原创 宋宝华 Linux阅码场 2018-04-01
2012年的文章,重新在微信公众号发表。 第一章: 硬实时Linux(RT-Preempt Patch)在PC上的编译、使用和测试 第二章: 硬实时Linux(RT-Preempt Patch)的中断线程化
硬实时Linux(RT-Preempt Patch)在PC上的编译、使用和测试
Vanilla kernel的问题 Linux kernel在spinlock、irq上下文方面无法抢占,因此高优先级任务被唤醒到得以执行的时间并不能完全确定。同时,Linux kernel本身也不处理优先级反转。RT-Preempt Patch是在Linux社区kernel的基础上,加上相关的补丁,以使得Linux满足硬实时的需求。本文描述了该patch在PC上的实践。我们的 测试环境为Ubuntu 10.10,默认情况下使用Ubuntu 10.10自带的kernel: 在Ubuntu 10.10,apt-get install rt-tests安装rt测试工具集,运行其中的cyclictest测试工具,默认创建5个SCHED_FIFO策略的realtime线程,优先级 76-80,运行周期是1000,1500,2000,2500,3000微秒: 由此可见在标准Linux内,rt线程投入运行的jitter非常不稳定,最小值在26-37微秒,平均值为68-889微秒,而最大值则分布在9481-13673微秒之间。 我们还是运行这个测试,但是在运行这个测试的过程中引入更多干扰,如mount /dev/sdb1 ~/development,则结果变为: mount过程中引入的irq、softirq和spinlock导致最大jitter明显地加大甚至达到了331482us,充分显示出了标准Linux内核中RT线程投入运行时间的不可预期性(硬实时要求意味着可预期)。 如果我们编译一份kernel,选择的是“Voluntary Kernel Preemption (Desktop)“,这类似于2.4不支持kernel抢占的情况,我们运行同样的case,时间的不确定性大地几乎让我们无法接受: RT-Preempt Patch使能 RT-Preempt Patch对Linux kernel的主要改造包括:
* Making in-kernel locking-primitives (using spinlocks) preemptible though reimplementation with rtmutexes:
* Critical sections protected by i.e. spinlock_t and rwlock_t are now preemptible. The creation of non-preemptible sections (in kernel) is still possible with raw_spinlock_t (same APIs like spinlock_t)
* Implementing priority inheritance for in-kernel spinlocks and semaphores. For more information on priority inversion and priority inheritance please consult Introduction to Priority Inversion (http://www.embedded.com/electronics-blogs/beginner-s-corner/4023947/Introduction-to-Priority-Inversion)
* Converting interrupt handlers into preemptible kernel threads: The RT-Preempt patch treats soft interrupt handlers in kernel thread context, which is represented by a task_struct like a common userspace process. However it is also possible to register an IRQ in kernel context.
* Converting the old Linux timer API into separate infrastructures for high resolution kernel timers plus one for timeouts, leading to userspace POSIX timers with high resolution.
在本试验中,我们取的带RT- Preempt Patch的kernel tree是git://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-stable- rt.git,使用其v3.4-rt-rebase branch,编译kernel时选中了"Fully Preemptible Kernel"抢占模型: make modules_install、make install、mkintramfs后,我们得到一个可以在Ubuntu中启动的RT kernel。具体编译方法可详见http://www.linuxidc.com/Linux/2012-01/50749.htm,根据该文修改版本 号等信息即可,我们运行的命令包括: 安装模块 安装kernel
barry@barry-VirtualBox:~/development/linux-2.6$ sudo make install
sh /home/barry/development/linux-2.6/arch/x86/boot/install.sh 3.4.11-rt19 arch/x86/boot/bzImage \
System.map "/boot"
制作initrd
barry@barry-VirtualBox:~/development/linux-2.6$ sudo mkinitramfs 3.4.11-rt19 -o /boot/initrd.img-3.4.11-rt19
修改grub配置
在grub.conf中增加新的启动entry,仿照现有的menuentry,增加一个新的,把其中的相关版本号都变更为3.4.11-rt19,我们的修改如下:
menuentry 'Ubuntu, with Linux 3.4.11-rt19' --class ubuntu --class gnu-linux --class gnu --class os {
recordfail
insmod part_msdos
insmod ext2
set root='(hd0,msdos1)'
search --no-floppy --fs-uuid --set a0db5cf0-6ce3-404f-9808-88ce18f0177a
linux /boot/vmlinuz-3.4.11-rt19 root=UUID=a0db5cf0-6ce3-404f-9808-88ce18f0177a ro quiet splash
initrd /boot/initrd.img-3.4.11-rt19
}
开机时选择3.4.11-rt19启动: RT-Preempt Patch试用 运行同样的测试cyclictest benchmark工具,结果迥异: 我们还是运行这个测试,但是在运行这个测试的过程中引入更多干扰,如mount /dev/sdb1 ~/development,则结果变为: 时间在可预期的范围内,没有出现标准kernel里面jitter达到331482的情况。需要说明的是,这个jitter大到超过了我们的预期,达到了10ms量级,相信是受到了我们的测试都是在Virtualbox虚拟机进行的影响。按照其他文档显示,这个jitter应该在数十us左右。 我们在这个kernel里面运行ps aux命令,可以看出线程化了的irq: 在其中编写一个RT 线程的应用程序,通常需要如下步骤:
- Setting a real time scheduling policy and priority.
- Locking memory so that page faults caused by virtual memory will not undermine deterministic behavior
- Pre-faulting the stack, so that a future stack fault will not undermine deterministic behavior 例 子test_rt.c,其中的mlockall是为了防止进程的虚拟地址空间对应的物理页面被swap出去,而stack_prefault()则故意提 前导致stack往下增长8KB,因此其后的函数调用和局部变量的使用将不再导致栈增长(依赖于page fault和内存申请):
#include <stdlib.h>
#include <stdio.h>
#include <time.h>
#include <sched.h>
#include <sys/mman.h>
#include <string.h>
#define MY_PRIORITY (49) /* we use 49 as the PRREMPT_RT use 50
as the priority of kernel tasklets
and interrupt handler by default */
#define MAX_SAFE_STACK (8*1024) /* The maximum stack size which is
guaranteed safe to access without
faulting */
#define NSEC_PER_SEC (1000000000) /* The number of nsecs per sec. */
void stack_prefault(void) {
unsigned char dummy[MAX_SAFE_STACK];
memset(dummy, 0, MAX_SAFE_STACK);
return;
}
int main(int argc, char* argv[])
{
struct timespec t;
struct sched_param param;
int interval = 50000; /* 50us*/
/* Declare ourself as a real time task */
param.sched_priority = MY_PRIORITY;
if(sched_setscheduler(0, SCHED_FIFO, ¶m) == -1) {
perror("sched_setscheduler failed");
exit(-1);
}
/* Lock memory */
if(mlockall(MCL_CURRENT|MCL_FUTURE) == -1) {
perror("mlockall failed");
exit(-2);
}
/* Pre-fault our stack */
stack_prefault();
clock_gettime(CLOCK_MONOTONIC ,&t);
/* start after one second */
t.tv_sec++;
while(1) {
/* wait until next shot */
clock_nanosleep(CLOCK_MONOTONIC, TIMER_ABSTIME, &t, NULL);
/* do the stuff */
/* calculate next shot */
t.tv_nsec += interval;
while (t.tv_nsec >= NSEC_PER_SEC) {
t.tv_nsec -= NSEC_PER_SEC;
t.tv_sec++;
}
}
}
编译之:gcc -o test_rt test_rt.c -lrt。本节就到这里,后续我们会有一系列博文来描述RT-Preempt Patch对kernel的主要改动,以及其工作原理。
硬实时Linux(RT-Preempt Patch)的中断线程化
底半部:线程化IRQ 线程化中断的支持在2009年已经进入Linux官方内核,详见Thomas Gleixner的patch: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=3aa551c9b4c40018f0e261a178e3d25478dc04a9 该patch提供一个能力,驱动可以通过 申请一个线程化的IRQ,kernel会为中断的底半部创建一个名字为irq/%d-%s的线程,%d对应着中断号。其中顶半部(硬中断)handler在做完必要的处理工作之后,会返回IRQ_WAKE_THREAD,之后kernel会唤醒irq/%d-%s线程,而该kernel线程会调用thread_fn函数,因此,该线程成为底半部。在后续维护的过程中,笔者曾参与进一步完善该功能的讨论,后续patch包括nested、oneshot等的支持,详见patch: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=399b5da29b9f851eb7b96e2882097127f003e87c http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=70aedd24d20e75198f5a0b11750faabbb56924e2 http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=b25c340c195447afb1860da580fe2a85a6b652c5 该机制目前在kernel中使用已经十分广泛,可以认为是继softirq(含tasklet)和workqueue之后的又一大中断底半部方式。 顶半部:强制线程化 在使能Linux RT-Preempt后,默认情况下会强制透过request_irq()申请的IRQ的顶半部函数在线程中执行,我们都知道request_irq的原型为: 这意味着通过request_irq()申请的IRQ,在没有Rt-Preepmt的情况下,kernel并不会为其创建irq线程,因为它在最终调用request_threaded_irq()的时候传递的thread_fn为NULL。 如果使能了RT-Preempt Patch的情况下,其中的genirq-force-threading.patch会强制ARM使用threaded irq: 在RT-Preempt Patch中,会针对使能了IRQ_FORCED_THREADING的情况,对这一原先没有线程化IRQ的case进行强制线程化,代码见_ _setup_irq():
887 static int
888 __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
889 {
890 ...
903
904 /*
905 * Check whether the interrupt nests into another interrupt
906 * thread.
907 */
908 nested = irq_settings_is_nested_thread(desc);
909 if (nested) {
910 ...
920 } else {
921 if (irq_settings_can_thread(desc))
922 irq_setup_forced_threading(new);
923 }
925 /*
926 * Create a handler thread when a thread function is supplied
927 * and the interrupt does not nest into another interrupt
928 * thread.
929 */
930 if (new->thread_fn && !nested) {
931 struct task_struct *t;
932
933 t = kthread_create(irq_thread, new, "irq/%d-%s", irq,
934 new->name);
935 ...
939 /*
940 * We keep the reference to the task struct even if
941 * the thread dies to avoid that the interrupt code
942 * references an already freed task_struct.
943 */
944 get_task_struct(t);
945 new->thread = t;
946 }
我们重点看一下其中的922行:
第878行和879行,强制将原先的handler复制给thread_fn,而又强制把原来的handler变更为irq_default_primary_handler(),而这个函数,其实神马都不做,只是直接返回IRQ_WAKE_THREAD: 第874的IRQF_ONESHOT就用到了我们前面说的oneshot功能。 所以,RT-Preempt实际上是把原先的顶半部底半部化了,而现在伪造了一个假的顶半部,它只是直接返回一个IRQ_WAKE_THREAD标记而已。 我们来看一下一个中断发生后,Linux RT-Preempt处理的全过程,首先是会跳到 arch/arm/kernel/entry-armv.S arch/arm/include/asm/entry-macro-multi.S 中的汇编入口,再进入arm/kernel/irq.c下的asm_do_IRQ 、handle_IRQ,之后generic的handle_irq_event_percpu()被调用:
133 handle_irq_event_percpu(struct irq_desc *desc, struct irqaction *action)
134 {
135 irqreturn_t retval = IRQ_NONE;
136 unsigned int flags = 0, irq = desc->irq_data.irq;
137
138 do {
139 irqreturn_t res;
140
141 trace_irq_handler_entry(irq, action);
142 res = action->handler(irq, action->dev_id);
143 trace_irq_handler_exit(irq, action, res);
144
145 if (WARN_ONCE(!irqs_disabled(),"irq %u handler %pF enabled interrupts\n",
146 irq, action->handler))
147 local_irq_disable();
148
149 switch (res) {
150 case IRQ_WAKE_THREAD:
151 /*
152 * Catch drivers which return WAKE_THREAD but
153 * did not set up a thread function
154 */
155 if (unlikely(!action->thread_fn)) {
156 warn_no_thread(irq, action);
157 break;
158 }
159
160 irq_wake_thread(desc, action);
161
162 /* Fall through to add to randomness */
163 case IRQ_HANDLED:
164 flags |= action->flags;
165 break;
166
167 default:
我们关注其中的第142行,本质上是调用irq_default_primary_handler(),接到150行,由于 irq_default_primary_handler()返回了IRQ_WAKE_THREAD,因此,generic的中断处理流程会执行 irq_wake_thread(desc, action);去唤醒前面的irq/%d-%s线程,该线程的代码是
789 static int irq_thread(void *data)
790 {
791 static const struct sched_param param = {
792 .sched_priority = MAX_USER_RT_PRIO/2,
793 };
794 struct irqaction *action = data;
795 struct irq_desc *desc = irq_to_desc(action->irq);
796 irqreturn_t (*handler_fn)(struct irq_desc *desc,
797 struct irqaction *action);
798
799 if (force_irqthreads && test_bit(IRQTF_FORCED_THREAD,
800 &action->thread_flags))
801 handler_fn = irq_forced_thread_fn;
802 else
803 handler_fn = irq_thread_fn;
804
805 sched_setscheduler(current, SCHED_FIFO, ¶m);
806 current->irq_thread = 1;
807
808 while (!irq_wait_for_interrupt(action)) {
809 irqreturn_t action_ret;
810
811 irq_thread_check_affinity(desc, action);
812
813 action_ret = handler_fn(desc, action);
814 if (!noirqdebug)
815 note_interrupt(action->irq, desc, action_ret);
816
817 wake_threads_waitq(desc);
818 }
819
820 /*
821 * This is the regular exit path. __fr
其中的813行会调用最终的被赋值给thread_fn的原来的handler,这样原来的中断顶半部就整个在irq_thread里面执行了,实现了所谓的顶半部的线程化。 绕开顶半部线程化 当然,在使能了RT-Preempt的情况之下,我们仍然可以绕开顶半部线程化的过程,避免前面的强势变更,只需要申请中断的时候设置IRQ_NOTHREAD标志,如其中的patch:
Subject: arm: Mark pmu interupt IRQF_NO_THREAD
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 16 Mar 2011 14:45:31 +0100
PMU interrupt must not be threaded. Remove IRQF_DISABLED while at it
as we run all handlers with interrupts disabled anyway.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/kernel/perf_event.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Index: linux-stable/arch/arm/kernel/perf_event.c
===================================================================
--- linux-stable.orig/arch/arm/kernel/perf_event.c
+++ linux-stable/arch/arm/kernel/perf_event.c
@@ -430,7 +430,7 @@ armpmu_reserve_hardware(struct arm_pmu *
}
err = request_irq(irq, handle_irq,
- IRQF_DISABLED | IRQF_NOBALANCING,
+ IRQF_NOBALANCING | IRQF_NO_THREAD,
"arm-pmu", armpmu);
if (err) {
r_err("unable to request IRQ%d for ARM PMU counters\n",
```