Rob Clark [Wed, 13 Sep 2017 22:05:38 +0000 (18:05 -0400)]
efi_loader: add bootmgr
Similar to a "real" UEFI implementation, the bootmgr looks at the
BootOrder and BootXXXX variables to try to find an EFI payload to load
and boot. This is added as a sub-command of bootefi.
The idea is that the distro bootcmd would first try loading a payload
via the bootmgr, and then if that fails (ie. first boot or corrupted
EFI variables) it would fallback to loading bootaa64.efi. (Which
would then load fallback.efi which would look for \EFI\*\boot.csv and
populate BootOrder and BootXXXX based on what it found.)
Signed-off-by: Rob Clark <robdclark@gmail.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Rob Clark [Wed, 13 Sep 2017 22:05:37 +0000 (18:05 -0400)]
efi_loader: efi variable support
Add EFI variable support, mapping to u-boot environment variables.
Variables are pretty important for setting up boot order, among other
things. If the board supports saveenv, then it will be called in
ExitBootServices() to persist variables set by the efi payload. (For
example, fallback.efi configuring BootOrder and BootXXXX load-option
variables.)
Variables are *not* currently exposed at runtime, post ExitBootServices.
On boards without a dedicated device for storage, which the loaded OS
is not trying to also use, this is rather tricky. One idea, at least
for boards that can persist RAM across reboot, is to keep a "journal"
of modified variables in RAM, and then turn halt into a reboot into
u-boot, plus store variables, plus halt. Whatever the solution, it
likely involves some per-board support.
Mapping between EFI variables and u-boot variables:
Rob Clark [Wed, 13 Sep 2017 22:05:36 +0000 (18:05 -0400)]
efi_loader: make pool allocations cacheline aligned
This avoids printf() spam about file reads (such as loading an image)
into unaligned buffers (and the associated memcpy()). And generally
seems like a good idea.
Signed-off-by: Rob Clark <robdclark@gmail.com>
[agraf: use __aligned] Signed-off-by: Alexander Graf <agraf@suse.de>
Rob Clark [Wed, 13 Sep 2017 22:05:33 +0000 (18:05 -0400)]
efi_loader: refactor boot device and loaded_image handling
Get rid of the hacky fake boot-device and duplicate device-path
constructing (which needs to match what efi_disk and efi_net do).
Instead convert over to use efi_device_path helpers to construct
device-paths, and use that to look up the actual boot device.
Also, extract out a helper to plug things in properly to the
loaded_image. In a following patch we'll want to re-use this in
efi_load_image() to handle the case of loading an image from a
file_path.
Signed-off-by: Rob Clark <robdclark@gmail.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Rob Clark [Wed, 13 Sep 2017 22:05:31 +0000 (18:05 -0400)]
efi_loader: use proper device-paths for partitions
Also, create disk objects for the disk itself, in addition to the
partitions. (UEFI terminology is a bit confusing, a "disk" object is
really a partition.) This helps grub properly identify the boot device
since it is trying to match up partition "disk" object with it's parent
device.
This is on a board with single USB disk and single sd-card. The
UnknownMessaging(1d) node in the device-path is the MMC device,
but grub_efi_print_device_path() hasn't been updated yet for some
of the newer device-path sub-types.
This patch is inspired by a patch originally from Peter Jones, but
re-worked to use efi_device_path, so it doesn't much resemble the
original.
Signed-off-by: Rob Clark <robdclark@gmail.com>
[agraf: s/unsigned/unsigned int/] Signed-off-by: Alexander Graf <agraf@suse.de>
Rob Clark [Wed, 13 Sep 2017 22:05:30 +0000 (18:05 -0400)]
efi_loader: flesh out device-path to text
It needs to handle more device-path node types, and also multiple levels
of path hierarchy. To simplify this, initially construct utf8 string to
a temporary buffer, and then allocate the real utf16 buffer that is
returned. This should be mostly for debugging or at least not critical-
path so an extra copy won't hurt, and is saner than the alternative.
Signed-off-by: Rob Clark <robdclark@gmail.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Rob Clark [Wed, 13 Sep 2017 22:05:28 +0000 (18:05 -0400)]
efi_loader: add device-path utils
Helpers to construct device-paths from devices, partitions, files, and
for parsing and manipulating device-paths.
For non-legacy devices, this will use u-boot's device-model to construct
device-paths which include bus hierarchy to construct device-paths. For
legacy devices we still fake it, but slightly more convincingly.
Signed-off-by: Rob Clark <robdclark@gmail.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Peter Jones [Wed, 13 Sep 2017 22:05:25 +0000 (18:05 -0400)]
part: extract MBR signature from partitions
EFI client programs need the signature information from the partition
table to determine the disk a partition is on, so we need to fill that
in here.
Signed-off-by: Peter Jones <pjones@redhat.com>
[separated from efi_loader part, and fixed build-errors for non-
CONFIG_EFI_PARTITION case] Signed-off-by: Rob Clark <robdclark@gmail.com> Signed-off-by: Alexander Graf <agraf@suse.de>
efi_selftest: provide unit test for event services
This unit test uses timer events to check the implementation
of the following boottime services:
CreateEvent, CloseEvent, WaitForEvent, CheckEvent, SetTimer
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
efi_loader: implement queueing of the notification function
For the correct implementation of the task priority level (TPL)
calling the notification function must be queued.
Add a status field 'queued' to events.
In function efi_signal_event set status queued if a notification
function exists and reset it after we have called the function.
A later patch will add a check of the TPL here.
In efi_create_event and efi_close_event unset the queued status.
In function efi_wait_for_event and efi_check_event
queue the notification function.
In efi_timer_check call the efi_notify_event
if the status queued is set.
For all timer events set status signaled.
In efi_console_timer_notify set the signaled state of the
WaitForKey event.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
Andre Przywara [Thu, 6 Jul 2017 09:14:03 +0000 (10:14 +0100)]
EFI: find EFI system partition by legacy MBR partition type
The UEFI spec allows an EFI system partition (ESP, with the bootloader or
kernel EFI apps on it) to reside on a disk using a "legacy" MBR
partitioning scheme.
But in contrast to actual legacy disks the ESP is not marked as
"bootable" using bit 7 in byte 0 of the legacy partition entry, but is
instead using partition *type* 0xef (in contrast to 0x0b or 0x0c for a
normal FAT partition). The EFI spec isn't 100% clear on this, but it even
seems to discourage the use of the bootable flag for ESPs.
Also it seems that some EFI implementations (EDK2?) even seem to ignore
partitions marked as bootable (probably since they believe they contain
legacy boot code).
The Debian installer [1] (*not* mini.iso), for instance, contains such an
MBR, where none of the two partitions are marked bootable, but the ESP
has clearly type 0xef.
Now U-Boot cannot find the ESP on such a disk (USB flash drive) and
fails to load the EFI grub and thus the installer.
Since it all boils down to the distro bootcmds eventually calling
"part list -bootable" to find potential boot partitions, it seems logical
to just add this "partition type is 0xef" condition to the is_bootable()
implementation.
This allows the bog standard arm64 Debian-testing installer to boot from
an USB pen drive on Allwinner A64 boards (Pine64, BananaPi-M64).
(Ubuntu and other distribution installers don't have a legacy MBR, so
U-Boot falls back to El Torito there).
[1] https://cdimage.debian.org/cdimage/daily-builds/daily/arch-latest/arm64/iso-cd/ Signed-off-by: Andre Przywara <andre.przywara@arm.com> Reviewed-by: Alexander Graf <agraf@suse.de> Signed-off-by: Alexander Graf <agraf@suse.de>
scripts/Makefile.lib: generalize building built in EFI app
Replace all occurences of helloworld by generalized forms.
This allows us to build additional EFI applications that are
included into the U-Boot binary without loading
scripts/Makefile.lib with specific filenames.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
In scripts/Makefile.lib we build section including helloworld.efi.
This allows to load the EFI binary with command 'bootefi hello'.
scripts/Makefile.lib contains explicit references to strings
containing helloworld and hello_world. This makes it impossible
to generalize the coding to accomodate additional built in
EFI binaries.
Let us rename the variables __efi_hello_world_* to
__efi_helloworld_*.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
The target
$(obj)/helloworld.so:
exists twice in Makefile.lib.
If you add an echo command to each of the two recipes you get
warnings like:
scripts/Makefile.lib:383: warning:
overriding recipe for target 'drivers/power/battery/helloworld.so'
scripts/Makefile.lib:379: warning:
ignoring old recipe for target 'drivers/power/battery/helloworld.so'
This patch removes the obsolete target.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
Alexander Graf [Sun, 3 Sep 2017 12:14:17 +0000 (14:14 +0200)]
efi_loader: Fix efi_exit gd clobbering
Commit f494950b (efi_loader: call __efi_exit_check in efi_exit) added a call
to __efi_exit_check inside efi_exit to account for the fact that we're exiting
the efi_exit function via a longjmp call.
However, __efi_exit_check also swizzles gd to the application gd while the
longjmp will put us back into EFI context, so we need the efi (u-boot) gd.
This patch fixes that up by explicitly setting gd back to efi_gd before
doing the longjmp. It also adds a few comments on why it does that.
Simon Glass [Wed, 6 Sep 2017 01:49:46 +0000 (19:49 -0600)]
dm: x86: Allow TSC timer to be used before DM is ready
With bootstage we need access to the timer before driver model is set up.
To handle this, put the required state in global_data and provide a new
function to set up the device, separate from the driver's probe() method.
This will be used by the 'early' timer also.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Wed, 6 Sep 2017 01:49:45 +0000 (19:49 -0600)]
board_f: Drop the timer after relocation
Once U-Boot relocates itself the existing driver-model timer (if any) is
no-longer valid until the device is reinitialised. Any use of the device
may cause a crash. To handle this, set the timer to NULL after relocation.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Bin Meng [Wed, 16 Aug 2017 05:42:02 +0000 (22:42 -0700)]
x86: Support Intel Cherry Hill board
This adds support to Intel Cherry Hill board, a board based on
Intel Braswell SoC. The following devices are validated:
- serial port as the serial console
- on-board Realtek 8169 ethernet controller
- SATA AHCI controller
- EMMC/SDHC controller
- USB 3.0 xHCI controller
- PCIe x1 slot with a graphics card
- ICH SPI controller with an 8MB Macronix SPI flash
- Integrated graphics device as the video console
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 16 Aug 2017 05:41:56 +0000 (22:41 -0700)]
x86: dm: video: Add a framebuffer driver that utilizes VBT
When a VBT is given to an FSP that supports graphics initialization,
the FSP will produce a graphics info HOB that contains all necessary
information for the linear frame buffer of the integrated graphics
device. This adds a DM video driver for it.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 16 Aug 2017 05:41:53 +0000 (22:41 -0700)]
x86: Add Video BIOS Table (VBT) related Kconfig options
This adds Kconfig options for Video BIOS Table which is normally
required if you are using an Intel FSP firmware that is complaint
with spec 1.1 or later to initialize the integrated graphics device.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 16 Aug 2017 05:41:52 +0000 (22:41 -0700)]
x86: fsp: Add FSP_GRAPHICS_INFO_HOB
This adds a new HOB type for graphics information introduced in FSP
spec 1.1. When graphics capability is included in FSP and enabled,
FSP produces an FSP_GRAPHICS_INFO_HOB as described in the EFI PI
specification which provides information about the graphics mode and
framebuffer.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 16 Aug 2017 05:41:51 +0000 (22:41 -0700)]
x86: fsp: Update struct common_buf for FSP spec 1.1
FSP spec 1.1 adds one more member to the struct common_buf to
determine the memory size that can be reserved by FSP below "top
of low usable memory" for bootloader usage. This new member uses
the reserved space so that it is still compatible with previous
FSP spec 1.0.
A new HOB (FSP_HOB_RESOURCE_OWNER_BOOTLOADER_TOLUM_GUID) is also
published when common_buf.tolum_size is valid and non zero.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 16 Aug 2017 05:41:50 +0000 (22:41 -0700)]
x86: tsc: Add Airmont reference clock values
Per the Intel 64 and IA-32 Architecture Software Developer's Manual,
add the reference clock for Intel Atom Processors based on the Airmont
Microarchitecture (Braswell).
This keeps in sync with Linux kernel commit: 6fcb41c: x86/tsc_msr: Add Airmont reference clock values
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Stephen Warren [Fri, 15 Sep 2017 18:19:38 +0000 (12:19 -0600)]
test/py: gpt: make use of infra-structure
Make various changes to the GPT test:
1) Reference the disk image using an absolute path in all cases. This
allows test/py to operate correctly if it's run from a directory other
than the root of the U-Boot source tree.
2) Store the disk image in the teswt/py persistent data directory. This
removes the need to re-generate it every time the tests are run.
3) Execute sgdisk using u_boot_utils.run_and_log() so that its output is
captured in the test log. This allows debugging any problems running it.
4) Make the disk image a test fixture. This removes the requirement to
always run all GPT tests, and run them in order. The current code doesn't
create the disk image if e.g. just test_gpt_uuid() is executed via the
test.py -k command-line option.
5) Use @pytest.mark.buildconfigspec for all feature dependencies, rather
than manually implementing some of them.
6) Make all tests depend on sandbox, since they use the sandbox-specific
host command.
Fixes: a2f422555fc8 ("add pytests for 'gpt guid' command in sandbox") Fixes: c5772188ede9 ("add pytests for 'gpt rename' and 'gpt swap'") Signed-off-by: Stephen Warren <swarren@nvidia.com> Tested-by: Tom Rini <trini@konsulko.com>
arm: dts: Add missing u-boot specific dtsi for dra7-evm
commit b63b995 (arm: dts: Add u-boot specific compatibles) separated out
u-boot specific compatibles from dts files.
The u-boot specifics were moved in *-u-boot.dtsi files for all dra7
platforms except dra7-evm.
Without it the SD card is broken on DRA7-evm because the regulator cannot
be enabled.
Fix it by adding the missing dra7-evm-u-boot.dtsi with the needed
properties.
Rob Clark [Tue, 12 Sep 2017 20:40:01 +0000 (16:40 -0400)]
fs/fat: fix fatbuf leak
A new fatbuf was allocated by get_fs_info() (called by fat_itr_root()),
but not freed, resulting in eventually running out of memory. Spotted
by running 'ls -r' in a large FAT filesystem from Shell.efi.
fatbuf is mainly used to cache FAT entry lookups (get_fatent())..
possibly once fat_write.c it can move into the iterator to simplify
this.
Signed-off-by: Rob Clark <robdclark@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Łukasz Majewski <lukma@denx.de>
Rob Clark [Sat, 9 Sep 2017 17:16:00 +0000 (13:16 -0400)]
fs/fat: Clean up open-coded sector <-> cluster conversions
Use the clust_to_sect() helper that was introduced earlier, and add an
inverse sect_to_clust(), plus update the various spots that open-coded
this conversion previously.
Signed-off-by: Rob Clark <robdclark@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Rob Clark [Sat, 9 Sep 2017 17:15:58 +0000 (13:15 -0400)]
fat/fs: move ls to generic implementation
Add a generic implementation of 'ls' using opendir/readdir/closedir, and
replace fat's custom implementation. Other filesystems should move to
the generic implementation after they add opendir/readdir/closedir
support.
Signed-off-by: Rob Clark <robdclark@gmail.com> Reviewed-by: Łukasz Majewski <lukma@denx.de> Reviewed-by: Simon Glass <sjg@chromium.org>
Rob Clark [Sat, 9 Sep 2017 17:15:55 +0000 (13:15 -0400)]
fs: add fs_readdir()
Needed to support efi file protocol. The fallback.efi loader wants
to be able to read the contents of the /EFI directory to find an OS
to boot.
Modelled after POSIX opendir()/readdir()/closedir(). Unlike the other
fs APIs, this is stateful (ie. state is held in the FS_DIR "directory
stream"), to avoid re-traversing of the directory structure at each
step. The directory stream must be released with closedir() when it
is no longer needed.
Signed-off-by: Rob Clark <robdclark@gmail.com> Reviewed-by: Łukasz Majewski <lukma@denx.de> Reviewed-by: Simon Glass <sjg@chromium.org>
Bin Meng [Wed, 13 Sep 2017 02:00:23 +0000 (19:00 -0700)]
blk: Remove various places that do flush cache after read
All these places seem to inherit the codes from the MMC driver where
a FIXME was put in the comment. However the correct operation after
read should be cache invalidate, not flush.
The underlying drivers should be responsible for the cache operation.
Remove these codes completely.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de> Reviewed-by: York Sun <york.sun@nxp.com> Reviewed-by: Joe Hershberger <joe.hershberger@ni.com> Reviewed-by: Simon Glass <sjg@chromium.org> Tested-by: York Sun <york.sun@nxp.com>
fdt_getprop_u32 is not exported and it's different than what the
unit test uses. Rename u32 prop access methods to something that's
unit test specific.
Signed-off-by: Pantelis Antoniou <pantelis.antoniou@konsulko.com> Acked-by: Simon Glass <sjg@chromium.org>
Pull some information regarding overlays from commit messages and
put them directly within the documentation. Also add some information
regarding required dtc version to properly use overlays.
Signed-off-by: Franklin S Cooper Jr <fcooper@ti.com> Acked-by: Simon Glass <sjg@chromium.org>
This patch enables an overlay to refer to a previous overlay's
labels by performing a merge of symbol information at application
time.
In a nutshell it allows an overlay to refer to a symbol that a previous
overlay has defined. It requires both the base and all the overlays
to be compiled with the -@ command line switch so that symbol
information is included.
Introduce fdt_overlay_apply_verbose, a method that applies an
overlay but in the case of an error produces a helpful message.
In addition if a base tree is found to be missing the __symbols__
node the message will point out that the probable reason is that
the base tree was miscompiled without the -@ option.
Signed-off-by: Pantelis Antoniou <pantelis.antoniou@konsulko.com> Acked-by: Simon Glass <sjg@chromium.org>
Simon Glass [Tue, 29 Aug 2017 20:16:00 +0000 (14:16 -0600)]
dtoc: Rename the auto-generated dt-structs.h file
The filename of the auto-generated file is the same as the file that
includes it. Even though the form is in the generated/ subdirectory, this
could be confused.
Rename the generated file to something that makes it clear it is
auto-generated.
Signed-off-by: Simon Glass <sjg@chromium.org> Tested-by: Kever Yang <kever.yang@rock-chips.com>
Simon Glass [Tue, 29 Aug 2017 20:15:59 +0000 (14:15 -0600)]
dtoc: Support properties containing multiple phandle values
At present dtoc has a very simplistic view of phandles. It assumes that
a property has only a single phandle with a single argument (i.e. two
cells per property).
This is not true in many cases. Enhance the implementation to scan all
phandles in a property and to use the correct number of arguments (which
can be 0, 1, 2 or more) when generating the C code. For the struct
definitions, use a struct which can hold the maximum number of arguments
used by the property.
Signed-off-by: Simon Glass <sjg@chromium.org> Tested-by: Kever Yang <kever.yang@rock-chips.com>
Simon Glass [Tue, 29 Aug 2017 20:15:58 +0000 (14:15 -0600)]
dtoc: Put phandle args in an array
We want to support more than one phandle argument. It makes sense to use
an array for this rather than discrete struct members. Adjust the code to
support this. Rename the member to 'arg' instead of 'id'.
Signed-off-by: Simon Glass <sjg@chromium.org> Tested-by: Kever Yang <kever.yang@rock-chips.com>
Simon Glass [Tue, 29 Aug 2017 20:15:57 +0000 (14:15 -0600)]
dtoc: Put each phandle on a separate line
When writing values from properties which contain phandles, dtoc currently
writes 8 phandles per line. Change this to write one phandle per line.
This helps reduce line length, since phandles are generally longer and may
have arguments.
Signed-off-by: Simon Glass <sjg@chromium.org> Tested-by: Kever Yang <kever.yang@rock-chips.com>
projects / u-boot.git / log