| Age | Commit message (Collapse) | Author |
|---|
|
Now we can build efi_loader with block device support (CONFIG_BLK) and
without CONFIG_PARTITIONS.
So change Makefile.
Signed-off-by: AKASHI Takahiro <[email protected]>
|
|
In the current implementation, partition table support (either GPT or DOS)
is not mandatory. So CONFIG_PARTITION_UUIDS should not be enabled
(selected) unconditionally.
Fixes: commit 17f8cda505e3 ("efi_loader: set partition GUID in device path for SIG_TYPE_GUID")
Signed-off-by: AKASHI Takahiro <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
Use u16_strsize().
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
Use u16_strsize().
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
The capsule update code has been modified for getting the image GUID
values from the platform code. With this, each image now has a unique
GUID value. With this change, there is no longer a need for defining
GUIDs for FIT and raw images. Remove these GUID values.
Signed-off-by: Sughosh Ganu <[email protected]>
|
|
Currently, there are a bunch of boards which enable the UEFI capsule
update feature. The actual update of the firmware images is done
through the dfu framework which uses the dfu_alt_info environment
variable for getting information on the update, like device, partition
number/address etc. The dfu framework allows the variable to be set
through the set_dfu_alt_info function defined by the platform, or if
the function is not defined, it gets the variable from the
environment. Using the value set in the environment is not very
robust, since the variable can be modified from the u-boot command
line and this can cause an incorrect update.
To prevent this from happening, define the set_dfu_alt_info function
when the capsule update feature is enabled. A weak function is defined
which sets the dfu_alt_info environment variable by getting the string
for the variable from the platform.
Signed-off-by: Sughosh Ganu <[email protected]>
Reviewed-by: Masami Hiramatsu <[email protected]>
|
|
The current capsule update code compares the image GUID value in the
capsule header with the image GUID value obtained from the
GetImageInfo function of the Firmware Management Protocol(FMP). This
comparison is done to ascertain if the FMP's SetImage function can be
called for the update. Make this checking more robust by comparing the
image_index value passed through the capsule with that returned by the
FMP's GetImageInfo function. This protects against the scenario of the
firmware being updated in a wrong partition/location on the storage
device if an incorrect value has been passed through the capsule,
since the image_index is used to determine the location of the update
on the storage device.
Signed-off-by: Sughosh Ganu <[email protected]>
Reviewed-by: Masami Hiramatsu <[email protected]>
Reviewed-by: Ilias Apalodimas <[email protected]>
|
|
Currently, the image descriptor array that has been passed to the
GetImageInfo function of the Firmware Management Protocol(FMP) gets
populated through the data stored with the dfu framework. The
dfu data is not restricted to contain information only of the images
updatable through the capsule update mechanism, but it also contains
information on other images. The image descriptor array is also parsed
by the ESRT generation code, and thus the ESRT table contains entries
for other images that are not being handled by the FMP for the capsule
updates. Fix this by populating the image descriptor array from the
structure initialised in the board file.
The other issue fixed is assignment of a separate GUID for all images
in the image descriptor array. The UEFI specification mandates that
all entries in the ESRT table should have a unique GUID value as part
of the FwClass member of the EFI_SYSTEM_RESOURCE_ENTRY. Currently, all
images are assigned a single GUID value, either an FIT GUID or a raw
image GUID. This is fixed by obtaining the GUID values from the
efi_fw_images array defined per platform.
Signed-off-by: Sughosh Ganu <[email protected]>
Reviewed-by: Masami Hiramatsu <[email protected]>
|
|
Use sysreset_walk_halt() directly from reset-after-capsule-on-disk
feature to reboot (cold reset) machine instead of using reset command
interface, since this is not a command.
Note that this will make CONFIG_EFI_CAPSULE_ON_DISK depending on
the CONFIG_SYSRESET.
Signed-off-by: Masami Hiramatsu <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
EFI_CAPSULE_ON_DISK=n
Make efi_load_capsule_drivers() available even if EFI_CAPSULE_ON_DISK
is disabled because the caller (efi_init_obj_list()) expects it only
relays on EFI_HAVE_CAPSULE_SUPPORT.
Suggested-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
|
|
Since be66b89da306 ("efi_loader: configuration of variables store")
the choice of EFI_VARIABLE_FILE_STORE or EFI_MM_COMM_TEE
is mutually-exclusive, however efi_var_to_file also allows
for "neither". Provide an additional Kconfig option.
Signed-off-by: Tom Saeger <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
Signed-off-by: Tom Rini <[email protected]>
|
|
It seems a copy'n'paste typo when tool had been introduced.
It has never had the 'exit' suffix in the file name. Hence,
the custom CFLAGS never been applied and, for example, BFD
linker complains:
LD lib/efi_loader/initrddump_efi.so
ld.bfd: lib/efi_loader/initrddump.o: warning: relocation in read-only section `.text.efi_main'
ld.bfd: warning: creating DT_TEXTREL in a shared object
Remove wrong 'exit' suffix from the custom CFLAGS variable.
Fixes: 65ab48d69ddb ("efi_selftest: provide initrddump test tool")
Fixes: 9c045a49a9c9 ("efi_loader: move dtbdump.c, initrddump.c to lib/efi_loader")
Signed-off-by: Andy Shevchenko <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
Drop the Kconfigs which are not used and all references to them. In
particular, this drops CONFIG_VIDEO to avoid confusion and allow us to
eventually rename CONFIG_DM_VIDEO to CONFIG_VIDEO.
Also drop the prototype for video_get_info_str() which is no-longer used.
Signed-off-by: Simon Glass <[email protected]>
Acked-by: Jason Liu <[email protected]>
|
|
Up to now the initrddump.efi application has drained the input after
showing the prompt. This works for humans but leads to problems when
automating testing. If the input is drained, this should be done before
showing the prompt.
Signed-off-by: Heinrich Schuchardt <[email protected]>
Acked-by: Ilias Apalodimas <[email protected]>
|
|
initrddump.efi uses colored output and clear the screen. This is not
helpful for integration into Python tests. Allow specifying 'nocolor' in
the load option data to suppress color output and clearing the screen.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
%s/devie-path/device-path/
Reported-by: Ilias Apalodimas <[email protected]>
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
The tools dtbdump.efi and initrddump.efi are useful for Python testing even
if CONFIG_EFI_SELFTEST=n.
Don't clear the screen as it is incompatible with Python testing.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
The block IO protocol may be installed on any handle. We should make
no assumption about the structure the handle points to.
efi_disk_is_system_part() makes an illegal widening cast from a handle
to a struct efi_disk_obj. Remove the function.
Fixes: Fixes: 41fd506842c2 ("efi_loader: disk: add efi_disk_is_system_part()")
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
The boot manager must support loading from boot options using a short-form
device-path, e.g. one where the first element is a hard drive media path.
See '3.1.2 Load Options Processing' in UEFI specification version 2.9.
Fixes: 0e074d12393b ("efi_loader: carve out efi_load_image_from_file()")
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
Let function efi_dp_find_obj() additionally check if a given protocol is
installed on the handle relating to the device-path.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
efi_dp_find_obj() should not return any handle with a partially matching
device path but the handle with the maximum matching device path.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
Rename function shorten_path() to efi_dp_shorten() and export it.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
Starting UEFI Spec 2.8 we must fill in the variable attributes when
GetVariable() returns EFI_BUFFER_TOO_SMALL and Attributes is non-NULL.
This code was written with 2.7 in mind so let's move the code around a
bit and fill in the attributes EFI_BUFFER_TOO_SMALL is returned
Signed-off-by: Ilias Apalodimas <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
InstallProtocolInterface() is called with a pointer to the protocol GUID.
There is not guarantee that the memory used by the caller for the protocol
GUID stays allocated. To play it safe the GUID should be copied to U-Boot's
internal structures.
Reported-by: Joerie de Gram <[email protected]>
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
UEFI specification requires that 5 minutes watchdog timer is
armed before the firmware's boot manager invokes an EFI boot option.
This watchdog timer is updated as follows, according to the
UEFI specification.
1) The EFI Image may reset or disable the watchdog timer as needed.
2) If control is returned to the firmware's boot manager,
the watchdog timer must be disabled.
3) On successful completion of EFI_BOOT_SERVICES.ExitBootServices()
the watchdog timer is disabled.
1) is up to the EFI image, and 3) is already implemented in U-Boot.
This patch implements 2), the watchdog is disabled when control is
returned to U-Boot.
In addition, current implementation arms the EFI watchdog at only
the first "bootefi" invocation. The EFI watchdog must be armed
in every EFI boot option invocation.
Signed-off-by: Masahisa Kojima <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
Add a cold reset soon after processing capsule update on disk.
This is required in UEFI specification 2.9 Section 8.5.5
"Delivery of Capsules via file on Mass Storage device" as;
In all cases that a capsule is identified for processing the system is
restarted after capsule processing is completed.
This also reports the result of each capsule update so that the user can
notice that the capsule update has been succeeded or not from console log.
Signed-off-by: Masami Hiramatsu <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
Since the efi_update_capsule() represents the UpdateCapsule() runtime
service, it has to handle the capsule flags and update ESRT. However
the capsule-on-disk doesn't need to care about such things.
Thus, the capsule-on-disk should use the efi_capsule_update_firmware()
directly instead of calling efi_update_capsule().
This means the roles of the efi_update_capsule() and capsule-on-disk
are different. We have to keep the efi_update_capsule() for providing
runtime service API at boot time.
Suggested-by: AKASHI Takahiro <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
The general rule of accepting or rejecting an image is
1. Is the sha256 of the image in dbx
2. Is the image signed with a certificate that's found in db and
not in dbx
3. The image carries a cert which is signed by a cert in db (and
not in dbx) and the image can be verified against the former
4. Is the sha256 of the image in db
For example SHIM is signed by "CN=Microsoft Windows UEFI Driver Publisher",
which is issued by "CN=Microsoft Corporation UEFI CA 2011", which in it's
turn is issued by "CN=Microsoft Corporation Third Party Marketplace Root".
The latter is a self-signed CA certificate and with our current implementation
allows shim to execute if we insert it in db.
However it's the CA cert in the middle of the chain which usually ends up
in the system's db. pkcs7_verify_one() might or might not return the root
certificate for a given chain. But when verifying executables in UEFI, the
trust anchor can be in the middle of the chain, as long as that certificate
is present in db. Currently we only allow this check on self-signed
certificates, so let's remove that check and allow all certs to try a
match an entry in db.
Open questions:
- Does this break any aspect of variable authentication since
efi_signature_verify() is used on those as well?
Signed-off-by: Ilias Apalodimas <[email protected]>
|
|
The EFI spec allows for images to carry multiple signatures. Currently
we don't adhere to the verification process for such images.
The spec says:
"Multiple signatures are allowed to exist in the binary's certificate
table (as per PE/COFF Section "Attribute Certificate Table"). Only one
hash or signature is required to be present in db in order to pass
validation, so long as neither the SHA-256 hash of the binary nor any
present signature is reflected in dbx."
With our current implementation signing the image with two certificates
and inserting both of them in db and one of them dbx doesn't always reject
the image. The rejection depends on the order that the image was signed
and the order the certificates are read (and checked) in db.
While at it move the sha256 hash verification outside the signature
checking loop, since it only needs to run once per image and get simplify
the logic for authenticating an unsigned imahe using sha256 hashes.
Signed-off-by: Ilias Apalodimas <[email protected]>
|
|
When we have the block descriptor we can simply access the device. Drop
the unnecessary function call.
Signed-off-by: Simon Glass <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
Use this function rather than following the pointers, since it is there
for this purpose.
Add the uclass name to the debug call at the end of dp_fill() since it is
quite useful.
Signed-off-by: Simon Glass <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
When loading an EFI binary via the UART we assign a UART device path to it.
But we lack a handle with that device path.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
The UEFI specification concerning Uart() device path nodes has been
clarified:
Parity and stop bits can either both use keywords or both use
numbers but numbers and keywords should not be mixed.
Let's go for keywords as this is what EDK II does. For illegal
values fall back to numbers.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
For printing an unsigned value we should use %u and not %d.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
efi_uintnt_t is an unsigned type. We should avoid showing negative numbers.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
The printing of a file path node must properly handle:
* odd length of the device path node
* UTF-16 character only partially contained in device path node
* buffer overflow due to very long file path
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
At present we use wide characters for unicode but this is not necessary.
Change the code to use the 'u' literal instead. This helps to fix build
warnings for sandbox on rpi.
Signed-off-by: Simon Glass <[email protected]>
Suggested-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
We don't have to recalculate the image hash every time we check against a
new db/dbx entry. So let's add a flag forcing it to run once since we only
support sha256 hashes
Suggested-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Ilias Apalodimas <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
A mix of signatures and hashes in db doesn't always work as intended.
Currently if the digest algorithm is not explicitly set to sha256 we
stop walking the security database and reject the image.
That's problematic in case we find and try to check a signature before
inspecting the sha256 hash. If the image is unsigned we will reject it
even if the digest matches.
Since we no longer reject the image on unknown algorithms add an explicit
check and reject the image if any other hash algorithm apart from sha256
is detected on dbx.
Suggested-by: Heinrich Schuchardt <[email protected]>
Signed-off-by: Ilias Apalodimas <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
This adds support for new RISCV_EFI_BOOT_PROTOCOL to
communicate the boot hart ID to bootloader/kernel on RISC-V
UEFI platforms.
The specification of the protocol is hosted at:
https://github.com/riscv-non-isa/riscv-uefi
Signed-off-by: Sunil V L <[email protected]>
Reviewed-by: Heinrich Schuchardt <[email protected]>
|
|
Replace @return and @param.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
The current EFI implementation confuses pointers and addresses. Normally
we can get away with this but in the case of sandbox it causes failures.
Despite the fact that efi_allocate_pages() returns a u64, it is actually
a pointer, not an address. Add special handling to avoid a crash when
running 'bootefi hello'.
Signed-off-by: Simon Glass <[email protected]>
|
|
This must be passed a ulong, not a u64. Fix it to avoid LTO warnings on
sandbox.
Signed-off-by: Simon Glass <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-efi
Pull request for efi-2022-04-rc1-2
Documentation:
* describe printf() format codes
UEFI
* enable more algorithms for UEFI image verification, e.g. SHA256-RSA2048
General
* simplify printing short texts for GUIDs
* provide a unit test for printing GUIDs
|
|
Sphinx expects Return: and not @return to indicate a return value.
find . -name '*.c' -exec \
sed -i 's/^\(\s\)\*\(\s*\)@return\(\s\)/\1*\2Return:\3/' {} \;
find . -name '*.h' -exec \
sed -i 's/^\(\s\)\*\(\s*\)@return\(\s\)/\1*\2Return:\3/' {} \;
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
For printing GUIDs with macro EFI_ENTRY use %pUs instead of %pUl to provide
readable debug output.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
If variable dfu_alt_info is not defined duplicate messages are displayed.
=> efidebug boot dump
Scanning disk mmc2.blk...
Scanning disk mmc1.blk...
Scanning disk mmc0.blk...
Found 3 disks
No EFI system partition
"dfu_alt_info" env variable not defined!
Probably dfu_alt_info not defined
"dfu_alt_info" env variable not defined!
Probably dfu_alt_info not defined
Remove the 'Probably dfu_alt_info not defined' message.
Instead write a warning if the variable contains no entities.
Signed-off-by: Heinrich Schuchardt <[email protected]>
|
|
U-Boot, in some occasions, injects a 'kaslr-seed' property on the /chosen
node. That would be problematic in case we want to measure the DTB we
install in the configuration table, since it would change across reboots.
The Linux kernel EFI-stub completely ignores it and only relies on
EFI_RNG_PROTOCOL for it's own randomness needs (i.e the randomization
of the physical placement of the kernel). In fact it (blindly) overwrites
the existing seed if the protocol is installed. However it still uses it
for randomizing it's virtual placement.
So let's get rid of it in the presence of the RNG protocol.
It's worth noting that TPMs also provide an RNG. So if we tweak our
EFI_RNG_PROTOCOL slightly and install the protocol when a TPM device
is present the 'kaslr-seed' property will always be removed, allowing
us to reliably measure our DTB.
Acked-by: Ard Biesheuvel <[email protected]>
Signed-off-by: Ilias Apalodimas <[email protected]>
Reviewed-by: Mark Kettenis <[email protected]>
|
|
Currently we allow and explicitly check a single shared page with
StandAloneMM. This is dictated by OP-TEE which runs the application.
However there's no way for us dynamically discover the number of pages we
are allowed to use. Since writing big EFI signature list variable
requires more than a page, OP-TEE has bumped the number of shared pages to
four.
Let's remove our explicit check and allow the request to reach OP-TEE even
if it's bigger than what it supports. There's no need to sanitize the
number of pages internally. OP-TEE will fail if we try to write more
than it's allowed. The error will just trigger later on, during the
StMM access.
While at it add an error message to help users figure out what failed.
Signed-off-by: Ilias Apalodimas <[email protected]>
Tested-by: Ying-Chun Liu (PaulLiu) <[email protected]>
Signed-off-by: Ilias Apalodimas <[email protected]>
|
