mirror of
https://github.com/zephyrproject-rtos/zephyr
synced 2025-09-01 12:07:04 +00:00
6b3c5e8bb2
Binutils ld has an annoying misfeature (apparently a regression from a
few years ago) that alignment directives (and alignment specifiers on
symbols) apply only to the runtime addresses and not, apparently, to
the load address region specified with the "AT>" syntax. The net
result is that by default the LMA output ends up too small for the
addresses generated in RAM. See here for some details:
https://sourceware.org/ml/binutils/2013-06/msg00246.html
https://sourceware.org/ml/binutils/2014-01/msg00350.html
The required workaround/fix is that AFAICT any section which can have
inherit a separate VMA vs. LMA from a previous section must specify an
"ALIGN_WITH_INPUT" attribute. Otherwise the sections will get out of
sync and the XIP data will be wrong at runtime.
No, I don't know why this isn't the default behavior.
A further complexity is that this feature only works as advertised
when the section is declared with the "AT> region" syntax after the
block and not "AT(address)" in the header. If you use the header
syntax (with or without ALIGN_WITH_INPUT), ld appears to DOUBLE-apply
padding and the LMA ends up to big. This is almost certainly a
binutils bug, but it's trivial to work around (and the working syntax
is actually cleaner) so we adjust the usage here.
Note finally that this patch includes an effective reversion of commit
d82e9dd9 ("x86: HACK force alignment for _k_task_list section"), which
was an earlier workaround for what seems to be the same issue.
Jira: ZEP-955
Change-Id: I2accd92901cb61fb546658b87d6752c1cd14de3a
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
116 lines
3.8 KiB
C
116 lines
3.8 KiB
C
/*
|
|
* Copyright (c) 2013-2014, Wind River Systems, Inc.
|
|
*
|
|
* Licensed under the Apache License, Version 2.0 (the "License");
|
|
* you may not use this file except in compliance with the License.
|
|
* You may obtain a copy of the License at
|
|
*
|
|
* http://www.apache.org/licenses/LICENSE-2.0
|
|
*
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
* distributed under the License is distributed on an "AS IS" BASIS,
|
|
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
* See the License for the specific language governing permissions and
|
|
* limitations under the License.
|
|
*/
|
|
|
|
/**
|
|
* @file
|
|
* @brief GCC toolchain linker defs
|
|
*
|
|
* This header file defines the necessary macros used by the linker script for
|
|
* use with the GCC linker.
|
|
*/
|
|
|
|
#ifndef __LINKER_TOOL_GCC_H
|
|
#define __LINKER_TOOL_GCC_H
|
|
|
|
#if defined(CONFIG_ARM)
|
|
OUTPUT_FORMAT("elf32-littlearm", "elf32-bigarm", "elf32-littlearm")
|
|
#elif defined(CONFIG_ARC)
|
|
OUTPUT_FORMAT("elf32-littlearc", "elf32-bigarc", "elf32-littlearc")
|
|
#elif defined(CONFIG_X86)
|
|
#if defined(__IAMCU)
|
|
OUTPUT_FORMAT("elf32-iamcu")
|
|
OUTPUT_ARCH(iamcu:intel)
|
|
#else
|
|
OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
|
|
OUTPUT_ARCH(i386)
|
|
#endif
|
|
#elif defined(CONFIG_NIOS2)
|
|
OUTPUT_FORMAT("elf32-littlenios2", "elf32-bignios2", "elf32-littlenios2")
|
|
#else
|
|
#error Arch not supported.
|
|
#endif
|
|
|
|
/*
|
|
* The GROUP_START() and GROUP_END() macros are used to define a group
|
|
* of sections located in one memory area, such as RAM, ROM, etc.
|
|
* The <where> parameter is the name of the memory area.
|
|
*/
|
|
#define GROUP_START(where)
|
|
#define GROUP_END(where)
|
|
|
|
/*
|
|
* The GROUP_LINK_IN() macro is located at the end of the section
|
|
* description and tells the linker that this section is located in
|
|
* the memory area specified by <where> argument.
|
|
*/
|
|
#define GROUP_LINK_IN(where) > where
|
|
|
|
/*
|
|
* As GROUP_LINK_IN(), but takes a second argument indicating the
|
|
* memory region (e.g. "ROM") for the load address. Used for
|
|
* initialized data sections that on XIP platforms must be copied at
|
|
* startup.
|
|
*
|
|
* And, because output directives in GNU ld are "sticky", this must
|
|
* also be used on the first section *after* such an initialized data
|
|
* section, specifying the same memory region (e.g. "RAM") for both
|
|
* vregion and lregion.
|
|
*/
|
|
#ifdef CONFIG_XIP
|
|
#define GROUP_DATA_LINK_IN(vregion, lregion) > vregion AT> lregion
|
|
#else
|
|
#define GROUP_DATA_LINK_IN(vregion, lregion) > vregion
|
|
#endif
|
|
|
|
/*
|
|
* The GROUP_FOLLOWS_AT() macro is located at the end of the section
|
|
* and indicates that the section does not specify an address at which
|
|
* it is to be loaded, but that it follows a section which did specify
|
|
* such an address
|
|
*/
|
|
#define GROUP_FOLLOWS_AT(where) AT > where
|
|
|
|
/*
|
|
* The SECTION_PROLOGUE() macro is used to define the beginning of a section.
|
|
* The <name> parameter is the name of the section, and the <option> parameter
|
|
* is to include any special options such as (NOLOAD). Page alignment has its
|
|
* own parameter since it needs abstraction across the different toolchains.
|
|
* If not required, the <options> and <align> parameters should be left blank.
|
|
*/
|
|
|
|
#define SECTION_PROLOGUE(name, options, align) name options : align
|
|
|
|
/*
|
|
* As for SECTION_PROLOGUE(), except that this one must (!) be used
|
|
* for data sections which on XIP platforms will have differeing
|
|
* virtual and load addresses (i.e. they'll be copied into RAM at
|
|
* program startup). Such a section must (!) also use
|
|
* GROUP_LINK_IN_LMA to specify the correct output load address.
|
|
*/
|
|
#ifdef CONFIG_XIP
|
|
#define SECTION_DATA_PROLOGUE(name, options, align) \
|
|
name options : ALIGN_WITH_INPUT align
|
|
#else
|
|
#define SECTION_DATA_PROLOGUE(name, options, align) name options : align
|
|
#endif
|
|
|
|
#define SORT_BY_NAME(x) SORT(x)
|
|
#define OPTIONAL
|
|
|
|
#define COMMON_SYMBOLS *(COMMON)
|
|
|
|
#endif /* !__LINKER_TOOL_GCC_H */
|