linux/arch/arm/mm/proc-arm946.S
Lorenzo Pieralisi 031bd879f7 ARM: mm: implement LoUIS API for cache maintenance ops
ARM v7 architecture introduced the concept of cache levels and related
control registers. New processors like A7 and A15 embed an L2 unified cache
controller that becomes part of the cache level hierarchy. Some operations in
the kernel like cpu_suspend and __cpu_disable do not require a flush of the
entire cache hierarchy to DRAM but just the cache levels belonging to the
Level of Unification Inner Shareable (LoUIS), which in most of ARM v7 systems
correspond to L1.

The current cache flushing API used in cpu_suspend and __cpu_disable,
flush_cache_all(), ends up flushing the whole cache hierarchy since for
v7 it cleans and invalidates all cache levels up to Level of Coherency
(LoC) which cripples system performance when used in hot paths like hotplug
and cpuidle.

Therefore a new kernel cache maintenance API must be added to cope with
latest ARM system requirements.

This patch adds flush_cache_louis() to the ARM kernel cache maintenance API.

This function cleans and invalidates all data cache levels up to the
Level of Unification Inner Shareable (LoUIS) and invalidates the instruction
cache for processors that support it (> v7).

This patch also creates an alias of the cache LoUIS function to flush_kern_all
for all processor versions prior to v7, so that the current cache flushing
behaviour is unchanged for those processors.

v7 cache maintenance code implements a cache LoUIS function that cleans and
invalidates the D-cache up to LoUIS and invalidates the I-cache, according
to the new API.

Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-by: Nicolas Pitre <nico@linaro.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Tested-by: Shawn Guo <shawn.guo@linaro.org>
2012-09-25 11:20:25 +01:00

432 lines
11 KiB
ArmAsm

/*
* linux/arch/arm/mm/arm946.S: utility functions for ARM946E-S
*
* Copyright (C) 2004-2006 Hyok S. Choi (hyok.choi@samsung.com)
*
* (Many of cache codes are from proc-arm926.S)
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*
*/
#include <linux/linkage.h>
#include <linux/init.h>
#include <asm/assembler.h>
#include <asm/hwcap.h>
#include <asm/pgtable-hwdef.h>
#include <asm/pgtable.h>
#include <asm/ptrace.h>
#include "proc-macros.S"
/*
* ARM946E-S is synthesizable to have 0KB to 1MB sized D-Cache,
* comprising 256 lines of 32 bytes (8 words).
*/
#define CACHE_DSIZE (CONFIG_CPU_DCACHE_SIZE) /* typically 8KB. */
#define CACHE_DLINESIZE 32 /* fixed */
#define CACHE_DSEGMENTS 4 /* fixed */
#define CACHE_DENTRIES (CACHE_DSIZE / CACHE_DSEGMENTS / CACHE_DLINESIZE)
#define CACHE_DLIMIT (CACHE_DSIZE * 4) /* benchmark needed */
.text
/*
* cpu_arm946_proc_init()
* cpu_arm946_switch_mm()
*
* These are not required.
*/
ENTRY(cpu_arm946_proc_init)
ENTRY(cpu_arm946_switch_mm)
mov pc, lr
/*
* cpu_arm946_proc_fin()
*/
ENTRY(cpu_arm946_proc_fin)
mrc p15, 0, r0, c1, c0, 0 @ ctrl register
bic r0, r0, #0x00001000 @ i-cache
bic r0, r0, #0x00000004 @ d-cache
mcr p15, 0, r0, c1, c0, 0 @ disable caches
mov pc, lr
/*
* cpu_arm946_reset(loc)
* Params : r0 = address to jump to
* Notes : This sets up everything for a reset
*/
.pushsection .idmap.text, "ax"
ENTRY(cpu_arm946_reset)
mov ip, #0
mcr p15, 0, ip, c7, c5, 0 @ flush I cache
mcr p15, 0, ip, c7, c6, 0 @ flush D cache
mcr p15, 0, ip, c7, c10, 4 @ drain WB
mrc p15, 0, ip, c1, c0, 0 @ ctrl register
bic ip, ip, #0x00000005 @ .............c.p
bic ip, ip, #0x00001000 @ i-cache
mcr p15, 0, ip, c1, c0, 0 @ ctrl register
mov pc, r0
ENDPROC(cpu_arm946_reset)
.popsection
/*
* cpu_arm946_do_idle()
*/
.align 5
ENTRY(cpu_arm946_do_idle)
mcr p15, 0, r0, c7, c0, 4 @ Wait for interrupt
mov pc, lr
/*
* flush_icache_all()
*
* Unconditionally clean and invalidate the entire icache.
*/
ENTRY(arm946_flush_icache_all)
mov r0, #0
mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache
mov pc, lr
ENDPROC(arm946_flush_icache_all)
/*
* flush_user_cache_all()
*/
ENTRY(arm946_flush_user_cache_all)
/* FALLTHROUGH */
/*
* flush_kern_cache_all()
*
* Clean and invalidate the entire cache.
*/
ENTRY(arm946_flush_kern_cache_all)
mov r2, #VM_EXEC
mov ip, #0
__flush_whole_cache:
#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
mcr p15, 0, ip, c7, c6, 0 @ flush D cache
#else
mov r1, #(CACHE_DSEGMENTS - 1) << 29 @ 4 segments
1: orr r3, r1, #(CACHE_DENTRIES - 1) << 4 @ n entries
2: mcr p15, 0, r3, c7, c14, 2 @ clean/flush D index
subs r3, r3, #1 << 4
bcs 2b @ entries n to 0
subs r1, r1, #1 << 29
bcs 1b @ segments 3 to 0
#endif
tst r2, #VM_EXEC
mcrne p15, 0, ip, c7, c5, 0 @ flush I cache
mcrne p15, 0, ip, c7, c10, 4 @ drain WB
mov pc, lr
/*
* flush_user_cache_range(start, end, flags)
*
* Clean and invalidate a range of cache entries in the
* specified address range.
*
* - start - start address (inclusive)
* - end - end address (exclusive)
* - flags - vm_flags describing address space
* (same as arm926)
*/
ENTRY(arm946_flush_user_cache_range)
mov ip, #0
sub r3, r1, r0 @ calculate total size
cmp r3, #CACHE_DLIMIT
bhs __flush_whole_cache
1: tst r2, #VM_EXEC
#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
mcr p15, 0, r0, c7, c6, 1 @ invalidate D entry
mcrne p15, 0, r0, c7, c5, 1 @ invalidate I entry
add r0, r0, #CACHE_DLINESIZE
mcr p15, 0, r0, c7, c6, 1 @ invalidate D entry
mcrne p15, 0, r0, c7, c5, 1 @ invalidate I entry
add r0, r0, #CACHE_DLINESIZE
#else
mcr p15, 0, r0, c7, c14, 1 @ clean and invalidate D entry
mcrne p15, 0, r0, c7, c5, 1 @ invalidate I entry
add r0, r0, #CACHE_DLINESIZE
mcr p15, 0, r0, c7, c14, 1 @ clean and invalidate D entry
mcrne p15, 0, r0, c7, c5, 1 @ invalidate I entry
add r0, r0, #CACHE_DLINESIZE
#endif
cmp r0, r1
blo 1b
tst r2, #VM_EXEC
mcrne p15, 0, ip, c7, c10, 4 @ drain WB
mov pc, lr
/*
* coherent_kern_range(start, end)
*
* Ensure coherency between the Icache and the Dcache in the
* region described by start, end. If you have non-snooping
* Harvard caches, you need to implement this function.
*
* - start - virtual start address
* - end - virtual end address
*/
ENTRY(arm946_coherent_kern_range)
/* FALLTHROUGH */
/*
* coherent_user_range(start, end)
*
* Ensure coherency between the Icache and the Dcache in the
* region described by start, end. If you have non-snooping
* Harvard caches, you need to implement this function.
*
* - start - virtual start address
* - end - virtual end address
* (same as arm926)
*/
ENTRY(arm946_coherent_user_range)
bic r0, r0, #CACHE_DLINESIZE - 1
1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry
mcr p15, 0, r0, c7, c5, 1 @ invalidate I entry
add r0, r0, #CACHE_DLINESIZE
cmp r0, r1
blo 1b
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov r0, #0
mov pc, lr
/*
* flush_kern_dcache_area(void *addr, size_t size)
*
* Ensure no D cache aliasing occurs, either with itself or
* the I cache
*
* - addr - kernel address
* - size - region size
* (same as arm926)
*/
ENTRY(arm946_flush_kern_dcache_area)
add r1, r0, r1
1: mcr p15, 0, r0, c7, c14, 1 @ clean+invalidate D entry
add r0, r0, #CACHE_DLINESIZE
cmp r0, r1
blo 1b
mov r0, #0
mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
/*
* dma_inv_range(start, end)
*
* Invalidate (discard) the specified virtual address range.
* May not write back any entries. If 'start' or 'end'
* are not cache line aligned, those lines must be written
* back.
*
* - start - virtual start address
* - end - virtual end address
* (same as arm926)
*/
arm946_dma_inv_range:
#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
tst r0, #CACHE_DLINESIZE - 1
mcrne p15, 0, r0, c7, c10, 1 @ clean D entry
tst r1, #CACHE_DLINESIZE - 1
mcrne p15, 0, r1, c7, c10, 1 @ clean D entry
#endif
bic r0, r0, #CACHE_DLINESIZE - 1
1: mcr p15, 0, r0, c7, c6, 1 @ invalidate D entry
add r0, r0, #CACHE_DLINESIZE
cmp r0, r1
blo 1b
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
/*
* dma_clean_range(start, end)
*
* Clean the specified virtual address range.
*
* - start - virtual start address
* - end - virtual end address
*
* (same as arm926)
*/
arm946_dma_clean_range:
#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
bic r0, r0, #CACHE_DLINESIZE - 1
1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry
add r0, r0, #CACHE_DLINESIZE
cmp r0, r1
blo 1b
#endif
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
/*
* dma_flush_range(start, end)
*
* Clean and invalidate the specified virtual address range.
*
* - start - virtual start address
* - end - virtual end address
*
* (same as arm926)
*/
ENTRY(arm946_dma_flush_range)
bic r0, r0, #CACHE_DLINESIZE - 1
1:
#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
mcr p15, 0, r0, c7, c14, 1 @ clean+invalidate D entry
#else
mcr p15, 0, r0, c7, c6, 1 @ invalidate D entry
#endif
add r0, r0, #CACHE_DLINESIZE
cmp r0, r1
blo 1b
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
/*
* dma_map_area(start, size, dir)
* - start - kernel virtual start address
* - size - size of region
* - dir - DMA direction
*/
ENTRY(arm946_dma_map_area)
add r1, r1, r0
cmp r2, #DMA_TO_DEVICE
beq arm946_dma_clean_range
bcs arm946_dma_inv_range
b arm946_dma_flush_range
ENDPROC(arm946_dma_map_area)
/*
* dma_unmap_area(start, size, dir)
* - start - kernel virtual start address
* - size - size of region
* - dir - DMA direction
*/
ENTRY(arm946_dma_unmap_area)
mov pc, lr
ENDPROC(arm946_dma_unmap_area)
.globl arm946_flush_kern_cache_louis
.equ arm946_flush_kern_cache_louis, arm946_flush_kern_cache_all
@ define struct cpu_cache_fns (see <asm/cacheflush.h> and proc-macros.S)
define_cache_functions arm946
ENTRY(cpu_arm946_dcache_clean_area)
#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry
add r0, r0, #CACHE_DLINESIZE
subs r1, r1, #CACHE_DLINESIZE
bhi 1b
#endif
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
__CPUINIT
.type __arm946_setup, #function
__arm946_setup:
mov r0, #0
mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache
mcr p15, 0, r0, c7, c6, 0 @ invalidate D cache
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mcr p15, 0, r0, c6, c3, 0 @ disable memory region 3~7
mcr p15, 0, r0, c6, c4, 0
mcr p15, 0, r0, c6, c5, 0
mcr p15, 0, r0, c6, c6, 0
mcr p15, 0, r0, c6, c7, 0
mov r0, #0x0000003F @ base = 0, size = 4GB
mcr p15, 0, r0, c6, c0, 0 @ set region 0, default
ldr r0, =(CONFIG_DRAM_BASE & 0xFFFFF000) @ base[31:12] of RAM
ldr r1, =(CONFIG_DRAM_SIZE >> 12) @ size of RAM (must be >= 4KB)
mov r2, #10 @ 11 is the minimum (4KB)
1: add r2, r2, #1 @ area size *= 2
mov r1, r1, lsr #1
bne 1b @ count not zero r-shift
orr r0, r0, r2, lsl #1 @ the region register value
orr r0, r0, #1 @ set enable bit
mcr p15, 0, r0, c6, c1, 0 @ set region 1, RAM
ldr r0, =(CONFIG_FLASH_MEM_BASE & 0xFFFFF000) @ base[31:12] of FLASH
ldr r1, =(CONFIG_FLASH_SIZE >> 12) @ size of FLASH (must be >= 4KB)
mov r2, #10 @ 11 is the minimum (4KB)
1: add r2, r2, #1 @ area size *= 2
mov r1, r1, lsr #1
bne 1b @ count not zero r-shift
orr r0, r0, r2, lsl #1 @ the region register value
orr r0, r0, #1 @ set enable bit
mcr p15, 0, r0, c6, c2, 0 @ set region 2, ROM/FLASH
mov r0, #0x06
mcr p15, 0, r0, c2, c0, 0 @ region 1,2 d-cacheable
mcr p15, 0, r0, c2, c0, 1 @ region 1,2 i-cacheable
#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
mov r0, #0x00 @ disable whole write buffer
#else
mov r0, #0x02 @ region 1 write bufferred
#endif
mcr p15, 0, r0, c3, c0, 0
/*
* Access Permission Settings for future permission control by PU.
*
* priv. user
* region 0 (whole) rw -- : b0001
* region 1 (RAM) rw rw : b0011
* region 2 (FLASH) rw r- : b0010
* region 3~7 (none) -- -- : b0000
*/
mov r0, #0x00000031
orr r0, r0, #0x00000200
mcr p15, 0, r0, c5, c0, 2 @ set data access permission
mcr p15, 0, r0, c5, c0, 3 @ set inst. access permission
mrc p15, 0, r0, c1, c0 @ get control register
orr r0, r0, #0x00001000 @ I-cache
orr r0, r0, #0x00000005 @ MPU/D-cache
#ifdef CONFIG_CPU_CACHE_ROUND_ROBIN
orr r0, r0, #0x00004000 @ .1.. .... .... ....
#endif
mov pc, lr
.size __arm946_setup, . - __arm946_setup
__INITDATA
@ define struct processor (see <asm/proc-fns.h> and proc-macros.S)
define_processor_functions arm946, dabort=nommu_early_abort, pabort=legacy_pabort, nommu=1
.section ".rodata"
string cpu_arch_name, "armv5te"
string cpu_elf_name, "v5t"
string cpu_arm946_name, "ARM946E-S"
.align
.section ".proc.info.init", #alloc, #execinstr
.type __arm946_proc_info,#object
__arm946_proc_info:
.long 0x41009460
.long 0xff00fff0
.long 0
.long 0
b __arm946_setup
.long cpu_arch_name
.long cpu_elf_name
.long HWCAP_SWP | HWCAP_HALF | HWCAP_THUMB
.long cpu_arm946_name
.long arm946_processor_functions
.long 0
.long 0
.long arm946_cache_fns
.size __arm946_proc_info, . - __arm946_proc_info