kernel_optimize_test/arch/arm/mm/proc-macros.S
Will Deacon b6ccb9803e ARM: 7954/1: mm: remove remaining domain support from ARMv6
CPU_32v6 currently selects CPU_USE_DOMAINS if CPU_V6 and MMU. This is
because ARM 1136 r0pX CPUs lack the v6k extensions, and therefore do
not have hardware thread registers. The lack of these registers requires
the kernel to update the vectors page at each context switch in order to
write a new TLS pointer. This write must be done via the userspace
mapping, since aliasing caches can lead to expensive flushing when using
kmap. Finally, this requires the vectors page to be mapped r/w for
kernel and r/o for user, which has implications for things like put_user
which must trigger CoW appropriately when targetting user pages.

The upshot of all this is that a v6/v7 kernel makes use of domains to
segregate kernel and user memory accesses. This has the nasty
side-effect of making device mappings executable, which has been
observed to cause subtle bugs on recent cores (e.g. Cortex-A15
performing a speculative instruction fetch from the GIC and acking an
interrupt in the process).

This patch solves this problem by removing the remaining domain support
from ARMv6. A new memory type is added specifically for the vectors page
which allows that page (and only that page) to be mapped as user r/o,
kernel r/w. All other user r/o pages are mapped also as kernel r/o.
Patch co-developed with Russell King.

Cc: <stable@vger.kernel.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-02-10 11:48:13 +00:00

334 lines
7.7 KiB
ArmAsm

/*
* We need constants.h for:
* VMA_VM_MM
* VMA_VM_FLAGS
* VM_EXEC
*/
#include <asm/asm-offsets.h>
#include <asm/thread_info.h>
/*
* vma_vm_mm - get mm pointer from vma pointer (vma->vm_mm)
*/
.macro vma_vm_mm, rd, rn
ldr \rd, [\rn, #VMA_VM_MM]
.endm
/*
* vma_vm_flags - get vma->vm_flags
*/
.macro vma_vm_flags, rd, rn
ldr \rd, [\rn, #VMA_VM_FLAGS]
.endm
.macro tsk_mm, rd, rn
ldr \rd, [\rn, #TI_TASK]
ldr \rd, [\rd, #TSK_ACTIVE_MM]
.endm
/*
* act_mm - get current->active_mm
*/
.macro act_mm, rd
bic \rd, sp, #8128
bic \rd, \rd, #63
ldr \rd, [\rd, #TI_TASK]
ldr \rd, [\rd, #TSK_ACTIVE_MM]
.endm
/*
* mmid - get context id from mm pointer (mm->context.id)
* note, this field is 64bit, so in big-endian the two words are swapped too.
*/
.macro mmid, rd, rn
#ifdef __ARMEB__
ldr \rd, [\rn, #MM_CONTEXT_ID + 4 ]
#else
ldr \rd, [\rn, #MM_CONTEXT_ID]
#endif
.endm
/*
* mask_asid - mask the ASID from the context ID
*/
.macro asid, rd, rn
and \rd, \rn, #255
.endm
.macro crval, clear, mmuset, ucset
#ifdef CONFIG_MMU
.word \clear
.word \mmuset
#else
.word \clear
.word \ucset
#endif
.endm
/*
* dcache_line_size - get the minimum D-cache line size from the CTR register
* on ARMv7.
*/
.macro dcache_line_size, reg, tmp
mrc p15, 0, \tmp, c0, c0, 1 @ read ctr
lsr \tmp, \tmp, #16
and \tmp, \tmp, #0xf @ cache line size encoding
mov \reg, #4 @ bytes per word
mov \reg, \reg, lsl \tmp @ actual cache line size
.endm
/*
* icache_line_size - get the minimum I-cache line size from the CTR register
* on ARMv7.
*/
.macro icache_line_size, reg, tmp
mrc p15, 0, \tmp, c0, c0, 1 @ read ctr
and \tmp, \tmp, #0xf @ cache line size encoding
mov \reg, #4 @ bytes per word
mov \reg, \reg, lsl \tmp @ actual cache line size
.endm
/*
* Sanity check the PTE configuration for the code below - which makes
* certain assumptions about how these bits are laid out.
*/
#ifdef CONFIG_MMU
#if L_PTE_SHARED != PTE_EXT_SHARED
#error PTE shared bit mismatch
#endif
#if !defined (CONFIG_ARM_LPAE) && \
(L_PTE_XN+L_PTE_USER+L_PTE_RDONLY+L_PTE_DIRTY+L_PTE_YOUNG+\
L_PTE_FILE+L_PTE_PRESENT) > L_PTE_SHARED
#error Invalid Linux PTE bit settings
#endif
#endif /* CONFIG_MMU */
/*
* The ARMv6 and ARMv7 set_pte_ext translation function.
*
* Permission translation:
* YUWD APX AP1 AP0 SVC User
* 0xxx 0 0 0 no acc no acc
* 100x 1 0 1 r/o no acc
* 10x0 1 0 1 r/o no acc
* 1011 0 0 1 r/w no acc
* 110x 1 1 1 r/o r/o
* 11x0 1 1 1 r/o r/o
* 1111 0 1 1 r/w r/w
*/
.macro armv6_mt_table pfx
\pfx\()_mt_table:
.long 0x00 @ L_PTE_MT_UNCACHED
.long PTE_EXT_TEX(1) @ L_PTE_MT_BUFFERABLE
.long PTE_CACHEABLE @ L_PTE_MT_WRITETHROUGH
.long PTE_CACHEABLE | PTE_BUFFERABLE @ L_PTE_MT_WRITEBACK
.long PTE_BUFFERABLE @ L_PTE_MT_DEV_SHARED
.long 0x00 @ unused
.long 0x00 @ L_PTE_MT_MINICACHE (not present)
.long PTE_EXT_TEX(1) | PTE_CACHEABLE | PTE_BUFFERABLE @ L_PTE_MT_WRITEALLOC
.long 0x00 @ unused
.long PTE_EXT_TEX(1) @ L_PTE_MT_DEV_WC
.long 0x00 @ unused
.long PTE_CACHEABLE | PTE_BUFFERABLE @ L_PTE_MT_DEV_CACHED
.long PTE_EXT_TEX(2) @ L_PTE_MT_DEV_NONSHARED
.long 0x00 @ unused
.long 0x00 @ unused
.long PTE_CACHEABLE | PTE_BUFFERABLE | PTE_EXT_APX @ L_PTE_MT_VECTORS
.endm
.macro armv6_set_pte_ext pfx
str r1, [r0], #2048 @ linux version
bic r3, r1, #0x000003fc
bic r3, r3, #PTE_TYPE_MASK
orr r3, r3, r2
orr r3, r3, #PTE_EXT_AP0 | 2
adr ip, \pfx\()_mt_table
and r2, r1, #L_PTE_MT_MASK
ldr r2, [ip, r2]
eor r1, r1, #L_PTE_DIRTY
tst r1, #L_PTE_DIRTY|L_PTE_RDONLY
orrne r3, r3, #PTE_EXT_APX
tst r1, #L_PTE_USER
orrne r3, r3, #PTE_EXT_AP1
tstne r3, #PTE_EXT_APX
@ user read-only -> kernel read-only
bicne r3, r3, #PTE_EXT_AP0
tst r1, #L_PTE_XN
orrne r3, r3, #PTE_EXT_XN
eor r3, r3, r2
tst r1, #L_PTE_YOUNG
tstne r1, #L_PTE_PRESENT
moveq r3, #0
tstne r1, #L_PTE_NONE
movne r3, #0
str r3, [r0]
mcr p15, 0, r0, c7, c10, 1 @ flush_pte
.endm
/*
* The ARMv3, ARMv4 and ARMv5 set_pte_ext translation function,
* covering most CPUs except Xscale and Xscale 3.
*
* Permission translation:
* YUWD AP SVC User
* 0xxx 0x00 no acc no acc
* 100x 0x00 r/o no acc
* 10x0 0x00 r/o no acc
* 1011 0x55 r/w no acc
* 110x 0xaa r/w r/o
* 11x0 0xaa r/w r/o
* 1111 0xff r/w r/w
*/
.macro armv3_set_pte_ext wc_disable=1
str r1, [r0], #2048 @ linux version
eor r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY
bic r2, r1, #PTE_SMALL_AP_MASK @ keep C, B bits
bic r2, r2, #PTE_TYPE_MASK
orr r2, r2, #PTE_TYPE_SMALL
tst r3, #L_PTE_USER @ user?
orrne r2, r2, #PTE_SMALL_AP_URO_SRW
tst r3, #L_PTE_RDONLY | L_PTE_DIRTY @ write and dirty?
orreq r2, r2, #PTE_SMALL_AP_UNO_SRW
tst r3, #L_PTE_PRESENT | L_PTE_YOUNG @ present and young?
movne r2, #0
.if \wc_disable
#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
tst r2, #PTE_CACHEABLE
bicne r2, r2, #PTE_BUFFERABLE
#endif
.endif
str r2, [r0] @ hardware version
.endm
/*
* Xscale set_pte_ext translation, split into two halves to cope
* with work-arounds. r3 must be preserved by code between these
* two macros.
*
* Permission translation:
* YUWD AP SVC User
* 0xxx 00 no acc no acc
* 100x 00 r/o no acc
* 10x0 00 r/o no acc
* 1011 01 r/w no acc
* 110x 10 r/w r/o
* 11x0 10 r/w r/o
* 1111 11 r/w r/w
*/
.macro xscale_set_pte_ext_prologue
str r1, [r0] @ linux version
eor r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY
bic r2, r1, #PTE_SMALL_AP_MASK @ keep C, B bits
orr r2, r2, #PTE_TYPE_EXT @ extended page
tst r3, #L_PTE_USER @ user?
orrne r2, r2, #PTE_EXT_AP_URO_SRW @ yes -> user r/o, system r/w
tst r3, #L_PTE_RDONLY | L_PTE_DIRTY @ write and dirty?
orreq r2, r2, #PTE_EXT_AP_UNO_SRW @ yes -> user n/a, system r/w
@ combined with user -> user r/w
.endm
.macro xscale_set_pte_ext_epilogue
tst r3, #L_PTE_PRESENT | L_PTE_YOUNG @ present and young?
movne r2, #0 @ no -> fault
str r2, [r0, #2048]! @ hardware version
mov ip, #0
mcr p15, 0, r0, c7, c10, 1 @ clean L1 D line
mcr p15, 0, ip, c7, c10, 4 @ data write barrier
.endm
.macro define_processor_functions name:req, dabort:req, pabort:req, nommu=0, suspend=0
.type \name\()_processor_functions, #object
.align 2
ENTRY(\name\()_processor_functions)
.word \dabort
.word \pabort
.word cpu_\name\()_proc_init
.word cpu_\name\()_proc_fin
.word cpu_\name\()_reset
.word cpu_\name\()_do_idle
.word cpu_\name\()_dcache_clean_area
.word cpu_\name\()_switch_mm
.if \nommu
.word 0
.else
.word cpu_\name\()_set_pte_ext
.endif
.if \suspend
.word cpu_\name\()_suspend_size
#ifdef CONFIG_PM_SLEEP
.word cpu_\name\()_do_suspend
.word cpu_\name\()_do_resume
#else
.word 0
.word 0
#endif
.else
.word 0
.word 0
.word 0
.endif
.size \name\()_processor_functions, . - \name\()_processor_functions
.endm
.macro define_cache_functions name:req
.align 2
.type \name\()_cache_fns, #object
ENTRY(\name\()_cache_fns)
.long \name\()_flush_icache_all
.long \name\()_flush_kern_cache_all
.long \name\()_flush_kern_cache_louis
.long \name\()_flush_user_cache_all
.long \name\()_flush_user_cache_range
.long \name\()_coherent_kern_range
.long \name\()_coherent_user_range
.long \name\()_flush_kern_dcache_area
.long \name\()_dma_map_area
.long \name\()_dma_unmap_area
.long \name\()_dma_flush_range
.size \name\()_cache_fns, . - \name\()_cache_fns
.endm
.macro define_tlb_functions name:req, flags_up:req, flags_smp
.type \name\()_tlb_fns, #object
ENTRY(\name\()_tlb_fns)
.long \name\()_flush_user_tlb_range
.long \name\()_flush_kern_tlb_range
.ifnb \flags_smp
ALT_SMP(.long \flags_smp )
ALT_UP(.long \flags_up )
.else
.long \flags_up
.endif
.size \name\()_tlb_fns, . - \name\()_tlb_fns
.endm
.macro globl_equ x, y
.globl \x
.equ \x, \y
.endm