linux/arch/arm/mm/proc-mohawk.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

458 lines
12 KiB
ArmAsm

/*
* linux/arch/arm/mm/proc-mohawk.S: MMU functions for Marvell PJ1 core
*
* PJ1 (codename Mohawk) is a hybrid of the xscale3 and Marvell's own core.
*
* Heavily based on proc-arm926.S and proc-xsc3.S
*
* 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; either version 2 of the License, or
* (at your option) any later version.
*
* 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
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
#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/page.h>
#include <asm/ptrace.h>
#include "proc-macros.S"
/*
* This is the maximum size of an area which will be flushed. If the
* area is larger than this, then we flush the whole cache.
*/
#define CACHE_DLIMIT 32768
/*
* The cache line size of the L1 D cache.
*/
#define CACHE_DLINESIZE 32
/*
* cpu_mohawk_proc_init()
*/
ENTRY(cpu_mohawk_proc_init)
mov pc, lr
/*
* cpu_mohawk_proc_fin()
*/
ENTRY(cpu_mohawk_proc_fin)
mrc p15, 0, r0, c1, c0, 0 @ ctrl register
bic r0, r0, #0x1800 @ ...iz...........
bic r0, r0, #0x0006 @ .............ca.
mcr p15, 0, r0, c1, c0, 0 @ disable caches
mov pc, lr
/*
* cpu_mohawk_reset(loc)
*
* Perform a soft reset of the system. Put the CPU into the
* same state as it would be if it had been reset, and branch
* to what would be the reset vector.
*
* loc: location to jump to for soft reset
*
* (same as arm926)
*/
.align 5
.pushsection .idmap.text, "ax"
ENTRY(cpu_mohawk_reset)
mov ip, #0
mcr p15, 0, ip, c7, c7, 0 @ invalidate I,D caches
mcr p15, 0, ip, c7, c10, 4 @ drain WB
mcr p15, 0, ip, c8, c7, 0 @ invalidate I & D TLBs
mrc p15, 0, ip, c1, c0, 0 @ ctrl register
bic ip, ip, #0x0007 @ .............cam
bic ip, ip, #0x1100 @ ...i...s........
mcr p15, 0, ip, c1, c0, 0 @ ctrl register
mov pc, r0
ENDPROC(cpu_mohawk_reset)
.popsection
/*
* cpu_mohawk_do_idle()
*
* Called with IRQs disabled
*/
.align 5
ENTRY(cpu_mohawk_do_idle)
mov r0, #0
mcr p15, 0, r0, c7, c10, 4 @ drain write buffer
mcr p15, 0, r0, c7, c0, 4 @ wait for interrupt
mov pc, lr
/*
* flush_icache_all()
*
* Unconditionally clean and invalidate the entire icache.
*/
ENTRY(mohawk_flush_icache_all)
mov r0, #0
mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache
mov pc, lr
ENDPROC(mohawk_flush_icache_all)
/*
* flush_user_cache_all()
*
* Clean and invalidate all cache entries in a particular
* address space.
*/
ENTRY(mohawk_flush_user_cache_all)
/* FALLTHROUGH */
/*
* flush_kern_cache_all()
*
* Clean and invalidate the entire cache.
*/
ENTRY(mohawk_flush_kern_cache_all)
mov r2, #VM_EXEC
mov ip, #0
__flush_whole_cache:
mcr p15, 0, ip, c7, c14, 0 @ clean & invalidate all D cache
tst r2, #VM_EXEC
mcrne p15, 0, ip, c7, c5, 0 @ invalidate I cache
mcrne p15, 0, ip, c7, c10, 0 @ drain write buffer
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(mohawk_flush_user_cache_range)
mov ip, #0
sub r3, r1, r0 @ calculate total size
cmp r3, #CACHE_DLIMIT
bgt __flush_whole_cache
1: tst r2, #VM_EXEC
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
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(mohawk_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(mohawk_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
*/
ENTRY(mohawk_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 v4wb)
*/
mohawk_dma_inv_range:
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
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 v4wb)
*/
mohawk_dma_clean_range:
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
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
*/
ENTRY(mohawk_dma_flush_range)
bic r0, r0, #CACHE_DLINESIZE - 1
1:
mcr p15, 0, r0, c7, c14, 1 @ clean+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_map_area(start, size, dir)
* - start - kernel virtual start address
* - size - size of region
* - dir - DMA direction
*/
ENTRY(mohawk_dma_map_area)
add r1, r1, r0
cmp r2, #DMA_TO_DEVICE
beq mohawk_dma_clean_range
bcs mohawk_dma_inv_range
b mohawk_dma_flush_range
ENDPROC(mohawk_dma_map_area)
/*
* dma_unmap_area(start, size, dir)
* - start - kernel virtual start address
* - size - size of region
* - dir - DMA direction
*/
ENTRY(mohawk_dma_unmap_area)
mov pc, lr
ENDPROC(mohawk_dma_unmap_area)
.globl mohawk_flush_kern_cache_louis
.equ mohawk_flush_kern_cache_louis, mohawk_flush_kern_cache_all
@ define struct cpu_cache_fns (see <asm/cacheflush.h> and proc-macros.S)
define_cache_functions mohawk
ENTRY(cpu_mohawk_dcache_clean_area)
1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry
add r0, r0, #CACHE_DLINESIZE
subs r1, r1, #CACHE_DLINESIZE
bhi 1b
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
/*
* cpu_mohawk_switch_mm(pgd)
*
* Set the translation base pointer to be as described by pgd.
*
* pgd: new page tables
*/
.align 5
ENTRY(cpu_mohawk_switch_mm)
mov ip, #0
mcr p15, 0, ip, c7, c14, 0 @ clean & invalidate all D cache
mcr p15, 0, ip, c7, c5, 0 @ invalidate I cache
mcr p15, 0, ip, c7, c10, 4 @ drain WB
orr r0, r0, #0x18 @ cache the page table in L2
mcr p15, 0, r0, c2, c0, 0 @ load page table pointer
mcr p15, 0, ip, c8, c7, 0 @ invalidate I & D TLBs
mov pc, lr
/*
* cpu_mohawk_set_pte_ext(ptep, pte, ext)
*
* Set a PTE and flush it out
*/
.align 5
ENTRY(cpu_mohawk_set_pte_ext)
armv3_set_pte_ext
mov r0, r0
mcr p15, 0, r0, c7, c10, 1 @ clean D entry
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
.globl cpu_mohawk_suspend_size
.equ cpu_mohawk_suspend_size, 4 * 6
#ifdef CONFIG_PM_SLEEP
ENTRY(cpu_mohawk_do_suspend)
stmfd sp!, {r4 - r9, lr}
mrc p14, 0, r4, c6, c0, 0 @ clock configuration, for turbo mode
mrc p15, 0, r5, c15, c1, 0 @ CP access reg
mrc p15, 0, r6, c13, c0, 0 @ PID
mrc p15, 0, r7, c3, c0, 0 @ domain ID
mrc p15, 0, r8, c1, c0, 1 @ auxiliary control reg
mrc p15, 0, r9, c1, c0, 0 @ control reg
bic r4, r4, #2 @ clear frequency change bit
stmia r0, {r4 - r9} @ store cp regs
ldmia sp!, {r4 - r9, pc}
ENDPROC(cpu_mohawk_do_suspend)
ENTRY(cpu_mohawk_do_resume)
ldmia r0, {r4 - r9} @ load cp regs
mov ip, #0
mcr p15, 0, ip, c7, c7, 0 @ invalidate I & D caches, BTB
mcr p15, 0, ip, c7, c10, 4 @ drain write (&fill) buffer
mcr p15, 0, ip, c7, c5, 4 @ flush prefetch buffer
mcr p15, 0, ip, c8, c7, 0 @ invalidate I & D TLBs
mcr p14, 0, r4, c6, c0, 0 @ clock configuration, turbo mode.
mcr p15, 0, r5, c15, c1, 0 @ CP access reg
mcr p15, 0, r6, c13, c0, 0 @ PID
mcr p15, 0, r7, c3, c0, 0 @ domain ID
orr r1, r1, #0x18 @ cache the page table in L2
mcr p15, 0, r1, c2, c0, 0 @ translation table base addr
mcr p15, 0, r8, c1, c0, 1 @ auxiliary control reg
mov r0, r9 @ control register
b cpu_resume_mmu
ENDPROC(cpu_mohawk_do_resume)
#endif
__CPUINIT
.type __mohawk_setup, #function
__mohawk_setup:
mov r0, #0
mcr p15, 0, r0, c7, c7 @ invalidate I,D caches
mcr p15, 0, r0, c7, c10, 4 @ drain write buffer
mcr p15, 0, r0, c8, c7 @ invalidate I,D TLBs
orr r4, r4, #0x18 @ cache the page table in L2
mcr p15, 0, r4, c2, c0, 0 @ load page table pointer
mov r0, #0 @ don't allow CP access
mcr p15, 0, r0, c15, c1, 0 @ write CP access register
adr r5, mohawk_crval
ldmia r5, {r5, r6}
mrc p15, 0, r0, c1, c0 @ get control register
bic r0, r0, r5
orr r0, r0, r6
mov pc, lr
.size __mohawk_setup, . - __mohawk_setup
/*
* R
* .RVI ZFRS BLDP WCAM
* .011 1001 ..00 0101
*
*/
.type mohawk_crval, #object
mohawk_crval:
crval clear=0x00007f3f, mmuset=0x00003905, ucset=0x00001134
__INITDATA
@ define struct processor (see <asm/proc-fns.h> and proc-macros.S)
define_processor_functions mohawk, dabort=v5t_early_abort, pabort=legacy_pabort
.section ".rodata"
string cpu_arch_name, "armv5te"
string cpu_elf_name, "v5"
string cpu_mohawk_name, "Marvell 88SV331x"
.align
.section ".proc.info.init", #alloc, #execinstr
.type __88sv331x_proc_info,#object
__88sv331x_proc_info:
.long 0x56158000 @ Marvell 88SV331x (MOHAWK)
.long 0xfffff000
.long PMD_TYPE_SECT | \
PMD_SECT_BUFFERABLE | \
PMD_SECT_CACHEABLE | \
PMD_BIT4 | \
PMD_SECT_AP_WRITE | \
PMD_SECT_AP_READ
.long PMD_TYPE_SECT | \
PMD_BIT4 | \
PMD_SECT_AP_WRITE | \
PMD_SECT_AP_READ
b __mohawk_setup
.long cpu_arch_name
.long cpu_elf_name
.long HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
.long cpu_mohawk_name
.long mohawk_processor_functions
.long v4wbi_tlb_fns
.long v4wb_user_fns
.long mohawk_cache_fns
.size __88sv331x_proc_info, . - __88sv331x_proc_info