From a2e2725541fad72416326798c2d7fa4dafb7d337 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Mon, 12 Oct 2009 23:40:10 -0700 Subject: net: Introduce recvmmsg socket syscall MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Meaning receive multiple messages, reducing the number of syscalls and net stack entry/exit operations. Next patches will introduce mechanisms where protocols that want to optimize this operation will provide an unlocked_recvmsg operation. This takes into account comments made by: . Paul Moore: sock_recvmsg is called only for the first datagram, sock_recvmsg_nosec is used for the rest. . Caitlin Bestler: recvmmsg now has a struct timespec timeout, that works in the same fashion as the ppoll one. If the underlying protocol returns a datagram with MSG_OOB set, this will make recvmmsg return right away with as many datagrams (+ the OOB one) it has received so far. . Rémi Denis-Courmont & Steven Whitehouse: If we receive N < vlen datagrams and then recvmsg returns an error, recvmmsg will return the successfully received datagrams, store the error and return it in the next call. This paves the way for a subsequent optimization, sk_prot->unlocked_recvmsg, where we will be able to acquire the lock only at batch start and end, not at every underlying recvmsg call. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- arch/arm/kernel/calls.S | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index fafce1b5c69f..f58c1156e779 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -374,6 +374,7 @@ CALL(sys_pwritev) CALL(sys_rt_tgsigqueueinfo) CALL(sys_perf_event_open) +/* 365 */ CALL(sys_recvmmsg) #ifndef syscalls_counted .equ syscalls_padding, ((NR_syscalls + 3) & ~3) - NR_syscalls #define syscalls_counted -- cgit v1.2.2 From 50469619999a0bc2ba8fa1365dc443b7aed190af Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Fri, 3 Apr 2009 05:11:21 -0700 Subject: sysctl arm: Remove binary sysctl support Now that sys_sysctl is a generic wrapper around /proc/sys .ctl_name and .strategy members of sysctl tables are dead code. Remove them. Remove a smattering of ctl_names used in sysctl paths, and kill the ctl_names in the recently added mach-bcmring. mach-bcmring never should have had sysctl entries with .ctl_name set. The binary sysctl interface has been frozen for a long time before that code was merged, to prevent probmes with conflicts and lack of testing. The sysctl_check code would have caught this if anyone had ever tested it that way. So I have simply dropped the binary sysctl support instead of adding another compat entry into sysctl_binary.c. Going through /proc/sys/reboot/warm will still work. Cc: Leo Chen Cc: Russell King Signed-off-by: Eric W. Biederman --- arch/arm/kernel/isa.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/isa.c b/arch/arm/kernel/isa.c index 8ac9b8424007..738dfcc658ca 100644 --- a/arch/arm/kernel/isa.c +++ b/arch/arm/kernel/isa.c @@ -22,47 +22,42 @@ static unsigned int isa_membase, isa_portbase, isa_portshift; static ctl_table ctl_isa_vars[4] = { { - .ctl_name = BUS_ISA_MEM_BASE, .procname = "membase", .data = &isa_membase, .maxlen = sizeof(isa_membase), .mode = 0444, .proc_handler = &proc_dointvec, }, { - .ctl_name = BUS_ISA_PORT_BASE, .procname = "portbase", .data = &isa_portbase, .maxlen = sizeof(isa_portbase), .mode = 0444, .proc_handler = &proc_dointvec, }, { - .ctl_name = BUS_ISA_PORT_SHIFT, .procname = "portshift", .data = &isa_portshift, .maxlen = sizeof(isa_portshift), .mode = 0444, .proc_handler = &proc_dointvec, - }, {0} + }, {} }; static struct ctl_table_header *isa_sysctl_header; static ctl_table ctl_isa[2] = { { - .ctl_name = CTL_BUS_ISA, .procname = "isa", .mode = 0555, .child = ctl_isa_vars, - }, {0} + }, {} }; static ctl_table ctl_bus[2] = { { - .ctl_name = CTL_BUS, .procname = "bus", .mode = 0555, .child = ctl_isa, - }, {0} + }, {} }; void __init -- cgit v1.2.2 From 6d4561110a3e9fa742aeec6717248a491dfb1878 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 16 Nov 2009 03:11:48 -0800 Subject: sysctl: Drop & in front of every proc_handler. For consistency drop & in front of every proc_handler. Explicity taking the address is unnecessary and it prevents optimizations like stubbing the proc_handlers to NULL. Cc: Alexey Dobriyan Cc: Ingo Molnar Cc: Joe Perches Signed-off-by: Eric W. Biederman --- arch/arm/kernel/isa.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/isa.c b/arch/arm/kernel/isa.c index 738dfcc658ca..346485910732 100644 --- a/arch/arm/kernel/isa.c +++ b/arch/arm/kernel/isa.c @@ -26,19 +26,19 @@ static ctl_table ctl_isa_vars[4] = { .data = &isa_membase, .maxlen = sizeof(isa_membase), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { .procname = "portbase", .data = &isa_portbase, .maxlen = sizeof(isa_portbase), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { .procname = "portshift", .data = &isa_portshift, .maxlen = sizeof(isa_portshift), .mode = 0444, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, {} }; -- cgit v1.2.2 From 2abc1c50b6be81233e0b79478dc04d8fec737ed5 Mon Sep 17 00:00:00 2001 From: Tim Abbott Date: Fri, 2 Oct 2009 16:32:46 -0400 Subject: ARM: convert to use __HEAD and HEAD_TEXT macros. This has the consequence of changing the section name used for head code from ".text.head" to ".head.text". Since this commit changes all users in the architecture, this change should be harmless. The .text.head output section is eliminated and the head text code is included at the start of the .init output section. Signed-off-by: Tim Abbott Cc: Linus Torvalds Acked-by: Sam Ravnborg Signed-off-by: Russell King --- arch/arm/kernel/head-nommu.S | 2 +- arch/arm/kernel/head.S | 2 +- arch/arm/kernel/vmlinux.lds.S | 8 +++----- 3 files changed, 5 insertions(+), 7 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S index e5dfc2895e24..573b803dc6bf 100644 --- a/arch/arm/kernel/head-nommu.S +++ b/arch/arm/kernel/head-nommu.S @@ -32,7 +32,7 @@ * numbers for r1. * */ - .section ".text.head", "ax" + __HEAD ENTRY(stext) setmode PSR_F_BIT | PSR_I_BIT | SVC_MODE, r9 @ ensure svc mode @ and irqs disabled diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 38ccbe1d3b2c..eb62bf947212 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -74,7 +74,7 @@ * crap here - that's what the boot loader (or in extreme, well justified * circumstances, zImage) is for. */ - .section ".text.head", "ax" + __HEAD ENTRY(stext) setmode PSR_F_BIT | PSR_I_BIT | SVC_MODE, r9 @ ensure svc mode @ and irqs disabled diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index aecf87dfbaec..0902f806cd37 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -24,13 +24,11 @@ SECTIONS #else . = PAGE_OFFSET + TEXT_OFFSET; #endif - .text.head : { - _stext = .; - _sinittext = .; - *(.text.head) - } .init : { /* Init code and data */ + _stext = .; + _sinittext = .; + HEAD_TEXT INIT_TEXT _einittext = .; __proc_info_begin = .; -- cgit v1.2.2 From 78d7530ac338e458fa513527ee7bb4278d0d8017 Mon Sep 17 00:00:00 2001 From: Nelson Elhage Date: Fri, 2 Oct 2009 16:32:47 -0400 Subject: ARM: Clean up linker script using new linker script macros. This patch is mostly a straightforward translation. The primary side effect to the resulting vmlinux should be to increase the alignment on the initramfs to the standard PAGE_SIZE from 32 bytes. Signed-off-by: Nelson Elhage Signed-off-by: Tim Abbott Acked-by: Sam Ravnborg Signed-off-by: Russell King --- arch/arm/kernel/vmlinux.lds.S | 77 +++++++++++++------------------------------ 1 file changed, 23 insertions(+), 54 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 0902f806cd37..71151bd87a36 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -40,43 +40,31 @@ SECTIONS __tagtable_begin = .; *(.taglist.init) __tagtable_end = .; - . = ALIGN(16); - __setup_start = .; - *(.init.setup) - __setup_end = .; + + INIT_SETUP(16) + __early_begin = .; *(.early_param.init) __early_end = .; - __initcall_start = .; - INITCALLS - __initcall_end = .; - __con_initcall_start = .; - *(.con_initcall.init) - __con_initcall_end = .; - __security_initcall_start = .; - *(.security_initcall.init) - __security_initcall_end = .; -#ifdef CONFIG_BLK_DEV_INITRD - . = ALIGN(32); - __initramfs_start = .; - usr/built-in.o(.init.ramfs) - __initramfs_end = .; -#endif - . = ALIGN(PAGE_SIZE); - __per_cpu_load = .; - __per_cpu_start = .; - *(.data.percpu.page_aligned) - *(.data.percpu) - *(.data.percpu.shared_aligned) - __per_cpu_end = .; + + INIT_CALLS + CON_INITCALL + SECURITY_INITCALL + INIT_RAM_FS + #ifndef CONFIG_XIP_KERNEL __init_begin = _stext; INIT_DATA - . = ALIGN(PAGE_SIZE); - __init_end = .; #endif } + PERCPU(PAGE_SIZE) + +#ifndef CONFIG_XIP_KERNEL + . = ALIGN(PAGE_SIZE); + __init_end = .; +#endif + /DISCARD/ : { /* Exit code and data */ EXIT_TEXT EXIT_DATA @@ -155,7 +143,7 @@ SECTIONS * first, the init task union, aligned * to an 8192 byte boundary. */ - *(.data.init_task) + INIT_TASK_DATA(THREAD_SIZE) #ifdef CONFIG_XIP_KERNEL . = ALIGN(PAGE_SIZE); @@ -165,17 +153,8 @@ SECTIONS __init_end = .; #endif - . = ALIGN(PAGE_SIZE); - __nosave_begin = .; - *(.data.nosave) - . = ALIGN(PAGE_SIZE); - __nosave_end = .; - - /* - * then the cacheline aligned data - */ - . = ALIGN(32); - *(.data.cacheline_aligned) + NOSAVE_DATA + CACHELINE_ALIGNED_DATA(32) /* * The exception fixup table (might need resorting at runtime) @@ -254,20 +233,10 @@ SECTIONS } #endif - .bss : { - __bss_start = .; /* BSS */ - *(.bss) - *(COMMON) - __bss_stop = .; - _end = .; - } - /* Stabs debugging sections. */ - .stab 0 : { *(.stab) } - .stabstr 0 : { *(.stabstr) } - .stab.excl 0 : { *(.stab.excl) } - .stab.exclstr 0 : { *(.stab.exclstr) } - .stab.index 0 : { *(.stab.index) } - .stab.indexstr 0 : { *(.stab.indexstr) } + BSS_SECTION(0, 0, 0) + _end = .; + + STABS_DEBUG .comment 0 : { *(.comment) } } -- cgit v1.2.2 From c5d6c7708c3e58015b2e4e13e6cea02c8567a94e Mon Sep 17 00:00:00 2001 From: Alexander Shishkin Date: Tue, 1 Dec 2009 14:00:51 +0100 Subject: ARM: 5841/1: a driver for on-chip ETM and ETB This driver implements support for on-chip Embedded Tracing Macrocell and Embedded Trace Buffer. It allows to trigger tracing of kernel execution flow and exporting trace output to userspace via character device and a sysrq combo. Trace output can then be decoded by a fairly simple open source tool [1] which is already sufficient to get the idea of what the kernel is doing. [1]: http://github.com/virtuoso/etm2human Signed-off-by: Alexander Shishkin Signed-off-by: Russell King --- arch/arm/kernel/Makefile | 2 + arch/arm/kernel/etm.c | 641 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 643 insertions(+) create mode 100644 arch/arm/kernel/etm.c (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index 79087dd6d869..e7ccf7e697ce 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -17,6 +17,8 @@ obj-y := compat.o elf.o entry-armv.o entry-common.o irq.o \ process.o ptrace.o return_address.o setup.o signal.o \ sys_arm.o stacktrace.o time.o traps.o +obj-$(CONFIG_OC_ETM) += etm.o + obj-$(CONFIG_ISA_DMA_API) += dma.o obj-$(CONFIG_ARCH_ACORN) += ecard.o obj-$(CONFIG_FIQ) += fiq.o diff --git a/arch/arm/kernel/etm.c b/arch/arm/kernel/etm.c new file mode 100644 index 000000000000..827753966301 --- /dev/null +++ b/arch/arm/kernel/etm.c @@ -0,0 +1,641 @@ +/* + * linux/arch/arm/kernel/etm.c + * + * Driver for ARM's Embedded Trace Macrocell and Embedded Trace Buffer. + * + * Copyright (C) 2009 Nokia Corporation. + * Alexander Shishkin + * + * 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 +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Alexander Shishkin"); + +static struct tracectx tracer; + +static inline bool trace_isrunning(struct tracectx *t) +{ + return !!(t->flags & TRACER_RUNNING); +} + +static int etm_setup_address_range(struct tracectx *t, int n, + unsigned long start, unsigned long end, int exclude, int data) +{ + u32 flags = ETMAAT_ARM | ETMAAT_IGNCONTEXTID | ETMAAT_NSONLY | \ + ETMAAT_NOVALCMP; + + if (n < 1 || n > t->ncmppairs) + return -EINVAL; + + /* comparators and ranges are numbered starting with 1 as opposed + * to bits in a word */ + n--; + + if (data) + flags |= ETMAAT_DLOADSTORE; + else + flags |= ETMAAT_IEXEC; + + /* first comparator for the range */ + etm_writel(t, flags, ETMR_COMP_ACC_TYPE(n * 2)); + etm_writel(t, start, ETMR_COMP_VAL(n * 2)); + + /* second comparator is right next to it */ + etm_writel(t, flags, ETMR_COMP_ACC_TYPE(n * 2 + 1)); + etm_writel(t, end, ETMR_COMP_VAL(n * 2 + 1)); + + flags = exclude ? ETMTE_INCLEXCL : 0; + etm_writel(t, flags | (1 << n), ETMR_TRACEENCTRL); + + return 0; +} + +static int trace_start(struct tracectx *t) +{ + u32 v; + unsigned long timeout = TRACER_TIMEOUT; + + etb_unlock(t); + + etb_writel(t, 0, ETBR_FORMATTERCTRL); + etb_writel(t, 1, ETBR_CTRL); + + etb_lock(t); + + /* configure etm */ + v = ETMCTRL_OPTS | ETMCTRL_PROGRAM | ETMCTRL_PORTSIZE(t->etm_portsz); + + if (t->flags & TRACER_CYCLE_ACC) + v |= ETMCTRL_CYCLEACCURATE; + + etm_unlock(t); + + etm_writel(t, v, ETMR_CTRL); + + while (!(etm_readl(t, ETMR_CTRL) & ETMCTRL_PROGRAM) && --timeout) + ; + if (!timeout) { + dev_dbg(t->dev, "Waiting for progbit to assert timed out\n"); + etm_lock(t); + return -EFAULT; + } + + etm_setup_address_range(t, 1, (unsigned long)_stext, + (unsigned long)_etext, 0, 0); + etm_writel(t, 0, ETMR_TRACEENCTRL2); + etm_writel(t, 0, ETMR_TRACESSCTRL); + etm_writel(t, 0x6f, ETMR_TRACEENEVT); + + v &= ~ETMCTRL_PROGRAM; + v |= ETMCTRL_PORTSEL; + + etm_writel(t, v, ETMR_CTRL); + + timeout = TRACER_TIMEOUT; + while (etm_readl(t, ETMR_CTRL) & ETMCTRL_PROGRAM && --timeout) + ; + if (!timeout) { + dev_dbg(t->dev, "Waiting for progbit to deassert timed out\n"); + etm_lock(t); + return -EFAULT; + } + + etm_lock(t); + + t->flags |= TRACER_RUNNING; + + return 0; +} + +static int trace_stop(struct tracectx *t) +{ + unsigned long timeout = TRACER_TIMEOUT; + + etm_unlock(t); + + etm_writel(t, 0x440, ETMR_CTRL); + while (!(etm_readl(t, ETMR_CTRL) & ETMCTRL_PROGRAM) && --timeout) + ; + if (!timeout) { + dev_dbg(t->dev, "Waiting for progbit to assert timed out\n"); + etm_lock(t); + return -EFAULT; + } + + etm_lock(t); + + etb_unlock(t); + etb_writel(t, ETBFF_MANUAL_FLUSH, ETBR_FORMATTERCTRL); + + timeout = TRACER_TIMEOUT; + while (etb_readl(t, ETBR_FORMATTERCTRL) & + ETBFF_MANUAL_FLUSH && --timeout) + ; + if (!timeout) { + dev_dbg(t->dev, "Waiting for formatter flush to commence " + "timed out\n"); + etb_lock(t); + return -EFAULT; + } + + etb_writel(t, 0, ETBR_CTRL); + + etb_lock(t); + + t->flags &= ~TRACER_RUNNING; + + return 0; +} + +static int etb_getdatalen(struct tracectx *t) +{ + u32 v; + int rp, wp; + + v = etb_readl(t, ETBR_STATUS); + + if (v & 1) + return t->etb_bufsz; + + rp = etb_readl(t, ETBR_READADDR); + wp = etb_readl(t, ETBR_WRITEADDR); + + if (rp > wp) { + etb_writel(t, 0, ETBR_READADDR); + etb_writel(t, 0, ETBR_WRITEADDR); + + return 0; + } + + return wp - rp; +} + +/* sysrq+v will always stop the running trace and leave it at that */ +static void etm_dump(void) +{ + struct tracectx *t = &tracer; + u32 first = 0; + int length; + + if (!t->etb_regs) { + printk(KERN_INFO "No tracing hardware found\n"); + return; + } + + if (trace_isrunning(t)) + trace_stop(t); + + etb_unlock(t); + + length = etb_getdatalen(t); + + if (length == t->etb_bufsz) + first = etb_readl(t, ETBR_WRITEADDR); + + etb_writel(t, first, ETBR_READADDR); + + printk(KERN_INFO "Trace buffer contents length: %d\n", length); + printk(KERN_INFO "--- ETB buffer begin ---\n"); + for (; length; length--) + printk("%08x", cpu_to_be32(etb_readl(t, ETBR_READMEM))); + printk(KERN_INFO "\n--- ETB buffer end ---\n"); + + /* deassert the overflow bit */ + etb_writel(t, 1, ETBR_CTRL); + etb_writel(t, 0, ETBR_CTRL); + + etb_writel(t, 0, ETBR_TRIGGERCOUNT); + etb_writel(t, 0, ETBR_READADDR); + etb_writel(t, 0, ETBR_WRITEADDR); + + etb_lock(t); +} + +static void sysrq_etm_dump(int key, struct tty_struct *tty) +{ + dev_dbg(tracer.dev, "Dumping ETB buffer\n"); + etm_dump(); +} + +static struct sysrq_key_op sysrq_etm_op = { + .handler = sysrq_etm_dump, + .help_msg = "ETM buffer dump", + .action_msg = "etm", +}; + +static int etb_open(struct inode *inode, struct file *file) +{ + if (!tracer.etb_regs) + return -ENODEV; + + file->private_data = &tracer; + + return nonseekable_open(inode, file); +} + +static ssize_t etb_read(struct file *file, char __user *data, + size_t len, loff_t *ppos) +{ + int total, i; + long length; + struct tracectx *t = file->private_data; + u32 first = 0; + u32 *buf; + + mutex_lock(&t->mutex); + + if (trace_isrunning(t)) { + length = 0; + goto out; + } + + etb_unlock(t); + + total = etb_getdatalen(t); + if (total == t->etb_bufsz) + first = etb_readl(t, ETBR_WRITEADDR); + + etb_writel(t, first, ETBR_READADDR); + + length = min(total * 4, (int)len); + buf = vmalloc(length); + + dev_dbg(t->dev, "ETB buffer length: %d\n", total); + dev_dbg(t->dev, "ETB status reg: %x\n", etb_readl(t, ETBR_STATUS)); + for (i = 0; i < length / 4; i++) + buf[i] = etb_readl(t, ETBR_READMEM); + + /* the only way to deassert overflow bit in ETB status is this */ + etb_writel(t, 1, ETBR_CTRL); + etb_writel(t, 0, ETBR_CTRL); + + etb_writel(t, 0, ETBR_WRITEADDR); + etb_writel(t, 0, ETBR_READADDR); + etb_writel(t, 0, ETBR_TRIGGERCOUNT); + + etb_lock(t); + + length -= copy_to_user(data, buf, length); + vfree(buf); + +out: + mutex_unlock(&t->mutex); + + return length; +} + +static int etb_release(struct inode *inode, struct file *file) +{ + /* there's nothing to do here, actually */ + return 0; +} + +static const struct file_operations etb_fops = { + .owner = THIS_MODULE, + .read = etb_read, + .open = etb_open, + .release = etb_release, +}; + +static struct miscdevice etb_miscdev = { + .name = "tracebuf", + .minor = 0, + .fops = &etb_fops, +}; + +static int __init etb_probe(struct amba_device *dev, struct amba_id *id) +{ + struct tracectx *t = &tracer; + int ret = 0; + + ret = amba_request_regions(dev, NULL); + if (ret) + goto out; + + t->etb_regs = ioremap_nocache(dev->res.start, resource_size(&dev->res)); + if (!t->etb_regs) { + ret = -ENOMEM; + goto out_release; + } + + amba_set_drvdata(dev, t); + + etb_miscdev.parent = &dev->dev; + + ret = misc_register(&etb_miscdev); + if (ret) + goto out_unmap; + + t->emu_clk = clk_get(&dev->dev, "emu_src_ck"); + if (IS_ERR(t->emu_clk)) { + dev_dbg(&dev->dev, "Failed to obtain emu_src_ck.\n"); + return -EFAULT; + } + + clk_enable(t->emu_clk); + + etb_unlock(t); + t->etb_bufsz = etb_readl(t, ETBR_DEPTH); + dev_dbg(&dev->dev, "Size: %x\n", t->etb_bufsz); + + /* make sure trace capture is disabled */ + etb_writel(t, 0, ETBR_CTRL); + etb_writel(t, 0x1000, ETBR_FORMATTERCTRL); + etb_lock(t); + + dev_dbg(&dev->dev, "ETB AMBA driver initialized.\n"); + +out: + return ret; + +out_unmap: + amba_set_drvdata(dev, NULL); + iounmap(t->etb_regs); + +out_release: + amba_release_regions(dev); + + return ret; +} + +static int etb_remove(struct amba_device *dev) +{ + struct tracectx *t = amba_get_drvdata(dev); + + amba_set_drvdata(dev, NULL); + + iounmap(t->etb_regs); + t->etb_regs = NULL; + + clk_disable(t->emu_clk); + clk_put(t->emu_clk); + + amba_release_regions(dev); + + return 0; +} + +static struct amba_id etb_ids[] = { + { + .id = 0x0003b907, + .mask = 0x0007ffff, + }, + { 0, 0 }, +}; + +static struct amba_driver etb_driver = { + .drv = { + .name = "etb", + .owner = THIS_MODULE, + }, + .probe = etb_probe, + .remove = etb_remove, + .id_table = etb_ids, +}; + +/* use a sysfs file "trace_running" to start/stop tracing */ +static ssize_t trace_running_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%x\n", trace_isrunning(&tracer)); +} + +static ssize_t trace_running_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t n) +{ + unsigned int value; + int ret; + + if (sscanf(buf, "%u", &value) != 1) + return -EINVAL; + + mutex_lock(&tracer.mutex); + ret = value ? trace_start(&tracer) : trace_stop(&tracer); + mutex_unlock(&tracer.mutex); + + return ret ? : n; +} + +static struct kobj_attribute trace_running_attr = + __ATTR(trace_running, 0644, trace_running_show, trace_running_store); + +static ssize_t trace_info_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + u32 etb_wa, etb_ra, etb_st, etb_fc, etm_ctrl, etm_st; + int datalen; + + etb_unlock(&tracer); + datalen = etb_getdatalen(&tracer); + etb_wa = etb_readl(&tracer, ETBR_WRITEADDR); + etb_ra = etb_readl(&tracer, ETBR_READADDR); + etb_st = etb_readl(&tracer, ETBR_STATUS); + etb_fc = etb_readl(&tracer, ETBR_FORMATTERCTRL); + etb_lock(&tracer); + + etm_unlock(&tracer); + etm_ctrl = etm_readl(&tracer, ETMR_CTRL); + etm_st = etm_readl(&tracer, ETMR_STATUS); + etm_lock(&tracer); + + return sprintf(buf, "Trace buffer len: %d\nComparator pairs: %d\n" + "ETBR_WRITEADDR:\t%08x\n" + "ETBR_READADDR:\t%08x\n" + "ETBR_STATUS:\t%08x\n" + "ETBR_FORMATTERCTRL:\t%08x\n" + "ETMR_CTRL:\t%08x\n" + "ETMR_STATUS:\t%08x\n", + datalen, + tracer.ncmppairs, + etb_wa, + etb_ra, + etb_st, + etb_fc, + etm_ctrl, + etm_st + ); +} + +static struct kobj_attribute trace_info_attr = + __ATTR(trace_info, 0444, trace_info_show, NULL); + +static ssize_t trace_mode_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%d %d\n", + !!(tracer.flags & TRACER_CYCLE_ACC), + tracer.etm_portsz); +} + +static ssize_t trace_mode_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t n) +{ + unsigned int cycacc, portsz; + + if (sscanf(buf, "%u %u", &cycacc, &portsz) != 2) + return -EINVAL; + + mutex_lock(&tracer.mutex); + if (cycacc) + tracer.flags |= TRACER_CYCLE_ACC; + else + tracer.flags &= ~TRACER_CYCLE_ACC; + + tracer.etm_portsz = portsz & 0x0f; + mutex_unlock(&tracer.mutex); + + return n; +} + +static struct kobj_attribute trace_mode_attr = + __ATTR(trace_mode, 0644, trace_mode_show, trace_mode_store); + +static int __init etm_probe(struct amba_device *dev, struct amba_id *id) +{ + struct tracectx *t = &tracer; + int ret = 0; + + if (t->etm_regs) { + dev_dbg(&dev->dev, "ETM already initialized\n"); + ret = -EBUSY; + goto out; + } + + ret = amba_request_regions(dev, NULL); + if (ret) + goto out; + + t->etm_regs = ioremap_nocache(dev->res.start, resource_size(&dev->res)); + if (!t->etm_regs) { + ret = -ENOMEM; + goto out_release; + } + + amba_set_drvdata(dev, t); + + mutex_init(&t->mutex); + t->dev = &dev->dev; + t->flags = TRACER_CYCLE_ACC; + t->etm_portsz = 1; + + etm_unlock(t); + ret = etm_readl(t, CSCR_PRSR); + + t->ncmppairs = etm_readl(t, ETMR_CONFCODE) & 0xf; + etm_writel(t, 0x440, ETMR_CTRL); + etm_lock(t); + + ret = sysfs_create_file(&dev->dev.kobj, + &trace_running_attr.attr); + if (ret) + goto out_unmap; + + /* failing to create any of these two is not fatal */ + ret = sysfs_create_file(&dev->dev.kobj, &trace_info_attr.attr); + if (ret) + dev_dbg(&dev->dev, "Failed to create trace_info in sysfs\n"); + + ret = sysfs_create_file(&dev->dev.kobj, &trace_mode_attr.attr); + if (ret) + dev_dbg(&dev->dev, "Failed to create trace_mode in sysfs\n"); + + dev_dbg(t->dev, "ETM AMBA driver initialized.\n"); + +out: + return ret; + +out_unmap: + amba_set_drvdata(dev, NULL); + iounmap(t->etm_regs); + +out_release: + amba_release_regions(dev); + + return ret; +} + +static int etm_remove(struct amba_device *dev) +{ + struct tracectx *t = amba_get_drvdata(dev); + + amba_set_drvdata(dev, NULL); + + iounmap(t->etm_regs); + t->etm_regs = NULL; + + amba_release_regions(dev); + + sysfs_remove_file(&dev->dev.kobj, &trace_running_attr.attr); + sysfs_remove_file(&dev->dev.kobj, &trace_info_attr.attr); + sysfs_remove_file(&dev->dev.kobj, &trace_mode_attr.attr); + + return 0; +} + +static struct amba_id etm_ids[] = { + { + .id = 0x0003b921, + .mask = 0x0007ffff, + }, + { 0, 0 }, +}; + +static struct amba_driver etm_driver = { + .drv = { + .name = "etm", + .owner = THIS_MODULE, + }, + .probe = etm_probe, + .remove = etm_remove, + .id_table = etm_ids, +}; + +static int __init etm_init(void) +{ + int retval; + + retval = amba_driver_register(&etb_driver); + if (retval) { + printk(KERN_ERR "Failed to register etb\n"); + return retval; + } + + retval = amba_driver_register(&etm_driver); + if (retval) { + amba_driver_unregister(&etb_driver); + printk(KERN_ERR "Failed to probe etm\n"); + return retval; + } + + /* not being able to install this handler is not fatal */ + (void)register_sysrq_key('v', &sysrq_etm_op); + + return 0; +} + +device_initcall(etm_init); + -- cgit v1.2.2 From 93fd03a8c6728b58879f8af20ffd55d9c32a778b Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 9 Dec 2009 10:02:18 +0000 Subject: ARM: Add an earlyprintk debug console MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch allows an earlyprintk console if CONFIG_DEBUG_LL is enabled, using the printch asm function. The patch is based on the original work by Sascha Hauer. Signed-off-by: Catalin Marinas Cc: Sascha Hauer Acked-by: Uwe Kleine-König Acked-by: Pavel Machek --- arch/arm/kernel/Makefile | 1 + arch/arm/kernel/early_printk.c | 57 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 58 insertions(+) create mode 100644 arch/arm/kernel/early_printk.c (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index 79087dd6d869..d0e40b7d5f42 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -52,5 +52,6 @@ endif head-y := head$(MMUEXT).o obj-$(CONFIG_DEBUG_LL) += debug.o +obj-$(CONFIG_EARLY_PRINTK) += early_printk.o extra-y := $(head-y) init_task.o vmlinux.lds diff --git a/arch/arm/kernel/early_printk.c b/arch/arm/kernel/early_printk.c new file mode 100644 index 000000000000..85aa2b292692 --- /dev/null +++ b/arch/arm/kernel/early_printk.c @@ -0,0 +1,57 @@ +/* + * linux/arch/arm/kernel/early_printk.c + * + * Copyright (C) 2009 Sascha Hauer + * + * 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 + +extern void printch(int); + +static void early_write(const char *s, unsigned n) +{ + while (n-- > 0) { + if (*s == '\n') + printch('\r'); + printch(*s); + s++; + } +} + +static void early_console_write(struct console *con, const char *s, unsigned n) +{ + early_write(s, n); +} + +static struct console early_console = { + .name = "earlycon", + .write = early_console_write, + .flags = CON_PRINTBUFFER | CON_BOOT, + .index = -1, +}; + +asmlinkage void early_printk(const char *fmt, ...) +{ + char buf[512]; + int n; + va_list ap; + + va_start(ap, fmt); + n = vscnprintf(buf, sizeof(buf), fmt, ap); + early_write(buf, n); + va_end(ap); +} + +static int __init setup_early_printk(char *buf) +{ + register_console(&early_console); + return 0; +} + +early_param("earlyprintk", setup_early_printk); -- cgit v1.2.2 From 10b496832efa046ab9222a8cf132681730886db3 Mon Sep 17 00:00:00 2001 From: Varun Swara Date: Wed, 9 Dec 2009 10:02:19 +0000 Subject: ARM: Do not allow the probing of the local timer Since this IRQ descriptor doesn't have an action registered, it is allowed for probing via probe_irq_on/off() and it will be disabled by the latter function. This patch sets the IRQ_NOPROBE status bit for the local timer descriptor. Signed-off-by: Varun Swara Signed-off-by: Catalin Marinas --- arch/arm/kernel/smp_twd.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp_twd.c b/arch/arm/kernel/smp_twd.c index a73a34dccf2a..ea02a7b1c244 100644 --- a/arch/arm/kernel/smp_twd.c +++ b/arch/arm/kernel/smp_twd.c @@ -160,6 +160,7 @@ void __cpuinit twd_timer_setup(struct clock_event_device *clk) /* Make sure our local interrupt controller has this enabled */ local_irq_save(flags); + irq_to_desc(clk->irq)->status |= IRQ_NOPROBE; get_irq_chip(clk->irq)->unmask(clk->irq); local_irq_restore(flags); -- cgit v1.2.2 From 2ea1d13f64efdf49319e86c87d9ba38c30902782 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Nov 2009 13:06:51 -0500 Subject: arm: add arch_mmap_check(), get rid of sys_arm_mremap() Acked-by: Russell King Signed-off-by: Al Viro --- arch/arm/kernel/calls.S | 2 +- arch/arm/kernel/sys_arm.c | 25 ------------------------- 2 files changed, 1 insertion(+), 26 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index f58c1156e779..9314a2d681f1 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -172,7 +172,7 @@ /* 160 */ CALL(sys_sched_get_priority_min) CALL(sys_sched_rr_get_interval) CALL(sys_nanosleep) - CALL(sys_arm_mremap) + CALL(sys_mremap) CALL(sys_setresuid16) /* 165 */ CALL(sys_getresuid16) CALL(sys_ni_syscall) /* vm86 */ diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index 78ecaac65206..3b897444a9bd 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -28,10 +28,6 @@ #include #include -extern unsigned long do_mremap(unsigned long addr, unsigned long old_len, - unsigned long new_len, unsigned long flags, - unsigned long new_addr); - /* common code for old and new mmaps */ inline long do_mmap2( unsigned long addr, unsigned long len, @@ -43,9 +39,6 @@ inline long do_mmap2( flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); - if (flags & MAP_FIXED && addr < FIRST_USER_ADDRESS) - goto out; - error = -EBADF; if (!(flags & MAP_ANONYMOUS)) { file = fget(fd); @@ -89,24 +82,6 @@ out: return error; } -asmlinkage unsigned long -sys_arm_mremap(unsigned long addr, unsigned long old_len, - unsigned long new_len, unsigned long flags, - unsigned long new_addr) -{ - unsigned long ret = -EINVAL; - - if (flags & MREMAP_FIXED && new_addr < FIRST_USER_ADDRESS) - goto out; - - down_write(¤t->mm->mmap_sem); - ret = do_mremap(addr, old_len, new_len, flags, new_addr); - up_write(¤t->mm->mmap_sem); - -out: - return ret; -} - /* * Perform the select(nd, in, out, ex, tv) and mmap() system * calls. -- cgit v1.2.2 From f8b7256096a20436f6d0926747e3ac3d64c81d24 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 30 Nov 2009 17:37:04 -0500 Subject: Unify sys_mmap* New helper - sys_mmap_pgoff(); switch syscalls to using it. Acked-by: David S. Miller Signed-off-by: Al Viro --- arch/arm/kernel/entry-common.S | 4 ++-- arch/arm/kernel/sys_arm.c | 30 +----------------------------- 2 files changed, 3 insertions(+), 31 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index f0fe95b7085d..2c1db77d7848 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -416,12 +416,12 @@ sys_mmap2: tst r5, #PGOFF_MASK moveq r5, r5, lsr #PAGE_SHIFT - 12 streq r5, [sp, #4] - beq do_mmap2 + beq sys_mmap_pgoff mov r0, #-EINVAL mov pc, lr #else str r5, [sp, #4] - b do_mmap2 + b sys_mmap_pgoff #endif ENDPROC(sys_mmap2) diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index 3b897444a9bd..ae4027bd01bd 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -28,34 +28,6 @@ #include #include -/* common code for old and new mmaps */ -inline long do_mmap2( - unsigned long addr, unsigned long len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long pgoff) -{ - int error = -EINVAL; - struct file * file = NULL; - - flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); - - error = -EBADF; - if (!(flags & MAP_ANONYMOUS)) { - file = fget(fd); - if (!file) - goto out; - } - - down_write(¤t->mm->mmap_sem); - error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); - up_write(¤t->mm->mmap_sem); - - if (file) - fput(file); -out: - return error; -} - struct mmap_arg_struct { unsigned long addr; unsigned long len; @@ -77,7 +49,7 @@ asmlinkage int old_mmap(struct mmap_arg_struct __user *arg) if (a.offset & ~PAGE_MASK) goto out; - error = do_mmap2(a.addr, a.len, a.prot, a.flags, a.fd, a.offset >> PAGE_SHIFT); + error = sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, a.offset >> PAGE_SHIFT); out: return error; } -- cgit v1.2.2 From 239007b8440abff689632f50cdf0f2b9e895b534 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 16:46:45 +0100 Subject: genirq: Convert irq_desc.lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- arch/arm/kernel/irq.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c index c9a8619f3856..b7cb45bb91e8 100644 --- a/arch/arm/kernel/irq.c +++ b/arch/arm/kernel/irq.c @@ -69,7 +69,7 @@ int show_interrupts(struct seq_file *p, void *v) } if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); + raw_spin_lock_irqsave(&irq_desc[i].lock, flags); action = irq_desc[i].action; if (!action) goto unlock; @@ -84,7 +84,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); unlock: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); + raw_spin_unlock_irqrestore(&irq_desc[i].lock, flags); } else if (i == NR_IRQS) { #ifdef CONFIG_FIQ show_fiq_list(p, v); @@ -139,7 +139,7 @@ void set_irq_flags(unsigned int irq, unsigned int iflags) } desc = irq_desc + irq; - spin_lock_irqsave(&desc->lock, flags); + raw_spin_lock_irqsave(&desc->lock, flags); desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN; if (iflags & IRQF_VALID) desc->status &= ~IRQ_NOREQUEST; @@ -147,7 +147,7 @@ void set_irq_flags(unsigned int irq, unsigned int iflags) desc->status &= ~IRQ_NOPROBE; if (!(iflags & IRQF_NOAUTOEN)) desc->status &= ~IRQ_NOAUTOEN; - spin_unlock_irqrestore(&desc->lock, flags); + raw_spin_unlock_irqrestore(&desc->lock, flags); } void __init init_IRQ(void) @@ -166,9 +166,9 @@ static void route_irq(struct irq_desc *desc, unsigned int irq, unsigned int cpu) { pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->node, cpu); - spin_lock_irq(&desc->lock); + raw_spin_lock_irq(&desc->lock); desc->chip->set_affinity(irq, cpumask_of(cpu)); - spin_unlock_irq(&desc->lock); + raw_spin_unlock_irq(&desc->lock); } /* -- cgit v1.2.2 From e3f28c1333acfe9e4109b836bc15caf794e94d09 Mon Sep 17 00:00:00 2001 From: Alan Jenkins Date: Sat, 7 Nov 2009 21:03:52 +0000 Subject: ARM: use unified discard definition in linker script Commit 023bf6f "linker script: unify usage of discard definition" changed the linker scripts for all architectures except for ARM. I can find no discussion about this ommision, so here are the changes for ARM. These changes are exactly parallel to the ia64 case. "ia64 is notable because it first throws away some ia64 specific subsections and then include the rest of the sections into the final image, so those sections must be discarded before the inclusion." Not boot-tested. In build testing, the modified linker script generated an identical vmlinux file. [I would like to be able to rely on this unified discard definition. I want to sort the kernel symbol tables to allow faster symbol resolution during module loading. The simplest way appears to be to generate sorted versions from vmlinux.o, link them in to vmlinux, _and discard the original unsorted tables_. This work is driven by my x86 netbook, but it is implemented at a generic level. It is possible it will benefit some ARM systems also.] Signed-off-by: Alan Jenkins Acked-by: Russell King Acked-by-without-testing: Tejun Heo Signed-off-by: Rusty Russell --- arch/arm/kernel/vmlinux.lds.S | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 71151bd87a36..4957e13ef55b 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -65,11 +65,11 @@ SECTIONS __init_end = .; #endif - /DISCARD/ : { /* Exit code and data */ - EXIT_TEXT - EXIT_DATA - *(.exitcall.exit) - *(.discard) + /* + * unwind exit sections must be discarded before the rest of the + * unwind sections get included. + */ + /DISCARD/ : { *(.ARM.exidx.exit.text) *(.ARM.extab.exit.text) #ifndef CONFIG_HOTPLUG_CPU @@ -238,6 +238,9 @@ SECTIONS STABS_DEBUG .comment 0 : { *(.comment) } + + /* Default discards */ + DISCARDS } /* -- cgit v1.2.2 From 3e7b19efe621bcf8bfef896c9c4cc5c99c52c3ec Mon Sep 17 00:00:00 2001 From: Alan Jenkins Date: Sat, 7 Nov 2009 21:03:53 +0000 Subject: ARM: unexport symbols used to implement floating point emulation The Kconfigs for in-tree floating point emulation do not allow building as modules. That leaves the Acorn FPEmulator module. I found two public releases of this as a binary module for 2.1 and 2.2 kernels, optimized for ARMV4.[1] If there is a resurgence of interest in this, the symbols can always be re-exported. This allows the EXPORT_SYMBOL_ALIAS() hack to be removed. The ulterior motive here is that EXPORT_SYMBOL_ALIAS() makes it harder to sort the resulting kernel symbol tables. Sorted symbol tables will allow faster symbol resolution during module loading. Note that fp_send_sigs() and fp_printk() are simply aliases for existing exports and add no obvious value. Similarly fp_enter could easily be renamed to kern_fp_enter at the point of definition. Therefore removing EXPORT_SYMBOL_ALIAS will not serve as a material obstacle to re-adding the exports should they be desired in future. Build tested only. [1] http://ftp.arm.linux.org.uk/pub/linux/arm/fpemulator/ Signed-off-by: Alan Jenkins CC: Russell King Signed-off-by: Rusty Russell --- arch/arm/kernel/armksyms.c | 20 -------------------- 1 file changed, 20 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c index 0e627705f746..8214bfebfaca 100644 --- a/arch/arm/kernel/armksyms.c +++ b/arch/arm/kernel/armksyms.c @@ -48,27 +48,7 @@ extern void __aeabi_uidivmod(void); extern void __aeabi_ulcmp(void); extern void fpundefinstr(void); -extern void fp_enter(void); -/* - * This has a special calling convention; it doesn't - * modify any of the usual registers, except for LR. - */ -#define EXPORT_CRC_ALIAS(sym) __CRC_SYMBOL(sym, "") - -#define EXPORT_SYMBOL_ALIAS(sym,orig) \ - EXPORT_CRC_ALIAS(sym) \ - static const struct kernel_symbol __ksymtab_##sym \ - __used __attribute__((section("__ksymtab"))) = \ - { (unsigned long)&orig, #sym }; - -/* - * floating point math emulator support. - * These symbols will never change their calling convention... - */ -EXPORT_SYMBOL_ALIAS(kern_fp_enter,fp_enter); -EXPORT_SYMBOL_ALIAS(fp_printk,printk); -EXPORT_SYMBOL_ALIAS(fp_send_sig,send_sig); EXPORT_SYMBOL(__backtrace); -- cgit v1.2.2 From 797245f5da543074ee7db0e0516da744c89aa17f Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 18 Dec 2009 14:34:43 +0000 Subject: ARM: Convert VFP/Crunch/XscaleCP thread_release() to exit_thread() This avoids races in the VFP code where the dead thread may have state on another CPU. By moving this code to exit_thread(), we will be running as the thread, and therefore be running on the current CPU. This means that we can ensure that the only local state is accessed in the thread notifiers. Acked-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/kernel/crunch.c | 2 +- arch/arm/kernel/process.c | 12 +++++------- arch/arm/kernel/xscale-cp0.c | 2 +- 3 files changed, 7 insertions(+), 9 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/crunch.c b/arch/arm/kernel/crunch.c index 769abe15cf91..25ef223ba7f3 100644 --- a/arch/arm/kernel/crunch.c +++ b/arch/arm/kernel/crunch.c @@ -51,7 +51,7 @@ static int crunch_do(struct notifier_block *self, unsigned long cmd, void *t) * initialised state information on the first fault. */ - case THREAD_NOTIFY_RELEASE: + case THREAD_NOTIFY_EXIT: crunch_task_release(thread); break; diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 0d96d0171c05..67304138a2ca 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -274,17 +274,18 @@ void show_regs(struct pt_regs * regs) __backtrace(); } +ATOMIC_NOTIFIER_HEAD(thread_notify_head); + +EXPORT_SYMBOL_GPL(thread_notify_head); + /* * Free current thread data structures etc.. */ void exit_thread(void) { + thread_notify(THREAD_NOTIFY_EXIT, current_thread_info()); } -ATOMIC_NOTIFIER_HEAD(thread_notify_head); - -EXPORT_SYMBOL_GPL(thread_notify_head); - void flush_thread(void) { struct thread_info *thread = current_thread_info(); @@ -299,9 +300,6 @@ void flush_thread(void) void release_thread(struct task_struct *dead_task) { - struct thread_info *thread = task_thread_info(dead_task); - - thread_notify(THREAD_NOTIFY_RELEASE, thread); } asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); diff --git a/arch/arm/kernel/xscale-cp0.c b/arch/arm/kernel/xscale-cp0.c index 17127db906fa..1796157e3dd5 100644 --- a/arch/arm/kernel/xscale-cp0.c +++ b/arch/arm/kernel/xscale-cp0.c @@ -70,7 +70,7 @@ static int iwmmxt_do(struct notifier_block *self, unsigned long cmd, void *t) * initialised state information on the first fault. */ - case THREAD_NOTIFY_RELEASE: + case THREAD_NOTIFY_EXIT: iwmmxt_task_release(thread); break; -- cgit v1.2.2 From e8b8f5ef90b934bc9eaef7c8960ecd9dda9f7a62 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 24 Dec 2009 14:45:39 +0000 Subject: ARM: dma-isa: request cascade channel after registering it We can't request the cascade channel before it's been registered, so move it afterwards. Signed-off-by: Russell King --- arch/arm/kernel/dma-isa.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/dma-isa.c b/arch/arm/kernel/dma-isa.c index 0e88e46fc732..360bb6d701f5 100644 --- a/arch/arm/kernel/dma-isa.c +++ b/arch/arm/kernel/dma-isa.c @@ -207,8 +207,6 @@ void __init isa_init_dma(void) outb(0x32, 0x4d6); outb(0x33, 0x4d6); - request_dma(DMA_ISA_CASCADE, "cascade"); - for (i = 0; i < ARRAY_SIZE(dma_resources); i++) request_resource(&ioport_resource, dma_resources + i); @@ -218,5 +216,7 @@ void __init isa_init_dma(void) printk(KERN_ERR "ISADMA%u: unable to register: %d\n", chan, ret); } + + request_dma(DMA_ISA_CASCADE, "cascade"); } } -- cgit v1.2.2 From 22325525d8bb1478daddefec1b762e7882bcd515 Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Fri, 8 Jan 2010 16:59:34 +0100 Subject: ARM: 5868/1: ARM: fix "BUG: using smp_processor_id() in preemptible code" Fix the following warning, which appears when the register dump for a faulting process is printed in a kernel with SMP, DEBUG_PREEMPT, and DEBUG_USER (with user_debug=31) enabled: BUG: using smp_processor_id() in preemptible [00000000] code: init/1 caller is __show_regs+0x18/0x234 Backtrace: [] (dump_backtrace+0x0/0x114) from [] (dump_stack+0x18/0x1c) r6:c781a000 r5:c0157544 r4:00000001 r3:00000000 [] (dump_stack+0x0/0x1c) from [] (debug_smp_processor_id+0xc4/0xf8) [] (debug_smp_processor_id+0x0/0xf8) from [] (__show_regs+0x18/0x234) r6:c781bfb0 r5:00000000 r4:c781bfb0 r3:00000000 [] (__show_regs+0x0/0x234) from [] (show_regs+0x40/0x50) [] (show_regs+0x0/0x50) from [] (__do_user_fault+0x5c/0xa4) r4:c781c000 r3:00000000 [] (__do_user_fault+0x0/0xa4) from [] (do_page_fault+0x1b4/0x1e4) r7:00000000 r6:00010000 r5:c781bfb0 r4:c781c000 [] (do_page_fault+0x0/0x1e4) from [] (do_DataAbort+0x3c/0xa0) [] (do_DataAbort+0x0/0xa0) from [] (ret_from_exception+0x0/0x10) Signed-off-by: Rabin Vincent Signed-off-by: Russell King --- arch/arm/kernel/process.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 67304138a2ca..ba2adefa53f7 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -212,7 +212,8 @@ void __show_regs(struct pt_regs *regs) char buf[64]; printk("CPU: %d %s (%s %.*s)\n", - smp_processor_id(), print_tainted(), init_utsname()->release, + raw_smp_processor_id(), print_tainted(), + init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); print_symbol("PC is at %s\n", instruction_pointer(regs)); -- cgit v1.2.2 From 7511bce4069de39ea04b14c1e1d55c249f9ce808 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 12 Jan 2010 18:59:16 +0000 Subject: ARM: Fix wrong dmb The __kuser_cmpxchg code uses an ARMv6 dmb instruction, rather than one based upon the architecture being built for. Switch to using the macro provided for this purpose, which also eliminates the need for an ifdef. Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index d2903e3bc861..6c5cf369183b 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -957,9 +957,7 @@ kuser_cmpxchg_fixup: #else -#ifdef CONFIG_SMP - mcr p15, 0, r0, c7, c10, 5 @ dmb -#endif + smp_dmb 1: ldrex r3, [r2] subs r3, r3, r0 strexeq r3, r1, [r2] -- cgit v1.2.2 From a7b22962776109b5bfe3a3efa4f42622318c547d Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Mon, 18 Jan 2010 15:04:32 +0100 Subject: ARM: 5883/1: Revert "disable NX support for OABI-supporting kernels" This reverts commit 14f0aa359365e8a93a77b71e3b840274b9b4dcb1. That commit was needed earlier because system call restarting for OABI (compat) required an executable stack and thus had problems with NX. Since ab72b00734ae4d0b ("ARM: Fix signal restart issues with NX and OABI compat") has reworked the code to not require an executable stack anymore, we can re-enable NX support for kernels with OABI (compat) support. Signed-off-by: Rabin Vincent Signed-off-by: Russell King --- arch/arm/kernel/elf.c | 9 --------- 1 file changed, 9 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/elf.c b/arch/arm/kernel/elf.c index 950391f194c4..d4a0da1e48f4 100644 --- a/arch/arm/kernel/elf.c +++ b/arch/arm/kernel/elf.c @@ -78,15 +78,6 @@ int arm_elf_read_implies_exec(const struct elf32_hdr *x, int executable_stack) return 1; if (cpu_architecture() < CPU_ARCH_ARMv6) return 1; -#if !defined(CONFIG_AEABI) || defined(CONFIG_OABI_COMPAT) - /* - * If we have support for OABI programs, we can never allow NX - * support - our signal syscall restart mechanism relies upon - * being able to execute code placed on the user stack. - */ - return 1; -#else return 0; -#endif } EXPORT_SYMBOL(arm_elf_read_implies_exec); -- cgit v1.2.2 From 200b7a8dc09504bc0aedac567a307a6e533f39e5 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 19 Jan 2010 16:40:07 +0100 Subject: ARM: 5884/1: arm: Fix DCC console for v7 Without this patch arch/arm/compressed/head.S defaults to generic DCC code that does not work for v7. For more information on the v7 DCC, see Cortex-A8 TRM "12.11.1 Debug communications channel". To use it with post 2.6.33-rc1 or later, you need to have: CONFIG_DEBUG_LL=y ONFIG_DEBUG_ICEDCC=y CONFIG_EARLY_PRINTK=y Earlier kernels need commit 93fd03a8c6728b58879f8af20ffd55d9c32a778b backported. Tested on omap3430. Signed-off-by: Tony Lindgren Signed-off-by: Russell King --- arch/arm/kernel/debug.S | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/debug.S b/arch/arm/kernel/debug.S index b121b6053cce..5c91addcaebc 100644 --- a/arch/arm/kernel/debug.S +++ b/arch/arm/kernel/debug.S @@ -49,6 +49,26 @@ 1002: .endm +#elif defined(CONFIG_CPU_V7) + + .macro addruart, rx + .endm + + .macro senduart, rd, rx + mcr p14, 0, \rd, c0, c5, 0 + .endm + + .macro busyuart, rd, rx +busy: mrc p14, 0, pc, c0, c1, 0 + bcs busy + .endm + + .macro waituart, rd, rx +wait: mrc p14, 0, pc, c0, c1, 0 + bcs wait + + .endm + #elif defined(CONFIG_CPU_XSCALE) .macro addruart, rx -- cgit v1.2.2 From 0f4f0672ac950c96cffaf84a666d35e817d7c3ca Mon Sep 17 00:00:00 2001 From: Jamie Iles Date: Tue, 2 Feb 2010 20:23:15 +0100 Subject: ARM: 5899/2: arm: provide a mechanism to reserve performance counters To add support for perf events and to allow the hardware counters to be shared with oprofile, we need a way to reserve access to the pmu (performance monitor unit). Platforms with PMU interrupts should register the interrupts in arch/arm/kernel/pmu.c Signed-off-by: Jamie Iles Signed-off-by: Russell King --- arch/arm/kernel/Makefile | 1 + arch/arm/kernel/pmu.c | 103 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 104 insertions(+) create mode 100644 arch/arm/kernel/pmu.c (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index dd00f747e2ad..216890d804c2 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -46,6 +46,7 @@ obj-$(CONFIG_CPU_XSCALE) += xscale-cp0.o obj-$(CONFIG_CPU_XSC3) += xscale-cp0.o obj-$(CONFIG_CPU_MOHAWK) += xscale-cp0.o obj-$(CONFIG_IWMMXT) += iwmmxt.o +obj-$(CONFIG_CPU_HAS_PMU) += pmu.o AFLAGS_iwmmxt.o := -Wa,-mcpu=iwmmxt ifneq ($(CONFIG_ARCH_EBSA110),y) diff --git a/arch/arm/kernel/pmu.c b/arch/arm/kernel/pmu.c new file mode 100644 index 000000000000..a124312e343f --- /dev/null +++ b/arch/arm/kernel/pmu.c @@ -0,0 +1,103 @@ +/* + * linux/arch/arm/kernel/pmu.c + * + * Copyright (C) 2009 picoChip Designs Ltd, Jamie Iles + * + * 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 + +/* + * Define the IRQs for the system. We could use something like a platform + * device but that seems fairly heavyweight for this. Also, the performance + * counters can't be removed or hotplugged. + * + * Ordering is important: init_pmu() will use the ordering to set the affinity + * to the corresponding core. e.g. the first interrupt will go to cpu 0, the + * second goes to cpu 1 etc. + */ +static const int irqs[] = { +#if defined(CONFIG_ARCH_OMAP2) + 3, +#elif defined(CONFIG_ARCH_BCMRING) + IRQ_PMUIRQ, +#elif defined(CONFIG_MACH_REALVIEW_EB) + IRQ_EB11MP_PMU_CPU0, + IRQ_EB11MP_PMU_CPU1, + IRQ_EB11MP_PMU_CPU2, + IRQ_EB11MP_PMU_CPU3, +#elif defined(CONFIG_ARCH_OMAP3) + INT_34XX_BENCH_MPU_EMUL, +#elif defined(CONFIG_ARCH_IOP32X) + IRQ_IOP32X_CORE_PMU, +#elif defined(CONFIG_ARCH_IOP33X) + IRQ_IOP33X_CORE_PMU, +#elif defined(CONFIG_ARCH_PXA) + IRQ_PMU, +#endif +}; + +static const struct pmu_irqs pmu_irqs = { + .irqs = irqs, + .num_irqs = ARRAY_SIZE(irqs), +}; + +static volatile long pmu_lock; + +const struct pmu_irqs * +reserve_pmu(void) +{ + return test_and_set_bit_lock(0, &pmu_lock) ? ERR_PTR(-EBUSY) : + &pmu_irqs; +} +EXPORT_SYMBOL_GPL(reserve_pmu); + +int +release_pmu(const struct pmu_irqs *irqs) +{ + if (WARN_ON(irqs != &pmu_irqs)) + return -EINVAL; + clear_bit_unlock(0, &pmu_lock); + return 0; +} +EXPORT_SYMBOL_GPL(release_pmu); + +static int +set_irq_affinity(int irq, + unsigned int cpu) +{ +#ifdef CONFIG_SMP + int err = irq_set_affinity(irq, cpumask_of(cpu)); + if (err) + pr_warning("unable to set irq affinity (irq=%d, cpu=%u)\n", + irq, cpu); + return err; +#else + return 0; +#endif +} + +int +init_pmu(void) +{ + int i, err = 0; + + for (i = 0; i < pmu_irqs.num_irqs; ++i) { + err = set_irq_affinity(pmu_irqs.irqs[i], i); + if (err) + break; + } + + return err; +} +EXPORT_SYMBOL_GPL(init_pmu); -- cgit v1.2.2 From 1b8873a0c6ec511870c106c80b94658f857c47f2 Mon Sep 17 00:00:00 2001 From: Jamie Iles Date: Tue, 2 Feb 2010 20:25:44 +0100 Subject: ARM: 5902/4: arm/perfevents: implement perf event support for ARMv6 This patch implements support for ARMv6 performance counters in the Linux performance events subsystem. ARMv6 architectures that have the performance counters should enable HW_PERF_EVENTS to get hardware performance events support in addition to the software events. Note: only ARM Ltd ARM cores are supported. This implementation also provides an ARM PMU abstraction layer to allow ARMv7 and others to be supported in the future by adding new a 'struct arm_pmu'. Cc: Jean Pihet Cc: Will Deacon Signed-off-by: Jamie Iles Signed-off-by: Russell King --- arch/arm/kernel/Makefile | 1 + arch/arm/kernel/perf_event.c | 1348 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 1349 insertions(+) create mode 100644 arch/arm/kernel/perf_event.c (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index 216890d804c2..c76e6d2679b8 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -47,6 +47,7 @@ obj-$(CONFIG_CPU_XSC3) += xscale-cp0.o obj-$(CONFIG_CPU_MOHAWK) += xscale-cp0.o obj-$(CONFIG_IWMMXT) += iwmmxt.o obj-$(CONFIG_CPU_HAS_PMU) += pmu.o +obj-$(CONFIG_HW_PERF_EVENTS) += perf_event.o AFLAGS_iwmmxt.o := -Wa,-mcpu=iwmmxt ifneq ($(CONFIG_ARCH_EBSA110),y) diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c new file mode 100644 index 000000000000..7b1022b9aa52 --- /dev/null +++ b/arch/arm/kernel/perf_event.c @@ -0,0 +1,1348 @@ +#undef DEBUG + +/* + * ARM performance counter support. + * + * Copyright (C) 2009 picoChip Designs, Ltd., Jamie Iles + * + * This code is based on the sparc64 perf event code, which is in turn based + * on the x86 code. Callchain code is based on the ARM OProfile backtrace + * code. + */ +#define pr_fmt(fmt) "hw perfevents: " fmt + +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +static const struct pmu_irqs *pmu_irqs; + +/* + * Hardware lock to serialize accesses to PMU registers. Needed for the + * read/modify/write sequences. + */ +DEFINE_SPINLOCK(pmu_lock); + +/* + * ARMv6 supports a maximum of 3 events, starting from index 1. If we add + * another platform that supports more, we need to increase this to be the + * largest of all platforms. + */ +#define ARMPMU_MAX_HWEVENTS 4 + +/* The events for a given CPU. */ +struct cpu_hw_events { + /* + * The events that are active on the CPU for the given index. Index 0 + * is reserved. + */ + struct perf_event *events[ARMPMU_MAX_HWEVENTS]; + + /* + * A 1 bit for an index indicates that the counter is being used for + * an event. A 0 means that the counter can be used. + */ + unsigned long used_mask[BITS_TO_LONGS(ARMPMU_MAX_HWEVENTS)]; + + /* + * A 1 bit for an index indicates that the counter is actively being + * used. + */ + unsigned long active_mask[BITS_TO_LONGS(ARMPMU_MAX_HWEVENTS)]; +}; +DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events); + +struct arm_pmu { + const char *name; + irqreturn_t (*handle_irq)(int irq_num, void *dev); + void (*enable)(struct hw_perf_event *evt, int idx); + void (*disable)(struct hw_perf_event *evt, int idx); + int (*event_map)(int evt); + u64 (*raw_event)(u64); + int (*get_event_idx)(struct cpu_hw_events *cpuc, + struct hw_perf_event *hwc); + u32 (*read_counter)(int idx); + void (*write_counter)(int idx, u32 val); + void (*start)(void); + void (*stop)(void); + int num_events; + u64 max_period; +}; + +/* Set at runtime when we know what CPU type we are. */ +static const struct arm_pmu *armpmu; + +#define HW_OP_UNSUPPORTED 0xFFFF + +#define C(_x) \ + PERF_COUNT_HW_CACHE_##_x + +#define CACHE_OP_UNSUPPORTED 0xFFFF + +static unsigned armpmu_perf_cache_map[PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX]; + +static int +armpmu_map_cache_event(u64 config) +{ + unsigned int cache_type, cache_op, cache_result, ret; + + cache_type = (config >> 0) & 0xff; + if (cache_type >= PERF_COUNT_HW_CACHE_MAX) + return -EINVAL; + + cache_op = (config >> 8) & 0xff; + if (cache_op >= PERF_COUNT_HW_CACHE_OP_MAX) + return -EINVAL; + + cache_result = (config >> 16) & 0xff; + if (cache_result >= PERF_COUNT_HW_CACHE_RESULT_MAX) + return -EINVAL; + + ret = (int)armpmu_perf_cache_map[cache_type][cache_op][cache_result]; + + if (ret == CACHE_OP_UNSUPPORTED) + return -ENOENT; + + return ret; +} + +static int +armpmu_event_set_period(struct perf_event *event, + struct hw_perf_event *hwc, + int idx) +{ + s64 left = atomic64_read(&hwc->period_left); + s64 period = hwc->sample_period; + int ret = 0; + + if (unlikely(left <= -period)) { + left = period; + atomic64_set(&hwc->period_left, left); + hwc->last_period = period; + ret = 1; + } + + if (unlikely(left <= 0)) { + left += period; + atomic64_set(&hwc->period_left, left); + hwc->last_period = period; + ret = 1; + } + + if (left > (s64)armpmu->max_period) + left = armpmu->max_period; + + atomic64_set(&hwc->prev_count, (u64)-left); + + armpmu->write_counter(idx, (u64)(-left) & 0xffffffff); + + perf_event_update_userpage(event); + + return ret; +} + +static u64 +armpmu_event_update(struct perf_event *event, + struct hw_perf_event *hwc, + int idx) +{ + int shift = 64 - 32; + s64 prev_raw_count, new_raw_count; + s64 delta; + +again: + prev_raw_count = atomic64_read(&hwc->prev_count); + new_raw_count = armpmu->read_counter(idx); + + if (atomic64_cmpxchg(&hwc->prev_count, prev_raw_count, + new_raw_count) != prev_raw_count) + goto again; + + delta = (new_raw_count << shift) - (prev_raw_count << shift); + delta >>= shift; + + atomic64_add(delta, &event->count); + atomic64_sub(delta, &hwc->period_left); + + return new_raw_count; +} + +static void +armpmu_disable(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + int idx = hwc->idx; + + WARN_ON(idx < 0); + + clear_bit(idx, cpuc->active_mask); + armpmu->disable(hwc, idx); + + barrier(); + + armpmu_event_update(event, hwc, idx); + cpuc->events[idx] = NULL; + clear_bit(idx, cpuc->used_mask); + + perf_event_update_userpage(event); +} + +static void +armpmu_read(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + /* Don't read disabled counters! */ + if (hwc->idx < 0) + return; + + armpmu_event_update(event, hwc, hwc->idx); +} + +static void +armpmu_unthrottle(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + /* + * Set the period again. Some counters can't be stopped, so when we + * were throttled we simply disabled the IRQ source and the counter + * may have been left counting. If we don't do this step then we may + * get an interrupt too soon or *way* too late if the overflow has + * happened since disabling. + */ + armpmu_event_set_period(event, hwc, hwc->idx); + armpmu->enable(hwc, hwc->idx); +} + +static int +armpmu_enable(struct perf_event *event) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + int idx; + int err = 0; + + /* If we don't have a space for the counter then finish early. */ + idx = armpmu->get_event_idx(cpuc, hwc); + if (idx < 0) { + err = idx; + goto out; + } + + /* + * If there is an event in the counter we are going to use then make + * sure it is disabled. + */ + event->hw.idx = idx; + armpmu->disable(hwc, idx); + cpuc->events[idx] = event; + set_bit(idx, cpuc->active_mask); + + /* Set the period for the event. */ + armpmu_event_set_period(event, hwc, idx); + + /* Enable the event. */ + armpmu->enable(hwc, idx); + + /* Propagate our changes to the userspace mapping. */ + perf_event_update_userpage(event); + +out: + return err; +} + +static struct pmu pmu = { + .enable = armpmu_enable, + .disable = armpmu_disable, + .unthrottle = armpmu_unthrottle, + .read = armpmu_read, +}; + +static int +validate_event(struct cpu_hw_events *cpuc, + struct perf_event *event) +{ + struct hw_perf_event fake_event = event->hw; + + if (event->pmu && event->pmu != &pmu) + return 0; + + return armpmu->get_event_idx(cpuc, &fake_event) >= 0; +} + +static int +validate_group(struct perf_event *event) +{ + struct perf_event *sibling, *leader = event->group_leader; + struct cpu_hw_events fake_pmu; + + memset(&fake_pmu, 0, sizeof(fake_pmu)); + + if (!validate_event(&fake_pmu, leader)) + return -ENOSPC; + + list_for_each_entry(sibling, &leader->sibling_list, group_entry) { + if (!validate_event(&fake_pmu, sibling)) + return -ENOSPC; + } + + if (!validate_event(&fake_pmu, event)) + return -ENOSPC; + + return 0; +} + +static int +armpmu_reserve_hardware(void) +{ + int i; + int err; + + pmu_irqs = reserve_pmu(); + if (IS_ERR(pmu_irqs)) { + pr_warning("unable to reserve pmu\n"); + return PTR_ERR(pmu_irqs); + } + + init_pmu(); + + if (pmu_irqs->num_irqs < 1) { + pr_err("no irqs for PMUs defined\n"); + return -ENODEV; + } + + for (i = 0; i < pmu_irqs->num_irqs; ++i) { + err = request_irq(pmu_irqs->irqs[i], armpmu->handle_irq, + IRQF_DISABLED, "armpmu", NULL); + if (err) { + pr_warning("unable to request IRQ%d for ARM " + "perf counters\n", pmu_irqs->irqs[i]); + break; + } + } + + if (err) { + for (i = i - 1; i >= 0; --i) + free_irq(pmu_irqs->irqs[i], NULL); + release_pmu(pmu_irqs); + pmu_irqs = NULL; + } + + return err; +} + +static void +armpmu_release_hardware(void) +{ + int i; + + for (i = pmu_irqs->num_irqs - 1; i >= 0; --i) + free_irq(pmu_irqs->irqs[i], NULL); + armpmu->stop(); + + release_pmu(pmu_irqs); + pmu_irqs = NULL; +} + +static atomic_t active_events = ATOMIC_INIT(0); +static DEFINE_MUTEX(pmu_reserve_mutex); + +static void +hw_perf_event_destroy(struct perf_event *event) +{ + if (atomic_dec_and_mutex_lock(&active_events, &pmu_reserve_mutex)) { + armpmu_release_hardware(); + mutex_unlock(&pmu_reserve_mutex); + } +} + +static int +__hw_perf_event_init(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + int mapping, err; + + /* Decode the generic type into an ARM event identifier. */ + if (PERF_TYPE_HARDWARE == event->attr.type) { + mapping = armpmu->event_map(event->attr.config); + } else if (PERF_TYPE_HW_CACHE == event->attr.type) { + mapping = armpmu_map_cache_event(event->attr.config); + } else if (PERF_TYPE_RAW == event->attr.type) { + mapping = armpmu->raw_event(event->attr.config); + } else { + pr_debug("event type %x not supported\n", event->attr.type); + return -EOPNOTSUPP; + } + + if (mapping < 0) { + pr_debug("event %x:%llx not supported\n", event->attr.type, + event->attr.config); + return mapping; + } + + /* + * Check whether we need to exclude the counter from certain modes. + * The ARM performance counters are on all of the time so if someone + * has asked us for some excludes then we have to fail. + */ + if (event->attr.exclude_kernel || event->attr.exclude_user || + event->attr.exclude_hv || event->attr.exclude_idle) { + pr_debug("ARM performance counters do not support " + "mode exclusion\n"); + return -EPERM; + } + + /* + * We don't assign an index until we actually place the event onto + * hardware. Use -1 to signify that we haven't decided where to put it + * yet. For SMP systems, each core has it's own PMU so we can't do any + * clever allocation or constraints checking at this point. + */ + hwc->idx = -1; + + /* + * Store the event encoding into the config_base field. config and + * event_base are unused as the only 2 things we need to know are + * the event mapping and the counter to use. The counter to use is + * also the indx and the config_base is the event type. + */ + hwc->config_base = (unsigned long)mapping; + hwc->config = 0; + hwc->event_base = 0; + + if (!hwc->sample_period) { + hwc->sample_period = armpmu->max_period; + hwc->last_period = hwc->sample_period; + atomic64_set(&hwc->period_left, hwc->sample_period); + } + + err = 0; + if (event->group_leader != event) { + err = validate_group(event); + if (err) + return -EINVAL; + } + + return err; +} + +const struct pmu * +hw_perf_event_init(struct perf_event *event) +{ + int err = 0; + + if (!armpmu) + return ERR_PTR(-ENODEV); + + event->destroy = hw_perf_event_destroy; + + if (!atomic_inc_not_zero(&active_events)) { + if (atomic_read(&active_events) > perf_max_events) { + atomic_dec(&active_events); + return ERR_PTR(-ENOSPC); + } + + mutex_lock(&pmu_reserve_mutex); + if (atomic_read(&active_events) == 0) { + err = armpmu_reserve_hardware(); + } + + if (!err) + atomic_inc(&active_events); + mutex_unlock(&pmu_reserve_mutex); + } + + if (err) + return ERR_PTR(err); + + err = __hw_perf_event_init(event); + if (err) + hw_perf_event_destroy(event); + + return err ? ERR_PTR(err) : &pmu; +} + +void +hw_perf_enable(void) +{ + /* Enable all of the perf events on hardware. */ + int idx; + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + if (!armpmu) + return; + + for (idx = 0; idx <= armpmu->num_events; ++idx) { + struct perf_event *event = cpuc->events[idx]; + + if (!event) + continue; + + armpmu->enable(&event->hw, idx); + } + + armpmu->start(); +} + +void +hw_perf_disable(void) +{ + if (armpmu) + armpmu->stop(); +} + +/* + * ARMv6 Performance counter handling code. + * + * ARMv6 has 2 configurable performance counters and a single cycle counter. + * They all share a single reset bit but can be written to zero so we can use + * that for a reset. + * + * The counters can't be individually enabled or disabled so when we remove + * one event and replace it with another we could get spurious counts from the + * wrong event. However, we can take advantage of the fact that the + * performance counters can export events to the event bus, and the event bus + * itself can be monitored. This requires that we *don't* export the events to + * the event bus. The procedure for disabling a configurable counter is: + * - change the counter to count the ETMEXTOUT[0] signal (0x20). This + * effectively stops the counter from counting. + * - disable the counter's interrupt generation (each counter has it's + * own interrupt enable bit). + * Once stopped, the counter value can be written as 0 to reset. + * + * To enable a counter: + * - enable the counter's interrupt generation. + * - set the new event type. + * + * Note: the dedicated cycle counter only counts cycles and can't be + * enabled/disabled independently of the others. When we want to disable the + * cycle counter, we have to just disable the interrupt reporting and start + * ignoring that counter. When re-enabling, we have to reset the value and + * enable the interrupt. + */ + +enum armv6_perf_types { + ARMV6_PERFCTR_ICACHE_MISS = 0x0, + ARMV6_PERFCTR_IBUF_STALL = 0x1, + ARMV6_PERFCTR_DDEP_STALL = 0x2, + ARMV6_PERFCTR_ITLB_MISS = 0x3, + ARMV6_PERFCTR_DTLB_MISS = 0x4, + ARMV6_PERFCTR_BR_EXEC = 0x5, + ARMV6_PERFCTR_BR_MISPREDICT = 0x6, + ARMV6_PERFCTR_INSTR_EXEC = 0x7, + ARMV6_PERFCTR_DCACHE_HIT = 0x9, + ARMV6_PERFCTR_DCACHE_ACCESS = 0xA, + ARMV6_PERFCTR_DCACHE_MISS = 0xB, + ARMV6_PERFCTR_DCACHE_WBACK = 0xC, + ARMV6_PERFCTR_SW_PC_CHANGE = 0xD, + ARMV6_PERFCTR_MAIN_TLB_MISS = 0xF, + ARMV6_PERFCTR_EXPL_D_ACCESS = 0x10, + ARMV6_PERFCTR_LSU_FULL_STALL = 0x11, + ARMV6_PERFCTR_WBUF_DRAINED = 0x12, + ARMV6_PERFCTR_CPU_CYCLES = 0xFF, + ARMV6_PERFCTR_NOP = 0x20, +}; + +enum armv6_counters { + ARMV6_CYCLE_COUNTER = 1, + ARMV6_COUNTER0, + ARMV6_COUNTER1, +}; + +/* + * The hardware events that we support. We do support cache operations but + * we have harvard caches and no way to combine instruction and data + * accesses/misses in hardware. + */ +static const unsigned armv6_perf_map[PERF_COUNT_HW_MAX] = { + [PERF_COUNT_HW_CPU_CYCLES] = ARMV6_PERFCTR_CPU_CYCLES, + [PERF_COUNT_HW_INSTRUCTIONS] = ARMV6_PERFCTR_INSTR_EXEC, + [PERF_COUNT_HW_CACHE_REFERENCES] = HW_OP_UNSUPPORTED, + [PERF_COUNT_HW_CACHE_MISSES] = HW_OP_UNSUPPORTED, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV6_PERFCTR_BR_EXEC, + [PERF_COUNT_HW_BRANCH_MISSES] = ARMV6_PERFCTR_BR_MISPREDICT, + [PERF_COUNT_HW_BUS_CYCLES] = HW_OP_UNSUPPORTED, +}; + +static const unsigned armv6_perf_cache_map[PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = { + [C(L1D)] = { + /* + * The performance counters don't differentiate between read + * and write accesses/misses so this isn't strictly correct, + * but it's the best we can do. Writes and reads get + * combined. + */ + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = ARMV6_PERFCTR_DCACHE_ACCESS, + [C(RESULT_MISS)] = ARMV6_PERFCTR_DCACHE_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = ARMV6_PERFCTR_DCACHE_ACCESS, + [C(RESULT_MISS)] = ARMV6_PERFCTR_DCACHE_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6_PERFCTR_ICACHE_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6_PERFCTR_ICACHE_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(DTLB)] = { + /* + * The ARM performance counters can count micro DTLB misses, + * micro ITLB misses and main TLB misses. There isn't an event + * for TLB misses, so use the micro misses here and if users + * want the main TLB misses they can use a raw counter. + */ + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6_PERFCTR_DTLB_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6_PERFCTR_DTLB_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6_PERFCTR_ITLB_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6_PERFCTR_ITLB_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, +}; + +enum armv6mpcore_perf_types { + ARMV6MPCORE_PERFCTR_ICACHE_MISS = 0x0, + ARMV6MPCORE_PERFCTR_IBUF_STALL = 0x1, + ARMV6MPCORE_PERFCTR_DDEP_STALL = 0x2, + ARMV6MPCORE_PERFCTR_ITLB_MISS = 0x3, + ARMV6MPCORE_PERFCTR_DTLB_MISS = 0x4, + ARMV6MPCORE_PERFCTR_BR_EXEC = 0x5, + ARMV6MPCORE_PERFCTR_BR_NOTPREDICT = 0x6, + ARMV6MPCORE_PERFCTR_BR_MISPREDICT = 0x7, + ARMV6MPCORE_PERFCTR_INSTR_EXEC = 0x8, + ARMV6MPCORE_PERFCTR_DCACHE_RDACCESS = 0xA, + ARMV6MPCORE_PERFCTR_DCACHE_RDMISS = 0xB, + ARMV6MPCORE_PERFCTR_DCACHE_WRACCESS = 0xC, + ARMV6MPCORE_PERFCTR_DCACHE_WRMISS = 0xD, + ARMV6MPCORE_PERFCTR_DCACHE_EVICTION = 0xE, + ARMV6MPCORE_PERFCTR_SW_PC_CHANGE = 0xF, + ARMV6MPCORE_PERFCTR_MAIN_TLB_MISS = 0x10, + ARMV6MPCORE_PERFCTR_EXPL_MEM_ACCESS = 0x11, + ARMV6MPCORE_PERFCTR_LSU_FULL_STALL = 0x12, + ARMV6MPCORE_PERFCTR_WBUF_DRAINED = 0x13, + ARMV6MPCORE_PERFCTR_CPU_CYCLES = 0xFF, +}; + +/* + * The hardware events that we support. We do support cache operations but + * we have harvard caches and no way to combine instruction and data + * accesses/misses in hardware. + */ +static const unsigned armv6mpcore_perf_map[PERF_COUNT_HW_MAX] = { + [PERF_COUNT_HW_CPU_CYCLES] = ARMV6MPCORE_PERFCTR_CPU_CYCLES, + [PERF_COUNT_HW_INSTRUCTIONS] = ARMV6MPCORE_PERFCTR_INSTR_EXEC, + [PERF_COUNT_HW_CACHE_REFERENCES] = HW_OP_UNSUPPORTED, + [PERF_COUNT_HW_CACHE_MISSES] = HW_OP_UNSUPPORTED, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV6MPCORE_PERFCTR_BR_EXEC, + [PERF_COUNT_HW_BRANCH_MISSES] = ARMV6MPCORE_PERFCTR_BR_MISPREDICT, + [PERF_COUNT_HW_BUS_CYCLES] = HW_OP_UNSUPPORTED, +}; + +static const unsigned armv6mpcore_perf_cache_map[PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = { + [C(L1D)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = + ARMV6MPCORE_PERFCTR_DCACHE_RDACCESS, + [C(RESULT_MISS)] = + ARMV6MPCORE_PERFCTR_DCACHE_RDMISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = + ARMV6MPCORE_PERFCTR_DCACHE_WRACCESS, + [C(RESULT_MISS)] = + ARMV6MPCORE_PERFCTR_DCACHE_WRMISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6MPCORE_PERFCTR_ICACHE_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6MPCORE_PERFCTR_ICACHE_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(DTLB)] = { + /* + * The ARM performance counters can count micro DTLB misses, + * micro ITLB misses and main TLB misses. There isn't an event + * for TLB misses, so use the micro misses here and if users + * want the main TLB misses they can use a raw counter. + */ + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6MPCORE_PERFCTR_DTLB_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6MPCORE_PERFCTR_DTLB_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6MPCORE_PERFCTR_ITLB_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV6MPCORE_PERFCTR_ITLB_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, +}; + +static inline unsigned long +armv6_pmcr_read(void) +{ + u32 val; + asm volatile("mrc p15, 0, %0, c15, c12, 0" : "=r"(val)); + return val; +} + +static inline void +armv6_pmcr_write(unsigned long val) +{ + asm volatile("mcr p15, 0, %0, c15, c12, 0" : : "r"(val)); +} + +#define ARMV6_PMCR_ENABLE (1 << 0) +#define ARMV6_PMCR_CTR01_RESET (1 << 1) +#define ARMV6_PMCR_CCOUNT_RESET (1 << 2) +#define ARMV6_PMCR_CCOUNT_DIV (1 << 3) +#define ARMV6_PMCR_COUNT0_IEN (1 << 4) +#define ARMV6_PMCR_COUNT1_IEN (1 << 5) +#define ARMV6_PMCR_CCOUNT_IEN (1 << 6) +#define ARMV6_PMCR_COUNT0_OVERFLOW (1 << 8) +#define ARMV6_PMCR_COUNT1_OVERFLOW (1 << 9) +#define ARMV6_PMCR_CCOUNT_OVERFLOW (1 << 10) +#define ARMV6_PMCR_EVT_COUNT0_SHIFT 20 +#define ARMV6_PMCR_EVT_COUNT0_MASK (0xFF << ARMV6_PMCR_EVT_COUNT0_SHIFT) +#define ARMV6_PMCR_EVT_COUNT1_SHIFT 12 +#define ARMV6_PMCR_EVT_COUNT1_MASK (0xFF << ARMV6_PMCR_EVT_COUNT1_SHIFT) + +#define ARMV6_PMCR_OVERFLOWED_MASK \ + (ARMV6_PMCR_COUNT0_OVERFLOW | ARMV6_PMCR_COUNT1_OVERFLOW | \ + ARMV6_PMCR_CCOUNT_OVERFLOW) + +static inline int +armv6_pmcr_has_overflowed(unsigned long pmcr) +{ + return (pmcr & ARMV6_PMCR_OVERFLOWED_MASK); +} + +static inline int +armv6_pmcr_counter_has_overflowed(unsigned long pmcr, + enum armv6_counters counter) +{ + int ret = 0; + + if (ARMV6_CYCLE_COUNTER == counter) + ret = pmcr & ARMV6_PMCR_CCOUNT_OVERFLOW; + else if (ARMV6_COUNTER0 == counter) + ret = pmcr & ARMV6_PMCR_COUNT0_OVERFLOW; + else if (ARMV6_COUNTER1 == counter) + ret = pmcr & ARMV6_PMCR_COUNT1_OVERFLOW; + else + WARN_ONCE(1, "invalid counter number (%d)\n", counter); + + return ret; +} + +static inline u32 +armv6pmu_read_counter(int counter) +{ + unsigned long value = 0; + + if (ARMV6_CYCLE_COUNTER == counter) + asm volatile("mrc p15, 0, %0, c15, c12, 1" : "=r"(value)); + else if (ARMV6_COUNTER0 == counter) + asm volatile("mrc p15, 0, %0, c15, c12, 2" : "=r"(value)); + else if (ARMV6_COUNTER1 == counter) + asm volatile("mrc p15, 0, %0, c15, c12, 3" : "=r"(value)); + else + WARN_ONCE(1, "invalid counter number (%d)\n", counter); + + return value; +} + +static inline void +armv6pmu_write_counter(int counter, + u32 value) +{ + if (ARMV6_CYCLE_COUNTER == counter) + asm volatile("mcr p15, 0, %0, c15, c12, 1" : : "r"(value)); + else if (ARMV6_COUNTER0 == counter) + asm volatile("mcr p15, 0, %0, c15, c12, 2" : : "r"(value)); + else if (ARMV6_COUNTER1 == counter) + asm volatile("mcr p15, 0, %0, c15, c12, 3" : : "r"(value)); + else + WARN_ONCE(1, "invalid counter number (%d)\n", counter); +} + +void +armv6pmu_enable_event(struct hw_perf_event *hwc, + int idx) +{ + unsigned long val, mask, evt, flags; + + if (ARMV6_CYCLE_COUNTER == idx) { + mask = 0; + evt = ARMV6_PMCR_CCOUNT_IEN; + } else if (ARMV6_COUNTER0 == idx) { + mask = ARMV6_PMCR_EVT_COUNT0_MASK; + evt = (hwc->config_base << ARMV6_PMCR_EVT_COUNT0_SHIFT) | + ARMV6_PMCR_COUNT0_IEN; + } else if (ARMV6_COUNTER1 == idx) { + mask = ARMV6_PMCR_EVT_COUNT1_MASK; + evt = (hwc->config_base << ARMV6_PMCR_EVT_COUNT1_SHIFT) | + ARMV6_PMCR_COUNT1_IEN; + } else { + WARN_ONCE(1, "invalid counter number (%d)\n", idx); + return; + } + + /* + * Mask out the current event and set the counter to count the event + * that we're interested in. + */ + spin_lock_irqsave(&pmu_lock, flags); + val = armv6_pmcr_read(); + val &= ~mask; + val |= evt; + armv6_pmcr_write(val); + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static irqreturn_t +armv6pmu_handle_irq(int irq_num, + void *dev) +{ + unsigned long pmcr = armv6_pmcr_read(); + struct perf_sample_data data; + struct cpu_hw_events *cpuc; + struct pt_regs *regs; + int idx; + + if (!armv6_pmcr_has_overflowed(pmcr)) + return IRQ_NONE; + + regs = get_irq_regs(); + + /* + * The interrupts are cleared by writing the overflow flags back to + * the control register. All of the other bits don't have any effect + * if they are rewritten, so write the whole value back. + */ + armv6_pmcr_write(pmcr); + + data.addr = 0; + + cpuc = &__get_cpu_var(cpu_hw_events); + for (idx = 0; idx <= armpmu->num_events; ++idx) { + struct perf_event *event = cpuc->events[idx]; + struct hw_perf_event *hwc; + + if (!test_bit(idx, cpuc->active_mask)) + continue; + + /* + * We have a single interrupt for all counters. Check that + * each counter has overflowed before we process it. + */ + if (!armv6_pmcr_counter_has_overflowed(pmcr, idx)) + continue; + + hwc = &event->hw; + armpmu_event_update(event, hwc, idx); + data.period = event->hw.last_period; + if (!armpmu_event_set_period(event, hwc, idx)) + continue; + + if (perf_event_overflow(event, 0, &data, regs)) + armpmu->disable(hwc, idx); + } + + /* + * Handle the pending perf events. + * + * Note: this call *must* be run with interrupts enabled. For + * platforms that can have the PMU interrupts raised as a PMI, this + * will not work. + */ + perf_event_do_pending(); + + return IRQ_HANDLED; +} + +static void +armv6pmu_start(void) +{ + unsigned long flags, val; + + spin_lock_irqsave(&pmu_lock, flags); + val = armv6_pmcr_read(); + val |= ARMV6_PMCR_ENABLE; + armv6_pmcr_write(val); + spin_unlock_irqrestore(&pmu_lock, flags); +} + +void +armv6pmu_stop(void) +{ + unsigned long flags, val; + + spin_lock_irqsave(&pmu_lock, flags); + val = armv6_pmcr_read(); + val &= ~ARMV6_PMCR_ENABLE; + armv6_pmcr_write(val); + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static inline int +armv6pmu_event_map(int config) +{ + int mapping = armv6_perf_map[config]; + if (HW_OP_UNSUPPORTED == mapping) + mapping = -EOPNOTSUPP; + return mapping; +} + +static inline int +armv6mpcore_pmu_event_map(int config) +{ + int mapping = armv6mpcore_perf_map[config]; + if (HW_OP_UNSUPPORTED == mapping) + mapping = -EOPNOTSUPP; + return mapping; +} + +static u64 +armv6pmu_raw_event(u64 config) +{ + return config & 0xff; +} + +static int +armv6pmu_get_event_idx(struct cpu_hw_events *cpuc, + struct hw_perf_event *event) +{ + /* Always place a cycle counter into the cycle counter. */ + if (ARMV6_PERFCTR_CPU_CYCLES == event->config_base) { + if (test_and_set_bit(ARMV6_CYCLE_COUNTER, cpuc->used_mask)) + return -EAGAIN; + + return ARMV6_CYCLE_COUNTER; + } else { + /* + * For anything other than a cycle counter, try and use + * counter0 and counter1. + */ + if (!test_and_set_bit(ARMV6_COUNTER1, cpuc->used_mask)) { + return ARMV6_COUNTER1; + } + + if (!test_and_set_bit(ARMV6_COUNTER0, cpuc->used_mask)) { + return ARMV6_COUNTER0; + } + + /* The counters are all in use. */ + return -EAGAIN; + } +} + +static void +armv6pmu_disable_event(struct hw_perf_event *hwc, + int idx) +{ + unsigned long val, mask, evt, flags; + + if (ARMV6_CYCLE_COUNTER == idx) { + mask = ARMV6_PMCR_CCOUNT_IEN; + evt = 0; + } else if (ARMV6_COUNTER0 == idx) { + mask = ARMV6_PMCR_COUNT0_IEN | ARMV6_PMCR_EVT_COUNT0_MASK; + evt = ARMV6_PERFCTR_NOP << ARMV6_PMCR_EVT_COUNT0_SHIFT; + } else if (ARMV6_COUNTER1 == idx) { + mask = ARMV6_PMCR_COUNT1_IEN | ARMV6_PMCR_EVT_COUNT1_MASK; + evt = ARMV6_PERFCTR_NOP << ARMV6_PMCR_EVT_COUNT1_SHIFT; + } else { + WARN_ONCE(1, "invalid counter number (%d)\n", idx); + return; + } + + /* + * Mask out the current event and set the counter to count the number + * of ETM bus signal assertion cycles. The external reporting should + * be disabled and so this should never increment. + */ + spin_lock_irqsave(&pmu_lock, flags); + val = armv6_pmcr_read(); + val &= ~mask; + val |= evt; + armv6_pmcr_write(val); + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static void +armv6mpcore_pmu_disable_event(struct hw_perf_event *hwc, + int idx) +{ + unsigned long val, mask, flags, evt = 0; + + if (ARMV6_CYCLE_COUNTER == idx) { + mask = ARMV6_PMCR_CCOUNT_IEN; + } else if (ARMV6_COUNTER0 == idx) { + mask = ARMV6_PMCR_COUNT0_IEN; + } else if (ARMV6_COUNTER1 == idx) { + mask = ARMV6_PMCR_COUNT1_IEN; + } else { + WARN_ONCE(1, "invalid counter number (%d)\n", idx); + return; + } + + /* + * Unlike UP ARMv6, we don't have a way of stopping the counters. We + * simply disable the interrupt reporting. + */ + spin_lock_irqsave(&pmu_lock, flags); + val = armv6_pmcr_read(); + val &= ~mask; + val |= evt; + armv6_pmcr_write(val); + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static const struct arm_pmu armv6pmu = { + .name = "v6", + .handle_irq = armv6pmu_handle_irq, + .enable = armv6pmu_enable_event, + .disable = armv6pmu_disable_event, + .event_map = armv6pmu_event_map, + .raw_event = armv6pmu_raw_event, + .read_counter = armv6pmu_read_counter, + .write_counter = armv6pmu_write_counter, + .get_event_idx = armv6pmu_get_event_idx, + .start = armv6pmu_start, + .stop = armv6pmu_stop, + .num_events = 3, + .max_period = (1LLU << 32) - 1, +}; + +/* + * ARMv6mpcore is almost identical to single core ARMv6 with the exception + * that some of the events have different enumerations and that there is no + * *hack* to stop the programmable counters. To stop the counters we simply + * disable the interrupt reporting and update the event. When unthrottling we + * reset the period and enable the interrupt reporting. + */ +static const struct arm_pmu armv6mpcore_pmu = { + .name = "v6mpcore", + .handle_irq = armv6pmu_handle_irq, + .enable = armv6pmu_enable_event, + .disable = armv6mpcore_pmu_disable_event, + .event_map = armv6mpcore_pmu_event_map, + .raw_event = armv6pmu_raw_event, + .read_counter = armv6pmu_read_counter, + .write_counter = armv6pmu_write_counter, + .get_event_idx = armv6pmu_get_event_idx, + .start = armv6pmu_start, + .stop = armv6pmu_stop, + .num_events = 3, + .max_period = (1LLU << 32) - 1, +}; + +static int __init +init_hw_perf_events(void) +{ + unsigned long cpuid = read_cpuid_id(); + unsigned long implementor = (cpuid & 0xFF000000) >> 24; + unsigned long part_number = (cpuid & 0xFFF0); + + /* We only support ARM CPUs implemented by ARM at the moment. */ + if (0x41 == implementor) { + switch (part_number) { + case 0xB360: /* ARM1136 */ + case 0xB560: /* ARM1156 */ + case 0xB760: /* ARM1176 */ + armpmu = &armv6pmu; + memcpy(armpmu_perf_cache_map, armv6_perf_cache_map, + sizeof(armv6_perf_cache_map)); + perf_max_events = armv6pmu.num_events; + break; + case 0xB020: /* ARM11mpcore */ + armpmu = &armv6mpcore_pmu; + memcpy(armpmu_perf_cache_map, + armv6mpcore_perf_cache_map, + sizeof(armv6mpcore_perf_cache_map)); + perf_max_events = armv6mpcore_pmu.num_events; + break; + default: + pr_info("no hardware support available\n"); + perf_max_events = -1; + } + } + + if (armpmu) + pr_info("enabled with %s PMU driver\n", + armpmu->name); + + return 0; +} +arch_initcall(init_hw_perf_events); + +/* + * Callchain handling code. + */ +static inline void +callchain_store(struct perf_callchain_entry *entry, + u64 ip) +{ + if (entry->nr < PERF_MAX_STACK_DEPTH) + entry->ip[entry->nr++] = ip; +} + +/* + * The registers we're interested in are at the end of the variable + * length saved register structure. The fp points at the end of this + * structure so the address of this struct is: + * (struct frame_tail *)(xxx->fp)-1 + * + * This code has been adapted from the ARM OProfile support. + */ +struct frame_tail { + struct frame_tail *fp; + unsigned long sp; + unsigned long lr; +} __attribute__((packed)); + +/* + * Get the return address for a single stackframe and return a pointer to the + * next frame tail. + */ +static struct frame_tail * +user_backtrace(struct frame_tail *tail, + struct perf_callchain_entry *entry) +{ + struct frame_tail buftail; + + /* Also check accessibility of one struct frame_tail beyond */ + if (!access_ok(VERIFY_READ, tail, sizeof(buftail))) + return NULL; + if (__copy_from_user_inatomic(&buftail, tail, sizeof(buftail))) + return NULL; + + callchain_store(entry, buftail.lr); + + /* + * Frame pointers should strictly progress back up the stack + * (towards higher addresses). + */ + if (tail >= buftail.fp) + return NULL; + + return buftail.fp - 1; +} + +static void +perf_callchain_user(struct pt_regs *regs, + struct perf_callchain_entry *entry) +{ + struct frame_tail *tail; + + callchain_store(entry, PERF_CONTEXT_USER); + + if (!user_mode(regs)) + regs = task_pt_regs(current); + + tail = (struct frame_tail *)regs->ARM_fp - 1; + + while (tail && !((unsigned long)tail & 0x3)) + tail = user_backtrace(tail, entry); +} + +/* + * Gets called by walk_stackframe() for every stackframe. This will be called + * whist unwinding the stackframe and is like a subroutine return so we use + * the PC. + */ +static int +callchain_trace(struct stackframe *fr, + void *data) +{ + struct perf_callchain_entry *entry = data; + callchain_store(entry, fr->pc); + return 0; +} + +static void +perf_callchain_kernel(struct pt_regs *regs, + struct perf_callchain_entry *entry) +{ + struct stackframe fr; + + callchain_store(entry, PERF_CONTEXT_KERNEL); + fr.fp = regs->ARM_fp; + fr.sp = regs->ARM_sp; + fr.lr = regs->ARM_lr; + fr.pc = regs->ARM_pc; + walk_stackframe(&fr, callchain_trace, entry); +} + +static void +perf_do_callchain(struct pt_regs *regs, + struct perf_callchain_entry *entry) +{ + int is_user; + + if (!regs) + return; + + is_user = user_mode(regs); + + if (!current || !current->pid) + return; + + if (is_user && current->state != TASK_RUNNING) + return; + + if (!is_user) + perf_callchain_kernel(regs, entry); + + if (current->mm) + perf_callchain_user(regs, entry); +} + +static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry); + +struct perf_callchain_entry * +perf_callchain(struct pt_regs *regs) +{ + struct perf_callchain_entry *entry = &__get_cpu_var(pmc_irq_entry); + + entry->nr = 0; + perf_do_callchain(regs, entry); + return entry; +} -- cgit v1.2.2 From 796d12959ad374cae8eb77faaf4243455a305433 Mon Sep 17 00:00:00 2001 From: Jean PIHET Date: Tue, 26 Jan 2010 18:51:05 +0100 Subject: ARM: 5903/1: arm/perfevents: add support for ARMv7 Adds the Performance Events support for ARMv7 processor, using the PMNC unit in HW. Supports the following: - Cortex-A8 and Cortex-A9 processors, - dynamic detection of the number of available counters, based on the PMCR value, - runtime detection of the CPU arch (v6 or v7) and model (Cortex-A8 or Cortex-A9) Tested on OMAP3 (Cortex-A8) only. Signed-off-by: Jean Pihet Signed-off-by: Russell King --- arch/arm/kernel/perf_event.c | 936 ++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 932 insertions(+), 4 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 7b1022b9aa52..c54ceb3d1f97 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -5,6 +5,9 @@ * * Copyright (C) 2009 picoChip Designs, Ltd., Jamie Iles * + * ARMv7 support: Jean Pihet + * 2010 (c) MontaVista Software, LLC. + * * This code is based on the sparc64 perf event code, which is in turn based * on the x86 code. Callchain code is based on the ARM OProfile backtrace * code. @@ -35,8 +38,12 @@ DEFINE_SPINLOCK(pmu_lock); * ARMv6 supports a maximum of 3 events, starting from index 1. If we add * another platform that supports more, we need to increase this to be the * largest of all platforms. + * + * ARMv7 supports up to 32 events: + * cycle counter CCNT + 31 events counters CNT0..30. + * Cortex-A8 has 1+4 counters, Cortex-A9 has 1+6 counters. */ -#define ARMPMU_MAX_HWEVENTS 4 +#define ARMPMU_MAX_HWEVENTS 33 /* The events for a given CPU. */ struct cpu_hw_events { @@ -61,7 +68,7 @@ struct cpu_hw_events { DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events); struct arm_pmu { - const char *name; + char *name; irqreturn_t (*handle_irq)(int irq_num, void *dev); void (*enable)(struct hw_perf_event *evt, int idx); void (*disable)(struct hw_perf_event *evt, int idx); @@ -1174,6 +1181,903 @@ static const struct arm_pmu armv6mpcore_pmu = { .max_period = (1LLU << 32) - 1, }; +/* + * ARMv7 Cortex-A8 and Cortex-A9 Performance Events handling code. + * + * Copied from ARMv6 code, with the low level code inspired + * by the ARMv7 Oprofile code. + * + * Cortex-A8 has up to 4 configurable performance counters and + * a single cycle counter. + * Cortex-A9 has up to 31 configurable performance counters and + * a single cycle counter. + * + * All counters can be enabled/disabled and IRQ masked separately. The cycle + * counter and all 4 performance counters together can be reset separately. + */ + +#define ARMV7_PMU_CORTEX_A8_NAME "ARMv7 Cortex-A8" + +#define ARMV7_PMU_CORTEX_A9_NAME "ARMv7 Cortex-A9" + +/* Common ARMv7 event types */ +enum armv7_perf_types { + ARMV7_PERFCTR_PMNC_SW_INCR = 0x00, + ARMV7_PERFCTR_IFETCH_MISS = 0x01, + ARMV7_PERFCTR_ITLB_MISS = 0x02, + ARMV7_PERFCTR_DCACHE_REFILL = 0x03, + ARMV7_PERFCTR_DCACHE_ACCESS = 0x04, + ARMV7_PERFCTR_DTLB_REFILL = 0x05, + ARMV7_PERFCTR_DREAD = 0x06, + ARMV7_PERFCTR_DWRITE = 0x07, + + ARMV7_PERFCTR_EXC_TAKEN = 0x09, + ARMV7_PERFCTR_EXC_EXECUTED = 0x0A, + ARMV7_PERFCTR_CID_WRITE = 0x0B, + /* ARMV7_PERFCTR_PC_WRITE is equivalent to HW_BRANCH_INSTRUCTIONS. + * It counts: + * - all branch instructions, + * - instructions that explicitly write the PC, + * - exception generating instructions. + */ + ARMV7_PERFCTR_PC_WRITE = 0x0C, + ARMV7_PERFCTR_PC_IMM_BRANCH = 0x0D, + ARMV7_PERFCTR_UNALIGNED_ACCESS = 0x0F, + ARMV7_PERFCTR_PC_BRANCH_MIS_PRED = 0x10, + ARMV7_PERFCTR_CLOCK_CYCLES = 0x11, + + ARMV7_PERFCTR_PC_BRANCH_MIS_USED = 0x12, + + ARMV7_PERFCTR_CPU_CYCLES = 0xFF +}; + +/* ARMv7 Cortex-A8 specific event types */ +enum armv7_a8_perf_types { + ARMV7_PERFCTR_INSTR_EXECUTED = 0x08, + + ARMV7_PERFCTR_PC_PROC_RETURN = 0x0E, + + ARMV7_PERFCTR_WRITE_BUFFER_FULL = 0x40, + ARMV7_PERFCTR_L2_STORE_MERGED = 0x41, + ARMV7_PERFCTR_L2_STORE_BUFF = 0x42, + ARMV7_PERFCTR_L2_ACCESS = 0x43, + ARMV7_PERFCTR_L2_CACH_MISS = 0x44, + ARMV7_PERFCTR_AXI_READ_CYCLES = 0x45, + ARMV7_PERFCTR_AXI_WRITE_CYCLES = 0x46, + ARMV7_PERFCTR_MEMORY_REPLAY = 0x47, + ARMV7_PERFCTR_UNALIGNED_ACCESS_REPLAY = 0x48, + ARMV7_PERFCTR_L1_DATA_MISS = 0x49, + ARMV7_PERFCTR_L1_INST_MISS = 0x4A, + ARMV7_PERFCTR_L1_DATA_COLORING = 0x4B, + ARMV7_PERFCTR_L1_NEON_DATA = 0x4C, + ARMV7_PERFCTR_L1_NEON_CACH_DATA = 0x4D, + ARMV7_PERFCTR_L2_NEON = 0x4E, + ARMV7_PERFCTR_L2_NEON_HIT = 0x4F, + ARMV7_PERFCTR_L1_INST = 0x50, + ARMV7_PERFCTR_PC_RETURN_MIS_PRED = 0x51, + ARMV7_PERFCTR_PC_BRANCH_FAILED = 0x52, + ARMV7_PERFCTR_PC_BRANCH_TAKEN = 0x53, + ARMV7_PERFCTR_PC_BRANCH_EXECUTED = 0x54, + ARMV7_PERFCTR_OP_EXECUTED = 0x55, + ARMV7_PERFCTR_CYCLES_INST_STALL = 0x56, + ARMV7_PERFCTR_CYCLES_INST = 0x57, + ARMV7_PERFCTR_CYCLES_NEON_DATA_STALL = 0x58, + ARMV7_PERFCTR_CYCLES_NEON_INST_STALL = 0x59, + ARMV7_PERFCTR_NEON_CYCLES = 0x5A, + + ARMV7_PERFCTR_PMU0_EVENTS = 0x70, + ARMV7_PERFCTR_PMU1_EVENTS = 0x71, + ARMV7_PERFCTR_PMU_EVENTS = 0x72, +}; + +/* ARMv7 Cortex-A9 specific event types */ +enum armv7_a9_perf_types { + ARMV7_PERFCTR_JAVA_HW_BYTECODE_EXEC = 0x40, + ARMV7_PERFCTR_JAVA_SW_BYTECODE_EXEC = 0x41, + ARMV7_PERFCTR_JAZELLE_BRANCH_EXEC = 0x42, + + ARMV7_PERFCTR_COHERENT_LINE_MISS = 0x50, + ARMV7_PERFCTR_COHERENT_LINE_HIT = 0x51, + + ARMV7_PERFCTR_ICACHE_DEP_STALL_CYCLES = 0x60, + ARMV7_PERFCTR_DCACHE_DEP_STALL_CYCLES = 0x61, + ARMV7_PERFCTR_TLB_MISS_DEP_STALL_CYCLES = 0x62, + ARMV7_PERFCTR_STREX_EXECUTED_PASSED = 0x63, + ARMV7_PERFCTR_STREX_EXECUTED_FAILED = 0x64, + ARMV7_PERFCTR_DATA_EVICTION = 0x65, + ARMV7_PERFCTR_ISSUE_STAGE_NO_INST = 0x66, + ARMV7_PERFCTR_ISSUE_STAGE_EMPTY = 0x67, + ARMV7_PERFCTR_INST_OUT_OF_RENAME_STAGE = 0x68, + + ARMV7_PERFCTR_PREDICTABLE_FUNCT_RETURNS = 0x6E, + + ARMV7_PERFCTR_MAIN_UNIT_EXECUTED_INST = 0x70, + ARMV7_PERFCTR_SECOND_UNIT_EXECUTED_INST = 0x71, + ARMV7_PERFCTR_LD_ST_UNIT_EXECUTED_INST = 0x72, + ARMV7_PERFCTR_FP_EXECUTED_INST = 0x73, + ARMV7_PERFCTR_NEON_EXECUTED_INST = 0x74, + + ARMV7_PERFCTR_PLD_FULL_DEP_STALL_CYCLES = 0x80, + ARMV7_PERFCTR_DATA_WR_DEP_STALL_CYCLES = 0x81, + ARMV7_PERFCTR_ITLB_MISS_DEP_STALL_CYCLES = 0x82, + ARMV7_PERFCTR_DTLB_MISS_DEP_STALL_CYCLES = 0x83, + ARMV7_PERFCTR_MICRO_ITLB_MISS_DEP_STALL_CYCLES = 0x84, + ARMV7_PERFCTR_MICRO_DTLB_MISS_DEP_STALL_CYCLES = 0x85, + ARMV7_PERFCTR_DMB_DEP_STALL_CYCLES = 0x86, + + ARMV7_PERFCTR_INTGR_CLK_ENABLED_CYCLES = 0x8A, + ARMV7_PERFCTR_DATA_ENGINE_CLK_EN_CYCLES = 0x8B, + + ARMV7_PERFCTR_ISB_INST = 0x90, + ARMV7_PERFCTR_DSB_INST = 0x91, + ARMV7_PERFCTR_DMB_INST = 0x92, + ARMV7_PERFCTR_EXT_INTERRUPTS = 0x93, + + ARMV7_PERFCTR_PLE_CACHE_LINE_RQST_COMPLETED = 0xA0, + ARMV7_PERFCTR_PLE_CACHE_LINE_RQST_SKIPPED = 0xA1, + ARMV7_PERFCTR_PLE_FIFO_FLUSH = 0xA2, + ARMV7_PERFCTR_PLE_RQST_COMPLETED = 0xA3, + ARMV7_PERFCTR_PLE_FIFO_OVERFLOW = 0xA4, + ARMV7_PERFCTR_PLE_RQST_PROG = 0xA5 +}; + +/* + * Cortex-A8 HW events mapping + * + * The hardware events that we support. We do support cache operations but + * we have harvard caches and no way to combine instruction and data + * accesses/misses in hardware. + */ +static const unsigned armv7_a8_perf_map[PERF_COUNT_HW_MAX] = { + [PERF_COUNT_HW_CPU_CYCLES] = ARMV7_PERFCTR_CPU_CYCLES, + [PERF_COUNT_HW_INSTRUCTIONS] = ARMV7_PERFCTR_INSTR_EXECUTED, + [PERF_COUNT_HW_CACHE_REFERENCES] = HW_OP_UNSUPPORTED, + [PERF_COUNT_HW_CACHE_MISSES] = HW_OP_UNSUPPORTED, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE, + [PERF_COUNT_HW_BRANCH_MISSES] = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED, + [PERF_COUNT_HW_BUS_CYCLES] = ARMV7_PERFCTR_CLOCK_CYCLES, +}; + +static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = { + [C(L1D)] = { + /* + * The performance counters don't differentiate between read + * and write accesses/misses so this isn't strictly correct, + * but it's the best we can do. Writes and reads get + * combined. + */ + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_DCACHE_ACCESS, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DCACHE_REFILL, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_DCACHE_ACCESS, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DCACHE_REFILL, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_L1_INST, + [C(RESULT_MISS)] = ARMV7_PERFCTR_L1_INST_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_L1_INST, + [C(RESULT_MISS)] = ARMV7_PERFCTR_L1_INST_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_L2_ACCESS, + [C(RESULT_MISS)] = ARMV7_PERFCTR_L2_CACH_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_L2_ACCESS, + [C(RESULT_MISS)] = ARMV7_PERFCTR_L2_CACH_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(DTLB)] = { + /* + * Only ITLB misses and DTLB refills are supported. + * If users want the DTLB refills misses a raw counter + * must be used. + */ + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DTLB_REFILL, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DTLB_REFILL, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_ITLB_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_ITLB_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_PC_WRITE, + [C(RESULT_MISS)] + = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_PC_WRITE, + [C(RESULT_MISS)] + = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, +}; + +/* + * Cortex-A9 HW events mapping + */ +static const unsigned armv7_a9_perf_map[PERF_COUNT_HW_MAX] = { + [PERF_COUNT_HW_CPU_CYCLES] = ARMV7_PERFCTR_CPU_CYCLES, + [PERF_COUNT_HW_INSTRUCTIONS] = + ARMV7_PERFCTR_INST_OUT_OF_RENAME_STAGE, + [PERF_COUNT_HW_CACHE_REFERENCES] = ARMV7_PERFCTR_COHERENT_LINE_HIT, + [PERF_COUNT_HW_CACHE_MISSES] = ARMV7_PERFCTR_COHERENT_LINE_MISS, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE, + [PERF_COUNT_HW_BRANCH_MISSES] = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED, + [PERF_COUNT_HW_BUS_CYCLES] = ARMV7_PERFCTR_CLOCK_CYCLES, +}; + +static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = { + [C(L1D)] = { + /* + * The performance counters don't differentiate between read + * and write accesses/misses so this isn't strictly correct, + * but it's the best we can do. Writes and reads get + * combined. + */ + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_DCACHE_ACCESS, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DCACHE_REFILL, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_DCACHE_ACCESS, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DCACHE_REFILL, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_IFETCH_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_IFETCH_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(DTLB)] = { + /* + * Only ITLB misses and DTLB refills are supported. + * If users want the DTLB refills misses a raw counter + * must be used. + */ + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DTLB_REFILL, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_DTLB_REFILL, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_ITLB_MISS, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = ARMV7_PERFCTR_ITLB_MISS, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, + [C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_PC_WRITE, + [C(RESULT_MISS)] + = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = ARMV7_PERFCTR_PC_WRITE, + [C(RESULT_MISS)] + = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = CACHE_OP_UNSUPPORTED, + [C(RESULT_MISS)] = CACHE_OP_UNSUPPORTED, + }, + }, +}; + +/* + * Perf Events counters + */ +enum armv7_counters { + ARMV7_CYCLE_COUNTER = 1, /* Cycle counter */ + ARMV7_COUNTER0 = 2, /* First event counter */ +}; + +/* + * The cycle counter is ARMV7_CYCLE_COUNTER. + * The first event counter is ARMV7_COUNTER0. + * The last event counter is (ARMV7_COUNTER0 + armpmu->num_events - 1). + */ +#define ARMV7_COUNTER_LAST (ARMV7_COUNTER0 + armpmu->num_events - 1) + +/* + * ARMv7 low level PMNC access + */ + +/* + * Per-CPU PMNC: config reg + */ +#define ARMV7_PMNC_E (1 << 0) /* Enable all counters */ +#define ARMV7_PMNC_P (1 << 1) /* Reset all counters */ +#define ARMV7_PMNC_C (1 << 2) /* Cycle counter reset */ +#define ARMV7_PMNC_D (1 << 3) /* CCNT counts every 64th cpu cycle */ +#define ARMV7_PMNC_X (1 << 4) /* Export to ETM */ +#define ARMV7_PMNC_DP (1 << 5) /* Disable CCNT if non-invasive debug*/ +#define ARMV7_PMNC_N_SHIFT 11 /* Number of counters supported */ +#define ARMV7_PMNC_N_MASK 0x1f +#define ARMV7_PMNC_MASK 0x3f /* Mask for writable bits */ + +/* + * Available counters + */ +#define ARMV7_CNT0 0 /* First event counter */ +#define ARMV7_CCNT 31 /* Cycle counter */ + +/* Perf Event to low level counters mapping */ +#define ARMV7_EVENT_CNT_TO_CNTx (ARMV7_COUNTER0 - ARMV7_CNT0) + +/* + * CNTENS: counters enable reg + */ +#define ARMV7_CNTENS_P(idx) (1 << (idx - ARMV7_EVENT_CNT_TO_CNTx)) +#define ARMV7_CNTENS_C (1 << ARMV7_CCNT) + +/* + * CNTENC: counters disable reg + */ +#define ARMV7_CNTENC_P(idx) (1 << (idx - ARMV7_EVENT_CNT_TO_CNTx)) +#define ARMV7_CNTENC_C (1 << ARMV7_CCNT) + +/* + * INTENS: counters overflow interrupt enable reg + */ +#define ARMV7_INTENS_P(idx) (1 << (idx - ARMV7_EVENT_CNT_TO_CNTx)) +#define ARMV7_INTENS_C (1 << ARMV7_CCNT) + +/* + * INTENC: counters overflow interrupt disable reg + */ +#define ARMV7_INTENC_P(idx) (1 << (idx - ARMV7_EVENT_CNT_TO_CNTx)) +#define ARMV7_INTENC_C (1 << ARMV7_CCNT) + +/* + * EVTSEL: Event selection reg + */ +#define ARMV7_EVTSEL_MASK 0x7f /* Mask for writable bits */ + +/* + * SELECT: Counter selection reg + */ +#define ARMV7_SELECT_MASK 0x1f /* Mask for writable bits */ + +/* + * FLAG: counters overflow flag status reg + */ +#define ARMV7_FLAG_P(idx) (1 << (idx - ARMV7_EVENT_CNT_TO_CNTx)) +#define ARMV7_FLAG_C (1 << ARMV7_CCNT) +#define ARMV7_FLAG_MASK 0xffffffff /* Mask for writable bits */ +#define ARMV7_OVERFLOWED_MASK ARMV7_FLAG_MASK + +static inline unsigned long armv7_pmnc_read(void) +{ + u32 val; + asm volatile("mrc p15, 0, %0, c9, c12, 0" : "=r"(val)); + return val; +} + +static inline void armv7_pmnc_write(unsigned long val) +{ + val &= ARMV7_PMNC_MASK; + asm volatile("mcr p15, 0, %0, c9, c12, 0" : : "r"(val)); +} + +static inline int armv7_pmnc_has_overflowed(unsigned long pmnc) +{ + return pmnc & ARMV7_OVERFLOWED_MASK; +} + +static inline int armv7_pmnc_counter_has_overflowed(unsigned long pmnc, + enum armv7_counters counter) +{ + int ret; + + if (counter == ARMV7_CYCLE_COUNTER) + ret = pmnc & ARMV7_FLAG_C; + else if ((counter >= ARMV7_COUNTER0) && (counter <= ARMV7_COUNTER_LAST)) + ret = pmnc & ARMV7_FLAG_P(counter); + else + pr_err("CPU%u checking wrong counter %d overflow status\n", + smp_processor_id(), counter); + + return ret; +} + +static inline int armv7_pmnc_select_counter(unsigned int idx) +{ + u32 val; + + if ((idx < ARMV7_COUNTER0) || (idx > ARMV7_COUNTER_LAST)) { + pr_err("CPU%u selecting wrong PMNC counter" + " %d\n", smp_processor_id(), idx); + return -1; + } + + val = (idx - ARMV7_EVENT_CNT_TO_CNTx) & ARMV7_SELECT_MASK; + asm volatile("mcr p15, 0, %0, c9, c12, 5" : : "r" (val)); + + return idx; +} + +static inline u32 armv7pmu_read_counter(int idx) +{ + unsigned long value = 0; + + if (idx == ARMV7_CYCLE_COUNTER) + asm volatile("mrc p15, 0, %0, c9, c13, 0" : "=r" (value)); + else if ((idx >= ARMV7_COUNTER0) && (idx <= ARMV7_COUNTER_LAST)) { + if (armv7_pmnc_select_counter(idx) == idx) + asm volatile("mrc p15, 0, %0, c9, c13, 2" + : "=r" (value)); + } else + pr_err("CPU%u reading wrong counter %d\n", + smp_processor_id(), idx); + + return value; +} + +static inline void armv7pmu_write_counter(int idx, u32 value) +{ + if (idx == ARMV7_CYCLE_COUNTER) + asm volatile("mcr p15, 0, %0, c9, c13, 0" : : "r" (value)); + else if ((idx >= ARMV7_COUNTER0) && (idx <= ARMV7_COUNTER_LAST)) { + if (armv7_pmnc_select_counter(idx) == idx) + asm volatile("mcr p15, 0, %0, c9, c13, 2" + : : "r" (value)); + } else + pr_err("CPU%u writing wrong counter %d\n", + smp_processor_id(), idx); +} + +static inline void armv7_pmnc_write_evtsel(unsigned int idx, u32 val) +{ + if (armv7_pmnc_select_counter(idx) == idx) { + val &= ARMV7_EVTSEL_MASK; + asm volatile("mcr p15, 0, %0, c9, c13, 1" : : "r" (val)); + } +} + +static inline u32 armv7_pmnc_enable_counter(unsigned int idx) +{ + u32 val; + + if ((idx != ARMV7_CYCLE_COUNTER) && + ((idx < ARMV7_COUNTER0) || (idx > ARMV7_COUNTER_LAST))) { + pr_err("CPU%u enabling wrong PMNC counter" + " %d\n", smp_processor_id(), idx); + return -1; + } + + if (idx == ARMV7_CYCLE_COUNTER) + val = ARMV7_CNTENS_C; + else + val = ARMV7_CNTENS_P(idx); + + asm volatile("mcr p15, 0, %0, c9, c12, 1" : : "r" (val)); + + return idx; +} + +static inline u32 armv7_pmnc_disable_counter(unsigned int idx) +{ + u32 val; + + + if ((idx != ARMV7_CYCLE_COUNTER) && + ((idx < ARMV7_COUNTER0) || (idx > ARMV7_COUNTER_LAST))) { + pr_err("CPU%u disabling wrong PMNC counter" + " %d\n", smp_processor_id(), idx); + return -1; + } + + if (idx == ARMV7_CYCLE_COUNTER) + val = ARMV7_CNTENC_C; + else + val = ARMV7_CNTENC_P(idx); + + asm volatile("mcr p15, 0, %0, c9, c12, 2" : : "r" (val)); + + return idx; +} + +static inline u32 armv7_pmnc_enable_intens(unsigned int idx) +{ + u32 val; + + if ((idx != ARMV7_CYCLE_COUNTER) && + ((idx < ARMV7_COUNTER0) || (idx > ARMV7_COUNTER_LAST))) { + pr_err("CPU%u enabling wrong PMNC counter" + " interrupt enable %d\n", smp_processor_id(), idx); + return -1; + } + + if (idx == ARMV7_CYCLE_COUNTER) + val = ARMV7_INTENS_C; + else + val = ARMV7_INTENS_P(idx); + + asm volatile("mcr p15, 0, %0, c9, c14, 1" : : "r" (val)); + + return idx; +} + +static inline u32 armv7_pmnc_disable_intens(unsigned int idx) +{ + u32 val; + + if ((idx != ARMV7_CYCLE_COUNTER) && + ((idx < ARMV7_COUNTER0) || (idx > ARMV7_COUNTER_LAST))) { + pr_err("CPU%u disabling wrong PMNC counter" + " interrupt enable %d\n", smp_processor_id(), idx); + return -1; + } + + if (idx == ARMV7_CYCLE_COUNTER) + val = ARMV7_INTENC_C; + else + val = ARMV7_INTENC_P(idx); + + asm volatile("mcr p15, 0, %0, c9, c14, 2" : : "r" (val)); + + return idx; +} + +static inline u32 armv7_pmnc_getreset_flags(void) +{ + u32 val; + + /* Read */ + asm volatile("mrc p15, 0, %0, c9, c12, 3" : "=r" (val)); + + /* Write to clear flags */ + val &= ARMV7_FLAG_MASK; + asm volatile("mcr p15, 0, %0, c9, c12, 3" : : "r" (val)); + + return val; +} + +#ifdef DEBUG +static void armv7_pmnc_dump_regs(void) +{ + u32 val; + unsigned int cnt; + + printk(KERN_INFO "PMNC registers dump:\n"); + + asm volatile("mrc p15, 0, %0, c9, c12, 0" : "=r" (val)); + printk(KERN_INFO "PMNC =0x%08x\n", val); + + asm volatile("mrc p15, 0, %0, c9, c12, 1" : "=r" (val)); + printk(KERN_INFO "CNTENS=0x%08x\n", val); + + asm volatile("mrc p15, 0, %0, c9, c14, 1" : "=r" (val)); + printk(KERN_INFO "INTENS=0x%08x\n", val); + + asm volatile("mrc p15, 0, %0, c9, c12, 3" : "=r" (val)); + printk(KERN_INFO "FLAGS =0x%08x\n", val); + + asm volatile("mrc p15, 0, %0, c9, c12, 5" : "=r" (val)); + printk(KERN_INFO "SELECT=0x%08x\n", val); + + asm volatile("mrc p15, 0, %0, c9, c13, 0" : "=r" (val)); + printk(KERN_INFO "CCNT =0x%08x\n", val); + + for (cnt = ARMV7_COUNTER0; cnt < ARMV7_COUNTER_LAST; cnt++) { + armv7_pmnc_select_counter(cnt); + asm volatile("mrc p15, 0, %0, c9, c13, 2" : "=r" (val)); + printk(KERN_INFO "CNT[%d] count =0x%08x\n", + cnt-ARMV7_EVENT_CNT_TO_CNTx, val); + asm volatile("mrc p15, 0, %0, c9, c13, 1" : "=r" (val)); + printk(KERN_INFO "CNT[%d] evtsel=0x%08x\n", + cnt-ARMV7_EVENT_CNT_TO_CNTx, val); + } +} +#endif + +void armv7pmu_enable_event(struct hw_perf_event *hwc, int idx) +{ + unsigned long flags; + + /* + * Enable counter and interrupt, and set the counter to count + * the event that we're interested in. + */ + spin_lock_irqsave(&pmu_lock, flags); + + /* + * Disable counter + */ + armv7_pmnc_disable_counter(idx); + + /* + * Set event (if destined for PMNx counters) + * We don't need to set the event if it's a cycle count + */ + if (idx != ARMV7_CYCLE_COUNTER) + armv7_pmnc_write_evtsel(idx, hwc->config_base); + + /* + * Enable interrupt for this counter + */ + armv7_pmnc_enable_intens(idx); + + /* + * Enable counter + */ + armv7_pmnc_enable_counter(idx); + + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static void armv7pmu_disable_event(struct hw_perf_event *hwc, int idx) +{ + unsigned long flags; + + /* + * Disable counter and interrupt + */ + spin_lock_irqsave(&pmu_lock, flags); + + /* + * Disable counter + */ + armv7_pmnc_disable_counter(idx); + + /* + * Disable interrupt for this counter + */ + armv7_pmnc_disable_intens(idx); + + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev) +{ + unsigned long pmnc; + struct perf_sample_data data; + struct cpu_hw_events *cpuc; + struct pt_regs *regs; + int idx; + + /* + * Get and reset the IRQ flags + */ + pmnc = armv7_pmnc_getreset_flags(); + + /* + * Did an overflow occur? + */ + if (!armv7_pmnc_has_overflowed(pmnc)) + return IRQ_NONE; + + /* + * Handle the counter(s) overflow(s) + */ + regs = get_irq_regs(); + + data.addr = 0; + + cpuc = &__get_cpu_var(cpu_hw_events); + for (idx = 0; idx <= armpmu->num_events; ++idx) { + struct perf_event *event = cpuc->events[idx]; + struct hw_perf_event *hwc; + + if (!test_bit(idx, cpuc->active_mask)) + continue; + + /* + * We have a single interrupt for all counters. Check that + * each counter has overflowed before we process it. + */ + if (!armv7_pmnc_counter_has_overflowed(pmnc, idx)) + continue; + + hwc = &event->hw; + armpmu_event_update(event, hwc, idx); + data.period = event->hw.last_period; + if (!armpmu_event_set_period(event, hwc, idx)) + continue; + + if (perf_event_overflow(event, 0, &data, regs)) + armpmu->disable(hwc, idx); + } + + /* + * Handle the pending perf events. + * + * Note: this call *must* be run with interrupts enabled. For + * platforms that can have the PMU interrupts raised as a PMI, this + * will not work. + */ + perf_event_do_pending(); + + return IRQ_HANDLED; +} + +static void armv7pmu_start(void) +{ + unsigned long flags; + + spin_lock_irqsave(&pmu_lock, flags); + /* Enable all counters */ + armv7_pmnc_write(armv7_pmnc_read() | ARMV7_PMNC_E); + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static void armv7pmu_stop(void) +{ + unsigned long flags; + + spin_lock_irqsave(&pmu_lock, flags); + /* Disable all counters */ + armv7_pmnc_write(armv7_pmnc_read() & ~ARMV7_PMNC_E); + spin_unlock_irqrestore(&pmu_lock, flags); +} + +static inline int armv7_a8_pmu_event_map(int config) +{ + int mapping = armv7_a8_perf_map[config]; + if (HW_OP_UNSUPPORTED == mapping) + mapping = -EOPNOTSUPP; + return mapping; +} + +static inline int armv7_a9_pmu_event_map(int config) +{ + int mapping = armv7_a9_perf_map[config]; + if (HW_OP_UNSUPPORTED == mapping) + mapping = -EOPNOTSUPP; + return mapping; +} + +static u64 armv7pmu_raw_event(u64 config) +{ + return config & 0xff; +} + +static int armv7pmu_get_event_idx(struct cpu_hw_events *cpuc, + struct hw_perf_event *event) +{ + int idx; + + /* Always place a cycle counter into the cycle counter. */ + if (event->config_base == ARMV7_PERFCTR_CPU_CYCLES) { + if (test_and_set_bit(ARMV7_CYCLE_COUNTER, cpuc->used_mask)) + return -EAGAIN; + + return ARMV7_CYCLE_COUNTER; + } else { + /* + * For anything other than a cycle counter, try and use + * the events counters + */ + for (idx = ARMV7_COUNTER0; idx <= armpmu->num_events; ++idx) { + if (!test_and_set_bit(idx, cpuc->used_mask)) + return idx; + } + + /* The counters are all in use. */ + return -EAGAIN; + } +} + +static struct arm_pmu armv7pmu = { + .handle_irq = armv7pmu_handle_irq, + .enable = armv7pmu_enable_event, + .disable = armv7pmu_disable_event, + .raw_event = armv7pmu_raw_event, + .read_counter = armv7pmu_read_counter, + .write_counter = armv7pmu_write_counter, + .get_event_idx = armv7pmu_get_event_idx, + .start = armv7pmu_start, + .stop = armv7pmu_stop, + .max_period = (1LLU << 32) - 1, +}; + +static u32 __init armv7_reset_read_pmnc(void) +{ + u32 nb_cnt; + + /* Initialize & Reset PMNC: C and P bits */ + armv7_pmnc_write(ARMV7_PMNC_P | ARMV7_PMNC_C); + + /* Read the nb of CNTx counters supported from PMNC */ + nb_cnt = (armv7_pmnc_read() >> ARMV7_PMNC_N_SHIFT) & ARMV7_PMNC_N_MASK; + + /* Add the CPU cycles counter and return */ + return nb_cnt + 1; +} + static int __init init_hw_perf_events(void) { @@ -1199,6 +2103,30 @@ init_hw_perf_events(void) sizeof(armv6mpcore_perf_cache_map)); perf_max_events = armv6mpcore_pmu.num_events; break; + case 0xC080: /* Cortex-A8 */ + armv7pmu.name = ARMV7_PMU_CORTEX_A8_NAME; + memcpy(armpmu_perf_cache_map, armv7_a8_perf_cache_map, + sizeof(armv7_a8_perf_cache_map)); + armv7pmu.event_map = armv7_a8_pmu_event_map; + armpmu = &armv7pmu; + + /* Reset PMNC and read the nb of CNTx counters + supported */ + armv7pmu.num_events = armv7_reset_read_pmnc(); + perf_max_events = armv7pmu.num_events; + break; + case 0xC090: /* Cortex-A9 */ + armv7pmu.name = ARMV7_PMU_CORTEX_A9_NAME; + memcpy(armpmu_perf_cache_map, armv7_a9_perf_cache_map, + sizeof(armv7_a9_perf_cache_map)); + armv7pmu.event_map = armv7_a9_pmu_event_map; + armpmu = &armv7pmu; + + /* Reset PMNC and read the nb of CNTx counters + supported */ + armv7pmu.num_events = armv7_reset_read_pmnc(); + perf_max_events = armv7pmu.num_events; + break; default: pr_info("no hardware support available\n"); perf_max_events = -1; @@ -1206,8 +2134,8 @@ init_hw_perf_events(void) } if (armpmu) - pr_info("enabled with %s PMU driver\n", - armpmu->name); + pr_info("enabled with %s PMU driver, %d counters available\n", + armpmu->name, armpmu->num_events); return 0; } -- cgit v1.2.2 From 4e6d488af37980d224cbf298224db6173673f362 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Mon, 1 Feb 2010 23:26:53 +0100 Subject: ARM: 5910/1: ARM: Add tmp register for addruart and loadsp Otherwise more complicated uart configuration won't be possible. We can use r1 for tmp register for both head.S and debug.S. NOTE: This patch depends on another patch to add the the tmp register into all debug-macro.S files. That can be done with: $ sed -i -e "s/addruart,rx|addruart, rx/addruart, rx, tmp/" arch/arm/*/include/*/debug-macro.S Signed-off-by: Tony Lindgren Signed-off-by: Russell King --- arch/arm/kernel/debug.S | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/debug.S b/arch/arm/kernel/debug.S index 5c91addcaebc..a38b4879441d 100644 --- a/arch/arm/kernel/debug.S +++ b/arch/arm/kernel/debug.S @@ -24,7 +24,7 @@ #if defined(CONFIG_CPU_V6) - .macro addruart, rx + .macro addruart, rx, tmp .endm .macro senduart, rd, rx @@ -51,7 +51,7 @@ #elif defined(CONFIG_CPU_V7) - .macro addruart, rx + .macro addruart, rx, tmp .endm .macro senduart, rd, rx @@ -71,7 +71,7 @@ wait: mrc p14, 0, pc, c0, c1, 0 #elif defined(CONFIG_CPU_XSCALE) - .macro addruart, rx + .macro addruart, rx, tmp .endm .macro senduart, rd, rx @@ -98,7 +98,7 @@ wait: mrc p14, 0, pc, c0, c1, 0 #else - .macro addruart, rx + .macro addruart, rx, tmp .endm .macro senduart, rd, rx @@ -164,7 +164,7 @@ ENDPROC(printhex2) .ltorg ENTRY(printascii) - addruart r3 + addruart r3, r1 b 2f 1: waituart r2, r3 senduart r1, r3 @@ -180,7 +180,7 @@ ENTRY(printascii) ENDPROC(printascii) ENTRY(printch) - addruart r3 + addruart r3, r1 mov r1, r0 mov r0, #0 b 1b -- cgit v1.2.2 From a9c9147eb9b1dba0ce567a41897c7773b4d1b0bc Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 26 Nov 2009 16:19:58 +0000 Subject: ARM: dma-mapping: provide per-cpu type map/unmap functions Signed-off-by: Russell King Tested-By: Santosh Shilimkar --- arch/arm/kernel/asm-offsets.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/asm-offsets.c b/arch/arm/kernel/asm-offsets.c index 4a881258bb17..883511522fca 100644 --- a/arch/arm/kernel/asm-offsets.c +++ b/arch/arm/kernel/asm-offsets.c @@ -12,6 +12,7 @@ */ #include #include +#include #include #include #include @@ -112,5 +113,9 @@ int main(void) #ifdef MULTI_PABORT DEFINE(PROCESSOR_PABT_FUNC, offsetof(struct processor, _prefetch_abort)); #endif + BLANK(); + DEFINE(DMA_BIDIRECTIONAL, DMA_BIDIRECTIONAL); + DEFINE(DMA_TO_DEVICE, DMA_TO_DEVICE); + DEFINE(DMA_FROM_DEVICE, DMA_FROM_DEVICE); return 0; } -- cgit v1.2.2 From 7921fc4a25800f4210bca96c05dea67a6b736d32 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 10 Jan 2010 17:05:08 +0000 Subject: ARM: remove old RTC support All RTC drivers have been converted to rtclib, so the old code providing the set_rtc function pointer, save_time_delta() and restore_time_delta() functions is obsolete. Remove it. Signed-off-by: Russell King --- arch/arm/kernel/time.c | 74 -------------------------------------------------- 1 file changed, 74 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c index d38cdf2c8276..e26f966d117b 100644 --- a/arch/arm/kernel/time.c +++ b/arch/arm/kernel/time.c @@ -10,11 +10,6 @@ * * This file contains the ARM-specific time handling details: * reading the RTC at bootup, etc... - * - * 1994-07-02 Alan Modra - * fixed set_rtc_mmss, fixed time.year for >= 2000, new mktime - * 1998-12-20 Updated NTP code according to technical memorandum Jan '96 - * "A Kernel Model for Precision Timekeeping" by Dave Mills */ #include #include @@ -77,11 +72,6 @@ unsigned long profile_pc(struct pt_regs *regs) EXPORT_SYMBOL(profile_pc); #endif -/* - * hook for setting the RTC's idea of the current time. - */ -int (*set_rtc)(void); - #ifndef CONFIG_GENERIC_TIME static unsigned long dummy_gettimeoffset(void) { @@ -89,36 +79,6 @@ static unsigned long dummy_gettimeoffset(void) } #endif -static unsigned long next_rtc_update; - -/* - * If we have an externally synchronized linux clock, then update - * CMOS clock accordingly every ~11 minutes. set_rtc() has to be - * called as close as possible to 500 ms before the new second - * starts. - */ -static inline void do_set_rtc(void) -{ - if (!ntp_synced() || set_rtc == NULL) - return; - - if (next_rtc_update && - time_before((unsigned long)xtime.tv_sec, next_rtc_update)) - return; - - if (xtime.tv_nsec < 500000000 - ((unsigned) tick_nsec >> 1) && - xtime.tv_nsec >= 500000000 + ((unsigned) tick_nsec >> 1)) - return; - - if (set_rtc()) - /* - * rtc update failed. Try again in 60s - */ - next_rtc_update = xtime.tv_sec + 60; - else - next_rtc_update = xtime.tv_sec + 660; -} - #ifdef CONFIG_LEDS static void dummy_leds_event(led_event_t evt) @@ -295,39 +255,6 @@ int do_settimeofday(struct timespec *tv) EXPORT_SYMBOL(do_settimeofday); #endif /* !CONFIG_GENERIC_TIME */ -/** - * save_time_delta - Save the offset between system time and RTC time - * @delta: pointer to timespec to store delta - * @rtc: pointer to timespec for current RTC time - * - * Return a delta between the system time and the RTC time, such - * that system time can be restored later with restore_time_delta() - */ -void save_time_delta(struct timespec *delta, struct timespec *rtc) -{ - set_normalized_timespec(delta, - xtime.tv_sec - rtc->tv_sec, - xtime.tv_nsec - rtc->tv_nsec); -} -EXPORT_SYMBOL(save_time_delta); - -/** - * restore_time_delta - Restore the current system time - * @delta: delta returned by save_time_delta() - * @rtc: pointer to timespec for current RTC time - */ -void restore_time_delta(struct timespec *delta, struct timespec *rtc) -{ - struct timespec ts; - - set_normalized_timespec(&ts, - delta->tv_sec + rtc->tv_sec, - delta->tv_nsec + rtc->tv_nsec); - - do_settimeofday(&ts); -} -EXPORT_SYMBOL(restore_time_delta); - #ifndef CONFIG_GENERIC_CLOCKEVENTS /* * Kernel system timer support. @@ -336,7 +263,6 @@ void timer_tick(void) { profile_tick(CPU_PROFILING); do_leds(); - do_set_rtc(); write_seqlock(&xtime_lock); do_timer(1); write_sequnlock(&xtime_lock); -- cgit v1.2.2 From 373b32abf9b33616ad94283461bb2ab49e8371d5 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 10 Jan 2010 17:15:32 +0000 Subject: ARM: move LED support code out of arch/arm/kernel/time.c Signed-off-by: Russell King --- arch/arm/kernel/Makefile | 1 + arch/arm/kernel/leds.c | 115 +++++++++++++++++++++++++++++++++++++++++++++++ arch/arm/kernel/time.c | 104 ------------------------------------------ 3 files changed, 116 insertions(+), 104 deletions(-) create mode 100644 arch/arm/kernel/leds.c (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile index dd00f747e2ad..eb0284cd8661 100644 --- a/arch/arm/kernel/Makefile +++ b/arch/arm/kernel/Makefile @@ -17,6 +17,7 @@ obj-y := compat.o elf.o entry-armv.o entry-common.o irq.o \ process.o ptrace.o return_address.o setup.o signal.o \ sys_arm.o stacktrace.o time.o traps.o +obj-$(CONFIG_LEDS) += leds.o obj-$(CONFIG_OC_ETM) += etm.o obj-$(CONFIG_ISA_DMA_API) += dma.o diff --git a/arch/arm/kernel/leds.c b/arch/arm/kernel/leds.c new file mode 100644 index 000000000000..31a316c1777b --- /dev/null +++ b/arch/arm/kernel/leds.c @@ -0,0 +1,115 @@ +/* + * LED support code, ripped out of arch/arm/kernel/time.c + * + * Copyright (C) 1994-2001 Russell King + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include + +#include + +static void dummy_leds_event(led_event_t evt) +{ +} + +void (*leds_event)(led_event_t) = dummy_leds_event; + +struct leds_evt_name { + const char name[8]; + int on; + int off; +}; + +static const struct leds_evt_name evt_names[] = { + { "amber", led_amber_on, led_amber_off }, + { "blue", led_blue_on, led_blue_off }, + { "green", led_green_on, led_green_off }, + { "red", led_red_on, led_red_off }, +}; + +static ssize_t leds_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + int ret = -EINVAL, len = strcspn(buf, " "); + + if (len > 0 && buf[len] == '\0') + len--; + + if (strncmp(buf, "claim", len) == 0) { + leds_event(led_claim); + ret = size; + } else if (strncmp(buf, "release", len) == 0) { + leds_event(led_release); + ret = size; + } else { + int i; + + for (i = 0; i < ARRAY_SIZE(evt_names); i++) { + if (strlen(evt_names[i].name) != len || + strncmp(buf, evt_names[i].name, len) != 0) + continue; + if (strncmp(buf+len, " on", 3) == 0) { + leds_event(evt_names[i].on); + ret = size; + } else if (strncmp(buf+len, " off", 4) == 0) { + leds_event(evt_names[i].off); + ret = size; + } + break; + } + } + return ret; +} + +static SYSDEV_ATTR(event, 0200, NULL, leds_store); + +static int leds_suspend(struct sys_device *dev, pm_message_t state) +{ + leds_event(led_stop); + return 0; +} + +static int leds_resume(struct sys_device *dev) +{ + leds_event(led_start); + return 0; +} + +static int leds_shutdown(struct sys_device *dev) +{ + leds_event(led_halted); + return 0; +} + +static struct sysdev_class leds_sysclass = { + .name = "leds", + .shutdown = leds_shutdown, + .suspend = leds_suspend, + .resume = leds_resume, +}; + +static struct sys_device leds_device = { + .id = 0, + .cls = &leds_sysclass, +}; + +static int __init leds_init(void) +{ + int ret; + ret = sysdev_class_register(&leds_sysclass); + if (ret == 0) + ret = sysdev_register(&leds_device); + if (ret == 0) + ret = sysdev_create_file(&leds_device, &attr_event); + return ret; +} + +device_initcall(leds_init); + +EXPORT_SYMBOL(leds_event); diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c index e26f966d117b..28753805d2d1 100644 --- a/arch/arm/kernel/time.c +++ b/arch/arm/kernel/time.c @@ -79,110 +79,6 @@ static unsigned long dummy_gettimeoffset(void) } #endif -#ifdef CONFIG_LEDS - -static void dummy_leds_event(led_event_t evt) -{ -} - -void (*leds_event)(led_event_t) = dummy_leds_event; - -struct leds_evt_name { - const char name[8]; - int on; - int off; -}; - -static const struct leds_evt_name evt_names[] = { - { "amber", led_amber_on, led_amber_off }, - { "blue", led_blue_on, led_blue_off }, - { "green", led_green_on, led_green_off }, - { "red", led_red_on, led_red_off }, -}; - -static ssize_t leds_store(struct sys_device *dev, - struct sysdev_attribute *attr, - const char *buf, size_t size) -{ - int ret = -EINVAL, len = strcspn(buf, " "); - - if (len > 0 && buf[len] == '\0') - len--; - - if (strncmp(buf, "claim", len) == 0) { - leds_event(led_claim); - ret = size; - } else if (strncmp(buf, "release", len) == 0) { - leds_event(led_release); - ret = size; - } else { - int i; - - for (i = 0; i < ARRAY_SIZE(evt_names); i++) { - if (strlen(evt_names[i].name) != len || - strncmp(buf, evt_names[i].name, len) != 0) - continue; - if (strncmp(buf+len, " on", 3) == 0) { - leds_event(evt_names[i].on); - ret = size; - } else if (strncmp(buf+len, " off", 4) == 0) { - leds_event(evt_names[i].off); - ret = size; - } - break; - } - } - return ret; -} - -static SYSDEV_ATTR(event, 0200, NULL, leds_store); - -static int leds_suspend(struct sys_device *dev, pm_message_t state) -{ - leds_event(led_stop); - return 0; -} - -static int leds_resume(struct sys_device *dev) -{ - leds_event(led_start); - return 0; -} - -static int leds_shutdown(struct sys_device *dev) -{ - leds_event(led_halted); - return 0; -} - -static struct sysdev_class leds_sysclass = { - .name = "leds", - .shutdown = leds_shutdown, - .suspend = leds_suspend, - .resume = leds_resume, -}; - -static struct sys_device leds_device = { - .id = 0, - .cls = &leds_sysclass, -}; - -static int __init leds_init(void) -{ - int ret; - ret = sysdev_class_register(&leds_sysclass); - if (ret == 0) - ret = sysdev_register(&leds_device); - if (ret == 0) - ret = sysdev_create_file(&leds_device, &attr_event); - return ret; -} - -device_initcall(leds_init); - -EXPORT_SYMBOL(leds_event); -#endif - #ifdef CONFIG_LEDS_TIMER static inline void do_leds(void) { -- cgit v1.2.2 From e119bfff1f102f8d1505910cd6c09df55c776b43 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 10 Jan 2010 17:23:29 +0000 Subject: ARM: Move creation of /proc/cpu out of alignment.c Always creating this directory avoids other users having to jump through silly hoops when they want to share this directory. Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index c6c57b640b6b..5357e48f2c39 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include @@ -782,9 +783,21 @@ static int __init topology_init(void) return 0; } - subsys_initcall(topology_init); +#ifdef CONFIG_HAVE_PROC_CPU +static int __init proc_cpu_init(void) +{ + struct proc_dir_entry *res; + + res = proc_mkdir("cpu", NULL); + if (!res) + return -ENOMEM; + return 0; +} +fs_initcall(proc_cpu_init); +#endif + static const char *hwcap_str[] = { "swp", "half", -- cgit v1.2.2 From 2b0d8c251b8876d530a6bf671eb5425838fa698a Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Mon, 11 Jan 2010 23:17:34 +0100 Subject: ARM: 5880/1: arm: use generic infrastructure for early params The ARM setup code includes its own parser for early params, there's also one in the generic init code. This patch removes __early_init (and related code) from arch/arm/kernel/setup.c, and changes users to the generic early_init macro instead. The generic macro takes a char * argument, rather than char **, so we need to update the parser functions a little. Signed-off-by: Jeremy Kerr Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 62 +++++++++++-------------------------------- arch/arm/kernel/vmlinux.lds.S | 4 --- 2 files changed, 16 insertions(+), 50 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 5357e48f2c39..b01a56a03ed8 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -418,10 +418,11 @@ static int __init arm_add_memory(unsigned long start, unsigned long size) * Pick out the memory size. We look for mem=size@start, * where start and size are "size[KkMm]" */ -static void __init early_mem(char **p) +static int __init early_mem(char *p) { static int usermem __initdata = 0; unsigned long size, start; + char *endp; /* * If the user specifies memory size, we @@ -434,52 +435,15 @@ static void __init early_mem(char **p) } start = PHYS_OFFSET; - size = memparse(*p, p); - if (**p == '@') - start = memparse(*p + 1, p); + size = memparse(p, &endp); + if (*endp == '@') + start = memparse(endp + 1, NULL); arm_add_memory(start, size); -} -__early_param("mem=", early_mem); -/* - * Initial parsing of the command line. - */ -static void __init parse_cmdline(char **cmdline_p, char *from) -{ - char c = ' ', *to = command_line; - int len = 0; - - for (;;) { - if (c == ' ') { - extern struct early_params __early_begin, __early_end; - struct early_params *p; - - for (p = &__early_begin; p < &__early_end; p++) { - int arglen = strlen(p->arg); - - if (memcmp(from, p->arg, arglen) == 0) { - if (to != command_line) - to -= 1; - from += arglen; - p->fn(&from); - - while (*from != ' ' && *from != '\0') - from++; - break; - } - } - } - c = *from++; - if (!c) - break; - if (COMMAND_LINE_SIZE <= ++len) - break; - *to++ = c; - } - *to = '\0'; - *cmdline_p = command_line; + return 0; } +early_param("mem", early_mem); static void __init setup_ramdisk(int doload, int prompt, int image_start, unsigned int rd_sz) @@ -740,9 +704,15 @@ void __init setup_arch(char **cmdline_p) init_mm.end_data = (unsigned long) _edata; init_mm.brk = (unsigned long) _end; - memcpy(boot_command_line, from, COMMAND_LINE_SIZE); - boot_command_line[COMMAND_LINE_SIZE-1] = '\0'; - parse_cmdline(cmdline_p, from); + /* parse_early_param needs a boot_command_line */ + strlcpy(boot_command_line, from, COMMAND_LINE_SIZE); + + /* populate command_line too for later use, preserving boot_command_line */ + strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); + *cmdline_p = command_line; + + parse_early_param(); + paging_init(mdesc); request_standard_resources(&meminfo, mdesc); diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 4957e13ef55b..b16c07914b55 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -43,10 +43,6 @@ SECTIONS INIT_SETUP(16) - __early_begin = .; - *(.early_param.init) - __early_end = .; - INIT_CALLS CON_INITCALL SECURITY_INITCALL -- cgit v1.2.2 From a9221de66d2d94e6e34c3f56bbdd744935020737 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 20 Jan 2010 17:02:54 +0000 Subject: ARM: add notify_die() support Kernel debuggers want to be informed of die() events, so that they can take some action to allow the problem to be inspected. Provide the hook in a similar manner to x86. Note that we currently don't implement the individual trap hooks. Acked-by: Jason Wessel Signed-off-by: Russell King --- arch/arm/kernel/traps.c | 35 +++++++++++++++++++++++++---------- 1 file changed, 25 insertions(+), 10 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 3f361a783f43..1621e5327b2a 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -12,15 +12,17 @@ * 'linux/arch/arm/lib/traps.S'. Mostly a debugging aid, but will probably * kill the offending process. */ -#include #include -#include #include #include -#include +#include +#include #include +#include +#include +#include +#include #include -#include #include #include @@ -224,14 +226,21 @@ void show_stack(struct task_struct *tsk, unsigned long *sp) #define S_SMP "" #endif -static void __die(const char *str, int err, struct thread_info *thread, struct pt_regs *regs) +static int __die(const char *str, int err, struct thread_info *thread, struct pt_regs *regs) { struct task_struct *tsk = thread->task; static int die_counter; + int ret; printk(KERN_EMERG "Internal error: %s: %x [#%d]" S_PREEMPT S_SMP "\n", str, err, ++die_counter); sysfs_printk_last_file(); + + /* trap and error numbers are mostly meaningless on ARM */ + ret = notify_die(DIE_OOPS, str, regs, err, tsk->thread.trap_no, SIGSEGV); + if (ret == NOTIFY_STOP) + return ret; + print_modules(); __show_regs(regs); printk(KERN_EMERG "Process %.*s (pid: %d, stack limit = 0x%p)\n", @@ -243,6 +252,8 @@ static void __die(const char *str, int err, struct thread_info *thread, struct p dump_backtrace(regs, tsk); dump_instr(KERN_EMERG, regs); } + + return ret; } DEFINE_SPINLOCK(die_lock); @@ -250,16 +261,21 @@ DEFINE_SPINLOCK(die_lock); /* * This function is protected against re-entrancy. */ -NORET_TYPE void die(const char *str, struct pt_regs *regs, int err) +void die(const char *str, struct pt_regs *regs, int err) { struct thread_info *thread = current_thread_info(); + int ret; oops_enter(); spin_lock_irq(&die_lock); console_verbose(); bust_spinlocks(1); - __die(str, err, thread, regs); + ret = __die(str, err, thread, regs); + + if (regs && kexec_should_crash(thread->task)) + crash_kexec(regs); + bust_spinlocks(0); add_taint(TAINT_DIE); spin_unlock_irq(&die_lock); @@ -267,11 +283,10 @@ NORET_TYPE void die(const char *str, struct pt_regs *regs, int err) if (in_interrupt()) panic("Fatal exception in interrupt"); - if (panic_on_oops) panic("Fatal exception"); - - do_exit(SIGSEGV); + if (ret != NOTIFY_STOP) + do_exit(SIGSEGV); } void arm_notify_die(const char *str, struct pt_regs *regs, -- cgit v1.2.2 From 48ab7e09e0a7c00a217f87e4b57dfbee9c603e79 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Wed, 27 Jan 2010 01:13:31 +0100 Subject: ARM: 5906/1: arm: change command_line to cmd_line drivers/of/fdt expects a cmd_line symbol, while arm uses command_line. Change to the former, so that we can eventually share with the fdt code. Signed-off-by: Jeremy Kerr Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index b01a56a03ed8..baf5959d639a 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -118,7 +118,7 @@ EXPORT_SYMBOL(elf_platform); static const char *cpu_name; static const char *machine_name; -static char __initdata command_line[COMMAND_LINE_SIZE]; +static char __initdata cmd_line[COMMAND_LINE_SIZE]; static char default_command_line[COMMAND_LINE_SIZE] __initdata = CONFIG_CMDLINE; static union { char c[4]; unsigned long l; } endian_test __initdata = { { 'l', '?', '?', 'b' } }; @@ -707,9 +707,9 @@ void __init setup_arch(char **cmdline_p) /* parse_early_param needs a boot_command_line */ strlcpy(boot_command_line, from, COMMAND_LINE_SIZE); - /* populate command_line too for later use, preserving boot_command_line */ - strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); - *cmdline_p = command_line; + /* populate cmd_line too for later use, preserving boot_command_line */ + strlcpy(cmd_line, boot_command_line, COMMAND_LINE_SIZE); + *cmdline_p = cmd_line; parse_early_param(); -- cgit v1.2.2 From d23bc1b3a7e6db935acb9a949a5985d9b77dfd13 Mon Sep 17 00:00:00 2001 From: Daniel Jacobowitz Date: Tue, 2 Feb 2010 18:22:16 +0100 Subject: ARM: 5912/1: Define a 32-bit Thumb-2 breakpoint instruction Recognize 0xf7f0 0xa000 as a 32-bit breakpoint instruction for Thumb-2. Signed-off-by: Daniel Jacobowitz Signed-off-by: Russell King --- arch/arm/kernel/ptrace.c | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index a2ea3854cb3c..bd56673c6a69 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -499,10 +499,41 @@ static struct undef_hook thumb_break_hook = { .fn = break_trap, }; +static int thumb2_break_trap(struct pt_regs *regs, unsigned int instr) +{ + unsigned int instr2; + void __user *pc; + + /* Check the second half of the instruction. */ + pc = (void __user *)(instruction_pointer(regs) + 2); + + if (processor_mode(regs) == SVC_MODE) { + instr2 = *(u16 *) pc; + } else { + get_user(instr2, (u16 __user *)pc); + } + + if (instr2 == 0xa000) { + ptrace_break(current, regs); + return 0; + } else { + return 1; + } +} + +static struct undef_hook thumb2_break_hook = { + .instr_mask = 0xffff, + .instr_val = 0xf7f0, + .cpsr_mask = PSR_T_BIT, + .cpsr_val = PSR_T_BIT, + .fn = thumb2_break_trap, +}; + static int __init ptrace_break_init(void) { register_undef_hook(&arm_break_hook); register_undef_hook(&thumb_break_hook); + register_undef_hook(&thumb2_break_hook); return 0; } -- cgit v1.2.2 From 462f39a8c7496c95f4de91ef46d875f46e0fa271 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 6 Feb 2010 10:35:23 +0000 Subject: ARM: ptrace: get rid of PTRACE_{PEEK,POKE}{TEXT,DATA} The generic ptrace_request() handles these for us, so there's no need to duplicate them in arch code. Signed-off-by: Russell King --- arch/arm/kernel/ptrace.c | 16 ---------------- 1 file changed, 16 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index bd56673c6a69..bdf002bab6a5 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -743,26 +743,10 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) int ret; switch (request) { - /* - * read word at location "addr" in the child process. - */ - case PTRACE_PEEKTEXT: - case PTRACE_PEEKDATA: - ret = generic_ptrace_peekdata(child, addr, data); - break; - case PTRACE_PEEKUSR: ret = ptrace_read_user(child, addr, (unsigned long __user *)data); break; - /* - * write the word at location addr. - */ - case PTRACE_POKETEXT: - case PTRACE_POKEDATA: - ret = generic_ptrace_pokedata(child, addr, data); - break; - case PTRACE_POKEUSR: ret = ptrace_write_user(child, addr, data); break; -- cgit v1.2.2 From ad187f956108e1c56b444706212bf08d84c0bee0 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 6 Feb 2010 11:36:23 +0000 Subject: ARM: vfp ptrace: no point flushing hw context for PTRACE_GETVFPREGS If we're only reading the VFP context via the ptrace call, there's no need to invalidate the hardware context - we only need to do that on PTRACE_SETVFPREGS. This allows more efficient monitoring of a traced task. Signed-off-by: Russell King --- arch/arm/kernel/ptrace.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index bdf002bab6a5..08f899fb76a6 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -700,7 +700,7 @@ static int ptrace_getvfpregs(struct task_struct *tsk, void __user *data) union vfp_state *vfp = &thread->vfpstate; struct user_vfp __user *ufp = data; - vfp_sync_state(thread); + vfp_sync_hwstate(thread); /* copy the floating point registers */ if (copy_to_user(&ufp->fpregs, &vfp->hard.fpregs, @@ -723,7 +723,7 @@ static int ptrace_setvfpregs(struct task_struct *tsk, void __user *data) union vfp_state *vfp = &thread->vfpstate; struct user_vfp __user *ufp = data; - vfp_sync_state(thread); + vfp_sync_hwstate(thread); /* copy the floating point registers */ if (copy_from_user(&vfp->hard.fpregs, &ufp->fpregs, @@ -734,6 +734,8 @@ static int ptrace_setvfpregs(struct task_struct *tsk, void __user *data) if (get_user(vfp->hard.fpscr, &ufp->fpscr)) return -EFAULT; + vfp_flush_hwstate(thread); + return 0; } #endif -- cgit v1.2.2 From 6c09f09d44690d341d970559b64779bef8b9236b Mon Sep 17 00:00:00 2001 From: Santosh Shilimkar Date: Tue, 16 Feb 2010 07:57:43 +0100 Subject: ARM: 5938/1: ARM: L2: export outer_cache_fns The 'outer_cache' variable is needed by the outer_inv_range(), outer_clean_range() and outer_flush_range() functions, which are declared as inline in asm/cacheflush.h. Otherwise drivers built as a loadable module, which access these functions, will have an undefined symbol. Signed-off-by: Santosh Shilimkar 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 c6c57b640b6b..621acad8ea43 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -102,6 +102,7 @@ struct cpu_cache_fns cpu_cache; #endif #ifdef CONFIG_OUTER_CACHE struct outer_cache_fns outer_cache; +EXPORT_SYMBOL(outer_cache); #endif struct stack { -- cgit v1.2.2 From b26b2d494b659f988b4d75eb394dfa0ddac415c9 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Fri, 1 Jan 2010 17:40:49 +0100 Subject: resource/PCI: align functions now return start of resource As suggested by Linus, align functions should return the start of a resource, not void. An update of "res->start" is no longer necessary. Cc: Bjorn Helgaas Cc: Yinghai Lu Signed-off-by: Dominik Brodowski Signed-off-by: Jesse Barnes --- arch/arm/kernel/bios32.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/bios32.c b/arch/arm/kernel/bios32.c index 809681900ec8..a7c85f84f06c 100644 --- a/arch/arm/kernel/bios32.c +++ b/arch/arm/kernel/bios32.c @@ -616,15 +616,17 @@ char * __init pcibios_setup(char *str) * but we want to try to avoid allocating at 0x2900-0x2bff * which might be mirrored at 0x0100-0x03ff.. */ -void pcibios_align_resource(void *data, struct resource *res, - resource_size_t size, resource_size_t align) +resource_size_t pcibios_align_resource(void *data, struct resource *res, + resource_size_t size, resource_size_t align) { resource_size_t start = res->start; if (res->flags & IORESOURCE_IO && start & 0x300) start = (start + 0x3ff) & ~0x3ff; - res->start = (start + align - 1) & ~(align - 1); + start = (start + align - 1) & ~(align - 1); + + return start; } /** -- cgit v1.2.2 From 3b7a17fcdae532d29dffab9d564a28be08960988 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Fri, 1 Jan 2010 17:40:50 +0100 Subject: resource/PCI: mark struct resource as const Now that we return the new resource start position, there is no need to update "struct resource" inside the align function. Therefore, mark the struct resource as const. Cc: Bjorn Helgaas Cc: Yinghai Lu Signed-off-by: Dominik Brodowski Signed-off-by: Jesse Barnes --- arch/arm/kernel/bios32.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/bios32.c b/arch/arm/kernel/bios32.c index a7c85f84f06c..bd397e0b663e 100644 --- a/arch/arm/kernel/bios32.c +++ b/arch/arm/kernel/bios32.c @@ -616,7 +616,7 @@ char * __init pcibios_setup(char *str) * but we want to try to avoid allocating at 0x2900-0x2bff * which might be mirrored at 0x0100-0x03ff.. */ -resource_size_t pcibios_align_resource(void *data, struct resource *res, +resource_size_t pcibios_align_resource(void *data, const struct resource *res, resource_size_t size, resource_size_t align) { resource_size_t start = res->start; -- cgit v1.2.2 From 51d47999b9452a8dc7ae58a11423c5db28f21ae1 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Thu, 4 Mar 2010 15:33:16 +0100 Subject: ARM: 5977/1: arm: Enable backtrace printing on oops when PC is corrupted If PC points outside kernel text, start printing the backtrace at LR instead. Signed-off-by: Laurent Pinchart Signed-off-by: Russell King --- arch/arm/kernel/unwind.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c index 786ac2b6914a..50292cd9c120 100644 --- a/arch/arm/kernel/unwind.c +++ b/arch/arm/kernel/unwind.c @@ -359,7 +359,9 @@ void unwind_backtrace(struct pt_regs *regs, struct task_struct *tsk) frame.fp = regs->ARM_fp; frame.sp = regs->ARM_sp; frame.lr = regs->ARM_lr; - frame.pc = regs->ARM_pc; + /* PC might be corrupted, use LR in that case. */ + frame.pc = kernel_text_address(regs->ARM_pc) + ? regs->ARM_pc : regs->ARM_lr; } else if (tsk == current) { frame.fp = (unsigned long)__builtin_frame_address(0); frame.sp = current_sp; -- cgit v1.2.2 From dc1d628a67a8f042e711ea5accc0beedc3ef0092 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 3 Mar 2010 15:55:04 +0100 Subject: perf: Provide generic perf_sample_data initialization This makes it easier to extend perf_sample_data and fixes a bug on arm and sparc, which failed to set ->raw to NULL, which can cause crashes when combined with PERF_SAMPLE_RAW. It also optimizes PowerPC and tracepoint, because the struct initialization is forced to zero out the whole structure. Signed-off-by: Peter Zijlstra Acked-by: Jean Pihet Reviewed-by: Frederic Weisbecker Acked-by: David S. Miller Cc: Jamie Iles Cc: Paul Mackerras Cc: Stephane Eranian Cc: stable@kernel.org LKML-Reference: <20100304140100.315416040@chello.nl> Signed-off-by: Ingo Molnar --- arch/arm/kernel/perf_event.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index c54ceb3d1f97..3875d99cc40f 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -965,7 +965,7 @@ armv6pmu_handle_irq(int irq_num, */ armv6_pmcr_write(pmcr); - data.addr = 0; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); for (idx = 0; idx <= armpmu->num_events; ++idx) { @@ -1945,7 +1945,7 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev) */ regs = get_irq_regs(); - data.addr = 0; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); for (idx = 0; idx <= armpmu->num_events; ++idx) { -- cgit v1.2.2 From 5d0e52830e9ae09b872567f4aca3dfb5b5918079 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 10 Mar 2010 15:21:13 -0800 Subject: Add generic sys_old_select() Add a generic implementation of the old select() syscall, which expects its argument in a memory block and switch all architectures over to use it. Signed-off-by: Christoph Hellwig Cc: Ralf Baechle Cc: Benjamin Herrenschmidt Cc: Paul Mundt Cc: Jeff Dike Cc: Hirokazu Takata Cc: Thomas Gleixner Cc: Ingo Molnar Reviewed-by: H. Peter Anvin Cc: Al Viro Cc: Arnd Bergmann Cc: Heiko Carstens Cc: Martin Schwidefsky Cc: "Luck, Tony" Cc: James Morris Acked-by: Andreas Schwab Acked-by: Russell King Acked-by: Greg Ungerer Acked-by: David Howells Cc: Andreas Schwab Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/calls.S | 2 +- arch/arm/kernel/sys_arm.c | 21 --------------------- 2 files changed, 1 insertion(+), 22 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index 9314a2d681f1..7671e9a75449 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -91,7 +91,7 @@ CALL(sys_settimeofday) /* 80 */ CALL(sys_getgroups16) CALL(sys_setgroups16) - CALL(OBSOLETE(old_select)) /* used by libc4 */ + CALL(OBSOLETE(sys_old_select)) /* used by libc4 */ CALL(sys_symlink) CALL(sys_ni_syscall) /* was sys_lstat */ /* 85 */ CALL(sys_readlink) diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index ae4027bd01bd..e59cddedcbba 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -54,27 +54,6 @@ out: return error; } -/* - * Perform the select(nd, in, out, ex, tv) and mmap() system - * calls. - */ - -struct sel_arg_struct { - unsigned long n; - fd_set __user *inp, *outp, *exp; - struct timeval __user *tvp; -}; - -asmlinkage int old_select(struct sel_arg_struct __user *arg) -{ - struct sel_arg_struct a; - - if (copy_from_user(&a, arg, sizeof(a))) - return -EFAULT; - /* sys_select() does the appropriate kernel locking */ - return sys_select(a.n, a.inp, a.outp, a.exp, a.tvp); -} - #if !defined(CONFIG_AEABI) || defined(CONFIG_OABI_COMPAT) /* * sys_ipc() is the de-multiplexer for the SysV IPC calls.. -- cgit v1.2.2 From a4679373cf4ee0e7792dc56205365732b725c2c1 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 10 Mar 2010 15:21:15 -0800 Subject: Add generic sys_old_mmap() Add a generic implementation of the old mmap() syscall, which expects its argument in a memory block and switch all architectures over to use it. Signed-off-by: Christoph Hellwig Cc: Ralf Baechle Cc: Benjamin Herrenschmidt Cc: Paul Mundt Cc: Jeff Dike Cc: Hirokazu Takata Cc: Thomas Gleixner Cc: Ingo Molnar Reviewed-by: H. Peter Anvin Cc: Al Viro Cc: Arnd Bergmann Cc: Heiko Carstens Cc: Martin Schwidefsky Cc: "Luck, Tony" Cc: James Morris Cc: Andreas Schwab Acked-by: Jesper Nilsson Acked-by: Russell King Acked-by: Greg Ungerer Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/calls.S | 2 +- arch/arm/kernel/sys_arm.c | 26 -------------------------- 2 files changed, 1 insertion(+), 27 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index 7671e9a75449..37ae301cc47c 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -99,7 +99,7 @@ CALL(sys_swapon) CALL(sys_reboot) CALL(OBSOLETE(sys_old_readdir)) /* used by libc4 */ -/* 90 */ CALL(OBSOLETE(old_mmap)) /* used by libc4 */ +/* 90 */ CALL(OBSOLETE(sys_old_mmap)) /* used by libc4 */ CALL(sys_munmap) CALL(sys_truncate) CALL(sys_ftruncate) diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index e59cddedcbba..a2e0e6f2ea7f 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -28,32 +28,6 @@ #include #include -struct mmap_arg_struct { - unsigned long addr; - unsigned long len; - unsigned long prot; - unsigned long flags; - unsigned long fd; - unsigned long offset; -}; - -asmlinkage int old_mmap(struct mmap_arg_struct __user *arg) -{ - int error = -EFAULT; - struct mmap_arg_struct a; - - if (copy_from_user(&a, arg, sizeof(a))) - goto out; - - error = -EINVAL; - if (a.offset & ~PAGE_MASK) - goto out; - - error = sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, a.offset >> PAGE_SHIFT); -out: - return error; -} - #if !defined(CONFIG_AEABI) || defined(CONFIG_OABI_COMPAT) /* * sys_ipc() is the de-multiplexer for the SysV IPC calls.. -- cgit v1.2.2 From baed7fc9b580bd3fb8252ff1d9b36eaf1f86b670 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 10 Mar 2010 15:21:18 -0800 Subject: Add generic sys_ipc wrapper Add a generic implementation of the ipc demultiplexer syscall. Except for s390 and sparc64 all implementations of the sys_ipc are nearly identical. There are slight differences in the types of the parameters, where mips and powerpc as the only 64-bit architectures with sys_ipc use unsigned long for the "third" argument as it gets casted to a pointer later, while it traditionally is an "int" like most other paramters. frv goes even further and uses unsigned long for all parameters execept for "ptr" which is a pointer type everywhere. The change from int to unsigned long for "third" and back to "int" for the others on frv should be fine due to the in-register calling conventions for syscalls (we already had a similar issue with the generic sys_ptrace), but I'd prefer to have the arch maintainers looks over this in details. Except for that h8300, m68k and m68knommu lack an impplementation of the semtimedop sub call which this patch adds, and various architectures have gets used - at least on i386 it seems superflous as the compat code on x86-64 and ia64 doesn't even bother to implement it. [akpm@linux-foundation.org: add sys_ipc to sys_ni.c] Signed-off-by: Christoph Hellwig Cc: Ralf Baechle Cc: Benjamin Herrenschmidt Cc: Paul Mundt Cc: Jeff Dike Cc: Hirokazu Takata Cc: Thomas Gleixner Cc: Ingo Molnar Reviewed-by: H. Peter Anvin Cc: Al Viro Cc: Arnd Bergmann Cc: Heiko Carstens Cc: Martin Schwidefsky Cc: "Luck, Tony" Cc: James Morris Cc: Andreas Schwab Acked-by: Jesper Nilsson Acked-by: Russell King Acked-by: David Howells Acked-by: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/sys_arm.c | 82 --------------------------------------- arch/arm/kernel/sys_oabi-compat.c | 3 -- 2 files changed, 85 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index a2e0e6f2ea7f..4350f75e578c 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -28,88 +28,6 @@ #include #include -#if !defined(CONFIG_AEABI) || defined(CONFIG_OABI_COMPAT) -/* - * sys_ipc() is the de-multiplexer for the SysV IPC calls.. - * - * This is really horribly ugly. - */ -asmlinkage int sys_ipc(uint call, int first, int second, int third, - void __user *ptr, long fifth) -{ - int version, ret; - - version = call >> 16; /* hack for backward compatibility */ - call &= 0xffff; - - switch (call) { - case SEMOP: - return sys_semtimedop (first, (struct sembuf __user *)ptr, second, NULL); - case SEMTIMEDOP: - return sys_semtimedop(first, (struct sembuf __user *)ptr, second, - (const struct timespec __user *)fifth); - - case SEMGET: - return sys_semget (first, second, third); - case SEMCTL: { - union semun fourth; - if (!ptr) - return -EINVAL; - if (get_user(fourth.__pad, (void __user * __user *) ptr)) - return -EFAULT; - return sys_semctl (first, second, third, fourth); - } - - case MSGSND: - return sys_msgsnd(first, (struct msgbuf __user *) ptr, - second, third); - case MSGRCV: - switch (version) { - case 0: { - struct ipc_kludge tmp; - if (!ptr) - return -EINVAL; - if (copy_from_user(&tmp,(struct ipc_kludge __user *)ptr, - sizeof (tmp))) - return -EFAULT; - return sys_msgrcv (first, tmp.msgp, second, - tmp.msgtyp, third); - } - default: - return sys_msgrcv (first, - (struct msgbuf __user *) ptr, - second, fifth, third); - } - case MSGGET: - return sys_msgget ((key_t) first, second); - case MSGCTL: - return sys_msgctl(first, second, (struct msqid_ds __user *)ptr); - - case SHMAT: - switch (version) { - default: { - ulong raddr; - ret = do_shmat(first, (char __user *)ptr, second, &raddr); - if (ret) - return ret; - return put_user(raddr, (ulong __user *)third); - } - case 1: /* Of course, we don't support iBCS2! */ - return -EINVAL; - } - case SHMDT: - return sys_shmdt ((char __user *)ptr); - case SHMGET: - return sys_shmget (first, second, third); - case SHMCTL: - return sys_shmctl (first, second, - (struct shmid_ds __user *) ptr); - default: - return -ENOSYS; - } -} -#endif - /* Fork a new task - this creates a new program thread. * This is called indirectly via a small wrapper */ diff --git a/arch/arm/kernel/sys_oabi-compat.c b/arch/arm/kernel/sys_oabi-compat.c index d59a0cd537f0..33ff678e32f2 100644 --- a/arch/arm/kernel/sys_oabi-compat.c +++ b/arch/arm/kernel/sys_oabi-compat.c @@ -346,9 +346,6 @@ asmlinkage long sys_oabi_semop(int semid, struct oabi_sembuf __user *tsops, return sys_oabi_semtimedop(semid, tsops, nsops, NULL); } -extern asmlinkage int sys_ipc(uint call, int first, int second, int third, - void __user *ptr, long fifth); - asmlinkage int sys_oabi_ipc(uint call, int first, int second, int third, void __user *ptr, long fifth) { -- cgit v1.2.2 From 440e6ca79aebdc274ce4c625a6f42c8bf3c7bc91 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 10 Mar 2010 15:22:50 -0800 Subject: arm: use generic ptrace_resume code Use the generic ptrace_resume code for PTRACE_SYSCALL, PTRACE_CONT, PTRACE_KILL and PTRACE_SINGLESTEP. This implies defining arch_has_single_step in and implementing the user_enable_single_step and user_disable_single_step functions, which also causes the breakpoint information to be cleared on fork, which could be considered a bug fix. Also the TIF_SYSCALL_TRACE thread flag is now cleared on PTRACE_KILL which it previously wasn't and the single stepping disable only happens if the tracee process isn't a zombie yet, which is consistent with all architectures using the modern ptrace code. Signed-off-by: Christoph Hellwig Cc: Oleg Nesterov Cc: Roland McGrath Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/ptrace.c | 60 ++++++++++-------------------------------------- arch/arm/kernel/ptrace.h | 14 ----------- 2 files changed, 12 insertions(+), 62 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index 08f899fb76a6..3f562a7c0a99 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -452,12 +452,23 @@ void ptrace_cancel_bpt(struct task_struct *child) clear_breakpoint(child, &child->thread.debug.bp[i]); } +void user_disable_single_step(struct task_struct *task) +{ + task->ptrace &= ~PT_SINGLESTEP; + ptrace_cancel_bpt(task); +} + +void user_enable_single_step(struct task_struct *task) +{ + task->ptrace |= PT_SINGLESTEP; +} + /* * Called by kernel/ptrace.c when detaching.. */ void ptrace_disable(struct task_struct *child) { - single_step_disable(child); + user_disable_single_step(child); } /* @@ -753,53 +764,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) ret = ptrace_write_user(child, addr, data); break; - /* - * continue/restart and stop at next (return from) syscall - */ - case PTRACE_SYSCALL: - case PTRACE_CONT: - ret = -EIO; - if (!valid_signal(data)) - break; - if (request == PTRACE_SYSCALL) - set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - else - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - child->exit_code = data; - single_step_disable(child); - wake_up_process(child); - ret = 0; - break; - - /* - * make the child exit. Best I can do is send it a sigkill. - * perhaps it should be put in the status that it wants to - * exit. - */ - case PTRACE_KILL: - single_step_disable(child); - if (child->exit_state != EXIT_ZOMBIE) { - child->exit_code = SIGKILL; - wake_up_process(child); - } - ret = 0; - break; - - /* - * execute single instruction. - */ - case PTRACE_SINGLESTEP: - ret = -EIO; - if (!valid_signal(data)) - break; - single_step_enable(child); - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - child->exit_code = data; - /* give it a chance to run. */ - wake_up_process(child); - ret = 0; - break; - case PTRACE_GETREGS: ret = ptrace_getregs(child, (void __user *)data); break; diff --git a/arch/arm/kernel/ptrace.h b/arch/arm/kernel/ptrace.h index def3b6184a79..3926605b82ea 100644 --- a/arch/arm/kernel/ptrace.h +++ b/arch/arm/kernel/ptrace.h @@ -13,20 +13,6 @@ extern void ptrace_cancel_bpt(struct task_struct *); extern void ptrace_set_bpt(struct task_struct *); extern void ptrace_break(struct task_struct *, struct pt_regs *); -/* - * make sure single-step breakpoint is gone. - */ -static inline void single_step_disable(struct task_struct *task) -{ - task->ptrace &= ~PT_SINGLESTEP; - ptrace_cancel_bpt(task); -} - -static inline void single_step_enable(struct task_struct *task) -{ - task->ptrace |= PT_SINGLESTEP; -} - /* * Send SIGTRAP if we're single-stepping */ -- cgit v1.2.2 From ddee87f208b6229d2910dd5930c87089dc56c87e Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 25 Feb 2010 15:04:14 +0100 Subject: ARM: 5959/1: ARM: perf-events: request PMU interrupts with IRQF_NOBALANCING If IRQ balancing is used on a multicore ARM system, PMU interrupt lines may be relocated onto CPUs other than the one causing the counter overflow. This can result in misattribution of events to the wrong core and, in the case that the CPU handling the interrupt has not experience counter overflow, the interrupt can be disabled because the handler returns IRQ_NONE. This patch adds the IRQF_NOBALANCING flag to the request_irq call in perf_events.c. Acked-by: Jamie Iles Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/perf_event.c | 3 ++- 1 file changed, 2 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 c54ceb3d1f97..b44d15948b56 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -332,7 +332,8 @@ armpmu_reserve_hardware(void) for (i = 0; i < pmu_irqs->num_irqs; ++i) { err = request_irq(pmu_irqs->irqs[i], armpmu->handle_irq, - IRQF_DISABLED, "armpmu", NULL); + IRQF_DISABLED | IRQF_NOBALANCING, + "armpmu", NULL); if (err) { pr_warning("unable to request IRQ%d for ARM " "perf counters\n", pmu_irqs->irqs[i]); -- cgit v1.2.2 From d10fca9f39238b07cc670b441d2b423de30359d2 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 26 Feb 2010 10:46:15 +0100 Subject: ARM: 5960/1: ARM: perf-events: fix v7 event selection mask The event selection mask for ARMv7 cores [ARMV7_EVTSEL_MASK] is incorrectly set to 0x7f. This means that the top bit of an event ID is ignored, so counting branch misses (id=0x10) and ISBs (id=0x90) give the same results. This patch sets the event selection mask to the correct value of 0xff. Signed-off-by: Jean Pihet Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/perf_event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index b44d15948b56..c45a155a73dc 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -1625,7 +1625,7 @@ enum armv7_counters { /* * EVTSEL: Event selection reg */ -#define ARMV7_EVTSEL_MASK 0x7f /* Mask for writable bits */ +#define ARMV7_EVTSEL_MASK 0xff /* Mask for writable bits */ /* * SELECT: Counter selection reg -- cgit v1.2.2 From 1027247f6eb727db6c600b9eb02796f0766ae704 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 12 Feb 2010 14:36:24 +0000 Subject: ARM: Add L2 cache handling to smp boot support The page table and secondary data which we're asking the secondary CPU to make use of has to hit RAM to ensure that the secondary CPU can see it since it may not be taking part in coherency or cache searches at this point. Acked-by: Santosh Shilimkar Signed-off-by: Russell King --- arch/arm/kernel/smp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 57162af53dc9..577543f3857f 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -99,6 +99,7 @@ int __cpuinit __cpu_up(unsigned int cpu) *pmd = __pmd((PHYS_OFFSET & PGDIR_MASK) | PMD_TYPE_SECT | PMD_SECT_AP_WRITE); flush_pmd_entry(pmd); + outer_clean_range(__pa(pmd), __pa(pmd + 1)); /* * We need to tell the secondary core where to find @@ -106,7 +107,8 @@ int __cpuinit __cpu_up(unsigned int cpu) */ secondary_data.stack = task_stack_page(idle) + THREAD_START_SP; secondary_data.pgdir = virt_to_phys(pgd); - wmb(); + __cpuc_flush_dcache_area(&secondary_data, sizeof(secondary_data)); + outer_clean_range(__pa(&secondary_data), __pa(&secondary_data + 1)); /* * Now bring the CPU into our world. -- cgit v1.2.2 From 5d8614cc5d6c5c02b7995ed97303481d4e3a8cc7 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 12 Mar 2010 11:03:58 +0100 Subject: ARM: 5989/1: ARM: KGDB: add support for SMP platforms To support SMP platforms, KGDB requires the architecture backend to implement the kgdb_roundup_cpus function. This patch, taken against 2.6.33, implements the function for ARM based on the MIPS port. Cc: Jason Wessel Cc: Jean-Michel Hautbois Cc: KGDB Mailing List Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/kgdb.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/kgdb.c b/arch/arm/kernel/kgdb.c index ba8ccfede964..a5b846b9895d 100644 --- a/arch/arm/kernel/kgdb.c +++ b/arch/arm/kernel/kgdb.c @@ -9,6 +9,7 @@ * Authors: George Davis * Deepak Saxena */ +#include #include #include @@ -158,6 +159,18 @@ static struct undef_hook kgdb_compiled_brkpt_hook = { .fn = kgdb_compiled_brk_fn }; +static void kgdb_call_nmi_hook(void *ignored) +{ + kgdb_nmicallback(raw_smp_processor_id(), get_irq_regs()); +} + +void kgdb_roundup_cpus(unsigned long flags) +{ + local_irq_enable(); + smp_call_function(kgdb_call_nmi_hook, NULL, 0); + local_irq_disable(); +} + /** * kgdb_arch_init - Perform any architecture specific initalization. * -- cgit v1.2.2 From 8e4971f2fb2380ce66196136e113d04196b80fcd Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anders=20Grafstr=C3=B6m?= Date: Mon, 15 Mar 2010 16:04:14 +0100 Subject: ARM: 5991/1: Fix regression in restore_user_regs macro MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ARMv5T and earlier require that a ldm {}^ instruction is not followed by an instruction that accesses banked registers. This patch restores the nop that was lost in commit b86040a59feb255a8193173caa4d5199464433d5. Signed-off-by: Anders Grafström Signed-off-by: Russell King --- arch/arm/kernel/entry-header.S | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-header.S b/arch/arm/kernel/entry-header.S index 7e9ed1eea40a..d93f976fb389 100644 --- a/arch/arm/kernel/entry-header.S +++ b/arch/arm/kernel/entry-header.S @@ -102,6 +102,8 @@ .else ldmdb sp, {r0 - lr}^ @ get calling r0 - lr .endif + mov r0, r0 @ ARMv5T and earlier require a nop + @ after ldm {}^ add sp, sp, #S_FRAME_SIZE - S_PC movs pc, lr @ return & move spsr_svc into cpsr .endm -- cgit v1.2.2 From 782a0fd16723bfc0e765d789e82853d5dc424e76 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Mon, 29 Mar 2010 06:59:16 +0100 Subject: ARM: 6005/1: arm: kprobes: fix register corruption with jprobes Current implementation of jprobes allocates empty pt_regs from the stack which is then passed to kprobe_handler() and eventually to singlestep(). Now when instruction being simulated is STMFD (like in normal function prologues without CONFIG_FRAME_POINTER), stores using SP actually write over top of the fabricated pt_regs structure. This can be reproduced for example by using LKDTM module: # modprobe lkdtm # mount -t debugfs none /sys/kernel/debug # echo PANIC > /sys/kernel/debug/provoke-crash/INT_HW_IRQ_EN after this, it fails with corrupted registers (before the requested crash would occur): lkdtm: Crash point INT_HW_IRQ_EN of type PANIC hit, trigger in 9 rounds lkdtm: Crash point INT_HW_IRQ_EN of type PANIC hit, trigger in 8 rounds Internal error: Oops - undefined instruction: 0 [#1] last sysfs file: /sys/devices/platform/serial8250.0/sleep_timeout Modules linked in: lkdtm CPU: 0 Not tainted (2.6.34-rc2 #69) PC is at irq_desc+0x1638/0xeeb0 LR is at 0x25 pc : [] lr : [<00000025>] psr: c80a0013 sp : ce94bd60 ip : c050b3e8 fp : a0000013 r10: c0aa453c r9 : cf5d4000 r8 : ce9a1822 r7 : c050b424 r6 : 00000025 r5 : c039d8f8 r4 : c050b3e8 r3 : 00000001 r2 : cf4d0440 r1 : c039d8f8 r0 : 00000020 Flags: NZcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 10c5387d Table: 8e804019 DAC: 00000015 Process sh (pid: 496, stack limit = 0xce94a2e8) Stack: (0xce94bd60 to 0xce94c000) [...] Code: 000002cd 00000000 00000000 00000001 (dead4ead) ---[ end trace 2b46d5f2b682f370 ]--- Kernel panic - not syncing: Fatal exception in interrupt This patch allocates enough space (2 * sizeof(struct pt_regs)) from the stack to prevent such corruption. Signed-off-by: Mika Westerberg Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/kernel/kprobes.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c index 60c62c377fa9..610e0f561c32 100644 --- a/arch/arm/kernel/kprobes.c +++ b/arch/arm/kernel/kprobes.c @@ -393,6 +393,14 @@ void __kprobes jprobe_return(void) /* * Setup an empty pt_regs. Fill SP and PC fields as * they're needed by longjmp_break_handler. + * + * We allocate some slack between the original SP and start of + * our fabricated regs. To be precise we want to have worst case + * covered which is STMFD with all 16 regs so we allocate 2 * + * sizeof(struct_pt_regs)). + * + * This is to prevent any simulated instruction from writing + * over the regs when they are accessing the stack. */ "sub sp, %0, %1 \n\t" "ldr r0, ="__stringify(JPROBE_MAGIC_ADDR)"\n\t" @@ -410,7 +418,7 @@ void __kprobes jprobe_return(void) "ldmia sp, {r0 - pc} \n\t" : : "r" (kcb->jprobe_saved_regs.ARM_sp), - "I" (sizeof(struct pt_regs)), + "I" (sizeof(struct pt_regs) * 2), "J" (offsetof(struct pt_regs, ARM_sp)), "J" (offsetof(struct pt_regs, ARM_pc)), "J" (offsetof(struct pt_regs, ARM_cpsr)) -- cgit v1.2.2 From 5a0e3ad6af8660be21ca98a971cd00f331318c05 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 24 Mar 2010 17:04:11 +0900 Subject: include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo Guess-its-ok-by: Christoph Lameter Cc: Ingo Molnar Cc: Lee Schermerhorn --- arch/arm/kernel/irq.c | 1 - arch/arm/kernel/kprobes.c | 1 + arch/arm/kernel/module.c | 2 +- arch/arm/kernel/process.c | 1 - arch/arm/kernel/sys_arm.c | 2 +- 5 files changed, 3 insertions(+), 4 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c index b7cb45bb91e8..3b3d2c80509c 100644 --- a/arch/arm/kernel/irq.c +++ b/arch/arm/kernel/irq.c @@ -27,7 +27,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c index 60c62c377fa9..1fb932b4fece 100644 --- a/arch/arm/kernel/kprobes.c +++ b/arch/arm/kernel/kprobes.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c index f28c5e9c51ea..c628bdf6c430 100644 --- a/arch/arm/kernel/module.c +++ b/arch/arm/kernel/module.c @@ -16,9 +16,9 @@ #include #include #include -#include #include #include +#include #include #include diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index ba2adefa53f7..0e12e0acbf26 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index 4350f75e578c..c23501842b98 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include #include @@ -27,6 +26,7 @@ #include #include #include +#include /* Fork a new task - this creates a new program thread. * This is called indirectly via a small wrapper -- cgit v1.2.2 From 82c6f5a5b3e91ef4d2fb8725de4b8cf7affd4d61 Mon Sep 17 00:00:00 2001 From: Imre Deak Date: Sun, 11 Apr 2010 15:58:27 +0100 Subject: ARM: 6051/1: VFP: preserve the HW context when calling signal handlers From: Imre Deak Signal handlers can use floating point, so prevent them to corrupt the main thread's VFP context. So far there were two signal stack frame formats defined based on the VFP implementation, but the user struct used for ptrace covers all posibilities, so use it for the signal stack too. Introduce also a new user struct for VFP exception registers. In this too fields not relevant to the current VFP architecture are ignored. Support to save / restore the exception registers was added by Will Deacon. Signed-off-by: Imre Deak Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/signal.c | 93 +++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 89 insertions(+), 4 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index e7714f367eb8..907d5a620bca 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "ptrace.h" #include "signal.h" @@ -175,6 +176,90 @@ static int restore_iwmmxt_context(struct iwmmxt_sigframe *frame) #endif +#ifdef CONFIG_VFP + +static int preserve_vfp_context(struct vfp_sigframe __user *frame) +{ + struct thread_info *thread = current_thread_info(); + struct vfp_hard_struct *h = &thread->vfpstate.hard; + const unsigned long magic = VFP_MAGIC; + const unsigned long size = VFP_STORAGE_SIZE; + int err = 0; + + vfp_sync_hwstate(thread); + __put_user_error(magic, &frame->magic, err); + __put_user_error(size, &frame->size, err); + + /* + * Copy the floating point registers. There can be unused + * registers see asm/hwcap.h for details. + */ + err |= __copy_to_user(&frame->ufp.fpregs, &h->fpregs, + sizeof(h->fpregs)); + /* + * Copy the status and control register. + */ + __put_user_error(h->fpscr, &frame->ufp.fpscr, err); + + /* + * Copy the exception registers. + */ + __put_user_error(h->fpexc, &frame->ufp_exc.fpexc, err); + __put_user_error(h->fpinst, &frame->ufp_exc.fpinst, err); + __put_user_error(h->fpinst2, &frame->ufp_exc.fpinst2, err); + + return err ? -EFAULT : 0; +} + +static int restore_vfp_context(struct vfp_sigframe __user *frame) +{ + struct thread_info *thread = current_thread_info(); + struct vfp_hard_struct *h = &thread->vfpstate.hard; + unsigned long magic; + unsigned long size; + unsigned long fpexc; + int err = 0; + + __get_user_error(magic, &frame->magic, err); + __get_user_error(size, &frame->size, err); + + if (err) + return -EFAULT; + if (magic != VFP_MAGIC || size != VFP_STORAGE_SIZE) + return -EINVAL; + + /* + * Copy the floating point registers. There can be unused + * registers see asm/hwcap.h for details. + */ + err |= __copy_from_user(&h->fpregs, &frame->ufp.fpregs, + sizeof(h->fpregs)); + /* + * Copy the status and control register. + */ + __get_user_error(h->fpscr, &frame->ufp.fpscr, err); + + /* + * Sanitise and restore the exception registers. + */ + __get_user_error(fpexc, &frame->ufp_exc.fpexc, err); + /* Ensure the VFP is enabled. */ + fpexc |= FPEXC_EN; + /* Ensure FPINST2 is invalid and the exception flag is cleared. */ + fpexc &= ~(FPEXC_EX | FPEXC_FP2V); + h->fpexc = fpexc; + + __get_user_error(h->fpinst, &frame->ufp_exc.fpinst, err); + __get_user_error(h->fpinst2, &frame->ufp_exc.fpinst2, err); + + if (!err) + vfp_flush_hwstate(thread); + + return err ? -EFAULT : 0; +} + +#endif + /* * Do a signal return; undo the signal stack. These are aligned to 64-bit. */ @@ -233,8 +318,8 @@ static int restore_sigframe(struct pt_regs *regs, struct sigframe __user *sf) err |= restore_iwmmxt_context(&aux->iwmmxt); #endif #ifdef CONFIG_VFP -// if (err == 0) -// err |= vfp_restore_state(&sf->aux.vfp); + if (err == 0) + err |= restore_vfp_context(&aux->vfp); #endif return err; @@ -348,8 +433,8 @@ setup_sigframe(struct sigframe __user *sf, struct pt_regs *regs, sigset_t *set) err |= preserve_iwmmxt_context(&aux->iwmmxt); #endif #ifdef CONFIG_VFP -// if (err == 0) -// err |= vfp_save_state(&sf->aux.vfp); + if (err == 0) + err |= preserve_vfp_context(&aux->vfp); #endif __put_user_error(0, &aux->end_magic, err); -- cgit v1.2.2 From 4260415f6a3b92c5c986398d96c314df37a4ccbf Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 19 Apr 2010 10:15:03 +0100 Subject: ARM: fix build error in arch/arm/kernel/process.c /tmp/ccJ3ssZW.s: Assembler messages: /tmp/ccJ3ssZW.s:1952: Error: can't resolve `.text' {.text section} - `.LFB1077' This is caused because: .section .data .section .text .section .text .previous does not return us to the .text section, but the .data section; this makes use of .previous dangerous if the ordering of previous sections is not known. Fix up the other users of .previous; .pushsection and .popsection are a safer pairing to use than .section and .previous. Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 10 +++++----- arch/arm/kernel/ftrace.c | 8 ++++---- arch/arm/kernel/process.c | 8 ++++---- 3 files changed, 13 insertions(+), 13 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 6c5cf369183b..e6a0fb0f392e 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -523,16 +523,16 @@ ENDPROC(__und_usr) /* * The out of line fixup for the ldrt above. */ - .section .fixup, "ax" + .pushsection .fixup, "ax" 4: mov pc, r9 - .previous - .section __ex_table,"a" + .popsection + .pushsection __ex_table,"a" .long 1b, 4b #if __LINUX_ARM_ARCH__ >= 7 .long 2b, 4b .long 3b, 4b #endif - .previous + .popsection /* * Check whether the instruction is a co-processor instruction. @@ -679,7 +679,7 @@ do_fpe: .data ENTRY(fp_enter) .word no_fp - .previous + .text ENTRY(no_fp) mov pc, lr diff --git a/arch/arm/kernel/ftrace.c b/arch/arm/kernel/ftrace.c index c63842766229..0298286ad4ad 100644 --- a/arch/arm/kernel/ftrace.c +++ b/arch/arm/kernel/ftrace.c @@ -62,15 +62,15 @@ int ftrace_modify_code(unsigned long pc, unsigned char *old_code, " movne %0, #2 \n" "3:\n" - ".section .fixup, \"ax\"\n" + ".pushsection .fixup, \"ax\"\n" "4: mov %0, #1 \n" " b 3b \n" - ".previous\n" + ".popsection\n" - ".section __ex_table, \"a\"\n" + ".pushsection __ex_table, \"a\"\n" " .long 1b, 4b \n" " .long 2b, 4b \n" - ".previous\n" + ".popsection\n" : "=r"(err), "=r"(replaced) : "r"(pc), "r"(new), "r"(old), "0"(err), "1"(replaced) diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 0e12e0acbf26..acf5e6fdb6dc 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -355,7 +355,7 @@ EXPORT_SYMBOL(dump_fpu); * the thread function, and r3 points to the exit function. */ extern void kernel_thread_helper(void); -asm( ".section .text\n" +asm( ".pushsection .text\n" " .align\n" " .type kernel_thread_helper, #function\n" "kernel_thread_helper:\n" @@ -363,11 +363,11 @@ asm( ".section .text\n" " mov lr, r3\n" " mov pc, r2\n" " .size kernel_thread_helper, . - kernel_thread_helper\n" -" .previous"); +" .popsection"); #ifdef CONFIG_ARM_UNWIND extern void kernel_thread_exit(long code); -asm( ".section .text\n" +asm( ".pushsection .text\n" " .align\n" " .type kernel_thread_exit, #function\n" "kernel_thread_exit:\n" @@ -377,7 +377,7 @@ asm( ".section .text\n" " nop\n" " .fnend\n" " .size kernel_thread_exit, . - kernel_thread_exit\n" -" .previous"); +" .popsection"); #else #define kernel_thread_exit do_exit #endif -- cgit v1.2.2 From 124efc27a7090d4aaab68b28f7e7a5137f4ecec9 Mon Sep 17 00:00:00 2001 From: Santosh Shilimkar Date: Fri, 30 Apr 2010 10:45:46 +0100 Subject: ARM: 6068/1: Fix build break with KPROBES enabled With CONFIG_KPROBES enabled two section are getting created which leads to below build break. LOG: AS arch/arm/kernel/entry-armv.o arch/arm/kernel/entry-armv.S: Assembler messages: arch/arm/kernel/entry-armv.S:431: Error: symbol ret_from_exception is in a different section arch/arm/kernel/entry-armv.S:490: Error: symbol ret_from_exception is in a different section arch/arm/kernel/entry-armv.S:491: Error: symbol __und_usr_unknown is in a different section This was introduced by commit 4260415f6a3b92c5c986398d96c314df37a4ccbf Reported-by: Anand Gadiyar Signed-off-by: Santosh Shilimkar Signed-off-by: Russell King --- arch/arm/kernel/entry-armv.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index e6a0fb0f392e..7ee48e7f8f31 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -676,10 +676,10 @@ do_fpe: * lr = unrecognised FP instruction return address */ - .data + .pushsection .data ENTRY(fp_enter) .word no_fp - .text + .popsection ENTRY(no_fp) mov pc, lr -- cgit v1.2.2 From 13ea9cc82138691856d7cd855dff9aef1479adb9 Mon Sep 17 00:00:00 2001 From: Santosh Shilimkar Date: Fri, 30 Apr 2010 06:51:20 +0100 Subject: ARM: 6066/1: Fix "BUG: scheduling while atomic: swapper/0/0x00000002 This patch fixes the preempt leak in the cpuidle path invoked from cpu-hotplug. The fix is suggested by Russell King and is based on x86 idea of calling init_idle() on the idle task when it's re-used which also resets the preempt count amongst other things dump: BUG: scheduling while atomic: swapper/0/0x00000002 Modules linked in: Backtrace: [] (dump_backtrace+0x0/0x110) from [] (dump_stack+0x18/0x1c) r7:c02149e4 r6:c033df00 r5:c7836000 r4:00000000 [] (dump_stack+0x0/0x1c) from [] (__schedule_bug+0x60/0x70) [] (__schedule_bug+0x0/0x70) from [] (schedule+0x98/0x7b8) r5:c7836000 r4:c7836000 [] (schedule+0x0/0x7b8) from [] (cpu_idle+0xb4/0xd4) # [] (cpu_idle+0x0/0xd4) from [] (secondary_start_kernel+0xe0/0xf0) r5:c7836000 r4:c0205f40 [] (secondary_start_kernel+0x0/0xf0) from [] (prm_rmw_mod_reg_bits+0x88/0xa4) r7:c02149e4 r6:00000001 r5:00000001 r4:c7836000 Backtrace aborted due to bad frame pointer Cc: Catalin Marinas Signed-off-by: Santosh Shilimkar Signed-off-by: Russell King --- arch/arm/kernel/smp.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 577543f3857f..a01194e583ff 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -86,6 +86,12 @@ int __cpuinit __cpu_up(unsigned int cpu) return PTR_ERR(idle); } ci->idle = idle; + } else { + /* + * Since this idle thread is being re-used, call + * init_idle() to reinitialize the thread structure. + */ + init_idle(idle, cpu); } /* -- cgit v1.2.2 From f4a27aed48584b9e4699e9ee76288618a8574d77 Mon Sep 17 00:00:00 2001 From: Srinidhi Kasagar Date: Wed, 12 May 2010 05:52:18 +0100 Subject: ARM: 6125/1: ARM TWD: move TWD registers to common header This moves the TWD register set of MPcore to a common existing file so that watchdog driver can access it Signed-off-by: srinidhi kasagar Acked-by: Linus Walleij Signed-off-by: Russell King --- arch/arm/kernel/smp_twd.c | 17 ----------------- 1 file changed, 17 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/smp_twd.c b/arch/arm/kernel/smp_twd.c index ea02a7b1c244..7c5f0c024db7 100644 --- a/arch/arm/kernel/smp_twd.c +++ b/arch/arm/kernel/smp_twd.c @@ -21,23 +21,6 @@ #include #include -#define TWD_TIMER_LOAD 0x00 -#define TWD_TIMER_COUNTER 0x04 -#define TWD_TIMER_CONTROL 0x08 -#define TWD_TIMER_INTSTAT 0x0C - -#define TWD_WDOG_LOAD 0x20 -#define TWD_WDOG_COUNTER 0x24 -#define TWD_WDOG_CONTROL 0x28 -#define TWD_WDOG_INTSTAT 0x2C -#define TWD_WDOG_RESETSTAT 0x30 -#define TWD_WDOG_DISABLE 0x34 - -#define TWD_TIMER_CONTROL_ENABLE (1 << 0) -#define TWD_TIMER_CONTROL_ONESHOT (0 << 1) -#define TWD_TIMER_CONTROL_PERIODIC (1 << 1) -#define TWD_TIMER_CONTROL_IT_ENABLE (1 << 2) - /* set up by the platform code */ void __iomem *twd_base; -- cgit v1.2.2