After loading compressed files in CBFS, we should check the decompressed
size is equal to the expected size. This might help us detect file
content corruption or compressor/decompressor bugs.
BUG=none
BRANCH=none
TEST=manually (we can still boot into kernel on Kukui, and verify that
loading files from CBFS still works by seeing ChromiumOS firmware
screen).
Change-Id: Ia756cc5477670dd0d1d8aa59d4160ab4233c6795
Signed-off-by: You-Cheng Syu <youcheng@google.com>
Reviewed-on: https://review.coreboot.org/c/31564
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Daisuke Nojiri <dnojiri@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Currently, cbfs_decompress() calls ulzma() and ulz4f() for LZMA/LZ4
decompression. These two functions don't accept input/output size as
parameters. We can make cbfs_decompress more robust by calling ulzman()
and ulz4fn() instead. This could prevent us from overflowing destination
buffer.
BUG=none
BRANCH=none
TEST=boot into kernel on Kukui with COMPRESSED_PAYLOAD_LZMA /
COMPRESSED_PAYLOAD_LZ4.
Change-Id: Ibe617825bd000ed618791d8e3c5f65bbbd5f7e33
Signed-off-by: You-Cheng Syu <youcheng@google.com>
Reviewed-on: https://review.coreboot.org/c/31606
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Daisuke Nojiri <dnojiri@chromium.org>
cbfs_get_handle() allocates memory for a handle and doesn't free it if
it errors out later, leaving the memory permanently leaked. Fix.
Change-Id: Ide198105ce3ad6237672ff152b4490c768909564
Reported-by: Coverity
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/16207
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The libpayload CBFS APIs are pretty old and clunky, primarily because of
the way the cbfs_media struct may or may not be passed in and may be
initialized inside the API calls in a way that cannot be passed back out
again. Due to this, the only real CBFS access function we have always
reads a whole file with all metadata, and everything else has to build
on top of that. This makes certain tasks like reading just a file
attribute very inefficient on non-memory-mapped platforms (because you
always have to map the whole file).
This patch isn't going to fix the world, but will allow a bit more
flexibility by bolting a new API on top which uses a struct cbfs_handle
to represent a found but not yet read file. A cbfs_handle contains a
copy of the cbfs_media needed to read the file, so it can be kept and
passed around to read individual parts of it after the initial lookup.
The existing (non-media) legacy API is retained for backwards
compatibility, as is cbfs_file_get_contents() (which is most likely what
more recent payloads would have used, and also a good convenience
wrapper for the most simple use case), but they are now implemented on
top of the new API.
TEST=Booted Oak, made sure that firmware screens and software sync
worked okay.
Change-Id: I269f3979e77ae691ee9d4e1ab564eff6d45b7cbe
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/14810
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This patch ports the LZ4 decompression code that debuted in libpayload
last year to coreboot for use in CBFS stages (upgrading the base
algorithm to LZ4's dev branch to access the new in-place decompression
checks). This is especially useful for pre-RAM stages in constrained
SRAM-based systems, which previously could not be compressed due to
the size requirements of the LZMA scratchpad and bounce buffer. The
LZ4 algorithm offers a very lean decompressor function and in-place
decompression support to achieve roughly the same boot speed gains
(trading compression ratio for decompression time) with nearly no
memory overhead.
For now we only activate it for the stages that had previously not been
compressed at all on non-XIP (read: non-x86) boards. In the future we
may also consider replacing LZMA completely for certain boards, since
which algorithm wins out on boot speed depends on board-specific
parameters (architecture, processor speed, SPI transfer rate, etc.).
BRANCH=None
BUG=None
TEST=Built and booted Oak, Jerry, Nyan and Falco. Measured boot time on
Oak to be about ~20ms faster (cutting load times for affected stages
almost in half).
Change-Id: Iec256c0e6d585d1b69985461939884a54e3ab900
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/13638
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This change revives the path which was made inert by CL:308520. When
media == CBFS_DEFAULT_MEDIA, cbfs_get_file replaces it with a pointer
to a default media. Thus, get_cbfs_range does not set cbfs offset &
size from lib_sysinfo.
BUG=chrome-os-partner:47772
BRANCH=tot
TEST=Tested on Jerry and Glados
Change-Id: I012f7871336dd24b8eada5c96c4d72117921b0d2
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: 279ba344788b4ba85f500e6cfcca8199af6d0a89
Original-Change-Id: I7f0798881519026a23d0801d0a790332ab878ff0
Original-Signed-off-by: Daisuke Nojiri <dnojiri@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/313205
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/12583
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
CBFS requests were always fulfilled using the CBFS specified in
cbtables. That's a great policy when default requests are sought, but
not so great when the user deliberately asked for something else.
So check if they want default CBFS media information, otherwise ignore
cbtables data.
BUG=chromium:445938
BRANCH=none
TEST=none
Change-Id: I01b63049eebfba6f467808ac84ef77385840c204
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: 621c916ab14c0de4bae3dde09c05060c4f3c63c5
Original-Change-Id: Ia4a8848fd7db9d9a2bf9f5c226566fe3936ff543
Original-Signed-off-by: Patrick Georgi <pgeorgi@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/308520
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/12232
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This change allows libpayload to read cbfs offset and size from sysinfo.
Legacy way of locating cbfs reagion is still supported in case sysinfo
doesn't store the offset and the size.
BUG=none
BRANCH=master
TEST=tested on samus and smaug
Change-Id: I86434fd249467e7c90d59d6b82f0e6c514bc2d05
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: 548a74b7a0758c3f9ba6809425d0fb9c6a5e9d7c
Original-Change-Id: I190d5545a65228483204bf1aa1cbf5a80db31ae0
Original-Signed-off-by: Daisuke Nojiri <dnojiri@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/296993
Original-Commit-Ready: Daisuke Nojiri <dnojiri@google.com>
Original-Tested-by: Daisuke Nojiri <dnojiri@google.com>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/11557
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The error-prone interface of cbfs_get_file_content() led to another
possible NULL dereferencing. So check for CBFS_DEFAULT_MEDIA here
like the other functions do.
Change-Id: Ib8732160d389e9ecceb44f28be0e7de9a1d66e04
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/11796
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
cbfs_file_find_attr(file, tag) finds the first attribute of file with
the given tag.
Change-Id: I78ee3b996b4b086605244c5d7d57ef7e3fc1db47
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/11678
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
Decompression is handled transparently within cbfs_get_file_content:
const char *name = "foo.bmp";
void *dst = cbfs_get_file_content(media, name, type, NULL);
To keep things consistent, a couple of API changes were necessary:
- cbfs_get_file_content always returns a copy of the data, even for
uncompressed files. It's the callers responsibility to free the
memory.
- same for cbfs_load_payload and cbfs_find_file.
- cbfs_load_optionrom doesn't take a "dest" argument anymore but always
returns a copy of the data, for compressed and uncompressed files.
Like with cbfs_get_file_content, the caller is responsible to free it.
It also decompresses based on extended file attributes instead of the
cbfs_optionrom subheader that libpayload specified but that (AFAIK)
nobody ever used, given that there's not even tooling for that.
Change-Id: If959e3dff9b93c6ae45ec7358afcc7840bc17218
Signed-off-by: Daisuke Nojiri <dnojiri@chromium.org>
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/10938
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
The dynamically sized region after struct cbfs_file doesn't contain only
the file name anymore.
Change-Id: I3241cb2f0cbec3fcf4d3c27d638e2847e43f4761
Signed-off-by: Daisuke Nojiri <dnojiri@chromium.org>
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/11676
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This comes from cbfstool (GPL) into libpayload (BSD-l), but I could have
just as well written it in libpayload first.
Change-Id: I86baefe5c299125a4733fa20523efd5d06de7182
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/11675
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
This will make the code work with the different styles
of Kconfig (emit unset bools vs don't emit unset bools)
Roughly, the patch does this, and a little bit of fixing up:
perl -pi -e 's,ifdef (CONFIG_LP_.+?)\b,if IS_ENABLED\($1\),g' `find . -name *.[ch]`
perl -pi -e 's,ifndef (CONFIG_LP_.+?)\b,if !IS_ENABLED\($1\),g' `find . -name *.[ch]`
Change-Id: Ib8a839b056a1f806a8597052e1b571ea3d18a79f
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/10711
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Used command line to remove empty lines at end of file:
find . -type f -exec sed -i -e :a -e '/^\n*$/{$d;N;};/\n$/ba' {} \;
Change-Id: I816ac9666b6dbb7c7e47843672f0d5cc499766a3
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: http://review.coreboot.org/10446
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Non-x86 boards currently need to hardcode the position of their CBFS
master header in a Kconfig. This is very brittle because it is usually
put in between the bootblock and the first CBFS entry, without any
checks to guarantee that it won't overlap either of those. It is not fun
to debug random failures that move and disappear with tiny alignment
changes because someone decided to write "ORBC1112" over some part of
your data section (in a way that is not visible in the symbolized .elf
binaries, only in the final image). This patch seeks to prevent those
issues and reduce the need for manual configuration by making the image
layout a completely automated part of cbfstool.
Since automated placement of the CBFS header means we can no longer
hardcode its position into coreboot, this patch takes the existing x86
solution of placing a pointer to the header at the very end of the
CBFS-managed section of the ROM and generalizes it to all architectures.
This is now even possible with the read-only/read-write split in
ChromeOS, since coreboot knows how large that section is from the
CBFS_SIZE Kconfig (which is by default equal to ROM_SIZE, but can be
changed on systems that place other data next to coreboot/CBFS in ROM).
Also adds a feature to cbfstool that makes the -B (bootblock file name)
argument on image creation optional, since we have recently found valid
use cases for CBFS images that are not the first boot medium of the
device (instead opened by an earlier bootloader that can already
interpret CBFS) and therefore don't really need a bootblock.
BRANCH=None
BUG=None
TEST=Built and booted on Veyron_Pinky, Nyan_Blaze and Falco.
Change-Id: Ib715bb8db258e602991b34f994750a2d3e2d5adf
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: e9879c0fbd57f105254c54bacb3e592acdcad35c
Original-Change-Id: Ifcc755326832755cfbccd6f0a12104cba28a20af
Original-Signed-off-by: Julius Werner <jwerner@chromium.org>
Original-Reviewed-on: https://chromium-review.googlesource.com/229975
Reviewed-on: http://review.coreboot.org/9620
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
A bootblock overwalk was occuring when deriving the actual
length, the bootblock size was not taken into account and bootblock
size was not aligned.
Resolved merge conflict.
Change-Id: I7eb42f8deaaf223dcf07b37bb7dde4643acd508f
Signed-off-by: Steven Sherk <steven.sherk@se-eng.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/65989
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Commit-Queue: Steve Sherk <ssherk70@gmail.com>
Tested-by: Steve Sherk <ssherk70@gmail.com>
(cherry picked from commit 20b0ba479b01755fbdc7f3dd9214e8af923402ba)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6539
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
When libpayload header files are included in the payload itself, it's possible
that the payloads config settings will conflict with the ones in libpayload.
It's also possible for the libpayload config settings to conflict with the
payloads. To avoid that, the libpayload config settings have _LP_ (for
libpayload) added to them. The symbols themselves as defined in the Config.in files
are still the same, but the prefix added to them is now CONFIG_LP_ instead of just
CONFIG_.
Change-Id: Ib8a46d202e7880afdeac7924d69a949bfbcc5f97
Signed-off-by: Gabe Black <gabeblack@google.com>
Reviewed-on: https://gerrit.chromium.org/gerrit/65303
Reviewed-by: Stefan Reinauer <reinauer@google.com>
Tested-by: Gabe Black <gabeblack@chromium.org>
Commit-Queue: Gabe Black <gabeblack@chromium.org>
(cherry picked from commit 23e866da20862cace0ed2a67d6fb74056bc9ea9a)
Signed-off-by: Isaac Christensen <isaac.christensen@se-eng.com>
Reviewed-on: http://review.coreboot.org/6427
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Instead of returning 0 on success and -1 on error, return the decompressed
size of the data on success and 0 on error. The decompressed size is useful
information to have that was being thrown away in that function.
Change-Id: If787201aa61456b1e47feaf3a0071c753fa299a3
Signed-off-by: Gabe Black <gabeblack@chromium.org>
Reviewed-on: http://review.coreboot.org/3578
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
The cbfs core code would print out the name of the file it is
searching for and when it is found would print out the name
again. This contributes to a lot of unnecessary messages in a
functioning payload’s output. Change this message to a DEBUG one
so that it will only be printed when CONFIG_DEBUG_CBFS is enabled.
Change-Id: Ib238ff174bedba8eaaad8d1d452721fcac339b1a
Signed-off-by: Dave Frodin <dave.frodin@se-eng.com>
Reviewed-on: http://review.coreboot.org/3208
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Bruce Griffith <Bruce.Griffith@se-eng.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The cbfs core code would print out all unmatched file
names when searching for a file. This contributes to a lot
of unnecessary messages in the boot log. Change this
message to a DEBUG one so that it will only be printed when
CONFIG_DEBUG_CBFS is enabled.
Change-Id: I34c747e0d3406351318abf70994dbc0bb3fa6c01
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/3164
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Marc Jones <marc.jones@se-eng.com>
Tested-by: build bot (Jenkins)
Change "ERROR" to "WARNING" -- not finding the indicated file is usually
not a fatal error.
Change-Id: I0600964360ee27484c393125823e833f29aaa7e7
Signed-off-by: Shawn Nematbakhsh <shawnn@google.com>
Reviewed-on: http://review.coreboot.org/2833
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Cherry-picking CBFS fix from http://review.coreboot.org/#/c/2292/
For x86, the old CBFS search behavior was to bypass bootblock and we should keep
that. This will speed up searching if a file does not exist in CBFS.
For arm, the size in header is correct now so we can remove the hack by
CONFIG_ROM_SIZE.
Change-Id: I286ecda73bd781550e03b0b817ed3fb567d6b8d7
Signed-off-by: Hung-Te Lin <hungte@chromium.org>
Reviewed-on: http://review.coreboot.org/2458
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
get_cbfs_header expects CBFS_HEADER_INVALID_ADDRESS (0xffffffff)
instead of NULL when something is wrong.
Also, fix typo.
Change-Id: Ibe56c9eab3b9fdfc6d0b14bc848ca75f3a4fc2f1
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/2455
Tested-by: build bot (Jenkins)
Reviewed-by: Peter Stuge <peter@stuge.se>
Upgrade CBFS in libpayload to use new media-based implementation from coreboot
( http://review.coreboot.org/#/c/2182/ ).
Old CBFS functions (cbfs_find, cbfs_find_file, get_cbfs_header) are still
supported, although the recommended way is to use new CBFS API.
To migrate your existing x86 payload source:
- Change cbfs_find to cbfs_get_file
- Change cbfs_find_file to cbfs_get_file_content
- Prefix every CBFS call with a CBFS_DEFAULT_MEDIA argument.
Ex, char *jpeg_data = cbfs_find_file("splash.jpg", CBFS_TYPE_BOOTSPLASH);
=> char *jpeg_data = cbfs_get_file_content(
CBFS_DEFAULT_MEDIA, "splash.jpg", CBFS_TYPE_BOOTSPLASH);
The legacy setup_cbfs_from_{ram,flash} is also supported, although the better
equivalent is to make a new media instance:
struct cbfs_media ram_media;
init_cbfs_ram_media(&ram_media, start, size);
char *data = cbfs_get_file_content(&ram_media, "myfile", my_type);
Verified by being successfully linked with filo.
Change-Id: If797bc7e3ba975d7e3be905c59424f7a93b8ce11
Signed-off-by: Hung-Te Lin <hungte@chromium.org>
Reviewed-on: http://review.coreboot.org/2191
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Anton Kochkov <anton.kochkov@gmail.com>
Keep in sync with coreboot's version.
Change-Id: I8a253446bd3b2ce9d05c6076a3f49f0260ecd5f9
Signed-off-by: Patrick Georgi <patrick.georgi@secunet.com>
Reviewed-on: http://review.coreboot.org/158
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <patrick@georgi-clan.de>
Add cbfs core from coreboot into libpayload, and to support lzma decode,
add coreboot's lzma code, too. Carl-Daniel agreed to relicense the
lzmadecode wrapper as BSD-l, solving licensing problems.
Change-Id: Id28990fe7e951d99447e265a4880d70a8f208dd2
Signed-off-by: Patrick Georgi <patrick.georgi@secunet.com>
Reviewed-on: http://review.coreboot.org/115
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Marc Jones <marcj303@gmail.com>
You-Cheng Syu