From 55bdd694116597d2f16510b121463cd579ba78da Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Fri, 21 May 2010 18:06:41 +0100 Subject: ARM: Add base support for ARMv7-M MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds the base support for the ARMv7-M architecture. It consists of the corresponding arch/arm/mm/ files and various #ifdef's around the kernel. Exception handling is implemented by a subsequent patch. [ukleinek: squash in some changes originating from commit b5717ba (Cortex-M3: Add support for the Microcontroller Prototyping System) from the v2.6.33-arm1 patch stack, port to post 3.6, drop zImage support, drop reorganisation of pt_regs, assert CONFIG_CPU_V7M doesn't leak into installed headers and a few cosmetic changes] Signed-off-by: Catalin Marinas Reviewed-by: Jonathan Austin Tested-by: Jonathan Austin Signed-off-by: Uwe Kleine-König --- arch/arm/kernel/head-nommu.S | 10 +++++++--- arch/arm/kernel/setup.c | 17 +++++++++++++++-- arch/arm/kernel/traps.c | 8 ++++++++ 3 files changed, 30 insertions(+), 5 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index 6a2e09c952c7..8812ce88f7a1 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -19,6 +19,7 @@ #include #include #include +#include /* * Kernel startup entry point. @@ -50,10 +51,13 @@ ENTRY(stext) setmode PSR_F_BIT | PSR_I_BIT | SVC_MODE, r9 @ ensure svc mode @ and irqs disabled -#ifndef CONFIG_CPU_CP15 - ldr r9, =CONFIG_PROCESSOR_ID -#else +#if defined(CONFIG_CPU_CP15) mrc p15, 0, r9, c0, c0 @ get processor id +#elif defined(CONFIG_CPU_V7M) + ldr r9, =BASEADDR_V7M_SCB + ldr r9, [r9, V7M_SCB_CPUID] +#else + ldr r9, =CONFIG_PROCESSOR_ID #endif bl __lookup_processor_type @ r5=procinfo r9=cpuid movs r10, r5 @ invalid processor (r5=0)? diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 1cc9e1796415..829124590e4c 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -128,7 +128,9 @@ struct stack { u32 und[3]; } ____cacheline_aligned; +#ifndef CONFIG_CPU_V7M static struct stack stacks[NR_CPUS]; +#endif char elf_platform[ELF_PLATFORM_SIZE]; EXPORT_SYMBOL(elf_platform); @@ -207,7 +209,7 @@ static const char *proc_arch[] = { "5TEJ", "6TEJ", "7", - "?(11)", + "7M", "?(12)", "?(13)", "?(14)", @@ -216,6 +218,12 @@ static const char *proc_arch[] = { "?(17)", }; +#ifdef CONFIG_CPU_V7M +static int __get_cpu_architecture(void) +{ + return CPU_ARCH_ARMv7M; +} +#else static int __get_cpu_architecture(void) { int cpu_arch; @@ -248,6 +256,7 @@ static int __get_cpu_architecture(void) return cpu_arch; } +#endif int __pure cpu_architecture(void) { @@ -293,7 +302,9 @@ static void __init cacheid_init(void) { unsigned int arch = cpu_architecture(); - if (arch >= CPU_ARCH_ARMv6) { + if (arch == CPU_ARCH_ARMv7M) { + cacheid = 0; + } else if (arch >= CPU_ARCH_ARMv6) { unsigned int cachetype = read_cpuid_cachetype(); if ((cachetype & (7 << 29)) == 4 << 29) { /* ARMv7 register format */ @@ -375,6 +386,7 @@ static void __init feat_v6_fixup(void) */ void cpu_init(void) { +#ifndef CONFIG_CPU_V7M unsigned int cpu = smp_processor_id(); struct stack *stk = &stacks[cpu]; @@ -425,6 +437,7 @@ void cpu_init(void) "I" (offsetof(struct stack, und[0])), PLC (PSR_F_BIT | PSR_I_BIT | SVC_MODE) : "r14"); +#endif } int __cpu_logical_map[NR_CPUS]; diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 1c089119b2d7..ec64571462d2 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -819,6 +819,7 @@ static void __init kuser_get_tls_init(unsigned long vectors) void __init early_trap_init(void *vectors_base) { +#ifndef CONFIG_CPU_V7M unsigned long vectors = (unsigned long)vectors_base; extern char __stubs_start[], __stubs_end[]; extern char __vectors_start[], __vectors_end[]; @@ -850,4 +851,11 @@ void __init early_trap_init(void *vectors_base) flush_icache_range(vectors, vectors + PAGE_SIZE); modify_domain(DOMAIN_USER, DOMAIN_CLIENT); +#else /* ifndef CONFIG_CPU_V7M */ + /* + * on V7-M there is no need to copy the vector table to a dedicated + * memory area. The address is configurable and so a table in the kernel + * image can be used. + */ +#endif } -- cgit v1.2.2 From 19c4d593f0b4bd46f6d923a3e514719982a22058 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Fri, 21 May 2010 18:06:42 +0100 Subject: ARM: ARMv7-M: Add support for exception handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch implements the exception handling for the ARMv7-M architecture (pretty different from the A or R profiles). It bases on work done earlier by Catalin for 2.6.33 but was nearly completely rewritten to use a pt_regs layout compatible to the A profile. Signed-off-by: Catalin Marinas Reviewed-by: Jonathan Austin Tested-by: Jonathan Austin Signed-off-by: Uwe Kleine-König --- arch/arm/kernel/entry-common.S | 4 ++ arch/arm/kernel/entry-header.S | 124 +++++++++++++++++++++++++++++++++++ arch/arm/kernel/entry-v7m.S | 143 +++++++++++++++++++++++++++++++++++++++++ 3 files changed, 271 insertions(+) create mode 100644 arch/arm/kernel/entry-v7m.S (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index 3248cde504ed..c45e0027613b 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -339,6 +339,9 @@ ENDPROC(ftrace_stub) .align 5 ENTRY(vector_swi) +#ifdef CONFIG_CPU_V7M + v7m_exception_entry +#else sub sp, sp, #S_FRAME_SIZE stmia sp, {r0 - r12} @ Calling r0 - r12 ARM( add r8, sp, #S_PC ) @@ -349,6 +352,7 @@ ENTRY(vector_swi) str lr, [sp, #S_PC] @ Save calling PC str r8, [sp, #S_PSR] @ Save CPSR str r0, [sp, #S_OLD_R0] @ Save OLD_R0 +#endif zero_fp /* diff --git a/arch/arm/kernel/entry-header.S b/arch/arm/kernel/entry-header.S index 9a8531eadd3d..b02e1394dc54 100644 --- a/arch/arm/kernel/entry-header.S +++ b/arch/arm/kernel/entry-header.S @@ -5,6 +5,7 @@ #include #include #include +#include @ Bad Abort numbers @ ----------------- @@ -44,6 +45,116 @@ #endif .endm +#ifdef CONFIG_CPU_V7M +/* + * ARMv7-M exception entry/exit macros. + * + * xPSR, ReturnAddress(), LR (R14), R12, R3, R2, R1, and R0 are + * automatically saved on the current stack (32 words) before + * switching to the exception stack (SP_main). + * + * If exception is taken while in user mode, SP_main is + * empty. Otherwise, SP_main is aligned to 64 bit automatically + * (CCR.STKALIGN set). + * + * Linux assumes that the interrupts are disabled when entering an + * exception handler and it may BUG if this is not the case. Interrupts + * are disabled during entry and reenabled in the exit macro. + * + * v7m_exception_slow_exit is used when returning from SVC or PendSV. + * When returning to kernel mode, we don't return from exception. + */ + .macro v7m_exception_entry + @ determine the location of the registers saved by the core during + @ exception entry. Depending on the mode the cpu was in when the + @ exception happend that is either on the main or the process stack. + @ Bit 2 of EXC_RETURN stored in the lr register specifies which stack + @ was used. + tst lr, #EXC_RET_STACK_MASK + mrsne r12, psp + moveq r12, sp + + @ we cannot rely on r0-r3 and r12 matching the value saved in the + @ exception frame because of tail-chaining. So these have to be + @ reloaded. + ldmia r12!, {r0-r3} + + @ Linux expects to have irqs off. Do it here before taking stack space + cpsid i + + sub sp, #S_FRAME_SIZE-S_IP + stmdb sp!, {r0-r11} + + @ load saved r12, lr, return address and xPSR. + @ r0-r7 are used for signals and never touched from now on. Clobbering + @ r8-r12 is OK. + mov r9, r12 + ldmia r9!, {r8, r10-r12} + + @ calculate the original stack pointer value. + @ r9 currently points to the memory location just above the auto saved + @ xPSR. + @ The cpu might automatically 8-byte align the stack. Bit 9 + @ of the saved xPSR specifies if stack aligning took place. In this case + @ another 32-bit value is included in the stack. + + tst r12, V7M_xPSR_FRAMEPTRALIGN + addne r9, r9, #4 + + @ store saved r12 using str to have a register to hold the base for stm + str r8, [sp, #S_IP] + add r8, sp, #S_SP + @ store r13-r15, xPSR + stmia r8!, {r9-r12} + @ store old_r0 + str r0, [r8] + .endm + + /* + * PENDSV and SVCALL are configured to have the same exception + * priorities. As a kernel thread runs at SVCALL execution priority it + * can never be preempted and so we will never have to return to a + * kernel thread here. + */ + .macro v7m_exception_slow_exit ret_r0 + cpsid i + ldr lr, =EXC_RET_THREADMODE_PROCESSSTACK + + @ read original r12, sp, lr, pc and xPSR + add r12, sp, #S_IP + ldmia r12, {r1-r5} + + @ an exception frame is always 8-byte aligned. To tell the hardware if + @ the sp to be restored is aligned or not set bit 9 of the saved xPSR + @ accordingly. + tst r2, #4 + subne r2, r2, #4 + orrne r5, V7M_xPSR_FRAMEPTRALIGN + biceq r5, V7M_xPSR_FRAMEPTRALIGN + + @ write basic exception frame + stmdb r2!, {r1, r3-r5} + ldmia sp, {r1, r3-r5} + .if \ret_r0 + stmdb r2!, {r0, r3-r5} + .else + stmdb r2!, {r1, r3-r5} + .endif + + @ restore process sp + msr psp, r2 + + @ restore original r4-r11 + ldmia sp!, {r0-r11} + + @ restore main sp + add sp, sp, #S_FRAME_SIZE-S_IP + + cpsie i + bx lr + .endm +#endif /* CONFIG_CPU_V7M */ + @ @ Store/load the USER SP and LR registers by switching to the SYS @ mode. Useful in Thumb-2 mode where "stm/ldm rd, {sp, lr}^" is not @@ -131,6 +242,18 @@ rfeia sp! .endm +#ifdef CONFIG_CPU_V7M + /* + * Note we don't need to do clrex here as clearing the local monitor is + * part of each exception entry and exit sequence. + */ + .macro restore_user_regs, fast = 0, offset = 0 + .if \offset + add sp, #\offset + .endif + v7m_exception_slow_exit ret_r0 = \fast + .endm +#else /* ifdef CONFIG_CPU_V7M */ .macro restore_user_regs, fast = 0, offset = 0 clrex @ clear the exclusive monitor mov r2, sp @@ -147,6 +270,7 @@ add sp, sp, #S_FRAME_SIZE - S_SP movs pc, lr @ return & move spsr_svc into cpsr .endm +#endif /* ifdef CONFIG_CPU_V7M / else */ .macro get_thread_info, rd mov \rd, sp diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S new file mode 100644 index 000000000000..e00621f1403f --- /dev/null +++ b/arch/arm/kernel/entry-v7m.S @@ -0,0 +1,143 @@ +/* + * linux/arch/arm/kernel/entry-v7m.S + * + * Copyright (C) 2008 ARM Ltd. + * + * 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 for the ARMv7-M architecture + */ +#include +#include +#include +#include + +#include + +#include "entry-header.S" + +#ifdef CONFIG_TRACE_IRQFLAGS +#error "CONFIG_TRACE_IRQFLAGS not supported on the current ARMv7M implementation" +#endif + +__invalid_entry: + v7m_exception_entry + adr r0, strerr + mrs r1, ipsr + mov r2, lr + bl printk + mov r0, sp + bl show_regs +1: b 1b +ENDPROC(__invalid_entry) + +strerr: .asciz "\nUnhandled exception: IPSR = %08lx LR = %08lx\n" + + .align 2 +__irq_entry: + v7m_exception_entry + + @ + @ Invoke the IRQ handler + @ + mrs r0, ipsr + ldr r1, =V7M_xPSR_EXCEPTIONNO + and r0, r1 + sub r0, #16 + mov r1, sp + stmdb sp!, {lr} + @ routine called with r0 = irq number, r1 = struct pt_regs * + bl nvic_do_IRQ + + pop {lr} + @ + @ Check for any pending work if returning to user + @ + ldr r1, =BASEADDR_V7M_SCB + ldr r0, [r1, V7M_SCB_ICSR] + tst r0, V7M_SCB_ICSR_RETTOBASE + beq 2f + + get_thread_info tsk + ldr r2, [tsk, #TI_FLAGS] + tst r2, #_TIF_WORK_MASK + beq 2f @ no work pending + mov r0, #V7M_SCB_ICSR_PENDSVSET + str r0, [r1, V7M_SCB_ICSR] @ raise PendSV + +2: + @ registers r0-r3 and r12 are automatically restored on exception + @ return. r4-r7 were not clobbered in v7m_exception_entry so for + @ correctness they don't need to be restored. So only r8-r11 must be + @ restored here. The easiest way to do so is to restore r0-r7, too. + ldmia sp!, {r0-r11} + add sp, #S_FRAME_SIZE-S_IP + cpsie i + bx lr +ENDPROC(__irq_entry) + +__pendsv_entry: + v7m_exception_entry + + ldr r1, =BASEADDR_V7M_SCB + mov r0, #V7M_SCB_ICSR_PENDSVCLR + str r0, [r1, V7M_SCB_ICSR] @ clear PendSV + + @ execute the pending work, including reschedule + get_thread_info tsk + mov why, #0 + b ret_to_user +ENDPROC(__pendsv_entry) + +/* + * Register switch for ARMv7-M 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) + .fnstart + .cantunwind + add ip, r1, #TI_CPU_SAVE + stmia ip!, {r4 - r11} @ Store most regs on stack + str sp, [ip], #4 + str lr, [ip], #4 + mov r5, r0 + add r4, r2, #TI_CPU_SAVE + ldr r0, =thread_notify_head + mov r1, #THREAD_NOTIFY_SWITCH + bl atomic_notifier_call_chain + mov ip, r4 + mov r0, r5 + ldmia ip!, {r4 - r11} @ Load all regs saved previously + ldr sp, [ip] + ldr pc, [ip, #4]! + .fnend +ENDPROC(__switch_to) + + .data + .align 8 +/* + * Vector table (64 words => 256 bytes natural alignment) + */ +ENTRY(vector_table) + .long 0 @ 0 - Reset stack pointer + .long __invalid_entry @ 1 - Reset + .long __invalid_entry @ 2 - NMI + .long __invalid_entry @ 3 - HardFault + .long __invalid_entry @ 4 - MemManage + .long __invalid_entry @ 5 - BusFault + .long __invalid_entry @ 6 - UsageFault + .long __invalid_entry @ 7 - Reserved + .long __invalid_entry @ 8 - Reserved + .long __invalid_entry @ 9 - Reserved + .long __invalid_entry @ 10 - Reserved + .long vector_swi @ 11 - SVCall + .long __invalid_entry @ 12 - Debug Monitor + .long __invalid_entry @ 13 - Reserved + .long __pendsv_entry @ 14 - PendSV + .long __invalid_entry @ 15 - SysTick + .rept 64 - 16 + .long __irq_entry @ 16..64 - External Interrupts + .endr -- cgit v1.2.2 From 4477ca45fb368880bf77b10ed3b24b03f0cc82da Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Thu, 21 Mar 2013 21:02:37 +0100 Subject: ARM: ARMv7-M: Allow the building of new kernel port MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch modifies the required Kconfig and Makefile files to allow the building of kernel for Cortex-M3. Signed-off-by: Catalin Marinas Reviewed-by: Jonathan Austin Tested-by: Jonathan Austin Signed-off-by: Uwe Kleine-König --- arch/arm/kernel/Makefile | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index 5f3338eacad2..00d703c49f82 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -15,7 +15,7 @@ CFLAGS_REMOVE_return_address.o = -pg # Object file lists. -obj-y := elf.o entry-armv.o entry-common.o irq.o opcodes.o \ +obj-y := elf.o entry-common.o irq.o opcodes.o \ process.o ptrace.o return_address.o sched_clock.o \ setup.o signal.o stacktrace.o sys_arm.o time.o traps.o @@ -23,6 +23,12 @@ obj-$(CONFIG_ATAGS) += atags_parse.o obj-$(CONFIG_ATAGS_PROC) += atags_proc.o obj-$(CONFIG_DEPRECATED_PARAM_STRUCT) += atags_compat.o +ifeq ($(CONFIG_CPU_V7M),y) +obj-y += entry-v7m.o +else +obj-y += entry-armv.o +endif + obj-$(CONFIG_OC_ETM) += etm.o obj-$(CONFIG_CPU_IDLE) += cpuidle.o obj-$(CONFIG_ISA_DMA_API) += dma.o -- cgit v1.2.2 From 9394c1c65e61eb6f4c1c99f342b49e451ec337b6 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Mon, 11 Mar 2013 13:52:12 +0100 Subject: ARM: 7669/1: keep __my_cpu_offset consistent with generic one Commit 14318efb(ARM: 7587/1: implement optimized percpu variable access) introduces arm's __my_cpu_offset to optimize percpu vaiable access, which really works well on hackbench, but will cause __my_cpu_offset to return garbage value before it is initialized in cpu_init() called by setup_arch, so accessing percpu variable before setup_arch may cause kernel hang. But generic __my_cpu_offset always returns zero before percpu area is brought up, and won't hang kernel. So the patch tries to clear __my_cpu_offset on boot CPU early to avoid boot hang. At least now percpu variable is accessed by lockdep before setup_arch(), and enabling CONFIG_LOCK_STAT or CONFIG_DEBUG_LOCKDEP can trigger kernel hang. Signed-off-by: Ming Lei Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 1522c7ae31b0..dd1c6aacbaf9 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -456,6 +456,13 @@ void __init smp_setup_processor_id(void) for (i = 1; i < nr_cpu_ids; ++i) cpu_logical_map(i) = i == cpu ? 0 : i; + /* + * clear __my_cpu_offset on boot CPU to avoid hang caused by + * using percpu variable early, for example, lockdep will + * access percpu variable inside lock_release + */ + set_my_cpu_offset(0); + printk(KERN_INFO "Booting Linux on physical CPU 0x%x\n", mpidr); } -- cgit v1.2.2 From 05774088391c7430f6a4c1d5d18196ef17bb3ba9 Mon Sep 17 00:00:00 2001 From: Stefano Stabellini Date: Tue, 21 May 2013 14:24:11 +0000 Subject: arm: introduce psci_smp_ops Rename virt_smp_ops to psci_smp_ops and move them to arch/arm/kernel/psci_smp.c. Remove mach-virt/platsmp.c, now unused. Compile psci_smp if CONFIG_ARM_PSCI and CONFIG_SMP. Add a cpu_die smp_op based on psci_ops.cpu_off. Initialize PSCI before setting smp_ops in setup_arch. If PSCI is available on the platform, prefer psci_smp_ops over the platform smp_ops. Signed-off-by: Stefano Stabellini Acked-by: Will Deacon CC: arnd@arndb.de CC: marc.zyngier@arm.com CC: linux@arm.linux.org.uk CC: nico@linaro.org CC: rob.herring@calxeda.com --- arch/arm/kernel/Makefile | 5 ++- arch/arm/kernel/psci.c | 7 ++-- arch/arm/kernel/psci_smp.c | 84 ++++++++++++++++++++++++++++++++++++++++++++++ arch/arm/kernel/setup.c | 7 +++- 4 files changed, 97 insertions(+), 6 deletions(-) create mode 100644 arch/arm/kernel/psci_smp.c (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index 5f3338eacad2..dd9d90ab65d0 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -82,6 +82,9 @@ obj-$(CONFIG_DEBUG_LL) += debug.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_ARM_VIRT_EXT) += hyp-stub.o -obj-$(CONFIG_ARM_PSCI) += psci.o +ifeq ($(CONFIG_ARM_PSCI),y) +obj-y += psci.o +obj-$(CONFIG_SMP) += psci_smp.o +endif extra-y := $(head-y) vmlinux.lds diff --git a/arch/arm/kernel/psci.c b/arch/arm/kernel/psci.c index 36531643cc2c..46931880093d 100644 --- a/arch/arm/kernel/psci.c +++ b/arch/arm/kernel/psci.c @@ -158,7 +158,7 @@ static const struct of_device_id psci_of_match[] __initconst = { {}, }; -static int __init psci_init(void) +void __init psci_init(void) { struct device_node *np; const char *method; @@ -166,7 +166,7 @@ static int __init psci_init(void) np = of_find_matching_node(NULL, psci_of_match); if (!np) - return 0; + return; pr_info("probing function IDs from device-tree\n"); @@ -206,6 +206,5 @@ static int __init psci_init(void) out_put_node: of_node_put(np); - return 0; + return; } -early_initcall(psci_init); diff --git a/arch/arm/kernel/psci_smp.c b/arch/arm/kernel/psci_smp.c new file mode 100644 index 000000000000..23a11424c568 --- /dev/null +++ b/arch/arm/kernel/psci_smp.c @@ -0,0 +1,84 @@ +/* + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Copyright (C) 2012 ARM Limited + * + * Author: Will Deacon + */ + +#include +#include +#include +#include + +#include +#include + +/* + * psci_smp assumes that the following is true about PSCI: + * + * cpu_suspend Suspend the execution on a CPU + * @state we don't currently describe affinity levels, so just pass 0. + * @entry_point the first instruction to be executed on return + * returns 0 success, < 0 on failure + * + * cpu_off Power down a CPU + * @state we don't currently describe affinity levels, so just pass 0. + * no return on successful call + * + * cpu_on Power up a CPU + * @cpuid cpuid of target CPU, as from MPIDR + * @entry_point the first instruction to be executed on return + * returns 0 success, < 0 on failure + * + * migrate Migrate the context to a different CPU + * @cpuid cpuid of target CPU, as from MPIDR + * returns 0 success, < 0 on failure + * + */ + +extern void secondary_startup(void); + +static int __cpuinit psci_boot_secondary(unsigned int cpu, + struct task_struct *idle) +{ + if (psci_ops.cpu_on) + return psci_ops.cpu_on(cpu_logical_map(cpu), + __pa(secondary_startup)); + return -ENODEV; +} + +#ifdef CONFIG_HOTPLUG_CPU +void __ref psci_cpu_die(unsigned int cpu) +{ + const struct psci_power_state ps = { + .type = PSCI_POWER_STATE_TYPE_POWER_DOWN, + }; + + if (psci_ops.cpu_off) + psci_ops.cpu_off(ps); + + /* We should never return */ + panic("psci: cpu %d failed to shutdown\n", cpu); +} +#else +#define psci_cpu_die NULL +#endif + +bool __init psci_smp_available(void) +{ + /* is cpu_on available at least? */ + return (psci_ops.cpu_on != NULL); +} + +struct smp_operations __initdata psci_smp_ops = { + .smp_boot_secondary = psci_boot_secondary, + .cpu_die = psci_cpu_die, +}; diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 1522c7ae31b0..8f7a6e9926a8 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include #include @@ -796,9 +797,13 @@ void __init setup_arch(char **cmdline_p) unflatten_device_tree(); arm_dt_init_cpu_maps(); + psci_init(); #ifdef CONFIG_SMP if (is_smp()) { - smp_set_ops(mdesc->smp); + if (psci_smp_available()) + smp_set_ops(&psci_smp_ops); + else if (mdesc->smp) + smp_set_ops(mdesc->smp); smp_init_cpus(); } #endif -- cgit v1.2.2 From b382b940f821784107ca22de3455bb90e4512557 Mon Sep 17 00:00:00 2001 From: Jon Medhurst Date: Tue, 21 May 2013 13:40:51 +0000 Subject: ARM: Enable selection of SMP operations at boot time Add a new 'smp_init' hook to machine_desc so platforms can specify a function to be used to setup smp ops instead of having a statically defined value. The hook must return true when smp_ops are initialized. If false the static mdesc->smp_ops will be used by default. Add the definition of "bool" by including the linux/types.h file to asm/mach/arch.h and make it self-contained. Signed-off-by: Jon Medhurst Signed-off-by: Nicolas Pitre Signed-off-by: Stefano Stabellini Signed-off-by: Nicolas Ferre Reviewed-by: Santosh Shilimkar --- arch/arm/kernel/setup.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 8f7a6e9926a8..2b3ba15408bd 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -800,10 +800,12 @@ void __init setup_arch(char **cmdline_p) psci_init(); #ifdef CONFIG_SMP if (is_smp()) { - if (psci_smp_available()) - smp_set_ops(&psci_smp_ops); - else if (mdesc->smp) - smp_set_ops(mdesc->smp); + if (!mdesc->smp_init || !mdesc->smp_init()) { + if (psci_smp_available()) + smp_set_ops(&psci_smp_ops); + else if (mdesc->smp) + smp_set_ops(mdesc->smp); + } smp_init_cpus(); } #endif -- cgit v1.2.2 From 4756dcbfd37819a8359d3c69a22be2ee41666d0f Mon Sep 17 00:00:00 2001 From: Cyril Chemparathy Date: Sat, 21 Jul 2012 15:55:04 -0400 Subject: ARM: LPAE: accomodate >32-bit addresses for page table base This patch redefines the early boot time use of the R4 register to steal a few low order bits (ARCH_PGD_SHIFT bits) on LPAE systems. This allows for up to 38-bit physical addresses. Signed-off-by: Cyril Chemparathy Signed-off-by: Vitaly Andrianov Acked-by: Nicolas Pitre Tested-by: Santosh Shilimkar Tested-by: Subash Patel Signed-off-by: Will Deacon --- arch/arm/kernel/head.S | 10 ++++------ arch/arm/kernel/smp.c | 11 +++++++++-- 2 files changed, 13 insertions(+), 8 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 8bac553fe213..45e8935cae4e 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -156,7 +156,7 @@ ENDPROC(stext) * * Returns: * r0, r3, r5-r7 corrupted - * r4 = physical page table address + * r4 = page table (see ARCH_PGD_SHIFT in asm/memory.h) */ __create_page_tables: pgtbl r4, r8 @ page table address @@ -331,6 +331,7 @@ __create_page_tables: #endif #ifdef CONFIG_ARM_LPAE sub r4, r4, #0x1000 @ point to the PGD table + mov r4, r4, lsr #ARCH_PGD_SHIFT #endif mov pc, lr ENDPROC(__create_page_tables) @@ -408,7 +409,7 @@ __secondary_data: * r0 = cp#15 control register * r1 = machine ID * r2 = atags or dtb pointer - * r4 = page table pointer + * r4 = page table (see ARCH_PGD_SHIFT in asm/memory.h) * r9 = processor ID * r13 = *virtual* address to jump to upon completion */ @@ -427,10 +428,7 @@ __enable_mmu: #ifdef CONFIG_CPU_ICACHE_DISABLE bic r0, r0, #CR_I #endif -#ifdef CONFIG_ARM_LPAE - mov r5, #0 - mcrr p15, 0, r4, r5, c2 @ load TTBR0 -#else +#ifndef CONFIG_ARM_LPAE mov r5, #(domain_val(DOMAIN_USER, DOMAIN_MANAGER) | \ domain_val(DOMAIN_KERNEL, DOMAIN_MANAGER) | \ domain_val(DOMAIN_TABLE, DOMAIN_MANAGER) | \ diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 550d63cef68e..217b755aadd4 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -78,6 +78,13 @@ void __init smp_set_ops(struct smp_operations *ops) smp_ops = *ops; }; +static unsigned long get_arch_pgd(pgd_t *pgd) +{ + phys_addr_t pgdir = virt_to_phys(pgd); + BUG_ON(pgdir & ARCH_PGD_MASK); + return pgdir >> ARCH_PGD_SHIFT; +} + int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle) { int ret; @@ -87,8 +94,8 @@ int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle) * its stack and the page tables. */ secondary_data.stack = task_stack_page(idle) + THREAD_START_SP; - secondary_data.pgdir = virt_to_phys(idmap_pgd); - secondary_data.swapper_pg_dir = virt_to_phys(swapper_pg_dir); + secondary_data.pgdir = get_arch_pgd(idmap_pgd); + secondary_data.swapper_pg_dir = get_arch_pgd(swapper_pg_dir); __cpuc_flush_dcache_area(&secondary_data, sizeof(secondary_data)); outer_clean_range(__pa(&secondary_data), __pa(&secondary_data + 1)); -- cgit v1.2.2 From a469abd0f868c902b75532579bf87553dcf1b360 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 8 Apr 2013 17:13:12 +0100 Subject: ARM: elf: add new hwcap for identifying atomic ldrd/strd instructions CPUs implementing LPAE have atomic ldrd/strd instructions, meaning that userspace software can avoid having to use the exclusive variants of these instructions if they wish. This patch advertises the atomicity of these instructions via the hwcaps, so userspace can detect this CPU feature. Reported-by: Vladimir Danushevsky Signed-off-by: Will Deacon --- arch/arm/kernel/setup.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 1522c7ae31b0..bdcd4dd13230 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -355,7 +355,7 @@ void __init early_print(const char *str, ...) static void __init cpuid_init_hwcaps(void) { - unsigned int divide_instrs; + unsigned int divide_instrs, vmsa; if (cpu_architecture() < CPU_ARCH_ARMv7) return; @@ -368,6 +368,11 @@ static void __init cpuid_init_hwcaps(void) case 1: elf_hwcap |= HWCAP_IDIVT; } + + /* LPAE implies atomic ldrd/strd instructions */ + vmsa = (read_cpuid_ext(CPUID_EXT_MMFR0) & 0xf) >> 0; + if (vmsa >= 5) + elf_hwcap |= HWCAP_LPAE; } static void __init feat_v6_fixup(void) @@ -872,6 +877,7 @@ static const char *hwcap_str[] = { "vfpv4", "idiva", "idivt", + "lpae", NULL }; -- cgit v1.2.2 From 14d86e725ed034917bc721cf5deea019857b6cf0 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Thu, 11 Apr 2013 23:32:28 +0200 Subject: ARM: pci: create pci_common_init_dev() When working with device tree support for PCI on ARM you run into a problem when mapping IRQs from the device tree irqmaps: doing this the code in drivers/of/of_pci_irq.c will try to find the OF node on the root bridge and this fails, because bus->dev.of_node is NULL, and that in turn boils down to the fact that pci_set_bus_of_node() has called pcibios_get_phb_of_node() from drivers/pci/of.c to obtain the OF node of the bridge or its parent and none is set and thus NULL is returned. Fix this by adding an additional parent argument API for registering PCI bridges on the ARM architecture called pci_common_init_dev(), and pass along this parent to pci_scan_root_bus() called from pcibios_init_hw() in bios32.c and voila: the IRQ mappings start working: the OF node can be retrieved from the parent. Create the old pci_common_init() as a wrapper around the new call. Cc: Mike Rapoport Cc: Russell King Cc: Ralf Baechle Cc: Arnd Bergmann Cc: Benjamin Herrenschmitt Reviewed-by: Andrew Murray Reviewed-by: Thierry Reding Signed-off-by: Linus Walleij --- arch/arm/kernel/bios32.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/bios32.c b/arch/arm/kernel/bios32.c index b2ed73c45489..261fcc826169 100644 --- a/arch/arm/kernel/bios32.c +++ b/arch/arm/kernel/bios32.c @@ -445,7 +445,8 @@ static int pcibios_init_resources(int busnr, struct pci_sys_data *sys) return 0; } -static void pcibios_init_hw(struct hw_pci *hw, struct list_head *head) +static void pcibios_init_hw(struct device *parent, struct hw_pci *hw, + struct list_head *head) { struct pci_sys_data *sys = NULL; int ret; @@ -480,7 +481,7 @@ static void pcibios_init_hw(struct hw_pci *hw, struct list_head *head) if (hw->scan) sys->bus = hw->scan(nr, sys); else - sys->bus = pci_scan_root_bus(NULL, sys->busnr, + sys->bus = pci_scan_root_bus(parent, sys->busnr, hw->ops, sys, &sys->resources); if (!sys->bus) @@ -497,7 +498,7 @@ static void pcibios_init_hw(struct hw_pci *hw, struct list_head *head) } } -void pci_common_init(struct hw_pci *hw) +void pci_common_init_dev(struct device *parent, struct hw_pci *hw) { struct pci_sys_data *sys; LIST_HEAD(head); @@ -505,7 +506,7 @@ void pci_common_init(struct hw_pci *hw) pci_add_flags(PCI_REASSIGN_ALL_RSRC); if (hw->preinit) hw->preinit(); - pcibios_init_hw(hw, &head); + pcibios_init_hw(parent, hw, &head); if (hw->postinit) hw->postinit(); -- cgit v1.2.2 From 40b313608ad4ea655addd2ec6cdd106477ae8e15 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Tue, 21 May 2013 13:49:35 +1000 Subject: Finally eradicate CONFIG_HOTPLUG Ever since commit 45f035ab9b8f ("CONFIG_HOTPLUG should be always on"), it has been basically impossible to build a kernel with CONFIG_HOTPLUG turned off. Remove all the remaining references to it. Cc: Russell King Cc: Doug Thompson Cc: Bjorn Helgaas Cc: Steven Whitehouse Cc: Arnd Bergmann Cc: Pavel Machek Cc: "Rafael J. Wysocki" Cc: Andrew Morton Signed-off-by: Stephen Rothwell Acked-by: Mauro Carvalho Chehab Acked-by: Hans Verkuil Signed-off-by: Greg Kroah-Hartman --- arch/arm/kernel/module.c | 8 -------- arch/arm/kernel/vmlinux.lds.S | 4 ---- 2 files changed, 12 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c index 1e9be5d25e56..85c3fb6c93c2 100644 --- a/arch/arm/kernel/module.c +++ b/arch/arm/kernel/module.c @@ -288,24 +288,16 @@ int module_finalize(const Elf32_Ehdr *hdr, const Elf_Shdr *sechdrs, if (strcmp(".ARM.exidx.init.text", secname) == 0) maps[ARM_SEC_INIT].unw_sec = s; - else if (strcmp(".ARM.exidx.devinit.text", secname) == 0) - maps[ARM_SEC_DEVINIT].unw_sec = s; else if (strcmp(".ARM.exidx", secname) == 0) maps[ARM_SEC_CORE].unw_sec = s; else if (strcmp(".ARM.exidx.exit.text", secname) == 0) maps[ARM_SEC_EXIT].unw_sec = s; - else if (strcmp(".ARM.exidx.devexit.text", secname) == 0) - maps[ARM_SEC_DEVEXIT].unw_sec = s; else if (strcmp(".init.text", secname) == 0) maps[ARM_SEC_INIT].txt_sec = s; - else if (strcmp(".devinit.text", secname) == 0) - maps[ARM_SEC_DEVINIT].txt_sec = s; else if (strcmp(".text", secname) == 0) maps[ARM_SEC_CORE].txt_sec = s; else if (strcmp(".exit.text", secname) == 0) maps[ARM_SEC_EXIT].txt_sec = s; - else if (strcmp(".devexit.text", secname) == 0) - maps[ARM_SEC_DEVEXIT].txt_sec = s; } for (i = 0; i < ARM_SEC_MAX; i++) diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index a871b8e00fca..fa25e4e425f6 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -70,10 +70,6 @@ SECTIONS ARM_EXIT_DISCARD(EXIT_TEXT) ARM_EXIT_DISCARD(EXIT_DATA) EXIT_CALL -#ifndef CONFIG_HOTPLUG - *(.ARM.exidx.devexit.text) - *(.ARM.extab.devexit.text) -#endif #ifndef CONFIG_MMU *(.fixup) *(__ex_table) -- cgit v1.2.2 From fdeb94b5dc5bf9db7b3e36f3f38089a554f6a108 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 3 Jun 2013 23:09:14 +0100 Subject: ARM: 7745/1: psci: fix building without HOTPLUG_CPU The cpu_die field in smp_operations is not valid with CONFIG_HOTPLUG_CPU, so we must enclose it in #ifdef, but at least that lets us remove two other lines. Signed-off-by: Arnd Bergmann Cc: Stefano Stabellini Cc: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/psci_smp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/psci_smp.c b/arch/arm/kernel/psci_smp.c index 23a11424c568..219f1d73572a 100644 --- a/arch/arm/kernel/psci_smp.c +++ b/arch/arm/kernel/psci_smp.c @@ -68,8 +68,6 @@ void __ref psci_cpu_die(unsigned int cpu) /* We should never return */ panic("psci: cpu %d failed to shutdown\n", cpu); } -#else -#define psci_cpu_die NULL #endif bool __init psci_smp_available(void) @@ -80,5 +78,7 @@ bool __init psci_smp_available(void) struct smp_operations __initdata psci_smp_ops = { .smp_boot_secondary = psci_boot_secondary, +#ifdef CONFIG_HOTPLUG_CPU .cpu_die = psci_cpu_die, +#endif }; -- cgit v1.2.2 From 0af0b189abf73d232af782df2f999235cd2fed7f Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Wed, 30 Jan 2013 18:17:49 +0000 Subject: ARM: hyp: initialize CNTVOFF to zero In order to be able to use the virtual counter in a safe way, make sure it is initialized to zero before dropping to SVC. Signed-off-by: Marc Zyngier Signed-off-by: Mark Rutland Acked-by: Santosh Shilimkar Cc: Dave Martin --- arch/arm/kernel/hyp-stub.S | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S index 1315c4ccfa56..dbe21107945a 100644 --- a/arch/arm/kernel/hyp-stub.S +++ b/arch/arm/kernel/hyp-stub.S @@ -153,6 +153,8 @@ THUMB( orr r7, #(1 << 30) ) @ HSCTLR.TE mrc p15, 4, r7, c14, c1, 0 @ CNTHCTL orr r7, r7, #3 @ PL1PCEN | PL1PCTEN mcr p15, 4, r7, c14, c1, 0 @ CNTHCTL + mov r7, #0 + mcrr p15, 4, r7, r7, c14 @ CNTVOFF 1: #endif -- cgit v1.2.2 From 3f71be237ce37e0131973ebfa33b326bc51d743e Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 12 Mar 2013 14:56:12 +0000 Subject: ARM: arch_timer: stop virtual timer when booted in HYP mode When booting the kernel, a bootloader could have left the virtual timer ticking away, potentially generating interrupts. This could be troublesome if the user of the virtual timer is not careful when enabling the interrupt. In order to avoid any surprise, stop the virtual timer from interrupting us when booted in HYP mode, as we'll use the physical timer in this case. Reported-by: Giridhar Maruthy Signed-off-by: Marc Zyngier Signed-off-by: Mark Rutland Acked-by: Santosh Shilimkar Cc: Dave Martin --- arch/arm/kernel/hyp-stub.S | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S index dbe21107945a..4910232c4833 100644 --- a/arch/arm/kernel/hyp-stub.S +++ b/arch/arm/kernel/hyp-stub.S @@ -155,6 +155,11 @@ THUMB( orr r7, #(1 << 30) ) @ HSCTLR.TE mcr p15, 4, r7, c14, c1, 0 @ CNTHCTL mov r7, #0 mcrr p15, 4, r7, r7, c14 @ CNTVOFF + + @ Disable virtual timer in case it was counting + mrc p15, 0, r7, c14, c3, 1 @ CNTV_CTL + bic r7, #1 @ Clear ENABLE + mcr p15, 0, r7, c14, c3, 1 @ CNTV_CTL 1: #endif -- cgit v1.2.2 From 01fafcab20fbbd2930691c7fdcf177eaa190d499 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 28 Feb 2012 11:50:32 +0000 Subject: ARM: nommu: add entry point for secondary CPUs to head-nommu.S MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds a secondary_startup entry point to head-nommu.S so that we can boot secondary CPUs on an SMP nommu configuration. Signed-off-by: Will Deacon CC: Uwe Kleine-König CC: Nicolas Pitre --- arch/arm/kernel/head-nommu.S | 50 ++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 46 insertions(+), 4 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index 8812ce88f7a1..06ba9c8e62be 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -63,12 +63,56 @@ ENTRY(stext) movs r10, r5 @ invalid processor (r5=0)? beq __error_p @ yes, error 'p' - adr lr, BSYM(__after_proc_init) @ return (PIC) address + ldr r13, =__mmap_switched @ address to jump to after + @ initialising sctlr + adr lr, BSYM(1f) @ return (PIC) address ARM( add pc, r10, #PROCINFO_INITFUNC ) THUMB( add r12, r10, #PROCINFO_INITFUNC ) THUMB( mov pc, r12 ) + 1: b __after_proc_init ENDPROC(stext) +#ifdef CONFIG_SMP + __CPUINIT +ENTRY(secondary_startup) + /* + * Common entry point for secondary CPUs. + * + * Ensure that we're in SVC mode, and IRQs are disabled. Lookup + * the processor type - there is no need to check the machine type + * as it has already been validated by the primary processor. + */ + setmode PSR_F_BIT | PSR_I_BIT | SVC_MODE, r9 +#ifndef CONFIG_CPU_CP15 + ldr r9, =CONFIG_PROCESSOR_ID +#else + mrc p15, 0, r9, c0, c0 @ get processor id +#endif + bl __lookup_processor_type @ r5=procinfo r9=cpuid + movs r10, r5 @ invalid processor? + beq __error_p @ yes, error 'p' + + adr r4, __secondary_data + ldmia r4, {r7, r12} + adr lr, BSYM(__after_proc_init) @ return address + mov r13, r12 @ __secondary_switched address + ARM( add pc, r10, #PROCINFO_INITFUNC ) + THUMB( add r12, r10, #PROCINFO_INITFUNC ) + THUMB( mov pc, r12 ) +ENDPROC(secondary_startup) + +ENTRY(__secondary_switched) + ldr sp, [r7, #8] @ set up the stack pointer + mov fp, #0 + b secondary_start_kernel +ENDPROC(__secondary_switched) + + .type __secondary_data, %object +__secondary_data: + .long secondary_data + .long __secondary_switched +#endif /* CONFIG_SMP */ + /* * Set the Control Register and Read the process ID. */ @@ -99,9 +143,7 @@ __after_proc_init: #endif mcr p15, 0, r0, c1, c0, 0 @ write control reg #endif /* CONFIG_CPU_CP15 */ - - b __mmap_switched @ clear the BSS and jump - @ to start_kernel + mov pc, r13 ENDPROC(__after_proc_init) .ltorg -- cgit v1.2.2 From c4a1f032ed35d744e3d74b8aebe8d85f29aecd88 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 28 Feb 2012 13:02:59 +0000 Subject: ARM: nommu: do not initialise page tables in secondary_data structure nommu systems do not require any page tables, so don't try to initialise them when bringing up secondary cores. Signed-off-by: Will Deacon --- arch/arm/kernel/smp.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 550d63cef68e..44d1c00dc45f 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -87,8 +87,10 @@ int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle) * its stack and the page tables. */ secondary_data.stack = task_stack_page(idle) + THREAD_START_SP; +#ifdef CONFIG_MMU secondary_data.pgdir = virt_to_phys(idmap_pgd); secondary_data.swapper_pg_dir = virt_to_phys(swapper_pg_dir); +#endif __cpuc_flush_dcache_area(&secondary_data, sizeof(secondary_data)); outer_clean_range(__pa(&secondary_data), __pa(&secondary_data + 1)); -- cgit v1.2.2 From aa1aadc3305c4917c39f0291613a5ec81dd4c73b Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 23 Feb 2012 13:51:38 +0000 Subject: ARM: suspend: fix CPU suspend code for !CONFIG_MMU configurations MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The ARM CPU suspend code can be selected even for a !CONFIG_MMU configuration. The resulting kernel will not compile and, even if it did, would access undefined co-processor registers when executing. This patch fixes the v6 and v7 CPU suspend code for the nommu case. Signed-off-by: Will Deacon Tested-by: Jonathan Austin CC: Lorenzo Pieralisi (commit_signer:1/3=33%) CC: Santosh Shilimkar (commit_signer:1/3=33%) CC: Uwe Kleine-König --- arch/arm/kernel/suspend.c | 64 ++++++++++++++++++++++++++--------------------- 1 file changed, 36 insertions(+), 28 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/suspend.c b/arch/arm/kernel/suspend.c index c59c97ea8268..38a50676213b 100644 --- a/arch/arm/kernel/suspend.c +++ b/arch/arm/kernel/suspend.c @@ -10,6 +10,42 @@ extern int __cpu_suspend(unsigned long, int (*)(unsigned long)); extern void cpu_resume_mmu(void); +#ifdef CONFIG_MMU +/* + * Hide the first two arguments to __cpu_suspend - these are an implementation + * detail which platform code shouldn't have to know about. + */ +int cpu_suspend(unsigned long arg, int (*fn)(unsigned long)) +{ + struct mm_struct *mm = current->active_mm; + int ret; + + if (!idmap_pgd) + return -EINVAL; + + /* + * Provide a temporary page table with an identity mapping for + * the MMU-enable code, required for resuming. On successful + * resume (indicated by a zero return code), we need to switch + * back to the correct page tables. + */ + ret = __cpu_suspend(arg, fn); + if (ret == 0) { + cpu_switch_mm(mm->pgd, mm); + local_flush_bp_all(); + local_flush_tlb_all(); + } + + return ret; +} +#else +int cpu_suspend(unsigned long arg, int (*fn)(unsigned long)) +{ + return __cpu_suspend(arg, fn); +} +#define idmap_pgd NULL +#endif + /* * This is called by __cpu_suspend() to save the state, and do whatever * flushing is required to ensure that when the CPU goes to sleep we have @@ -46,31 +82,3 @@ void __cpu_suspend_save(u32 *ptr, u32 ptrsz, u32 sp, u32 *save_ptr) outer_clean_range(virt_to_phys(save_ptr), virt_to_phys(save_ptr) + sizeof(*save_ptr)); } - -/* - * Hide the first two arguments to __cpu_suspend - these are an implementation - * detail which platform code shouldn't have to know about. - */ -int cpu_suspend(unsigned long arg, int (*fn)(unsigned long)) -{ - struct mm_struct *mm = current->active_mm; - int ret; - - if (!idmap_pgd) - return -EINVAL; - - /* - * Provide a temporary page table with an identity mapping for - * the MMU-enable code, required for resuming. On successful - * resume (indicated by a zero return code), we need to switch - * back to the correct page tables. - */ - ret = __cpu_suspend(arg, fn); - if (ret == 0) { - cpu_switch_mm(mm->pgd, mm); - local_flush_bp_all(); - local_flush_tlb_all(); - } - - return ret; -} -- cgit v1.2.2 From 8006b4d1a7c70f27a87cb753b5ed90483f0cfe26 Mon Sep 17 00:00:00 2001 From: Jonathan Austin Date: Wed, 24 Apr 2013 11:51:38 +0100 Subject: ARM: nommu: Don't build smp_tlb.c for !CONFIG_MMU Without an MMU we don't need to do any TLB maintenance. Until the addition of 93dc68876b60 (ARM: 7684/1: errata: Workaround for Cortex-A15 erratum 798181 (TLBI/DSB operations)) building the tlb maintenance ops in smp_tlb.c worked, though none of the contents were used. Since that commit, however, SMP NOMMU has not been able to build. This patch restores that ability by making the building of smp_tlb.c dependent on MMU. Signed-off-by: Jonathan Austin Acked-by: Catalin Marinas CC: Will Deacon --- arch/arm/kernel/Makefile | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index f4285b5ffb05..fccfbdb03df1 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -38,7 +38,10 @@ obj-$(CONFIG_ARTHUR) += arthur.o obj-$(CONFIG_ISA_DMA) += dma-isa.o obj-$(CONFIG_PCI) += bios32.o isa.o obj-$(CONFIG_ARM_CPU_SUSPEND) += sleep.o suspend.o -obj-$(CONFIG_SMP) += smp.o smp_tlb.o +obj-$(CONFIG_SMP) += smp.o +ifdef CONFIG_MMU +obj-$(CONFIG_SMP) += smp_tlb.o +endif obj-$(CONFIG_HAVE_ARM_SCU) += smp_scu.o obj-$(CONFIG_HAVE_ARM_TWD) += smp_twd.o obj-$(CONFIG_ARM_ARCH_TIMER) += arch_timer.o -- cgit v1.2.2 From 67c9845beab16a0c97b9c07f72a4b36b7175bb86 Mon Sep 17 00:00:00 2001 From: Jonathan Austin Date: Fri, 22 Feb 2013 17:48:56 +0000 Subject: ARM: mpu: add early bring-up code for the ARMv7 PMSA-compliant MPU This patch adds initial support for using the MPU, which is necessary for SMP operation on PMSAv7 processors because it is the only way to ensure memory is shared. This is an initial patch and full SMP support is added later in this series. The setup of the MPU is performed in a way analagous to that for the MMU: Very early initialisation before the C environment is brought up, followed by a sanity check and more complete initialisation in C. This patch provides the simplest possible memory region configuration: MPU_PROBE_REGION: Reserved for probing MPU details, not enabled MPU_BG_REGION: A 'background' region that specifies all memory strongly ordered MPU_RAM_REGION: A single shared, cacheable, normal region for the valid RAM. In this early initialisation code we simply map the whole of the address space with the BG_REGION and (at least) the kernel with the RAM_REGION. The MPU has region alignment constraints that require us to round past the end of the kernel. As region 2 has a higher priority than region 1, it overrides the strongly- ordered behaviour for RAM only. Subsequent patches will add more complete initialisation from the C-world and support for bringing up secondary CPUs. Signed-off-by: Jonathan Austin Reviewed-by: Will Deacon CC: Hyok S. Choi --- arch/arm/kernel/head-nommu.S | 87 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 87 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index 06ba9c8e62be..659912c49571 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -17,9 +17,11 @@ #include #include #include +#include #include #include #include +#include /* * Kernel startup entry point. @@ -63,6 +65,17 @@ ENTRY(stext) movs r10, r5 @ invalid processor (r5=0)? beq __error_p @ yes, error 'p' +#ifdef CONFIG_ARM_MPU + /* Calculate the size of a region covering just the kernel */ + ldr r5, =PHYS_OFFSET @ Region start: PHYS_OFFSET + ldr r6, =(_end) @ Cover whole kernel + sub r6, r6, r5 @ Minimum size of region to map + clz r6, r6 @ Region size must be 2^N... + rsb r6, r6, #31 @ ...so round up region size + lsl r6, r6, #MPU_RSR_SZ @ Put size in right field + orr r6, r6, #(1 << MPU_RSR_EN) @ Set region enabled bit + bl __setup_mpu +#endif ldr r13, =__mmap_switched @ address to jump to after @ initialising sctlr adr lr, BSYM(1f) @ return (PIC) address @@ -147,4 +160,78 @@ __after_proc_init: ENDPROC(__after_proc_init) .ltorg +#ifdef CONFIG_ARM_MPU + + +/* Set which MPU region should be programmed */ +.macro set_region_nr tmp, rgnr + mov \tmp, \rgnr @ Use static region numbers + mcr p15, 0, \tmp, c6, c2, 0 @ Write RGNR +.endm + +/* Setup a single MPU region, either D or I side (D-side for unified) */ +.macro setup_region bar, acr, sr, side = MPU_DATA_SIDE + mcr p15, 0, \bar, c6, c1, (0 + \side) @ I/DRBAR + mcr p15, 0, \acr, c6, c1, (4 + \side) @ I/DRACR + mcr p15, 0, \sr, c6, c1, (2 + \side) @ I/DRSR +.endm + +/* + * Setup the MPU and initial MPU Regions. We create the following regions: + * Region 0: Use this for probing the MPU details, so leave disabled. + * Region 1: Background region - covers the whole of RAM as strongly ordered + * Region 2: Normal, Shared, cacheable for RAM. From PHYS_OFFSET, size from r6 + * + * r6: Value to be written to DRSR (and IRSR if required) for MPU_RAM_REGION +*/ + +ENTRY(__setup_mpu) + + /* Probe for v7 PMSA compliance */ + mrc p15, 0, r0, c0, c1, 4 @ Read ID_MMFR0 + and r0, r0, #(MMFR0_PMSA) @ PMSA field + teq r0, #(MMFR0_PMSAv7) @ PMSA v7 + bne __error_p @ Fail: ARM_MPU on NOT v7 PMSA + + /* Determine whether the D/I-side memory map is unified. We set the + * flags here and continue to use them for the rest of this function */ + mrc p15, 0, r0, c0, c0, 4 @ MPUIR + ands r5, r0, #MPUIR_DREGION_SZMASK @ 0 size d region => No MPU + beq __error_p @ Fail: ARM_MPU and no MPU + tst r0, #MPUIR_nU @ MPUIR_nU = 0 for unified + + /* Setup second region first to free up r6 */ + set_region_nr r0, #MPU_RAM_REGION + isb + /* Full access from PL0, PL1, shared for CONFIG_SMP, cacheable */ + ldr r0, =PHYS_OFFSET @ RAM starts at PHYS_OFFSET + ldr r5,=(MPU_AP_PL1RW_PL0RW | MPU_RGN_NORMAL) + + setup_region r0, r5, r6, MPU_DATA_SIDE @ PHYS_OFFSET, shared, enabled + beq 1f @ Memory-map not unified + setup_region r0, r5, r6, MPU_INSTR_SIDE @ PHYS_OFFSET, shared, enabled +1: isb + + /* First/background region */ + set_region_nr r0, #MPU_BG_REGION + isb + /* Execute Never, strongly ordered, inaccessible to PL0, rw PL1 */ + mov r0, #0 @ BG region starts at 0x0 + ldr r5,=(MPU_ACR_XN | MPU_RGN_STRONGLY_ORDERED | MPU_AP_PL1RW_PL0NA) + mov r6, #MPU_RSR_ALL_MEM @ 4GB region, enabled + + setup_region r0, r5, r6, MPU_DATA_SIDE @ 0x0, BG region, enabled + beq 2f @ Memory-map not unified + setup_region r0, r5, r6, MPU_INSTR_SIDE @ 0x0, BG region, enabled +2: isb + + /* Enable the MPU */ + mrc p15, 0, r0, c1, c0, 0 @ Read SCTLR + bic r0, r0, #CR_BR @ Disable the 'default mem-map' + orr r0, r0, #CR_M @ Set SCTRL.M (MPU on) + mcr p15, 0, r0, c1, c0, 0 @ Enable MPU + isb + mov pc,lr +ENDPROC(__setup_mpu) +#endif #include "head-common.S" -- cgit v1.2.2 From eb08375ea66e63c5e11dea69b43c5633d531ce81 Mon Sep 17 00:00:00 2001 From: Jonathan Austin Date: Fri, 22 Feb 2013 18:51:30 +0000 Subject: ARM: mpu: add MPU initialisation for secondary cores The MPU initialisation on the primary core is performed in two stages, one minimal stage to ensure the CPU can boot and a second one after sanity_check_meminfo. As the memory configuration is known by the time we boot secondary cores only a single step is necessary, provided the values for DRSR are passed to secondaries. This patch implements this arrangement. The configuration generated for the MPU regions is made available to the secondary core, which can then use the asm MPU intialisation code to program a complete region configuration. This is necessary for SMP configurations without an MMU, as the MPU initialisation is the only way to ensure that memory is specified as 'shared'. Signed-off-by: Jonathan Austin Reviewed-by: Will Deacon CC: Nicolas Pitre --- arch/arm/kernel/head-nommu.S | 7 +++++++ arch/arm/kernel/smp.c | 8 ++++++-- 2 files changed, 13 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index 659912c49571..13741d004de5 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -107,6 +107,13 @@ ENTRY(secondary_startup) adr r4, __secondary_data ldmia r4, {r7, r12} + +#ifdef CONFIG_ARM_MPU + /* Use MPU region info supplied by __cpu_up */ + ldr r6, [r7] @ get secondary_data.mpu_szr + bl __setup_mpu @ Initialize the MPU +#endif + adr lr, BSYM(__after_proc_init) @ return address mov r13, r12 @ __secondary_switched address ARM( add pc, r10, #PROCINFO_INITFUNC ) diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 44d1c00dc45f..e17d9346baee 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -45,6 +45,7 @@ #include #include #include +#include /* * as from 2.5, kernels no longer have an init_tasks structure @@ -87,6 +88,10 @@ int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle) * its stack and the page tables. */ secondary_data.stack = task_stack_page(idle) + THREAD_START_SP; +#ifdef CONFIG_ARM_MPU + secondary_data.mpu_rgn_szr = mpu_rgn_info.rgns[MPU_RAM_REGION].drsr; +#endif + #ifdef CONFIG_MMU secondary_data.pgdir = virt_to_phys(idmap_pgd); secondary_data.swapper_pg_dir = virt_to_phys(swapper_pg_dir); @@ -114,9 +119,8 @@ int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle) pr_err("CPU%u: failed to boot: %d\n", cpu, ret); } - secondary_data.stack = NULL; - secondary_data.pgdir = 0; + memset(&secondary_data, 0, sizeof(secondary_data)); return ret; } -- cgit v1.2.2 From 5a9b5855c248d0298eac4d5490c7bc11c2f1983b Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Sat, 1 Jun 2013 23:39:38 -0700 Subject: ARM: sched_clock: Remove unused needs_suspend member The needs_suspend member is unused now that we always do the suspend/resume handling (see 6a4dae5 (ARM: 7565/1: sched: stop sched_clock() during suspend, 2012-10-23)). Acked-by: Will Deacon Signed-off-by: Stephen Boyd Signed-off-by: John Stultz --- arch/arm/kernel/sched_clock.c | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/sched_clock.c b/arch/arm/kernel/sched_clock.c index e8edcaa0e432..45efe86e8b08 100644 --- a/arch/arm/kernel/sched_clock.c +++ b/arch/arm/kernel/sched_clock.c @@ -24,7 +24,6 @@ struct clock_data { u32 mult; u32 shift; bool suspended; - bool needs_suspend; }; static void sched_clock_poll(unsigned long wrap_ticks); -- cgit v1.2.2 From ffbfb5e316f0db486798ccf1db36a577ffe79637 Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Sat, 1 Jun 2013 23:39:39 -0700 Subject: ARM: sched_clock: Return suspended count earlier If we're suspended and sched_clock() is called we're going to read the hardware one more time and throw away that value and return back the cached value we saved during the suspend callback. This is wasteful. Let's short circuit all that and return the cached value as early as possible if we're suspended. Acked-by: Will Deacon Signed-off-by: Stephen Boyd Signed-off-by: John Stultz --- arch/arm/kernel/sched_clock.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/sched_clock.c b/arch/arm/kernel/sched_clock.c index 45efe86e8b08..a781c59b93c0 100644 --- a/arch/arm/kernel/sched_clock.c +++ b/arch/arm/kernel/sched_clock.c @@ -55,9 +55,6 @@ static unsigned long long notrace cyc_to_sched_clock(u32 cyc, u32 mask) u64 epoch_ns; u32 epoch_cyc; - if (cd.suspended) - return cd.epoch_ns; - /* * Load the epoch_cyc and epoch_ns atomically. We do this by * ensuring that we always write epoch_cyc, epoch_ns and @@ -174,6 +171,9 @@ unsigned long long __read_mostly (*sched_clock_func)(void) = sched_clock_32; unsigned long long notrace sched_clock(void) { + if (cd.suspended) + return cd.epoch_ns; + return sched_clock_func(); } -- cgit v1.2.2 From 38ff87f77af0b5a93fc8581cff1d6e5692ab8970 Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Sat, 1 Jun 2013 23:39:40 -0700 Subject: sched_clock: Make ARM's sched_clock generic for all architectures Nothing about the sched_clock implementation in the ARM port is specific to the architecture. Generalize the code so that other architectures can use it by selecting GENERIC_SCHED_CLOCK. Signed-off-by: Stephen Boyd [jstultz: Merge minor collisions with other patches in my tree] Signed-off-by: John Stultz --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/arch_timer.c | 2 +- arch/arm/kernel/sched_clock.c | 216 ------------------------------------------ arch/arm/kernel/time.c | 4 +- 4 files changed, 3 insertions(+), 221 deletions(-) delete mode 100644 arch/arm/kernel/sched_clock.c (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index 5f3338eacad2..97cb0576d07c 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -16,7 +16,7 @@ CFLAGS_REMOVE_return_address.o = -pg # Object file lists. obj-y := elf.o entry-armv.o entry-common.o irq.o opcodes.o \ - process.o ptrace.o return_address.o sched_clock.o \ + process.o ptrace.o return_address.o \ setup.o signal.o stacktrace.o sys_arm.o time.o traps.o obj-$(CONFIG_ATAGS) += atags_parse.o diff --git a/arch/arm/kernel/arch_timer.c b/arch/arm/kernel/arch_timer.c index 59dcdced6e30..221f07b11ccb 100644 --- a/arch/arm/kernel/arch_timer.c +++ b/arch/arm/kernel/arch_timer.c @@ -11,9 +11,9 @@ #include #include #include +#include #include -#include #include diff --git a/arch/arm/kernel/sched_clock.c b/arch/arm/kernel/sched_clock.c deleted file mode 100644 index a781c59b93c0..000000000000 --- a/arch/arm/kernel/sched_clock.c +++ /dev/null @@ -1,216 +0,0 @@ -/* - * sched_clock.c: support for extending counters to full 64-bit ns counter - * - * 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 -#include -#include -#include -#include -#include -#include -#include - -#include - -struct clock_data { - u64 epoch_ns; - u32 epoch_cyc; - u32 epoch_cyc_copy; - unsigned long rate; - u32 mult; - u32 shift; - bool suspended; -}; - -static void sched_clock_poll(unsigned long wrap_ticks); -static DEFINE_TIMER(sched_clock_timer, sched_clock_poll, 0, 0); -static int irqtime = -1; - -core_param(irqtime, irqtime, int, 0400); - -static struct clock_data cd = { - .mult = NSEC_PER_SEC / HZ, -}; - -static u32 __read_mostly sched_clock_mask = 0xffffffff; - -static u32 notrace jiffy_sched_clock_read(void) -{ - return (u32)(jiffies - INITIAL_JIFFIES); -} - -static u32 __read_mostly (*read_sched_clock)(void) = jiffy_sched_clock_read; - -static inline u64 notrace cyc_to_ns(u64 cyc, u32 mult, u32 shift) -{ - return (cyc * mult) >> shift; -} - -static unsigned long long notrace cyc_to_sched_clock(u32 cyc, u32 mask) -{ - u64 epoch_ns; - u32 epoch_cyc; - - /* - * Load the epoch_cyc and epoch_ns atomically. We do this by - * ensuring that we always write epoch_cyc, epoch_ns and - * epoch_cyc_copy in strict order, and read them in strict order. - * If epoch_cyc and epoch_cyc_copy are not equal, then we're in - * the middle of an update, and we should repeat the load. - */ - do { - epoch_cyc = cd.epoch_cyc; - smp_rmb(); - epoch_ns = cd.epoch_ns; - smp_rmb(); - } while (epoch_cyc != cd.epoch_cyc_copy); - - return epoch_ns + cyc_to_ns((cyc - epoch_cyc) & mask, cd.mult, cd.shift); -} - -/* - * Atomically update the sched_clock epoch. - */ -static void notrace update_sched_clock(void) -{ - unsigned long flags; - u32 cyc; - u64 ns; - - cyc = read_sched_clock(); - ns = cd.epoch_ns + - cyc_to_ns((cyc - cd.epoch_cyc) & sched_clock_mask, - cd.mult, cd.shift); - /* - * Write epoch_cyc and epoch_ns in a way that the update is - * detectable in cyc_to_fixed_sched_clock(). - */ - raw_local_irq_save(flags); - cd.epoch_cyc_copy = cyc; - smp_wmb(); - cd.epoch_ns = ns; - smp_wmb(); - cd.epoch_cyc = cyc; - raw_local_irq_restore(flags); -} - -static void sched_clock_poll(unsigned long wrap_ticks) -{ - mod_timer(&sched_clock_timer, round_jiffies(jiffies + wrap_ticks)); - update_sched_clock(); -} - -void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate) -{ - unsigned long r, w; - u64 res, wrap; - char r_unit; - - if (cd.rate > rate) - return; - - BUG_ON(bits > 32); - WARN_ON(!irqs_disabled()); - read_sched_clock = read; - sched_clock_mask = (1 << bits) - 1; - cd.rate = rate; - - /* calculate the mult/shift to convert counter ticks to ns. */ - clocks_calc_mult_shift(&cd.mult, &cd.shift, rate, NSEC_PER_SEC, 0); - - r = rate; - if (r >= 4000000) { - r /= 1000000; - r_unit = 'M'; - } else if (r >= 1000) { - r /= 1000; - r_unit = 'k'; - } else - r_unit = ' '; - - /* calculate how many ns until we wrap */ - wrap = cyc_to_ns((1ULL << bits) - 1, cd.mult, cd.shift); - do_div(wrap, NSEC_PER_MSEC); - w = wrap; - - /* calculate the ns resolution of this counter */ - res = cyc_to_ns(1ULL, cd.mult, cd.shift); - pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lums\n", - bits, r, r_unit, res, w); - - /* - * Start the timer to keep sched_clock() properly updated and - * sets the initial epoch. - */ - sched_clock_timer.data = msecs_to_jiffies(w - (w / 10)); - update_sched_clock(); - - /* - * Ensure that sched_clock() starts off at 0ns - */ - cd.epoch_ns = 0; - - /* Enable IRQ time accounting if we have a fast enough sched_clock */ - if (irqtime > 0 || (irqtime == -1 && rate >= 1000000)) - enable_sched_clock_irqtime(); - - pr_debug("Registered %pF as sched_clock source\n", read); -} - -static unsigned long long notrace sched_clock_32(void) -{ - u32 cyc = read_sched_clock(); - return cyc_to_sched_clock(cyc, sched_clock_mask); -} - -unsigned long long __read_mostly (*sched_clock_func)(void) = sched_clock_32; - -unsigned long long notrace sched_clock(void) -{ - if (cd.suspended) - return cd.epoch_ns; - - return sched_clock_func(); -} - -void __init sched_clock_postinit(void) -{ - /* - * If no sched_clock function has been provided at that point, - * make it the final one one. - */ - if (read_sched_clock == jiffy_sched_clock_read) - setup_sched_clock(jiffy_sched_clock_read, 32, HZ); - - sched_clock_poll(sched_clock_timer.data); -} - -static int sched_clock_suspend(void) -{ - sched_clock_poll(sched_clock_timer.data); - cd.suspended = true; - return 0; -} - -static void sched_clock_resume(void) -{ - cd.epoch_cyc = read_sched_clock(); - cd.epoch_cyc_copy = cd.epoch_cyc; - cd.suspended = false; -} - -static struct syscore_ops sched_clock_ops = { - .suspend = sched_clock_suspend, - .resume = sched_clock_resume, -}; - -static int __init sched_clock_syscore_init(void) -{ - register_syscore_ops(&sched_clock_ops); - return 0; -} -device_initcall(sched_clock_syscore_init); diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c index abff4e9aaee0..98aee3258398 100644 --- a/arch/arm/kernel/time.c +++ b/arch/arm/kernel/time.c @@ -24,9 +24,9 @@ #include #include #include +#include #include -#include #include #include #include @@ -120,6 +120,4 @@ void __init time_init(void) machine_desc->init_time(); else clocksource_of_init(); - - sched_clock_postinit(); } -- cgit v1.2.2 From 1aa2b3b7a6c4f3dbd3671171113a20e6a6190e3b Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Wed, 5 Jun 2013 11:25:13 +0100 Subject: ARM: 7748/1: oabi: handle faults when loading swi instruction from userspace Running an OABI_COMPAT kernel on an SMP platform can lead to fun and games with page aging. If one CPU issues a swi instruction immediately before another CPU decides to mkold the page containing the swi instruction, then we will fault attempting to load the instruction during the vector_swi handler in order to retrieve its immediate field. Since this fault is not currently dealt with by our exception tables, this results in a panic: Unable to handle kernel paging request at virtual address 4020841c pgd = c490c000 [4020841c] *pgd=84451831, *pte=bf05859d, *ppte=00000000 Internal error: Oops: 17 [#1] PREEMPT SMP ARM Modules linked in: hid_sony(O) CPU: 1 Tainted: G W O (3.4.0-perf-gf496dca-01162-gcbcc62b #1) PC is at vector_swi+0x28/0x88 LR is at 0x40208420 This patch wraps all of the swi instruction loads with the USER macro and provides a shared exception table entry which simply rewinds the saved user PC and returns from the system call (without setting tbl, so there's no worries with tracing or syscall restarting). Returning to userspace will re-enter the page fault handler, from where we will probably send SIGSEGV to the current task. Reported-by: Wang, Yalin Reviewed-by: Nicolas Pitre Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/entry-common.S | 42 +++++++++++++++++++++++++++++------------- 1 file changed, 29 insertions(+), 13 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index bc5bc0a97131..4bc816a74a2e 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -362,6 +362,16 @@ ENTRY(vector_swi) str r0, [sp, #S_OLD_R0] @ Save OLD_R0 zero_fp +#ifdef CONFIG_ALIGNMENT_TRAP + ldr ip, __cr_alignment + ldr ip, [ip] + mcr p15, 0, ip, c1, c0 @ update control register +#endif + + enable_irq + ct_user_exit + get_thread_info tsk + /* * Get the system call number. */ @@ -375,9 +385,9 @@ ENTRY(vector_swi) #ifdef CONFIG_ARM_THUMB tst r8, #PSR_T_BIT movne r10, #0 @ no thumb OABI emulation - ldreq r10, [lr, #-4] @ get SWI instruction + USER( ldreq r10, [lr, #-4] ) @ get SWI instruction #else - ldr r10, [lr, #-4] @ get SWI instruction + USER( ldr r10, [lr, #-4] ) @ get SWI instruction #endif #ifdef CONFIG_CPU_ENDIAN_BE8 rev r10, r10 @ little endian instruction @@ -392,22 +402,13 @@ ENTRY(vector_swi) /* Legacy ABI only, possibly thumb mode. */ tst r8, #PSR_T_BIT @ this is SPSR from save_user_regs addne scno, r7, #__NR_SYSCALL_BASE @ put OS number in - ldreq scno, [lr, #-4] + USER( ldreq scno, [lr, #-4] ) #else /* Legacy ABI only. */ - ldr scno, [lr, #-4] @ get SWI instruction + USER( ldr scno, [lr, #-4] ) @ get SWI instruction #endif -#ifdef CONFIG_ALIGNMENT_TRAP - ldr ip, __cr_alignment - ldr ip, [ip] - mcr p15, 0, ip, c1, c0 @ update control register -#endif - enable_irq - ct_user_exit - - get_thread_info tsk adr tbl, sys_call_table @ load syscall table pointer #if defined(CONFIG_OABI_COMPAT) @@ -442,6 +443,21 @@ local_restart: eor r0, scno, #__NR_SYSCALL_BASE @ put OS number back bcs arm_syscall b sys_ni_syscall @ not private func + +#if defined(CONFIG_OABI_COMPAT) || !defined(CONFIG_AEABI) + /* + * We failed to handle a fault trying to access the page + * containing the swi instruction, but we're not really in a + * position to return -EFAULT. Instead, return back to the + * instruction and re-enter the user fault handling path trying + * to page it in. This will likely result in sending SEGV to the + * current task. + */ +9001: + sub lr, lr, #4 + str lr, [sp, #S_PC] + b ret_fast_syscall +#endif ENDPROC(vector_swi) /* -- cgit v1.2.2 From 9dfc28b6308096e48b54c28259825a1200f60742 Mon Sep 17 00:00:00 2001 From: Jonathan Austin Date: Thu, 18 Apr 2013 18:37:24 +0100 Subject: ARM: mpu: protect the vectors page with an MPU region Without an MMU it is possible for userspace programs to start executing code in places that they have no business executing. The MPU allows some level of protection against this. This patch protects the vectors page from access by userspace processes. Userspace tasks that dereference a null pointer are already protected by an svc at 0x0 that kills them. However when tasks use an offset from a null pointer (eg a function in a null struct) they miss this carefully placed svc and enter the exception vectors in user mode, ending up in the kernel. This patch causes programs that do this to receive a SEGV instead of happily entering the kernel in user-mode, and hence avoid a 'Bad Mode' panic. As part of this change it is necessary to make sigreturn happen via the stack when there is not an sa_restorer function. This change is invisible to userspace, and irrelevant to code compiled using a uClibc toolchain, which always uses an sa_restorer function. Because we don't get to remap the vectors in !MMU kuser_helpers are not in a defined location, and hence aren't usable. This means we don't need to worry about keeping them accessible from PL0 Signed-off-by: Jonathan Austin Reviewed-by: Will Deacon CC: Nicolas Pitre CC: Catalin Marinas --- arch/arm/kernel/head-nommu.S | 16 ++++++++++++++++ arch/arm/kernel/signal.c | 9 +++++++-- 2 files changed, 23 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index 13741d004de5..75f14cc3e073 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -22,6 +22,7 @@ #include #include #include +#include /* * Kernel startup entry point. @@ -188,6 +189,7 @@ ENDPROC(__after_proc_init) * Region 0: Use this for probing the MPU details, so leave disabled. * Region 1: Background region - covers the whole of RAM as strongly ordered * Region 2: Normal, Shared, cacheable for RAM. From PHYS_OFFSET, size from r6 + * Region 3: Normal, shared, inaccessible from PL0 to protect the vectors page * * r6: Value to be written to DRSR (and IRSR if required) for MPU_RAM_REGION */ @@ -232,6 +234,20 @@ ENTRY(__setup_mpu) setup_region r0, r5, r6, MPU_INSTR_SIDE @ 0x0, BG region, enabled 2: isb + /* Vectors region */ + set_region_nr r0, #MPU_VECTORS_REGION + isb + /* Shared, inaccessible to PL0, rw PL1 */ + mov r0, #CONFIG_VECTORS_BASE @ Cover from VECTORS_BASE + ldr r5,=(MPU_AP_PL1RW_PL0NA | MPU_RGN_NORMAL) + /* Writing N to bits 5:1 (RSR_SZ) --> region size 2^N+1 */ + mov r6, #(((PAGE_SHIFT - 1) << MPU_RSR_SZ) | 1 << MPU_RSR_EN) + + setup_region r0, r5, r6, MPU_DATA_SIDE @ VECTORS_BASE, PL0 NA, enabled + beq 3f @ Memory-map not unified + setup_region r0, r5, r6, MPU_INSTR_SIDE @ VECTORS_BASE, PL0 NA, enabled +3: isb + /* Enable the MPU */ mrc p15, 0, r0, c1, c0, 0 @ Read SCTLR bic r0, r0, #CR_BR @ Disable the 'default mem-map' diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 296786bdbb73..1c16c35c271a 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -392,14 +392,19 @@ setup_return(struct pt_regs *regs, struct ksignal *ksig, if (ksig->ka.sa.sa_flags & SA_SIGINFO) idx += 3; + /* + * Put the sigreturn code on the stack no matter which return + * mechanism we use in order to remain ABI compliant + */ if (__put_user(sigreturn_codes[idx], rc) || __put_user(sigreturn_codes[idx+1], rc+1)) return 1; - if (cpsr & MODE32_BIT) { + if ((cpsr & MODE32_BIT) && !IS_ENABLED(CONFIG_ARM_MPU)) { /* * 32-bit code can use the new high-page - * signal return code support. + * signal return code support except when the MPU has + * protected the vectors page from PL0 */ retcode = KERN_SIGRETURN_CODE + (idx << 2) + thumb; } else { -- cgit v1.2.2 From 8cf72172d739639f2699131821a3ebc291287cf2 Mon Sep 17 00:00:00 2001 From: Lorenzo Pieralisi Date: Thu, 16 May 2013 10:32:09 +0100 Subject: ARM: kernel: build MPIDR hash function data structure On ARM SMP systems, cores are identified by their MPIDR register. The MPIDR guidelines in the ARM ARM do not provide strict enforcement of MPIDR layout, only recommendations that, if followed, split the MPIDR on ARM 32 bit platforms in three affinity levels. In multi-cluster systems like big.LITTLE, if the affinity guidelines are followed, the MPIDR can not be considered an index anymore. This means that the association between logical CPU in the kernel and the HW CPU identifier becomes somewhat more complicated requiring methods like hashing to associate a given MPIDR to a CPU logical index, in order for the look-up to be carried out in an efficient and scalable way. This patch provides a function in the kernel that starting from the cpu_logical_map, implement collision-free hashing of MPIDR values by checking all significative bits of MPIDR affinity level bitfields. The hashing can then be carried out through bits shifting and ORing; the resulting hash algorithm is a collision-free though not minimal hash that can be executed with few assembly instructions. The mpidr is filtered through a mpidr mask that is built by checking all bits that toggle in the set of MPIDRs corresponding to possible CPUs. Bits that do not toggle do not carry information so they do not contribute to the resulting hash. Pseudo code: /* check all bits that toggle, so they are required */ for (i = 1, mpidr_mask = 0; i < num_possible_cpus(); i++) mpidr_mask |= (cpu_logical_map(i) ^ cpu_logical_map(0)); /* * Build shifts to be applied to aff0, aff1, aff2 values to hash the mpidr * fls() returns the last bit set in a word, 0 if none * ffs() returns the first bit set in a word, 0 if none */ fs0 = mpidr_mask[7:0] ? ffs(mpidr_mask[7:0]) - 1 : 0; fs1 = mpidr_mask[15:8] ? ffs(mpidr_mask[15:8]) - 1 : 0; fs2 = mpidr_mask[23:16] ? ffs(mpidr_mask[23:16]) - 1 : 0; ls0 = fls(mpidr_mask[7:0]); ls1 = fls(mpidr_mask[15:8]); ls2 = fls(mpidr_mask[23:16]); bits0 = ls0 - fs0; bits1 = ls1 - fs1; bits2 = ls2 - fs2; aff0_shift = fs0; aff1_shift = 8 + fs1 - bits0; aff2_shift = 16 + fs2 - (bits0 + bits1); u32 hash(u32 mpidr) { u32 l0, l1, l2; u32 mpidr_masked = mpidr & mpidr_mask; l0 = mpidr_masked & 0xff; l1 = mpidr_masked & 0xff00; l2 = mpidr_masked & 0xff0000; return (l0 >> aff0_shift | l1 >> aff1_shift | l2 >> aff2_shift); } The hashing algorithm relies on the inherent properties set in the ARM ARM recommendations for the MPIDR. Exotic configurations, where for instance the MPIDR values at a given affinity level have large holes, can end up requiring big hash tables since the compression of values that can be achieved through shifting is somewhat crippled when holes are present. Kernel warns if the number of buckets of the resulting hash table exceeds the number of possible CPUs by a factor of 4, which is a symptom of a very sparse HW MPIDR configuration. The hash algorithm is quite simple and can easily be implemented in assembly code, to be used in code paths where the kernel virtual address space is not set-up (ie cpu_resume) and instruction and data fetches are strongly ordered so code must be compact and must carry out few data accesses. Cc: Will Deacon Cc: Catalin Marinas Cc: Russell King Cc: Colin Cross Cc: Santosh Shilimkar Cc: Daniel Lezcano Cc: Amit Kucheria Signed-off-by: Lorenzo Pieralisi Reviewed-by: Dave Martin Reviewed-by: Nicolas Pitre Tested-by: Shawn Guo Tested-by: Kevin Hilman Tested-by: Stephen Warren --- arch/arm/kernel/setup.c | 67 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 67 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index ca34224f891f..9048513cbe0d 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -478,6 +478,72 @@ void __init smp_setup_processor_id(void) printk(KERN_INFO "Booting Linux on physical CPU 0x%x\n", mpidr); } +struct mpidr_hash mpidr_hash; +#ifdef CONFIG_SMP +/** + * smp_build_mpidr_hash - Pre-compute shifts required at each affinity + * level in order to build a linear index from an + * MPIDR value. Resulting algorithm is a collision + * free hash carried out through shifting and ORing + */ +static void __init smp_build_mpidr_hash(void) +{ + u32 i, affinity; + u32 fs[3], bits[3], ls, mask = 0; + /* + * Pre-scan the list of MPIDRS and filter out bits that do + * not contribute to affinity levels, ie they never toggle. + */ + for_each_possible_cpu(i) + mask |= (cpu_logical_map(i) ^ cpu_logical_map(0)); + pr_debug("mask of set bits 0x%x\n", mask); + /* + * Find and stash the last and first bit set at all affinity levels to + * check how many bits are required to represent them. + */ + for (i = 0; i < 3; i++) { + affinity = MPIDR_AFFINITY_LEVEL(mask, i); + /* + * Find the MSB bit and LSB bits position + * to determine how many bits are required + * to express the affinity level. + */ + ls = fls(affinity); + fs[i] = affinity ? ffs(affinity) - 1 : 0; + bits[i] = ls - fs[i]; + } + /* + * An index can be created from the MPIDR by isolating the + * significant bits at each affinity level and by shifting + * them in order to compress the 24 bits values space to a + * compressed set of values. This is equivalent to hashing + * the MPIDR through shifting and ORing. It is a collision free + * hash though not minimal since some levels might contain a number + * of CPUs that is not an exact power of 2 and their bit + * representation might contain holes, eg MPIDR[7:0] = {0x2, 0x80}. + */ + mpidr_hash.shift_aff[0] = fs[0]; + mpidr_hash.shift_aff[1] = MPIDR_LEVEL_BITS + fs[1] - bits[0]; + mpidr_hash.shift_aff[2] = 2*MPIDR_LEVEL_BITS + fs[2] - + (bits[1] + bits[0]); + mpidr_hash.mask = mask; + mpidr_hash.bits = bits[2] + bits[1] + bits[0]; + pr_debug("MPIDR hash: aff0[%u] aff1[%u] aff2[%u] mask[0x%x] bits[%u]\n", + mpidr_hash.shift_aff[0], + mpidr_hash.shift_aff[1], + mpidr_hash.shift_aff[2], + mpidr_hash.mask, + mpidr_hash.bits); + /* + * 4x is an arbitrary value used to warn on a hash table much bigger + * than expected on most systems. + */ + if (mpidr_hash_size() > 4 * num_possible_cpus()) + pr_warn("Large number of MPIDR hash buckets detected\n"); + sync_cache_w(&mpidr_hash); +} +#endif + static void __init setup_processor(void) { struct proc_info_list *list; @@ -825,6 +891,7 @@ void __init setup_arch(char **cmdline_p) smp_set_ops(mdesc->smp); } smp_init_cpus(); + smp_build_mpidr_hash(); } #endif -- cgit v1.2.2 From 7604537bbb5720376e8c9e6bc74a8e6305e3094d Mon Sep 17 00:00:00 2001 From: Lorenzo Pieralisi Date: Thu, 16 May 2013 10:34:30 +0100 Subject: ARM: kernel: implement stack pointer save array through MPIDR hashing Current implementation of cpu_{suspend}/cpu_{resume} relies on the MPIDR to index the array of pointers where the context is saved and restored. The current approach works as long as the MPIDR can be considered a linear index, so that the pointers array can simply be dereferenced by using the MPIDR[7:0] value. On ARM multi-cluster systems, where the MPIDR may not be a linear index, to properly dereference the stack pointer array, a mapping function should be applied to it so that it can be used for arrays look-ups. This patch adds code in the cpu_{suspend}/cpu_{resume} implementation that relies on shifting and ORing hashing method to map a MPIDR value to a set of buckets precomputed at boot to have a collision free mapping from MPIDR to context pointers. The hashing algorithm must be simple, fast, and implementable with few instructions since in the cpu_resume path the mapping is carried out with the MMU off and the I-cache off, hence code and data are fetched from DRAM with no-caching available. Simplicity is counterbalanced with a little increase of memory (allocated dynamically) for stack pointers buckets, that should be anyway fairly limited on most systems. Memory for context pointers is allocated in a early_initcall with size precomputed and stashed previously in kernel data structures. Memory for context pointers is allocated through kmalloc; this guarantees contiguous physical addresses for the allocated memory which is fundamental to the correct functioning of the resume mechanism that relies on the context pointer array to be a chunk of contiguous physical memory. Virtual to physical address conversion for the context pointer array base is carried out at boot to avoid fiddling with virt_to_phys conversions in the cpu_resume path which is quite fragile and should be optimized to execute as few instructions as possible. Virtual and physical context pointer base array addresses are stashed in a struct that is accessible from assembly using values generated through the asm-offsets.c mechanism. Cc: Will Deacon Cc: Catalin Marinas Cc: Russell King Cc: Colin Cross Cc: Santosh Shilimkar Cc: Daniel Lezcano Cc: Amit Kucheria Signed-off-by: Lorenzo Pieralisi Reviewed-by: Dave Martin Reviewed-by: Nicolas Pitre Tested-by: Shawn Guo Tested-by: Kevin Hilman Tested-by: Stephen Warren --- arch/arm/kernel/asm-offsets.c | 6 +++ arch/arm/kernel/sleep.S | 97 +++++++++++++++++++++++++++++++++++-------- arch/arm/kernel/suspend.c | 20 +++++++++ 3 files changed, 105 insertions(+), 18 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/asm-offsets.c b/arch/arm/kernel/asm-offsets.c index ee68cce6b48e..ded041711beb 100644 --- a/arch/arm/kernel/asm-offsets.c +++ b/arch/arm/kernel/asm-offsets.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -144,6 +145,11 @@ int main(void) #endif #ifdef MULTI_CACHE DEFINE(CACHE_FLUSH_KERN_ALL, offsetof(struct cpu_cache_fns, flush_kern_all)); +#endif +#ifdef CONFIG_ARM_CPU_SUSPEND + DEFINE(SLEEP_SAVE_SP_SZ, sizeof(struct sleep_save_sp)); + DEFINE(SLEEP_SAVE_SP_PHYS, offsetof(struct sleep_save_sp, save_ptr_stash_phys)); + DEFINE(SLEEP_SAVE_SP_VIRT, offsetof(struct sleep_save_sp, save_ptr_stash)); #endif BLANK(); DEFINE(DMA_BIDIRECTIONAL, DMA_BIDIRECTIONAL); diff --git a/arch/arm/kernel/sleep.S b/arch/arm/kernel/sleep.S index 987dcf33415c..db1536b8b30b 100644 --- a/arch/arm/kernel/sleep.S +++ b/arch/arm/kernel/sleep.S @@ -6,6 +6,49 @@ #include .text +/* + * Implementation of MPIDR hash algorithm through shifting + * and OR'ing. + * + * @dst: register containing hash result + * @rs0: register containing affinity level 0 bit shift + * @rs1: register containing affinity level 1 bit shift + * @rs2: register containing affinity level 2 bit shift + * @mpidr: register containing MPIDR value + * @mask: register containing MPIDR mask + * + * Pseudo C-code: + * + *u32 dst; + * + *compute_mpidr_hash(u32 rs0, u32 rs1, u32 rs2, u32 mpidr, u32 mask) { + * u32 aff0, aff1, aff2; + * u32 mpidr_masked = mpidr & mask; + * aff0 = mpidr_masked & 0xff; + * aff1 = mpidr_masked & 0xff00; + * aff2 = mpidr_masked & 0xff0000; + * dst = (aff0 >> rs0 | aff1 >> rs1 | aff2 >> rs2); + *} + * Input registers: rs0, rs1, rs2, mpidr, mask + * Output register: dst + * Note: input and output registers must be disjoint register sets + (eg: a macro instance with mpidr = r1 and dst = r1 is invalid) + */ + .macro compute_mpidr_hash dst, rs0, rs1, rs2, mpidr, mask + and \mpidr, \mpidr, \mask @ mask out MPIDR bits + and \dst, \mpidr, #0xff @ mask=aff0 + ARM( mov \dst, \dst, lsr \rs0 ) @ dst=aff0>>rs0 + THUMB( lsr \dst, \dst, \rs0 ) + and \mask, \mpidr, #0xff00 @ mask = aff1 + ARM( orr \dst, \dst, \mask, lsr \rs1 ) @ dst|=(aff1>>rs1) + THUMB( lsr \mask, \mask, \rs1 ) + THUMB( orr \dst, \dst, \mask ) + and \mask, \mpidr, #0xff0000 @ mask = aff2 + ARM( orr \dst, \dst, \mask, lsr \rs2 ) @ dst|=(aff2>>rs2) + THUMB( lsr \mask, \mask, \rs2 ) + THUMB( orr \dst, \dst, \mask ) + .endm + /* * Save CPU state for a suspend. This saves the CPU general purpose * registers, and allocates space on the kernel stack to save the CPU @@ -29,12 +72,18 @@ ENTRY(__cpu_suspend) mov r1, r4 @ size of save block mov r2, r5 @ virtual SP ldr r3, =sleep_save_sp -#ifdef CONFIG_SMP - ALT_SMP(mrc p15, 0, lr, c0, c0, 5) - ALT_UP(mov lr, #0) - and lr, lr, #15 + ldr r3, [r3, #SLEEP_SAVE_SP_VIRT] + ALT_SMP(mrc p15, 0, r9, c0, c0, 5) + ALT_UP_B(1f) + ldr r8, =mpidr_hash + /* + * This ldmia relies on the memory layout of the mpidr_hash + * struct mpidr_hash. + */ + ldmia r8, {r4-r7} @ r4 = mpidr mask (r5,r6,r7) = l[0,1,2] shifts + compute_mpidr_hash lr, r5, r6, r7, r9, r4 add r3, r3, lr, lsl #2 -#endif +1: bl __cpu_suspend_save adr lr, BSYM(cpu_suspend_abort) ldmfd sp!, {r0, pc} @ call suspend fn @@ -81,15 +130,23 @@ ENDPROC(cpu_resume_after_mmu) .data .align ENTRY(cpu_resume) -#ifdef CONFIG_SMP - adr r0, sleep_save_sp - ALT_SMP(mrc p15, 0, r1, c0, c0, 5) - ALT_UP(mov r1, #0) - and r1, r1, #15 - ldr r0, [r0, r1, lsl #2] @ stack phys addr -#else - ldr r0, sleep_save_sp @ stack phys addr -#endif + mov r1, #0 + ALT_SMP(mrc p15, 0, r0, c0, c0, 5) + ALT_UP_B(1f) + adr r2, mpidr_hash_ptr + ldr r3, [r2] + add r2, r2, r3 @ r2 = struct mpidr_hash phys address + /* + * This ldmia relies on the memory layout of the mpidr_hash + * struct mpidr_hash. + */ + ldmia r2, { r3-r6 } @ r3 = mpidr mask (r4,r5,r6) = l[0,1,2] shifts + compute_mpidr_hash r1, r4, r5, r6, r0, r3 +1: + adr r0, _sleep_save_sp + ldr r0, [r0, #SLEEP_SAVE_SP_PHYS] + ldr r0, [r0, r1, lsl #2] + setmode PSR_I_BIT | PSR_F_BIT | SVC_MODE, r1 @ set SVC, irqs off @ load phys pgd, stack, resume fn ARM( ldmia r0!, {r1, sp, pc} ) @@ -98,7 +155,11 @@ THUMB( mov sp, r2 ) THUMB( bx r3 ) ENDPROC(cpu_resume) -sleep_save_sp: - .rept CONFIG_NR_CPUS - .long 0 @ preserve stack phys ptr here - .endr + .align 2 +mpidr_hash_ptr: + .long mpidr_hash - . @ mpidr_hash struct offset + + .type sleep_save_sp, #object +ENTRY(sleep_save_sp) +_sleep_save_sp: + .space SLEEP_SAVE_SP_SZ @ struct sleep_save_sp diff --git a/arch/arm/kernel/suspend.c b/arch/arm/kernel/suspend.c index 38a50676213b..41cf3cbf756d 100644 --- a/arch/arm/kernel/suspend.c +++ b/arch/arm/kernel/suspend.c @@ -1,9 +1,12 @@ #include +#include +#include #include #include #include #include +#include #include #include @@ -82,3 +85,20 @@ void __cpu_suspend_save(u32 *ptr, u32 ptrsz, u32 sp, u32 *save_ptr) outer_clean_range(virt_to_phys(save_ptr), virt_to_phys(save_ptr) + sizeof(*save_ptr)); } + +extern struct sleep_save_sp sleep_save_sp; + +static int cpu_suspend_alloc_sp(void) +{ + void *ctx_ptr; + /* ctx_ptr is an array of physical addresses */ + ctx_ptr = kcalloc(mpidr_hash_size(), sizeof(u32), GFP_KERNEL); + + if (WARN_ON(!ctx_ptr)) + return -ENOMEM; + sleep_save_sp.save_ptr_stash = ctx_ptr; + sleep_save_sp.save_ptr_stash_phys = virt_to_phys(ctx_ptr); + sync_cache_w(&sleep_save_sp); + return 0; +} +early_initcall(cpu_suspend_alloc_sp); -- cgit v1.2.2 From a4780adeefd042482f624f5e0d577bf9cdcbb760 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Andr=C3=A9=20Hentschel?= Date: Tue, 18 Jun 2013 23:23:26 +0100 Subject: ARM: 7735/2: Preserve the user r/w register TPIDRURW on context switch and fork MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit 6a1c53124aa1 the user writeable TLS register was zeroed to prevent it from being used as a covert channel between two tasks. There are more and more applications coming to Windows RT, Wine could support them, but mostly they expect to have the thread environment block (TEB) in TPIDRURW. This patch preserves that register per thread instead of clearing it. Unlike the TPIDRURO, which is already switched, the TPIDRURW can be updated from userspace so needs careful treatment in the case that we modify TPIDRURW and call fork(). To avoid this we must always read TPIDRURW in copy_thread. Signed-off-by: André Hentschel Signed-off-by: Will Deacon Signed-off-by: Jonathan Austin Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 5 +++-- arch/arm/kernel/process.c | 4 +++- arch/arm/kernel/ptrace.c | 2 +- arch/arm/kernel/traps.c | 4 ++-- 4 files changed, 9 insertions(+), 6 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 582b405befc5..a39cfc2a1f90 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -685,15 +685,16 @@ 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 ) + ldr r4, [r2, #TI_TP_VALUE] + ldr r5, [r2, #TI_TP_VALUE + 4] #ifdef CONFIG_CPU_USE_DOMAINS ldr r6, [r2, #TI_CPU_DOMAIN] #endif - set_tls r3, r4, r5 + switch_tls r1, r4, r5, r3, r7 #if defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_SMP) ldr r7, [r2, #TI_TASK] ldr r8, =__stack_chk_guard diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index f21970316836..087064148ebf 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -39,6 +39,7 @@ #include #include #include +#include #ifdef CONFIG_CC_STACKPROTECTOR #include @@ -343,7 +344,8 @@ copy_thread(unsigned long clone_flags, unsigned long stack_start, clear_ptrace_hw_breakpoint(p); if (clone_flags & CLONE_SETTLS) - thread->tp_value = childregs->ARM_r3; + thread->tp_value[0] = childregs->ARM_r3; + thread->tp_value[1] = get_tpuser(); thread_notify(THREAD_NOTIFY_COPY, thread); diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index 03deeffd9f6d..2bc1514d6dbe 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -849,7 +849,7 @@ long arch_ptrace(struct task_struct *child, long request, #endif case PTRACE_GET_THREAD_AREA: - ret = put_user(task_thread_info(child)->tp_value, + ret = put_user(task_thread_info(child)->tp_value[0], datap); break; diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 18b32e8e4497..517bfd4da1c9 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -581,7 +581,7 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs) return regs->ARM_r0; case NR(set_tls): - thread->tp_value = regs->ARM_r0; + thread->tp_value[0] = regs->ARM_r0; if (tls_emu) return 0; if (has_tls_reg) { @@ -699,7 +699,7 @@ static int get_tp_trap(struct pt_regs *regs, unsigned int instr) int reg = (instr >> 12) & 15; if (reg == 15) return 1; - regs->uregs[reg] = current_thread_info()->tp_value; + regs->uregs[reg] = current_thread_info()->tp_value[0]; regs->ARM_pc += 4; return 0; } -- cgit v1.2.2 From c5f927a6f62196226915f12194c9d0df4e2210d7 Mon Sep 17 00:00:00 2001 From: Jed Davis Date: Thu, 20 Jun 2013 10:16:29 +0100 Subject: ARM: 7765/1: perf: Record the user-mode PC in the call chain. With this change, we no longer lose the innermost entry in the user-mode part of the call chain. See also the x86 port, which includes the ip. It's possible to partially work around this problem by post-processing the data to use the PERF_SAMPLE_IP value, but this works only if the CPU wasn't in the kernel when the sample was taken. Cc: Signed-off-by: Jed Davis Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/perf_event.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 8c3094d0f7b7..d9f5cd4e533f 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -569,6 +569,7 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) return; } + perf_callchain_store(entry, regs->ARM_pc); tail = (struct frame_tail __user *)regs->ARM_fp - 1; while ((entry->nr < PERF_MAX_STACK_DEPTH) && -- cgit v1.2.2 From 0d0752bca1f9a91fb646647aa4abbb21156f316c Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Fri, 21 Jun 2013 12:07:27 +0100 Subject: ARM: 7769/1: Cortex-A15: fix erratum 798181 implementation Looking into the active_asids array is not enough, as we also need to look into the reserved_asids array (they both represent processes that are currently running). Also, not holding the ASID allocator lock is racy, as another CPU could schedule that process and trigger a rollover, making the erratum workaround miss an IPI. Exposing this outside of context.c is a little ugly on the side, so let's define a new entry point that the erratum workaround can call to obtain the cpumask. Cc: # 3.9 Acked-by: Will Deacon Acked-by: Catalin Marinas Signed-off-by: Marc Zyngier Signed-off-by: Russell King --- arch/arm/kernel/smp_tlb.c | 18 ++---------------- 1 file changed, 2 insertions(+), 16 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp_tlb.c b/arch/arm/kernel/smp_tlb.c index 9a52a07aa40e..a98b62dca2fa 100644 --- a/arch/arm/kernel/smp_tlb.c +++ b/arch/arm/kernel/smp_tlb.c @@ -103,7 +103,7 @@ static void broadcast_tlb_a15_erratum(void) static void broadcast_tlb_mm_a15_erratum(struct mm_struct *mm) { - int cpu, this_cpu; + int this_cpu; cpumask_t mask = { CPU_BITS_NONE }; if (!erratum_a15_798181()) @@ -111,21 +111,7 @@ static void broadcast_tlb_mm_a15_erratum(struct mm_struct *mm) dummy_flush_tlb_a15_erratum(); this_cpu = get_cpu(); - for_each_online_cpu(cpu) { - if (cpu == this_cpu) - continue; - /* - * We only need to send an IPI if the other CPUs are running - * the same ASID as the one being invalidated. There is no - * need for locking around the active_asids check since the - * switch_mm() function has at least one dmb() (as required by - * this workaround) in case a context switch happens on - * another CPU after the condition below. - */ - if (atomic64_read(&mm->context.id) == - atomic64_read(&per_cpu(active_asids, cpu))) - cpumask_set_cpu(cpu, &mask); - } + a15_erratum_get_cpumask(this_cpu, mm, &mask); smp_call_function_many(&mask, ipi_flush_tlb_a15_erratum, NULL, 1); put_cpu(); } -- cgit v1.2.2 From 8c69d7af12265e5be9326110b695e010b47505bd Mon Sep 17 00:00:00 2001 From: Stephen Warren Date: Fri, 5 Jul 2013 22:59:42 +0100 Subject: ARM: 7780/1: add missing linker section markup to head-common.S Macro __INIT is used to place various code in head-common.S into the init section. This should be matched by a closing __FINIT. Also, add an explicit ".text" to ensure subsequent code is placed into the correct section; __FINIT is simply a closing marker to match __INIT and doesn't guarantee to revert to .text. This historically caused no problem, because macro __CPUINIT was used at the exact location where __FINIT was missing, which then placed following code into the cpuinit section. However, with commit 22f0a2736 "init.h: remove __cpuinit sections from the kernel" applied, __CPUINIT becomes a no-op, thus leaving all this code in the init section, rather than the regular text section. This caused issues such as secondary CPU boot failures or crashes. Signed-off-by: Stephen Warren Acked-by: Paul Gortmaker Signed-off-by: Russell King --- arch/arm/kernel/head-common.S | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-common.S b/arch/arm/kernel/head-common.S index 5b391a689b47..76ab5ca50610 100644 --- a/arch/arm/kernel/head-common.S +++ b/arch/arm/kernel/head-common.S @@ -133,6 +133,9 @@ ENTRY(lookup_processor_type) ldmfd sp!, {r4 - r6, r9, pc} ENDPROC(lookup_processor_type) + __FINIT + .text + /* * Read processor ID register (CP#15, CR0), and look up in the linker-built * supported processor list. Note that we can't use the absolute addresses -- cgit v1.2.2 From 6af9df7f5ba35806a5919d3a36d95fd40e210b89 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 8 Jul 2013 16:00:51 -0700 Subject: ptrace/arm: revert "hw_breakpoints: Fix racy access to ptrace breakpoints" This reverts commit bf0b8f4b55e5 ("hw_breakpoints: Fix racy access to ptrace breakpoints"). The patch was fine but we can no longer race with SIGKILL after commit 9899d11f6544 ("ptrace: ensure arch_ptrace/ptrace_request can never race with SIGKILL"), the __TASK_TRACED tracee can't be woken up and ->ptrace_bps[] can't go away. Signed-off-by: Oleg Nesterov Acked-by: Will Deacon Cc: Benjamin Herrenschmidt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jan Kratochvil Cc: Michael Neuling Cc: Paul Mackerras Cc: Paul Mundt Cc: Prasad Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/ptrace.c | 8 -------- 1 file changed, 8 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index 2bc1514d6dbe..0dd3b79b15c3 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -886,20 +886,12 @@ long arch_ptrace(struct task_struct *child, long request, #ifdef CONFIG_HAVE_HW_BREAKPOINT case PTRACE_GETHBPREGS: - if (ptrace_get_breakpoints(child) < 0) - return -ESRCH; - ret = ptrace_gethbpregs(child, addr, (unsigned long __user *)data); - ptrace_put_breakpoints(child); break; case PTRACE_SETHBPREGS: - if (ptrace_get_breakpoints(child) < 0) - return -ESRCH; - ret = ptrace_sethbpregs(child, addr, (unsigned long __user *)data); - ptrace_put_breakpoints(child); break; #endif -- cgit v1.2.2 From 16d6d5b00ee75307bab7e4ede9452c97b28f30e2 Mon Sep 17 00:00:00 2001 From: Robin Holt Date: Mon, 8 Jul 2013 16:01:39 -0700 Subject: reboot: arm: prepare reboot_mode for moving to generic kernel code Prepare for the moving the parsing of reboot= to the generic kernel code by making reboot_mode into a more generic form. Signed-off-by: Robin Holt Cc: Russell King Cc: Russ Anderson Cc: Robin Holt Cc: H. Peter Anvin Cc: Guan Xuetao Acked-by: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/process.c | 8 ++++---- arch/arm/kernel/setup.c | 6 +++--- 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 7f1efcd4a6e9..2d544062fd7d 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -175,14 +175,14 @@ void arch_cpu_idle(void) default_idle(); } -static char reboot_mode = 'h'; +enum reboot_mode reboot_mode = REBOOT_HARD; -int __init reboot_setup(char *str) +static int __init reboot_setup(char *str) { - reboot_mode = str[0]; + if ('s' == str[0]) + reboot_mode = REBOOT_SOFT; return 1; } - __setup("reboot=", reboot_setup); /* diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 9b653278c9e8..63af9a7ae512 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -74,7 +74,7 @@ __setup("fpe=", fpe_setup); extern void paging_init(struct machine_desc *desc); extern void sanity_check_meminfo(void); -extern void reboot_setup(char *str); +extern enum reboot_mode reboot_mode; extern void setup_dma_zone(struct machine_desc *desc); unsigned int processor_id; @@ -861,8 +861,8 @@ void __init setup_arch(char **cmdline_p) setup_dma_zone(mdesc); - if (mdesc->restart_mode) - reboot_setup(&mdesc->restart_mode); + if (mdesc->reboot_mode != REBOOT_HARD) + reboot_mode = mdesc->reboot_mode; init_mm.start_code = (unsigned long) _text; init_mm.end_code = (unsigned long) _etext; -- cgit v1.2.2 From 7b6d864b48d95e6ea1df7df64475b9cb9616dcf9 Mon Sep 17 00:00:00 2001 From: Robin Holt Date: Mon, 8 Jul 2013 16:01:40 -0700 Subject: reboot: arm: change reboot_mode to use enum reboot_mode Preparing to move the parsing of reboot= to generic kernel code forces the change in reboot_mode handling to use the enum. [akpm@linux-foundation.org: fix arch/arm/mach-socfpga/socfpga.c] Signed-off-by: Robin Holt Cc: Russell King Cc: Russ Anderson Cc: Robin Holt Cc: H. Peter Anvin Cc: Guan Xuetao Acked-by: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/process.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 2d544062fd7d..b7fdd864c839 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include @@ -113,7 +114,7 @@ void soft_restart(unsigned long addr) BUG(); } -static void null_restart(char mode, const char *cmd) +static void null_restart(enum reboot_mode reboot_mode, const char *cmd) { } @@ -123,7 +124,7 @@ static void null_restart(char mode, const char *cmd) void (*pm_power_off)(void); EXPORT_SYMBOL(pm_power_off); -void (*arm_pm_restart)(char str, const char *cmd) = null_restart; +void (*arm_pm_restart)(enum reboot_mode reboot_mode, const char *cmd) = null_restart; EXPORT_SYMBOL_GPL(arm_pm_restart); /* -- cgit v1.2.2 From 1b3a5d02ee070c8f9943333b9b6370f486601e0f Mon Sep 17 00:00:00 2001 From: Robin Holt Date: Mon, 8 Jul 2013 16:01:42 -0700 Subject: reboot: move arch/x86 reboot= handling to generic kernel Merge together the unicore32, arm, and x86 reboot= command line parameter handling. Signed-off-by: Robin Holt Cc: H. Peter Anvin Cc: Russell King Cc: Guan Xuetao Cc: Russ Anderson Cc: Robin Holt Acked-by: Ingo Molnar Acked-by: Guan Xuetao Acked-by: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/process.c | 10 ---------- 1 file changed, 10 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index b7fdd864c839..d3ca4f6915af 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -176,16 +176,6 @@ void arch_cpu_idle(void) default_idle(); } -enum reboot_mode reboot_mode = REBOOT_HARD; - -static int __init reboot_setup(char *str) -{ - if ('s' == str[0]) - reboot_mode = REBOOT_SOFT; - return 1; -} -__setup("reboot=", reboot_setup); - /* * Called by kexec, immediately prior to machine_kexec(). * -- cgit v1.2.2 From cbbe6f82b489e7ceba4ad7c833bd3a76cd0084cb Mon Sep 17 00:00:00 2001 From: Jason Liu Date: Mon, 1 Jul 2013 09:53:30 +0100 Subject: ARM: 7778/1: smp_twd: twd_update_frequency need be run on all online CPUs When the local timer freq changed, the twd_update_frequency function should be run all the CPUs include itself, otherwise, the twd freq will not get updated and the local timer will not run correcttly. smp_call_function will run functions on all other CPUs, but not include himself, this is not correct,use on_each_cpu instead to fix this issue. Cc: Linus Walleij Cc: Rob Herring Cc: Shawn Guo Cc: Arnd Bergmann Cc: stable@vger.kernel.org Acked-by: Linus Walleij Acked-by: Shawn Guo Signed-off-by: Jason Liu Signed-off-by: Russell King --- arch/arm/kernel/smp_twd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp_twd.c b/arch/arm/kernel/smp_twd.c index 90525d9d290b..f6fd1d4398c6 100644 --- a/arch/arm/kernel/smp_twd.c +++ b/arch/arm/kernel/smp_twd.c @@ -120,7 +120,7 @@ static int twd_rate_change(struct notifier_block *nb, * changing cpu. */ if (flags == POST_RATE_CHANGE) - smp_call_function(twd_update_frequency, + on_each_cpu(twd_update_frequency, (void *)&cnd->new_rate, 1); return NOTIFY_OK; -- cgit v1.2.2 From 8bd26e3a7e49af2697449bbcb7187a39dc85d672 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 17 Jun 2013 15:43:14 -0400 Subject: arm: delete __cpuinit/__CPUINIT usage from all ARM users The __cpuinit type of throwaway sections might have made sense some time ago when RAM was more constrained, but now the savings do not offset the cost and complications. For example, the fix in commit 5e427ec2d0 ("x86: Fix bit corruption at CPU resume time") is a good example of the nasty type of bugs that can be created with improper use of the various __init prefixes. After a discussion on LKML[1] it was decided that cpuinit should go the way of devinit and be phased out. Once all the users are gone, we can then finally remove the macros themselves from linux/init.h. Note that some harmless section mismatch warnings may result, since notify_cpu_starting() and cpu_up() are arch independent (kernel/cpu.c) and are flagged as __cpuinit -- so if we remove the __cpuinit from the arch specific callers, we will also get section mismatch warnings. As an intermediate step, we intend to turn the linux/init.h cpuinit related content into no-ops as early as possible, since that will get rid of these warnings. In any case, they are temporary and harmless. This removes all the ARM uses of the __cpuinit macros from C code, and all __CPUINIT from assembly code. It also had two ".previous" section statements that were paired off against __CPUINIT (aka .section ".cpuinit.text") that also get removed here. [1] https://lkml.org/lkml/2013/5/20/589 Cc: Russell King Cc: Will Deacon Cc: linux-arm-kernel@lists.infradead.org Signed-off-by: Paul Gortmaker --- arch/arm/kernel/head-common.S | 1 - arch/arm/kernel/head-nommu.S | 1 - arch/arm/kernel/head.S | 1 - arch/arm/kernel/hw_breakpoint.c | 4 ++-- arch/arm/kernel/perf_event_cpu.c | 6 +++--- arch/arm/kernel/psci_smp.c | 3 +-- arch/arm/kernel/smp.c | 18 +++++++++--------- arch/arm/kernel/smp_twd.c | 6 +++--- 8 files changed, 18 insertions(+), 22 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-common.S b/arch/arm/kernel/head-common.S index 76ab5ca50610..47cd974e57ea 100644 --- a/arch/arm/kernel/head-common.S +++ b/arch/arm/kernel/head-common.S @@ -149,7 +149,6 @@ ENDPROC(lookup_processor_type) * r5 = proc_info pointer in physical address space * r9 = cpuid (preserved) */ - __CPUINIT __lookup_processor_type: adr r3, __lookup_processor_type_data ldmia r3, {r4 - r6} diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index 75f14cc3e073..b361de143756 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -87,7 +87,6 @@ ENTRY(stext) ENDPROC(stext) #ifdef CONFIG_SMP - __CPUINIT ENTRY(secondary_startup) /* * Common entry point for secondary CPUs. diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 45e8935cae4e..9cf6063020ae 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -343,7 +343,6 @@ __turn_mmu_on_loc: .long __turn_mmu_on_end #if defined(CONFIG_SMP) - __CPUINIT ENTRY(secondary_startup) /* * Common entry point for secondary CPUs. diff --git a/arch/arm/kernel/hw_breakpoint.c b/arch/arm/kernel/hw_breakpoint.c index 1fd749ee4a1b..7b95de601357 100644 --- a/arch/arm/kernel/hw_breakpoint.c +++ b/arch/arm/kernel/hw_breakpoint.c @@ -1020,7 +1020,7 @@ out_mdbgen: cpumask_or(&debug_err_mask, &debug_err_mask, cpumask_of(cpu)); } -static int __cpuinit dbg_reset_notify(struct notifier_block *self, +static int dbg_reset_notify(struct notifier_block *self, unsigned long action, void *cpu) { if ((action & ~CPU_TASKS_FROZEN) == CPU_ONLINE) @@ -1029,7 +1029,7 @@ static int __cpuinit dbg_reset_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __cpuinitdata dbg_reset_nb = { +static struct notifier_block dbg_reset_nb = { .notifier_call = dbg_reset_notify, }; diff --git a/arch/arm/kernel/perf_event_cpu.c b/arch/arm/kernel/perf_event_cpu.c index 1f2740e3dbc0..aebe0e99c153 100644 --- a/arch/arm/kernel/perf_event_cpu.c +++ b/arch/arm/kernel/perf_event_cpu.c @@ -157,8 +157,8 @@ static void cpu_pmu_init(struct arm_pmu *cpu_pmu) * UNKNOWN at reset, the PMU must be explicitly reset to avoid reading * junk values out of them. */ -static int __cpuinit cpu_pmu_notify(struct notifier_block *b, - unsigned long action, void *hcpu) +static int cpu_pmu_notify(struct notifier_block *b, unsigned long action, + void *hcpu) { if ((action & ~CPU_TASKS_FROZEN) != CPU_STARTING) return NOTIFY_DONE; @@ -171,7 +171,7 @@ static int __cpuinit cpu_pmu_notify(struct notifier_block *b, return NOTIFY_OK; } -static struct notifier_block __cpuinitdata cpu_pmu_hotplug_notifier = { +static struct notifier_block cpu_pmu_hotplug_notifier = { .notifier_call = cpu_pmu_notify, }; diff --git a/arch/arm/kernel/psci_smp.c b/arch/arm/kernel/psci_smp.c index 219f1d73572a..70ded3fb42d9 100644 --- a/arch/arm/kernel/psci_smp.c +++ b/arch/arm/kernel/psci_smp.c @@ -46,8 +46,7 @@ extern void secondary_startup(void); -static int __cpuinit psci_boot_secondary(unsigned int cpu, - struct task_struct *idle) +static int psci_boot_secondary(unsigned int cpu, struct task_struct *idle) { if (psci_ops.cpu_on) return psci_ops.cpu_on(cpu_logical_map(cpu), diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index c5fb5469054b..c2b4f8f0be9a 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -58,7 +58,7 @@ struct secondary_data secondary_data; * control for which core is the next to come out of the secondary * boot "holding pen" */ -volatile int __cpuinitdata pen_release = -1; +volatile int pen_release = -1; enum ipi_msg_type { IPI_WAKEUP, @@ -86,7 +86,7 @@ static unsigned long get_arch_pgd(pgd_t *pgd) return pgdir >> ARCH_PGD_SHIFT; } -int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle) +int __cpu_up(unsigned int cpu, struct task_struct *idle) { int ret; @@ -138,7 +138,7 @@ void __init smp_init_cpus(void) smp_ops.smp_init_cpus(); } -int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle) +int boot_secondary(unsigned int cpu, struct task_struct *idle) { if (smp_ops.smp_boot_secondary) return smp_ops.smp_boot_secondary(cpu, idle); @@ -170,7 +170,7 @@ static int platform_cpu_disable(unsigned int cpu) /* * __cpu_disable runs on the processor to be shutdown. */ -int __cpuinit __cpu_disable(void) +int __cpu_disable(void) { unsigned int cpu = smp_processor_id(); int ret; @@ -216,7 +216,7 @@ static DECLARE_COMPLETION(cpu_died); * called on the thread which is asking for a CPU to be shutdown - * waits until shutdown has completed, or it is timed out. */ -void __cpuinit __cpu_die(unsigned int cpu) +void __cpu_die(unsigned int cpu) { if (!wait_for_completion_timeout(&cpu_died, msecs_to_jiffies(5000))) { pr_err("CPU%u: cpu didn't die\n", cpu); @@ -306,7 +306,7 @@ void __ref cpu_die(void) * Called by both boot and secondaries to move global data into * per-processor storage. */ -static void __cpuinit smp_store_cpu_info(unsigned int cpuid) +static void smp_store_cpu_info(unsigned int cpuid) { struct cpuinfo_arm *cpu_info = &per_cpu(cpu_data, cpuid); @@ -322,7 +322,7 @@ static void percpu_timer_setup(void); * This is the secondary CPU boot entry. We're using this CPUs * idle thread stack, but a set of temporary page tables. */ -asmlinkage void __cpuinit secondary_start_kernel(void) +asmlinkage void secondary_start_kernel(void) { struct mm_struct *mm = &init_mm; unsigned int cpu; @@ -521,7 +521,7 @@ static void broadcast_timer_set_mode(enum clock_event_mode mode, { } -static void __cpuinit broadcast_timer_setup(struct clock_event_device *evt) +static void broadcast_timer_setup(struct clock_event_device *evt) { evt->name = "dummy_timer"; evt->features = CLOCK_EVT_FEAT_ONESHOT | @@ -550,7 +550,7 @@ int local_timer_register(struct local_timer_ops *ops) } #endif -static void __cpuinit percpu_timer_setup(void) +static void percpu_timer_setup(void) { unsigned int cpu = smp_processor_id(); struct clock_event_device *evt = &per_cpu(percpu_clockevent, cpu); diff --git a/arch/arm/kernel/smp_twd.c b/arch/arm/kernel/smp_twd.c index f6fd1d4398c6..25956204ef23 100644 --- a/arch/arm/kernel/smp_twd.c +++ b/arch/arm/kernel/smp_twd.c @@ -187,7 +187,7 @@ core_initcall(twd_cpufreq_init); #endif -static void __cpuinit twd_calibrate_rate(void) +static void twd_calibrate_rate(void) { unsigned long count; u64 waitjiffies; @@ -265,7 +265,7 @@ static void twd_get_clock(struct device_node *np) /* * Setup the local clock events for a CPU. */ -static int __cpuinit twd_timer_setup(struct clock_event_device *clk) +static int twd_timer_setup(struct clock_event_device *clk) { struct clock_event_device **this_cpu_clk; int cpu = smp_processor_id(); @@ -308,7 +308,7 @@ static int __cpuinit twd_timer_setup(struct clock_event_device *clk) return 0; } -static struct local_timer_ops twd_lt_ops __cpuinitdata = { +static struct local_timer_ops twd_lt_ops = { .setup = twd_timer_setup, .stop = twd_timer_stop, }; -- cgit v1.2.2 From b60d5db690f72a7cfe5e39f191f0719a9ef3a096 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Thu, 18 Jul 2013 17:20:32 +0100 Subject: ARM: 7786/1: hyp: fix macro parameterisation Currently, compare_cpu_mode_with_primary uses a mixture of macro arguments and hardcoded registers, and does so incorrectly, as it stores (__boot_cpu_mode_offset | BOOT_CPU_MODE_MISMATCH) to (__boot_cpu_mode + &__boot_cpu_mode_offset), which could corrupt an arbitrary portion of memory. This patch fixes up compare_cpu_mode_with_primary to use the macro arguments, correctly updating __boot_cpu_mode. Signed-off-by: Mark Rutland Acked-by: Dave Martin Acked-by: Marc Zyngier Cc: Christoffer Dall Signed-off-by: Russell King --- arch/arm/kernel/hyp-stub.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S index 4910232c4833..797b1a6a4906 100644 --- a/arch/arm/kernel/hyp-stub.S +++ b/arch/arm/kernel/hyp-stub.S @@ -56,8 +56,8 @@ ENTRY(__boot_cpu_mode) ldr \reg3, [\reg2] ldr \reg1, [\reg2, \reg3] cmp \mode, \reg1 @ matches primary CPU boot mode? - orrne r7, r7, #BOOT_CPU_MODE_MISMATCH - strne r7, [r5, r6] @ record what happened and give up + orrne \reg1, \reg1, #BOOT_CPU_MODE_MISMATCH + strne \reg1, [\reg2, \reg3] @ record what happened and give up .endm #else /* ZIMAGE */ -- cgit v1.2.2 From ab8d46c0609843a83aef3f486365ca5e7c21d537 Mon Sep 17 00:00:00 2001 From: Tetsuyuki Kobayashi Date: Mon, 22 Jul 2013 14:58:17 +0100 Subject: ARM: 7788/1: elf: fix lpae hwcap feature reporting in proc/cpuinfo Commit a469abd0f868 ("ARM: elf: add new hwcap for identifying atomic ldrd/strd instructions") added a new hwcap to identify LPAE on CPUs which support it. Whilst the hwcap data is correct, the string reported in /proc/cpuinfo actually matches on HWCAP_VFPD32, which was missing an entry in the string table. This patch fixes this problem by adding a "vfpd32" string at the correct offset, preventing us from falsely advertising LPAE on CPUs which do not support it. [will: added commit message] Acked-by: Will Deacon Tested-by: Will Deacon Signed-off-by: Tetsuyuki Kobayashi Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 63af9a7ae512..96286cb383bd 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -971,6 +971,7 @@ static const char *hwcap_str[] = { "vfpv4", "idiva", "idivt", + "vfpd32", "lpae", NULL }; -- cgit v1.2.2 From 8fbac214e5c594a0c2fe78c14adf2cdbb1febc92 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Thu, 18 Jul 2013 17:20:33 +0100 Subject: ARM: 7787/1: virt: ensure visibility of __boot_cpu_mode Secondary CPUs write to __boot_cpu_mode with caches disabled, and thus a cached value of __boot_cpu_mode may be incoherent with that in memory. This could lead to a failure to detect mismatched boot modes. This patch adds flushing to ensure that writes by secondaries to __boot_cpu_mode are made visible before we test against it. Signed-off-by: Mark Rutland Acked-by: Dave Martin Acked-by: Marc Zyngier Cc: Christoffer Dall Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 96286cb383bd..afc2489ee13b 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -836,6 +836,8 @@ static int __init meminfo_cmp(const void *_a, const void *_b) void __init hyp_mode_check(void) { #ifdef CONFIG_ARM_VIRT_EXT + sync_boot_mode(); + if (is_hyp_mode_available()) { pr_info("CPU: All CPU(s) started in HYP mode.\n"); pr_info("CPU: Virtualization extensions available.\n"); -- cgit v1.2.2 From 1f49856bb029779d8f1b63517a3a3b34ffe672c7 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Tue, 23 Jul 2013 15:13:06 +0100 Subject: ARM: 7789/1: Do not run dummy_flush_tlb_a15_erratum() on non-Cortex-A15 Commit 93dc688 (ARM: 7684/1: errata: Workaround for Cortex-A15 erratum 798181 (TLBI/DSB operations)) causes the following undefined instruction error on a mx53 (Cortex-A8): Internal error: Oops - undefined instruction: 0 [#1] SMP ARM CPU: 0 PID: 275 Comm: modprobe Not tainted 3.11.0-rc2-next-20130722-00009-g9b0f371 #881 task: df46cc00 ti: df48e000 task.ti: df48e000 PC is at check_and_switch_context+0x17c/0x4d0 LR is at check_and_switch_context+0xdc/0x4d0 This problem happens because check_and_switch_context() calls dummy_flush_tlb_a15_erratum() without checking if we are really running on a Cortex-A15 or not. To avoid this issue, only call dummy_flush_tlb_a15_erratum() inside check_and_switch_context() if erratum_a15_798181() returns true, which means that we are really running on a Cortex-A15. Signed-off-by: Fabio Estevam Acked-by: Catalin Marinas Reviewed-by: Roger Quadros Signed-off-by: Russell King --- arch/arm/kernel/smp_tlb.c | 17 ----------------- 1 file changed, 17 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp_tlb.c b/arch/arm/kernel/smp_tlb.c index a98b62dca2fa..c2edfff573c2 100644 --- a/arch/arm/kernel/smp_tlb.c +++ b/arch/arm/kernel/smp_tlb.c @@ -70,23 +70,6 @@ static inline void ipi_flush_bp_all(void *ignored) local_flush_bp_all(); } -#ifdef CONFIG_ARM_ERRATA_798181 -static int erratum_a15_798181(void) -{ - unsigned int midr = read_cpuid_id(); - - /* Cortex-A15 r0p0..r3p2 affected */ - if ((midr & 0xff0ffff0) != 0x410fc0f0 || midr > 0x413fc0f2) - return 0; - return 1; -} -#else -static int erratum_a15_798181(void) -{ - return 0; -} -#endif - static void ipi_flush_tlb_a15_erratum(void *arg) { dmb(); -- cgit v1.2.2 From bed859c1eea89feaba2acbb57d967a6d6f68af09 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Tue, 30 Jul 2013 11:33:00 +0100 Subject: ARM: 7800/1: ARMv7-M: Fix name of NVIC handler function MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The name changed in response to review comments for the nvic irqchip driver when the original name was already accepted into Russell King's tree. Signed-off-by: Uwe Kleine-König Signed-off-by: Russell King --- arch/arm/kernel/entry-v7m.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S index e00621f1403f..52b26432c9a9 100644 --- a/arch/arm/kernel/entry-v7m.S +++ b/arch/arm/kernel/entry-v7m.S @@ -49,7 +49,7 @@ __irq_entry: mov r1, sp stmdb sp!, {lr} @ routine called with r0 = irq number, r1 = struct pt_regs * - bl nvic_do_IRQ + bl nvic_handle_irq pop {lr} @ -- cgit v1.2.2 From f928d4f2a86f46b030fa0850385b4391fc2b5918 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 4 Jul 2013 11:00:23 +0100 Subject: ARM: poison the vectors page Fill the empty regions of the vectors page with an exception generating instruction. This ensures that any inappropriate branch to the vector page is appropriately trapped, rather than just encountering some code to execute. (The vectors page was filled with zero before, which corresponds with the "andeq r0, r0, r0" instruction - a no-op.) Cc: Acked-by Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/traps.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index cab094c234ee..9433e8a12b5e 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -818,9 +818,19 @@ void __init early_trap_init(void *vectors_base) extern char __vectors_start[], __vectors_end[]; extern char __kuser_helper_start[], __kuser_helper_end[]; int kuser_sz = __kuser_helper_end - __kuser_helper_start; + unsigned i; vectors_page = vectors_base; + /* + * Poison the vectors page with an undefined instruction. This + * instruction is chosen to be undefined for both ARM and Thumb + * ISAs. The Thumb version is an undefined instruction with a + * branch back to the undefined instruction. + */ + for (i = 0; i < PAGE_SIZE / sizeof(u32); i++) + ((u32 *)vectors_base)[i] = 0xe7fddef1; + /* * Copy the vectors, stubs and kuser helpers (in entry-armv.S) * into the vector page, mapped at 0xffff0000, and ensure these -- cgit v1.2.2 From 5b43e7a383d69381ffe53423e46dd0fafae07da3 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 4 Jul 2013 11:32:04 +0100 Subject: ARM: poison memory between kuser helpers Poison the memory between each kuser helper. This ensures that any branch between the kuser helpers will be appropriately trapped. Cc: Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index a39cfc2a1f90..02437345289a 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -742,6 +742,17 @@ ENDPROC(__switch_to) #endif .endm + .macro kuser_pad, sym, size + .if (. - \sym) & 3 + .rept 4 - (. - \sym) & 3 + .byte 0 + .endr + .endif + .rept (\size - (. - \sym)) / 4 + .word 0xe7fddef1 + .endr + .endm + .align 5 .globl __kuser_helper_start __kuser_helper_start: @@ -832,18 +843,13 @@ kuser_cmpxchg64_fixup: #error "incoherent kernel configuration" #endif - /* pad to next slot */ - .rept (16 - (. - __kuser_cmpxchg64)/4) - .word 0 - .endr - - .align 5 + kuser_pad __kuser_cmpxchg64, 64 __kuser_memory_barrier: @ 0xffff0fa0 smp_dmb arm usr_ret lr - .align 5 + kuser_pad __kuser_memory_barrier, 32 __kuser_cmpxchg: @ 0xffff0fc0 @@ -916,13 +922,14 @@ kuser_cmpxchg32_fixup: #endif - .align 5 + kuser_pad __kuser_cmpxchg, 32 __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 + kuser_pad __kuser_get_tls, 16 + .rep 3 .word 0 @ 0xffff0ff0 software TLS value, then .endr @ pad up to __kuser_helper_version -- cgit v1.2.2 From 19accfd373847ac3d10623c5d20f948846299741 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 4 Jul 2013 11:40:32 +0100 Subject: ARM: move vector stubs Move the machine vector stubs into the page above the vector page, which we can prevent from being visible to userspace. Also move the reset stub, and place the swi vector at a location that the 'ldr' can get to it. This hides pointers into the kernel which could give valuable information to attackers, and reduces the number of exploitable instructions at a fixed address. Cc: Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 50 +++++++++++++++++++++----------------------- arch/arm/kernel/traps.c | 4 ++-- 2 files changed, 26 insertions(+), 28 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 02437345289a..79a41fa978f5 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -944,9 +944,9 @@ __kuser_helper_end: /* * 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. + * This code is copied to 0xffff1000 so we can use branches in the + * vectors, rather than ldr's. Note that this code must not exceed + * a page size. * * Common stub entry macro: * Enter in IRQ mode, spsr = SVC/USR CPSR, lr = SVC/USR PC @@ -995,6 +995,15 @@ ENDPROC(vector_\name) .globl __stubs_start __stubs_start: + @ This must be the first word + .word vector_swi + +vector_rst: + ARM( swi SYS_ERROR0 ) + THUMB( svc #0 ) + THUMB( nop ) + b vector_und + /* * Interrupt dispatcher */ @@ -1088,6 +1097,16 @@ __stubs_start: .align 5 +/*============================================================================= + * 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 + /*============================================================================= * Undefined FIQs *----------------------------------------------------------------------------- @@ -1101,35 +1120,14 @@ __stubs_start: vector_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 + .equ stubs_offset, __vectors_start + 0x1000 - __stubs_start .globl __vectors_start __vectors_start: - ARM( swi SYS_ERROR0 ) - THUMB( svc #0 ) - THUMB( nop ) + W(b) vector_rst + stubs_offset W(b) vector_und + stubs_offset W(ldr) pc, .LCvswi + stubs_offset W(b) vector_pabt + stubs_offset diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 9433e8a12b5e..2c8c7fa78b8c 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -837,7 +837,7 @@ void __init early_trap_init(void *vectors_base) * are visible to the instruction stream. */ memcpy((void *)vectors, __vectors_start, __vectors_end - __vectors_start); - memcpy((void *)vectors + 0x200, __stubs_start, __stubs_end - __stubs_start); + memcpy((void *)vectors + 0x1000, __stubs_start, __stubs_end - __stubs_start); memcpy((void *)vectors + 0x1000 - kuser_sz, __kuser_helper_start, kuser_sz); /* @@ -852,7 +852,7 @@ void __init early_trap_init(void *vectors_base) memcpy((void *)(vectors + KERN_SIGRETURN_CODE - CONFIG_VECTORS_BASE), sigreturn_codes, sizeof(sigreturn_codes)); - flush_icache_range(vectors, vectors + PAGE_SIZE); + flush_icache_range(vectors, vectors + PAGE_SIZE * 2); modify_domain(DOMAIN_USER, DOMAIN_CLIENT); #else /* ifndef CONFIG_CPU_V7M */ /* -- cgit v1.2.2 From b9b32bf70f2fb710b07c94e13afbc729afe221da Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 4 Jul 2013 12:03:31 +0100 Subject: ARM: use linker magic for vectors and vector stubs Use linker magic to create the vectors and vector stubs: we can tell the linker to place them at an appropriate VMA, but keep the LMA within the kernel. This gets rid of some unnecessary symbol manipulation, and have the linker calculate the relocations appropriately. Cc: Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 28 ++++++++++------------------ arch/arm/kernel/vmlinux.lds.S | 17 +++++++++++++++++ 2 files changed, 27 insertions(+), 18 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 79a41fa978f5..7e1ae91e5b1a 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -993,7 +993,7 @@ ENDPROC(vector_\name) 1: .endm - .globl __stubs_start + .section .stubs, "ax", %progbits __stubs_start: @ This must be the first word .word vector_swi @@ -1120,24 +1120,16 @@ vector_addrexcptn: vector_fiq: subs pc, lr, #4 - .globl __stubs_end -__stubs_end: - - .equ stubs_offset, __vectors_start + 0x1000 - __stubs_start - - .globl __vectors_start + .section .vectors, "ax", %progbits __vectors_start: - W(b) vector_rst + stubs_offset - 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: + W(b) vector_rst + W(b) vector_und + W(ldr) pc, __vectors_start + 0x1000 + W(b) vector_pabt + W(b) vector_dabt + W(b) vector_addrexcptn + W(b) vector_irq + W(b) vector_fiq .data diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index fa25e4e425f6..7bcee5c9b604 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -148,6 +148,23 @@ SECTIONS . = ALIGN(PAGE_SIZE); __init_begin = .; #endif + /* + * The vectors and stubs are relocatable code, and the + * only thing that matters is their relative offsets + */ + __vectors_start = .; + .vectors 0 : AT(__vectors_start) { + *(.vectors) + } + . = __vectors_start + SIZEOF(.vectors); + __vectors_end = .; + + __stubs_start = .; + .stubs 0x1000 : AT(__stubs_start) { + *(.stubs) + } + . = __stubs_start + SIZEOF(.stubs); + __stubs_end = .; INIT_TEXT_SECTION(8) .exit.text : { -- cgit v1.2.2 From e39e3f3ebfef03450cf7bfa7a974a8c61f7980c8 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 9 Jul 2013 01:03:17 +0100 Subject: ARM: update FIQ support for relocation of vectors FIQ should no longer copy the FIQ code into the user visible vector page. Instead, it should use the hidden page. This change makes that happen. Cc: Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 3 +++ arch/arm/kernel/fiq.c | 19 ++++++++++++++----- 2 files changed, 17 insertions(+), 5 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 7e1ae91e5b1a..94caefb550bf 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -1120,6 +1120,9 @@ vector_addrexcptn: vector_fiq: subs pc, lr, #4 + .globl vector_fiq_offset + .equ vector_fiq_offset, vector_fiq + .section .vectors, "ax", %progbits __vectors_start: W(b) vector_rst diff --git a/arch/arm/kernel/fiq.c b/arch/arm/kernel/fiq.c index 2adda11f712f..25442f451148 100644 --- a/arch/arm/kernel/fiq.c +++ b/arch/arm/kernel/fiq.c @@ -47,6 +47,11 @@ #include #include +#define FIQ_OFFSET ({ \ + extern void *vector_fiq_offset; \ + (unsigned)&vector_fiq_offset; \ + }) + static unsigned long no_fiq_insn; /* Default reacquire function @@ -80,13 +85,16 @@ int show_fiq_list(struct seq_file *p, int prec) void set_fiq_handler(void *start, unsigned int length) { #if defined(CONFIG_CPU_USE_DOMAINS) - memcpy((void *)0xffff001c, start, length); + void *base = (void *)0xffff0000; #else - memcpy(vectors_page + 0x1c, start, length); + void *base = vectors_page; #endif - flush_icache_range(0xffff001c, 0xffff001c + length); + unsigned offset = FIQ_OFFSET; + + memcpy(base + offset, start, length); + flush_icache_range(0xffff0000 + offset, 0xffff0000 + offset + length); if (!vectors_high()) - flush_icache_range(0x1c, 0x1c + length); + flush_icache_range(offset, offset + length); } int claim_fiq(struct fiq_handler *f) @@ -144,6 +152,7 @@ EXPORT_SYMBOL(disable_fiq); void __init init_FIQ(int start) { - no_fiq_insn = *(unsigned long *)0xffff001c; + unsigned offset = FIQ_OFFSET; + no_fiq_insn = *(unsigned long *)(0xffff0000 + offset); fiq_start = start; } -- cgit v1.2.2 From f6f91b0d9fd971c630cef908dde8fe8795aefbf8 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 23 Jul 2013 18:37:00 +0100 Subject: ARM: allow kuser helpers to be removed from the vector page Provide a kernel configuration option to allow the kernel user helpers to be removed from the vector page, thereby preventing their use with ROP (return orientated programming) attacks. This option is only visible for CPU architectures which natively support all the operations which kernel user helpers would normally provide, and must be enabled with caution. Cc: Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 3 +++ arch/arm/kernel/traps.c | 23 ++++++++++++++--------- 2 files changed, 17 insertions(+), 9 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 94caefb550bf..d40d0ef389db 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -753,6 +753,7 @@ ENDPROC(__switch_to) .endr .endm +#ifdef CONFIG_KUSER_HELPERS .align 5 .globl __kuser_helper_start __kuser_helper_start: @@ -939,6 +940,8 @@ __kuser_helper_version: @ 0xffff0ffc .globl __kuser_helper_end __kuser_helper_end: +#endif + THUMB( .thumb ) /* diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 2c8c7fa78b8c..e3ca35ccd38e 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -800,15 +800,26 @@ void __init trap_init(void) return; } -static void __init kuser_get_tls_init(unsigned long vectors) +#ifdef CONFIG_KUSER_HELPERS +static void __init kuser_init(void *vectors) { + extern char __kuser_helper_start[], __kuser_helper_end[]; + int kuser_sz = __kuser_helper_end - __kuser_helper_start; + + memcpy(vectors + 0x1000 - kuser_sz, __kuser_helper_start, kuser_sz); + /* * vectors + 0xfe0 = __kuser_get_tls * vectors + 0xfe8 = hardware TLS instruction at 0xffff0fe8 */ if (tls_emu || has_tls_reg) - memcpy((void *)vectors + 0xfe0, (void *)vectors + 0xfe8, 4); + memcpy(vectors + 0xfe0, vectors + 0xfe8, 4); +} +#else +static void __init kuser_init(void *vectors) +{ } +#endif void __init early_trap_init(void *vectors_base) { @@ -816,8 +827,6 @@ void __init early_trap_init(void *vectors_base) unsigned long vectors = (unsigned long)vectors_base; extern char __stubs_start[], __stubs_end[]; extern char __vectors_start[], __vectors_end[]; - extern char __kuser_helper_start[], __kuser_helper_end[]; - int kuser_sz = __kuser_helper_end - __kuser_helper_start; unsigned i; vectors_page = vectors_base; @@ -838,12 +847,8 @@ void __init early_trap_init(void *vectors_base) */ memcpy((void *)vectors, __vectors_start, __vectors_end - __vectors_start); memcpy((void *)vectors + 0x1000, __stubs_start, __stubs_end - __stubs_start); - memcpy((void *)vectors + 0x1000 - kuser_sz, __kuser_helper_start, kuser_sz); - /* - * Do processor specific fixups for the kuser helpers - */ - kuser_get_tls_init(vectors); + kuser_init(vectors_base); /* * Copy signal return handlers into the vector page, and -- cgit v1.2.2 From 48be69a026b2c17350a5ef18a1959a919f60be7d Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 24 Jul 2013 00:29:18 +0100 Subject: ARM: move signal handlers into a vdso-like page Move the signal handlers into a VDSO page rather than keeping them in the vectors page. This allows us to place them randomly within this page, and also map the page at a random location within userspace further protecting these code fragments from ROP attacks. The new VDSO page is also poisoned in the same way as the vector page. Signed-off-by: Russell King --- arch/arm/kernel/process.c | 40 +++++++++++++++++++++++++++++++++--- arch/arm/kernel/signal.c | 52 ++++++++++++++++++++++++++++++++++++++++------- arch/arm/kernel/signal.h | 12 ----------- arch/arm/kernel/traps.c | 9 -------- 4 files changed, 82 insertions(+), 31 deletions(-) delete mode 100644 arch/arm/kernel/signal.h (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index d3ca4f6915af..566d0d71a1e7 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -428,8 +428,8 @@ unsigned long arch_randomize_brk(struct mm_struct *mm) #ifdef CONFIG_MMU /* * The vectors page is always readable from user space for the - * atomic helpers and the signal restart code. Insert it into the - * gate_vma so that it is visible through ptrace and /proc//mem. + * atomic helpers. Insert it into the gate_vma so that it is visible + * through ptrace and /proc//mem. */ static struct vm_area_struct gate_vma = { .vm_start = 0xffff0000, @@ -461,6 +461,40 @@ int in_gate_area_no_mm(unsigned long addr) const char *arch_vma_name(struct vm_area_struct *vma) { - return (vma == &gate_vma) ? "[vectors]" : NULL; + return (vma == &gate_vma) ? "[vectors]" : + (vma->vm_mm && vma->vm_start == vma->vm_mm->context.sigpage) ? + "[sigpage]" : NULL; +} + +extern struct page *get_signal_page(void); + +int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) +{ + struct mm_struct *mm = current->mm; + struct page *page; + unsigned long addr; + int ret; + + page = get_signal_page(); + if (!page) + return -ENOMEM; + + down_write(&mm->mmap_sem); + addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0); + if (IS_ERR_VALUE(addr)) { + ret = addr; + goto up_fail; + } + + ret = install_special_mapping(mm, addr, PAGE_SIZE, + VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC, + &page); + + if (ret == 0) + mm->context.sigpage = addr; + + up_fail: + up_write(&mm->mmap_sem); + return ret; } #endif diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 1c16c35c271a..0f17e06d51e6 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -8,6 +8,7 @@ * published by the Free Software Foundation. */ #include +#include #include #include #include @@ -15,12 +16,11 @@ #include #include +#include #include #include #include -#include "signal.h" - /* * For ARM syscalls, we encode the syscall number into the instruction. */ @@ -40,11 +40,13 @@ #define SWI_THUMB_SIGRETURN (0xdf00 << 16 | 0x2700 | (__NR_sigreturn - __NR_SYSCALL_BASE)) #define SWI_THUMB_RT_SIGRETURN (0xdf00 << 16 | 0x2700 | (__NR_rt_sigreturn - __NR_SYSCALL_BASE)) -const unsigned long sigreturn_codes[7] = { +static const unsigned long sigreturn_codes[7] = { MOV_R7_NR_SIGRETURN, SWI_SYS_SIGRETURN, SWI_THUMB_SIGRETURN, MOV_R7_NR_RT_SIGRETURN, SWI_SYS_RT_SIGRETURN, SWI_THUMB_RT_SIGRETURN, }; +static unsigned long signal_return_offset; + #ifdef CONFIG_CRUNCH static int preserve_crunch_context(struct crunch_sigframe __user *frame) { @@ -401,12 +403,15 @@ setup_return(struct pt_regs *regs, struct ksignal *ksig, return 1; if ((cpsr & MODE32_BIT) && !IS_ENABLED(CONFIG_ARM_MPU)) { + struct mm_struct *mm = current->mm; + /* - * 32-bit code can use the new high-page - * signal return code support except when the MPU has - * protected the vectors page from PL0 + * 32-bit code can use the signal return page + * except when the MPU has protected the vectors + * page from PL0 */ - retcode = KERN_SIGRETURN_CODE + (idx << 2) + thumb; + retcode = mm->context.sigpage + signal_return_offset + + (idx << 2) + thumb; } else { /* * Ensure that the instruction cache sees @@ -608,3 +613,36 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) } while (thread_flags & _TIF_WORK_MASK); return 0; } + +static struct page *signal_page; + +struct page *get_signal_page(void) +{ + if (!signal_page) { + unsigned long ptr; + unsigned offset; + void *addr; + + signal_page = alloc_pages(GFP_KERNEL, 0); + + if (!signal_page) + return NULL; + + addr = page_address(signal_page); + + /* Give the signal return code some randomness */ + offset = 0x200 + (get_random_int() & 0x7fc); + signal_return_offset = offset; + + /* + * Copy signal return handlers into the vector page, and + * set sigreturn to be a pointer to these. + */ + memcpy(addr + offset, sigreturn_codes, sizeof(sigreturn_codes)); + + ptr = (unsigned long)addr + offset; + flush_icache_range(ptr, ptr + sizeof(sigreturn_codes)); + } + + return signal_page; +} diff --git a/arch/arm/kernel/signal.h b/arch/arm/kernel/signal.h deleted file mode 100644 index 5ff067b7c752..000000000000 --- a/arch/arm/kernel/signal.h +++ /dev/null @@ -1,12 +0,0 @@ -/* - * linux/arch/arm/kernel/signal.h - * - * Copyright (C) 2005-2009 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. - */ -#define KERN_SIGRETURN_CODE (CONFIG_VECTORS_BASE + 0x00000500) - -extern const unsigned long sigreturn_codes[7]; diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index e3ca35ccd38e..ab517fcce21b 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -35,8 +35,6 @@ #include #include -#include "signal.h" - static const char *handler[]= { "prefetch abort", "data abort", "address exception", "interrupt" }; void *vectors_page; @@ -850,13 +848,6 @@ void __init early_trap_init(void *vectors_base) kuser_init(vectors_base); - /* - * Copy signal return handlers into the vector page, and - * set sigreturn to be a pointer to these. - */ - memcpy((void *)(vectors + KERN_SIGRETURN_CODE - CONFIG_VECTORS_BASE), - sigreturn_codes, sizeof(sigreturn_codes)); - flush_icache_range(vectors, vectors + PAGE_SIZE * 2); modify_domain(DOMAIN_USER, DOMAIN_CLIENT); #else /* ifndef CONFIG_CPU_V7M */ -- cgit v1.2.2 From a5463cd3435475386cbbe7b06e01292ac169d36f Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 31 Jul 2013 21:58:56 +0100 Subject: ARM: make vectors page inaccessible from userspace If kuser helpers are not provided by the kernel, disable user access to the vectors page. With the kuser helpers gone, there is no reason for this page to be visible to userspace. Signed-off-by: Russell King --- arch/arm/kernel/process.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 566d0d71a1e7..1e6c33d01c05 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -426,6 +426,7 @@ unsigned long arch_randomize_brk(struct mm_struct *mm) } #ifdef CONFIG_MMU +#ifdef CONFIG_KUSER_HELPERS /* * The vectors page is always readable from user space for the * atomic helpers. Insert it into the gate_vma so that it is visible @@ -458,10 +459,14 @@ int in_gate_area_no_mm(unsigned long addr) { return in_gate_area(NULL, addr); } +#define is_gate_vma(vma) ((vma) = &gate_vma) +#else +#define is_gate_vma(vma) 0 +#endif const char *arch_vma_name(struct vm_area_struct *vma) { - return (vma == &gate_vma) ? "[vectors]" : + return is_gate_vma(vma) ? "[vectors]" : (vma->vm_mm && vma->vm_start == vma->vm_mm->context.sigpage) ? "[sigpage]" : NULL; } -- cgit v1.2.2 From 44424c34049f41123a3a8b4853822f47f4ff03a2 Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Tue, 30 Jul 2013 23:09:46 +0100 Subject: ARM: 7803/1: Fix deadlock scenario with smp_send_stop() If one process calls sys_reboot and that process then stops other CPUs while those CPUs are within a spin_lock() region we can potentially encounter a deadlock scenario like below. CPU 0 CPU 1 ----- ----- spin_lock(my_lock) smp_send_stop() handle_IPI() disable_preemption/irqs while(1); spin_lock(my_lock) <--- Waits forever We shouldn't attempt to run any other tasks after we send a stop IPI to a CPU so disable preemption so that this task runs to completion. We use local_irq_disable() here for cross-arch consistency with x86. Reported-by: Sundarajan Srinivasan Signed-off-by: Stephen Boyd Signed-off-by: Russell King --- arch/arm/kernel/process.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index d3ca4f6915af..08b47ebd3144 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -197,6 +197,7 @@ void machine_shutdown(void) */ void machine_halt(void) { + local_irq_disable(); smp_send_stop(); local_irq_disable(); @@ -211,6 +212,7 @@ void machine_halt(void) */ void machine_power_off(void) { + local_irq_disable(); smp_send_stop(); if (pm_power_off) @@ -230,6 +232,7 @@ void machine_power_off(void) */ void machine_restart(char *cmd) { + local_irq_disable(); smp_send_stop(); arm_pm_restart(reboot_mode, cmd); -- cgit v1.2.2 From 2449189bb7c73b5fe55a18bc0d289e39bdcd4998 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 31 Jul 2013 11:37:17 +0100 Subject: ARM: Add .text annotations where required after __CPUINIT removal Commit 8bd26e3a7 (arm: delete __cpuinit/__CPUINIT usage from all ARM users) caused some code to leak into sections which are discarded through the removal of __CPUINIT annotations. Add appropriate .text annotations to bring these back into the kernel text. Signed-off-by: Russell King --- arch/arm/kernel/head-nommu.S | 1 + arch/arm/kernel/head.S | 1 + 2 files changed, 2 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index b361de143756..14235ba64a90 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -87,6 +87,7 @@ ENTRY(stext) ENDPROC(stext) #ifdef CONFIG_SMP + .text ENTRY(secondary_startup) /* * Common entry point for secondary CPUs. diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 9cf6063020ae..2c7cc1e03473 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -343,6 +343,7 @@ __turn_mmu_on_loc: .long __turn_mmu_on_end #if defined(CONFIG_SMP) + .text ENTRY(secondary_startup) /* * Common entry point for secondary CPUs. -- cgit v1.2.2 From e0d407564b532d978b03ceccebd224a05d02f111 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 3 Aug 2013 10:30:05 +0100 Subject: ARM: fix a cockup in 48be69a02 (ARM: move signal handlers into a vdso-like page) Unfortunately, I never committed the fix to a nasty oops which can occur as a result of that commit: ------------[ cut here ]------------ kernel BUG at /home/olof/work/batch/include/linux/mm.h:414! Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM Modules linked in: CPU: 0 PID: 490 Comm: killall5 Not tainted 3.11.0-rc3-00288-gabe0308 #53 task: e90acac0 ti: e9be8000 task.ti: e9be8000 PC is at special_mapping_fault+0xa4/0xc4 LR is at __do_fault+0x68/0x48c This doesn't show up unless you do quite a bit of testing; a simple boot test does not do this, so all my nightly tests were passing fine. The reason for this is that install_special_mapping() expects the page array to stick around, and as this was only inserting one page which was stored on the kernel stack, that's why this was blowing up. Reported-by: Olof Johansson Tested-by: Olof Johansson Signed-off-by: Russell King --- arch/arm/kernel/process.c | 9 +++++---- arch/arm/kernel/signal.c | 41 +++++++++++++++++++---------------------- 2 files changed, 24 insertions(+), 26 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 1e6c33d01c05..d03b5bd889c5 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -471,17 +471,18 @@ const char *arch_vma_name(struct vm_area_struct *vma) "[sigpage]" : NULL; } +static struct page *signal_page; extern struct page *get_signal_page(void); int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) { struct mm_struct *mm = current->mm; - struct page *page; unsigned long addr; int ret; - page = get_signal_page(); - if (!page) + if (!signal_page) + signal_page = get_signal_page(); + if (!signal_page) return -ENOMEM; down_write(&mm->mmap_sem); @@ -493,7 +494,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) ret = install_special_mapping(mm, addr, PAGE_SIZE, VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC, - &page); + &signal_page); if (ret == 0) mm->context.sigpage = addr; diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 0f17e06d51e6..39e7105a9b70 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -614,35 +614,32 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) return 0; } -static struct page *signal_page; - struct page *get_signal_page(void) { - if (!signal_page) { - unsigned long ptr; - unsigned offset; - void *addr; + unsigned long ptr; + unsigned offset; + struct page *page; + void *addr; - signal_page = alloc_pages(GFP_KERNEL, 0); + page = alloc_pages(GFP_KERNEL, 0); - if (!signal_page) - return NULL; + if (!page) + return NULL; - addr = page_address(signal_page); + addr = page_address(page); - /* Give the signal return code some randomness */ - offset = 0x200 + (get_random_int() & 0x7fc); - signal_return_offset = offset; + /* Give the signal return code some randomness */ + offset = 0x200 + (get_random_int() & 0x7fc); + signal_return_offset = offset; - /* - * Copy signal return handlers into the vector page, and - * set sigreturn to be a pointer to these. - */ - memcpy(addr + offset, sigreturn_codes, sizeof(sigreturn_codes)); + /* + * Copy signal return handlers into the vector page, and + * set sigreturn to be a pointer to these. + */ + memcpy(addr + offset, sigreturn_codes, sizeof(sigreturn_codes)); - ptr = (unsigned long)addr + offset; - flush_icache_range(ptr, ptr + sizeof(sigreturn_codes)); - } + ptr = (unsigned long)addr + offset; + flush_icache_range(ptr, ptr + sizeof(sigreturn_codes)); - return signal_page; + return page; } -- cgit v1.2.2 From 8c0cc8a5d90bc7373a7a9e7f7a40eb41f51e03fc Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 3 Aug 2013 10:39:51 +0100 Subject: ARM: fix nommu builds with 48be69a02 (ARM: move signal handlers into a vdso-like page) Olof reports that noMMU builds error out with: arch/arm/kernel/signal.c: In function 'setup_return': arch/arm/kernel/signal.c:413:25: error: 'mm_context_t' has no member named 'sigpage' This shows one of the evilnesses of IS_ENABLED(). Get rid of it here and replace it with #ifdef's - and as no noMMU platform can make use of sigpage, depend on CONIFG_MMU not CONFIG_ARM_MPU. Reported-by: Olof Johansson Signed-off-by: Russell King --- arch/arm/kernel/signal.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 39e7105a9b70..ab3304225272 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -402,7 +402,8 @@ setup_return(struct pt_regs *regs, struct ksignal *ksig, __put_user(sigreturn_codes[idx+1], rc+1)) return 1; - if ((cpsr & MODE32_BIT) && !IS_ENABLED(CONFIG_ARM_MPU)) { +#ifdef CONFIG_MMU + if (cpsr & MODE32_BIT) { struct mm_struct *mm = current->mm; /* @@ -412,7 +413,9 @@ setup_return(struct pt_regs *regs, struct ksignal *ksig, */ retcode = mm->context.sigpage + signal_return_offset + (idx << 2) + thumb; - } else { + } else +#endif + { /* * Ensure that the instruction cache sees * the return code written onto the stack. -- cgit v1.2.2 From b88a2595b6d8aedbd275c07dfa784657b4f757eb Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Wed, 7 Aug 2013 16:18:08 -0700 Subject: perf/arm: Fix armpmu_map_hw_event() Fix constraint check in armpmu_map_hw_event(). Reported-and-tested-by: Vince Weaver Cc: Signed-off-by: Ingo Molnar Signed-off-by: Linus Torvalds --- arch/arm/kernel/perf_event.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index d9f5cd4e533f..21f77906602c 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -53,7 +53,12 @@ armpmu_map_cache_event(const unsigned (*cache_map) static int armpmu_map_hw_event(const unsigned (*event_map)[PERF_COUNT_HW_MAX], u64 config) { - int mapping = (*event_map)[config]; + int mapping; + + if (config >= PERF_COUNT_HW_MAX) + return -ENOENT; + + mapping = (*event_map)[config]; return mapping == HW_OP_UNSUPPORTED ? -ENOENT : mapping; } -- cgit v1.2.2