首页 > 其他分享 >-XX:MaxDirectMemorySize

-XX:MaxDirectMemorySize

时间:2022-10-28 14:59:28浏览次数:84  
标签:XX 内存 ByteBuffer MaxDirectMemorySize ptr size

 

1. 关于MaxDirectMemorySize的设置

https://www.jianshu.com/p/e1503204a059?u_atoken=95c81aa9-f78f-4fc2-b740-53ce5fef397f&u_asession=01Q8-suMVmBRarjM7BaZydEWImR2F7rLNNHL7zvDUxJjtYq-vQrLc0ILaSMinNxzM9X0KNBwm7Lovlpxjd_P_q4JsKWYrT3W_NKPr8w6oU7K8SAYuKkdP5ggJDCJOpAqUehUF3o-sVtq6Wun3JL3SJe2BkFo3NEHBv0PZUm6pbxQU&u_asig=05Z6NcKy0zBHRKYB83RslIJjEpRZahR2Pn_STZvKXGlpB6ESuoAFqCxa2Pmav9Xhb7jP7UykqqZ_pRDEgU4mX_ZJLB5yf9TJmo9UeeyjCTNFOSR_r0f5pr99y4TQkJEIE-PPFpjrPfVUxwhLOlgDfi80ApPyGxvJJeHMulJxXXMtn9JS7q8ZD7Xtz2Ly-b0kmuyAKRFSVJkkdwVUnyHAIJzSPRPdjWlTtXJ8SH4ysRuoa5PtozGbqYVBvWUQar3T9Jom7nzSzR1LP16f45fIKp-e3h9VXwMyh6PgyDIVSG1W80wJZp_M4CDHDaZyKcCA7nuT4N6mo9qseyRZVP0KJCdI7F-asVQwvQh0NcpOaYaXtDPY3j0zKSjlrUDzqj1ljDmWspDxyAEEo4kbsryBKb9Q&u_aref=GYhJNLU69cEjjA8m3QyIKhbkDZM%3D

 

 

2. JVM参数之MaxDirectMemorySize

 

1、Java_JVM参数-XX:MaxDirectMemorySize 与 两种 ByteBuffer: heap,direct ByteBuffer(参考:https://www.cnblogs.com/laoqing/p/10380536.html)

     ByteBuffer有两种:
     heap ByteBuffer -> -XX:Xmx
     1.1、一种是heap ByteBuffer,该类对象分配在JVM的堆内存里面,直接由Java虚拟机负责垃圾回收;

     direct ByteBuffer -> -XX:MaxDirectMemorySize
     1.2、一种是direct ByteBuffer是通过jni在虚拟机外内存中分配的。通过jmap无法查看该快内存的使用情况。只能通过top来看它的内存使用情况。
           1.2.1、JVM堆内存大小可以通过-Xmx来设置,同样的direct ByteBuffer可以通过-XX:MaxDirectMemorySize来设置,此参数的含义是当Direct ByteBuffer分配的堆外内存到达指定大小后,即触发Full GC。注意该值是有上限的,默认是64M,最大为sun.misc.VM.maxDirectMemory(),在程序中中可以获得-XX:MaxDirectMemorySize的设置的值。
           1.2.2、没有配置MaxDirectMemorySize的,因此MaxDirectMemorySize的大小即等于-Xmx
           1.2.3、Direct Memory的回收机制,Direct Memory是受GC控制的
           1.2.4、对于使用Direct Memory较多的场景,需要注意下MaxDirectMemorySize的设置,避免-Xmx + Direct Memory超出物理内存大小的现象

2、用JDK8的一定要配置:-Xms -Xmx -XX:MaxDirectMemorySize,【Xmx +(加) MaxDirectMemorySize】的值不能超过docker的最大内存,不然docker内存占满了会被oomkill掉;

     没配置参数导致的问题以及处理参考:http://hellojava.info/?tag=maxdirectmemorysize (物理内存耗尽、CMS GC碎片造成RT慢的两个Case
    分析:https://my.oschina.net/go4it/blog/3029481

3、dmesg排查消失的进程:
      3.1、适用场景:

              如果发现自己的java进程悄无声息的消失了,几乎没有留下任何线索,那么dmesg一发,很有可能有你想要的。

      3.2、具体操作
              sudo dmesg|grep -i kill | less 或者 dmesg | grep kill

      去找关键字oom_killer,找到的结果类似如下:
      [6710782.021013] java invoked oom-killer: gfp_mask=0xd0, order=0, oom_adj=0, oom_scoe_adj=0 
      [6710782.070639] [] ? oom_kill_process+0x68/0x140 
      [6710782.257588] Task in /LXC011175068174 killed as a result of limit of /LXC011175068174 
      [6710784.698347] Memory cgroup out of memory: Kill process 215701 (java) score 854 or sacrifice child 
      [6710784.707978] Killed process 215701, UID 679, (java) total-vm:11017300kB, anon-rss:7152432kB, file-rss:1232kB

       以上表明,对应的java进程被系统的OOM Killer给干掉了,得分为854. 
       解释一下OOM killer(Out-Of-Memory killer),该机制会监控机器的内存资源消耗。当机器内存耗尽前,该机制会扫描所有的进程(按照一定规则计算,内存占用,时间等),挑选出得分最高的进程,然后杀死,从而保护机器。

 

       dmesg日志时间转换公式: 
       log实际时间=格林威治1970-01-01+(当前时间秒数-系统启动至今的秒数+dmesg打印的log时间)秒数:
       date -d “1970-01-01 UTC echo "$(date +%s)-$(cat /proc/uptime|cut -f 1 -d' ')+12288812.926194"|bc seconds”
       剩下的,就是看看为什么内存这么大,触发了OOM-Killer了。

4、默认情况下,VM将用于直接字节缓冲区的堆内存量限制为最大堆大小的大约85%

 http://lovestblog.cn/blog/2015/05/12/direct-buffer/

 

附:几种内存溢出处理办法:cnblogs.com/leasonWang/p/11146030.html

 

————————————————————————————————————————————————————————————————————————————————————      https://www.zhihu.com/question/399007267/answer/1260691185 

先说结论:NIO的DirectBuffer的大小,和元空间没有关系。不能因为DirectBuffer和metaspace都是在native内存中分配的就将其混为一谈。两个走的完全是不相关的逻辑。

以OpenJDK 8的ByteBuffer为例,本质上分配内存的逻辑为:

DirectByteBuffer(int cap) {                   // package-private
    super(-1, 0, cap, cap);
    boolean pa = VM.isDirectMemoryPageAligned();
    int ps = Bits.pageSize();
    long size = Math.max(1L, (long) cap + (pa ? ps : 0));
    Bits.reserveMemory(size, cap);
    
    long base = 0;
    try {
        base = unsafe.allocateMemory(size);
    } catch (OutOfMemoryError x) {
        Bits.unreserveMemory(size, cap);
        throw x;
    }
    unsafe.setMemory(base, size, (byte) 0);
    if (pa && (base % ps != 0)) {
        // Round up to page boundary
        address = base + ps - (base & (ps - 1));
    } else {
        address = base;
    }
    cleaner = Cleaner.create(this, new DirectByteBuffer.Deallocator(base, size, cap));
    att = null;
}

核心还是用Unsafe的allocateMemory来分配的内存,而Unsafe的分配逻辑是:

UNSAFE_ENTRY(jlong, Unsafe_AllocateMemory(JNIEnv *env, jobject unsafe, jlong size))
  UnsafeWrapper("Unsafe_AllocateMemory");
  size_t sz = (size_t)size;
  if (sz != (julong)size || size < 0) {
    THROW_0(vmSymbols::java_lang_IllegalArgumentException());
  }
  if (sz == 0) {
    return 0;
  }
  sz = round_to(sz, HeapWordSize);
  void* x = os::malloc(sz, mtInternal);
  if (x == NULL) {
    THROW_0(vmSymbols::java_lang_OutOfMemoryError());
  }
  //Copy::fill_to_words((HeapWord*)x, sz / HeapWordSize);
  return addr_to_java(x);
UNSAFE_END

最核心的os::malloc的逻辑是:

void* os::malloc(size_t size, MEMFLAGS memflags, const NativeCallStack& stack) {
  NOT_PRODUCT(inc_stat_counter(&num_mallocs, 1));
  NOT_PRODUCT(inc_stat_counter(&alloc_bytes, size));

#ifdef ASSERT
  // checking for the WatcherThread and crash_protection first
  // since os::malloc can be called when the libjvm.{dll,so} is
  // first loaded and we don't have a thread yet.
  // try to find the thread after we see that the watcher thread
  // exists and has crash protection.
  WatcherThread *wt = WatcherThread::watcher_thread();
  if (wt != NULL && wt->has_crash_protection()) {
    Thread* thread = ThreadLocalStorage::get_thread_slow();
    if (thread == wt) {
      assert(!wt->has_crash_protection(),
          "Can't malloc with crash protection from WatcherThread");
    }
  }
#endif

  if (size == 0) {
    // return a valid pointer if size is zero
    // if NULL is returned the calling functions assume out of memory.
    size = 1;
  }

  // NMT support
  NMT_TrackingLevel level = MemTracker::tracking_level();
  size_t            nmt_header_size = MemTracker::malloc_header_size(level);

#ifndef ASSERT
  const size_t alloc_size = size + nmt_header_size;
#else
  const size_t alloc_size = GuardedMemory::get_total_size(size + nmt_header_size);
  if (size + nmt_header_size > alloc_size) { // Check for rollover.
    return NULL;
  }
#endif

  NOT_PRODUCT(if (MallocVerifyInterval > 0) check_heap());

  u_char* ptr;
  if (MallocMaxTestWords > 0) {
    ptr = testMalloc(alloc_size);
  } else {
    ptr = (u_char*)::malloc(alloc_size);
  }

#ifdef ASSERT
  if (ptr == NULL) {
    return NULL;
  }
  // Wrap memory with guard
  GuardedMemory guarded(ptr, size + nmt_header_size);
  ptr = guarded.get_user_ptr();
#endif
  if ((intptr_t)ptr == (intptr_t)MallocCatchPtr) {
    tty->print_cr("os::malloc caught, " SIZE_FORMAT " bytes --> " PTR_FORMAT, size, ptr);
    breakpoint();
  }
  debug_only(if (paranoid) verify_memory(ptr));
  if (PrintMalloc && tty != NULL) {
    tty->print_cr("os::malloc " SIZE_FORMAT " bytes --> " PTR_FORMAT, size, ptr);
  }

  // we do not track guard memory
  return MemTracker::record_malloc((address)ptr, size, memflags, stack, level);
}

所以本质上还是malloc的库函数分配的内存,而JVM参数-XX:MaxDirectMemorySize=会在启动JVM的时候以sun.nio.MaxDirectMemorySize加载进参数中。最终在ByteBuffer分配的时候经过:

private static boolean tryReserveMemory(long size, int cap) {

    // -XX:MaxDirectMemorySize limits the total capacity rather than the
    // actual memory usage, which will differ when buffers are page
    // aligned.
    long totalCap;
    while (cap <= maxMemory - (totalCap = totalCapacity.get())) {
        if (totalCapacity.compareAndSet(totalCap, totalCap + cap)) {
            reservedMemory.addAndGet(size);
            count.incrementAndGet();
            return true;
        }
    }
    return false;
}

当目前使用内存过多,触及参数上限时,这套逻辑分配的内存会在这里分配失败从而抛出异常。

因此直接内存的大小和元空间一点儿关系也没有,元空间的内存存储完全是JVM内部另一套内存分配逻辑,同时它们存储的内容也不相同,元空间应由参数:-XX:MaxMetaspaceSize=来设定,这个参数影响的逻辑是:

void MetaspaceGC::initialize() {
  // Set the high-water mark to MaxMetapaceSize during VM initializaton since
  // we can't do a GC during initialization.
  _capacity_until_GC = MaxMetaspaceSize;
}

从而影响元空间内存分配时的大小校验,两个逻辑没有关系。

标签:XX,内存,ByteBuffer,MaxDirectMemorySize,ptr,size
From: https://www.cnblogs.com/kelelipeng/p/16836058.html

相关文章