coreboot-kgpe-d16/payloads/libpayload/libcbfs/cbfs_core.c

365 lines
11 KiB
C
Raw Normal View History

/*
* This file is part of the libpayload project.
*
* Copyright (C) 2011 secunet Security Networks AG
* Copyright (C) 2013 Google, Inc.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
/* The CBFS core requires a couple of #defines or functions to adapt it to the
* target environment:
*
* CBFS_CORE_WITH_LZMA (must be #define)
* if defined, ulzma() must exist for decompression of data streams
*
* CBFS_CORE_WITH_LZ4 (must be #define)
* if defined, ulz4f() must exist for decompression of data streams
*
* ERROR(x...)
* print an error message x (in printf format)
*
* LOG(x...)
* print a message x (in printf format)
*
* DEBUG(x...)
* print a debug message x (in printf format)
*
*/
#include <libpayload.h>
#include <cbfs.h>
#include <string.h>
#include <sysinfo.h>
/* returns a pointer to CBFS master header, or CBFS_HEADER_INVALID_ADDRESS
* on failure */
const struct cbfs_header *cbfs_get_header(struct cbfs_media *media)
{
CBFS: Automate ROM image layout and remove hardcoded offsets 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>
2014-11-10 22:14:24 +01:00
int32_t rel_offset;
const struct cbfs_header *header;
struct cbfs_media default_media;
if (media == CBFS_DEFAULT_MEDIA) {
media = &default_media;
if (init_default_cbfs_media(media) != 0) {
ERROR("Failed to initialize default media.\n");
return CBFS_HEADER_INVALID_ADDRESS;
}
}
media->open(media);
CBFS: Automate ROM image layout and remove hardcoded offsets 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>
2014-11-10 22:14:24 +01:00
if (!media->read(media, &rel_offset, (size_t)(0 - sizeof(int32_t)),
sizeof(int32_t))) {
ERROR("Could not read CBFS master header offset!\n");
return CBFS_HEADER_INVALID_ADDRESS;
}
header = media->map(media, (size_t)rel_offset, sizeof(*header));
DEBUG("CBFS header at %#zx (-%#zx from end of image).\n",
(size_t)rel_offset, (size_t)-rel_offset);
media->close(media);
if (header == CBFS_MEDIA_INVALID_MAP_ADDRESS) {
CBFS: Automate ROM image layout and remove hardcoded offsets 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>
2014-11-10 22:14:24 +01:00
ERROR("Failed to load CBFS header from %#zx(-%#zx)\n",
(size_t)rel_offset, (size_t)-rel_offset);
return CBFS_HEADER_INVALID_ADDRESS;
}
if (CBFS_HEADER_MAGIC != ntohl(header->magic)) {
CBFS: Automate ROM image layout and remove hardcoded offsets 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>
2014-11-10 22:14:24 +01:00
ERROR("Could not find valid CBFS master header at %#zx(-%#zx): "
"magic %#.8x vs %#.8x.\n", (size_t)rel_offset,
(size_t)-rel_offset, CBFS_HEADER_MAGIC,
ntohl(header->magic));
if (header->magic == 0xffffffff) {
ERROR("Maybe ROM is not mapped properly?\n");
}
return CBFS_HEADER_INVALID_ADDRESS;
}
return header;
}
static int get_cbfs_range(uint32_t *offset, uint32_t *cbfs_end,
struct cbfs_media *media)
{
const struct cbfs_header *header;
if (media == CBFS_DEFAULT_MEDIA &&
lib_sysinfo.cbfs_offset && lib_sysinfo.cbfs_size) {
*offset = lib_sysinfo.cbfs_offset;
*cbfs_end = *offset + lib_sysinfo.cbfs_size;
return 0;
}
/* read offset and size from cbfs master header */
DEBUG("Read CBFS offset & size from master header\n");
header = cbfs_get_header(media);
if (header == CBFS_HEADER_INVALID_ADDRESS)
return -1;
// Logical offset (for source media) of first file.
*offset = ntohl(header->offset);
*cbfs_end = ntohl(header->romsize);
#if CONFIG(LP_ARCH_X86)
// resolve actual length of ROM used for CBFS components
// the bootblock size was not taken into account
*cbfs_end -= ntohl(header->bootblocksize);
// fine tune the length to handle alignment positioning.
// using (bootblock size) % align, to derive the
// number of bytes the bootblock is off from the alignment size.
if ((ntohl(header->bootblocksize) % CBFS_ALIGNMENT))
*cbfs_end -= (CBFS_ALIGNMENT -
(ntohl(header->bootblocksize) % CBFS_ALIGNMENT));
else
*cbfs_end -= 1;
#endif
return 0;
}
/* public API starts here*/
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
struct cbfs_handle *cbfs_get_handle(struct cbfs_media *media, const char *name)
{
const char *vardata;
uint32_t offset, cbfs_end, vardata_len;
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
struct cbfs_file file;
struct cbfs_handle *handle = malloc(sizeof(*handle));
if (!handle)
return NULL;
if (get_cbfs_range(&offset, &cbfs_end, media)) {
ERROR("Failed to find cbfs range\n");
free(handle);
return NULL;
}
if (media == CBFS_DEFAULT_MEDIA) {
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
media = &handle->media;
if (init_default_cbfs_media(media) != 0) {
ERROR("Failed to initialize default media.\n");
free(handle);
return NULL;
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
} else {
memcpy(&handle->media, media, sizeof(*media));
}
DEBUG("CBFS location: 0x%x~0x%x\n", offset, cbfs_end);
DEBUG("Looking for '%s' starting from 0x%x.\n", name, offset);
media->open(media);
while (offset < cbfs_end &&
media->read(media, &file, offset, sizeof(file)) == sizeof(file)) {
if (memcmp(CBFS_FILE_MAGIC, file.magic,
sizeof(file.magic)) != 0) {
uint32_t new_align = CBFS_ALIGNMENT;
if (offset % CBFS_ALIGNMENT)
new_align += CBFS_ALIGNMENT -
(offset % CBFS_ALIGNMENT);
ERROR("ERROR: No file header found at 0x%xx - "
"try next aligned address: 0x%x.\n", offset,
offset + new_align);
offset += new_align;
continue;
}
vardata_len = ntohl(file.offset) - sizeof(file);
DEBUG(" - load entry 0x%x variable data (%d bytes)...\n",
offset, vardata_len);
// load file name (arbitrary length).
vardata = (const char*)media->map(
media, offset + sizeof(file), vardata_len);
if (vardata == CBFS_MEDIA_INVALID_MAP_ADDRESS) {
ERROR("ERROR: Failed to get filename: 0x%x.\n", offset);
} else if (strcmp(vardata, name) == 0) {
int file_offset = ntohl(file.offset),
file_len = ntohl(file.len);
DEBUG("Found file (offset=0x%x, len=%d).\n",
offset + file_offset, file_len);
media->unmap(media, vardata);
media->close(media);
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
handle->type = ntohl(file.type);
handle->media_offset = offset;
handle->content_offset = file_offset;
handle->content_size = file_len;
handle->attribute_offset =
ntohl(file.attributes_offset);
return handle;
} else {
DEBUG(" (unmatched file @0x%x: %s)\n", offset,
vardata);
media->unmap(media, vardata);
}
// Move to next file.
uint32_t next_offset = offset + ntohl(file.len) + ntohl(file.offset);
if (next_offset % CBFS_ALIGNMENT)
next_offset += CBFS_ALIGNMENT - (next_offset % CBFS_ALIGNMENT);
// Check that offset is strictly monotonic to prevent infinite loop
if (next_offset <= offset) {
ERROR("ERROR: corrupted CBFS file header at 0x%x.\n", offset);
break;
}
offset = next_offset;
}
media->close(media);
LOG("WARNING: '%s' not found.\n", name);
free(handle);
return NULL;
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
void *cbfs_get_contents(struct cbfs_handle *handle, size_t *size, size_t limit)
{
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
struct cbfs_media *m = &handle->media;
size_t on_media_size = handle->content_size;
int algo = CBFS_COMPRESS_NONE;
void *ret = NULL;
size_t dummy_size;
if (!size)
size = &dummy_size;
struct cbfs_file_attr_compression *comp =
cbfs_get_attr(handle, CBFS_FILE_ATTR_TAG_COMPRESSION);
if (comp) {
algo = ntohl(comp->compression);
DEBUG("File '%s' is compressed (alg=%d)\n", name, algo);
*size = ntohl(comp->decompressed_size);
/* TODO: Implement partial decompression with |limit| */
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
if (algo == CBFS_COMPRESS_NONE) {
if (limit != 0 && limit < on_media_size)
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
on_media_size = limit;
*size = on_media_size;
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
void *data = m->map(m, handle->media_offset + handle->content_offset,
on_media_size);
if (data == CBFS_MEDIA_INVALID_MAP_ADDRESS)
return NULL;
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
ret = malloc(*size);
if (ret != NULL) {
size_t final_size = cbfs_decompress(algo, data, on_media_size,
ret, *size);
if (final_size != *size) {
ERROR("Expect %zu bytes but got %zu bytes after "
"decompression.\n", *size, final_size);
free(ret);
ret = NULL;
}
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
m->unmap(m, data);
return ret;
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
void *cbfs_get_file_content(struct cbfs_media *media, const char *name,
int type, size_t *sz)
{
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
void *ret = NULL;
struct cbfs_handle *handle = cbfs_get_handle(media, name);
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
if (!handle)
return NULL;
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
if (handle->type == type)
ret = cbfs_get_contents(handle, sz, 0);
else
ERROR("File '%s' is of type %x, but we requested %x.\n", name,
handle->type, type);
free(handle);
return ret;
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
void *cbfs_get_attr(struct cbfs_handle *handle, uint32_t tag)
{
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
struct cbfs_media *m = &handle->media;
uint32_t offset = handle->media_offset + handle->attribute_offset;
uint32_t end = handle->media_offset + handle->content_offset;
struct cbfs_file_attribute attr;
void *ret;
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
/* attribute_offset should be 0 when there is no attribute, but all
* values that point into the cbfs_file header are invalid, too. */
if (handle->attribute_offset <= sizeof(struct cbfs_file))
return NULL;
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
m->open(m);
while (offset + sizeof(attr) <= end) {
if (m->read(m, &attr, offset, sizeof(attr)) != sizeof(attr)) {
ERROR("Failed to read attribute header %#x\n", offset);
m->close(m);
return NULL;
}
if (ntohl(attr.tag) != tag) {
uint32_t next_offset = offset + ntohl(attr.len);
// Check that offset is strictly monotonic to prevent infinite loop
if (next_offset <= offset) {
ERROR("ERROR: corrupted CBFS attribute at 0x%x.\n", offset);
m->close(m);
return NULL;
}
offset = next_offset;
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
continue;
}
ret = m->map(m, offset, ntohl(attr.len));
if (ret == CBFS_MEDIA_INVALID_MAP_ADDRESS) {
ERROR("Failed to map attribute at %#x\n", offset);
m->close(m);
return NULL;
}
return ret;
}
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
m->close(m);
libpayload: cbfs: Add cbfs_handle API for more fine-grained accesses 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>
2016-05-12 23:40:57 +02:00
return NULL;
}
size_t cbfs_decompress(int algo, const void *src, size_t srcn, void *dst,
size_t dstn)
{
size_t len;
switch (algo) {
case CBFS_COMPRESS_NONE:
len = MIN(srcn, dstn);
memcpy(dst, src, len);
return len;
#ifdef CBFS_CORE_WITH_LZMA
case CBFS_COMPRESS_LZMA:
return ulzman(src, srcn, dst, dstn);
#endif
#ifdef CBFS_CORE_WITH_LZ4
case CBFS_COMPRESS_LZ4:
return ulz4fn(src, srcn, dst, dstn);
#endif
default:
ERROR("tried to decompress %zu bytes with algorithm "
"#%x, but that algorithm id is unsupported.\n",
srcn, algo);
return 0;
}
}