Some distribution compilers enable Position Independent Executable (PIE)
by default, causing a build failure.
So explicitly disable PIE by passing the flag `-fno-pie`, to fix the
build error.
Change-Id: I1b7d7168e34c5c93c25bc03ffa49b2eeac0e76f8
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reviewed-on: https://review.coreboot.org/17097
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
amdfwtool currently gets built for a specific size of ROM chip. This
should be updated to be passed in on the amdfwtool command line, but
until that's done, stop sharing the tool between builds.
This caused a problem for abuild when we tried changing the default
rom to one that used a 256KB rom chip. That wasn't large enough for
all of the files included by amdfwtool on several platforms, causing
build failures.
Change-Id: Ib08f3283e5be956f995a4a416a70b12a32462882
Signed-off-by: Martin Roth <martinroth@chromium.org>
Reviewed-on: https://review.coreboot.org/17070
Tested-by: build bot (Jenkins)
Reviewed-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This patch adds '--includes' option to 'git config --global' command
to allow user name and email to be defined in a file included from
the global gitconfig (~/.gitconfig) file.
BUG=none
BRANCH=none
TEST=make gitconfig with ~/.gitconfig including another file which
defines user.name and email.
Change-Id: I4fe61078b143c3a2e26b0be69c3ca8e6f069d8b0
Signed-off-by: Daisuke Nojiri <dnojiri@chromium.org>
Reviewed-on: https://review.coreboot.org/16912
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
All systems are building with IASL warnings as errors enabled.
Remove the option to disable it.
Remove the notification at the end of the build.
Change-Id: I5c6218c182fdf173b4026fd010d939a5fa36040e
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16606
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Updating submodules seem to give people headaches, so this adds a pair
of git aliases to update them.
'git sup' updates the submodules to the latest versions, but leaves any
locally modified files.
'git sup-destroy' will remove the current submodules and re-initialize
them. This deletes any local changes.
Change-Id: Id62a30d88b3b6d285b3f00555d7609509aa1561f
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16573
Tested-by: build bot (Jenkins)
Reviewed-by: Omar Pakker
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Add a stripped-down version of libgnat. This is somehow comparable to
libgcc but for Ada programs. It's licensed under GPLv3 but with the
runtime library exception. So it's totally fine to link it with our
GPLv2 code and keep it under GPLv2.
Change-Id: Ie6522abf093f0a516b9ae18ddc69131bd721dc0c
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/11836
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Edward O'Callaghan <edward.ocallaghan@koparo.com>
Some remarks on the make process:
o We usually leave Ada specs (.ads files which are like c headers)
together with the bodies (implementations in .adb files) in one
directory. So we have to know, where they live.
o If there is no matching .adb an .ads is a valid source file and
we'll generate an object file from it.
o Object files need to have the same basename as their source files :-/
That's why we put them in build/<class>/ dirs now.
o We track dependencies by looking at the compiler output (.ali files
which accompany every .o). This way we don't need any gnatmake
magic, or even more complex, less portable tools.
For ADAFLAGS_common, I simply copied the CFLAGS_common whilst dropping
everything unsupported and adding sane warning options.
The set of language features is highly restricted (see gnat.adc). This
should suit the embedded nature of coreboot and helps proving absence
of runtime errors with SPARK.
Change-Id: I70df9adbd467ecd2dc7c5c1cf418b7765aca4e93
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/13044
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Edward O'Callaghan <edward.ocallaghan@koparo.com>
Commit 93ef3ff makes the following only print the part number when
the ROM is built. In Makefile.inc, $(MAINBOARDDIR) is the variable
that has the quotes stripped off from $(CONFIG_MAINBOARD_DIR), so
use it instead of $(MAINBOARD_DIR).
build_complete:: coreboot
printf "\nBuilt %s (%s)\n" $(MAINBOARD_DIR) \
$(CONFIG_MAINBOARD_PART_NUMBER)
Change-Id: I729a583182937db7a926eb75aa28dfb53360046c
Signed-off-by: Iru Cai <mytbk920423@gmail.com>
Reviewed-on: https://review.coreboot.org/16410
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
This patch adds functionality to compile a C data structure into a raw
binary file, add it to CBFS and allow coreboot to load it at runtime.
This is useful in all cases where we need to be able to have several
larger data sets available in an image, but will only require a small
subset of them at boot (a classic example would be DRAM parameters) or
only require it in certain boot modes. This allows us to load less data
from flash and increase boot speed compared to solutions that compile
all data sets into a stage.
Each structure has to be defined in a separate .c file which contains no
functions and only a single global variable. The data type must be
serialization safe (composed of only fixed-width types, paying attention
to padding). It must be added to CBFS in a Makefile with the 'struct'
file processor.
Change-Id: Iab65c0b6ebea235089f741eaa8098743e54d6ccc
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16272
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
That's more useful than just COREBOOT for more complex scenarios
Change-Id: I93cd686d698799a3331ca2ea487cd6efb304caa0
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16143
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
It was a band-aid that isn't required any more.
Change-Id: Ib1793ae8fe25eecf9bd5ab8e5feef0d9380b43c2
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16142
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
To override fallback/foo's position or alignment in region BAR, use
fallback/foo-BAR-{position,align} = 0x1234
Like for the global settings, specifying both isn't allowed
because that's rather pointless.
Change-Id: I94f41ebc9f35108267265df4164f23b70e3d0bf6
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16140
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
They're now sorted later in the process after the per-region file lists
are determined.
Change-Id: I0bba381d09dc4b99e2fe5cae16ff7ffcb5b3aa82
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16138
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Make sure that files with a fixed position are placed first (whose order
doesn't matter: either they collide or they don't), then all aligned
files (where we just hope that the right thing happens) and finally the
files with no further requirements (again, hope).
It's still a pretty good heuristic given a typical coreboot image.
The global sorting that happens earlier in the build flow will be
removed in the future to make room for per-region requirements.
Change-Id: I269c00b2ece262c95d310b76a6651c9574badb58
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16137
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Instead of adding each file in all requested regions, sort by region,
then by file.
This is in preparation of per-region file options
(eg. position, alignment)
Change-Id: Ide09a1c8840279380294a059bbd5d2f9f0cba780
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/16130
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
The variable MAINBOARD_DIR already has the quotes stripped off.
Change-Id: Ib434ce92bdbc49180fb3f713b26d65ba4cf8c441
Signed-off-by: Martin Roth <martinroth@chromium.org>
Reviewed-on: https://review.coreboot.org/16117
Tested-by: build bot (Jenkins)
Reviewed-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Minor change - Instead of stripping the quotes from CONFIG_DEVICETREE
inline, add it to the location where we normalize all the other Kconfig
variables.
Change-Id: Idbc58179c7b45160afef7d7e44f9b3b334f8c4a7
Signed-off-by: Martin Roth <martinroth@chromium.org>
Reviewed-on: https://review.coreboot.org/16116
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The mainboard chip.h files were (mostly) removed long ago.
Change-Id: I1d5a9381945427c96868fa17756e6ecabb1048b2
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16080
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Instead of having directories and file names hardcoded, pass in the full
path and filename of both the input and output files.
In the makefile, create variables for these values, and use them in
places that previously had the names and paths written out.
Change-Id: Icb6f536547ce3193980ec5d60c786a29755c2813
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/16078
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Instead of forcing the hardcoded 'devicetree.cb' filename under the
mainboard directory, this allows mainboards to select a filename for
the devicetree file.
This allows mainboard variants that need to use different devicetree
files to live under the same directory.
Change-Id: I761e676ba5d5f70d1fb86656b528f63db169fcef
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12529
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The first attempt of providing a options-for-region function to call
to determining a file's cbfstool options would work, but it means there
can only be one instance which has to handle all of the files that may
need an override. That logic can be problematic in impelementation.
Instead, provide a mechanism to target cbfstool options for a given
CBFS region where the implementation is tightly coupled in the build
system to where the file as requested to be added to cbfs. This allows
there to be a base set of cbfstool options while more easily extending
arguments on specific regions.
Example which adds '-b 0x10000' only for the COREBOOT CBFS region:
cbfs-files-y += file.bin
file.bin-COREBOOT-cbfstool-opts := -b 0x10000
Change-Id: Idfafb0205be42768adb04bb0a30fe46a9ca1bd57
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/14640
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
With VBOOT_VERIFY_FIRMWARE separated from CHROMEOS, move recovery and
developer mode check functions to vboot. Thus, get rid of the
BOOTMODE_STRAPS option which controlled these functions under src/lib.
BUG=chrome-os-partner:55639
Change-Id: Ia2571026ce8976856add01095cc6be415d2be22e
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/15868
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
VBOOT_VERIFY_FIRMWARE should be independent of CHROMEOS. This allows use
of verified boot library without having to stick to CHROMEOS.
BUG=chrome-os-partner:55639
Change-Id: Ia2c328712caedd230ab295b8a613e3c1ed1532d9
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/15867
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@chromium.org>
This renames the VB_SOURCE variable to VBOOT_SOURCE in the build system,
providing increased clarity about what it represents.
Since the submodule itself is called "vboot", it makes sense to use that
name in full instead of a very shortened (and confusing) version of it.
Change-Id: Ib343b6642363665ec1205134832498a59b7c4a26
Signed-off-by: Paul Kocialkowski <contact@paulk.fr>
Reviewed-on: https://review.coreboot.org/15824
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Allow bootblock to get access to the static device tree like
other early stages. device_romstage.c was renamed to
device_simple.c to better articulate the usage since it's not
just being used in romstage.
BUG=chrome-os-partner:55357
Change-Id: I3d63d2754c737cc738c09a3e3b3b468362fb78d1
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/15837
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Subrata Banik <subrata.banik@intel.com>
Tested-by: build bot (Jenkins)
Currently, on Intel Skylake the uCode binary is added to
CBFS based on the config option CBFS_EXTERNAL_HEADER. But
the entry is missing into the Firmware Interface Table, so
add it there.
BRANCH=none
BUG=chrome-os-partner:55403, chrome-os-partner:53077
TEST=built and verified FIT table has ucode entry.
Change-Id: I7dd7459ff7d2468f0aff66eb3ee9c2e3d7eda501
Signed-off-by: Subrata Banik <subrata.banik@intel.com>
Reviewed-on: https://review.coreboot.org/15783
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Martin Roth <martinroth@google.com>
Currently there are two sources for the final size of the
flash image. One is defined as a Kconfig variable
(ROM_SIZE) and the other can be provided in a user defined
flashmap.fmd. This patch will enable the usage of CONFIG_ROM_SIZE
in flashmap.fmd to define the flash size. In this way, the
Kconfig variable is the only source of information for the
flash image size. This way is optional.
Change-Id: Id5298e06d360aaa6d94f2b5a2ffa65e45919853e
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/15219
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Build the <board>_checklist.html file which contains a checklist table
for each stage of coreboot. This processing builds a set of implemented
(done) routines which are marked green in the table. The remaining
required routines (work-to-do) are marked red in the table and the
optional routines are marked yellow in the table. The table heading
for each stage contains a completion percentage in terms of count of
routines (done .vs. required).
Add some Kconfig values:
* CREATE_BOARD_CHECKLIST - When selected creates the checklist file
* MAKE_CHECKLIST_PUBLIC - Copies the checklist file into the
Documenation directory
* CHECKLIST_DATA_FILE_LOCATION - Location of the checklist data files:
* <stage>_complete.dat - Lists all of the weak routines
* <stage>_optional.dat - Lists weak routines which may be optionally
implemented
TEST=Build with Galileo Gen2.
Change-Id: Ie056f8bb6d45ff7f3bc6390b5630b5063f54c527
Signed-off-by: Lee Leahy <leroy.p.leahy@intel.com>
Reviewed-on: https://review.coreboot.org/15011
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Add ifwitool to list of tools to be built so that it can be used by the
build system.
Change-Id: Ifcfbfd87ad9b7ba3ea11cfbcf40894f3e0dae694
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/15013
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The previous code harded the suffix of splash file as
"jpg". Actually, SeaBIOS supports both jpg and bmp.
Change-Id: I06c4b14aae7f75be3406652a94612b5f30ce91c2
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: https://review.coreboot.org/14932
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
On modern x86 platforms like apollolake, pre-RAM stages verstage and
romstage run within the cache-as-ram region. Thus, we do not need to
pass in the --xip parameter to cbfstool while adding these
stages. Introduce a new Kconfig variable NO_XIP_EARLY_STAGES which is
default false for all x86 platforms. Apollolake selects this option
since it supports code execution with CAR.
Change-Id: I2848046472f40f09ce7fc230c258b0389851b2ea
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/14623
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Previously, the XIP_ROM_SIZE Kconfig variable is used globally on
x86 platforms with the assumption that all chipsets utilize this
value. For the chipsets which do not use the variable it can lead
to unnecessary alignment constraints in cbfs for romstage. Therefore,
allow those chipsets a path to not be burdened by not passing
'-P $(XIP_ROM_SIZE)' to cbfstool when adding romstage.
Change-Id: Id8692df5ecec116a72b8e5886d86648ca959c78b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/14625
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
This fixes UPDATE_IMAGE builds, assuming that the fmap configuration in
the tree didn't change, at least as far as the CBFS regions are
concerned.
Another option would be to synthesize the fmap related files from the
existing image, but that comes with other issues (eg. what about
updating images old enough that there is no fmap?) and is more complex,
so keep it simple, stupid for now.
Change-Id: I036dab9f81f524f7d70bc0029b1ef835e6180a53
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/14601
Tested-by: build bot (Jenkins)
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Commit 785a31d67e
(Makefile.inc: Move payload code to payloads/) breaks the usage of
Linux kernel as payload. The reason for it is that cbfs-files-y is
evaluated before payloads/external/Makefile.inc is sourced and as a
consequence ADDITIONAL_PAYLOAD_CONFIG is empty when it is used for
payload options. That leads to missing command line and initrd for
the kernel which in turn leads to kernel panic when it boots.
To avoid it, move the code which adds payload to cbfs completely to
payloads/extranal/Makefile.inc. This way, ADDITIONAL_PAYLOAD_CONFIG is
set right before the payload itself is added to cbfs-files-y.
I have tested this patch with a Linux kernel as well as with SeaBIOS as
payload on mc_tcu3 and it works. If someone sees impact to other
payloads just let me know.
Change-Id: I7aad352f8b3fc1fdba1875b12648b07eba14e282
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/14579
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
There's an in-tree fmap.h, and the file generated by fmaptool is likely
used in tandem with it. To avoid problems, rename the generated file
(which so far isn't used).
Change-Id: I95dfde513a7f78677cf18ecd7ce8745e40af316b
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/14570
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reorder drivers to fit src/drivers/[X]/[Y]/ scheme to make
them pluggable.
Also, fix up the following driver subdirectories by switching
to the src/drivers/[X]/[Y]/ scheme as these are hard requirements
for the main change:
* drivers/intel
* drivers/pc80
* drivers/dec
Change-Id: I455d3089a317181d5b99bf658df759ec728a5f6b
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/14047
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Currently, the coreboot makefiles repeatedly run git to try to set
the KERNELVERSION variable and to fetch the submodules. This happens
three times for every build. By exporting a variable, we can catch
this on recursive makes and not run each of these steps again.
Change-Id: I85ab867b40e80c36bd94d48510ffe3252c6cf93f
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/14392
Tested-by: build bot (Jenkins)
Reviewed-by: Alexander Couzens <lynxis@fe80.eu>
This should help catch cases where the AML is not correct.
Change-Id: I48efb9ed0b62b3e17dcf3045ef9c32d813a412bc
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/14340
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
We already have the ability to add a pxe rom to cbfs, but it needs to be
configured and built separately.
This moves the existing Kconfig options for PXE from device/Kconfig and
the top level Makefile.inc to payloads, and adds the option to download
and build iPXE as part of the coreboot build process.
This configures the serial output of iPXE to match coreboot's serial
port configuration by editing the .h files. iPXE doesn't give any
real build-time method of setting these configuration options.
Change-Id: I3d77b2c6845b7f5f644440f6910c3b4533a0d415
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/14085
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Certain chipsets don't have a memory-mapped boot media
so their code execution for stages prior to DRAM initialization
is backed by SRAM or cache-as-ram. The postcar stage/phase
handles the cache-as-ram situation where in order to tear down
cache-as-ram one needs to be executing out of a backing
store that isn't transient. By current definition, cache-as-ram
is volatile and tearing it down leads to its contents disappearing.
Therefore provide a shim layer, postcar, that's loaded into
memory and executed which does 2 things:
1. Tears down cache-as-ram with a chipset helper function.
2. Loads and runs ramstage.
Because those 2 things are executed out of ram there's no issue
of the code's backing store while executing the code that
tears down cache-as-ram. The current implementation makes no
assumption regarding cacheability of the DRAM itself. If the
chipset code wishes to cache DRAM for loading of the postcar
stage/phase then it's also up to the chipset to handle any
coherency issues pertaining to cache-as-ram destruction.
Change-Id: Ia58efdadd0b48f20cfe7de2f49ab462306c3a19b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/14140
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Add a few additional macros that can be used throughout the coreboot
makefiles.
tolower: returns the value in all lowercase
toupper: returns the value in all uppercase
ws_to_under: returns the value with any whitespace changed to underscores
Change-Id: Icd0e6586481d8f229af0e899e0c94ef7c5c672c3
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/14093
Tested-by: build bot (Jenkins)
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
On non-x86 platforms, coreboot uses the memlayout.ld mechanism to
statically allocate the different memory regions it needs and guarantees
at build time that there are no dangerous overlaps between them. At the
end of its (ramstage) execution, however, it usually loads a payload
(and possibly other platform-specific components) that is not integrated
into the coreboot build system and therefore cannot provide the same
overlap guarantees through memlayout.ld. This creates a dangerous memory
hazard where a new component could be loaded over memory areas that are
still in use by the code-loading ramstage and lead to arbitrary memory
corruption bugs.
This patch fills this gap in our build-time correctness guarantees by
adding the necessary checks as a new intermediate Makefile target on
route to assembling the final image. It will parse the memory footprint
information of the payload (and other platform-specific post-ramstage
components) from CBFS and compare it to a list of memory areas known to
be still in use during late ramstage, generating a build failure in case
of a possible hazard.
BUG=chrome-os-partner:48008
TEST=Built Oak while moving critical regions in the way of BL31 or the
payload, observing the desired build-time errors. Built Nyan, Jerry and
Falco without issues for good measure.
Change-Id: I3ebd2c1caa4df959421265e26f9cab2c54909b68
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/13949
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Paul Menzel