on
his Odroid-C2 system. We definitely do not want to regress just yet,
so let's not expose the HII protocols by default.
Instead, let's make it a config option that people can play with
This way, we can stabilize the code in tree without breaking any
users.
Once someone figures out, why this breaks iPXE (probably a NULL
dereference), we can enable it by default.
Reported-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Remove HII selftest as well
v2 -> v3:
- Make config option
This patch is a place holder for HII configuration routing protocol and
HII configuration access protocol.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This patch provides enough implementation of the following protocols to
run EDKII's Shell.efi and UEFI SCT:
* EfiHiiDatabaseProtocol
* EfiHiiStringProtocol
Not implemented are:
* ExportPackageLists()
* RegisterPackageNotify()/UnregisterPackageNotify()
* SetKeyboardLayout() (i.e. *current* keyboard layout)
HII database protocol in this patch series can handle only:
* GUID package
* string package
* keyboard layout package
(The other packages, except Device path package, will be necessary
for interactive and graphical UI.)
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
CopyMem() must support overlapping buffers. So replace memcpy() by
memmove().
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
We use u16* for Unicode strings and efi_uintn_t for UINTN. Correct the
signature of efi_exit() and efi_start_image().
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
EFI_ENTRY in efi_connect_controller() should use %pD to print the remaining
device path.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
This reverts commit 3170db63c4.
It reportedly breaks OpenBSD/armv7 booting and I've already received
complaints from people that it breaks some Linux armv7 systems as well.
We'll have to give this whole caching story a good bit more thought.
Reported-by: Jonathan Gray <jsg@jsg.id.au>
Signed-off-by: Alexander Graf <agraf@suse.de>
If UninstallMultipleProtocolInterfaces fails, we sometimes return the wrong
status code. The UEFI spec mandates to always return EFI_INVALID_PARAMETER.
Update unit test.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
Rename the component parent of some EFI objects to header. This avoids
misunderstandings.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
A pointer to a struct efi_object is a handle. We do not need any handle
member in this structure. Let's eliminate it.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
All our handles point to a struct efi_object. So let's define the
efi_handle_t accordingly. This helps us to discover coding errors much
more easily. This becomes evident by the corrections to the usage of
handles in this patch.
Rename variable image_handle to image_obj where applicable.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
When the last protocol interface has been uninstalled remove the handle.
Adjust ReinstallProtocol so that it does not remove the handle.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
As observed with clang:
lib/efi_loader/efi_boottime.c:1624:7: warning: variable 'info'
is used uninitialized whenever 'if' condition is true [-Wsometimes-uninitialized]
if (ret != EFI_SUCCESS)
^~~~~~~~~~~~~~~~~~
lib/efi_loader/efi_boottime.c:1653:7: note: uninitialized use
occurs here
free(info);
^~~~
lib/efi_loader/efi_boottime.c:1624:3: note: remove the 'if' if
its condition is always false
if (ret != EFI_SUCCESS)
^~~~~~~~~~~~~~~~~~~~~~~
lib/efi_loader/efi_boottime.c:1602:31: note: initialize the
variable 'info' to silence this warning
struct efi_loaded_image *info;
^
= NULL
Rather than change how we unwind the function it makes the most sense to
initialize info to NULL so that we can continue to pass it to free().
Fixes: c982874e93 ("efi_loader: refactor efi_setup_loaded_image()")
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
The first parameter of efi_allocate_pool is a memory type. It cannot be
EFI_ALLOCATE_ANY_PAGES. Use EFI_BOOT_SERVICES_DATA instead.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
Create the handle of loaded images and the EFI_LOADED_IMAGE_PROTOCOL
inside efi_setup_loaded_image(). Do not use local variables.
Currently we expect the loaded image handle to point to the loaded image
protocol. Additionally we have appended private fields to the protocol.
With the patch the handle points to a loaded image object and the private
fields are added here. This matches how we handle the net and the gop
object.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
Since GRUB patch d0c070179d4d ("arm/efi: Switch to arm64 linux loader",
2018-07-09) we do not need a workaround for GRUB on 32bit ARM anymore.
So let's eliminate function efi_exit_caches().
This will require Linux distributions to update grub-efi-arm to the GRUB
git HEAD (a tag containing the aforementioned GRUB patch is not available
yet).
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
Currently we assign a lot of protocols to loaded images though
these protocols are not related to them. Instead they should be
installed on a separate handle. Via the device path it is the
parent to the devices like the network adapter.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
The patch implements the EFI_UNICODE_COLLATION_PROTOCOL.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
We have moved generating html documentation with Sphinx.
%s/Return Value/Return/g
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
The system table must be passed as a pointer in the loaded image
protocol.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
In int-ll64.h, we always use the following typedefs:
typedef unsigned int u32;
typedef unsigned long uintptr_t;
typedef unsigned long long u64;
This does not need to match to the compiler's <inttypes.h>.
Do not include it.
The use of PRI* makes the code super-ugly. You can simply use
"l" for printing uintptr_t, "ll" for u64, and no modifier for u32.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The crc32 of the runtime services table must be updated after detaching.
efi_update_table_header_crc32() must be __efi_runtime. So move it to
efi_runtime.c
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
The commit 21b3edfc96 ("efi_loader: check parameters of CreateEvent")
enforces a strict parameter check at CreateEvent(). On the other hand,
UEFI specification version 2.7, section 7.1, says:
The EVT_NOTIFY_WAIT and EVT_NOTIFY_SIGNAL flags are exclusive. If
neither flag is specified, the caller does not require any notification
concerning the event and the NotifyTpl, NotifyFunction, and
NotifyContext parameters are ignored.
So the check should be mitigated so as to comply with the specification.
Without this patch, EDK2's Shell.efi won't be started.
Fixes: 21b3edfc96 ("efi_loader: check parameters of CreateEvent")
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
If the number of installed tables is changed in
InstallConfigurationTable() update the crc32 of the system table.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
Use const for the buffer. We are not changing the buffer.
Use efi_uintn_t where prescribed by the UEFI spec.
Prefer u32 over uint32_t.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
The system table contains a link to the list of configurations tables.
These include the device tree, SMBIOS table, and the ACPI table.
This array is currently statically linked. With the patch it is allocated
as EFI_RUNTIME_SERVICES_DATA. Due to the structure of the system table we
cannot work with a linked list here.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
