cbfstool has a routine to deal with old images that may encourage it to
overwrite the master header. That routine is triggered for
"cbfstool add-master-header" prepared images even though these are not
at risk, and - worse - destroys the chain structure (through a negative
file length), so avoid touching such images.
Change-Id: I9d0bbe3e6300b9b9f3e50347737d1850f83ddad8
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/13672
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
While assembling CBFS images within the RW slots on Chrome OS
machines the current approach is to 'cbfstool copy' from the
RO CBFS to each RW CBFS. Additional fixups are required such
as removing unneeded files from the RW CBFS (e.g. verstage)
as well as removing and adding back files with the proper
arguments (FSP relocation as well as romstage XIP relocation).
This ends up leaving holes in the RW CBFS. To speed up RW
CBFS slot hashing it's beneficial to pack all non-empty files
together at the beginning of the CBFS. Therefore, provide
the 'compact' command which bubbles all the empty entries to
the end of the CBFS.
Change-Id: I8311172d71a2ccfccab384f8286cf9f21a17dec9
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/13479
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
In order to more easily process the output of 'cbfstool print'
with other tools provide a -k option which spits out the
tab-separated header and fields:
Name Offset Type Metadata Size Data Size Total Size
ALIGN_UP(Offset + Total Size, 64) would be the start
of the next entry. Also, one can analzye the overhead
and offsets of each file more easily.
Example output (note: tabs aren't in here):
$ ./coreboot-builds/sharedutils/cbfstool/cbfstool test.serial.bin print
-r FW_MAIN_A -k
Performing operation on 'FW_MAIN_A' region...
Name Offset Type Metadata Size Data Size Total Size
cmos_layout.bin 0x0 cmos_layout 0x38 0x48c 0x4c4
dmic-2ch-48khz-16b.bin 0x500 raw 0x48 0xb68 0xbb0
dmic-2ch-48khz-32b.bin 0x10c0 raw 0x48 0xb68 0xbb0
nau88l25-2ch-48khz-24b.bin 0x1c80 raw 0x48 0x54 0x9c
ssm4567-render-2ch-48khz-24b.bin 0x1d40 raw 0x58 0x54 0xac
ssm4567-capture-4ch-48khz-32b.bin 0x1e00 raw 0x58 0x54 0xac
vbt.bin 0x1ec0 optionrom 0x38 0x1000 0x1038
spd.bin 0x2f00 spd 0x38 0x600 0x638
config 0x3540 raw 0x38 0x1ab7 0x1aef
revision 0x5040 raw 0x38 0x25e 0x296
font.bin 0x5300 raw 0x38 0x77f 0x7b7
vbgfx.bin 0x5ac0 raw 0x38 0x32f8 0x3330
locales 0x8e00 raw 0x28 0x2 0x2a
locale_en.bin 0x8e40 raw 0x38 0x29f6 0x2a2e
u-boot.dtb 0xb880 mrc_cache 0x38 0xff1 0x1029
(empty) 0xc8c0 null 0x64 0xadf4 0xae58
fallback/ramstage 0x17740 stage 0x38 0x15238 0x15270
(empty) 0x2c9c0 null 0x64 0xd2c4 0xd328
fallback/payload 0x39d00 payload 0x38 0x12245 0x1227d
cpu_microcode_blob.bin 0x4bf80 microcode 0x60 0x17000 0x17060
(empty) 0x63000 null 0x28 0x37cf98 0x37cfc0
Change-Id: I1c5f8c1b5f2f980033d6c954c9840299c6268431
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/13475
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
With the introduction of flashmap cbfs alignment of files gets
broken because flashmap is located at the beginning of the flash
and cbfstool didn't take care about that offset.
This commit fixes the alignment in cbfs.
Change-Id: Idebb86d4c691b49a351a402ef79c62d31622c773
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/13417
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Instead of people open coding the offset field access within a
struct buffer provide buffer_offset() so that the implementation
can change if needed without high touch in the code base.
Change-Id: I751c7145687a8529ab549d87e412b7f2d1fb90ed
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/13468
Tested-by: build bot (Jenkins)
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
gcc 4.4.7 fails to compile due to the missing initializers
for all struct members. Add initializers for all fields.
Change-Id: If1ad4fff0f965ccd7e821820c0703853c1e5c590
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/13418
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
There can be an error when a cbfs file is added aligned or as
xip-stage and hashing of this file is enabled. This commit
resolves this error. Though adding a file to a fixed position
while hashing is used can still lead to errors.
Change-Id: Icd98d970891410538909db2830666bf159553133
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/13136
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Add functionality to cbfstool to generate file attributes
for position and alignment constraints. This new feature
can be activated with the -g option and will generate,
once the option has been enabled, additional attributes
for the files where position, xip or alignment was specified.
Change-Id: I3db9bd2c20d26b168bc7f320362ed41be349ae3a
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/12967
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
cbfs_add calculated a base address out of the alignment specification
and stored it in param.baseaddress.
This worked when every cbfstool invocation only added a single file, but
with -r REGION1,REGION2,... multiple additions can happen.
In that case, the second (and later) additions would have both alignment
and baseaddress set, which isn't allowed, aborting the process.
Change-Id: I8c5a512dbe3c97e08c5bcd92b5541b58f65c63b3
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/13063
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The CBFS flag in fmd files isn't stored in the fmap, so allow storing it
out of band using the -R option.
Change-Id: I342772878d7f8ce350de1a32dc7b2a5b07d6617d
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/13058
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
On Apple OS X, the ntohl and htonl need including header,
#include <arpa/inet.h>
Please refer the manpage for these command on OS X,
https://developer.apple.com/library/mac/documentation/Darwin/Reference/ManPages/man3/htonl.3.html
Change-Id: Ia942c58f34637c18222fbf985b93c48abf63c5b8
Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: https://review.coreboot.org/11672
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
cbfstool tries opening the input file for write access even if the
command does not require modifying the file.
Let's not request write access unless it is necessary, this way one
can examine write protected files without sudo.
BRANCH=none
BUG=none
TEST=running
cbfstool /build/<board>/firmware/image.bin print
in chroot does not require root access any more.
Change-Id: Ic4e4cc389b160da190e44a676808f5c4e6625567
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: ef6a8e25d9e257d7de4cc6b94e510234fe20a56d
Original-Change-Id: I871f32f0662221ffbdb13bf0482cb285ec184d07
Original-Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/317300
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/12931
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Instead of looking for an FMAP at every byte, only search down
to a granularity of 16 bytes, reducing the time for a cbfstool
call by 0.3s when no FMAP is found.
Signed-off-by: Stefan Reinauner <reinauer@chromium.org>
BUG=none
BRANCH=none
TEST=time ./cbfstool coreboot.rom add -f locale_de.bin -n locale_de.bin -t 0x50 -c lzma
is 0.3s faster than before.
Change-Id: Icb4937330e920ae09928ceda7c1af6a3c5130ac7
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: bc92d838ba9db7733870ea6e8423fa4fa41bf8fe
Original-Change-Id: Idbaec58a199df93bdc10e883c56675b419ab5b8e
Original-Reviewed-on: https://chromium-review.googlesource.com/317321
Original-Commit-Ready: Stefan Reinauer <reinauer@chromium.org>
Original-Tested-by: Stefan Reinauer <reinauer@chromium.org>
Original-Reviewed-by: Daisuke Nojiri <dnojiri@chromium.org>
Reviewed-on: https://review.coreboot.org/12932
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
In parse_elf_to_stage(), it uses 32-bit variable to handle address.
The correct address type is Elf64_Addr. Use uint64_t to prevent address
to be truncated.
BUG=none
TEST=emerge-oak coreboot
BRANCH=none
Change-Id: I1abcd16899a69b18dd10e9678e767b0564b2846e
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: ebc1aae0ae4ca30802a80a4a4e2ae0c0dad4d88a
Original-Change-Id: I21f8057ddf13e442f1cf55da6702c3449ba0cc35
Original-Reviewed-on: https://chromium-review.googlesource.com/292553
Original-Commit-Ready: Yidi Lin <yidi.lin@mediatek.com>
Original-Tested-by: Yidi Lin <yidi.lin@mediatek.com>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/12927
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
hashcbfs was spliced in a line early, mixing up 'extract' and 'cbfshash'
help texts.
Change-Id: I86d4edb9eec0685a290b2dd4c2dc45d3611eba9a
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/12922
Tested-by: build bot (Jenkins)
Reviewed-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Remove duplicate line which sets baseaddress parameter.
Change-Id: Idfbb0297e413344be892fa1ecc676a64d20352bf
Signed-off-by: Werner Zeh <werner.zeh@siemens.com>
Reviewed-on: https://review.coreboot.org/12904
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The cbfs master header's offset and romsize fields are absolute values
within the boot media proper. Therefore, when adding a master header
provide the offset of the CBFS region one is operating on as well as
the absolute end offset (romsize) to match expectations.
Built with and without CBFS_SIZE != ROM_SIZE on x86 and ARM device. Manually
inspected the master headers within the images to confirm proper caclulations.
Change-Id: Id0623fd713ee7a481ce3326f4770c81beda20f64
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/12825
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
For the purposes of maintaining integrity of a CBFS allow one to
hash a CBFS over a given region. The hash consists of all file
metadata and non-empty file data. The resulting digest is saved
to the requested destination region.
BUG=chrome-os-partner:48412
BUG=chromium:445938
BRANCH=None
TEST=Integrated with glados chrome os build. vboot verification
works using the same code to generate the hash in the tooling
as well as at runtime on the board in question.
Change-Id: Ib0d6bf668ffd6618f5f73e1217bdef404074dbfc
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/12790
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Adding new files overwrote the header with the empty file (ie 0xff),
so carve out some space.
BUG=chromium:445938
BRANCH=none
TEST=none
Change-Id: I91c292df381c2bac41c6cb9dda74dae99defd81d
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/12789
Tested-by: build bot (Jenkins)
These need to go together, so the commit became a bit larger than
typial.
- Add an option -R for the copy source fmap region.
Use: cbfstool copy -r target-region -R source-region.
- Don't generate a CBFS master header because for fmap regions, we
assume that the region starts with a file header.
Use cbfstool add-master-header to add it afterwards, if necessary.
- Don't copy files of type "cbfs master header" (which are what cbfstool
add-master-header creates)
- Leave room for the master header pointer
- Remove -D command line option as it's no longer used.
BUG=chromium:445938
BRANCH=none
TEST=Manual test on image and integration test w/ bundle_firmware
changes.
CQ-DEPEND=CL:313770,CL:313771
Change-Id: I2a11cda42caee96aa763f162b5f3bc11bb7992f9
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/12788
Tested-by: build bot (Jenkins)
This allows adding support for FMAP based cbfstool copy more easily.
BUG=chromium:445938
Change-Id: I72e7bc4da7d27853e324400f76f86136e3d8726e
Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-on: https://review.coreboot.org/12787
Tested-by: build bot (Jenkins)
The constant for ppc64 is 'hotstuff'. For many reasons.
Note that line 2894 of elf.h is not indented. This is because in the
original the line begins with a space. Checkpatch rejects that.
Checkpatch also rejects changing the space to a tab because that makes
it more than 80 chars. I rejected breaking the line because it makes it
even less readable. All the changes forced by checkpatch make the code
less readable.
Herman Hollerith would be proud.
Change-Id: I21f049fe8c655a30f17dff694b8f42789ad9efb7
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: https://review.coreboot.org/12711
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Tested-by: build bot (Jenkins)
We still add a master header for compatibility purposes, and the default
layouts don't cover anything non-coreboot (eg. IFD regions) yet.
The default layouts can be overridden by specifying an fmd file, from
which the fmap is generated.
Future work:
- map IFD regions to fmap regions
- non-x86: build minimalistic trampolines that jump into the first cbfs
file, so the bootblock can be part of CBFS instead of reserving a
whole 64K for it.
- teach coreboot's cbfs code to work without the master header
- teach coreboot's cbfs code to work on different fmap regions
Change-Id: Id1085dcd5107cf0e02e8dc1e77dc0dd9497a819c
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/11692
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
The former interpretation sprung from the x86 way of doing things
(assuming top-alignment to 4GB). Extend the mechanism to work with CBFS
regions residing elsewhere.
It's compatible with x86 because the default region there resides at the
old location, so things fall in place. It also makes more complex
layouts and non-x86 layouts work with negative base addresses.
Change-Id: Ibcde973d85bad5d1195d657559f527695478f46c
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/12683
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
The help text had gotten kind of sloppy. There was a missing newline
in the add-stage command, some of the lines were too long, etc.
Change-Id: If7bdc519ae062fb4ac6fc67e6b55af1e80eabe33
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12646
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Alexander Couzens <lynxis@fe80.eu>
ERROR: code indent should use tabs where possible
+^I trampoline_len);$
Change-Id: If46f977e2e07d73e6cfd3038912a172236a7e571
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/12620
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
It's not needed, so we can remove some extra file mangling, too.
Change-Id: I80d707708e70c07a29653258b4cb6e9cd88d3de3
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/12508
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Add the code necessary to create the linux trampoline blob.
Don't enforce this for the in-coreboot build or use objcopy
to produce linux_trampoline.o as it is a bit trickier to get
all the details right than I had hoped:
- you have to know the elf architecture of the host machine
- you might have to have more tools (xxd, perl, etc) installed
Change-Id: I9b7877c58d90f9fb21d16e0061a31e19fffa2470
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/12505
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
This makes the make process look like the one inside
of coreboot's build system.
Change-Id: I48be2df39cad47644e16ce583b27c33a1da81fc3
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/12509
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
If HOSTCC=clang, the -Wtautological-constant-out-of-range-compare is
set automaticaaly. That assume the value of type enum is in the defined
range. Then testing if a type enum is out of range causes build error.
Error:
coreboot/util/cbfstool/cbfs_image.c:1387:16: error:
comparison of constant 4 with expression of type 'enum vb2_hash_algorithm'
is always false [-Werror,-Wtautological-constant-out-of-range-compare]
if (hash_type >= CBFS_NUM_SUPPORTED_HASHES)
~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.
clang version:
FreeBSD clang version 3.4.1 (tags/RELEASE_34/dot1-final 208032) 20140512
Target: x86_64-unknown-freebsd10.2
Thread model: posix
Change-Id: I3e1722bf6f9553793a9f0c7f4e790706b6938522
Signed-off-by: zbao <fishbaozi@gmail.com>
Reviewed-on: http://review.coreboot.org/12330
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
When using FMAP regions (with option -r) that were generated with a
master header (as done by cbfstool copy, eg. in Chrome OS' build
system), there were differences in interpretation of the master header's
fields.
Normalize for that by not sanity-checking the master header's size field
(there are enough other tests) and by dealing with region offsets
properly.
BUG=chromium:445938
BRANCH=tot
TEST=`cbfstool /build/veyron_minnie/firmware/image.dev.bin print -r
FW_MAIN_A` shows that region's directory (instead of claiming that
there's no CBFS at all, or showing an empty directory).
Change-Id: Ia840c823739d4ca144a7f861573d6d1b4113d799
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: 0e5364d291f45e4705e83c0331e128e35ab226d3
Original-Change-Id: Ie28edbf55ec56b7c78160000290ef3c57fda0f0e
Original-Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/312210
Original-Commit-Ready: Patrick Georgi <pgeorgi@chromium.org>
Original-Tested-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12416
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This is required to handle certain relative-to-flash-start offsets.
BUG=none
BRANCH=tot
TEST=none
Change-Id: I8b30c7b532e330af5db4b8ed65b21774c6cbbd25
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: 596ba1aaa62aedb2b214ca55444e3068b9cb1044
Original-Change-Id: Idc9a5279f16951befec4d84aab35117988f7edb7
Original-Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/312220
Original-Commit-Ready: Patrick Georgi <pgeorgi@chromium.org>
Original-Tested-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12415
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
It encourages users from writing to the FSF without giving an address.
Linux also prefers to drop that and their checkpatch.pl (that we
imported) looks out for that.
This is the result of util/scripts/no-fsf-addresses.sh with no further
editing.
Change-Id: Ie96faea295fe001911d77dbc51e9a6789558fbd6
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/11888
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Those may collide with strings.h's index(), included transitively
through system headers.
Change-Id: I6b03236844509ea85cfcdc0a37acf1df97d4c5f3
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12279
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
With the previous ELF stage extract support the resulting
ELF files wouldn't handle rmodules correctly in that the
rmodule header as well as the relocations were a part of
the program proper. Instead, try an initial pass at
converting the stage as if it was an rmodule first. If it
doesn't work fall back on the normal ELF extraction.
TEST=Pulled an rmodule out of Chrome OS shellball. Manually
matched up the metadata and relocations.
Change-Id: Iaf222f92d145116ca4dfaa955fb7278e583161f2
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12222
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
In order to convert rmodules back into ELF files one needs
to add in the relocations so they can be converted back to
rmodules. Because of that requirement symbol tables need
to be present because the relocations reference the symbols.
Additionally, symbol tables reference a string table for the
symbol names. Provide the necessary support for adding all
of those things to an ELF writer.
TEST=Extracted rmodule from a cbfs and compared with the
source ELF file. Confirmed relocations and code sizes
are correct.
Change-Id: I07e87a30b3371ddedabcfc682046e3db8c956ff2
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12221
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Instead of creating a loadable segment for each section with
SHF_ALLOC flag merge those sections into a single program
segment. This makes more tidy readelf, but it also allows
one to extract an rmodule into an ELF and turn it back into
an rmodule.
TEST=Extracted both regular stages and rmodule stages. Compared
against original ELF files prior to cbfs insert.
Change-Id: I0a600d2e9db5ee6c11278d8ad673caab1af6c759
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12220
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Instead of dumping the raw stage data when cbfstool
extract is used on stage create an equivalent ELF file.
Because there isn't a lot of information within a stage
file only a rudimentary ELF can be created.
Note: this will break Chrome OS' current usage of extract
since the file is no longer a cbfs_stage. It's an ELF file.
TEST=Extracted romstage from rom.
Change-Id: I8d24a7fa4c5717e4bbba5963139d0d9af4ef8f52
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12219
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
In order for one to extract ELF files from cbfs it's
helpful to have common code which creates a default
executable ELF header for the provided constraints.
BUG=None
TEST=With follow up patch am able to extract out romstage
as an ELF file.
Change-Id: Ib8f2456f41b79c6c0430861e33e8b909725013f1
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12218
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
In order to prepare allowing for one to extract a stage
into an ELF file provide an optional -m ARCH option. This
allows one to indicate to cbfstool what architecture type
the ELF file should be in.
Longer term each stage and payload will have an attribute
associated with it which indicates the attributes of
the executable.
Change-Id: Id190c9719908afa85d5a3b2404ff818009eabb4c
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12217
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
In order to actually do something useful with the
resulting file after being extracted decompress stage
files' content. That way one can interrogate the
resulting file w/o having to decompress on the fly.
Note: This change will cause an unexpected change to
Chrome OS devices which package up individual stage
files in the RW slots w/o using cbfs. The result will
be that compressed stages are now decompressed.
Longer term is to turn these files into proper ELF
files on the way out.
Change-Id: I373ecc7b924ea21af8d891a8cb8f01fd64467360
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12174
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Currently cbfs stage files that are compressed do not have
the decompressed size readily available. Therefore there's
no good way to know actual size of data after it is
decompressed. Optionally return the decompressed data size
if requested.
Change-Id: If371753d28d0ff512118d8bc06fdd48f4a0aeae7
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12173
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
If one wants to use buffer_init() for initializing a
struct buffer all the fields should be initialized.
Change-Id: I791c90a406301d662fd333c5b65b2e35c934d0f7
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/12172
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This patch fixes compilation of cbfstool on Cygwin.
As reported in http://review.coreboot.org/#/c/10027
cbfstool on Cygwin likes to be compiled with -D_GNU_SOURCE.
That patch was abandoned because it would unwantedly turn on
more GNU extensions. Instead of doing that, only enable the
define on Cygwin, switch to -std=gnu99 instead of -std=c99 to
make fileno and strdup actually available.
A MINGW32 check that was forgotten in Makefile was copied over
from Makefile.inc to keep the two files in sync.
This patch has no impact on non-Windows builds.
Change-Id: I068b181d67daf9c7280110e64aefb634aa20c69b
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/11667
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Tested-by: build bot (Jenkins)
Cygwin complains:
cbfstool.c: 1075:5 error: array subscript has type 'char' [-Werror=char-subscripts]
so add an explicit cast.
Change-Id: Ie89153518d6af2bacce3f48fc7952fee17a688dd
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/11666
Tested-by: build bot (Jenkins)
Reviewed-by: Zheng Bao <zheng.bao@amd.com>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Currently, cbfstool regressed that removing a file from CBFS the space
is marked as empty but the filename is still shown, preventing adding a
file with the same name again. [1]
```
$ echo a > a
$ echo b > b
$ ./util/cbfstool/cbfstool test.rom create -m x86 -s 1024
Created CBFS (capacity = 920 bytes)
$ ./util/cbfstool/cbfstool test.rom add -f a -n a -t raw
$ ./util/cbfstool/cbfstool test.rom add -f b -n b -t raw
$ cp test.rom test.rom.original
$ ./util/cbfstool/cbfstool test.rom remove -n
$ diff -up <(hexdump -C test.rom.original) <(hexdump -C test.rom)
--- /dev/fd/63 2015-08-07 08:43:42.118430961 -0500
+++ /dev/fd/62 2015-08-07 08:43:42.114430961 -0500
@@ -1,4 +1,4 @@
-00000000 4c 41 52 43 48 49 56 45 00 00 00 02 00 00 00 50 |LARCHIVE.......P|
+00000000 4c 41 52 43 48 49 56 45 00 00 00 02 ff ff ff ff |LARCHIVE........|
00000010 00 00 00 00 00 00 00 28 61 00 00 00 00 00 00 00 |.......(a.......|
00000020 00 00 00 00 00 00 00 00 61 0a ff ff ff ff ff ff |........a.......|
00000030 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
$ ./util/cbfstool/cbfstool test.rom add -f c -n c -t raw
$ ./util/cbfstool/cbfstool test.rom print
test.rom: 1 kB, bootblocksize 0, romsize 1024, offset 0x0
alignment: 64 bytes, architecture: x86
Name Offset Type Size
c 0x0 raw 2
b 0x40 raw 2
(empty) 0x80 null 792
```
So it is “deteled” as the type changed. But the name was not changed to
match the *(empty)* heuristic.
So also adapt the name when removing a file by writing a null byte to
the beginning of the name, so that the heuristic works. (Though remove
doesn't really clear contents.)
```
$ ./util/cbfstool/cbfstool test.rom remove -n c
$ ./util/cbfstool/cbfstool test.rom print
test.rom: 1 kB, bootblocksize 0, romsize 1024, offset 0x0
alignment: 64 bytes, architecture: x86
Name Offset Type Size
(empty) 0x0 null 2
b 0x40 raw 2
(empty) 0x80 null 792
```
[1] http://www.coreboot.org/pipermail/coreboot/2015-August/080201.html
Change-Id: I033456ab10e3e1b402ac2374f3a887cefd3e5abf
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-on: http://review.coreboot.org/11632
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Patrick Georgi