提交 7ada189f 编写于 作者: J Jamie Iles 提交者: Russell King

ARM: 5900/2: arm: enable support for software perf events

The perf events subsystem allows counting of both hardware and
software events. This patch implements the bare minimum for software
performance events.

Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NJamie Iles <jamie.iles@picochip.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
上级 1618fdd9
......@@ -18,6 +18,8 @@ config ARM
select HAVE_KRETPROBES if (HAVE_KPROBES)
select HAVE_FUNCTION_TRACER if (!XIP_KERNEL)
select HAVE_GENERIC_DMA_COHERENT
select HAVE_PERF_EVENTS
select PERF_USE_VMALLOC
help
The ARM series is a line of low-power-consumption RISC chip designs
licensed by ARM Ltd and targeted at embedded applications and
......
/*
* linux/arch/arm/include/asm/perf_event.h
*
* 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.
*
*/
#ifndef __ARM_PERF_EVENT_H__
#define __ARM_PERF_EVENT_H__
/*
* NOP: on *most* (read: all supported) ARM platforms, the performance
* counter interrupts are regular interrupts and not an NMI. This
* means that when we receive the interrupt we can call
* perf_event_do_pending() that handles all of the work with
* interrupts enabled.
*/
static inline void
set_perf_event_pending(void)
{
}
/* ARM performance counters start from 1 (in the cp15 accesses) so use the
* same indexes here for consistency. */
#define PERF_EVENT_INDEX_OFFSET 1
#endif /* __ARM_PERF_EVENT_H__ */
......@@ -18,6 +18,7 @@
#include <linux/page-flags.h>
#include <linux/sched.h>
#include <linux/highmem.h>
#include <linux/perf_event.h>
#include <asm/system.h>
#include <asm/pgtable.h>
......@@ -302,6 +303,12 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
fault = __do_page_fault(mm, addr, fsr, tsk);
up_read(&mm->mmap_sem);
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, addr);
if (fault & VM_FAULT_MAJOR)
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0, regs, addr);
else if (fault & VM_FAULT_MINOR)
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0, regs, addr);
/*
* Handle the "normal" case first - VM_FAULT_MAJOR / VM_FAULT_MINOR
*/
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册