mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-04 08:26:55 +07:00
c8c90860cd
Commit 81d11955bf ("ARM: 6405/1: Handle __flush_icache_all for
CONFIG_SMP_ON_UP") added a new function to struct cpu_cache_fns:
flush_icache_all(). It also implemented this for v6 and v7 but not
for v5 and backwards. Without the function pointer in place, we
will be calling wrong cache functions.
For example with ep93xx we get following:
Unable to handle kernel paging request at virtual address ee070f38
pgd = c0004000
[ee070f38] *pgd=00000000
Internal error: Oops: 80000005 [#1] PREEMPT
last sysfs file:
Modules linked in:
CPU: 0 Not tainted (2.6.36+ #1)
PC is at 0xee070f38
LR is at __dma_alloc+0x11c/0x2d0
pc : [<ee070f38>] lr : [<c0032c8c>] psr: 60000013
sp : c581bde0 ip : 00000000 fp : c0472000
r10: c0472000 r9 : 000000d0 r8 : 00020000
r7 : 0001ffff r6 : 00000000 r5 : c0472400 r4 : c5980000
r3 : c03ab7e0 r2 : 00000000 r1 : c59a0000 r0 : c5980000
Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel
Control: c000717f Table: c0004000 DAC: 00000017
Process swapper (pid: 1, stack limit = 0xc581a270)
[<c0032c8c>] (__dma_alloc+0x11c/0x2d0)
[<c0032e5c>] (dma_alloc_writecombine+0x1c/0x24)
[<c0204148>] (ep93xx_pcm_preallocate_dma_buffer+0x44/0x60)
[<c02041c0>] (ep93xx_pcm_new+0x5c/0x88)
[<c01ff188>] (snd_soc_instantiate_cards+0x8a8/0xbc0)
[<c01ff59c>] (soc_probe+0xfc/0x134)
[<c01adafc>] (platform_drv_probe+0x18/0x1c)
[<c01acca4>] (driver_probe_device+0xb0/0x16c)
[<c01ac284>] (bus_for_each_drv+0x48/0x84)
[<c01ace90>] (device_attach+0x50/0x68)
[<c01ac0f8>] (bus_probe_device+0x24/0x44)
[<c01aad7c>] (device_add+0x2fc/0x44c)
[<c01adfa8>] (platform_device_add+0x104/0x15c)
[<c0015eb8>] (simone_init+0x60/0x94)
[<c0021410>] (do_one_initcall+0xd0/0x1a4)
__dma_alloc() calls (inlined) __dma_alloc_buffer() which ends up
calling dmac_flush_range(). Now since the entries in the
arm920_cache_fns are shifted by one, we jump into address 0xee070f38
which is actually next instruction after the arm920_cache_fns
structure.
So implement flush_icache_all() for the rest of the supported CPUs
using a generic 'invalidate I cache' instruction.
Signed-off-by: Mika Westerberg <mika.westerberg@iki.fi>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
269 lines
6.4 KiB
ArmAsm
269 lines
6.4 KiB
ArmAsm
/*
|
|
* linux/arch/arm/mm/cache-v4wb.S
|
|
*
|
|
* Copyright (C) 1997-2002 Russell king
|
|
*
|
|
* 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/memory.h>
|
|
#include <asm/page.h>
|
|
#include "proc-macros.S"
|
|
|
|
/*
|
|
* The size of one data cache line.
|
|
*/
|
|
#define CACHE_DLINESIZE 32
|
|
|
|
/*
|
|
* The total size of the data cache.
|
|
*/
|
|
#if defined(CONFIG_CPU_SA110)
|
|
# define CACHE_DSIZE 16384
|
|
#elif defined(CONFIG_CPU_SA1100)
|
|
# define CACHE_DSIZE 8192
|
|
#else
|
|
# error Unknown cache size
|
|
#endif
|
|
|
|
/*
|
|
* This is the size at which it becomes more efficient to
|
|
* clean the whole cache, rather than using the individual
|
|
* cache line maintainence instructions.
|
|
*
|
|
* Size Clean (ticks) Dirty (ticks)
|
|
* 4096 21 20 21 53 55 54
|
|
* 8192 40 41 40 106 100 102
|
|
* 16384 77 77 76 140 140 138
|
|
* 32768 150 149 150 214 216 212 <---
|
|
* 65536 296 297 296 351 358 361
|
|
* 131072 591 591 591 656 657 651
|
|
* Whole 132 136 132 221 217 207 <---
|
|
*/
|
|
#define CACHE_DLIMIT (CACHE_DSIZE * 4)
|
|
|
|
.data
|
|
flush_base:
|
|
.long FLUSH_BASE
|
|
.text
|
|
|
|
/*
|
|
* flush_icache_all()
|
|
*
|
|
* Unconditionally clean and invalidate the entire icache.
|
|
*/
|
|
ENTRY(v4wb_flush_icache_all)
|
|
mov r0, #0
|
|
mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache
|
|
mov pc, lr
|
|
ENDPROC(v4wb_flush_icache_all)
|
|
|
|
/*
|
|
* flush_user_cache_all()
|
|
*
|
|
* Clean and invalidate all cache entries in a particular address
|
|
* space.
|
|
*/
|
|
ENTRY(v4wb_flush_user_cache_all)
|
|
/* FALLTHROUGH */
|
|
/*
|
|
* flush_kern_cache_all()
|
|
*
|
|
* Clean and invalidate the entire cache.
|
|
*/
|
|
ENTRY(v4wb_flush_kern_cache_all)
|
|
mov ip, #0
|
|
mcr p15, 0, ip, c7, c5, 0 @ invalidate I cache
|
|
__flush_whole_cache:
|
|
ldr r3, =flush_base
|
|
ldr r1, [r3, #0]
|
|
eor r1, r1, #CACHE_DSIZE
|
|
str r1, [r3, #0]
|
|
add r2, r1, #CACHE_DSIZE
|
|
1: ldr r3, [r1], #32
|
|
cmp r1, r2
|
|
blo 1b
|
|
#ifdef FLUSH_BASE_MINICACHE
|
|
add r2, r2, #FLUSH_BASE_MINICACHE - FLUSH_BASE
|
|
sub r1, r2, #512 @ only 512 bytes
|
|
1: ldr r3, [r1], #32
|
|
cmp r1, r2
|
|
blo 1b
|
|
#endif
|
|
mcr p15, 0, ip, c7, c10, 4 @ drain write buffer
|
|
mov pc, lr
|
|
|
|
/*
|
|
* flush_user_cache_range(start, end, flags)
|
|
*
|
|
* Invalidate a range of cache entries in the specified
|
|
* address space.
|
|
*
|
|
* - start - start address (inclusive, page aligned)
|
|
* - end - end address (exclusive, page aligned)
|
|
* - flags - vma_area_struct flags describing address space
|
|
*/
|
|
ENTRY(v4wb_flush_user_cache_range)
|
|
mov ip, #0
|
|
sub r3, r1, r0 @ calculate total size
|
|
tst r2, #VM_EXEC @ executable region?
|
|
mcrne p15, 0, ip, c7, c5, 0 @ invalidate I cache
|
|
|
|
cmp r3, #CACHE_DLIMIT @ total size >= limit?
|
|
bhs __flush_whole_cache @ flush whole D cache
|
|
|
|
1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry
|
|
mcr p15, 0, r0, c7, c6, 1 @ invalidate D entry
|
|
add r0, r0, #CACHE_DLINESIZE
|
|
cmp r0, r1
|
|
blo 1b
|
|
tst r2, #VM_EXEC
|
|
mcrne p15, 0, ip, c7, c10, 4 @ drain write buffer
|
|
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(v4wb_flush_kern_dcache_area)
|
|
add r1, r0, r1
|
|
/* fall through */
|
|
|
|
/*
|
|
* coherent_kern_range(start, end)
|
|
*
|
|
* Ensure coherency between the Icache and the Dcache in the
|
|
* region described by start. If you have non-snooping
|
|
* Harvard caches, you need to implement this function.
|
|
*
|
|
* - start - virtual start address
|
|
* - end - virtual end address
|
|
*/
|
|
ENTRY(v4wb_coherent_kern_range)
|
|
/* fall through */
|
|
|
|
/*
|
|
* coherent_user_range(start, end)
|
|
*
|
|
* Ensure coherency between the Icache and the Dcache in the
|
|
* region described by start. If you have non-snooping
|
|
* Harvard caches, you need to implement this function.
|
|
*
|
|
* - start - virtual start address
|
|
* - end - virtual end address
|
|
*/
|
|
ENTRY(v4wb_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, c6, 1 @ invalidate D entry
|
|
add r0, r0, #CACHE_DLINESIZE
|
|
cmp r0, r1
|
|
blo 1b
|
|
mov ip, #0
|
|
mcr p15, 0, ip, c7, c5, 0 @ invalidate I cache
|
|
mcr p15, 0, ip, 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
|
|
*/
|
|
v4wb_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 write buffer
|
|
mov pc, lr
|
|
|
|
/*
|
|
* dma_clean_range(start, end)
|
|
*
|
|
* Clean (write back) the specified virtual address range.
|
|
*
|
|
* - start - virtual start address
|
|
* - end - virtual end address
|
|
*/
|
|
v4wb_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 write buffer
|
|
mov pc, lr
|
|
|
|
/*
|
|
* dma_flush_range(start, end)
|
|
*
|
|
* Clean and invalidate the specified virtual address range.
|
|
*
|
|
* - start - virtual start address
|
|
* - end - virtual end address
|
|
*
|
|
* This is actually the same as v4wb_coherent_kern_range()
|
|
*/
|
|
.globl v4wb_dma_flush_range
|
|
.set v4wb_dma_flush_range, v4wb_coherent_kern_range
|
|
|
|
/*
|
|
* dma_map_area(start, size, dir)
|
|
* - start - kernel virtual start address
|
|
* - size - size of region
|
|
* - dir - DMA direction
|
|
*/
|
|
ENTRY(v4wb_dma_map_area)
|
|
add r1, r1, r0
|
|
cmp r2, #DMA_TO_DEVICE
|
|
beq v4wb_dma_clean_range
|
|
bcs v4wb_dma_inv_range
|
|
b v4wb_dma_flush_range
|
|
ENDPROC(v4wb_dma_map_area)
|
|
|
|
/*
|
|
* dma_unmap_area(start, size, dir)
|
|
* - start - kernel virtual start address
|
|
* - size - size of region
|
|
* - dir - DMA direction
|
|
*/
|
|
ENTRY(v4wb_dma_unmap_area)
|
|
mov pc, lr
|
|
ENDPROC(v4wb_dma_unmap_area)
|
|
|
|
__INITDATA
|
|
|
|
.type v4wb_cache_fns, #object
|
|
ENTRY(v4wb_cache_fns)
|
|
.long v4wb_flush_icache_all
|
|
.long v4wb_flush_kern_cache_all
|
|
.long v4wb_flush_user_cache_all
|
|
.long v4wb_flush_user_cache_range
|
|
.long v4wb_coherent_kern_range
|
|
.long v4wb_coherent_user_range
|
|
.long v4wb_flush_kern_dcache_area
|
|
.long v4wb_dma_map_area
|
|
.long v4wb_dma_unmap_area
|
|
.long v4wb_dma_flush_range
|
|
.size v4wb_cache_fns, . - v4wb_cache_fns
|