The placement calculation logic in cbfs_add_component() has become quite
a mess, and this patch can only fix that to a limited degree. The
interaction between all the different pathways of how the `offset`
variable can be set and at what point exactly the final placement offset
is decided can get quite convoluted. In particular, one existing problem
is that the offset for a file added with the --align flag is decided
before the convert() function is called, which may change the form (and
thereby the size) of the file again after its location was found --
resulting in a location that ends up being too small, or being unable to
find a location for a file that should fit. This used to be okay under
the assumption that forced alignment should really only be necessary for
use cases like XIP where the file is directly "used" straight from its
location on flash in some way, and those cases can never be compressed
-- however, recent AMD platforms have started using the --align flag to
meet the requirements of their SPI DMA controller and broken this
assumption.
This patch fixes that particular problem and hopefully eliminates a bit
of the convolution by moving the offset decision point in the --align
case after the convert() step. This is safe when the steps in-between
(add_topswap_bootblock() and convert() itself) do not rely on the
location having already been decided by --align before that point. For
the topswap case this is easy, because in practice we always call it
with --base-address (and as far as I can tell that's the only way it was
ever meant to work?) -- so codify that assumption in the function. For
convert() this mostly means that the implementations that do touch the
offset variable (mkstage and FSP) need to ensure they take care of the
alignment themselves. The FSP case is particularly complex so I tried to
rewrite the code in a slightly more straight-forward way and clearly
document the supported cases, which should hopefully make it easier to
see that the offset variable is handled correctly in all of them. For
mkstage the best solution seems to be to only have it touch the offset
variable in the XIP case (where we know compression must be disabled, so
we can rely on it not changing the file size later), and have the extra
space for the stage header directly taken care of by do_cbfs_locate() so
that can happen after convert().
NOTE: This is changing the behavior of `cbfstool add -t fsp` when
neither --base-address nor --xip are passed (e.g. FSP-S). Previously,
cbfstool would implicitly force an alignment of 4K. As far as I can tell
from the comments, this is unnecessary because this binary is loaded
into RAM and CBFS placement does not matter, so I assume this is an
oversight caused by accidentally reusing code that was only meant for
the XIP case.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: Ia49a585988f7a74944a6630b77b3ebd79b3a9897
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59877
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <patrick@coreboot.org>
Reviewed-by: Raul Rangel <rrangel@chromium.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
endian.h wasn't included (although it probably came in as an indirect
include) but in some header sets _XOPEN_SOURCE overrides _DEFAULT_SOURCE
whereas the latter is a super set of the former:
We should get the same things as with _XOPEN_SOURCE (such as memccpy for
which it has been defined) but also extra features like htole32.
Change-Id: Iaee7495b2ae64fdc719ae0879ea95fe7df286212
Signed-off-by: Patrick Georgi <patrick@coreboot.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59891
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Rob Barnes <robbarnes@google.com>
Commit 796aeeba96 (util/cse_fpt: Add a new tool for managing Intel CSE
FPT binaries) and commit d7fb6a90e1 (util/cse_serger: Add a new tool
for stitching CSE components) add two utilities, and building cbfstool
also generates executables for them. When building cbfstool standalone,
these executables are placed in `util/cbfstool/`, and Git should never
track them.
Specify these executables' file names in .gitignore in order to prevent
unintentional inclusion of these files in commits, which is very likely
to happen when using `git add` on directories.
Change-Id: I285a4d7aeee642822eaae2eb69e5d52efb4bc8c0
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/59670
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch makes all accesses to the FMAP fields explicitly little endian.
It fixes issue where build on BE host produced different binary image than
on LE.
Signed-off-by: Marek Kasiewicz <marek.kasiewicz@3mdeb.com>
Signed-off-by: Krystian Hebel <krystian.hebel@3mdeb.com>
Change-Id: Ia88c0625cefa1e594ac1849271a71c3aacc8ce78
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55039
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Add events for Chrome OS diagnostics in eventlog tool:
* ELOG_TYPE_CROS_DIAGNOSTICS(0xb6): diagnostics-related events
* ELOG_CROS_LAUNCH_DIAGNOSTICS(0x01): sub-type for diagnostics boot
These events are not added anywhere currently. They will be added in
another separate commit.
Signed-off-by: Hsuan Ting Chen <roccochen@chromium.org>
Change-Id: I1b67fdb46f64db33f581cfb5635103c9f5bbb302
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58795
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jack Rosenthal <jrosenth@chromium.org>
Reviewed-by: Yu-Ping Wu <yupingso@google.com>
This CL adds a python test for elogtool.
It tests the basic functionality of elogtool: list, clear and add.
A future CL will include more complex tests.
BUG=b:172210863
TEST=pytest elogtool_test.py
Change-Id: If1241ad070d1c690c84f5ca61c0487ba27c2a287
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57869
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jack Rosenthal <jrosenth@chromium.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
When using `DECLARE_OPTIONAL_REGION`, it is assumed that
REGION_SIZE(name) == 0 if the region was not defined in the memlayout.
When using non-rmodule stages (i.e., bootblock, romstage, etc), this
assumption holds true, but breaks down in rmodule (i.e., ramstage)
stages.
The rmodule tool is not currently omitting undefined externals from the
relocation table. e.g.,
extern u8 _##name##_size[];
This means that when the rmodule loader runs, it will rewrite the symbol
from 0 (which is the default the linker assumed) to 0 + offset. This is
wrong since the symbol doesn't actually exist. Instead we need to omit
the relocation so it continues to keep the default value of 0.
BUG=b:179699789
TEST=Print out REGION_SIZE(cbfs_cache) in ramstage and verify it is set
to 0.
I also see the following printed by the rmodtool now:
DEBUG: Omitting relocation for undefined extern: _watchdog_tombstone_size
DEBUG: Omitting relocation for undefined extern: _watchdog_tombstone
DEBUG: Omitting relocation for undefined extern: _watchdog_tombstone
DEBUG: Omitting relocation for absolute symbol: _stack_size
DEBUG: Omitting relocation for absolute symbol: _program_size
DEBUG: Omitting relocation for absolute symbol: _cbmem_init_hooks_size
DEBUG: Omitting relocation for absolute symbol: _payload_preload_cache_size
DEBUG: Omitting relocation for absolute symbol: _payload_preload_cache
DEBUG: Omitting relocation for absolute symbol: _payload_preload_cache_size
DEBUG: Omitting relocation for absolute symbol: _payload_preload_cache
DEBUG: Omitting relocation for undefined extern: _cbfs_cache
DEBUG: Omitting relocation for undefined extern: _cbfs_cache_size
As you can see the _watchdog_tombstone will also be fixed by this CL.
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: Ib57e263fa9014da4f6854637000c1c8ad8eb351a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58376
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Karthik Ramasubramanian <kramasub@google.com>
This change adds a new command `create-cse-region` to cse_serger tool
which takes as inputs offset:size and file for different CSE
partitions and generates the entire CSE region image.
BUG=b:189177186
Change-Id: Ib087f5516e5beb6390831ef4e34b0b067d3fbc8b
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58215
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This change replaces `struct cse_layout_regions` with an array of
`struct region` and introduces enums for DP and BP[1-4]. This makes it
easier to loop over the different regions in following changes.
BUG=b:189177186
Change-Id: If3cced4506d26dc534047cb9c385aaa9418d8522
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58214
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This change moves `read_member` and `write_member` helper functions
out of cse_fpt.c and cse_serger.c into cse_helpers.c to avoid
duplication.
BUG=b:189177186,b:189167923
Change-Id: I7b646b29c9058d892bb0fc9824ef1b4340d2510c
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58201
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This change adds a new tool `cse_serger` which can be used to print,
dump and stitch together different components for the CSE region.
BUG=b:189177186
Change-Id: I90dd809b47fd16afdc80e66431312721082496aa
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55503
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This change adds a new tool `cse_fpt` which can be used to print and
dump CSE partitions in Flash Partition Table (FPT) format.
BUG=b:189167923
Change-Id: I93c8d33e9baa327cbdab918a14f2f7a039953be6
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55259
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
This change adds a helper function `buffer_from_file_aligned_size`
that loads a file into memory buffer by creating a memory buffer of
size rounded up to the provided `size_granularity` parameter.
BUG=b:189177186,b:189167923
Change-Id: Iad3430d476abcdad850505ac50e36cd5d5deecb4
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55989
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
UEFI payload is supported on some ARM64 platforms, for example MT8195.
As a result, add MACHINE_TYPE_ARM64 to support ARM SystemReady.
Signed-off-by: Rex-BC Chen <rex-bc.chen@mediatek.com>
Change-Id: I4c0c6e263bd2f518a62ff9db44d72dd31086756a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/58055
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Yu-Ping Wu <yupingso@google.com>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Hung-Te Lin <hungte@chromium.org>
elogtool was setting the timestamp with the wrong value in the month.
This CL fixes that by incrementing the month by one. This is needed
since gmtime() returns the month value starting at 0.
TEST=pytest elogtool_test.py (see next CL in relation chain)
Change-Id: I00f89ed99b049caafba2e47feae3c068245f9021
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57868
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Jack Rosenthal <jrosenth@chromium.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
This CL fixes a compilation error that happens in 32-bit platforms.
This error happens because printf() was using %ld instead of %zu to
print size_t variables.
This CL fixes it.
BUG=b:200608182
TEST=emerge-kevin (ARM 32-bit)
TEST=emerge-eve (Intel 64-bit)
Change-Id: I340e108361c052601f2b126db45caf2e35ee7ace
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57792
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Fagerburg <pfagerburg@chromium.org>
Adds "add" command to elogtool. This command allows adding elog events
manually. It supports event type and, optionally, event data.
If the free buffer space is < 1/4 of the total space, it shrinks the
buffer, making sure that ~1/4 of the free space is available.
BUG=b:172210863
TEST=./elogtool add 0x17 0101
./elogtool add 0x18
Repeated the same tests on buffers that needed to be shrunk.
Change-Id: Ia6fdf4f951565f842d1bff52173811b52f617f66
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57397
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
This function is "extracted" from cmd_clear().
This new function will be called from cmd_add(), and new command that
will be added in a future CL (see CL chain).
Additional minor fixes:
- calls usage() if no valid commands are passed.
- Slightly improves usage() output. Needed for cmd_clear()
BUG=b:172210863
TEST=elogtool clear
Change-Id: I0d8ecc893675758d7f90845282a588d367b55567
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57395
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Adds "clear" command to cbfsutil/elogtool tool.
"clear" clears the RW_ELOG region by using either:
* flashrom if no file is provided
* or using file write if an input file is provided.
The region is filled with ELOG_TYPE_EOL. And a
ELOG_TYPE_LOG_CLEAR event is inserted.
Additionally, it does a minor cleanup to command "list", like:
* use buffer_end()
* add "list" to the cmds struct
* and make elog_read() very similar to elog_write()
Usage:
$ elogtool clear
BUG=b:172210863
TEST=elogtool clear && elogtool list
elogtool clear -f invalid.raw
elogtool clear -f valid.raw
Change-Id: Ia28a6eb34c82103ab078a0841b022e2e5e430585
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/56883
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Jack Rosenthal <jrosenth@chromium.org>
The patch is to fix "Not a usable UEFI firmware volume" issue when
creating CBFS/flash image. This issue is caused by adding FvNameGuid
in UefiPayloadEntry.fdf in EDKII. There is an ext header between header
of Fv and header of PayloadEntry in Fv with FvNameGuid. The ext header
causes the UefiPayloadEntry to be found incorrectly when parsing Fv.
Commit in EDKII: 4bac086e8e007c7143e33f87bb96238326d1d6ba
Bugzila: https://bugzilla.tianocore.org/show_bug.cgi?id=3585
Signed-off-by: Dun Tan <dun.tan@intel.com>
Change-Id: Id063efb1c8e6c7a96ec2182e87b71c7e8b7b6423
Reviewed-on: https://review.coreboot.org/c/coreboot/+/57296
Reviewed-by: Ray Ni <ray.ni@intel.com>
Reviewed-by: King Sumo <kingsumos@gmail.com>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add buffer_end() function to common.h. This function returns a
pointer to the end of the buffer (exclusive).
This is needed by elogtool util. (See the next CL in the chain).
BUG=b:172210863
Change-Id: I380eecbc89c13f5fe5ab4c31d7a4fef97690a791
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/56987
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Add the binary output of the new elogtool to the .gitignore, so that
running "make -C util/cbfstool" keeps the tree clean.
Signed-off-by: Jack Rosenthal <jrosenth@chromium.org>
Change-Id: I806338a4b33abbc3d55e4edef2736c19d56fa005
Reviewed-on: https://review.coreboot.org/c/coreboot/+/56864
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Ricardo Quesada <ricardoq@google.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Add a new tool that that prints elog events.
The tool, as input, accepts either a file with the RW_ELOG contents, or
if the file is not provided it reads the contents of RW_ELOG by calling
the "flashrom" tool.
The tool is based on "mosys eventlog list"[1]. For the moment it only
supports "list", but future commits will add additional functionality.
This commit also adds missing ELOG defines needed for the tool. These
defines are added with the rest of the ELOG defines, in
include/commonlib/bsd/elog.h
The tool is placed inside util/cbfstool. The rationale behind the
decision, is that this tool shares a lot in common with the other tools
located in cbfstool: vboot dependency, shared files like common.o and
valstr.o, and in spirit is similar to some of the tools located in
cbfstool/.
As an example, you call the tool like the following:
$ elogtool list -f rw_elog_dump.bin
[1]: https://chromium.googlesource.com/chromiumos/platform/mosys/+/refs/heads/main/lib/eventlog/elog.c
BUG=b:172210863
Signed-off-by: Ricardo Quesada <ricardoq@google.com>
Change-Id: Ia1fe1c9ed3c4c6bda846055d4b10943b54463935
Reviewed-on: https://review.coreboot.org/c/coreboot/+/56406
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Jack Rosenthal <jrosenth@chromium.org>
The -a flag was already implemented, it just wasn't exposed for the
add-payload command.
Setting the alignment of the payload will enable using the SPI DMA
controller to read the payload on AMD devices.
BUG=b:179699789
TEST=cbfstool foo.bin add-payload -a 64 ...
Signed-off-by: Raul E Rangel <rrangel@chromium.org>
Change-Id: I9f4aea5f0cbeaa8e761212041099b37f4718ac39
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55973
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin Roth <martinroth@google.com>
The dereferced parameter is never updated so passing a copy would work
too.
Change-Id: Ie36f64f55d4fc7034780116c28aaed65aa304d5e
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55792
Reviewed-by: Julius Werner <jwerner@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Vboot's Makefile is controlled by a ${DEBUG} environment variable.
As the name is very generic, it may be set by accident without any
intention to change the build. Having it set would break reproduci-
bility at least but it also turns out that the hostlib build would
be incomplete so that linking cbfstool fails due to internal calls
to vb2api_fail() which is not built in.
Change-Id: I2a9eb9a645c70451a320c455b8f24bfed197117c
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55764
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
The null-termination of `filetypes` was added after the code was
written, obviously resulting in NULL dereferences. As some more
code has grown around the termination, it's hard to revert the
regression, so let's update the code that still used the array
length.
This fixes commit 7f5f9331d1 (util/cbfstool: fix buffer over-read)
which actually did fix something, but only one path while it broke
two others. We should be careful with fixes, they can always break
something else. Especially when a dumb tool triggered the patching
it seems likely that fewer people looked into related code.
Change-Id: If2ece1f5ad62952ed2e57769702e318ba5468f0c
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55763
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
This fixes a hard to debug hang that could occur in any stage, but in
the end it follows simple rules and is easy to fix.
In long mode the 32bit displacement addressing used on 'mov' and 'lea'
instructions is sign-extended. Those instructions can be found using
readelf on the stage and searching for relocation type R_X86_64_32S.
The sign extension is no issue when either running in protected mode or
the code module and thus the address is below 2GiB. If the address is
greater than 2GiB, as usually the case for code in TSEG, the higher
address bits [64:32] are all set to 1 and the effective address is
pointing to memory not paged. Accessing this memory will cause a page
fault, which isn't handled either.
To prevent such problems
- disable R_AMD64_32S relocations in rmodtool
- add comment explaining why it's not allowed
- use the pseudo op movabs, which doesn't use 32bit displacement addressing
- Print a useful error message if such a reloc is present in the code
Fixes a crash in TSEG and when in long mode seen on Intel Sandybridge.
Change-Id: Ia5f5a9cde7c325f67b12e3a8e9a76283cc3870a3
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55448
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
The e820 type don't fully match the LB_TAG_MEMORY types, so change all
unknown types to e820 to '2', reserved memory.
TESTED with Linuxboot: e820 now shows the CBMEM region as reserved.
Change-Id: Ie0e41c66e002919e41590327afe0f543e0037369
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55074
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Rocky Phagura
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
There is a function to fetch the fit table at both the regular address
and the TS address. So reuse that function instead of attempting to
find the TS fit using some pointer aritmetics that is incorrect.
Change-Id: I9114f5439202ede7e01cd0fcbb1e3c4cdb8698b0
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/54680
Reviewed-by: Rizwan Qureshi <rizwan.qureshi@intel.com>
Reviewed-by: Meera Ravindranath <meera.ravindranath@intel.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Fixes compilation on FreeBSD CURRENT, and possibly other releases.
The compiler, clang, complained about:
util/cbfstool/cbfstool.c:181:40: error: implicit declaration of function 'memmem' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
util/cbfstool/cbfstool.c:181:31: error: incompatible integer to pointer conversion initializing 'struct metadata_hash_anchor *' with an expression of type 'int' [-Werror,-Wint-conversion]
Signed-off-by: Idwer Vollering <vidwer@gmail.com>
Change-Id: I45c02a21709160df44fc8da329f6c4a9bad24478
Reviewed-on: https://review.coreboot.org/c/coreboot/+/53996
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The purpose of this is to eventually move the FIT table out of the
bootblock, generate it separately as a cbfs file and then have the FIT
pointer point to that cbfs file.
TESTED: extracted a FIT table using dd, added it as a cbfs file and see
that the FIT pointer correctly points to it. Also test that trying to
add a non valid FIT cbfs file results in an error.
Change-Id: I6e38b7df31e6b30f75b0ae57a5332f386e00f16b
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/50925
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-by: Christian Walter <christian.walter@9elements.com>
The CBFS stage header is part of the file data (not the header) from
CBFS's point of view, which is problematic for verification: in pre-RAM
environments, there's usually not enough scratch space in CBFS_CACHE to
load the full stage into memory, so it must be directly loaded into its
final destination. However, that destination is decided from reading the
stage header. There's no way we can verify the stage header without
loading the whole file and we can't load the file without trusting the
information in the stage header.
To solve this problem, this patch changes the CBFS stage format to move
the stage header out of the file contents and into a separate CBFS
attribute. Attributes are part of the metadata, so they have already
been verified before the file is loaded.
Since CBFS stages are generally only meant to be used by coreboot itself
and the coreboot build system builds cbfstool and all stages together in
one go, maintaining backwards-compatibility should not be necessary. An
older version of coreboot will build the old version of cbfstool and a
newer version of coreboot will build the new version of cbfstool before
using it to add stages to the final image, thus cbfstool and coreboot's
stage loader should stay in sync. This only causes problems when someone
stashes away a copy of cbfstool somewhere and later uses it to try to
extract stages from a coreboot image built from a different revision...
a debugging use-case that is hopefully rare enough that affected users
can manually deal with finding a matching version of cbfstool.
The SELF (payload) format, on the other hand, is designed to be used for
binaries outside of coreboot that may use independent build systems and
are more likely to be added with a potentially stale copy of cbfstool,
so it would be more problematic to make a similar change for SELFs. It
is not necessary for verification either, since they're usually only
used in post-RAM environments and selfload() already maps SELFs to
CBFS_CACHE before loading them to their final destination anyway (so
they can be hashed at that time).
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I8471ad7494b07599e24e82b81e507fcafbad808a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/46484
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The --alignment flag is currently only handled by cbfstool add, but
there seems little reason to not handle it for all file-adding commands
(the help text actually mentions it for add-stage as well but it doesn't
currently work there). This patch moves the related code (and the
related baseaddress handling) into cbfs_add_component(). As a nice side
effect this allows us to rearrange cbfs_add_component() such that we can
conclusively determine whether we need a hash attribute before trying to
align the file, allowing that code to correctly infer the final header
size even when a hash attribute was implicitly added (for an image built
with CBFS verification enabled).
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: Idc6d68b2c7f30e5d136433adb3aec5a87053f992
Reviewed-on: https://review.coreboot.org/c/coreboot/+/47823
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The 'x' option is not set up in the getopt options.
Change-Id: Ib4aa10b0ea2a3f97e8d2439152b708613bcf43db
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/50923
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
To support the new CONFIG_CBFS_VERIFICATION feature, cbfstool needs to
update the metadata hash embedded in the bootblock code every time it
adds or removes a CBFS file. This can lead to problems on certain
platforms where the bootblock needs to be specially wrapped in some
platform-specific data structure so that the platform's masked ROM can
recognize it. If that data structure contains any form of hash or
signature of the bootblock code that is checked on every boot, it will
no longer match if cbfstool modifies it after the fact.
In general, we should always try to disable these kinds of features
where possible (they're not super useful anyway). But for platforms
where the hardware simply doesn't allow that, this patch introduces the
concept of "platform fixups" to cbfstool. Whenever cbfstool finds a
metadata hash anchor in a CBFS image, it will run all built-in "fixup
probe" functions on that bootblock to check if it can recognize it as
the wrapper format for a platform known to have such an issue. If so, it
will register a corresponding fixup function that will run whenever it
tries to write back modified data to that bootblock. The function can
then modify any platform-specific headers as necessary.
As first supported platform, this patch adds a fixup for Qualcomm
platforms (specifically the header format used by sc7180), which
recalculates the bootblock body hash originally added by
util/qualcomm/createxbl.py.
(Note that this feature is not intended to support platform-specific
signature schemes like BootGuard directly in cbfstool. For anything that
requires an actual secret key, it should be okay if the user needs to
run a platform-specific signing tool on the final CBFS image before
flashing. This feature is intended for the normal unsigned case (which
on some platforms may be implemented as signing with a well-known key)
so that on a board that is not "locked down" in any way the normal use
case of manipulating an image with cbfstool and then directly flashing
the output file stays working with CONFIG_CBFS_VERIFICATION.)
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I02a83a40f1d0009e6f9561ae5d2d9f37a510549a
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41122
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch adds support for the new CONFIG_CBFS_VERIFICATION feature to
cbfstool. When CBFS verification is enabled, cbfstool must automatically
add a hash attribute to every CBFS file it adds (with a handful of
exceptions like bootblock and "header" pseudofiles that are never read
by coreboot code itself). It must also automatically update the metadata
hash that is embedded in the bootblock code. It will automatically find
the metadata hash by scanning the bootblock for its magic number and use
its presence to auto-detect whether CBFS verification is enabled for an
image (and which hash algorithm to use).
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I61a84add8654f60c683ef213b844a11b145a5cb7
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41121
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
cbfstool has always had a CBFS_FILENAME_ALIGN that forces the filename
field to be aligned upwards to the next 16-byte boundary. This was
presumably done to align the file contents (which used to come
immediately after the filename field).
However, this hasn't really worked right ever since we introduced CBFS
attributes. Attributes come between the filename and the contents, so
what this code currently does is fill up the filename field with extra
NUL-bytes to the boundary, and then just put the attributes behind it
with whatever size they may be. The file contents don't end up with any
alignment guarantee and the filename field is just wasting space.
This patch removes the old FILENAME_ALIGN, and instead adds a new
alignment of 4 for the attributes. 4 seems like a reasonable alignment
to enforce since all existing attributes (with the exception of weird
edge cases with the padding attribute) already use sizes divisible by 4
anyway, and the common attribute header fields have a natural alignment
of 4. This means file contents will also have a minimum alignment
guarantee of 4 -- files requiring a larger guarantee can still be added
with the --alignment flag as usual.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I43f3906977094df87fdc283221d8971a6df01b53
Reviewed-on: https://review.coreboot.org/c/coreboot/+/47827
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
In a rare placement edge case when adding a file with alignment
requirements, cbfstool may need to generate a CBFS header that's
slightly larger than it needs to be. The way we do this is by just
increasing the data offset field in the CBFS header until the data falls
to the desired value.
This approach works but it may confuse parsing code in the presence of
CBFS attributes. Normally, the whole area between the attribute offset
and the data offset is filled with valid attributes written back to
back, but when this header expansion occurs the attributes are followed
by some garbage data (usually 0xff). Parsers are resilient against this
but may show unexpected error messages.
This patch solves the problem by moving the attribute offset forwards
together with the data offset, so that the total area used for
attributes doesn't change. Instead, the filename field becomes the
expanded area, which is a closer match to how this worked when it was
originally implemented (before attributes existed) and is less confusing
for parsers since filenames are zero-terminated anyway.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I3dd503dd5c9e6c4be437f694a7f8993a57168c2b
Reviewed-on: https://review.coreboot.org/c/coreboot/+/47824
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
The *location argument to parse_elf_to_stage() is a relic from code all
the way back to 2009 where this function was still used to parse XIP
stages. Nowadays we have a separate parse_elf_to_xip_stage() for that,
so there is no need to heed XIP concerns here. Having a pointer to
represent the location in flash is absolutely irrelevant to a non-XIP
stage, and it is used incorrectly -- we just get lucky that no code path
in cbfstool can currently lead to that value being anything other than
0, otherwise the adjustment of data_start to be no lower than *location
could easily screw things up. This patch removes it.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: Ia7f850c0edd7536ed3bef643efaae7271599313d
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49369
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Memlayout is a mechanism to define memory areas outside the normal
program segment constructed by the linker. Therefore, it generally
doesn't make sense to relocate memlayout symbols when the program is
relocated. They tend to refer to things that are always in one specific
spot, independent of where the program is loaded.
This hasn't really hurt us in the past because the use case we have for
rmodules (ramstage on x86) just happens to not really need to refer to
any memlayout-defined areas at the moment. But that use case may come up
in the future so it's still worth fixing.
This patch declares all memlayout-defined symbols as ABSOLUTE() in the
linker, which is then reflected in the symbol table of the generated
ELF. We can then use that distinction to have rmodtool skip them when
generating the relocation table for an rmodule. (Also rearrange rmodtool
a little to make the primary string table more easily accessible to the
rest of the code, so we can refer to symbol names in debug output.)
A similar problem can come up with userspace unit tests, but we cannot
modify the userspace relocation toolchain (and for unfortunate
historical reasons, it tries to relocate even absolute symbols). We'll
just disable PIC and make those binaries fully static to avoid that
issue.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: Ic51d9add3dc463495282b365c1b6d4a9bf11dbf2
Reviewed-on: https://review.coreboot.org/c/coreboot/+/50629
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Fix regression from commit 0dcc0662f3 util/cbfstool: Introduce
concept of mmap_window.
Use of region_end() wraps around at 4 GiB, if utility is run in
32bit userspace. The build completes with an invalid coreboot.rom,
while one can find error message in stdout or make.log:
E: Host address(ffc002e4) not in any mmap window!
Change-Id: Ib9b6b60c7b5031122901aabad7b3aa8d59f1bc68
Signed-off-by: Furquan Shaikh <furquan@google.com>
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/50618
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Compilation has been broken in commit I022468f6957415ae68a7a7e70428ae6f82d23b06
Adding a missing define solved this. See https://cgit.freebsd.org/src/tree/sys/sys/fcntl.h#n319
Signed-off-by: Idwer Vollering <vidwer@gmail.com>
Change-Id: I3433e4c9269880d3202dd494e5b2e962757a6b87
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49314
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Trying to do multiple operations on the same CBFS image at the same time
likely leads to data corruption. For this reason, add BSD advisory file
locking (flock()) to cbfstool (and ifittool which is using the same file
I/O library), so that only one process will operate on the same file at
the same time and the others will wait in line. This should help resolve
parallel build issues with the INTERMEDIATE target on certain platforms.
Unfortunately, some platforms use the INTERMEDIATE target to do a direct
dd into the CBFS image. This should generally be discouraged and future
platforms should aim to clearly deliminate regions that need to be
written directly by platform scripts with custom FMAP sections, so that
they can be written with `cbfstool write`. For the time being, update
the legacy platforms that do this with explicit calls to the `flock`
utility.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Change-Id: I022468f6957415ae68a7a7e70428ae6f82d23b06
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49190
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Currently, use of the VPD driver to read VPD tables from flash
requires the use of a custom FMAP with one or more VPD regions.
Extend this funtionality to boards using the default FMAP by
creating a dedicated VPD region when the driver is selected.
Test: build qemu target with CONFIG_VPD selected, verify entry
added to build/fmap.fmd.
Change-Id: Ie9e3c7cf11a6337a43223a6037632a4d9c84d988
Signed-off-by: Matt DeVillier <matt.devillier@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/49049
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Michael Niewöhner <foss@mniewoehner.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Julius Werner