这是一个老生常谈的问题。
我们先简单说下什么是中断「因为最近在群里看到有人竟然不懂什么是中断」。中断是计算机里面非常核心的东西，我们可以跑OS，可以多任务运行都因为中断的存在。
假设你是一个CPU，你正在睡觉。
你突然觉得肚子疼，「这个事情就是一个中断你睡觉的事情」，然后你就去上厕所了。
然后你又回到床上睡觉。
你突然又觉得肚子饿了，「这个事情也是一个中断你睡觉的事情」，然后你就去吃了个汉堡。
中断可以认为是突发的事情，并且时间不是特别长，如果时间非常久，那就不能叫中断了，就叫主业了。
CPU怎么用中断做到多任务运行呢？
CPU不是普通的人，CPU比普通的人快得多得多。
就拿我手上最差的新唐单片机举例子，他的主频是24M，那执行一条指令周期的时间就是
1/24000000 = 41.6 纳秒
假设你现在有两个事情要做，玩王者荣耀和扫地
你在第一个时钟周期，拿着手机玩王者荣耀，第二个时钟周期扫地，第三个时钟周期玩王者荣耀，第四个时钟周期扫地……
这样，你会发现，CPU是一个闪电侠，可能比闪电侠还要快，根据这个方法，可以裂变出很多很多个人做很多很多件事情。
Linux 内核中断上文为什么不能休眠呢？
现在的Linux 内核中断是不能嵌套的，所以我们就只讨论单中断的问题。
进中断的第一件事情，就是关中断，这个是关所有中断。
然后就在中断里面干事情了，具体做什么事情，我也不知道。
做完事情然后就开中断，去做其他事情了。
如果休眠会引起什么呢？
比如这样的代码
static irqreturn_t tpd_eint_interrupt_handler(unsigned irq, struct irq_desc *desc)
{TPD_DEBUG_PRINT_INT;tpd_flag = 1;/* enter EINT handler disable INT, make sure INT is disable when handle touch event including top/bottom half *//* use _nosync to avoid deadlock */spin_lock(&irq_lock);if (tpd_irq_flag) {tpd_irq_flag = 0;disable_irq_nosync(tpd_touch_irq);}spin_unlock(&irq_lock);ssleep(1);//系统会死掉wake_up_interruptible(&waiter);return IRQ_HANDLED;
}
这个代码的问题在ssleep(1)上，ssleep(1)是OS里面的休眠函数，这个函数会让任务休眠，休眠的意思就是让这个任务睡觉不执行，CPU去执行其他的任务。
因为CPU是按时间片执行的，等下一次调度到这个中断的时候，就有可能再获取一次spin_lock自旋锁，两次获取锁的操作，肯定就引起系统挂壁了。
具体的可以去看看ssleep的代码
/*** msleep - sleep safely even with waitqueue interruptions* @msecs: Time in milliseconds to sleep for*/
void msleep(unsigned int msecs)
{unsigned long timeout = msecs_to_jiffies(msecs) + 1;while (timeout)timeout = schedule_timeout_uninterruptible(timeout);
}EXPORT_SYMBOL(msleep);
系统宕机日志
[ 1.564767] <3>.(3)[139:kworker/u8:2]<<GTP-INF>>[tpd_irq_registration:484] Device Tree Tpd_irq_registration!
[ 1.566520] <3>.(3)[139:kworker/u8:2]gpiod_set_debounce: invalid GPIO
[ 1.567319] <3>.(3)[139:kworker/u8:2]<<GTP-INF>>[tpd_irq_registration:495] Device gt1x_int_type = 1!
[ 1.568484] <0>-(0)[0:swapper/0]BUG: scheduling while atomic: swapper/0/0/0x00010002
[ 1.568493] <0>.(3)[139:kworker/u8:2]<<GTP-INF>>[tpd_irq_registration:514] irq:119, debounce:0-0:
[ 1.570534] Modules linked in:
[ 1.570912] <0>-(0)[0:swapper/0]CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.4.146 #41
[ 1.571879] <0>-(0)[0:swapper/0]Hardware name: Generic DT based system
[ 1.572689] Backtrace:
[ 1.572989] <0>-(0)[0:swapper/0][<c010d5d4>] (dump_backtrace) from [<c010d7fc>] (show_stack+0x18/0x1c)
[ 1.574143] r6:60000193 r5:c1136bc8 r4:00000000 r3:dc8ba692
[ 1.574841] <0>-(0)[0:swapper/0][<c010d7e4>] (show_stack) from [<c041b4c4>] (dump_stack+0x94/0xa8)
[ 1.575950] <0>-(0)[0:swapper/0][<c041b430>] (dump_stack) from [<c0155350>] (__schedule_bug+0x58/0x6c)
[ 1.577104] r6:c1101ce8 r5:c1085b40 r4:00000000 r3:60000193
[ 1.577802] <0>-(0)[0:swapper/0][<c01552f8>] (__schedule_bug) from [<c0b6dc88>] (__schedule+0x59c/0x808)
[ 1.578978] r4:c110bcf8 r3:00010002
[ 1.579418] <0>-(0)[0:swapper/0][<c0b6d6ec>] (__schedule) from [<c0b6df48>] (schedule+0x54/0xc4)
[ 1.580507] r10:00000000 r9:c1102100 r8:dfedfb80 r7:c1103948
[ 1.581216] <0>-(0)[0:swapper/0][<c0b6def4>] (schedule) from [<c0b712c4>] (schedule_timeout+0x148/0x264)
[ 1.582391] r4:fffea370 r3:c1100000
[ 1.582831] <0>-(0)[0:swapper/0][<c0b7117c>] (schedule_timeout) from [<c01ad8f8>] (msleep+0x34/0x40)
[ 1.583964] r10:c1100000 r9:00000000 r8:00000000 r7:00000077
[ 1.584673] <0>-(0)[0:swapper/0][<c01ad8c4>] (msleep) from [<c081e0d0>] (tpd_eint_interrupt_handler+0x2c/0xd8)
[ 1.585913] r5:ced00000 r4:cec8d540
[ 1.586353] <0>-(0)[0:swapper/0][<c081e0a4>] (tpd_eint_interrupt_handler) from [<c0199684>] (handle_irq_event_percpu+0x6c/0x268)
[ 1.587787] r5:d0344ae0 r4:cec8d540
[ 1.588227] <0>-(0)[0:swapper/0][<c0199618>] (handle_irq_event_percpu) from [<c01998c0>] (handle_irq_event+0x40/0x64)
[ 1.589542] r10:c0c487cc r9:00000077 r8:00000000 r7:00000000
[ 1.590251] <0>-(0)[0:swapper/0][<c0199880>] (handle_irq_event) from [<c019d388>] (handle_level_irq+0xb4/0x16c)
[ 1.591502] r6:d0344a90 r5:d0344ae0 r4:d0344a80 r3:00020002
[ 1.592201] <0>-(0)[0:swapper/0][<c019d2d4>] (handle_level_irq) from [<c0198b64>] (generic_handle_irq+0x2c/0x3c)
[ 1.593462] r6:d030e6d0 r5:02000000 r4:00000019 r3:c019d2d4
[ 1.594161] <0>-(0)[0:swapper/0][<c0198b38>] (generic_handle_irq) from [<c045d7e4>] (mtk_eint_irq_handler+0x288/0x36c)
[ 1.595486] <0>-(0)[0:swapper/0][<c045d55c>] (mtk_eint_irq_handler) from [<c0198b64>] (generic_handle_irq+0x2c/0x3c)
[ 1.596790] r10:00000001 r9:00000000 r8:00010001 r7:00000000
[ 1.597499] <0>-(0)[0:swapper/0][<c0198b38>] (generic_handle_irq) from [<c0198e7c>] (__handle_domain_irq+0xc0/0x25c)
[ 1.598802] <0>-(0)[0:swapper/0][<c0198dbc>] (__handle_domain_irq) from [<c01015ec>] (gic_handle_irq+0x50/0x94)
[ 1.600053] r9:e1003000 r8:e1002000 r7:c1101eb0 r6:e100200c
[ 1.600752] <0>-(0)[0:swapper/0][<c010159c>] (gic_handle_irq) from [<c010e314>] (__irq_svc+0x54/0x90)
[ 1.601893] <0>-(0)[0:swapper/0]Exception stack(0xc1101eb0 to 0xc1101ef8)
[ 1.602736] <0>-(0)[0:swapper/0]1ea0: 00000000 ffffffff 1ee35000 00000000
[ 1.603965] <0>-(0)[0:swapper/0]1ec0: c11ac9b4 cfbdf800 5cc9dbdf 00000000 5ccecf4c 00000000 00000001 c1101f44
[ 1.605192] <0>-(0)[0:swapper/0]1ee0: c1101f00 c1101f00 c08a2380 c08a2390 60000013 ffffffff
[ 1.606224] r9:00000000 r8:5ccecf4c r7:c1101ee4 r6:ffffffff
[ 1.606922] <0>-(0)[0:swapper/0][<c08a2220>] (cpuidle_enter_state) from [<c08a2670>] (cpuidle_enter+0x1c/0x20)
[ 1.608162] r10:c1103a1c r9:c1187308 r8:cfbdf800 r7:c11ac9b4
[ 1.608872] <0>-(0)[0:swapper/0][<c08a2654>] (cpuidle_enter) from [<c0183334>] (cpu_startup_entry+0x1e0/0x3d8)
[ 1.610110] <0>-(0)[0:swapper/0][<c0183154>] (cpu_startup_entry) from [<c0b6ac18>] (rest_init+0x90/0x94)
[ 1.611285] r7:ffffffff
[ 1.611596] <0>-(0)[0:swapper/0][<c0b6ab88>] (rest_init) from [<c1000e44>] (start_kernel+0x418/0x468)
[ 1.612739] r4:c11b1040 r3:dc8ba692
[ 1.613179] <0>-(0)[0:swapper/0]Backtrace aborted due to bad pc <c1000a30>
[ 1.614040] <0>-(0)[0:swapper/0]------------[ cut here ]------------
[ 1.614816] <0>-(0)[0:swapper/0]Kernel BUG at c0155360 [verbose debug info unavailable]
[ 1.615807] <0>-(0)[0:swapper/0]Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
[ 1.616746] <0>-(0)[0:swapper/0][LY]rtc_mark_aee_kernel_panic!!!
[ 1.617551] disable aee kernel api
[ 1.617935] <0>-(0)[0:swapper/0]Kernel Offset: disabled
[ 1.618603] Modules linked in:
Linux 内核中断函数能不能执行malloc函数？
这个是之前已经发过的文章
可以在中断服务程序执行malloc吗？
内核里面分配内存的函数是kmalloc ,其他不废话，我直接测试了下面的代码
static irqreturn_t tpd_eint_interrupt_handler(unsigned irq, struct irq_desc *desc)
{char *rbuff = NULL;rbuff = kmalloc(1024, GFP_KERNEL);TPD_DEBUG_PRINT_INT;tpd_flag = 1;/* enter EINT handler disable INT, make sure INT is disable when handle touch event including top/bottom half *//* use _nosync to avoid deadlock */spin_lock(&irq_lock);if (tpd_irq_flag) {tpd_irq_flag = 0;disable_irq_nosync(tpd_touch_irq);}spin_unlock(&irq_lock);ssleep(1);//系统会死掉wake_up_interruptible(&waiter);kfree(rbuff);return IRQ_HANDLED;
}
系统是没有挂掉的，然后去看看kmalloc的实现。
/*** kmalloc - allocate memory* @size: how many bytes of memory are required.* @flags: the type of memory to allocate.** kmalloc is the normal method of allocating memory* for objects smaller than page size in the kernel.** The @flags argument may be one of:** %GFP_USER - Allocate memory on behalf of user. May sleep.** %GFP_KERNEL - Allocate normal kernel ram. May sleep.** %GFP_ATOMIC - Allocation will not sleep. May use emergency pools.* For example, use this inside interrupt handlers.** %GFP_HIGHUSER - Allocate pages from high memory.** %GFP_NOIO - Do not do any I/O at all while trying to get memory.** %GFP_NOFS - Do not make any fs calls while trying to get memory.** %GFP_NOWAIT - Allocation will not sleep.** %__GFP_THISNODE - Allocate node-local memory only.** %GFP_DMA - Allocation suitable for DMA.* Should only be used for kmalloc() caches. Otherwise, use a* slab created with SLAB_DMA.** Also it is possible to set different flags by OR'ing* in one or more of the following additional @flags:** %__GFP_COLD - Request cache-cold pages instead of* trying to return cache-warm pages.** %__GFP_HIGH - This allocation has high priority and may use emergency pools.** %__GFP_NOFAIL - Indicate that this allocation is in no way allowed to fail* (think twice before using).** %__GFP_NORETRY - If memory is not immediately available,* then give up at once.** %__GFP_NOWARN - If allocation fails, don't issue any warnings.** %__GFP_REPEAT - If allocation fails initially, try once more before failing.** There are other flags available as well, but these are not intended* for general use, and so are not documented here. For a full list of* potential flags, always refer to linux/gfp.h.*/
static __always_inline void *kmalloc(size_t size, gfp_t flags)
{if (__builtin_constant_p(size)) {if (size > KMALLOC_MAX_CACHE_SIZE)return kmalloc_large(size, flags);
#ifndef CONFIG_SLOBif (!(flags & GFP_DMA)) {int index = kmalloc_index(size);if (!index)return ZERO_SIZE_PTR;return kmem_cache_alloc_trace(kmalloc_caches[index],flags, size);}
#endif}return __kmalloc(size, flags);
}
代码里面没有看到引起调度的代码，但是注释里面有说会引起休眠。
这个函数跟微信微信公众号留言提到的gfp一致。
推荐阅读：
专辑|Linux文章汇总
专辑|程序人生
专辑|C语言
我的知识小密圈
关注公众号，后台回复「1024」获取学习资料网盘链接。
欢迎点赞，关注，转发，在看，您的每一次鼓励，我都将铭记于心~
