| Age | Commit message (Collapse) | Author |
|---|
|
The UEFI spec requires that the memory map key is checked in
ExitBootServices().
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Check the parameters of boottime service GetMemoryMap().
Return EFI_INVALID_PARAMETER where required by the UEFI spec.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
If no pointer is provided throw an error.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Rigorously check the TPL level and the event type.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
We currently expose host addresses in the EFI memory map. That can be
bad if we ever want to use sandbox to boot strap a real kernel, because
then the kernel would fetch its memory table from our host virtual address
map. But to make that use case work, we would need to have full control
over the address space the EFI application sees.
So let's expose only U-Boot addresses to the guest until we get to the
point of allocation. EFI's allocation functions are fun - they can take
U-Boot addresses as input values for hints and return host addresses as
allocation results through the same uint64_t * parameter. So we need to
be extra careful on what to pass in when.
With this patch I am successfully able to run the efi selftest suite as
well as grub.efi on aarch64.
Signed-off-by: Alexander Graf <[email protected]>
|
|
With sandbox the U-Boot code is not mapped into the sandbox memory range
so does not need to be excluded when allocating EFI memory. Update the EFI
memory init code to take account of that.
Signed-off-by: Simon Glass <[email protected]>
[agraf: Remove map_sysmem() call and header reference]
Signed-off-by: Alexander Graf <[email protected]>
|
|
Now that elf.h contains relocation defines for all architectures
we care about, let's just include it unconditionally and refer to
the defines.
Signed-off-by: Alexander Graf <[email protected]>
|
|
Thanks to CONFIG_SANDBOX, we can not rely on config options to tell us
what CPU architecture we're running on.
The compiler however does know that, so let's just move the ifdefs over
to compiler based defines rather than kconfig based options.
Signed-off-by: Alexander Graf <[email protected]>
|
|
Varargs differ between sysv and ms abi. On x86_64 we have to follow the ms
abi though, so we also need to make sure we use x86_64 varargs helpers.
This patch introduces generic efi vararg helpers that adhere to the
respective EFI ABI. That way we can deal with them properly from efi
loader code and properly interpret variable arguments.
This fixes the InstallMultipleProtocolInterfaces tests in the efi selftests
on x86_64 for me.
Signed-off-by: Alexander Graf <[email protected]>
|
|
When running on the sandbox the stack is not necessarily at a higher memory
address than the highest free memory.
There is no reason why the checking of the highest memory address should be
more restrictive for EFI_ALLOCATE_ANY_PAGES than for
EFI_ALLOCATE_MAX_ADDRESS.
Signed-off-by: Heinrich Schuchardt <[email protected]>
[agraf: use -1ULL instead]
Signed-off-by: Alexander Graf <[email protected]>
|
|
We try hard to make sure that SMBIOS tables live in the lower 32bit.
However, when we can not find any space at all there, we should not
error out but instead just fall back to map them in the full address
space instead.
This can for example happen on systems that do not have any RAM mapped
in the lower 32bits of address space. In that case having any SMBIOS
tables at all is better than having none.
Signed-off-by: Alexander Graf <[email protected]>
|
|
The EFI image loader tries to determine which target architecture we're
working with to only load PE binaries that match.
So far this has worked based on CONFIG defines, because the target CPU
was always indicated by a config define. With sandbox however, this is
not longer true as all sandbox targets only encompass a single CONFIG
option and so we need to use compiler defines to determine the CPU
architecture.
Signed-off-by: Alexander Graf <[email protected]>
|
|
This reverts commit c524997acb3d322e1bbd36c06ad02ef589705e7c.
Booting ARMv7 in non-secure mode using bootefi works now.
Signed-off-by: Mark Kettenis <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
The Linux kernel moved to sphinx-based documentation and got rid of the
DocBook based documentation quite a while ago. Hence, the DocBook
documentation for U-Boot should be converted as well.
To achieve this, import the necessary files from Linux v4.17, and
convert the current DocBook documentation (three files altogether) to
sphinx/reStructuredText.
For now, all old DocBook documentation was merged into a single
handbook, tentatively named "U-Boot Hacker Manual".
For some source files, the documentation style was changed to comply
with kernel-doc; no functional changes were applied.
Signed-off-by: Mario Six <[email protected]>
|
|
Output ACPI configuration table if it exists.
Signed-off-by: Bin Meng <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
ACPI tables can be passed via EFI configuration table to an EFI
application. This is only supported on x86 so far.
Signed-off-by: Bin Meng <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
At present the number of configuration tables is set to 2. By
looking at which tables the Linux EFI stub or iPXE can process,
it looks 16 is a reasonable number.
Signed-off-by: Bin Meng <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
We store pixels as BGRA in memory, as can be seen from struct efi_gop_pixel.
So we need to expose the same format to UEFI payloads to actually have them
use the correct colors.
Reported-by: Fabian Vogt <[email protected]>
Tested-by: Peter Robinson <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
These constants are defined in arch-specific code but redefined here. Add
a TODO to clean this up.
Signed-off-by: Simon Glass <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
The PE standard allows for HI20/LOW12 relocations. Within the efi_loader
target we always know that our relocation target is 4k aligned, so we
don't need to worry about the LOW12 part.
This patch adds support for the respective relocations. With this and a
few grub patches I have cooking in parallel I'm able to run grub on RISC-V.
Signed-off-by: Alexander Graf <[email protected]>
|
|
The code to determine rows / cols on the screen could potentially run
into a case where it doesn't know how big the screen is. In that case,
assume 80x25.
Signed-off-by: Alexander Graf <[email protected]>
|
|
Handles are not used at runtime. They are freed by the firmware when the
last protocol interface is uninstalled. So there is no reason to use EFI
memory when creating handles.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
efi_mem_carve_out() is used to remove memory pages from a mapping.
As the number of pages to be removed is a 64bit type the return type
should be 64bit too.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Do not use anonymous constants when calling efi_allocage_pages.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
With 'make htmldocs' we can generate a documentation if the function
comments follow the DocBook conventions.
This patch adjusts the comments for EFI boot services and provides the
DocBook template for the EFI subsystem.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Clarify the operation of this code with some additional comments.
Signed-off-by: Simon Glass <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
The ReinstallProtocolInterface boot time service is implemented.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
The interface has to be checked in UninstallProtocolInterface.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
efi_get_variable() always stores an extra zero byte after the output data.
When the returned data size matches the output buffer size, the extra zero
byte is stored past the end of the output buffer.
Signed-off-by: Ivan Gorinov <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
If a request for the console size would be answered with a response
with less then three values, uninitialized stack memory would be
copied to the number of rows and columns of the terminal.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
The definitons of the variable services are adjusted:
- use efi_uintn_t instead of unsigned long
- use u16 * instead of s16 * for Unicode strings
- correct definition of QueryVariableInfo
- rename efi_get_next_variable to efi_get_next_variable_name
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
If we cannot determine the size of the serial terminal we still have
to check the parameters of efi_cout_query_mode().
Querying the size of the serial terminal drains the keyboard buffer.
So make sure we do this during the initialization and not in the midst
of an EFI application.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
The current coding advances the column by the number of UTF-8 bytes.
The column should be increased by one per unicode character.
The UEFI spec provides advance rules for U+0000, U+0008, U+000A,
and U000D. All other characters, including control characters
U+0007 (bel) and U+0009 (tab), have to increase the column by one.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
asm/global_data.h is already included via common.h.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
EFI applications like GRUB issue ANSI sequences for setting colors
and for positioning. So if the EFI subsystem is enabled, we should
also enable ANSI control sequences by default.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Instead of difference between preferred and actual image base, the
actual base is added to the fields specified in the .reloc section.
Use ImageBase from PE optional header to compute the delta,
exit early if the image is loaded at the preferred address.
Signed-off-by: Ivan Gorinov <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
We have almost all pieces needed to support RISC-V UEFI binaries in place already.
The only missing piece are ELF relocations for runtime code and
data.
This patch adds respective support in the linker script and the runtime
relocation code. It also allows users to enable the EFI_LOADER configuration
switch on RISC-V platforms.
Signed-off-by: Alexander Graf <[email protected]>
|
|
We were using our EFI_CACHELINE_SIZE define only in the runtime service
code, but left the image loader to use plain CONFIG_SYS_CACHELINE_SIZE.
This patch moves EFI_CACHELINE_SIZE into efi_loader.h and converts
the image loader to use it.
Signed-off-by: Alexander Graf <[email protected]>
|
|
As part of the main conversion a few files were missed. These files had
additional whitespace after the '*' and before the SPDX tag and my
previous regex was too strict. This time I did a grep for all SPDX tags
and then filtered out anything that matched the correct styles.
Fixes: 83d290c56fab ("SPDX: Convert all of our single license tags to Linux Kernel style")
Reported-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Tom Rini <[email protected]>
|
|
When U-Boot started using SPDX tags we were among the early adopters and
there weren't a lot of other examples to borrow from. So we picked the
area of the file that usually had a full license text and replaced it
with an appropriate SPDX-License-Identifier: entry. Since then, the
Linux Kernel has adopted SPDX tags and they place it as the very first
line in a file (except where shebangs are used, then it's second line)
and with slightly different comment styles than us.
In part due to community overlap, in part due to better tag visibility
and in part for other minor reasons, switch over to that style.
This commit changes all instances where we have a single declared
license in the tag as both the before and after are identical in tag
contents. There's also a few places where I found we did not have a tag
and have introduced one.
Signed-off-by: Tom Rini <[email protected]>
|
|
We have a large number of places where while we historically referenced
gd in the code we no longer do, as well as cases where the code added
that line "just in case" during development and never dropped it.
Signed-off-by: Tom Rini <[email protected]>
|
|
The missing services of the EFI_DEVICE_PATH_UTILITIES_PROTOCOL are
implemented.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Device paths may consist of multiple instances. Up to now we have only
considered the size of the first instance. For the services of the
EFI_DEVICE_PATH_UTILITIES_PROTOCOL in most cases the total length of the
device path is relevant.
So let's rename efi_dp_size() to efi_dp_instance_size() and create a new
function efi_dp_size() that calculates the total device path length.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
efi_dp_size() is meant to return the device path length without the end
node.
The length of a device path containing only an end node was incorrectly
reported as 4.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
The logic of the AppendDevicePath service of the
EFI_DEVICE_PATH_UTILITIES_PROTOCOL is incorrectly implemented:
* if both paths are NULL an end node has to be returned
* if both paths are not NULL the end node of the second device path has to
be kept
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Implement the CreateDeviceNode service of the device path utility protocol.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
We do not support bootefi booting ARMv7 in non-secure mode.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
This reverts the parts of commit 3b3ea2c56ec4bc5588281fd103c744e608f8b25c
where it changed the EFI dependency on NET.
Signed-off-by: Joe Hershberger <[email protected]>
Reviewed-by: Duncan Hare <[email protected]>
|
|
Check FileHeader.Machine to make sure the EFI executable image is built
for the same architecture. For example, 32-bit U-Boot on x86 will print
an error message instead of loading an x86_64 image and crashing.
Signed-off-by: Ivan Gorinov <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
|
Add missing network initialization code.
Before the patch the network was only usable if a network command like
dhcp or tftp had beed executed.
This was visible when interrupting the console countdown and executing
bootefi selftest for vexpress_ca15_tc2_defconfig.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Alexander Graf <[email protected]>
|
