Age | Commit message (Collapse) | Author |
|
lib/binman.c references the following 3 ofnode APIs:
ofnode_first_subnode(), ofnode_path() and ofnode_read_bool().
These APIs get built only when DM is on. Fix the dependency then.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
VenHw and VenMedia device path nodes may carry vendor defined data of
arbitrary length. When converting a device path node to text ensure that we
do not overrun our internal buffer.
In our implementation of
EFI_DEVICE_PATH_TO_TEXT_PROTOCOL.ConvertDevicePathToText() we could first
determine the output length and then allocate buffers but that would nearly
double the code size. Therefore keep the preallocated buffers and truncate
excessive device paths instead.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
The UEFI spec does not allow ACPI tables to be in runtime services memory.
It recommends EfiACPIReclaimMemory.
Remove a superfluous check that the allocated pages are 16 byte aligned.
EFI pages are 4 KiB aligned.
Fixes: 86df34d42b05 ("efi_loader: Install ACPI configuration tables")
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
Include "compiler.h" in sha512.c. This is needed by 'cpu_to_be64' macro
that used in 'sha512_base_do_finalize' function.
Signed-off-by: Siew Chin Lim <elly.siew.chin.lim@intel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
Currently we fail silently if there is an algorithm mismatch. To help
distinguish this failure condition.
Signed-off-by: Sean Anderson <sean.anderson@seco.com>
|
|
Our current implementation of
EFI_DEVICE_PATH_TO_TEXT_PROTOCOL.ConvertDevicePathToText() truncates multi
part device paths after the first part. We should convert all parts.
Render device path instance ends as commas. This is not explicitly
described in the UEFI spec but mimics what EDK II does.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
|
|
Test EFI_DEVICE_PATH_TO_TEXT_PROTOCOL.ConvertDevicePathToText() for a multi
part device path.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
|
|
Fix a typo. Apply the correct compiler flags to dtbdump.o.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
The last block is of size media->block_size
Signed-off-by: Jesper Schmitz Mouridsen <jesper@schmitz.computer>
Simplify expression.
Apply same change to efi_disk_write_blocks().
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
- Merge the patch to take <asm/global_data.h> out of <common.h>
|
|
On Rockchip platforms we need this area of code in TPL, but there is no
TPL_SEPARATE_BSS symbol.
This reverts commit 0a2aaab0b678fd1778ff2fc59d0770fc82995532.
Reported-by: Markus Reichl <m.reichl@fivetechno.de>
Reported-by: Jesper Schmitz Mouridsen <jesper@schmitz.computer>
Reported-by: Peter Robinson <pbrobinson@gmail.com>
Tested-by: Peter Robinson <pbrobinson@gmail.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
|
|
fix get_last_capsule() leads to writes beyond the stack allocated buffer.
This was indicated when enabling the stack protector.
utf16_utf8_strcpy() only stops copying when reaching '\0'. The current
invocation always writes beyond the end of value[].
The output length of utf16_utf8_strcpy() may be longer than the number of
UTF-16 tokens. E.g has "CapsuleКиев" has 11 UTF-16 tokens but 15 UTF-8
tokens. Hence, using utf16_utf8_strcpy() without checking the input may
lead to further writes beyond value[].
The current invocation of strict_strtoul() reads beyond the end of value[].
A non-hexadecimal value after "Capsule" (e.g. "CapsuleZZZZ") must result in
an error. We cat catch this by checking the return value of strict_strtoul().
A value that is too short after "Capsule" (e.g. "Capsule0") must result in
an error. We must check the string length of value[].
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
'.' and '..' are directories. So when looking for capsule files it is
sufficient to check that the attribute EFI_FILE_DIRECTORY is not set. We
don't have to check for these special names.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
PE section table entries' SizeOfRawData must be a multiple of
FileAlignment, and thus may be rounded up and larger than their
VirtualSize.
We should not load beyond the VirtualSize, which is "the total size of
the section when loaded into memory" -- we may clobber real data at the
target in some other section, since we load sections in reverse order
and sections are usually laid out sequentially.
Signed-off-by: Asherah Connor <ashe@kivikakk.ee>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
While static configuration is useful it cannot cover every case. Sometimes
board revisions are encoded in resistor straps and must be read at
runtime.
The easiest way to provide this information is via sysinfo, since the
board can then provide a driver to read whatever is needed.
Add some standard sysinfo options for this, and use them to obtain the
required information.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
|
|
At present the version string is obtained from PLAIN_VERSION. Some boards
may want to configure this using the device tree, since the build system
can more easily insert things there after U-Boot itself is built. Add this
option to the code.
Also in some cases the version needs to be generated programmatically,
such as when it is stored elsewhere in the ROM and must be read first.
To handle this, keep a pointer around so that it can be updated later.
This works by storing the last string in the context, since it is easier
than passing out a little-used extra parameter.
Provide a function to update the version string.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
|
|
Add a new member to the context struct which tracks the end of the string
table. This allows us to avoid recalculating this at the end.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
|
|
We can store this in the context and avoid passing it to each function.
This makes it easier to follow and will also allow keeping track of the
end of the string table (in future patches).
Add an 'eos' field to the context and create a function to set it up.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
|
|
At present we pass the ofnode to each function. We also pass the 'eos'
pointer for adding new strings. We don't track the current end of the
string table, so have smbios_string_table_len() to find that.
The code can be made more efficient if it keeps information in a
context struct. This also makes it easier to add more features.
As a first step, switch the ofnode parameter to be a context pointer.
Update smbios_add_prop() at the same time to avoid changing the same
lines of code in consecutive patches.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
|
|
We may as well include the U-Boot release information in the type-0 table
since it is designed for that purpose.
U-Boot uses release versions based on the year and month. The year cannot
fit in a byte, so drop the century.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
|
|
This type is not used outside the smbios.c file so there is no need for it
to be in the header file. Move it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
|
|
https://gitlab.denx.de/u-boot/custodians/u-boot-efi
Pull request for UEFI sub-system for efi-2021-04-rc2
Bug fixes:
* do not allow creating of files with filenames on FAT file system
* install UEFI System Partition GUID on ESP handle
* in dtbdump.efi test tool use GUID to find ESP handle
Documentation:
* man-page for load command
* describe end of life of plat_auto
|
|
The Load File2 protocol exposes a device path with a VenMedia() node. Hence
our implementation of the device path to text protocol should support this
node.
Cc: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
If dtbdump.efi is loaded from memory when calling LoadImage the loaded
image protocol will not indicate the partition from where it was loaded.
In this case use the EFI system partition for the 'load' and 'save'
commands.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
On the handle for the UEFI System Partition we must install the System
Partition GUID (with a NULL interface).
Let the efidebug command display the GUID.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
In the implementation of the EFI_DT_FIXUP_PROTOCOL:
* Only check the buffer size when EFI_DT_APPLY_FIXUPS is set.
* In this case the field totalsize of the device-tree may not exceed the
buffer size.
* Install device-tree only if EFI_DT_INSTALL_TABLE is set.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
At present prior_stage_fdt_address is declared as phys_addr_t. On
a 32-bit platform where phys_addr_t can be 64-bit, assigning its
value to gd->fdt_blob which is a pointer, can cause warnings.
Cast it to uintptr_t before the assignment.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
Move this out of the common header and include it only where needed. In
a number of cases this requires adding "struct udevice;" to avoid adding
another large header or in other cases replacing / adding missing header
files that had been pulled in, very indirectly. Finally, we have a few
cases where we did not need to include <asm/global_data.h> at all, so
remove that include.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
|
|
https://gitlab.denx.de/u-boot/custodians/u-boot-dm
tpm fixes for coral
binman fixes support for symbols in sub-sections
support for additional cros_ec commands
various minor fixes / tweaks
|
|
Add some debugging to indicate which node the binman library is looking
at.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
Some ARMv7 boards using PSCI require to be in secure-mode when booted via
'bootz' or 'bootm'. During distro-boot 'bootefi bootmgr' is called to check
if booting via UEFI is possible.
With the change we change the switch from secure mode to non-secure mode is
moved from the UEFI subsystem setup to just before calling StartImage().
Cc: Jernej Škrabec <jernej.skrabec@gmail.com>
Reported by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
If the field Media->IoAlign of the EFI block IO protocol is zero, no
alignment is required. Our code required 4 GiB alignment in this case.
Don't check buffer alignment if Media->IoAlign == 0.
Fixes: f59f0825e8b9 ("efi_loader: parameter checks BLOCK_IO_PROTOCOL")
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
The field Media->LastBlock of the EFI_BLOCK_IO_PROTOCOL must be filled
with the index of the last logical block (LBA) for the block device that
our test driver provides.
After calling ConnectController() U-Boot exposes the block IO protocol for
the partition check that the value of Media->LastBlock equals the partition
size minus one.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
Some images may have multiple copies of the same thing, e.g. two versions
of the read/write U-Boots. It is necessary to read data from one or other
of these under selection of the verified-boot logic. Add a function to
select the subnode to use.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
Move this logic out of the main init function so it is available for
other purpose.
Use a different error when multiple-images is in use but no subnode is
available. This makes it easier to determine what is wrong.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
Unfortunately the toolchain often brings in the crc32 table even if the
function is not actually used. For now, exclude it from the TPL build,
which is very sensitive to size.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
At present this code uses a simple printf() format if running in SPL. But
SPL can use the full printf. Use USE_TINY_PRINTF instead.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
This setting may be different in SPL and TPL. Update the code to check
the correct setting.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
This includes information about sandbox which is not relevant for most
boards. Drop it.
Also add the address to help figure out the problem.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
In the function rsa_verify_hash, if the "main" key doesn't
work, u-boot try others keys. But it searches those keys
in the FIT image instead of the u-boot device tree.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
The field Media.LastBlock of the EFI_BLOCK_IO_PROTOCOL.Media was filled
incorrectly both for block devices as well as for partitions.
The field must be filled with the index of the last logical block (LBA):
* block devices: device size minus 1
* partitions: partition size minus 1
Reported-by: Alexander von Gluck IV <kallisti5@unixzen.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
The UEFI specification requires event notification levels to be
> TPL_APPLICATION and <= TPL_HIGH_LEVEL.
With the patch the CreateEvent() and CreatedEventEx() services are changed
to check that the notification level is not TPL_APPLICATION.
Reference:
UEFI Specification 2.8 Errata B, table 27 "TPL Restrictions",
"Event Notification Levels"
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
CID 316361 says:
> Passing argument "count * 8UL /* sizeof (*files) */" to
> function "dlmalloc" and then casting the return value to "u16 **" is
> suspicious. In this particular case "sizeof (u16 **)" happens to be equal
> to "sizeof (u16 *)", but this is not a portable assumption.
> > 767 tmp_files = malloc(count * sizeof(*files));
The change in this patch fixes the ambiguity.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Fixes: CID 316361
For readability use *tmp_files instead of tmp_files[0]
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
Use the correct GUID when closing the file load protocol.
Remove an unnecessary check (Coverity CID 316363).
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
https://gitlab.denx.de/u-boot/custodians/u-boot-efi
Pull request for UEFI sub-system for efi-2021-04-rc1-2
* Provide a test tool for initial RAM disk provided via load file2 protocol.
* Make more items configurable to reduce code size:
* Boot manager
* EFI_DT_FIXUP_PROTOCOL
* EFI_DEVICE_PATH_UTILITIES_PROTOCOL
* Bug fixes
* avoid EFI runtime symbols in global symbol table
|
|
Atish reports that on RISC-V, accessing the EFI variables causes
a kernel panic. An objdump of the file verifies that, since the
global pointer for efi_var_buf ends up in .GOT section which is
not mapped in virtual address space for Linux.
<snip of efi_var_mem_find>
0000000000000084 <efi_var_mem_find>:
84: 715d addi sp,sp,-80
* objdump -dr
0000000000000086 <.LCFI2>:
86: e0a2 sd s0,64(sp)
88: fc26 sd s1,56(sp)
8a: e486 sd ra,72(sp)
8c: f84a sd s2,48(sp)
8e: f44e sd s3,40(sp)
90: f052 sd s4,32(sp)
92: ec56 sd s5,24(sp)
94: 00000497 auipc s1,0x0
94: R_RISCV_GOT_HI20 efi_var_buf
98: 0004b483 ld s1,0(s1) # 94 <.LCFI2+0xe>
98: R_RISCV_PCREL_LO12_I .L0
98: R_RISCV_RELAX *ABS*
* objdump -t
0000000000000084 g F .text.efi_runtime 00000000000000b8 efi_var_mem_find
With the patch applied:
* objdump -dr
0000000000000086 <.LCFI2>:
86: e0a2 sd s0,64(sp)
88: fc26 sd s1,56(sp)
8a: e486 sd ra,72(sp)
8c: f84a sd s2,48(sp)
8e: f44e sd s3,40(sp)
90: f052 sd s4,32(sp)
92: ec56 sd s5,24(sp)
94: 00000497 auipc s1,0x0
94: R_RISCV_PCREL_HI20 .LANCHOR0
94: R_RISCV_RELAX *ABS*
98: 00048493 mv s1,s1
98: R_RISCV_PCREL_LO12_I .L0
98: R_RISCV_RELAX *ABS*
* objdump -t
0000000000000008 l O .data.efi_runtime 0000000000000008 efi_var_buf
On arm64 this works, because there's no .GOT entries for this
and everything is converted to relative references.
* objdump -dr (identical pre-post patch, only the new function shows up)
00000000000000b4 <efi_var_mem_find>:
b4: aa0003ee mov x14, x0
b8: 9000000a adrp x10, 0 <efi_var_mem_compare>
b8: R_AARCH64_ADR_PREL_PG_HI21 .data.efi_runtime
bc: 91000140 add x0, x10, #0x0
bc: R_AARCH64_ADD_ABS_LO12_NC .data.efi_runtime
c0: aa0103ed mov x13, x1
c4: 79400021 ldrh w1, [x1]
c8: aa0203eb mov x11, x2
cc: f9400400 ldr x0, [x0, #8]
d0: b940100c ldr w12, [x0, #16]
d4: 8b0c000c add x12, x0, x12
So let's switch efi_var_buf to static and create a helper function for
anyone that needs to update it.
Fixes: e01aed47d6a0 ("efi_loader: Enable run-time variable support for tee based variables")
Reported-by: Atish Patra <atishp@atishpatra.org>
Tested-by: Atish Patra <atish.patra@wdc.com>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
EFI applications must use CR LF as line endings.
Provide a print() function to reduce code size.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
Provide an UEFI application to save the initial RAM disk provided by U-Boot
via the Load File2 protocol.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
|
|
If we are using ACPI tables instead of a device tree, we don't need the
dtbdump.efi test tool.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|
|
Before overwriting an existing file ask the user.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
|