Add a utility to generate a compressed BIOS image for AMD Family 17h.
If the input is an elf file, the utility extracts the program portion
for compression. Otherwise the file is compressed as-is.
In modern AMD systems, the PSP brings up DRAM then uncompresses the
BIOS image into memory prior to x86 beginning execution. The PSP
supports a zlib engine, and interprets the first 256 bytes as a
header, where offset 0x14 containing the uncompressed size. For
further details, see AMD Platform Security Processor BIOS Architecture
Design Guide for AMD Family 17h Processors (NDA only, #55758).
BUG=b:127766506
TEST=Use with WIP Picasso
Change-Id: Id1c54e0a6dae9e4a0362c6635fe8b8aa48a369d8
Signed-off-by: Marshall Dawson <marshalldawson3rd@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33401
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
Depend on ifittool and use it to update FIT instead cbfstool FIT code.
Move the TOPSWAP / microcode handling out of cbfstool and implement it
in the Makefile.
The new FIT looks like the old one and has been tested on Broadwell-DE.
The TOPSWAP / microcode code path needs test on real hardware.
Change-Id: I687469d62557f81e9d88398cfc93182164fdac95
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/31495
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Christian Walter <christian.walter@9elements.com>
Clang doesn't know `-Wlogical-op`, so let's move it into xcompile where
we can easily distinguish between the two. However, this requires us to
split out `GCC_ADAFLAGS*` from `GCC_CFLAGS*`.
Change-Id: I6a50de0bc5372f61337f237383d32645ba86b0fd
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33579
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch removes all possible dependencies in order to build platform
with CONFIG_RAMPAYLOAD enable(without ramstage).
A. Create coreboot separate stage kconfigs
This patch creates seperate stage configs as below
1. HAVE_BOOTBLOCK
2. HAVE_VERSTAGE
3. HAVE_ROMSTAGE
4. HAVE_POSTCAR
5. HAVE_RAMSTAGE
B. Also ensures below kconfigs are aligned with correct stage configs
1. COMPRESS_RAMSTAGE and RELOCATABLE_RAMSTAGE are now enable if
CONFIG_HAVE_RAMSTAGE is selected.
2. COMPRESS_BOOTBLOCK will enable if CONFIG_HAVE_BOOTBLOCK is set
3. COMPRESS_PRERAM_STAGES will enable if CONFIG_HAVE_VERSTAGE
|| CONFIG_HAVE_ROMSTAGE is selected.
C. Also fix compilation issue with !CONFIG_HAVE_RAMSTAGE
On x86 platform:
Case 1: ramstage do exist: CONFIG_HAVE_RAMSTAGE=1
>> rmodules_$(ARCH-ramstage-y) will evaluate as rmodules_x86_32
Case 2: ramstage doesn't exist: CONFIG_HAVE_RAMSTAGE=0
>> rmodules_$(ARCH-ramstage-y) will evaluate as rmodules_
This patch fixes Case 2 usecase where platform doesn't select
CONFIG_HAVE_RAMSTAGE.
Also add option to create sipi_vector.manual based on $(TARGET_STAGE)
variable.
$(TARGET_STAGE)=ramstage if user selects CONFIG_HAVE_RAMSTAGE
$(TARGET_STAGE)=postcar if user selects CONFIG_RAMPAYLOAD
Change-Id: I0f7e4174619016c5a54c28bedd52699df417a5b7
Signed-off-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/33142
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
With the version string "TIMELESS", binaries are slightly smaller than
for a regular build. This may lead to false positive build tests if the
space is limited (e.g. bootblock). So let's make the string a little
longer.
Change-Id: I3bbf6f71d5bcd74728a3fe39734312690901d0ec
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32986
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This reverts commit 99e836c843.
This relands commit c4ab50cdde.
The issues with -Wtype-limits in the vboot submodule have been resolved
now, so we can enable this flag again.
Change-Id: I32e8cc88e69072e7ee66cf443b578a9a8ea0ebe2
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32749
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Joel Kitching <kitching@google.com>
This reverts commit c4ab50cdde.
Reason for revert:
vboot recently was changed so that -Wtype-limits fails, and that
was just brought upstream. Since vboot internals are more likely
to be used elsewhere while -Wtype-limits is less critical, revert
this until vboot is resolved, then bring -Wtype-limits back again.
Change-Id: I9cce10462b9e57189513fa49e11fd27ebe35ba51
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32670
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch enables -Wtype-limits for the whole repository, which
disallows checking for a condition that must be always true or always
false based on type width (e.g. checking whether an unsigned variable is
negative or whether a 32-bit integer is larger than 4G). This helps
avoid easy to make and hard to find (because they often only affect
error paths) mistakes like
size_t size = fmap_read_area(...);
if (size < 0)
die("If only the compiler could've told me to use ssize_t instead");
Change-Id: I19edabfd092d09dad720e3fc47b44838163bfe25
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32536
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Rather than selectively update the fsp submodule based on
FSP version or platform selection, update it when building
for any FSP-enabled platform, so all have latest version available.
Change-Id: If07d55828a1863623e04a4ecdd1514c3cb6d9c11
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32526
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
We only grep'ed for "ACPI Warning" resulting in an actual more severe
"ACPI Error" being ignored.
Change-Id: I9cec8a388f5558b1ffc383cc2fc69405252cbb37
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/32469
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Frans Hendriks <fhendriks@eltan.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
We used .aml as the file extension for preprocessed ASL code. That
file gets overwritten with the compilation results, fix that.
Change-Id: I11a03dfbcebb0fd762da7b27862a7bdb9a581b92
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/31523
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
If a non aligned CONFIG_CBFS_SIZE is used the region RW_MRC_CACHE and
CONSOLE could end up non aligned. Currently this is only possible if
the user messes with CONFIG_CBFS_SIZE in menuconfig, but better be
safe than sorry.
Change-Id: Ieb7e3c7112bd4b3f9733c36af21b1d59b3836811
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/30420
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Put the FMAP FMAP region right above the coreboot CBFS region.
The other regions like RW_MRC_CACHE and CONSOLE often have alignment
requirements so it makes sense to put those on top. This also
simplifies the code the generate the default fmap a little.
Change-Id: I24fa6c89ecf85fb9002c0357f14aa970ee51b1df
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/30419
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
We generate a $(obj)/cb-config.ads once and copy it per stage that uses
it to $(obj)/<stage>/cb-config.ads (to simplify the gnat-bind step). The
Ada package is called `CB.Config`. As there was no `CB` package yet, add
that too.
Change-Id: I963a6517ef4bcf84f2c8e9ae8d24a0d6b971d2b0
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/30584
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Default to FSP binary and headers shiped in 3rdparty/fsp.
* Drop headers and code from vendorcode/intel/fsp1_0/broadwell_de
* Select HAVE_FSP_BIN to build test the platform
* Fetch FSP repo as submodule
* Make FSP_HEADER_PATH known from FSP2.0 useable on FSP1.0
* Introduce FSP_SRC_PATH for FSP source file
* Add sane defaults for FSP_FILE
Tested on wedge100s.
Change-Id: I46f201218d19cf34c43a04f57458f474d8c3340d
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/30742
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Jay Talbott <JayTalbott@sysproconsulting.com>
The option to specify a binary file name was added later for platforms
that do not provide microcode updates in our blobs repository. Alas,
it wasn't visible what platforms these are. And if you specified a file
for a platform that already had one, they were all included together.
Make it visible which platforms don't provide binaries with the new con-
figs MICROCODE_BLOB_NOT_IN_BLOB_REPO, MICROCODE_BLOB_NOT_HOOKED_UP and
MICROCODE_BLOB_UNDISCLOSED. Based on that we can decide if we want to
include binaries by default or explicitly show that no files are inclu-
ded (default to CPU_MICROCODE_CBFS_NONE).
Also split CPU_MICROCODE_CBFS_GENERATE into the more explicit
CPU_MICROCODE_CBFS_DEFAULT_BINS and CPU_MICROCODE_CBFS_EXTERNAL_BINS.
And clean up the visibility of options: Don't show CBFS related options
on platforms that don't support it and don't show external file options
if the platform uses special rules for multiple files (CPU_MICROCODE_
MULTIPLE_FILES).
Change-Id: Ib403402e240d3531640a62ce93b7a93b4ef6ca5e
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/29934
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Use the `git-format' tool to sanitise coreboot commits such that
they conform to coreboot's coding style.
This fancy piece of machinary allows one to have LibFormat from
Clang to automatically check your commit conforms to coreboot's
coding style, fix any issues automatically and provides you a
diff you may review and apply at your convenience.
N.B. When the `clang-format' binary is not found we issue a warning
that the test was skipped and carry on as usual. Hence, this is
strictly non-enforcing at this current time. You may use it at your
leisure.
Change-Id: If49017ea82f0707efd47cae5978a286a9af8f3b7
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: https://review.coreboot.org/c/8037
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: HAOUAS Elyes <ehaouas@noos.fr>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
When building coreboot from scratch with 'make -j4', I sometimes see
this error:
CREATE build/mainboard/emulation/qemu-riscv/cbfs-file.wblRgZ.out (from /.../coreboot/.config)
HOSTCC cbfstool/cbfstool (link)
make[1]: execvp: build/util/kconfig/conf: Permission denied
make[1]: *** [/.../coreboot/util/kconfig/Makefile:92: savedefconfig] Error 127
It happens, I think, because the rule generated by
cbfs-files-processor-defconfig runs 'make savedefconfig', which builds
build/util/kconfig/conf, and something also builds it, at the same time.
Fix this case, by making this rule depend on $(objutil)/kconfig/conf.
The same fix is also precautiously applied to the rule for
$(KCONFIG_AUTOHEADER) in Makefile.
Change-Id: Ie93eda567f88ca08c97df7e70cdff5b07442747d
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-on: https://review.coreboot.org/c/29984
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
The reason for this code cleanup is the legacy
Google Purin board which isn't available anymore
and AFAIK never made it into the stores.
* Remove broadcom cygnus SoC support
* Remove /util/broadcom tool
* Remove Google Purin mainboard
* Remove MAINTAINERS entries
Change-Id: I148dd7eb0192d396cb69bc26c4062f88a764771a
Signed-off-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/29905
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Check for PLATFORM_USES_FSP2_0 instead of MAINBOARD_USES_FSP2_0. The
latter is only valid for Skylake where we decide per mainboard if FSP2.0
is used. PLATFORM_USES_FSP2_0 is the one that actually enables the
FSP2.0 integration.
Change-Id: I3f16e5f4454c0bf02d51db5d1c267a921917f377
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/29291
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Lijian Zhao <lijian.zhao@intel.com>
Reviewed-by: Pratikkumar V Prajapati <pratikkumar.v.prajapati@intel.com>
Its spreading copies got out of sync. And as it is not a standard header
but used in commonlib code, it belongs into commonlib. While we are at
it, always include it via GCC's `-include` switch.
Some Windows and BSD quirk handling went into the util copies. We always
guard from redefinitions now to prevent further issues.
Change-Id: I850414e6db1d799dce71ff2dc044e6a000ad2552
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/28927
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
GNU make is too smart (or too stupid?) for empty recipes. In the case of
empty recipes, GNU make doesn't consider the target as updated even if
its prerequisites are. So if we told make to rebuild `build/romstage/
lib/cbfs.o` for instance, and the FMAP changed, it rerun the fmaptool
recipe (as a prerequisite) but only considered `cbfs.o` to be updated
by chance.
Just not leaving the recipes empty seems to help here. I seeemed to
remember that it wasn't that easy, but it fixes the issue for me...
Change-Id: Ic7ecb88cf7df7f2488defd47ea02255fc10a67e9
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/28198
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
It only happens if both USE_BLOBS and MAINBOARD_USES_FSP2_0 are enabled.
Change-Id: I46843c61d3ddf398a3c058bb571d285b596bf5c1
Signed-off-by: Patrick Georgi <patrick@georgi.software>
Reviewed-on: https://review.coreboot.org/28304
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Since ifdfake has been deprecated in favor of better alternatives, there
is no need to support it any further. Remove it from the build system.
Change-Id: Id62e95ba72004a1e15453e3eb75f09cb8194feb2
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/28233
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
There were so many pitfalls that I wrote my own version of this even-
tually. This version is inspired by the procedure of Alex Thiessen[1].
Instead of generating a `build.h` on demand, we always generate a tem-
porary version that, if it differs from the current one, is added as
a dependency.
As we use .SECONDEXPANSION on the prerequisites, special care is taken
that we won't generate the file twice. As it would be too late to add
the dependency if we'd run `genbuild_h.sh` inside a recipe, we have
to run it through the `$(shell)` function. But that brings us to the
next issue: The make variables used by `genbuild_h.sh` are not expor-
ted to this shell like they would be in a recipe. So we export them
manually. We could also make these variables explicit parameters of
`genbuild_h.sh` instead.
An alternative to always creating the temporary `build.h` would be
to add a phony target as dependency instead, and finally calling
`genbuild_h.sh` again in case we need an update. But, um, we create
so many files anyway...
[1] https://review.coreboot.org/25685
Change-Id: I311cf610eabae873c70f2985fc7a09acec8061f0
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/28197
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
To install this hook, run
make install-git-commit-clangfmt
This will install a pre-commit-msg hook that runs clang-format
on all .c and .h files that are staged.
It will add a clang-formatted-by: <git username>
line to the commit message to indicate that clang-format
was run on the files and that further processing of them
is not needed.
Change-Id: I1773f55b5b4677dad8f4bea017b6328fd93df20c
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: https://review.coreboot.org/27779
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
- Exclude build flags that generate warnings when scanbuild is running
- Add the SCANBUILD_ARGS variable to abuild so we can pass in arguments
to scanbuild.
- Set the default scanbuild argument to -k (--keep-going) so that even
if an error occurs it continues with the scan. This is similar to what
we do with coverity runs.
Change-Id: I82e7c13d7fd7432b43c17a31834ec82fca158a07
Signed-off-by: Martin Roth <gaumless@gmail.com>
Reviewed-on: https://review.coreboot.org/27595
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This config is used to provide the name of a region where a microcode
is located. The address of this will be added as the first entry in
the FIT of the topswap bootblock.
This adds a capability to associate two microcodes for each
of the two bootblocks, this allows for the CPU to boot with different
microcodes with 2 separate bootblocks.
Change-Id: I4ee41d90bae34862aa68c9b8bd69288de1335585
Signed-off-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Reviewed-on: https://review.coreboot.org/27151
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Intel PCH/Southbridges have feature that it is possible
to have the southbridge/PCH look for the bootblock at a 64K or
128K/256K/512K/1MB (in case of newer SoCs) offset instead of
the usual top of flash.
Add configs to create a second bootblock and configure its size.
Change-Id: I4bbd19c35871891b762a0673f840858d972e129e
Signed-off-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Signed-off-by: Aamir Bohra <aamir.bohra@intel.com>
Reviewed-on: https://review.coreboot.org/22533
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The `files-in-dir` macro is supposed to return all files (out of a
given set) that reside directly (non-recursive) in a given directory.
While the current solution worked splendidly, we can achieve the same
without recursive macros that look at each parent dir individually.
Beside providing better readability, this also fixes a future make
error, as make doesn't like the variable name ` ` anymore ;)
Change-Id: Iac0eacdf91b8b5098592ad301c1f3fdb632454e9
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/27324
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Clang doesn't know the warning `packed-not-aligned`, so only add it
when GCC is used.
error: unknown warning option '-Wno-packed-not-aligned'; did you \
mean '-Wno-over-aligned'? [-Werror,-Wunknown-warning-option]
Change-Id: I86ee12a12fc24a0b8b92c4a0e665103ee4c4003d
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: https://review.coreboot.org/26879
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This change adds a new config option OVERRIDE_DEVICETREE that allows
variants to provide an override devicetree file to override the
registers and/or add new devices on top of the ones provided by
baseboard devicetree using CONFIG_DEVICETREE.
BUG=b:80081934
Change-Id: Ica046b7e0d70d0f1e8d94da714d1e62032277916
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/26691
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Scan-build refuses to run if the -fconserve-stack flag is added to
cflags. It fails with the cryptic message "could not find clang line".
Change-Id: Ib1b56ef7d217138a1a195fe993d8e8dd965bd855
Signed-off-by: Martin Roth <gaumless@gmail.com>
Reviewed-on: https://review.coreboot.org/26878
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Masked ROMs are the silent killers of boot speed on devices without
memory-mapped SPI flash. They often contain awfully slow SPI drivers
(presumably bit-banged) that take hundreds of milliseconds to load our
bootblock, and every extra kilobyte of bootblock size has a hugely
disproportionate impact on boot speed. The coreboot timestamps can never
show that component, but it impacts our users all the same.
This patch tries to alleviate that issue a bit by allowing us to
compress the bootblock with LZ4, which can cut its size down to nearly
half. Of course, masked ROMs usually don't come with decompression
algorithms built in, so we need to introduce a little decompression stub
that can decompress the rest of the bootblock. This is done by creating
a new "decompressor" stage which runs before the bootblock, but includes
the compressed bootblock code in its data section. It needs to be as
small as possible to get a real benefit from this approach, which means
no device drivers, no console output, no exception handling, etc.
Besides the decompression algorithm itself we only include the timer
driver so that we can measure the boot speed impact of decompression. On
ARM and ARM64 systems, we also need to give SoC code a chance to
initialize the MMU, since running decompression without MMU is
prohibitively slow on these architectures.
This feature is implemented for ARM and ARM64 architectures for now,
although most of it is architecture-independent and it should be
relatively simple to port to other platforms where a masked ROM loads
the bootblock into SRAM. It is also supposed to be a clean starting
point from which later optimizations can hopefully cut down the
decompression stub size (currently ~4K on RK3399) a bit more.
NOTE: Bootblock compression is not for everyone. Possible side effects
include trying to run LZ4 on CPUs that come out of reset extremely
underclocked or enabling this too early in SoC bring-up and getting
frustrated trying to find issues in an undebuggable environment. Ask
your SoC vendor if bootblock compression is right for you.
Change-Id: I0dc1cad9ae7508892e477739e743cd1afb5945e8
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/26340
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This patch adds the CCACHE_EXTRAFILES variable to the list of exported
environment variables, which can be useful as a target-specific variable
to make ccache aware of extra dependencies that it cannot figure out on
its own. It also adds a CPPFLAGS parameter to define the __BUILD_DIR__
constant for the preprocessor so that the current output build directory
can be referenced in C code if necessary.
Change-Id: I4fdd08842972cfed8ef5e5a61ebf859c0571bcfb
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/26451
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This patch moves the objcopy invocation that changes the bootblock's
section flags to make sure .data and .bss are preserved in the binary
image from the generation of bootblock.raw.bin into a separate
bootblock.raw.elf file. Some SoCs (like SDM845) like to have an ELF
rather than a raw binary as input to their masked ROM wrapper
generation script.
Also move those objcopy flags out into a variable because I'll need them
again in a later patch.
Change-Id: I9557b184df7f753a442c7e0ceb58e81c5e19f2c5
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/26338
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
We use packed structs with unaligned members all the time, which is the
entire point of us using the packed attribute.
Change-Id: Ib26b422ba83257d1a7f26134ee20217fad5823cd
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/25996
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
As I mention in the comment, this is valid ASL, which was added as
a warning with the comment "This would appear to be worthless in
real-world ASL code." While code using empty resource templates
could probably be rewritten, this seems like an arbitrary choice
to generate this as a warning, since it's valid.
This gets rid of warnings such as this one:
dsdt.aml 2975: Return (ResourceTemplate() {})
Warning 3150 - Empty Resource Template (END_TAG only)
Which is generated by this code in google/rambi/acpi/mainboard.asl:
Method (_CRS)
{
/* Only return interrupt if I2C1 is PCI mode */
If (LEqual (\S1EN, 0)) {
Return (^RBUF)
}
/* Return empty resource template otherwise */
Return (ResourceTemplate() {})
}
Change-Id: I9cfe9069c738a284aa85feada9d58e1aee97e433
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/26352
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The `-t` argument was never required for `add-payload` and results in
a warning now because the type was renamed.
TEST=Built with BUILD_TIMELESS=1 and compared binaries with and without
this patch.
Change-Id: I6ccb70acc6e88a602b90c625040d4f05d8e3630a
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/26323
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Allow bootblock to get access to the entire static device tree
as other stages can access independently.
TEST=SMM code now can access devicetree.cb variables.
Change-Id: I59537c16f0a459e48d8b1efb5c1b196302f13381
Signed-off-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-on: https://review.coreboot.org/23823
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Currently, at the end of a build `CBFSPRINT` prints the content of all
CBFS regions. This is confusing, as they are identical. To avoid
confusion print the layout beforehand.
> layout [-w] – List mutable (or, with -w, readable) image regions
Change-Id: Ibf03b125ef6dae41c58b8ae867430047778cfff3
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reviewed-on: https://review.coreboot.org/22143
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Both GRUB and SeaBIOS can chainload lzma compressed payloads.
Therefore it is beneficial to compress secondary payloads
like Memtest86+, coreinfo, nvramcui,... for both size reasons and
often also speed reasons since the limiting factor is generally the
IO of the boot device.
Tested with SeaBIOS and memtest86+ master on Thinkpad X220.
Change-Id: Iddfd6fcf4112d255cc7b2b49b99bf5ea4d6f8db4
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/23639
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
Jacob Garber