arm-trusted-firmware/include/lib/utils.h

95 lines
2.9 KiB
C
Raw Normal View History

/*
* Copyright (c) 2016-2019, ARM Limited and Contributors. All rights reserved.
*
* SPDX-License-Identifier: BSD-3-Clause
*/
#ifndef UTILS_H
#define UTILS_H
Introduce unified API to zero memory Introduce zeromem_dczva function on AArch64 that can handle unaligned addresses and make use of DC ZVA instruction to zero a whole block at a time. This zeroing takes place directly in the cache to speed it up without doing external memory access. Remove the zeromem16 function on AArch64 and replace it with an alias to zeromem. This zeromem16 function is now deprecated. Remove the 16-bytes alignment constraint on __BSS_START__ in firmware-design.md as it is now not mandatory anymore (it used to comply with zeromem16 requirements). Change the 16-bytes alignment constraints in SP min's linker script to a 8-bytes alignment constraint as the AArch32 zeromem implementation is now more efficient on 8-bytes aligned addresses. Introduce zero_normalmem and zeromem helpers in platform agnostic header that are implemented this way: * AArch32: * zero_normalmem: zero using usual data access * zeromem: alias for zero_normalmem * AArch64: * zero_normalmem: zero normal memory using DC ZVA instruction (needs MMU enabled) * zeromem: zero using usual data access Usage guidelines: in most cases, zero_normalmem should be preferred. There are 2 scenarios where zeromem (or memset) must be used instead: * Code that must run with MMU disabled (which means all memory is considered device memory for data accesses). * Code that fills device memory with null bytes. Optionally, the following rule can be applied if performance is important: * Code zeroing small areas (few bytes) that are not secrets should use memset to take advantage of compiler optimizations. Note: Code zeroing security-related critical information should use zero_normalmem/zeromem instead of memset to avoid removal by compilers' optimizations in some cases or misbehaving versions of GCC. Fixes ARM-software/tf-issues#408 Change-Id: Iafd9663fc1070413c3e1904e54091cf60effaa82 Signed-off-by: Douglas Raillard <douglas.raillard@arm.com>
2016-12-02 13:51:54 +00:00
/*
* C code should be put in this part of the header to avoid breaking ASM files
* or linker scripts including it.
*/
#if !(defined(__LINKER__) || defined(__ASSEMBLER__))
Introduce unified API to zero memory Introduce zeromem_dczva function on AArch64 that can handle unaligned addresses and make use of DC ZVA instruction to zero a whole block at a time. This zeroing takes place directly in the cache to speed it up without doing external memory access. Remove the zeromem16 function on AArch64 and replace it with an alias to zeromem. This zeromem16 function is now deprecated. Remove the 16-bytes alignment constraint on __BSS_START__ in firmware-design.md as it is now not mandatory anymore (it used to comply with zeromem16 requirements). Change the 16-bytes alignment constraints in SP min's linker script to a 8-bytes alignment constraint as the AArch32 zeromem implementation is now more efficient on 8-bytes aligned addresses. Introduce zero_normalmem and zeromem helpers in platform agnostic header that are implemented this way: * AArch32: * zero_normalmem: zero using usual data access * zeromem: alias for zero_normalmem * AArch64: * zero_normalmem: zero normal memory using DC ZVA instruction (needs MMU enabled) * zeromem: zero using usual data access Usage guidelines: in most cases, zero_normalmem should be preferred. There are 2 scenarios where zeromem (or memset) must be used instead: * Code that must run with MMU disabled (which means all memory is considered device memory for data accesses). * Code that fills device memory with null bytes. Optionally, the following rule can be applied if performance is important: * Code zeroing small areas (few bytes) that are not secrets should use memset to take advantage of compiler optimizations. Note: Code zeroing security-related critical information should use zero_normalmem/zeromem instead of memset to avoid removal by compilers' optimizations in some cases or misbehaving versions of GCC. Fixes ARM-software/tf-issues#408 Change-Id: Iafd9663fc1070413c3e1904e54091cf60effaa82 Signed-off-by: Douglas Raillard <douglas.raillard@arm.com>
2016-12-02 13:51:54 +00:00
#include <stddef.h>
#include <stdint.h>
Introduce unified API to zero memory Introduce zeromem_dczva function on AArch64 that can handle unaligned addresses and make use of DC ZVA instruction to zero a whole block at a time. This zeroing takes place directly in the cache to speed it up without doing external memory access. Remove the zeromem16 function on AArch64 and replace it with an alias to zeromem. This zeromem16 function is now deprecated. Remove the 16-bytes alignment constraint on __BSS_START__ in firmware-design.md as it is now not mandatory anymore (it used to comply with zeromem16 requirements). Change the 16-bytes alignment constraints in SP min's linker script to a 8-bytes alignment constraint as the AArch32 zeromem implementation is now more efficient on 8-bytes aligned addresses. Introduce zero_normalmem and zeromem helpers in platform agnostic header that are implemented this way: * AArch32: * zero_normalmem: zero using usual data access * zeromem: alias for zero_normalmem * AArch64: * zero_normalmem: zero normal memory using DC ZVA instruction (needs MMU enabled) * zeromem: zero using usual data access Usage guidelines: in most cases, zero_normalmem should be preferred. There are 2 scenarios where zeromem (or memset) must be used instead: * Code that must run with MMU disabled (which means all memory is considered device memory for data accesses). * Code that fills device memory with null bytes. Optionally, the following rule can be applied if performance is important: * Code zeroing small areas (few bytes) that are not secrets should use memset to take advantage of compiler optimizations. Note: Code zeroing security-related critical information should use zero_normalmem/zeromem instead of memset to avoid removal by compilers' optimizations in some cases or misbehaving versions of GCC. Fixes ARM-software/tf-issues#408 Change-Id: Iafd9663fc1070413c3e1904e54091cf60effaa82 Signed-off-by: Douglas Raillard <douglas.raillard@arm.com>
2016-12-02 13:51:54 +00:00
typedef struct mem_region {
uintptr_t base;
size_t nbytes;
} mem_region_t;
/*
* zero_normalmem all the regions defined in tbl.
*/
void clear_mem_regions(mem_region_t *tbl, size_t nregions);
/*
* zero_normalmem all the regions defined in region. It dynamically
* maps chunks of 'chunk_size' in 'va' virtual address and clears them.
* For this reason memory regions must be multiple of chunk_size and
* must be aligned to it as well. chunk_size and va can be selected
* in a way that they minimize the number of entries used in the
* translation tables.
*/
void clear_map_dyn_mem_regions(struct mem_region *regions,
size_t nregions,
uintptr_t va,
size_t chunk);
/*
* checks that a region (addr + nbytes-1) of memory is totally covered by
* one of the regions defined in tbl. Caller must ensure that (addr+nbytes-1)
* doesn't overflow.
*/
int mem_region_in_array_chk(mem_region_t *tbl, size_t nregions,
uintptr_t addr, size_t nbytes);
Introduce unified API to zero memory Introduce zeromem_dczva function on AArch64 that can handle unaligned addresses and make use of DC ZVA instruction to zero a whole block at a time. This zeroing takes place directly in the cache to speed it up without doing external memory access. Remove the zeromem16 function on AArch64 and replace it with an alias to zeromem. This zeromem16 function is now deprecated. Remove the 16-bytes alignment constraint on __BSS_START__ in firmware-design.md as it is now not mandatory anymore (it used to comply with zeromem16 requirements). Change the 16-bytes alignment constraints in SP min's linker script to a 8-bytes alignment constraint as the AArch32 zeromem implementation is now more efficient on 8-bytes aligned addresses. Introduce zero_normalmem and zeromem helpers in platform agnostic header that are implemented this way: * AArch32: * zero_normalmem: zero using usual data access * zeromem: alias for zero_normalmem * AArch64: * zero_normalmem: zero normal memory using DC ZVA instruction (needs MMU enabled) * zeromem: zero using usual data access Usage guidelines: in most cases, zero_normalmem should be preferred. There are 2 scenarios where zeromem (or memset) must be used instead: * Code that must run with MMU disabled (which means all memory is considered device memory for data accesses). * Code that fills device memory with null bytes. Optionally, the following rule can be applied if performance is important: * Code zeroing small areas (few bytes) that are not secrets should use memset to take advantage of compiler optimizations. Note: Code zeroing security-related critical information should use zero_normalmem/zeromem instead of memset to avoid removal by compilers' optimizations in some cases or misbehaving versions of GCC. Fixes ARM-software/tf-issues#408 Change-Id: Iafd9663fc1070413c3e1904e54091cf60effaa82 Signed-off-by: Douglas Raillard <douglas.raillard@arm.com>
2016-12-02 13:51:54 +00:00
/*
* Fill a region of normal memory of size "length" in bytes with zero bytes.
*
* WARNING: This function can only operate on normal memory. This means that
* the MMU must be enabled when using this function. Otherwise, use
* zeromem.
*/
void zero_normalmem(void *mem, u_register_t length);
/*
* Fill a region of memory of size "length" in bytes with null bytes.
*
* Unlike zero_normalmem, this function has no restriction on the type of
* memory targeted and can be used for any device memory as well as normal
* memory. This function must be used instead of zero_normalmem when MMU is
* disabled.
*
* NOTE: When data cache and MMU are enabled, prefer zero_normalmem for faster
* zeroing.
*/
void zeromem(void *mem, u_register_t length);
/*
* Utility function to return the address of a symbol. By default, the
* compiler generates adr/adrp instruction pair to return the reference
* to the symbol and this utility is used to override this compiler
* generated to code to use `ldr` instruction.
*
* This helps when Position Independent Executable needs to reference a symbol
* which is constant and does not depend on the execute address of the binary.
*/
#define DEFINE_LOAD_SYM_ADDR(_name) \
static inline u_register_t load_addr_## _name(void) \
{ \
u_register_t v; \
__asm__ volatile ("ldr %0, =" #_name : "=r" (v) : "X" (#_name));\
return v; \
}
/* Helper to invoke the function defined by DEFINE_LOAD_SYM_ADDR() */
#define LOAD_ADDR_OF(_name) (typeof(_name) *) load_addr_## _name()
#endif /* !(defined(__LINKER__) || defined(__ASSEMBLER__)) */
Introduce unified API to zero memory Introduce zeromem_dczva function on AArch64 that can handle unaligned addresses and make use of DC ZVA instruction to zero a whole block at a time. This zeroing takes place directly in the cache to speed it up without doing external memory access. Remove the zeromem16 function on AArch64 and replace it with an alias to zeromem. This zeromem16 function is now deprecated. Remove the 16-bytes alignment constraint on __BSS_START__ in firmware-design.md as it is now not mandatory anymore (it used to comply with zeromem16 requirements). Change the 16-bytes alignment constraints in SP min's linker script to a 8-bytes alignment constraint as the AArch32 zeromem implementation is now more efficient on 8-bytes aligned addresses. Introduce zero_normalmem and zeromem helpers in platform agnostic header that are implemented this way: * AArch32: * zero_normalmem: zero using usual data access * zeromem: alias for zero_normalmem * AArch64: * zero_normalmem: zero normal memory using DC ZVA instruction (needs MMU enabled) * zeromem: zero using usual data access Usage guidelines: in most cases, zero_normalmem should be preferred. There are 2 scenarios where zeromem (or memset) must be used instead: * Code that must run with MMU disabled (which means all memory is considered device memory for data accesses). * Code that fills device memory with null bytes. Optionally, the following rule can be applied if performance is important: * Code zeroing small areas (few bytes) that are not secrets should use memset to take advantage of compiler optimizations. Note: Code zeroing security-related critical information should use zero_normalmem/zeromem instead of memset to avoid removal by compilers' optimizations in some cases or misbehaving versions of GCC. Fixes ARM-software/tf-issues#408 Change-Id: Iafd9663fc1070413c3e1904e54091cf60effaa82 Signed-off-by: Douglas Raillard <douglas.raillard@arm.com>
2016-12-02 13:51:54 +00:00
#endif /* UTILS_H */