Commit Graph

862 Commits

Author SHA1 Message Date
Tom Rini
f855a7bc12 Patch queue for efi - 2017-10-13
This is the second batch of amazing improvements for efi_loader in 2017.11:
 
   - New self tests to verify our own code
   - A few bug fixes
   - colored text support
   - event and SNP improvements, should get us close to iPXE working
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2
 
 iQIcBAABAgAGBQJZ4HqlAAoJECszeR4D/txgZ0YQAIwd158lVczf/cMnzf3UcIsH
 e4YUFvQJaGxJw5hccaewaEQJqtPhEEJuHnzc683XDoYISTgvzCyhNEs8o0f7LcX3
 41c50x2bQPz7oFO5N60m58RknHqKiGd5VAC6+r4vBM8C0zFHeNTULo4fzn1qRgb3
 YUHDiRMlToVZXCOkbCeFm+dEJvjkXWmDDDdJpFIgDs6Oj+jOuV+H1GRMF2d56V6r
 Dd4/QPAFl4sRprBAMlvsmuI7obwnGV7/aA4CHIlUqKOVcbddBq2KigX8ZpIYEpaf
 FhRfxRLuWqvDRKpn+ixKfgl4KiBE/CH7sg7F6Un0mIjQVvo8k22Jbetqi9m1j/+x
 YzIFnwzb9ZhQEFOmfHnH8M9+bMmHyW46wkS4gdO4OAd/W0SeyIZ0a0gB1prgb2LK
 RfYQ37WsZM5TRARlcvmJt4H5+EtRYLT8lLf+yPSuyX8/8ubVnYHWtVE1IMLCiCAG
 eVmTwfjLzo3c703RJd9rwDtE13lTQmIHczM4cWybQzooXA6ePLJpSZSWx7EJxdQq
 GC7jNIKyKpu2NjhM9fUssbU+SiYLaIvtUBSOlGrcP1TCVR/j8OQ0Y1eIc9xOprIi
 +VEZVB/Z59D8j2TMTn1Xr+hB9i5h9ZxcO88gdEwfs65uihfvJGVhV1AxTyiaxvNv
 novPhik2UyHG5buRHb2Y
 =PiNi
 -----END PGP SIGNATURE-----

Merge tag 'signed-efi-next' of git://github.com/agraf/u-boot

Patch queue for efi - 2017-10-13

This is the second batch of amazing improvements for efi_loader in 2017.11:

  - New self tests to verify our own code
  - A few bug fixes
  - colored text support
  - event and SNP improvements, should get us close to iPXE working
2017-10-13 09:53:58 -04:00
Heinrich Schuchardt
abe994633b efi_selftest: correctly check return values
When cancelling the timer we should check the return
value provided by the set_timer service.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-13 10:24:22 +02:00
Rob Clark
2d5dc2a52d efi_loader: console support for color attributes
Shell.efi uses this, and supporting color attributes makes things look
nicer.  Map the EFI fg/bg color attributes to ANSI escape sequences.
Not all colors have a perfect match, but spec just says "Devices
supporting a different number of text colors are required to emulate the
above colors to the best of the device’s capabilities".

Signed-off-by: Rob Clark <robdclark@gmail.com>
Tested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Alexander Graf <agraf@suse.de>
[agraf: s/unsigned/unsigned int/]
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-12 17:24:25 +02:00
Rob Clark
bf19273e81 efi_loader: Add mem-mapped for fallback
When we don't have a real device/image path, such as 'bootefi hello',
construct a mem-mapped device-path.

This fixes 'bootefi hello' after devicepath refactoring.

Fixes: 95c5553ea2 ("efi_loader: refactor boot device and loaded_image handling")
Signed-off-by: Rob Clark <robdclark@gmail.com>
Acked-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-12 17:22:58 +02:00
Jonathan Gray
bcbc4a8046 efi_loader: don't increment part twice per loop
Correct a mistake in the part number handling of commit
16a73b249d and only increment part once
per loop.

Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Tested-by: Peter Robinson <pbrobinson@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-10 18:14:05 +02:00
Jonathan Gray
16a73b249d efi_loader: search all possible disk partitions
When searching for partitions don't stop if a partition is not present
for a given partition number as there may be valid partitions after.

Search for up to MAX_SEARCH_PARTITIONS matching the other callers of
part_get_info().

This allows OpenBSD to boot via the efi_loader on rpi_3 again after
changes made after U-Boot 2017.09.  With MBR partitioning OpenBSD will
by default use the fourth partition for the 0xA6 (OpenBSD) partition.

Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-10 09:56:04 +02:00
Heinrich Schuchardt
fdd04563ce efi_selftest: error handling in SNP test
Avoid NULL pointer dereference after setup failed due to a
missing network.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:38 +02:00
Heinrich Schuchardt
10a08c4ed7 efi_loader: comments for functions add missing @return
For some functions the @return description is missing.

Fix typo.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:38 +02:00
Rob Clark
77511b3b4b efi_loader: Fix disk dp's for pre-DM/legacy devices
This fixes an issue with OpenBSD's bootloader, and I think should also
fix a similar issue with grub2 on legacy devices.  In the legacy case
we were creating disk objects for the partitions, but not also the
parent device.

Reported-by: Jonathan Gray <jsg@jsg.id.au>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:38 +02:00
Heinrich Schuchardt
1a2c8d2f60 efi_loader: avoid NULL dereference in efi_dp_match
When calling bootefi hello twice a kernel dump occurs.

Neither bootefi hello nor bootefi selftest have an image
device patch. So do not try to dereference the NULL
value.

Fixes: 95c5553ea2 efi_loader: refactor boot device and loaded_image handling
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:37 +02:00
Heinrich Schuchardt
7f8ec5b63e efi_selftest: avoid dereferencing NULL in tpl test
The task priority levels test uses two events one passes the
notification counter as context. The other passes NULL.
Both use the same notification function. So we need to check
for NULL here.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:37 +02:00
Heinrich Schuchardt
c155dfeb1e efi_selftest: efi_st_memcmp return difference of bytes
If the memory regions are different efi_st_memcmp currently
returns the difference of the addresses. Insted the
difference of the first differing byte pair should be
returned.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:36 +02:00
Heinrich Schuchardt
5ca23ed5bc efi_loader: supply EFI network test
This patch provides an EFI application to check the correct function
of the Simple Network Protocol implementation.

It sends a DHCP request and analyzes the DHCP offer.

Different error conditions including a 10s timeout are checked.

A successful execution will look like this:

=> bootefi nettest
Scanning disk ide.blk#0...
Found 1 disks
WARNING: Invalid device tree, expect boot to fail
Network test
DHCP Discover
DHCP reply received from 192.168.76.2 (52:55:c0:a8:4c:02)
as broadcast message.
OK. The test was completed successfully.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:36 +02:00
Heinrich Schuchardt
1b6332597f efi_selftest: allow printing MAC addresses
Add %pm as format string to print a MAC address.
This is helpful when analyzing network problems.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:35 +02:00
Heinrich Schuchardt
336d9dfc0a efi_loader: fill return values in SimpleNetworkProtocol
In the receive function all return values should be filled.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:35 +02:00
Heinrich Schuchardt
8db174d651 efi_loader: size fields in SimpleNetworkProtocol
The size fields in the Simple Network Protocol are all
UINTN in the UEFI spec. So use size_t.

Provide a function description of the receive function.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:34 +02:00
Heinrich Schuchardt
891b3d9051 efi_loader: fix efi_net_get_status
The returned interrupt status was wrong.

As out transmit buffer is empty we need to always set
EFI_SIMPLE_NETWORK_TRANSMIT_INTERRUPT.

When we have received a packet we need to set
EFI_SIMPLE_NETWORK_RECEIVE_INTERRUPT.

Furthermore we should call efi_timer_check() to handle events.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:34 +02:00
Heinrich Schuchardt
e5c21603fc efi_loader: implement WaitForPacket event
The WaitForPacket event informs that a network package has been
received by the SimpleNetworkProtocol.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
[agraf: Move is_signaled = true line into efi_net_push()]
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:33 +02:00
Heinrich Schuchardt
a0549ef607 efi_loader: use events for efi_net_receive
A timer event is defined. The timer handler cares for receiving new
packets.

efi_timer_check is called both in efi_net_transmit and efi_net_receive
to enable events during network communication.

Calling efi_timer_check in efi_net_get_status is implemented in a
separate patch.

[agraf] This patch is needed to make efi_net_get_status() actually
        report incoming packets.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
[agraf: fix spelling in comment]
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:33 +02:00
Heinrich Schuchardt
61da678c39 efi_net: return EFI_UNSUPPORTED where appropriate
U-Boot does not implement all functions of the simple network
protocol. The unimplemented functions return either of
EFI_SUCCESS and EFI_INVALID_PARAMETER.

The UEFI spec foresees to return EFI_UNSUPPORTED in these cases.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:32 +02:00
Heinrich Schuchardt
5d4a5ea964 efi_loader: efi_net: hwaddr_size = 6
The length of a MAC address is 6.
We have to set this length in the EFI_SIMPLE_NETWORK_MODE
structure of the EFI_SIMPLE_NETWORK_PROTOCOL.

Without this patch iPXE fails to initialize the network with
error message
SNP MAC(001e0633bcbf,0x0) has invalid hardware address length 0

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:32 +02:00
Heinrich Schuchardt
bdecf974f1 efi_loader: fill simple network protocol revision
Provide the simple network protocol revision.
This revision number could be used to identify backwards compatible
enhancements of the protocol.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:32 +02:00
Heinrich Schuchardt
ca379e1bf1 efi_loader: wrong type in wait_for_event
The UEFI spec defines parameter index of WaitForEvent as UINTN*.
So we should use size_t here.

I deliberately do not use UINTN because I hold a following patch
that will eliminate UINTN because uppercase types to not match
the U-Boot coding style.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:30 +02:00
Heinrich Schuchardt
5a9682d0dd efi_loader: pass GUIDs as const efi_guid_t *
We need to call some boottime services internally.
Our GUIDs are stored as const efi_guid_t *.

The boottime services never change GUIDs.
So we can define the parameters as const efi_guid_t *.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:30 +02:00
Heinrich Schuchardt
fc05a95906 efi_loader: parameters of CopyMem and SetMem
The UEFI spec defines the length parameters of CopyMem and SetMem
as UINTN. We should size_t here.

The source buffer of CopyMem should be marked as const.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:29 +02:00
Heinrich Schuchardt
f7c78176d6 efi_loader: call EFI_EXIT in efi_copy_mem, efi_set_mem
EFI_ENTRY and EFI_EXIT calls must match.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:29 +02:00
Heinrich Schuchardt
7d963323a2 efi_loader: replace efi_div10 by do_div
We should use the existing 64bit division instead of
reinventing the wheel.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:29 +02:00
Heinrich Schuchardt
332468f7fc efi_loader: provide function comments for boot services
Provide comments describing the boot service functions.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:28 +02:00
Heinrich Schuchardt
e67e7249c8 efi_selftest: make tests easier to read
Rename counter to more illustrative names.
Update notification function description.
Simplify notification function.
Add comment for arbitrary non-zero value.
Document @return.
Use constants for return values of setup, execute, teardown.

Reported-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 07:00:23 +02:00
Heinrich Schuchardt
e190e8972f efi_loader: use type bool for event states
Queued and signaled describe boolean states of events.
So let's use type bool and rename the structure members to is_queued
and is_signaled.

Update the comments for is_queued and is_signaled.

Reported-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 06:59:38 +02:00
Heinrich Schuchardt
037ee6f91b efi_selftest: use efi_st_error for all error messages
All error messages in the selftests should use efi_st_error.
efi_st_error will print the file name and line number of the error.

Splitting message texts due to lines being over 80
characters is avoided. This resolves the issue reported
by Simon Glass in
https://lists.denx.de/pipermail/u-boot/2017-September/307387.html

Reported-by: Simon Glass <sjg@chromium.org>
Fixes: 623b3a5797 efi_selftest: provide an EFI selftest application
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-09 06:59:24 +02:00
zijun_hu
01a8359989 ARMv8: get new GD address from gd->new_gd directly
the new GD address is calculated via board data BD currently
it require the new GD area locates below BD tightly, so a strict
constraint is imposed on memory layout which maybe make special
platform unpleasant.

fix it by getting new GD address from gd->new_gd directly.

Signed-off-by: zijun_hu <zijun_hu@htc.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
2017-10-08 16:19:56 -04:00
Jean-Jacques Hiblot
2f57c95100 spl: dm: Make it possible for the SPL to pick its own DTB from a FIT
u-boot can be embedded within a FIT image with multiple DTBs. It then
selects at run-time  which one is best suited for the platform.
Use the same principle here for the SPL: put the DTBs in a FIT image,
compress it (LZO, GZIP, or no compression) and append it at the end of the
SPL.

Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
[trini: Move default y of SPL_MULTI_DTB_FIT_DYN_ALLOC to it being the
default choice if SYS_MALLOC_F, drop spl.h include from lib/fdtdec.c
it's unused.]
Signed-off-by Tom Rini <trini@konsulko.com>
2017-10-06 11:26:41 -04:00
Jean-Jacques Hiblot
035d64025c fdtdec: sort include files
Sort include files in accordance to U-Boot coding style.

Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
2017-10-05 21:31:04 -04:00
Jean-Jacques Hiblot
f52bdf4b67 lib: allow building lzo for the SPL
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
2017-10-05 21:31:04 -04:00
Jean-Jacques Hiblot
376ddf9d4a gzip: add a function to parse the header
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
2017-10-05 21:31:04 -04:00
Jean-Jacques Hiblot
d753f942ec lzo: add a function to check the validity of the header
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
2017-10-05 21:31:04 -04:00
Jean-Jacques Hiblot
11955590a2 dts: renamed FIT_EMBED to MULTI_DTB_FIT and moved it to the dts Kconfig
CONFIG_FIT_EMBED might be confused with CONFIG_OF_EMBED, rename it
MULTI_DTB_FIT as it is able to get a DTB from a FIT image containing
multiple DTBs. Also move the option to the Kconfig dedicated to the DTS
options and create a README for this feature.

Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
2017-10-05 21:31:04 -04:00
Heinrich Schuchardt
3cc6e3fe95 efi_loader: allow creating new handles
In efi_install_protocol_interface support creating
a new handle.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-10-05 14:46:57 +02:00
Marek Behún
85d8bf5713 lib: Add CRC32-C
This is needed for BTRFS.

Signed-off-by: Marek Behun <marek.behun@nic.cz>

 create mode 100644 lib/crc32c.c
2017-10-02 20:31:25 -04:00
Rob Clark
3a45bc7faa efi_loader: Some console improvements for vidconsole
1) use fputs() to reduce cache flushes from once-per-char to
   once-per-string
2) handle \r, \t, and \b in addition to just \n for tracking
   cursor position
3) cursor row/col are zero based, not one based

Signed-off-by: Rob Clark <robdclark@gmail.com>
[agraf: s/unsigned/unsigned int/]
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 11:29:35 +02:00
Rob Clark
a18c5a8382 efi_loader: Correctly figure out size for vidconsole
If stdout is vidconsole, we cannot rely on ANSI escape sequences to
query the size, as vidconsole cannot reply on stdin.  Instead special-
case this if stdout is vidconsole.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 11:28:39 +02:00
Rob Clark
71cc25c35c efi_loader: split out escape sequence based size query
We need to do something different for vidconsole, since it cannot
respond to the query on stdin.  Prep work for next patch.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 11:14:12 +02:00
Rob Clark
778e6af8a6 efi_loader: print GUIDs
Utilize printf GUID support to print GUIDs.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 11:10:42 +02:00
Rob Clark
4657a3f1ac efi_loader: set loaded image code/data type properly
These should be set according to the image type.  Shell.efi and SCT.efi
use these fields to determine what sort of image they are loading.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 11:10:24 +02:00
Rob Clark
9975fe96b6 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>
2017-09-20 11:08:01 +02:00
Rob Clark
ad644e7c18 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:

  efi_$guid_$varname = {attributes}(type)value

For example:

  efi_8be4df61-93ca-11d2-aa0d-00e098032b8c_OsIndicationsSupported=
     "{ro,boot,run}(blob)0000000000000000"
  efi_8be4df61-93ca-11d2-aa0d-00e098032b8c_BootOrder=
     "(blob)00010000"

The attributes are a comma separated list of these possible
attributes:

  + ro   - read-only
  + boot - boot-services access
  + run  - runtime access

NOTE: with current implementation, no variables are available after
ExitBootServices, and all are persisted (if possible).

If not specified, the attributes default to "{boot}".

The required type is one of:

  + utf8 - raw utf8 string
  + blob - arbitrary length hex string

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 11:00:57 +02:00
Rob Clark
946160f334 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>
2017-09-20 10:48:09 +02:00
Rob Clark
838ee4b479 efi_loader: support load_image() from a file-path
Previously we only supported the case when the EFI application loaded
the image into memory for us.  But fallback.efi does not do this.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 10:45:28 +02:00
Rob Clark
2a92080d8c efi_loader: add file/filesys support
fallback.efi (and probably other things) use UEFI's simple-file-system
protocol and file support to search for OS's to boot.

Signed-off-by: Rob Clark <robdclark@gmail.com>
[agraf: whitespace fixes, unsigned fixes]
Signed-off-by: Alexander Graf <agraf@suse.de>
2017-09-20 10:43:54 +02:00