Remove the arch check for each stage as the arch for different stages can be
different based on the SoC. e.g.: Rush has arm32-based romstage whereas
arm64-based ramstage
BUG=None
BRANCH=None
TEST=Compiles successfully for nyan, link and rush
Original-Change-Id: I561dab5a5d87c6b93b8d667857d5e181ff72e35d
Original-Signed-off-by: Furquan Shaikh <furquan@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/205761
Original-Tested-by: Furquan Shaikh <furquan@chromium.org>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org>
Original-Reviewed-by: Ronald Minnich <rminnich@chromium.org>
(cherry picked from commit 6a6a87b65fcab5a7e8163258c7e8d704fa8d97c3)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ic412d60d8a72dac4f9807cae5d8c89499a157f96
Reviewed-on: http://review.coreboot.org/8179
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
For arm64, the machine type is arm64 in cbfstool, however it was displayed as
aarch64 in help message. This patch corrects it.
BUG=None
BRANCH=None
TEST=None
Original-Change-Id: I0319907d6c9d136707ed35d6e9686ba67da7dfb2
Original-Signed-off-by: Furquan Shaikh <furquan@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/204379
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Tested-by: Furquan Shaikh <furquan@chromium.org>
(cherry picked from commit 1f5f4c853efac5d842147ca0373cf9b5dd9f0ad0)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I00f51f1d4a9e336367f0619910fd8eb965b69bab
Reviewed-on: http://review.coreboot.org/8144
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
In https://chromium-review.googlesource.com/181272 the payload->type has been
changed to big-endian (network ordering) but the cbfs_image is still parsing
type as host ordering, which caused printing cbfs image verbosely
(cbfstool imge print -v) to fail to find entry field and print numerous
garbage output.
Payload fields should be always parsed in big-endian (network ordering).
BUG=none
TEST=make; cbfstool image.bin print -v -v -v # see payloads correctly
Original-Change-Id: If1ac355b8847fb54988069f694bd2f317ce49a1a
Original-Signed-off-by: Hung-Te Lin <hungte@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/200158
Original-Reviewed-by: Ronald Minnich <rminnich@chromium.org>
(cherry picked from commit 423f7dd28f8b071692d57401e144232d5ee2e479)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: I5a4694e887c7ff48d8d0713bb5808c29256141a9
Reviewed-on: http://review.coreboot.org/8005
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
Fixes regression caused by commit 405304ac
(cbfstool: Add option to ignore section in add-stage)
Change-Id: If9e3eea9ab2c05027f660d0057a635abf981b901
Signed-off-by: Francis Rowe <info@gluglug.org.uk>
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: http://review.coreboot.org/7545
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Allow add-stage to have an optional parameter for ignoring any section. This is
required to ensure proper operation of elf_to_stage in case of loadable segments
with zero filesize.
Change-Id: I49ad62c2a4260ab9cec173c80c0f16923fc66c79
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: http://review.coreboot.org/7304
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Change cbfs-mkstage to use parsed elf instead of calling elf_headers. That
allows us to have access to the complete elf including the string table.
Change-Id: Ie767d28bdf41af38d1df0bce54bc0ada45123136
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: http://review.coreboot.org/7303
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Add support for creating ARM rmodules. There are 3 expected
relocations for an ARM rmodule:
- R_ARM_ABS32
- R_ARM_THM_PC22
- R_ARM_THM_JUMP24
R_ARM_ABS32 is the only type that needs to emitted for relocation
as the other 2 are relative relocations.
BUG=chrome-os-partner:27094
BRANCH=None
TEST=Built vbootstub for ARM device.
Original-Change-Id: I0c22d4abca970e82ccd60b33fed700b96e3e52fb
Original-Signed-off-by: Aaron Durbin <adurbin@chromuim.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/190922
Original-Reviewed-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit a642102ba7ace5c1829abe7732199eda6646950a)
Signed-off-by: Marc Jones <marc.jones@se-eng.com>
Change-Id: Ib3b3c90ebb672d8d6a537df896b97dc82c6186cc
Reviewed-on: http://review.coreboot.org/7204
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
The LZMA compression algorithm, currently the only one available, will fail
if you ask it to write more data to the output than you've given it space for.
The code that calls into LZMA allocates an output buffer the same size as the
input, so if compression increases the size of the output the call will fail.
The caller(s) were written to assume that the call succeeded and check the
returned length to see if the size would have increased, but that will never
happen with LZMA.
Rather than try to rework the LZMA library to dynamically resize the output
buffer or try to guess what the maximal size the data could expand to is, this
change makes the caller simply print a warning and disable compression if the
call failed for some reason.
This may lead to images that are larger than necessary if compression fails
for some other reason and the user doesn't notice, but since compression
errors were ignored entirely until very recently that will hopefully not be
a problem in practice, and we should be guaranteed to at least produce a
correct image.
Original-Change-Id: I5f59529c2d48e9c4c2e011018b40ec336c4fcca8
Original-Signed-off-by: Gabe Black <gabeblack@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/187365
Original-Reviewed-by: David Hendricks <dhendrix@chromium.org>
Original-Tested-by: Gabe Black <gabeblack@chromium.org>
Original-Commit-Queue: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit b9f622a554d5fb9a9aff839c64e11acb27785f13)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Change-Id: I5f59529c2d48e9c4c2e011018b40ec336c4fcca8
Reviewed-on: http://review.coreboot.org/6958
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
When compression fails for whatever reason, the caller should know about it
rather than blindly assuming it worked correctly. That can prevent half
compressed data from ending up in the image.
This is currently happening for a segment of depthcharge which is triggering
a failure in LZMA. The size of the "compressed" data is never set and is
recorded as zero, and that segment effectively isn't loaded during boot.
Change-Id: Idbff01f5413d030bbf5382712780bbd0b9e83bc7
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://chromium-review.googlesource.com/187364
Reviewed-by: Hung-Te Lin <hungte@chromium.org>
Tested-by: Gabe Black <gabeblack@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit be48f3e41eaf0eaf6686c61c439095fc56883cec)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6960
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Currently, rmodules with 0 relocations are not allowed. Fix this by skipping
addition of .rmodules section on 0 relocs.
Change-Id: I7a39cf409a5f2bc808967d2b5334a15891c4748e
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: http://review.coreboot.org/6774
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@google.com>
Tested-by: build bot (Jenkins)
Add support for enabling different coreboot stages (bootblock, romstage and
ramstage) to have arm64 architecture. Most of the files have been copied over
from arm/ or arm64-generic work.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Reviewed-on: https://chromium-review.googlesource.com/197397
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Commit-Queue: Furquan Shaikh <furquan@chromium.org>
Tested-by: Furquan Shaikh <furquan@chromium.org>
(cherry picked from commit 033ba96516805502673ac7404bc97e6ce4e2a934)
This patch is essentially a squash of aarch64 changes made by
these patches:
d955885 coreboot: Rename coreboot_ram stage to ramstage
a492761 cbmem console: Locate the preram console with a symbol instead of a sect
96e7f0e aarch64: Enable early icache and migrate SCTLR from EL3
3f854dc aarch64: Pass coreboot table in jmp_to_elf_entry
ab3ecaf aarch64/foundation-armv8: Set up RAM area and enter ramstage
25fd2e9 aarch64: Remove CAR definitions from early_variables.h
65bf77d aarch64/foundation-armv8: Enable DYNAMIC_CBMEM
9484873 aarch64: Change default exception level to EL2
7a152c3 aarch64: Fix formatting of exception registers dump
6946464 aarch64: Implement basic exception handling
c732a9d aarch64/foundation-armv8: Basic bootblock implementation
3bc412c aarch64: Comment out some parts of code to allow build
ab5be71 Add initial aarch64 support
The ramstage support is the only portion that has been tested
on actual hardware. Bootblock and romstage support may require
modifications to run on hardware.
Change-Id: Icd59bec55c963a471a50e30972a8092e4c9d2fb2
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6915
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
The aarch64 is not really an arm variant, it's sufficiently
different that it can be considered (for purposes of cbfs, certainly)
to be a new architecture.
Add a constant in cbfs.h and strings to correspond to it.
Note that with the new cbfstool support that we added earlier,
the actual use of aarch64 ELF files actually "just works" (at
least when tested earlier).
Change-Id: Ib4900900d99c9aae6eef858d8ee097709368c4d4
Reviewed-on: https://chromium-review.googlesource.com/180221
Reviewed-by: Stefan Reinauer <reinauer@chromium.org>
Commit-Queue: Ronald Minnich <rminnich@chromium.org>
Tested-by: Ronald Minnich <rminnich@chromium.org>
(cherry picked from commit f836e14695827b2667804bc1058e08ec7b297921)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6896
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
In the process of rewriting cbfstool for ARM and using
a new internal API a regression was introduced that would
silently let you add an ARM payload into an x86 CBFS image
and the other way around. This patch fixes cbfstool to
produce an error in that case again.
Change-Id: I37ee65a467d9658d0846c2cf43b582e285f1a8f8
Signed-off-by: Stefan Reinauer <reinauer@google.com>
Reviewed-on: https://chromium-review.googlesource.com/176711
Reviewed-by: Hung-Te Lin <hungte@chromium.org>
Commit-Queue: Stefan Reinauer <reinauer@chromium.org>
Tested-by: Stefan Reinauer <reinauer@chromium.org>
(cherry picked from commit 8f74f3f5227e440ae46b59f8fd692f679f3ada2d)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6879
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
There are ARM systems which are essentially heterogeneous multicores where
some cores implement a different ARM architecture version than other cores. A
specific example is the tegra124 which boots on an ARMv4 coprocessor while
most code, including most of the firmware, runs on the main ARMv7 core. To
support SOCs like this, the plan is to generalize the ARM architecture so that
all versions are available, and an SOC/CPU can then select what architecture
variant should be used for each component of the firmware; bootblock,
romstage, and ramstage.
Old-Change-Id: I22e048c3bc72bd56371e14200942e436c1e312c2
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://chromium-review.googlesource.com/171338
Reviewed-by: Gabe Black <gabeblack@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
Tested-by: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 8423a41529da0ff67fb9873be1e2beb30b09ae2d)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
ARM: Split out ARMv7 code and make it possible to have other arch versions.
We don't always want to use ARMv7 code when building for ARM, so we should
separate out the ARMv7 code so it can be excluded, and also make it possible
to include code for some other version of the architecture instead, all per
build component for cases where we need more than one architecture version
at a time.
The tegra124 bootblock will ultimately need to be ARMv4, but until we have
some ARMv4 code to switch over to we can leave it set to ARMv7.
Old-Change-Id: Ia982c91057fac9c252397b7c866224f103761cc7
Reviewed-on: https://chromium-review.googlesource.com/171400
Reviewed-by: Gabe Black <gabeblack@chromium.org>
Tested-by: Gabe Black <gabeblack@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 799514e6060aa97acdcf081b5c48f965be134483)
Squashed two related patches for splitting ARM support into general
ARM support and ARMv7 specific pieces.
Change-Id: Ic6511507953a2223c87c55f90252c4a4e1dd6010
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6782
Tested-by: build bot (Jenkins)
The Linux trampoline code does not set up the segment descriptors for
__BOOT_CS and __BOOT_DS as described in the Linux kernel
documentation:
... a GDT must be loaded with the descriptors for selectors
__BOOT_CS(0x10) and __BOOT_DS(0x18); both descriptors must be 4G
flat segment; __BOOT_CS must have execute/read permission, and
__BOOT_DS must have read/write permission;
This is not a problem when launching a Linux payload from coreboot, as
coreboot configures the segment descriptors at selectors 0x10 and
0x18. Coreboot configures these selectors in the ramstage to match
what the Linux kernel expects (see
coreboot/src/arch/x86/lib/c_start.S).
When the cbfs payload is launched in other environments, SeaBIOS for
example, the segment descriptors are configured differently and the
cbfs Linux payload does not work.
If the cbfs Linux payload is to be used in multiple environments
should the trampoline needs to take care of the descriptors that Linux
requires.
This patch updates the Linux trampoline code to configure the 4G flat
descriptors that Linux expects. The configuration is borrowed from
the descriptor configs in coreboot/src/arch/x86/lib/c_start.S for
selectors 0x10 and 0x18.
The linux_trampoline code is slightly refractored by defining the
trampoline entry address, 0x40000, as TRAMPOLINE_ENTRY_LOC. This
definition is moved into a separate header file, linux_trampoline.h.
This header file is now included by both the trampoline assembly
language code and the trampoline loader C code.
The trampoline assembly language code can now use TRAMPOLINE_ENTRY_LOC
as scratch space for the sgdt CPU instruction.
Testing Done:
Verified the Linux payload is booted correctly in the following
environments:
1. Coreboot -> Linux Payload
2. Coreboot -> SeaBIOS -> Linux Payload: (previously did not work)
Change-Id: I888f74ff43073a6b7318f6713a8d4ecb804c0162
Signed-off-by: Curt Brune <curt@cumulusnetworks.com>
Reviewed-on: http://review.coreboot.org/6796
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Apparently when I originally wrote this I confused myself to no end.
The code/data of an rmodule has a set memory size which is associated
with the .payload section. The relocation entries may increase the
overall footprint of the memory size if the rmodule has no bss but
a lot of relocations. Therefore, just compare relocation entries size
plus the file size of the .payload section with the memory size of the
paylod section. The .empty section is added only when we have not met
the final target size.
Change-Id: I5521dff048ae64a9b6e3c8f84a390eba37c7d0f5
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/6767
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Furquan Shaikh <furquan@google.com>
It's a more direct approach to get the file size.
Change-Id: If49df26bf4996bd556c675f3a673d0003b4adf89
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/6594
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
The process probably terminates not much later, but in
case anyone reuses the function in something with
longer life-time, free unused resources.
Change-Id: I10c471ee3d9dc9a3ebf08fe4605f223ea59b990e
Found-by: Coverity Scan
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/6559
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
"cbfstool create -B bootblock -s size" (in this order)
would break bootblock selection.
Change-Id: I9a9f5660827c8bf60dae81b519c6f026f3aaa0f3
Found-by: Coverity Scan
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/6564
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
The printing routines of the cbfs_payload_segment assumed the type
could be accessed in host order. Each of the fields need to be
converted to the host order before inspecting the fields. In addition,
this removes all the ntoh*() calls while processing the
cbfs_payload_segment structures.
cbfstool would crash adding entries or just printing entries
containing a payload when -v was passed on the command line.
Change-Id: Iff41c64a99001b9e3920e2e26828c5fd6e671239
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/6498
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <gaumless@gmail.com>
Fix memory leak found by scan-build from clang version 3.2-11.
Change-Id: Id8f9db46cf42012a0eb0a632c9d83a4eec1989a2
Signed-off-by: Daniele Forsi <dforsi@gmail.com>
Reviewed-on: http://review.coreboot.org/6379
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
It just doesn't work to have files depend on their parent
directory: As soon as the files are written, the time stamp
of the directory changes, too.
This led to spurious updates of cbfstool and rmodtool, and
related "permission denied" errors when linker and build
system ran into each other.
Change-Id: I44a7d7b4b1d47a1567ece1f57dfd6745d05ee651
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/6276
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Spotted by building with Clang.
Change-Id: I7ab97278d8bd586a71e453c8cc9d26dd6938c8d2
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/6139
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Idwer Vollering <vidwer@gmail.com>
cbfstool fails to built under 32-bit platforms since commit
aa2f739a cbfs: fix issues with word size and endianness.
due to the use of '%ld' format specifier on size_t, which on these
platforms is only 32-bit.
No error is seen though, when cbfstool is built, when building a coreboot
image, where it is put in `build/cbfstool`.
Use the length modifier `z` for size_t arguments, and cast to size_t where
appropriate.
Change-Id: Id84a20fbf237376a31f7e4816bd139463800c977
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/5388
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@gmail.com>
For bzImages the trampoline segment is added unconditionally.
However, that segment wasn't properly being accounted for.
Explicitly add the trampoline segments like the other ones.
Change-Id: I74f6fcc2a65615bb87578a8a3a76cecf858fe856
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5702
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The magic number mismatch was introduced by commit a8a133
(Add section header parsing and use it in the mk-payload step).
Change-Id: I73b0adb969816e9d130f19f48e175c57124e2f3a
Signed-off-by: Wei Hu <wei@aristanetworks.com>
Reviewed-on: http://review.coreboot.org/5528
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
This fixes a double free crash that occurs when a call to
cbfs_image_from_file() fails in cbfs_extract() and falls though to
cbfs_image_delete() with a NULL-pointer.
To reproduce the crash pass the following arguments where the files
passed, in fact, do not exist. As follows:
./cbfstool build/coreboot.rom extract -n config -f /tmp/config.txt
Change-Id: I2213ff175d0703705a0ec10271b30bb26b6f8d0a
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: http://review.coreboot.org/5353
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
This change started with tracking down a bug where the trampoline
size was not being taken into account for sizing the output buffer
leading to a heap corruption. I was having a hard time keeping
track of what num_segments actually tracked as well as what parts
were being placed in the output buffer. Here's my attempt at
hopefully providing more clarity.
This change doesn't crash when adding a bzImage:
$ dd if=/dev/zero of=bb.bin bs=64 count=1
$ ./cbfstool tmp.rom create -s 4M -B bb.bin -m x86 -a 64
$ ./cbfstool tmp.rom add-payload -f ~/Downloads/bzImage -C "1" -n
"fallback"/payload
Change-Id: Ib1de1ddfec3c7102facffc5815c52b340fcdc628
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5408
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
The current implementation of creating rmodules relies
on invoking the linker in a certain manner with the
relocations overlaid on the BSS section. It's not really
surprising that the linker doesn't always behave the way
one wants depending on the linker used and the architecture.
Instead, introduce rmodtool which takes an ELF file as an
input, parses it, and creates a new ELF file in the format
the rmodule loader expects.
Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5378
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
In order to generate rmodules in the format of ELF files
there needs to be support for writing out ELF files. The
ELF writer is fairly simple. It accpets sections that can
be associated with an optional buffer (file data). For each
section flagged with SHF_ALLOC a PT_LOAD segment is generated.
There isn't smart merging of the sections into a single PT_LOAD
segment.
Change-Id: I4d1a11f2e65be2369fb3f8bff350cbb28e14c89d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5377
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Optionally parse the symbol table contained within an ELF
file. It currently assumes there is only one symbol table present,
and it errors out if more than one is found.
Change-Id: I4ac4ad03184a319562576d8ab24fa620e701672a
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5376
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Optionally parse the string tables within an ELF file.
Change-Id: I89f9da50b4fcf1fed7ac44f00c60b495c35555ef
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5375
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Optionally parse the relocation entries found within an ELF
file.
Change-Id: I343647f104901eb8a6a997ddf44aa5d36c31b44b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5374
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
In order to make the ELF parsing more flexible introduce
a parse_elf() function which takes a struct parsed_elf
parameter. In addition take a flags parameter which instructs
the ELF parser as to what data within the ELF file should be
parsed.
Change-Id: I3e30e84bf8043c3df96a6ab56cd077eef2632173
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5373
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
I was overzealous in checking the section size with respect
to the file size. That check makes no sense as the section only
deals with link sizes -- not on-disk sizes. Remove the check as
it doesn't make any sense.
Change-Id: I348e7847ae3a50badc22693439614f813462445a
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5384
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
While parsing the section and program headers ensure the
locations of their contents are within the elf file proper.
Change-Id: I856f7de45f82ac15977abc06e51bedb51c58dde1
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5372
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Though the result doesn't matter much, the callers of calloc()
should order the parameters correctly. i.e. the first paramter
is the number of elements in an array and the second is the
size of each element.
Change-Id: Ic7c2910d623d96f380feb4e5f6fa432376f49e9b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5371
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
elfparsing.h serves as the header to working with the elf
parser. Additionally, only include what is needed by the other
files. Many had no reason to be including elf.h aside from fixing
compilation problems when including cbfs.h.
Change-Id: I9eb5f09f3122aa18beeca52d2e4dc2102d70fb9d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5370
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
The only user of iself() was in elfheaders.c. Move it there,
and make it local to the compilation unit.
Change-Id: I0d919ce372f6e2fce75885fb4fcba20d985979b3
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5369
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
The elfheaders code was manipulating struct buffers. Use
the introduced buffer helper functions. Additionally fix
up offset and size checks for the program headers and section
headers by using common code paths.
Change-Id: I279c77f77aaa1860a0be43fb111df890dd1d84d5
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5368
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
There are some open-coded manipulation of the struct buffer
innards in the elf parsing code. Add helper functions to avoid
reaching into the struct itself.
Change-Id: I0d5300afa1a3549f87f588f976184e880d071682
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5367
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
There was already a bgets() function which operates on a buffer to
copy a byte stream. Provide bputs() to store a byte stream to a
buffer, thus making the API symmetrical.
Change-Id: I6166f6b68eacb822da38c9da61a3e44f4c67136d
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5366
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
In order to provide consistent usage provide the get8()
and put8() callbacks to xdr operations. That way no futzing
needs to be done to handle 8-bit reads and writes.
Change-Id: I1233d25df67134dc5c3bbd1a84206be77f0da417
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5365
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
In order for multiple tools to use the common code found
in common.c place the verbose variable within common.c's
compilation unit.
Change-Id: I71660a5fd4d186ddee81b0da8b57ce2abddf178a
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/5364
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Furquan Shaikh