GNUBoot
/
coreboot-kgpe-d16
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
coreboot-kgpe-d16/util/cbfstool/rmodule.c

697 lines
16 KiB
C
Raw Normal View History

util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
/*
;* Copyright (C) 2014 Google, Inc.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; version 2 of the License.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
Remove address from GPLv2 headers As per discussion with lawyers[tm], it's not a good idea to shorten the license header too much - not for legal reasons but because there are tools that look for them, and giving them a standard pattern simplifies things. However, we got confirmation that we don't have to update every file ever added to coreboot whenever the FSF gets a new lease, but can drop the address instead. util/kconfig is excluded because that's imported code that we may want to synchronize every now and then. $ find * -type f -exec sed -i "s:Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, *MA[, ]*02110-1301[, ]*USA:Foundation, Inc.:" {} + $ find * -type f -exec sed -i "s:Foundation, Inc., 51 Franklin Street, Suite 500, Boston, MA 02110-1335, USA:Foundation, Inc.:" {} + $ find * -type f -exec sed -i "s:Foundation, Inc., 59 Temple Place[-, ]*Suite 330, Boston, MA *02111-1307[, ]*USA:Foundation, Inc.:" {} + $ find * -type f -exec sed -i "s:Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.:Foundation, Inc.:" {} + $ find * -type f -a \! -name \*.patch \ -a \! -name \*_shipped \ -a \! -name LICENSE_GPL \ -a \! -name LGPL.txt \ -a \! -name COPYING \ -a \! -name DISCLAIMER \ -exec sed -i "/Foundation, Inc./ N;s:Foundation, Inc.* USA\.* *:Foundation, Inc. :;s:Foundation, Inc. $:Foundation, Inc.:" {} + Change-Id: Icc968a5a5f3a5df8d32b940f9cdb35350654bef9 Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Reviewed-on: http://review.coreboot.org/9233 Tested-by: build bot (Jenkins) Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
2015-03-26 15:17:45 +01:00
* Foundation, Inc.
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
*/
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
#include <inttypes.h>
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "elfparsing.h"
#include "rmodule.h"
#include "../../src/include/rmodule-defs.h"
struct rmod_context;
struct arch_ops {
int arch;
/* Determine if relocation is a valid type for the architecture. */
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
int (*valid_type)(Elf64_Rela *rel);
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
/* Determine if relocation should be emitted. */
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
int (*should_emit)(Elf64_Rela *rel);
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
};
struct rmod_context {
/* Ops to process relocations. */
struct arch_ops *ops;
/* endian conversion ops */
struct xdr *xdr;
/* Parsed ELF sturcture. */
struct parsed_elf pelf;
/* Program segment. */
Elf64_Phdr *phdr;
/* Collection of relocation addresses fixup in the module. */
Elf64_Xword nrelocs;
Elf64_Addr *emitted_relocs;
/* The following fields are addresses within the linked program. */
Elf64_Addr link_addr;
Elf64_Addr entry;
Elf64_Addr parameters_begin;
Elf64_Addr parameters_end;
Elf64_Addr bss_begin;
Elf64_Addr bss_end;
Elf64_Xword size;
};
/*
* Architecture specific support operations.
*/
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
static int valid_reloc_386(Elf64_Rela *rel)
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
{
int type;
type = ELF64_R_TYPE(rel->r_info);
/* Only these 2 relocations are expected to be found. */
return (type == R_386_32 || type == R_386_PC32);
}
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
static int should_emit_386(Elf64_Rela *rel)
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
{
int type;
type = ELF64_R_TYPE(rel->r_info);
/* R_386_32 relocations are absolute. Must emit these. */
return (type == R_386_32);
}
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
static int valid_reloc_arm(Elf64_Rela *rel)
rmodtool: add support for ARM Add support for creating ARM rmodules. There are 3 expected relocations for an ARM rmodule: - R_ARM_ABS32 - R_ARM_THM_PC22 - R_ARM_THM_JUMP24 R_ARM_ABS32 is the only type that needs to emitted for relocation as the other 2 are relative relocations. BUG=chrome-os-partner:27094 BRANCH=None TEST=Built vbootstub for ARM device. Original-Change-Id: I0c22d4abca970e82ccd60b33fed700b96e3e52fb Original-Signed-off-by: Aaron Durbin <adurbin@chromuim.org> Original-Reviewed-on: https://chromium-review.googlesource.com/190922 Original-Reviewed-by: Gabe Black <gabeblack@chromium.org> (cherry picked from commit a642102ba7ace5c1829abe7732199eda6646950a) Signed-off-by: Marc Jones <marc.jones@se-eng.com> Change-Id: Ib3b3c90ebb672d8d6a537df896b97dc82c6186cc Reviewed-on: http://review.coreboot.org/7204 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com> Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
2014-03-20 17:08:02 +01:00
{
int type;
type = ELF64_R_TYPE(rel->r_info);
cbfstool: Add relocation codes for arm mode Add relocation codes required for arm mode. These are required by armv4. BUG=chrome-os-partner:30784 BRANCH=None TEST=Compiles successfully for rush Original-Change-Id: Ie7c5b3e07689c85091036a619a65f9fea1918b6b Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/209973 Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org> (cherry picked from commit dc5f411f95e02a02b4a4ef4652303ce62aa220c2) Signed-off-by: Marc Jones <marc.jones@se-eng.com> Change-Id: Ie62ed730080299e474c256371ab88f605b54c10f Reviewed-on: http://review.coreboot.org/8675 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-07-23 22:42:22 +02:00
/* Only these 6 relocations are expected to be found. */
rmodtool: add support for ARM Add support for creating ARM rmodules. There are 3 expected relocations for an ARM rmodule: - R_ARM_ABS32 - R_ARM_THM_PC22 - R_ARM_THM_JUMP24 R_ARM_ABS32 is the only type that needs to emitted for relocation as the other 2 are relative relocations. BUG=chrome-os-partner:27094 BRANCH=None TEST=Built vbootstub for ARM device. Original-Change-Id: I0c22d4abca970e82ccd60b33fed700b96e3e52fb Original-Signed-off-by: Aaron Durbin <adurbin@chromuim.org> Original-Reviewed-on: https://chromium-review.googlesource.com/190922 Original-Reviewed-by: Gabe Black <gabeblack@chromium.org> (cherry picked from commit a642102ba7ace5c1829abe7732199eda6646950a) Signed-off-by: Marc Jones <marc.jones@se-eng.com> Change-Id: Ib3b3c90ebb672d8d6a537df896b97dc82c6186cc Reviewed-on: http://review.coreboot.org/7204 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com> Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
2014-03-20 17:08:02 +01:00
return (type == R_ARM_ABS32 || type == R_ARM_THM_PC22 ||
cbfstool: Add relocation codes for arm mode Add relocation codes required for arm mode. These are required by armv4. BUG=chrome-os-partner:30784 BRANCH=None TEST=Compiles successfully for rush Original-Change-Id: Ie7c5b3e07689c85091036a619a65f9fea1918b6b Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/209973 Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org> (cherry picked from commit dc5f411f95e02a02b4a4ef4652303ce62aa220c2) Signed-off-by: Marc Jones <marc.jones@se-eng.com> Change-Id: Ie62ed730080299e474c256371ab88f605b54c10f Reviewed-on: http://review.coreboot.org/8675 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-07-23 22:42:22 +02:00
type == R_ARM_THM_JUMP24 || type == R_ARM_V4BX ||
type == R_ARM_CALL || type == R_ARM_JUMP24);
rmodtool: add support for ARM Add support for creating ARM rmodules. There are 3 expected relocations for an ARM rmodule: - R_ARM_ABS32 - R_ARM_THM_PC22 - R_ARM_THM_JUMP24 R_ARM_ABS32 is the only type that needs to emitted for relocation as the other 2 are relative relocations. BUG=chrome-os-partner:27094 BRANCH=None TEST=Built vbootstub for ARM device. Original-Change-Id: I0c22d4abca970e82ccd60b33fed700b96e3e52fb Original-Signed-off-by: Aaron Durbin <adurbin@chromuim.org> Original-Reviewed-on: https://chromium-review.googlesource.com/190922 Original-Reviewed-by: Gabe Black <gabeblack@chromium.org> (cherry picked from commit a642102ba7ace5c1829abe7732199eda6646950a) Signed-off-by: Marc Jones <marc.jones@se-eng.com> Change-Id: Ib3b3c90ebb672d8d6a537df896b97dc82c6186cc Reviewed-on: http://review.coreboot.org/7204 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com> Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
2014-03-20 17:08:02 +01:00
}
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
static int should_emit_arm(Elf64_Rela *rel)
rmodtool: add support for ARM Add support for creating ARM rmodules. There are 3 expected relocations for an ARM rmodule: - R_ARM_ABS32 - R_ARM_THM_PC22 - R_ARM_THM_JUMP24 R_ARM_ABS32 is the only type that needs to emitted for relocation as the other 2 are relative relocations. BUG=chrome-os-partner:27094 BRANCH=None TEST=Built vbootstub for ARM device. Original-Change-Id: I0c22d4abca970e82ccd60b33fed700b96e3e52fb Original-Signed-off-by: Aaron Durbin <adurbin@chromuim.org> Original-Reviewed-on: https://chromium-review.googlesource.com/190922 Original-Reviewed-by: Gabe Black <gabeblack@chromium.org> (cherry picked from commit a642102ba7ace5c1829abe7732199eda6646950a) Signed-off-by: Marc Jones <marc.jones@se-eng.com> Change-Id: Ib3b3c90ebb672d8d6a537df896b97dc82c6186cc Reviewed-on: http://review.coreboot.org/7204 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com> Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
2014-03-20 17:08:02 +01:00
{
int type;
type = ELF64_R_TYPE(rel->r_info);
/* R_ARM_ABS32 relocations are absolute. Must emit these. */
return (type == R_ARM_ABS32);
}
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
static int valid_reloc_aarch64(Elf64_Rela *rel)
rmodtool: Add support for aarch64 BUG=chrome-os-partner:31615 BRANCH=None TEST=Compiles succesfully, rmodule created and loaded for ryu Change-Id: Icc80b845fe43a012665d77c3ef55bd30784fd3fc Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: 24ad4383a9ea75ba6beb084451b74e8a8735085b Original-Change-Id: I4f3a5dbb8fc8150aa670d2e6fed56a6774feb4a6 Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/214329 Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/8806 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-08-27 00:21:15 +02:00
{
int type;
type = ELF64_R_TYPE(rel->r_info);
return (type == R_AARCH64_ADR_PREL_PG_HI21 ||
type == R_AARCH64_ADD_ABS_LO12_NC ||
ARM64 rmodule: Add new reloc type R_AARCH64_LDST8_ABS_LO12_NC BUG=chrome-os-partner:33962 BRANCH=None TEST=Compiles and boots to kernel prompt. Change-Id: Id7b0dfb5a51c2f29bdb031b98606940c118959ec Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: 7a1c4d2f35c135d542708c4dabcca5e8c1d453c0 Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Change-Id: If132323885f23d75e1fcde064398e85c2c17f257 Original-Reviewed-on: https://chromium-review.googlesource.com/231560 Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Commit-Queue: Furquan Shaikh <furquan@chromium.org> Reviewed-on: http://review.coreboot.org/8809 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-11-22 00:41:10 +01:00
type == R_AARCH64_LDST8_ABS_LO12_NC ||
ARM64 rmodule: Add new reloc type R_AARCH64_CONDBR19 BUG=chrome-os-partner:41185 BRANCH=None TEST=Compiles successfully for ryu. Change-Id: I78ccc4b5ef8b49bae533e5a82323f07aaab01a7e Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: cf9d288d4bbc91301dc814408bf5e3686b869974 Original-Change-Id: I24df0eb51883a634ec3d26d46f79a059a4f8394a Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/275749 Original-Trybot-Ready: Furquan Shaikh <furquan@chromium.org> Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Commit-Queue: Furquan Shaikh <furquan@chromium.org> Reviewed-on: http://review.coreboot.org/10476 Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins)
2015-06-08 20:58:04 +02:00
type == R_AARCH64_CONDBR19 ||
rmodtool: Add support for aarch64 BUG=chrome-os-partner:31615 BRANCH=None TEST=Compiles succesfully, rmodule created and loaded for ryu Change-Id: Icc80b845fe43a012665d77c3ef55bd30784fd3fc Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: 24ad4383a9ea75ba6beb084451b74e8a8735085b Original-Change-Id: I4f3a5dbb8fc8150aa670d2e6fed56a6774feb4a6 Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/214329 Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/8806 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-08-27 00:21:15 +02:00
type == R_AARCH64_JUMP26 ||
type == R_AARCH64_LDST32_ABS_LO12_NC ||
rmodtool: add another aarch64 relocation BUG=chrome-os-partner:32112 BRANCH=None TEST=Built secmon which had this type of relocation. Change-Id: Ie367c348fbf59465e238e5fa60f217f5373501b3 Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: a754bc1fe39c19ab8b2f7be9648cccb06156b0ef Original-Change-Id: If170d9e270daf3153e92d16c06516915c727e930 Original-Signed-off-by: Aaron Durbin <adurbin@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/218843 Original-Reviewed-by: Furquan Shaikh <furquan@chromium.org> Reviewed-on: http://review.coreboot.org/8807 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-09-18 20:39:16 +02:00
type == R_AARCH64_LDST64_ABS_LO12_NC ||
rmodtool: Add support for aarch64 BUG=chrome-os-partner:31615 BRANCH=None TEST=Compiles succesfully, rmodule created and loaded for ryu Change-Id: Icc80b845fe43a012665d77c3ef55bd30784fd3fc Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: 24ad4383a9ea75ba6beb084451b74e8a8735085b Original-Change-Id: I4f3a5dbb8fc8150aa670d2e6fed56a6774feb4a6 Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/214329 Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/8806 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-08-27 00:21:15 +02:00
type == R_AARCH64_CALL26 ||
type == R_AARCH64_ABS64 ||
type == R_AARCH64_LD_PREL_LO19 ||
type == R_AARCH64_ADR_PREL_LO21);
}
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
static int should_emit_aarch64(Elf64_Rela *rel)
rmodtool: Add support for aarch64 BUG=chrome-os-partner:31615 BRANCH=None TEST=Compiles succesfully, rmodule created and loaded for ryu Change-Id: Icc80b845fe43a012665d77c3ef55bd30784fd3fc Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: 24ad4383a9ea75ba6beb084451b74e8a8735085b Original-Change-Id: I4f3a5dbb8fc8150aa670d2e6fed56a6774feb4a6 Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/214329 Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/8806 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-08-27 00:21:15 +02:00
{
int type;
type = ELF64_R_TYPE(rel->r_info);
return (type == R_AARCH64_ABS64);
}
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
static struct arch_ops reloc_ops[] = {
{
.arch = EM_386,
.valid_type = valid_reloc_386,
.should_emit = should_emit_386,
},
rmodtool: add support for ARM Add support for creating ARM rmodules. There are 3 expected relocations for an ARM rmodule: - R_ARM_ABS32 - R_ARM_THM_PC22 - R_ARM_THM_JUMP24 R_ARM_ABS32 is the only type that needs to emitted for relocation as the other 2 are relative relocations. BUG=chrome-os-partner:27094 BRANCH=None TEST=Built vbootstub for ARM device. Original-Change-Id: I0c22d4abca970e82ccd60b33fed700b96e3e52fb Original-Signed-off-by: Aaron Durbin <adurbin@chromuim.org> Original-Reviewed-on: https://chromium-review.googlesource.com/190922 Original-Reviewed-by: Gabe Black <gabeblack@chromium.org> (cherry picked from commit a642102ba7ace5c1829abe7732199eda6646950a) Signed-off-by: Marc Jones <marc.jones@se-eng.com> Change-Id: Ib3b3c90ebb672d8d6a537df896b97dc82c6186cc Reviewed-on: http://review.coreboot.org/7204 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com> Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
2014-03-20 17:08:02 +01:00
{
.arch = EM_ARM,
.valid_type = valid_reloc_arm,
.should_emit = should_emit_arm,
},
rmodtool: Add support for aarch64 BUG=chrome-os-partner:31615 BRANCH=None TEST=Compiles succesfully, rmodule created and loaded for ryu Change-Id: Icc80b845fe43a012665d77c3ef55bd30784fd3fc Signed-off-by: Patrick Georgi <pgeorgi@chromium.org> Original-Commit-Id: 24ad4383a9ea75ba6beb084451b74e8a8735085b Original-Change-Id: I4f3a5dbb8fc8150aa670d2e6fed56a6774feb4a6 Original-Signed-off-by: Furquan Shaikh <furquan@google.com> Original-Reviewed-on: https://chromium-review.googlesource.com/214329 Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org> Original-Tested-by: Furquan Shaikh <furquan@chromium.org> Original-Commit-Queue: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/8806 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2014-08-27 00:21:15 +02:00
{
.arch = EM_AARCH64,
.valid_type = valid_reloc_aarch64,
.should_emit = should_emit_aarch64,
},
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
};
/*
* Relocation processing loops.
*/
static int for_each_reloc(struct rmod_context *ctx, int do_emit)
{
Elf64_Half i;
struct parsed_elf *pelf = &ctx->pelf;
for (i = 0; i < pelf->ehdr.e_shnum; i++) {
Elf64_Shdr *shdr;
Elf64_Rela *relocs;
Elf64_Xword nrelocs;
Elf64_Xword j;
relocs = pelf->relocs[i];
/* No relocations in this section. */
if (relocs == NULL)
continue;
shdr = &pelf->shdr[i];
nrelocs = shdr->sh_size / shdr->sh_entsize;
for (j = 0; j < nrelocs; j++) {
Elf64_Rela *r = &relocs[j];
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
if (!ctx->ops->valid_type(r)) {
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
ERROR("Invalid reloc type: %u\n",
(unsigned int)ELF64_R_TYPE(r->r_info));
return -1;
}
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
if (ctx->ops->should_emit(r)) {
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
int n = ctx->nrelocs;
if (do_emit)
ctx->emitted_relocs[n] = r->r_offset;
ctx->nrelocs++;
}
}
}
return 0;
}
static int find_program_segment(struct rmod_context *ctx)
{
int i;
int nsegments;
struct parsed_elf *pelf;
Fix GCC 5.1 compile issue seen at Linux Arch rmodule.c: In function ‘rmodule_create’: rmodule.c:287:29: error: ‘phdr’ may be used uninitialized in this function [-Werror=maybe-uninitialized] (phdr->p_vaddr + phdr->p_memsz))) { ^ rmodule.c:204:14: note: ‘phdr’ was declared here Elf64_Phdr *phdr; ^ Change-Id: I94a235253610348484eef218ec855103a3bb5da5 Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com> Reviewed-on: http://review.coreboot.org/10881 Tested-by: build bot (Jenkins) Reviewed-by: Francis Rowe <info@gluglug.org.uk>
2015-07-11 02:30:01 +02:00
Elf64_Phdr *phdr = NULL;
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
pelf = &ctx->pelf;
/* There should only be a single loadable segment. */
nsegments = 0;
for (i = 0; i < pelf->ehdr.e_phnum; i++) {
if (pelf->phdr[i].p_type != PT_LOAD)
continue;
phdr = &pelf->phdr[i];
nsegments++;
}
if (nsegments != 1) {
ERROR("Unexepcted number of loadable segments: %d.\n",
nsegments);
return -1;
}
INFO("Segment at 0x%0llx, file size 0x%0llx, mem size 0x%0llx.\n",
(long long)phdr->p_vaddr, (long long)phdr->p_filesz,
(long long)phdr->p_memsz);
ctx->phdr = phdr;
return 0;
}
static int
filter_relocation_sections(struct rmod_context *ctx)
{
int i;
const char *shstrtab;
struct parsed_elf *pelf;
const Elf64_Phdr *phdr;
pelf = &ctx->pelf;
phdr = ctx->phdr;
shstrtab = buffer_get(pelf->strtabs[pelf->ehdr.e_shstrndx]);
/*
* Find all relocation sections that contain relocation entries
* for sections that fall within the bounds of the segment. For
* easier processing the pointer to the relocation array for the
* sections that don't fall within the loadable program are NULL'd
* out.
*/
for (i = 0; i < pelf->ehdr.e_shnum; i++) {
Elf64_Shdr *shdr;
Elf64_Word sh_info;
const char *section_name;
shdr = &pelf->shdr[i];
/* Ignore non-relocation sections. */
if (shdr->sh_type != SHT_RELA && shdr->sh_type != SHT_REL)
continue;
/* Obtain section which relocations apply. */
sh_info = shdr->sh_info;
shdr = &pelf->shdr[sh_info];
section_name = &shstrtab[shdr->sh_name];
DEBUG("Relocation section found for '%s' section.\n",
section_name);
/* Do not process relocations for debug sections. */
if (strstr(section_name, ".debug") != NULL) {
pelf->relocs[i] = NULL;
continue;
}
/*
* If relocations apply to a non program section ignore the
* relocations for future processing.
*/
if (shdr->sh_type != SHT_PROGBITS) {
pelf->relocs[i] = NULL;
continue;
}
if (shdr->sh_addr < phdr->p_vaddr ||
((shdr->sh_addr + shdr->sh_size) >
(phdr->p_vaddr + phdr->p_memsz))) {
ERROR("Relocations being applied to section %d not "
"within segment region.\n", sh_info);
return -1;
}
}
return 0;
}
static int vaddr_cmp(const void *a, const void *b)
{
const Elf64_Addr *pa = a;
const Elf64_Addr *pb = b;
if (*pa < *pb)
return -1;
if (*pa > *pb)
return 1;
return 0;
}
static int collect_relocations(struct rmod_context *ctx)
{
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
Elf64_Xword nrelocs;
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
/*
* The relocs array in the pelf should only contain relocations that
* apply to the program. Count the number relocations. Then collect
* them into the allocated buffer.
*/
if (for_each_reloc(ctx, 0))
return -1;
nrelocs = ctx->nrelocs;
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
INFO("%" PRIu64 " relocations to be emitted.\n", nrelocs);
rmodtool: Allow rmodules with 0 relocations Currently, rmodules with 0 relocations are not allowed. Fix this by skipping addition of .rmodules section on 0 relocs. Change-Id: I7a39cf409a5f2bc808967d2b5334a15891c4748e Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: http://review.coreboot.org/6774 Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@google.com> Tested-by: build bot (Jenkins)
2014-08-26 23:59:36 +02:00
if (!nrelocs)
return 0;
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
/* Reset the counter for indexing into the array. */
ctx->nrelocs = 0;
ctx->emitted_relocs = calloc(nrelocs, sizeof(Elf64_Addr));
/* Write out the relocations into the emitted_relocs array. */
if (for_each_reloc(ctx, 1))
return -1;
if (ctx->nrelocs != nrelocs) {
ERROR("Mismatch counted and emitted relocations: %zu vs %zu.\n",
(size_t)nrelocs, (size_t)ctx->nrelocs);
return -1;
}
/* Sort the relocations by their address. */
qsort(ctx->emitted_relocs, nrelocs, sizeof(Elf64_Addr), vaddr_cmp);
return 0;
}
static int
populate_sym(struct rmod_context *ctx, const char *sym_name, Elf64_Addr *addr,
rmodtool: make rmodule parameter section optional There are currently 2 uses for rmodule programs: stand alone programs that are separate from the coreboot stages and a relocatable ramstage. For the ramstage usage there's no reason to require a rmodule parameter section. Therefore make this optional. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built ramstage w/ normal linking (w/o a rmodule parameter section). No error. Change-Id: I5f8a415e86510be9409a28068e3d3a4d0ba8733e Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11523 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-06 17:39:10 +02:00
int nsyms, const char *strtab, int optional)
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
{
int i;
Elf64_Sym *syms;
syms = ctx->pelf.syms;
for (i = 0; i < nsyms; i++) {
if (syms[i].st_name == 0)
continue;
if (strcmp(sym_name, &strtab[syms[i].st_name]))
continue;
DEBUG("%s -> 0x%llx\n", sym_name, (long long)syms[i].st_value);
*addr = syms[i].st_value;
return 0;
}
rmodtool: make rmodule parameter section optional There are currently 2 uses for rmodule programs: stand alone programs that are separate from the coreboot stages and a relocatable ramstage. For the ramstage usage there's no reason to require a rmodule parameter section. Therefore make this optional. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built ramstage w/ normal linking (w/o a rmodule parameter section). No error. Change-Id: I5f8a415e86510be9409a28068e3d3a4d0ba8733e Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11523 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-06 17:39:10 +02:00
if (optional) {
DEBUG("optional symbol '%s' not found.\n", sym_name);
*addr = 0;
return 0;
}
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
ERROR("symbol '%s' not found.\n", sym_name);
return -1;
}
static int populate_program_info(struct rmod_context *ctx)
{
int i;
const char *strtab;
struct parsed_elf *pelf;
Elf64_Ehdr *ehdr;
int nsyms;
pelf = &ctx->pelf;
ehdr = &pelf->ehdr;
/* Obtain the string table. */
strtab = NULL;
for (i = 0; i < ehdr->e_shnum; i++) {
if (ctx->pelf.strtabs[i] == NULL)
continue;
/* Don't use the section headers' string table. */
if (i == ehdr->e_shstrndx)
continue;
strtab = buffer_get(ctx->pelf.strtabs[i]);
break;
}
if (strtab == NULL) {
ERROR("No string table found.\n");
return -1;
}
/* Determine number of symbols. */
nsyms = 0;
for (i = 0; i < ehdr->e_shnum; i++) {
if (pelf->shdr[i].sh_type != SHT_SYMTAB)
continue;
nsyms = pelf->shdr[i].sh_size / pelf->shdr[i].sh_entsize;
break;
}
rmodule: use program.ld for linking Bring rmodule linking into the common linking method. The __rmodule_entry symbol was removed while using a more common _start symbol. The rmodtool will honor the entry point found within the ELF header. Add ENV_RMODULE so that one can distinguish the environment when generating linker scripts for rmodules. Lastly, directly use program.ld for the rmodule.ld linker script. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built rambi and analyzed the relocatable ramstage, sipi_vector, and smm rmodules. Change-Id: Iaa499eb229d8171272add9ee6d27cff75e7534ac Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11517 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-05 19:59:26 +02:00
if (populate_sym(ctx, "_rmodule_params", &ctx->parameters_begin,
rmodtool: make rmodule parameter section optional There are currently 2 uses for rmodule programs: stand alone programs that are separate from the coreboot stages and a relocatable ramstage. For the ramstage usage there's no reason to require a rmodule parameter section. Therefore make this optional. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built ramstage w/ normal linking (w/o a rmodule parameter section). No error. Change-Id: I5f8a415e86510be9409a28068e3d3a4d0ba8733e Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11523 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-06 17:39:10 +02:00
nsyms, strtab, 1))
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
return -1;
rmodule: use program.ld for linking Bring rmodule linking into the common linking method. The __rmodule_entry symbol was removed while using a more common _start symbol. The rmodtool will honor the entry point found within the ELF header. Add ENV_RMODULE so that one can distinguish the environment when generating linker scripts for rmodules. Lastly, directly use program.ld for the rmodule.ld linker script. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built rambi and analyzed the relocatable ramstage, sipi_vector, and smm rmodules. Change-Id: Iaa499eb229d8171272add9ee6d27cff75e7534ac Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11517 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-05 19:59:26 +02:00
if (populate_sym(ctx, "_ermodule_params", &ctx->parameters_end,
rmodtool: make rmodule parameter section optional There are currently 2 uses for rmodule programs: stand alone programs that are separate from the coreboot stages and a relocatable ramstage. For the ramstage usage there's no reason to require a rmodule parameter section. Therefore make this optional. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built ramstage w/ normal linking (w/o a rmodule parameter section). No error. Change-Id: I5f8a415e86510be9409a28068e3d3a4d0ba8733e Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11523 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-06 17:39:10 +02:00
nsyms, strtab, 1))
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
return -1;
rmodtool: make rmodule parameter section optional There are currently 2 uses for rmodule programs: stand alone programs that are separate from the coreboot stages and a relocatable ramstage. For the ramstage usage there's no reason to require a rmodule parameter section. Therefore make this optional. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built ramstage w/ normal linking (w/o a rmodule parameter section). No error. Change-Id: I5f8a415e86510be9409a28068e3d3a4d0ba8733e Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11523 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-06 17:39:10 +02:00
if (populate_sym(ctx, "_bss", &ctx->bss_begin, nsyms, strtab, 0))
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
return -1;
rmodtool: make rmodule parameter section optional There are currently 2 uses for rmodule programs: stand alone programs that are separate from the coreboot stages and a relocatable ramstage. For the ramstage usage there's no reason to require a rmodule parameter section. Therefore make this optional. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built ramstage w/ normal linking (w/o a rmodule parameter section). No error. Change-Id: I5f8a415e86510be9409a28068e3d3a4d0ba8733e Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11523 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-06 17:39:10 +02:00
if (populate_sym(ctx, "_ebss", &ctx->bss_end, nsyms, strtab, 0))
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
return -1;
rmodule: use program.ld for linking Bring rmodule linking into the common linking method. The __rmodule_entry symbol was removed while using a more common _start symbol. The rmodtool will honor the entry point found within the ELF header. Add ENV_RMODULE so that one can distinguish the environment when generating linker scripts for rmodules. Lastly, directly use program.ld for the rmodule.ld linker script. BUG=chrome-os-partner:44827 BRANCH=None TEST=Built rambi and analyzed the relocatable ramstage, sipi_vector, and smm rmodules. Change-Id: Iaa499eb229d8171272add9ee6d27cff75e7534ac Signed-off-by: Aaron Durbin <adubin@chromium.org> Reviewed-on: http://review.coreboot.org/11517 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-09-05 19:59:26 +02:00
/* Honor the entry point within the ELF header. */
ctx->entry = ehdr->e_entry;
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
/* Link address is the virtual address of the program segment. */
ctx->link_addr = ctx->phdr->p_vaddr;
/* The program size is the memsz of the program segment. */
ctx->size = ctx->phdr->p_memsz;
return 0;
}
static int
add_section(struct elf_writer *ew, struct buffer *data, const char *name,
Elf64_Addr addr, Elf64_Word size)
{
Elf64_Shdr shdr;
int ret;
memset(&shdr, 0, sizeof(shdr));
if (data != NULL) {
shdr.sh_type = SHT_PROGBITS;
shdr.sh_flags = SHF_ALLOC | SHF_WRITE | SHF_EXECINSTR;
} else {
shdr.sh_type = SHT_NOBITS;
shdr.sh_flags = SHF_ALLOC;
}
shdr.sh_addr = addr;
shdr.sh_offset = addr;
shdr.sh_size = size;
ret = elf_writer_add_section(ew, &shdr, data, name);
if (ret)
ERROR("Could not add '%s' section.\n", name);
return ret;
}
static int
write_elf(const struct rmod_context *ctx, const struct buffer *in,
struct buffer *out)
{
int ret;
int bit64;
size_t loc;
size_t rmod_data_size;
struct elf_writer *ew;
struct buffer rmod_data;
struct buffer rmod_header;
struct buffer program;
struct buffer relocs;
Elf64_Xword total_size;
Elf64_Addr addr;
Elf64_Ehdr ehdr;
bit64 = ctx->pelf.ehdr.e_ident[EI_CLASS] == ELFCLASS64;
/*
* 3 sections will be added to the ELF file.
* +------------------+
* | rmodule header |
* +------------------+
* | program |
* +------------------+
* | relocations |
* +------------------+
*/
/* Create buffer for header and relocations. */
rmod_data_size = sizeof(struct rmodule_header);
if (bit64)
rmod_data_size += ctx->nrelocs * sizeof(Elf64_Addr);
else
rmod_data_size += ctx->nrelocs * sizeof(Elf32_Addr);
if (buffer_create(&rmod_data, rmod_data_size, "rmod"))
return -1;
buffer_splice(&rmod_header, &rmod_data,
0, sizeof(struct rmodule_header));
buffer_clone(&relocs, &rmod_data);
buffer_seek(&relocs, sizeof(struct rmodule_header));
/* Reset current location. */
buffer_set_size(&rmod_header, 0);
buffer_set_size(&relocs, 0);
/* Program contents. */
buffer_splice(&program, in, ctx->phdr->p_offset, ctx->phdr->p_filesz);
/* Create ELF writer with modified entry point. */
memcpy(&ehdr, &ctx->pelf.ehdr, sizeof(ehdr));
ehdr.e_entry = ctx->entry;
ew = elf_writer_init(&ehdr);
if (ew == NULL) {
ERROR("Failed to create ELF writer.\n");
buffer_delete(&rmod_data);
return -1;
}
/* Write out rmodule_header. */
ctx->xdr->put16(&rmod_header, RMODULE_MAGIC);
ctx->xdr->put8(&rmod_header, RMODULE_VERSION_1);
ctx->xdr->put8(&rmod_header, 0);
/* payload_begin_offset */
loc = sizeof(struct rmodule_header);
ctx->xdr->put32(&rmod_header, loc);
/* payload_end_offset */
loc += ctx->phdr->p_filesz;
ctx->xdr->put32(&rmod_header, loc);
/* relocations_begin_offset */
ctx->xdr->put32(&rmod_header, loc);
/* relocations_end_offset */
if (bit64)
loc += ctx->nrelocs * sizeof(Elf64_Addr);
else
loc += ctx->nrelocs * sizeof(Elf32_Addr);
ctx->xdr->put32(&rmod_header, loc);
/* module_link_start_address */
ctx->xdr->put32(&rmod_header, ctx->link_addr);
/* module_program_size */
ctx->xdr->put32(&rmod_header, ctx->size);
/* module_entry_point */
ctx->xdr->put32(&rmod_header, ctx->entry);
/* parameters_begin */
ctx->xdr->put32(&rmod_header, ctx->parameters_begin);
/* parameters_end */
ctx->xdr->put32(&rmod_header, ctx->parameters_end);
/* bss_begin */
ctx->xdr->put32(&rmod_header, ctx->bss_begin);
/* bss_end */
ctx->xdr->put32(&rmod_header, ctx->bss_end);
/* padding[4] */
ctx->xdr->put32(&rmod_header, 0);
ctx->xdr->put32(&rmod_header, 0);
ctx->xdr->put32(&rmod_header, 0);
ctx->xdr->put32(&rmod_header, 0);
/* Write the relocations. */
cbfstool: Clean up in preparation for adding new files This enables more warnings on the cbfstool codebase and fixes the issues that surface as a result. A memory leak that used to occur when compressing files with lzma is also found and fixed. Finally, there are several fixes for the Makefile: - Its autodependencies used to be broken because the target for the .dependencies file was misnamed; this meant that Make didn't know how to rebuild the file, and so would silently skip the step of updating it before including it. - The ability to build to a custom output directory by defining the obj variable had bitrotted. - The default value of the obj variable was causing implicit rules not to apply when specifying a file as a target without providing a custom value for obj. - Add a distclean target for removing the .dependencies file. BUG=chromium:461875 TEST=Build an image with cbfstool both before and after. BRANCH=None Change-Id: I951919d63443f2b053c2e67c1ac9872abc0a43ca Signed-off-by: Sol Boucher <solb@chromium.org> Original-Commit-Id: 49293443b4e565ca48d284e9a66f80c9c213975d Original-Change-Id: Ia7350c2c3306905984cfa711d5fc4631f0b43d5b Original-Signed-off-by: Sol Boucher <solb@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/257340 Original-Reviewed-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-by: Stefan Reinauer <reinauer@chromium.org> Reviewed-on: http://review.coreboot.org/9937 Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <pgeorgi@google.com>
2015-03-06 00:38:03 +01:00
for (unsigned i = 0; i < ctx->nrelocs; i++) {
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
if (bit64)
ctx->xdr->put64(&relocs, ctx->emitted_relocs[i]);
else
ctx->xdr->put32(&relocs, ctx->emitted_relocs[i]);
}
total_size = 0;
addr = 0;
/*
* There are 2 cases to deal with. The program has a large NOBITS
* section and the relocations can fit entirely within occupied memory
* region for the program. The other is that the relocations increase
* the memory footprint of the program if it was loaded directly into
* the region it would run. The rmdoule header is a fixed cost that
* is considered a part of the program.
*/
total_size += buffer_size(&rmod_header);
rmodtool: correct final memory size calculation Apparently when I originally wrote this I confused myself to no end. The code/data of an rmodule has a set memory size which is associated with the .payload section. The relocation entries may increase the overall footprint of the memory size if the rmodule has no bss but a lot of relocations. Therefore, just compare relocation entries size plus the file size of the .payload section with the memory size of the paylod section. The .empty section is added only when we have not met the final target size. Change-Id: I5521dff048ae64a9b6e3c8f84a390eba37c7d0f5 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/6767 Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Tested-by: build bot (Jenkins) Reviewed-by: Furquan Shaikh <furquan@google.com>
2014-08-26 20:52:30 +02:00
if (buffer_size(&relocs) + ctx->phdr->p_filesz > ctx->phdr->p_memsz) {
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
total_size += buffer_size(&relocs);
total_size += ctx->phdr->p_filesz;
rmodtool: correct final memory size calculation Apparently when I originally wrote this I confused myself to no end. The code/data of an rmodule has a set memory size which is associated with the .payload section. The relocation entries may increase the overall footprint of the memory size if the rmodule has no bss but a lot of relocations. Therefore, just compare relocation entries size plus the file size of the .payload section with the memory size of the paylod section. The .empty section is added only when we have not met the final target size. Change-Id: I5521dff048ae64a9b6e3c8f84a390eba37c7d0f5 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/6767 Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Tested-by: build bot (Jenkins) Reviewed-by: Furquan Shaikh <furquan@google.com>
2014-08-26 20:52:30 +02:00
} else {
total_size += ctx->phdr->p_memsz;
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
}
ret = add_section(ew, &rmod_header, ".header", addr,
buffer_size(&rmod_header));
if (ret < 0)
goto out;
addr += buffer_size(&rmod_header);
ret = add_section(ew, &program, ".program", addr, ctx->phdr->p_filesz);
if (ret < 0)
goto out;
addr += ctx->phdr->p_filesz;
rmodtool: Allow rmodules with 0 relocations Currently, rmodules with 0 relocations are not allowed. Fix this by skipping addition of .rmodules section on 0 relocs. Change-Id: I7a39cf409a5f2bc808967d2b5334a15891c4748e Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: http://review.coreboot.org/6774 Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org> Reviewed-by: Aaron Durbin <adurbin@google.com> Tested-by: build bot (Jenkins)
2014-08-26 23:59:36 +02:00
if (ctx->nrelocs) {
ret = add_section(ew, &relocs, ".relocs", addr,
buffer_size(&relocs));
if (ret < 0)
goto out;
addr += buffer_size(&relocs);
}
util: add rmodtool for parsing ELF files to rmodules The current implementation of creating rmodules relies on invoking the linker in a certain manner with the relocations overlaid on the BSS section. It's not really surprising that the linker doesn't always behave the way one wants depending on the linker used and the architecture. Instead, introduce rmodtool which takes an ELF file as an input, parses it, and creates a new ELF file in the format the rmodule loader expects. Change-Id: I31ac2d327d450ef841c3a7d9740b787278382bef Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: http://review.coreboot.org/5378 Tested-by: build bot (Jenkins) Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2014-03-07 22:11:53 +01:00
if (total_size != addr) {
ret = add_section(ew, NULL, ".empty", addr, total_size - addr);
if (ret < 0)
goto out;
}
/*
* Ensure last section has a memory usage that meets the required
* total size of the program in memory.
*/
ret = elf_writer_serialize(ew, out);
if (ret < 0)
ERROR("Failed to serialize ELF to buffer.\n");
out:
buffer_delete(&rmod_data);
elf_writer_destroy(ew);
return ret;
}
int rmodule_create(const struct buffer *elfin, struct buffer *elfout)
{
struct rmod_context ctx;
struct parsed_elf *pelf;
int i;
int ret;
ret = -1;
memset(&ctx, 0, sizeof(ctx));
pelf = &ctx.pelf;
if (parse_elf(elfin, pelf, ELF_PARSE_ALL)) {
ERROR("Couldn't parse ELF!\n");
return -1;
}
/* Only allow executables to be turned into rmodules. */
if (pelf->ehdr.e_type != ET_EXEC) {
ERROR("ELF is not an executable: %u.\n", pelf->ehdr.e_type);
goto out;
}
/* Determine if architecture is supported. */
for (i = 0; i < ARRAY_SIZE(reloc_ops); i++) {
if (reloc_ops[i].arch == pelf->ehdr.e_machine) {
ctx.ops = &reloc_ops[i];
break;
}
}
if (ctx.ops == NULL) {
ERROR("ELF is unsupported arch: %u.\n", pelf->ehdr.e_machine);
goto out;
}
/* Set the endian ops. */
if (ctx.pelf.ehdr.e_ident[EI_DATA] == ELFDATA2MSB)
ctx.xdr = &xdr_be;
else
ctx.xdr = &xdr_le;
if (find_program_segment(&ctx))
goto out;
if (filter_relocation_sections(&ctx))
goto out;
if (collect_relocations(&ctx))
goto out;
if (populate_program_info(&ctx))
goto out;
if (write_elf(&ctx, elfin, elfout))
goto out;
ret = 0;
out:
free(ctx.emitted_relocs);
parsed_elf_destroy(pelf);
return ret;
}