linux/arch/arm/kernel/entry-armv.S

1228 lines
28 KiB
ArmAsm
Raw Normal View History

/*
* linux/arch/arm/kernel/entry-armv.S
*
* Copyright (C) 1996,1997,1998 Russell King.
* ARM700 fix by Matthew Godbolt (linux-user@willothewisp.demon.co.uk)
* nommu support by Hyok S. Choi (hyok.choi@samsung.com)
*
* 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.
*
* Low-level vector interface routines
*
* Note: there is a StrongARM bug in the STMIA rn, {regs}^ instruction
* that causes it to save wrong values... Be aware!
*/
#include <asm/memory.h>
#include <asm/glue-df.h>
#include <asm/glue-pf.h>
#include <asm/vfpmacros.h>
#include <mach/entry-macro.S>
#include <asm/thread_notify.h>
#include <asm/unwind.h>
#include <asm/unistd.h>
#include <asm/tls.h>
#include "entry-header.S"
#include <asm/entry-macro-multi.S>
/*
* Interrupt handling. Preserves r7, r8, r9
*/
.macro irq_handler
#ifdef CONFIG_MULTI_IRQ_HANDLER
ldr r5, =handle_arch_irq
mov r0, sp
ldr r5, [r5]
adr lr, BSYM(9997f)
teq r5, #0
movne pc, r5
#endif
arch_irq_handler_default
9997:
.endm
.macro pabt_helper
mov r0, r2 @ pass address of aborted instruction.
#ifdef MULTI_PABORT
ldr r4, .LCprocfns
mov lr, pc
ldr pc, [r4, #PROCESSOR_PABT_FUNC]
#else
bl CPU_PABORT_HANDLER
#endif
.endm
.macro dabt_helper
@
@ Call the processor-specific abort handler:
@
@ r2 - aborted context pc
@ r3 - aborted context cpsr
@
@ The abort handler must return the aborted address in r0, and
@ the fault status register in r1. r9 must be preserved.
@
#ifdef MULTI_DABORT
ldr r4, .LCprocfns
mov lr, pc
ldr pc, [r4, #PROCESSOR_DABT_FUNC]
#else
bl CPU_DABORT_HANDLER
#endif
.endm
#ifdef CONFIG_KPROBES
.section .kprobes.text,"ax",%progbits
#else
.text
#endif
/*
* Invalid mode handlers
*/
.macro inv_entry, reason
sub sp, sp, #S_FRAME_SIZE
ARM( stmib sp, {r1 - lr} )
THUMB( stmia sp, {r0 - r12} )
THUMB( str sp, [sp, #S_SP] )
THUMB( str lr, [sp, #S_LR] )
mov r1, #\reason
.endm
__pabt_invalid:
inv_entry BAD_PREFETCH
b common_invalid
ENDPROC(__pabt_invalid)
__dabt_invalid:
inv_entry BAD_DATA
b common_invalid
ENDPROC(__dabt_invalid)
__irq_invalid:
inv_entry BAD_IRQ
b common_invalid
ENDPROC(__irq_invalid)
__und_invalid:
inv_entry BAD_UNDEFINSTR
@
@ XXX fall through to common_invalid
@
@
@ common_invalid - generic code for failed exception (re-entrant version of handlers)
@
common_invalid:
zero_fp
ldmia r0, {r4 - r6}
add r0, sp, #S_PC @ here for interlock avoidance
mov r7, #-1 @ "" "" "" ""
str r4, [sp] @ save preserved r0
stmia r0, {r5 - r7} @ lr_<exception>,
@ cpsr_<exception>, "old_r0"
mov r0, sp
b bad_mode
ENDPROC(__und_invalid)
/*
* SVC mode handlers
*/
#if defined(CONFIG_AEABI) && (__LINUX_ARM_ARCH__ >= 5)
#define SPFIX(code...) code
#else
#define SPFIX(code...)
#endif
.macro svc_entry, stack_hole=0
UNWIND(.fnstart )
UNWIND(.save {r0 - pc} )
sub sp, sp, #(S_FRAME_SIZE + \stack_hole - 4)
#ifdef CONFIG_THUMB2_KERNEL
SPFIX( str r0, [sp] ) @ temporarily saved
SPFIX( mov r0, sp )
SPFIX( tst r0, #4 ) @ test original stack alignment
SPFIX( ldr r0, [sp] ) @ restored
#else
SPFIX( tst sp, #4 )
#endif
SPFIX( subeq sp, sp, #4 )
stmia sp, {r1 - r12}
ldmia r0, {r1 - r3}
add r5, sp, #S_SP - 4 @ here for interlock avoidance
mov r4, #-1 @ "" "" "" ""
add r0, sp, #(S_FRAME_SIZE + \stack_hole - 4)
SPFIX( addeq r0, r0, #4 )
str r1, [sp, #-4]! @ save the "real" r0 copied
@ from the exception stack
mov r1, lr
@
@ We are now ready to fill in the remaining blanks on the stack:
@
@ r0 - sp_svc
@ r1 - lr_svc
@ r2 - lr_<exception>, already fixed up for correct return/restart
@ r3 - spsr_<exception>
@ r4 - orig_r0 (see pt_regs definition in ptrace.h)
@
stmia r5, {r0 - r4}
.endm
.align 5
__dabt_svc:
svc_entry
@
@ get ready to re-enable interrupts if appropriate
@
mrs r9, cpsr
tst r3, #PSR_I_BIT
biceq r9, r9, #PSR_I_BIT
dabt_helper
@
@ set desired IRQ state, then call main handler
@
debug_entry r1
msr cpsr_c, r9
mov r2, sp
bl do_DataAbort
@
@ IRQs off again before pulling preserved data off the stack
@
ARM: lockdep: fix unannotated irqs-on CPU: Testing write buffer coherency: ok ------------[ cut here ]------------ WARNING: at kernel/lockdep.c:3145 check_flags+0xcc/0x1dc() Modules linked in: [<c0035120>] (unwind_backtrace+0x0/0xf8) from [<c0355374>] (dump_stack+0x20/0x24) [<c0355374>] (dump_stack+0x20/0x24) from [<c0060c04>] (warn_slowpath_common+0x58/0x70) [<c0060c04>] (warn_slowpath_common+0x58/0x70) from [<c0060c3c>] (warn_slowpath_null+0x20/0x24) [<c0060c3c>] (warn_slowpath_null+0x20/0x24) from [<c008f224>] (check_flags+0xcc/0x1dc) [<c008f224>] (check_flags+0xcc/0x1dc) from [<c00945dc>] (lock_acquire+0x50/0x140) [<c00945dc>] (lock_acquire+0x50/0x140) from [<c0358434>] (_raw_spin_lock+0x50/0x88) [<c0358434>] (_raw_spin_lock+0x50/0x88) from [<c00fd114>] (set_task_comm+0x2c/0x60) [<c00fd114>] (set_task_comm+0x2c/0x60) from [<c007e184>] (kthreadd+0x30/0x108) [<c007e184>] (kthreadd+0x30/0x108) from [<c0030104>] (kernel_thread_exit+0x0/0x8) ---[ end trace 1b75b31a2719ed1c ]--- possible reason: unannotated irqs-on. irq event stamp: 3 hardirqs last enabled at (2): [<c0059bb0>] finish_task_switch+0x48/0xb0 hardirqs last disabled at (3): [<c002f0b0>] ret_slow_syscall+0xc/0x1c softirqs last enabled at (0): [<c005f3e0>] copy_process+0x394/0xe5c softirqs last disabled at (0): [<(null)>] (null) Fix this by ensuring that the lockdep interrupt state is manipulated in the appropriate places. We essentially treat userspace as an entirely separate environment which isn't relevant to lockdep (lockdep doesn't monitor userspace.) We don't tell lockdep that IRQs will be enabled in that environment. Instead, when creating kernel threads (which is a rare event compared to entering/leaving userspace) we have to update the lockdep state. Do this by starting threads with IRQs disabled, and in the kthread helper, tell lockdep that IRQs are enabled, and enable them. This provides lockdep with a consistent view of the current IRQ state in kernel space. This also revert portions of 0d928b0b616d1c5c5fe76019a87cba171ca91633 which didn't fix the problem. Tested-by: Ming Lei <tom.leiming@gmail.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-07-10 09:10:18 +00:00
disable_irq_notrace
@
@ restore SPSR and restart the instruction
@
ldr r2, [sp, #S_PSR]
svc_exit r2 @ return from exception
UNWIND(.fnend )
ENDPROC(__dabt_svc)
.align 5
__irq_svc:
svc_entry
ARM: lockdep: fix unannotated irqs-on CPU: Testing write buffer coherency: ok ------------[ cut here ]------------ WARNING: at kernel/lockdep.c:3145 check_flags+0xcc/0x1dc() Modules linked in: [<c0035120>] (unwind_backtrace+0x0/0xf8) from [<c0355374>] (dump_stack+0x20/0x24) [<c0355374>] (dump_stack+0x20/0x24) from [<c0060c04>] (warn_slowpath_common+0x58/0x70) [<c0060c04>] (warn_slowpath_common+0x58/0x70) from [<c0060c3c>] (warn_slowpath_null+0x20/0x24) [<c0060c3c>] (warn_slowpath_null+0x20/0x24) from [<c008f224>] (check_flags+0xcc/0x1dc) [<c008f224>] (check_flags+0xcc/0x1dc) from [<c00945dc>] (lock_acquire+0x50/0x140) [<c00945dc>] (lock_acquire+0x50/0x140) from [<c0358434>] (_raw_spin_lock+0x50/0x88) [<c0358434>] (_raw_spin_lock+0x50/0x88) from [<c00fd114>] (set_task_comm+0x2c/0x60) [<c00fd114>] (set_task_comm+0x2c/0x60) from [<c007e184>] (kthreadd+0x30/0x108) [<c007e184>] (kthreadd+0x30/0x108) from [<c0030104>] (kernel_thread_exit+0x0/0x8) ---[ end trace 1b75b31a2719ed1c ]--- possible reason: unannotated irqs-on. irq event stamp: 3 hardirqs last enabled at (2): [<c0059bb0>] finish_task_switch+0x48/0xb0 hardirqs last disabled at (3): [<c002f0b0>] ret_slow_syscall+0xc/0x1c softirqs last enabled at (0): [<c005f3e0>] copy_process+0x394/0xe5c softirqs last disabled at (0): [<(null)>] (null) Fix this by ensuring that the lockdep interrupt state is manipulated in the appropriate places. We essentially treat userspace as an entirely separate environment which isn't relevant to lockdep (lockdep doesn't monitor userspace.) We don't tell lockdep that IRQs will be enabled in that environment. Instead, when creating kernel threads (which is a rare event compared to entering/leaving userspace) we have to update the lockdep state. Do this by starting threads with IRQs disabled, and in the kthread helper, tell lockdep that IRQs are enabled, and enable them. This provides lockdep with a consistent view of the current IRQ state in kernel space. This also revert portions of 0d928b0b616d1c5c5fe76019a87cba171ca91633 which didn't fix the problem. Tested-by: Ming Lei <tom.leiming@gmail.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-07-10 09:10:18 +00:00
#ifdef CONFIG_TRACE_IRQFLAGS
bl trace_hardirqs_off
#endif
#ifdef CONFIG_PREEMPT
get_thread_info tsk
ldr r8, [tsk, #TI_PREEMPT] @ get preempt count
add r7, r8, #1 @ increment it
str r7, [tsk, #TI_PREEMPT]
#endif
irq_handler
#ifdef CONFIG_PREEMPT
str r8, [tsk, #TI_PREEMPT] @ restore preempt count
ldr r0, [tsk, #TI_FLAGS] @ get flags
teq r8, #0 @ if preempt count != 0
movne r0, #0 @ force flags to 0
tst r0, #_TIF_NEED_RESCHED
blne svc_preempt
#endif
ldr r4, [sp, #S_PSR] @ irqs are already disabled
#ifdef CONFIG_TRACE_IRQFLAGS
tst r4, #PSR_I_BIT
bleq trace_hardirqs_on
#endif
svc_exit r4 @ return from exception
UNWIND(.fnend )
ENDPROC(__irq_svc)
.ltorg
#ifdef CONFIG_PREEMPT
svc_preempt:
mov r8, lr
1: bl preempt_schedule_irq @ irq en/disable is done inside
ldr r0, [tsk, #TI_FLAGS] @ get new tasks TI_FLAGS
tst r0, #_TIF_NEED_RESCHED
moveq pc, r8 @ go again
b 1b
#endif
.align 5
__und_svc:
#ifdef CONFIG_KPROBES
@ If a kprobe is about to simulate a "stmdb sp..." instruction,
@ it obviously needs free stack space which then will belong to
@ the saved context.
svc_entry 64
#else
svc_entry
#endif
@
@ call emulation code, which returns using r9 if it has emulated
@ the instruction, or the more conventional lr if we are to treat
@ this as a real undefined instruction
@
@ r0 - instruction
@
#ifndef CONFIG_THUMB2_KERNEL
ldr r0, [r2, #-4]
#else
ldrh r0, [r2, #-2] @ Thumb instruction at LR - 2
and r9, r0, #0xf800
cmp r9, #0xe800 @ 32-bit instruction if xx >= 0
ldrhhs r9, [r2] @ bottom 16 bits
orrhs r0, r9, r0, lsl #16
#endif
adr r9, BSYM(1f)
bl call_fpe
mov r0, sp @ struct pt_regs *regs
bl do_undefinstr
@
@ IRQs off again before pulling preserved data off the stack
@
ARM: lockdep: fix unannotated irqs-on CPU: Testing write buffer coherency: ok ------------[ cut here ]------------ WARNING: at kernel/lockdep.c:3145 check_flags+0xcc/0x1dc() Modules linked in: [<c0035120>] (unwind_backtrace+0x0/0xf8) from [<c0355374>] (dump_stack+0x20/0x24) [<c0355374>] (dump_stack+0x20/0x24) from [<c0060c04>] (warn_slowpath_common+0x58/0x70) [<c0060c04>] (warn_slowpath_common+0x58/0x70) from [<c0060c3c>] (warn_slowpath_null+0x20/0x24) [<c0060c3c>] (warn_slowpath_null+0x20/0x24) from [<c008f224>] (check_flags+0xcc/0x1dc) [<c008f224>] (check_flags+0xcc/0x1dc) from [<c00945dc>] (lock_acquire+0x50/0x140) [<c00945dc>] (lock_acquire+0x50/0x140) from [<c0358434>] (_raw_spin_lock+0x50/0x88) [<c0358434>] (_raw_spin_lock+0x50/0x88) from [<c00fd114>] (set_task_comm+0x2c/0x60) [<c00fd114>] (set_task_comm+0x2c/0x60) from [<c007e184>] (kthreadd+0x30/0x108) [<c007e184>] (kthreadd+0x30/0x108) from [<c0030104>] (kernel_thread_exit+0x0/0x8) ---[ end trace 1b75b31a2719ed1c ]--- possible reason: unannotated irqs-on. irq event stamp: 3 hardirqs last enabled at (2): [<c0059bb0>] finish_task_switch+0x48/0xb0 hardirqs last disabled at (3): [<c002f0b0>] ret_slow_syscall+0xc/0x1c softirqs last enabled at (0): [<c005f3e0>] copy_process+0x394/0xe5c softirqs last disabled at (0): [<(null)>] (null) Fix this by ensuring that the lockdep interrupt state is manipulated in the appropriate places. We essentially treat userspace as an entirely separate environment which isn't relevant to lockdep (lockdep doesn't monitor userspace.) We don't tell lockdep that IRQs will be enabled in that environment. Instead, when creating kernel threads (which is a rare event compared to entering/leaving userspace) we have to update the lockdep state. Do this by starting threads with IRQs disabled, and in the kthread helper, tell lockdep that IRQs are enabled, and enable them. This provides lockdep with a consistent view of the current IRQ state in kernel space. This also revert portions of 0d928b0b616d1c5c5fe76019a87cba171ca91633 which didn't fix the problem. Tested-by: Ming Lei <tom.leiming@gmail.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-07-10 09:10:18 +00:00
1: disable_irq_notrace
@
@ restore SPSR and restart the instruction
@
ldr r2, [sp, #S_PSR] @ Get SVC cpsr
svc_exit r2 @ return from exception
UNWIND(.fnend )
ENDPROC(__und_svc)
.align 5
__pabt_svc:
svc_entry
@
@ re-enable interrupts if appropriate
@
mrs r9, cpsr
tst r3, #PSR_I_BIT
biceq r9, r9, #PSR_I_BIT
pabt_helper
debug_entry r1
msr cpsr_c, r9 @ Maybe enable interrupts
mov r2, sp @ regs
bl do_PrefetchAbort @ call abort handler
@
@ IRQs off again before pulling preserved data off the stack
@
ARM: lockdep: fix unannotated irqs-on CPU: Testing write buffer coherency: ok ------------[ cut here ]------------ WARNING: at kernel/lockdep.c:3145 check_flags+0xcc/0x1dc() Modules linked in: [<c0035120>] (unwind_backtrace+0x0/0xf8) from [<c0355374>] (dump_stack+0x20/0x24) [<c0355374>] (dump_stack+0x20/0x24) from [<c0060c04>] (warn_slowpath_common+0x58/0x70) [<c0060c04>] (warn_slowpath_common+0x58/0x70) from [<c0060c3c>] (warn_slowpath_null+0x20/0x24) [<c0060c3c>] (warn_slowpath_null+0x20/0x24) from [<c008f224>] (check_flags+0xcc/0x1dc) [<c008f224>] (check_flags+0xcc/0x1dc) from [<c00945dc>] (lock_acquire+0x50/0x140) [<c00945dc>] (lock_acquire+0x50/0x140) from [<c0358434>] (_raw_spin_lock+0x50/0x88) [<c0358434>] (_raw_spin_lock+0x50/0x88) from [<c00fd114>] (set_task_comm+0x2c/0x60) [<c00fd114>] (set_task_comm+0x2c/0x60) from [<c007e184>] (kthreadd+0x30/0x108) [<c007e184>] (kthreadd+0x30/0x108) from [<c0030104>] (kernel_thread_exit+0x0/0x8) ---[ end trace 1b75b31a2719ed1c ]--- possible reason: unannotated irqs-on. irq event stamp: 3 hardirqs last enabled at (2): [<c0059bb0>] finish_task_switch+0x48/0xb0 hardirqs last disabled at (3): [<c002f0b0>] ret_slow_syscall+0xc/0x1c softirqs last enabled at (0): [<c005f3e0>] copy_process+0x394/0xe5c softirqs last disabled at (0): [<(null)>] (null) Fix this by ensuring that the lockdep interrupt state is manipulated in the appropriate places. We essentially treat userspace as an entirely separate environment which isn't relevant to lockdep (lockdep doesn't monitor userspace.) We don't tell lockdep that IRQs will be enabled in that environment. Instead, when creating kernel threads (which is a rare event compared to entering/leaving userspace) we have to update the lockdep state. Do this by starting threads with IRQs disabled, and in the kthread helper, tell lockdep that IRQs are enabled, and enable them. This provides lockdep with a consistent view of the current IRQ state in kernel space. This also revert portions of 0d928b0b616d1c5c5fe76019a87cba171ca91633 which didn't fix the problem. Tested-by: Ming Lei <tom.leiming@gmail.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-07-10 09:10:18 +00:00
disable_irq_notrace
@
@ restore SPSR and restart the instruction
@
ldr r2, [sp, #S_PSR]
svc_exit r2 @ return from exception
UNWIND(.fnend )
ENDPROC(__pabt_svc)
.align 5
.LCcralign:
.word cr_alignment
#ifdef MULTI_DABORT
.LCprocfns:
.word processor
#endif
.LCfp:
.word fp_enter
/*
* User mode handlers
*
* EABI note: sp_svc is always 64-bit aligned here, so should S_FRAME_SIZE
*/
#if defined(CONFIG_AEABI) && (__LINUX_ARM_ARCH__ >= 5) && (S_FRAME_SIZE & 7)
#error "sizeof(struct pt_regs) must be a multiple of 8"
#endif
.macro usr_entry
UNWIND(.fnstart )
UNWIND(.cantunwind ) @ don't unwind the user space
sub sp, sp, #S_FRAME_SIZE
ARM( stmib sp, {r1 - r12} )
THUMB( stmia sp, {r0 - r12} )
ldmia r0, {r1 - r3}
add r0, sp, #S_PC @ here for interlock avoidance
mov r4, #-1 @ "" "" "" ""
str r1, [sp] @ save the "real" r0 copied
@ from the exception stack
@
@ We are now ready to fill in the remaining blanks on the stack:
@
@ r2 - lr_<exception>, already fixed up for correct return/restart
@ r3 - spsr_<exception>
@ r4 - orig_r0 (see pt_regs definition in ptrace.h)
@
@ Also, separately save sp_usr and lr_usr
@
stmia r0, {r2 - r4}
ARM( stmdb r0, {sp, lr}^ )
THUMB( store_user_sp_lr r0, r1, S_SP - S_PC )
@
@ Enable the alignment trap while in kernel mode
@
alignment_trap r0
@
@ Clear FP to mark the first stack frame
@
zero_fp
.endm
.macro kuser_cmpxchg_check
#if __LINUX_ARM_ARCH__ < 6 && !defined(CONFIG_NEEDS_SYSCALL_FOR_CMPXCHG)
#ifndef CONFIG_MMU
#warning "NPTL on non MMU needs fixing"
#else
@ Make sure our user space atomic helper is restarted
@ if it was interrupted in a critical region. Here we
@ perform a quick test inline since it should be false
@ 99.9999% of the time. The rest is done out of line.
cmp r2, #TASK_SIZE
blhs kuser_cmpxchg_fixup
#endif
#endif
.endm
.align 5
__dabt_usr:
usr_entry
kuser_cmpxchg_check
dabt_helper
@
@ IRQs on, then call the main handler
@
debug_entry r1
enable_irq
mov r2, sp
adr lr, BSYM(ret_from_exception)
b do_DataAbort
UNWIND(.fnend )
ENDPROC(__dabt_usr)
.align 5
__irq_usr:
usr_entry
kuser_cmpxchg_check
ARM: 6952/1: fix lockdep warning of "unannotated irqs-off" This patch fixes the lockdep warning of "unannotated irqs-off"[1]. After entering __irq_usr, arm core will disable interrupt automatically, but __irq_usr does not annotate the irq disable, so lockdep may complain the warning if it has chance to check this in irq handler. This patch adds trace_hardirqs_off in __irq_usr before entering irq_handler to handle the irq, also calls ret_to_user_from_irq to avoid calling disable_irq again. This is also a fix for irq off tracer. [1], lockdep warning log of "unannotated irqs-off" [ 13.804687] ------------[ cut here ]------------ [ 13.809570] WARNING: at kernel/lockdep.c:3335 check_flags+0x78/0x1d0() [ 13.816467] Modules linked in: [ 13.819732] Backtrace: [ 13.822357] [<c01cb42c>] (dump_backtrace+0x0/0x100) from [<c06abb14>] (dump_stack+0x20/0x24) [ 13.831268] r6:c07d8c2c r5:00000d07 r4:00000000 r3:00000000 [ 13.837280] [<c06abaf4>] (dump_stack+0x0/0x24) from [<c01ffc04>] (warn_slowpath_common+0x5c/0x74) [ 13.846649] [<c01ffba8>] (warn_slowpath_common+0x0/0x74) from [<c01ffc48>] (warn_slowpath_null+0x2c/0x34) [ 13.856781] r8:00000000 r7:00000000 r6:c18b8194 r5:60000093 r4:ef182000 [ 13.863708] r3:00000009 [ 13.866485] [<c01ffc1c>] (warn_slowpath_null+0x0/0x34) from [<c0237d84>] (check_flags+0x78/0x1d0) [ 13.875823] [<c0237d0c>] (check_flags+0x0/0x1d0) from [<c023afc8>] (lock_acquire+0x4c/0x150) [ 13.884704] [<c023af7c>] (lock_acquire+0x0/0x150) from [<c06af638>] (_raw_spin_lock+0x4c/0x84) [ 13.893798] [<c06af5ec>] (_raw_spin_lock+0x0/0x84) from [<c01f9a44>] (sched_ttwu_pending+0x58/0x8c) [ 13.903320] r6:ef92d040 r5:00000003 r4:c18b8180 [ 13.908233] [<c01f99ec>] (sched_ttwu_pending+0x0/0x8c) from [<c01f9a90>] (scheduler_ipi+0x18/0x1c) [ 13.917663] r6:ef183fb0 r5:00000003 r4:00000000 r3:00000001 [ 13.923645] [<c01f9a78>] (scheduler_ipi+0x0/0x1c) from [<c01bc458>] (do_IPI+0x9c/0xfc) [ 13.932006] [<c01bc3bc>] (do_IPI+0x0/0xfc) from [<c06b0888>] (__irq_usr+0x48/0xe0) [ 13.939971] Exception stack(0xef183fb0 to 0xef183ff8) [ 13.945281] 3fa0: ffffffc3 0001500c 00000001 0001500c [ 13.953948] 3fc0: 00000050 400b45f0 400d9000 00000000 00000001 400d9600 6474e552 bea05b3c [ 13.962585] 3fe0: 400d96c0 bea059c0 400b6574 400b65d8 20000010 ffffffff [ 13.969573] r6:00000403 r5:fa240100 r4:ffffffff r3:20000010 [ 13.975585] ---[ end trace efc4896ab0fb62cb ]--- [ 13.980468] possible reason: unannotated irqs-off. [ 13.985534] irq event stamp: 1610 [ 13.989044] hardirqs last enabled at (1610): [<c01c703c>] no_work_pending+0x8/0x2c [ 13.997131] hardirqs last disabled at (1609): [<c01c7024>] ret_slow_syscall+0xc/0x1c [ 14.005371] softirqs last enabled at (0): [<c01fe5e4>] copy_process+0x2cc/0xa24 [ 14.013183] softirqs last disabled at (0): [< (null)>] (null) Signed-off-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-06-05 01:24:58 +00:00
#ifdef CONFIG_IRQSOFF_TRACER
bl trace_hardirqs_off
#endif
get_thread_info tsk
#ifdef CONFIG_PREEMPT
ldr r8, [tsk, #TI_PREEMPT] @ get preempt count
add r7, r8, #1 @ increment it
str r7, [tsk, #TI_PREEMPT]
#endif
irq_handler
#ifdef CONFIG_PREEMPT
ldr r0, [tsk, #TI_PREEMPT]
str r8, [tsk, #TI_PREEMPT]
teq r0, r7
ARM( strne r0, [r0, -r0] )
THUMB( movne r0, #0 )
THUMB( strne r0, [r0] )
#endif
mov why, #0
ARM: 6952/1: fix lockdep warning of "unannotated irqs-off" This patch fixes the lockdep warning of "unannotated irqs-off"[1]. After entering __irq_usr, arm core will disable interrupt automatically, but __irq_usr does not annotate the irq disable, so lockdep may complain the warning if it has chance to check this in irq handler. This patch adds trace_hardirqs_off in __irq_usr before entering irq_handler to handle the irq, also calls ret_to_user_from_irq to avoid calling disable_irq again. This is also a fix for irq off tracer. [1], lockdep warning log of "unannotated irqs-off" [ 13.804687] ------------[ cut here ]------------ [ 13.809570] WARNING: at kernel/lockdep.c:3335 check_flags+0x78/0x1d0() [ 13.816467] Modules linked in: [ 13.819732] Backtrace: [ 13.822357] [<c01cb42c>] (dump_backtrace+0x0/0x100) from [<c06abb14>] (dump_stack+0x20/0x24) [ 13.831268] r6:c07d8c2c r5:00000d07 r4:00000000 r3:00000000 [ 13.837280] [<c06abaf4>] (dump_stack+0x0/0x24) from [<c01ffc04>] (warn_slowpath_common+0x5c/0x74) [ 13.846649] [<c01ffba8>] (warn_slowpath_common+0x0/0x74) from [<c01ffc48>] (warn_slowpath_null+0x2c/0x34) [ 13.856781] r8:00000000 r7:00000000 r6:c18b8194 r5:60000093 r4:ef182000 [ 13.863708] r3:00000009 [ 13.866485] [<c01ffc1c>] (warn_slowpath_null+0x0/0x34) from [<c0237d84>] (check_flags+0x78/0x1d0) [ 13.875823] [<c0237d0c>] (check_flags+0x0/0x1d0) from [<c023afc8>] (lock_acquire+0x4c/0x150) [ 13.884704] [<c023af7c>] (lock_acquire+0x0/0x150) from [<c06af638>] (_raw_spin_lock+0x4c/0x84) [ 13.893798] [<c06af5ec>] (_raw_spin_lock+0x0/0x84) from [<c01f9a44>] (sched_ttwu_pending+0x58/0x8c) [ 13.903320] r6:ef92d040 r5:00000003 r4:c18b8180 [ 13.908233] [<c01f99ec>] (sched_ttwu_pending+0x0/0x8c) from [<c01f9a90>] (scheduler_ipi+0x18/0x1c) [ 13.917663] r6:ef183fb0 r5:00000003 r4:00000000 r3:00000001 [ 13.923645] [<c01f9a78>] (scheduler_ipi+0x0/0x1c) from [<c01bc458>] (do_IPI+0x9c/0xfc) [ 13.932006] [<c01bc3bc>] (do_IPI+0x0/0xfc) from [<c06b0888>] (__irq_usr+0x48/0xe0) [ 13.939971] Exception stack(0xef183fb0 to 0xef183ff8) [ 13.945281] 3fa0: ffffffc3 0001500c 00000001 0001500c [ 13.953948] 3fc0: 00000050 400b45f0 400d9000 00000000 00000001 400d9600 6474e552 bea05b3c [ 13.962585] 3fe0: 400d96c0 bea059c0 400b6574 400b65d8 20000010 ffffffff [ 13.969573] r6:00000403 r5:fa240100 r4:ffffffff r3:20000010 [ 13.975585] ---[ end trace efc4896ab0fb62cb ]--- [ 13.980468] possible reason: unannotated irqs-off. [ 13.985534] irq event stamp: 1610 [ 13.989044] hardirqs last enabled at (1610): [<c01c703c>] no_work_pending+0x8/0x2c [ 13.997131] hardirqs last disabled at (1609): [<c01c7024>] ret_slow_syscall+0xc/0x1c [ 14.005371] softirqs last enabled at (0): [<c01fe5e4>] copy_process+0x2cc/0xa24 [ 14.013183] softirqs last disabled at (0): [< (null)>] (null) Signed-off-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-06-05 01:24:58 +00:00
b ret_to_user_from_irq
UNWIND(.fnend )
ENDPROC(__irq_usr)
.ltorg
.align 5
__und_usr:
usr_entry
@
@ fall through to the emulation code, which returns using r9 if
@ it has emulated the instruction, or the more conventional lr
@ if we are to treat this as a real undefined instruction
@
@ r0 - instruction
@
adr r9, BSYM(ret_from_exception)
adr lr, BSYM(__und_usr_unknown)
tst r3, #PSR_T_BIT @ Thumb mode?
itet eq @ explicit IT needed for the 1f label
subeq r4, r2, #4 @ ARM instr at LR - 4
subne r4, r2, #2 @ Thumb instr at LR - 2
1: ldreqt r0, [r4]
#ifdef CONFIG_CPU_ENDIAN_BE8
reveq r0, r0 @ little endian instruction
#endif
beq call_fpe
@ Thumb instruction
#if __LINUX_ARM_ARCH__ >= 7
2:
ARM( ldrht r5, [r4], #2 )
THUMB( ldrht r5, [r4] )
THUMB( add r4, r4, #2 )
and r0, r5, #0xf800 @ mask bits 111x x... .... ....
cmp r0, #0xe800 @ 32bit instruction if xx != 0
blo __und_usr_unknown
3: ldrht r0, [r4]
add r2, r2, #2 @ r2 is PC + 2, make it PC + 4
orr r0, r0, r5, lsl #16
#else
b __und_usr_unknown
#endif
UNWIND(.fnend )
ENDPROC(__und_usr)
@
@ fallthrough to call_fpe
@
/*
* The out of line fixup for the ldrt above.
*/
.pushsection .fixup, "ax"
4: mov pc, r9
.popsection
.pushsection __ex_table,"a"
.long 1b, 4b
#if __LINUX_ARM_ARCH__ >= 7
.long 2b, 4b
.long 3b, 4b
#endif
.popsection
/*
* Check whether the instruction is a co-processor instruction.
* If yes, we need to call the relevant co-processor handler.
*
* Note that we don't do a full check here for the co-processor
* instructions; all instructions with bit 27 set are well
* defined. The only instructions that should fault are the
* co-processor instructions. However, we have to watch out
* for the ARM6/ARM7 SWI bug.
*
* NEON is a special case that has to be handled here. Not all
* NEON instructions are co-processor instructions, so we have
* to make a special case of checking for them. Plus, there's
* five groups of them, so we have a table of mask/opcode pairs
* to check against, and if any match then we branch off into the
* NEON handler code.
*
* Emulators may wish to make use of the following registers:
* r0 = instruction opcode.
* r2 = PC+4
* r9 = normal "successful" return address
* r10 = this threads thread_info structure.
* lr = unrecognised instruction return address
*/
@
@ Fall-through from Thumb-2 __und_usr
@
#ifdef CONFIG_NEON
adr r6, .LCneon_thumb_opcodes
b 2f
#endif
call_fpe:
#ifdef CONFIG_NEON
adr r6, .LCneon_arm_opcodes
2:
ldr r7, [r6], #4 @ mask value
cmp r7, #0 @ end mask?
beq 1f
and r8, r0, r7
ldr r7, [r6], #4 @ opcode bits matching in mask
cmp r8, r7 @ NEON instruction?
bne 2b
get_thread_info r10
mov r7, #1
strb r7, [r10, #TI_USED_CP + 10] @ mark CP#10 as used
strb r7, [r10, #TI_USED_CP + 11] @ mark CP#11 as used
b do_vfp @ let VFP handler handle this
1:
#endif
tst r0, #0x08000000 @ only CDP/CPRT/LDC/STC have bit 27
tstne r0, #0x04000000 @ bit 26 set on both ARM and Thumb-2
#if defined(CONFIG_CPU_ARM610) || defined(CONFIG_CPU_ARM710)
and r8, r0, #0x0f000000 @ mask out op-code bits
teqne r8, #0x0f000000 @ SWI (ARM6/7 bug)?
#endif
moveq pc, lr
get_thread_info r10 @ get current thread
and r8, r0, #0x00000f00 @ mask out CP number
THUMB( lsr r8, r8, #8 )
mov r7, #1
add r6, r10, #TI_USED_CP
ARM( strb r7, [r6, r8, lsr #8] ) @ set appropriate used_cp[]
THUMB( strb r7, [r6, r8] ) @ set appropriate used_cp[]
#ifdef CONFIG_IWMMXT
@ Test if we need to give access to iWMMXt coprocessors
ldr r5, [r10, #TI_FLAGS]
rsbs r7, r8, #(1 << 8) @ CP 0 or 1 only
movcss r7, r5, lsr #(TIF_USING_IWMMXT + 1)
bcs iwmmxt_task_enable
#endif
ARM( add pc, pc, r8, lsr #6 )
THUMB( lsl r8, r8, #2 )
THUMB( add pc, r8 )
nop
movw_pc lr @ CP#0
W(b) do_fpe @ CP#1 (FPE)
W(b) do_fpe @ CP#2 (FPE)
movw_pc lr @ CP#3
#ifdef CONFIG_CRUNCH
b crunch_task_enable @ CP#4 (MaverickCrunch)
b crunch_task_enable @ CP#5 (MaverickCrunch)
b crunch_task_enable @ CP#6 (MaverickCrunch)
#else
movw_pc lr @ CP#4
movw_pc lr @ CP#5
movw_pc lr @ CP#6
#endif
movw_pc lr @ CP#7
movw_pc lr @ CP#8
movw_pc lr @ CP#9
#ifdef CONFIG_VFP
W(b) do_vfp @ CP#10 (VFP)
W(b) do_vfp @ CP#11 (VFP)
#else
movw_pc lr @ CP#10 (VFP)
movw_pc lr @ CP#11 (VFP)
#endif
movw_pc lr @ CP#12
movw_pc lr @ CP#13
movw_pc lr @ CP#14 (Debug)
movw_pc lr @ CP#15 (Control)
#ifdef CONFIG_NEON
.align 6
.LCneon_arm_opcodes:
.word 0xfe000000 @ mask
.word 0xf2000000 @ opcode
.word 0xff100000 @ mask
.word 0xf4000000 @ opcode
.word 0x00000000 @ mask
.word 0x00000000 @ opcode
.LCneon_thumb_opcodes:
.word 0xef000000 @ mask
.word 0xef000000 @ opcode
.word 0xff100000 @ mask
.word 0xf9000000 @ opcode
.word 0x00000000 @ mask
.word 0x00000000 @ opcode
#endif
do_fpe:
enable_irq
ldr r4, .LCfp
add r10, r10, #TI_FPSTATE @ r10 = workspace
ldr pc, [r4] @ Call FP module USR entry point
/*
* The FP module is called with these registers set:
* r0 = instruction
* r2 = PC+4
* r9 = normal "successful" return address
* r10 = FP workspace
* lr = unrecognised FP instruction return address
*/
.pushsection .data
ENTRY(fp_enter)
.word no_fp
.popsection
ENTRY(no_fp)
mov pc, lr
ENDPROC(no_fp)
__und_usr_unknown:
enable_irq
mov r0, sp
adr lr, BSYM(ret_from_exception)
b do_undefinstr
ENDPROC(__und_usr_unknown)
.align 5
__pabt_usr:
usr_entry
pabt_helper
debug_entry r1
enable_irq @ Enable interrupts
mov r2, sp @ regs
bl do_PrefetchAbort @ call abort handler
UNWIND(.fnend )
/* fall through */
/*
* This is the return code to user mode for abort handlers
*/
ENTRY(ret_from_exception)
UNWIND(.fnstart )
UNWIND(.cantunwind )
get_thread_info tsk
mov why, #0
b ret_to_user
UNWIND(.fnend )
ENDPROC(__pabt_usr)
ENDPROC(ret_from_exception)
/*
* Register switch for ARMv3 and ARMv4 processors
* r0 = previous task_struct, r1 = previous thread_info, r2 = next thread_info
* previous and next are guaranteed not to be the same.
*/
ENTRY(__switch_to)
UNWIND(.fnstart )
UNWIND(.cantunwind )
add ip, r1, #TI_CPU_SAVE
ldr r3, [r2, #TI_TP_VALUE]
ARM( stmia ip!, {r4 - sl, fp, sp, lr} ) @ Store most regs on stack
THUMB( stmia ip!, {r4 - sl, fp} ) @ Store most regs on stack
THUMB( str sp, [ip], #4 )
THUMB( str lr, [ip], #4 )
#ifdef CONFIG_CPU_USE_DOMAINS
ldr r6, [r2, #TI_CPU_DOMAIN]
#endif
set_tls r3, r4, r5
#if defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_SMP)
ldr r7, [r2, #TI_TASK]
ldr r8, =__stack_chk_guard
ldr r7, [r7, #TSK_STACK_CANARY]
#endif
#ifdef CONFIG_CPU_USE_DOMAINS
mcr p15, 0, r6, c3, c0, 0 @ Set domain register
#endif
mov r5, r0
add r4, r2, #TI_CPU_SAVE
ldr r0, =thread_notify_head
mov r1, #THREAD_NOTIFY_SWITCH
bl atomic_notifier_call_chain
#if defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_SMP)
str r7, [r8]
#endif
THUMB( mov ip, r4 )
mov r0, r5
ARM( ldmia r4, {r4 - sl, fp, sp, pc} ) @ Load all regs saved previously
THUMB( ldmia ip!, {r4 - sl, fp} ) @ Load all regs saved previously
THUMB( ldr sp, [ip], #4 )
THUMB( ldr pc, [ip] )
UNWIND(.fnend )
ENDPROC(__switch_to)
__INIT
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
/*
* User helpers.
*
* These are segment of kernel provided user code reachable from user space
* at a fixed address in kernel memory. This is used to provide user space
* with some operations which require kernel help because of unimplemented
* native feature and/or instructions in many ARM CPUs. The idea is for
* this code to be executed directly in user mode for best efficiency but
* which is too intimate with the kernel counter part to be left to user
* libraries. In fact this code might even differ from one CPU to another
* depending on the available instruction set and restrictions like on
* SMP systems. In other words, the kernel reserves the right to change
* this code as needed without warning. Only the entry points and their
* results are guaranteed to be stable.
*
* Each segment is 32-byte aligned and will be moved to the top of the high
* vector page. New segments (if ever needed) must be added in front of
* existing ones. This mechanism should be used only for things that are
* really small and justified, and not be abused freely.
*
* User space is expected to implement those things inline when optimizing
* for a processor that has the necessary native support, but only if such
* resulting binaries are already to be incompatible with earlier ARM
* processors due to the use of unsupported instructions other than what
* is provided here. In other words don't make binaries unable to run on
* earlier processors just for the sake of not using these kernel helpers
* if your compiled code is not going to use the new instructions for other
* purpose.
*/
THUMB( .arm )
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
.macro usr_ret, reg
#ifdef CONFIG_ARM_THUMB
bx \reg
#else
mov pc, \reg
#endif
.endm
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
.align 5
.globl __kuser_helper_start
__kuser_helper_start:
/*
* Reference prototype:
*
* void __kernel_memory_barrier(void)
*
* Input:
*
* lr = return address
*
* Output:
*
* none
*
* Clobbered:
*
* none
*
* Definition and user space usage example:
*
* typedef void (__kernel_dmb_t)(void);
* #define __kernel_dmb (*(__kernel_dmb_t *)0xffff0fa0)
*
* Apply any needed memory barrier to preserve consistency with data modified
* manually and __kuser_cmpxchg usage.
*
* This could be used as follows:
*
* #define __kernel_dmb() \
* asm volatile ( "mov r0, #0xffff0fff; mov lr, pc; sub pc, r0, #95" \
* : : : "r0", "lr","cc" )
*/
__kuser_memory_barrier: @ 0xffff0fa0
ARM: 6516/1: Allow SMP_ON_UP to work with Thumb-2 kernels. * __fixup_smp_on_up has been modified with support for the THUMB2_KERNEL case. For THUMB2_KERNEL only, fixups are split into halfwords in case of misalignment, since we can't rely on unaligned accesses working before turning the MMU on. No attempt is made to optimise the aligned case, since the number of fixups is typically small, and it seems best to keep the code as simple as possible. * Add a rotate in the fixup_smp code in order to support CPU_BIG_ENDIAN, as suggested by Nicolas Pitre. * Add an assembly-time sanity-check to ALT_UP() to ensure that the content really is the right size (4 bytes). (No check is done for ALT_SMP(). Possibly, this could be fixed by splitting the two uses ot ALT_SMP() (ALT_SMP...SMP_UP versus ALT_SMP...SMP_UP_B) into two macros. In the first case, ALT_SMP needs to expand to >= 4 bytes, not == 4.) * smp_mpidr.h (which implements ALT_SMP()/ALT_UP() manually due to macro limitations) has not been modified: the affected instruction (mov) has no 16-bit encoding, so the correct instruction size is satisfied in this case. * A "mode" parameter has been added to smp_dmb: smp_dmb arm @ assumes 4-byte instructions (for ARM code, e.g. kuser) smp_dmb @ uses W() to ensure 4-byte instructions for ALT_SMP() This avoids assembly failures due to use of W() inside smp_dmb, when assembling pure-ARM code in the vectors page. There might be a better way to achieve this. * Kconfig: make SMP_ON_UP depend on (!THUMB2_KERNEL || !BIG_ENDIAN) i.e., THUMB2_KERNEL is now supported, but only if !BIG_ENDIAN (The fixup code for Thumb-2 currently assumes little-endian order.) Tested using a single generic realview kernel on: ARM RealView PB-A8 (CONFIG_THUMB2_KERNEL={n,y}) ARM RealView PBX-A9 (SMP) Signed-off-by: Dave Martin <dave.martin@linaro.org> Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-01 14:39:23 +00:00
smp_dmb arm
usr_ret lr
.align 5
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
/*
* Reference prototype:
*
* int __kernel_cmpxchg(int oldval, int newval, int *ptr)
*
* Input:
*
* r0 = oldval
* r1 = newval
* r2 = ptr
* lr = return address
*
* Output:
*
* r0 = returned value (zero or non-zero)
* C flag = set if r0 == 0, clear if r0 != 0
*
* Clobbered:
*
* r3, ip, flags
*
* Definition and user space usage example:
*
* typedef int (__kernel_cmpxchg_t)(int oldval, int newval, int *ptr);
* #define __kernel_cmpxchg (*(__kernel_cmpxchg_t *)0xffff0fc0)
*
* Atomically store newval in *ptr if *ptr is equal to oldval for user space.
* Return zero if *ptr was changed or non-zero if no exchange happened.
* The C flag is also set if *ptr was changed to allow for assembly
* optimization in the calling code.
*
* Notes:
*
* - This routine already includes memory barriers as needed.
*
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
* For example, a user space atomic_add implementation could look like this:
*
* #define atomic_add(ptr, val) \
* ({ register unsigned int *__ptr asm("r2") = (ptr); \
* register unsigned int __result asm("r1"); \
* asm volatile ( \
* "1: @ atomic_add\n\t" \
* "ldr r0, [r2]\n\t" \
* "mov r3, #0xffff0fff\n\t" \
* "add lr, pc, #4\n\t" \
* "add r1, r0, %2\n\t" \
* "add pc, r3, #(0xffff0fc0 - 0xffff0fff)\n\t" \
* "bcc 1b" \
* : "=&r" (__result) \
* : "r" (__ptr), "rIL" (val) \
* : "r0","r3","ip","lr","cc","memory" ); \
* __result; })
*/
__kuser_cmpxchg: @ 0xffff0fc0
#if defined(CONFIG_NEEDS_SYSCALL_FOR_CMPXCHG)
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
/*
* Poor you. No fast solution possible...
* The kernel itself must perform the operation.
* A special ghost syscall is used for that (see traps.c).
*/
stmfd sp!, {r7, lr}
ldr r7, 1f @ it's 20 bits
swi __ARM_NR_cmpxchg
ldmfd sp!, {r7, pc}
1: .word __ARM_NR_cmpxchg
#elif __LINUX_ARM_ARCH__ < 6
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
#ifdef CONFIG_MMU
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
/*
* The only thing that can break atomicity in this cmpxchg
* implementation is either an IRQ or a data abort exception
* causing another process/thread to be scheduled in the middle
* of the critical sequence. To prevent this, code is added to
* the IRQ and data abort exception handlers to set the pc back
* to the beginning of the critical section if it is found to be
* within that critical section (see kuser_cmpxchg_fixup).
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
*/
1: ldr r3, [r2] @ load current val
subs r3, r3, r0 @ compare with oldval
2: streq r1, [r2] @ store newval if eq
rsbs r0, r3, #0 @ set return val and C flag
usr_ret lr
.text
kuser_cmpxchg_fixup:
@ Called from kuser_cmpxchg_check macro.
@ r2 = address of interrupted insn (must be preserved).
@ sp = saved regs. r7 and r8 are clobbered.
@ 1b = first critical insn, 2b = last critical insn.
@ If r2 >= 1b and r2 <= 2b then saved pc_usr is set to 1b.
mov r7, #0xffff0fff
sub r7, r7, #(0xffff0fff - (0xffff0fc0 + (1b - __kuser_cmpxchg)))
subs r8, r2, r7
rsbcss r8, r8, #(2b - 1b)
strcs r7, [sp, #S_PC]
mov pc, lr
.previous
#else
#warning "NPTL on non MMU needs fixing"
mov r0, #-1
adds r0, r0, #0
usr_ret lr
#endif
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
#else
ARM: 6516/1: Allow SMP_ON_UP to work with Thumb-2 kernels. * __fixup_smp_on_up has been modified with support for the THUMB2_KERNEL case. For THUMB2_KERNEL only, fixups are split into halfwords in case of misalignment, since we can't rely on unaligned accesses working before turning the MMU on. No attempt is made to optimise the aligned case, since the number of fixups is typically small, and it seems best to keep the code as simple as possible. * Add a rotate in the fixup_smp code in order to support CPU_BIG_ENDIAN, as suggested by Nicolas Pitre. * Add an assembly-time sanity-check to ALT_UP() to ensure that the content really is the right size (4 bytes). (No check is done for ALT_SMP(). Possibly, this could be fixed by splitting the two uses ot ALT_SMP() (ALT_SMP...SMP_UP versus ALT_SMP...SMP_UP_B) into two macros. In the first case, ALT_SMP needs to expand to >= 4 bytes, not == 4.) * smp_mpidr.h (which implements ALT_SMP()/ALT_UP() manually due to macro limitations) has not been modified: the affected instruction (mov) has no 16-bit encoding, so the correct instruction size is satisfied in this case. * A "mode" parameter has been added to smp_dmb: smp_dmb arm @ assumes 4-byte instructions (for ARM code, e.g. kuser) smp_dmb @ uses W() to ensure 4-byte instructions for ALT_SMP() This avoids assembly failures due to use of W() inside smp_dmb, when assembling pure-ARM code in the vectors page. There might be a better way to achieve this. * Kconfig: make SMP_ON_UP depend on (!THUMB2_KERNEL || !BIG_ENDIAN) i.e., THUMB2_KERNEL is now supported, but only if !BIG_ENDIAN (The fixup code for Thumb-2 currently assumes little-endian order.) Tested using a single generic realview kernel on: ARM RealView PB-A8 (CONFIG_THUMB2_KERNEL={n,y}) ARM RealView PBX-A9 (SMP) Signed-off-by: Dave Martin <dave.martin@linaro.org> Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-12-01 14:39:23 +00:00
smp_dmb arm
1: ldrex r3, [r2]
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
subs r3, r3, r0
strexeq r3, r1, [r2]
teqeq r3, #1
beq 1b
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
rsbs r0, r3, #0
/* beware -- each __kuser slot must be 8 instructions max */
ALT_SMP(b __kuser_memory_barrier)
ALT_UP(usr_ret lr)
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
#endif
.align 5
/*
* Reference prototype:
*
* int __kernel_get_tls(void)
*
* Input:
*
* lr = return address
*
* Output:
*
* r0 = TLS value
*
* Clobbered:
*
* none
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
*
* Definition and user space usage example:
*
* typedef int (__kernel_get_tls_t)(void);
* #define __kernel_get_tls (*(__kernel_get_tls_t *)0xffff0fe0)
*
* Get the TLS value as previously set via the __ARM_NR_set_tls syscall.
*
* This could be used as follows:
*
* #define __kernel_get_tls() \
* ({ register unsigned int __val asm("r0"); \
* asm( "mov r0, #0xffff0fff; mov lr, pc; sub pc, r0, #31" \
* : "=r" (__val) : : "lr","cc" ); \
* __val; })
*/
__kuser_get_tls: @ 0xffff0fe0
ldr r0, [pc, #(16 - 8)] @ read TLS, set in kuser_get_tls_init
usr_ret lr
mrc p15, 0, r0, c13, c0, 3 @ 0xffff0fe8 hardware TLS code
.rep 4
.word 0 @ 0xffff0ff0 software TLS value, then
.endr @ pad up to __kuser_helper_version
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
/*
* Reference declaration:
*
* extern unsigned int __kernel_helper_version;
*
* Definition and user space usage example:
*
* #define __kernel_helper_version (*(unsigned int *)0xffff0ffc)
*
* User space may read this to determine the curent number of helpers
* available.
*/
__kuser_helper_version: @ 0xffff0ffc
.word ((__kuser_helper_end - __kuser_helper_start) >> 5)
.globl __kuser_helper_end
__kuser_helper_end:
THUMB( .thumb )
[PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-04-29 21:08:33 +00:00
/*
* Vector stubs.
*
* This code is copied to 0xffff0200 so we can use branches in the
* vectors, rather than ldr's. Note that this code must not
* exceed 0x300 bytes.
*
* Common stub entry macro:
* Enter in IRQ mode, spsr = SVC/USR CPSR, lr = SVC/USR PC
*
* SP points to a minimal amount of processor-private memory, the address
* of which is copied into r0 for the mode specific abort handler.
*/
.macro vector_stub, name, mode, correction=0
.align 5
vector_\name:
.if \correction
sub lr, lr, #\correction
.endif
@
@ Save r0, lr_<exception> (parent PC) and spsr_<exception>
@ (parent CPSR)
@
stmia sp, {r0, lr} @ save r0, lr
mrs lr, spsr
str lr, [sp, #8] @ save spsr
@
@ Prepare for SVC32 mode. IRQs remain disabled.
@
mrs r0, cpsr
eor r0, r0, #(\mode ^ SVC_MODE | PSR_ISETSTATE)
msr spsr_cxsf, r0
@
@ the branch table must immediately follow this code
@
and lr, lr, #0x0f
THUMB( adr r0, 1f )
THUMB( ldr lr, [r0, lr, lsl #2] )
mov r0, sp
ARM( ldr lr, [pc, lr, lsl #2] )
movs pc, lr @ branch to handler in SVC mode
ENDPROC(vector_\name)
.align 2
@ handler addresses follow this label
1:
.endm
.globl __stubs_start
__stubs_start:
/*
* Interrupt dispatcher
*/
vector_stub irq, IRQ_MODE, 4
.long __irq_usr @ 0 (USR_26 / USR_32)
.long __irq_invalid @ 1 (FIQ_26 / FIQ_32)
.long __irq_invalid @ 2 (IRQ_26 / IRQ_32)
.long __irq_svc @ 3 (SVC_26 / SVC_32)
.long __irq_invalid @ 4
.long __irq_invalid @ 5
.long __irq_invalid @ 6
.long __irq_invalid @ 7
.long __irq_invalid @ 8
.long __irq_invalid @ 9
.long __irq_invalid @ a
.long __irq_invalid @ b
.long __irq_invalid @ c
.long __irq_invalid @ d
.long __irq_invalid @ e
.long __irq_invalid @ f
/*
* Data abort dispatcher
* Enter in ABT mode, spsr = USR CPSR, lr = USR PC
*/
vector_stub dabt, ABT_MODE, 8
.long __dabt_usr @ 0 (USR_26 / USR_32)
.long __dabt_invalid @ 1 (FIQ_26 / FIQ_32)
.long __dabt_invalid @ 2 (IRQ_26 / IRQ_32)
.long __dabt_svc @ 3 (SVC_26 / SVC_32)
.long __dabt_invalid @ 4
.long __dabt_invalid @ 5
.long __dabt_invalid @ 6
.long __dabt_invalid @ 7
.long __dabt_invalid @ 8
.long __dabt_invalid @ 9
.long __dabt_invalid @ a
.long __dabt_invalid @ b
.long __dabt_invalid @ c
.long __dabt_invalid @ d
.long __dabt_invalid @ e
.long __dabt_invalid @ f
/*
* Prefetch abort dispatcher
* Enter in ABT mode, spsr = USR CPSR, lr = USR PC
*/
vector_stub pabt, ABT_MODE, 4
.long __pabt_usr @ 0 (USR_26 / USR_32)
.long __pabt_invalid @ 1 (FIQ_26 / FIQ_32)
.long __pabt_invalid @ 2 (IRQ_26 / IRQ_32)
.long __pabt_svc @ 3 (SVC_26 / SVC_32)
.long __pabt_invalid @ 4
.long __pabt_invalid @ 5
.long __pabt_invalid @ 6
.long __pabt_invalid @ 7
.long __pabt_invalid @ 8
.long __pabt_invalid @ 9
.long __pabt_invalid @ a
.long __pabt_invalid @ b
.long __pabt_invalid @ c
.long __pabt_invalid @ d
.long __pabt_invalid @ e
.long __pabt_invalid @ f
/*
* Undef instr entry dispatcher
* Enter in UND mode, spsr = SVC/USR CPSR, lr = SVC/USR PC
*/
vector_stub und, UND_MODE
.long __und_usr @ 0 (USR_26 / USR_32)
.long __und_invalid @ 1 (FIQ_26 / FIQ_32)
.long __und_invalid @ 2 (IRQ_26 / IRQ_32)
.long __und_svc @ 3 (SVC_26 / SVC_32)
.long __und_invalid @ 4
.long __und_invalid @ 5
.long __und_invalid @ 6
.long __und_invalid @ 7
.long __und_invalid @ 8
.long __und_invalid @ 9
.long __und_invalid @ a
.long __und_invalid @ b
.long __und_invalid @ c
.long __und_invalid @ d
.long __und_invalid @ e
.long __und_invalid @ f
.align 5
/*=============================================================================
* Undefined FIQs
*-----------------------------------------------------------------------------
* Enter in FIQ mode, spsr = ANY CPSR, lr = ANY PC
* MUST PRESERVE SVC SPSR, but need to switch to SVC mode to show our msg.
* Basically to switch modes, we *HAVE* to clobber one register... brain
* damage alert! I don't think that we can execute any code in here in any
* other mode than FIQ... Ok you can switch to another mode, but you can't
* get out of that mode without clobbering one register.
*/
vector_fiq:
disable_fiq
subs pc, lr, #4
/*=============================================================================
* Address exception handler
*-----------------------------------------------------------------------------
* These aren't too critical.
* (they're not supposed to happen, and won't happen in 32-bit data mode).
*/
vector_addrexcptn:
b vector_addrexcptn
/*
* We group all the following data together to optimise
* for CPUs with separate I & D caches.
*/
.align 5
.LCvswi:
.word vector_swi
.globl __stubs_end
__stubs_end:
.equ stubs_offset, __vectors_start + 0x200 - __stubs_start
.globl __vectors_start
__vectors_start:
ARM( swi SYS_ERROR0 )
THUMB( svc #0 )
THUMB( nop )
W(b) vector_und + stubs_offset
W(ldr) pc, .LCvswi + stubs_offset
W(b) vector_pabt + stubs_offset
W(b) vector_dabt + stubs_offset
W(b) vector_addrexcptn + stubs_offset
W(b) vector_irq + stubs_offset
W(b) vector_fiq + stubs_offset
.globl __vectors_end
__vectors_end:
.data
.globl cr_alignment
.globl cr_no_alignment
cr_alignment:
.space 4
cr_no_alignment:
.space 4
#ifdef CONFIG_MULTI_IRQ_HANDLER
.globl handle_arch_irq
handle_arch_irq:
.space 4
#endif