AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-11-30 01:26:44 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
031bd879f7
linux_dsm_epyc7002/arch/arm/mm/proc-feroceon.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

602 lines
15 KiB
ArmAsm
Raw Blame History

/*
* linux/arch/arm/mm/proc-feroceon.S: MMU functions for Feroceon
*
* Heavily based on proc-arm926.S
* Maintainer: Assaf Hoffman <hoffman@marvell.com>
*
* 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 invalidated
* using the single invalidate entry instructions. Anything larger
* than this, and we go for the whole cache.
*
* This value should be chosen such that we choose the cheapest
* alternative.
*/
#define CACHE_DLIMIT 16384
/*
* the cache line size of the I and D cache
*/
#define CACHE_DLINESIZE 32
.bss
.align 3
__cache_params_loc:
.space 8
.text
__cache_params:
.word __cache_params_loc
/*
* cpu_feroceon_proc_init()
*/
ENTRY(cpu_feroceon_proc_init)
mrc p15, 0, r0, c0, c0, 1 @ read cache type register
ldr r1, __cache_params
mov r2, #(16 << 5)
tst r0, #(1 << 16) @ get way
mov r0, r0, lsr #18 @ get cache size order
movne r3, #((4 - 1) << 30) @ 4-way
and r0, r0, #0xf
moveq r3, #0 @ 1-way
mov r2, r2, lsl r0 @ actual cache size
movne r2, r2, lsr #2 @ turned into # of sets
sub r2, r2, #(1 << 5)
stmia r1, {r2, r3}
mov pc, lr
/*
* cpu_feroceon_proc_fin()
*/
ENTRY(cpu_feroceon_proc_fin)
#if defined(CONFIG_CACHE_FEROCEON_L2) && \
!defined(CONFIG_CACHE_FEROCEON_L2_WRITETHROUGH)
mov r0, #0
mcr p15, 1, r0, c15, c9, 0 @ clean L2
mcr p15, 0, r0, c7, c10, 4 @ drain WB
#endif
mrc p15, 0, r0, c1, c0, 0 @ ctrl register
bic r0, r0, #0x1000 @ ...i............
bic r0, r0, #0x000e @ ............wca.
mcr p15, 0, r0, c1, c0, 0 @ disable caches
mov pc, lr
/*
* cpu_feroceon_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
*/
.align 5
.pushsection .idmap.text, "ax"
ENTRY(cpu_feroceon_reset)
mov ip, #0
mcr p15, 0, ip, c7, c7, 0 @ invalidate I,D caches
mcr p15, 0, ip, c7, c10, 4 @ drain WB
#ifdef CONFIG_MMU
mcr p15, 0, ip, c8, c7, 0 @ invalidate I & D TLBs
#endif
mrc p15, 0, ip, c1, c0, 0 @ ctrl register
bic ip, ip, #0x000f @ ............wcam
bic ip, ip, #0x1100 @ ...i...s........
mcr p15, 0, ip, c1, c0, 0 @ ctrl register
mov pc, r0
ENDPROC(cpu_feroceon_reset)
.popsection
/*
* cpu_feroceon_do_idle()
*
* Called with IRQs disabled
*/
.align 5
ENTRY(cpu_feroceon_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(feroceon_flush_icache_all)
mov r0, #0
mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache
mov pc, lr
ENDPROC(feroceon_flush_icache_all)
/*
* flush_user_cache_all()
*
* Clean and invalidate all cache entries in a particular
* address space.
*/
.align 5
ENTRY(feroceon_flush_user_cache_all)
/* FALLTHROUGH */
/*
* flush_kern_cache_all()
*
* Clean and invalidate the entire cache.
*/
ENTRY(feroceon_flush_kern_cache_all)
mov r2, #VM_EXEC
__flush_whole_cache:
ldr r1, __cache_params
ldmia r1, {r1, r3}
1: orr ip, r1, r3
2: mcr p15, 0, ip, c7, c14, 2 @ clean + invalidate D set/way
subs ip, ip, #(1 << 30) @ next way
bcs 2b
subs r1, r1, #(1 << 5) @ next set
bcs 1b
tst r2, #VM_EXEC
mov ip, #0
mcrne p15, 0, ip, c7, c5, 0 @ invalidate 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
*/
.align 5
ENTRY(feroceon_flush_user_cache_range)
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
mov ip, #0
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
*/
.align 5
ENTRY(feroceon_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
*/
ENTRY(feroceon_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
*/
.align 5
ENTRY(feroceon_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
.align 5
ENTRY(feroceon_range_flush_kern_dcache_area)
mrs r2, cpsr
add r1, r0, #PAGE_SZ - CACHE_DLINESIZE @ top addr is inclusive
orr r3, r2, #PSR_I_BIT
msr cpsr_c, r3 @ disable interrupts
mcr p15, 5, r0, c15, c15, 0 @ D clean/inv range start
mcr p15, 5, r1, c15, c15, 1 @ D clean/inv range top
msr cpsr_c, r2 @ restore interrupts
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)
*/
.align 5
feroceon_dma_inv_range:
tst r0, #CACHE_DLINESIZE - 1
bic r0, 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
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
.align 5
feroceon_range_dma_inv_range:
mrs r2, cpsr
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
cmp r1, r0
subne r1, r1, #1 @ top address is inclusive
orr r3, r2, #PSR_I_BIT
msr cpsr_c, r3 @ disable interrupts
mcr p15, 5, r0, c15, c14, 0 @ D inv range start
mcr p15, 5, r1, c15, c14, 1 @ D inv range top
msr cpsr_c, r2 @ restore interrupts
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)
*/
.align 5
feroceon_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
.align 5
feroceon_range_dma_clean_range:
mrs r2, cpsr
cmp r1, r0
subne r1, r1, #1 @ top address is inclusive
orr r3, r2, #PSR_I_BIT
msr cpsr_c, r3 @ disable interrupts
mcr p15, 5, r0, c15, c13, 0 @ D clean range start
mcr p15, 5, r1, c15, c13, 1 @ D clean range top
msr cpsr_c, r2 @ restore interrupts
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
*/
.align 5
ENTRY(feroceon_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
.align 5
ENTRY(feroceon_range_dma_flush_range)
mrs r2, cpsr
cmp r1, r0
subne r1, r1, #1 @ top address is inclusive
orr r3, r2, #PSR_I_BIT
msr cpsr_c, r3 @ disable interrupts
mcr p15, 5, r0, c15, c15, 0 @ D clean/inv range start
mcr p15, 5, r1, c15, c15, 1 @ D clean/inv range top
msr cpsr_c, r2 @ restore interrupts
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(feroceon_dma_map_area)
add r1, r1, r0
cmp r2, #DMA_TO_DEVICE
beq feroceon_dma_clean_range
bcs feroceon_dma_inv_range
b feroceon_dma_flush_range
ENDPROC(feroceon_dma_map_area)
/*
* dma_map_area(start, size, dir)
* - start - kernel virtual start address
* - size - size of region
* - dir - DMA direction
*/
ENTRY(feroceon_range_dma_map_area)
add r1, r1, r0
cmp r2, #DMA_TO_DEVICE
beq feroceon_range_dma_clean_range
bcs feroceon_range_dma_inv_range
b feroceon_range_dma_flush_range
ENDPROC(feroceon_range_dma_map_area)
/*
* dma_unmap_area(start, size, dir)
* - start - kernel virtual start address
* - size - size of region
* - dir - DMA direction
*/
ENTRY(feroceon_dma_unmap_area)
mov pc, lr
ENDPROC(feroceon_dma_unmap_area)
.globl feroceon_flush_kern_cache_louis
.equ feroceon_flush_kern_cache_louis, feroceon_flush_kern_cache_all
@ define struct cpu_cache_fns (see <asm/cacheflush.h> and proc-macros.S)
define_cache_functions feroceon
.macro range_alias basename
.globl feroceon_range_\basename
.type feroceon_range_\basename , %function
.equ feroceon_range_\basename , feroceon_\basename
.endm
/*
* Most of the cache functions are unchanged for this case.
* Export suitable alias symbols for the unchanged functions:
*/
range_alias flush_icache_all
range_alias flush_user_cache_all
range_alias flush_kern_cache_all
range_alias flush_user_cache_range
range_alias coherent_kern_range
range_alias coherent_user_range
range_alias dma_unmap_area
define_cache_functions feroceon_range
.align 5
ENTRY(cpu_feroceon_dcache_clean_area)
#if defined(CONFIG_CACHE_FEROCEON_L2) && \
!defined(CONFIG_CACHE_FEROCEON_L2_WRITETHROUGH)
mov r2, r0
mov r3, r1
#endif
1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry
add r0, r0, #CACHE_DLINESIZE
subs r1, r1, #CACHE_DLINESIZE
bhi 1b
#if defined(CONFIG_CACHE_FEROCEON_L2) && \
!defined(CONFIG_CACHE_FEROCEON_L2_WRITETHROUGH)
1: mcr p15, 1, r2, c15, c9, 1 @ clean L2 entry
add r2, r2, #CACHE_DLINESIZE
subs r3, r3, #CACHE_DLINESIZE
bhi 1b
#endif
mcr p15, 0, r0, c7, c10, 4 @ drain WB
mov pc, lr
/* =============================== PageTable ============================== */
/*
* cpu_feroceon_switch_mm(pgd)
*
* Set the translation base pointer to be as described by pgd.
*
* pgd: new page tables
*/
.align 5
ENTRY(cpu_feroceon_switch_mm)
#ifdef CONFIG_MMU
/*
* Note: we wish to call __flush_whole_cache but we need to preserve
* lr to do so. The only way without touching main memory is to
* use r2 which is normally used to test the VM_EXEC flag, and
* compensate locally for the skipped ops if it is not set.
*/
mov r2, lr @ abuse r2 to preserve lr
bl __flush_whole_cache
@ if r2 contains the VM_EXEC bit then the next 2 ops are done already
tst r2, #VM_EXEC
mcreq p15, 0, ip, c7, c5, 0 @ invalidate I cache
mcreq p15, 0, ip, c7, c10, 4 @ drain WB
mcr p15, 0, r0, c2, c0, 0 @ load page table pointer
mcr p15, 0, ip, c8, c7, 0 @ invalidate I & D TLBs
mov pc, r2
#else
mov pc, lr
#endif
/*
* cpu_feroceon_set_pte_ext(ptep, pte, ext)
*
* Set a PTE and flush it out
*/
.align 5
ENTRY(cpu_feroceon_set_pte_ext)
#ifdef CONFIG_MMU
armv3_set_pte_ext wc_disable=0
mov r0, r0
mcr p15, 0, r0, c7, c10, 1 @ clean D entry
#if defined(CONFIG_CACHE_FEROCEON_L2) && \
!defined(CONFIG_CACHE_FEROCEON_L2_WRITETHROUGH)
mcr p15, 1, r0, c15, c9, 1 @ clean L2 entry
#endif
mcr p15, 0, r0, c7, c10, 4 @ drain WB
#endif
mov pc, lr
__CPUINIT
.type __feroceon_setup, #function
__feroceon_setup:
mov r0, #0
mcr p15, 0, r0, c7, c7 @ invalidate I,D caches on v4
mcr p15, 0, r0, c7, c10, 4 @ drain write buffer on v4
#ifdef CONFIG_MMU
mcr p15, 0, r0, c8, c7 @ invalidate I,D TLBs on v4
#endif
adr r5, feroceon_crval
ldmia r5, {r5, r6}
mrc p15, 0, r0, c1, c0 @ get control register v4
bic r0, r0, r5
orr r0, r0, r6
mov pc, lr
.size __feroceon_setup, . - __feroceon_setup
/*
* B
* R P
* .RVI UFRS BLDP WCAM
* .011 .001 ..11 0101
*
*/
.type feroceon_crval, #object
feroceon_crval:
crval clear=0x0000773f, mmuset=0x00003135, ucset=0x00001134
__INITDATA
@ define struct processor (see <asm/proc-fns.h> and proc-macros.S)
define_processor_functions feroceon, dabort=v5t_early_abort, pabort=legacy_pabort
.section ".rodata"
string cpu_arch_name, "armv5te"
string cpu_elf_name, "v5"
string cpu_feroceon_name, "Feroceon"
string cpu_88fr531_name, "Feroceon 88FR531-vd"
string cpu_88fr571_name, "Feroceon 88FR571-vd"
string cpu_88fr131_name, "Feroceon 88FR131"
.align
.section ".proc.info.init", #alloc, #execinstr
.macro feroceon_proc_info name:req, cpu_val:req, cpu_mask:req, cpu_name:req, cache:req
.type __\name\()_proc_info,#object
__\name\()_proc_info:
.long \cpu_val
.long \cpu_mask
.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 __feroceon_setup
.long cpu_arch_name
.long cpu_elf_name
.long HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
.long \cpu_name
.long feroceon_processor_functions
.long v4wbi_tlb_fns
.long feroceon_user_fns
.long \cache
.size __\name\()_proc_info, . - __\name\()_proc_info
.endm
#ifdef CONFIG_CPU_FEROCEON_OLD_ID
feroceon_proc_info feroceon_old_id, 0x41009260, 0xff00fff0, \
cpu_name=cpu_feroceon_name, cache=feroceon_cache_fns
#endif
feroceon_proc_info 88fr531, 0x56055310, 0xfffffff0, cpu_88fr531_name, \
cache=feroceon_cache_fns
feroceon_proc_info 88fr571, 0x56155710, 0xfffffff0, cpu_88fr571_name, \
cache=feroceon_range_cache_fns
feroceon_proc_info 88fr131, 0x56251310, 0xfffffff0, cpu_88fr131_name, \
cache=feroceon_range_cache_fns
Reference in New Issue View Git Blame Copy Permalink