1<html><head><meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"><title>JEMALLOC</title><meta name="generator" content="DocBook XSL Stylesheets V1.78.1"></head><body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"><div class="refentry"><a name="idm316394519664"></a><div class="titlepage"></div><div class="refnamediv"><h2>Name</h2><p>jemalloc — general purpose memory allocation functions</p></div><div class="refsect1"><a name="library"></a><h2>LIBRARY</h2><p>This manual describes jemalloc 3.6.0-0-g46c0af68bd248b04df75e4f92d5fb804c3d75340. More information 2 can be found at the <a class="ulink" href="http://www.canonware.com/jemalloc/" target="_top">jemalloc website</a>.</p></div><div class="refsynopsisdiv"><h2>SYNOPSIS</h2><div class="funcsynopsis"><pre class="funcsynopsisinfo">#include <<code class="filename">stdlib.h</code>> 3#include <<code class="filename">jemalloc/jemalloc.h</code>></pre><div class="refsect2"><a name="idm316394002288"></a><h3>Standard API</h3><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">malloc</b>(</code></td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">calloc</b>(</code></td><td>size_t <var class="pdparam">number</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">posix_memalign</b>(</code></td><td>void **<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">alignment</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">aligned_alloc</b>(</code></td><td>size_t <var class="pdparam">alignment</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">realloc</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">free</b>(</code></td><td>void *<var class="pdparam">ptr</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div></div><div class="refsect2"><a name="idm316393986160"></a><h3>Non-standard API</h3><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">mallocx</b>(</code></td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void *<b class="fsfunc">rallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">xallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">extra</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">sallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">dallocx</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">nallocx</b>(</code></td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">mallctl</b>(</code></td><td>const char *<var class="pdparam">name</var>, </td></tr><tr><td>�</td><td>void *<var class="pdparam">oldp</var>, </td></tr><tr><td>�</td><td>size_t *<var class="pdparam">oldlenp</var>, </td></tr><tr><td>�</td><td>void *<var class="pdparam">newp</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">newlen</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">mallctlnametomib</b>(</code></td><td>const char *<var class="pdparam">name</var>, </td></tr><tr><td>�</td><td>size_t *<var class="pdparam">mibp</var>, </td></tr><tr><td>�</td><td>size_t *<var class="pdparam">miblenp</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">mallctlbymib</b>(</code></td><td>const size_t *<var class="pdparam">mib</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">miblen</var>, </td></tr><tr><td>�</td><td>void *<var class="pdparam">oldp</var>, </td></tr><tr><td>�</td><td>size_t *<var class="pdparam">oldlenp</var>, </td></tr><tr><td>�</td><td>void *<var class="pdparam">newp</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">newlen</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">malloc_stats_print</b>(</code></td><td>void <var class="pdparam">(*write_cb)</var> 4 <code>(</code>void *, const char *<code>)</code> 5 , </td></tr><tr><td>�</td><td>void *<var class="pdparam">cbopaque</var>, </td></tr><tr><td>�</td><td>const char *<var class="pdparam">opts</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">size_t <b class="fsfunc">malloc_usable_size</b>(</code></td><td>const void *<var class="pdparam">ptr</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">void <b class="fsfunc">(*malloc_message)</b>(</code></td><td>void *<var class="pdparam">cbopaque</var>, </td></tr><tr><td>�</td><td>const char *<var class="pdparam">s</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><p><span class="type">const char *</span><code class="varname">malloc_conf</code>;</p></div><div class="refsect2"><a name="idm316388684112"></a><h3>Experimental API</h3><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">allocm</b>(</code></td><td>void **<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>size_t *<var class="pdparam">rsize</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">rallocm</b>(</code></td><td>void **<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>size_t *<var class="pdparam">rsize</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">extra</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">sallocm</b>(</code></td><td>const void *<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>size_t *<var class="pdparam">rsize</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">dallocm</b>(</code></td><td>void *<var class="pdparam">ptr</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div><table border="0" class="funcprototype-table" summary="Function synopsis" style="cellspacing: 0; cellpadding: 0;"><tr><td><code class="funcdef">int <b class="fsfunc">nallocm</b>(</code></td><td>size_t *<var class="pdparam">rsize</var>, </td></tr><tr><td>�</td><td>size_t <var class="pdparam">size</var>, </td></tr><tr><td>�</td><td>int <var class="pdparam">flags</var><code>)</code>;</td></tr></table><div class="funcprototype-spacer">�</div></div></div></div><div class="refsect1"><a name="description"></a><h2>DESCRIPTION</h2><div class="refsect2"><a name="idm316388663504"></a><h3>Standard API</h3><p>The <code class="function">malloc</code>(<em class="parameter"><code></code></em>) function allocates 6 <em class="parameter"><code>size</code></em> bytes of uninitialized memory. The allocated 7 space is suitably aligned (after possible pointer coercion) for storage 8 of any type of object.</p><p>The <code class="function">calloc</code>(<em class="parameter"><code></code></em>) function allocates 9 space for <em class="parameter"><code>number</code></em> objects, each 10 <em class="parameter"><code>size</code></em> bytes in length. The result is identical to 11 calling <code class="function">malloc</code>(<em class="parameter"><code></code></em>) with an argument of 12 <em class="parameter"><code>number</code></em> * <em class="parameter"><code>size</code></em>, with the 13 exception that the allocated memory is explicitly initialized to zero 14 bytes.</p><p>The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function 15 allocates <em class="parameter"><code>size</code></em> bytes of memory such that the 16 allocation's base address is an even multiple of 17 <em class="parameter"><code>alignment</code></em>, and returns the allocation in the value 18 pointed to by <em class="parameter"><code>ptr</code></em>. The requested 19 <em class="parameter"><code>alignment</code></em> must be a power of 2 at least as large 20 as <code class="code">sizeof(<span class="type">void *</span>)</code>.</p><p>The <code class="function">aligned_alloc</code>(<em class="parameter"><code></code></em>) function 21 allocates <em class="parameter"><code>size</code></em> bytes of memory such that the 22 allocation's base address is an even multiple of 23 <em class="parameter"><code>alignment</code></em>. The requested 24 <em class="parameter"><code>alignment</code></em> must be a power of 2. Behavior is 25 undefined if <em class="parameter"><code>size</code></em> is not an integral multiple of 26 <em class="parameter"><code>alignment</code></em>.</p><p>The <code class="function">realloc</code>(<em class="parameter"><code></code></em>) function changes the 27 size of the previously allocated memory referenced by 28 <em class="parameter"><code>ptr</code></em> to <em class="parameter"><code>size</code></em> bytes. The 29 contents of the memory are unchanged up to the lesser of the new and old 30 sizes. If the new size is larger, the contents of the newly allocated 31 portion of the memory are undefined. Upon success, the memory referenced 32 by <em class="parameter"><code>ptr</code></em> is freed and a pointer to the newly 33 allocated memory is returned. Note that 34 <code class="function">realloc</code>(<em class="parameter"><code></code></em>) may move the memory allocation, 35 resulting in a different return value than <em class="parameter"><code>ptr</code></em>. 36 If <em class="parameter"><code>ptr</code></em> is <code class="constant">NULL</code>, the 37 <code class="function">realloc</code>(<em class="parameter"><code></code></em>) function behaves identically to 38 <code class="function">malloc</code>(<em class="parameter"><code></code></em>) for the specified size.</p><p>The <code class="function">free</code>(<em class="parameter"><code></code></em>) function causes the 39 allocated memory referenced by <em class="parameter"><code>ptr</code></em> to be made 40 available for future allocations. If <em class="parameter"><code>ptr</code></em> is 41 <code class="constant">NULL</code>, no action occurs.</p></div><div class="refsect2"><a name="idm316388639904"></a><h3>Non-standard API</h3><p>The <code class="function">mallocx</code>(<em class="parameter"><code></code></em>), 42 <code class="function">rallocx</code>(<em class="parameter"><code></code></em>), 43 <code class="function">xallocx</code>(<em class="parameter"><code></code></em>), 44 <code class="function">sallocx</code>(<em class="parameter"><code></code></em>), 45 <code class="function">dallocx</code>(<em class="parameter"><code></code></em>), and 46 <code class="function">nallocx</code>(<em class="parameter"><code></code></em>) functions all have a 47 <em class="parameter"><code>flags</code></em> argument that can be used to specify 48 options. The functions only check the options that are contextually 49 relevant. Use bitwise or (<code class="code">|</code>) operations to 50 specify one or more of the following: 51 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="constant">MALLOCX_LG_ALIGN(<em class="parameter"><code>la</code></em>) 52 </code></span></dt><dd><p>Align the memory allocation to start at an address 53 that is a multiple of <code class="code">(1 << 54 <em class="parameter"><code>la</code></em>)</code>. This macro does not validate 55 that <em class="parameter"><code>la</code></em> is within the valid 56 range.</p></dd><dt><span class="term"><code class="constant">MALLOCX_ALIGN(<em class="parameter"><code>a</code></em>) 57 </code></span></dt><dd><p>Align the memory allocation to start at an address 58 that is a multiple of <em class="parameter"><code>a</code></em>, where 59 <em class="parameter"><code>a</code></em> is a power of two. This macro does not 60 validate that <em class="parameter"><code>a</code></em> is a power of 2. 61 </p></dd><dt><span class="term"><code class="constant">MALLOCX_ZERO</code></span></dt><dd><p>Initialize newly allocated memory to contain zero 62 bytes. In the growing reallocation case, the real size prior to 63 reallocation defines the boundary between untouched bytes and those 64 that are initialized to contain zero bytes. If this macro is 65 absent, newly allocated memory is uninitialized.</p></dd><dt><span class="term"><code class="constant">MALLOCX_ARENA(<em class="parameter"><code>a</code></em>) 66 </code></span></dt><dd><p>Use the arena specified by the index 67 <em class="parameter"><code>a</code></em> (and by necessity bypass the thread 68 cache). This macro has no effect for huge regions, nor for regions 69 that were allocated via an arena other than the one specified. 70 This macro does not validate that <em class="parameter"><code>a</code></em> 71 specifies an arena index in the valid range.</p></dd></dl></div><p> 72 </p><p>The <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function allocates at 73 least <em class="parameter"><code>size</code></em> bytes of memory, and returns a pointer 74 to the base address of the allocation. Behavior is undefined if 75 <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, or if request size 76 overflows due to size class and/or alignment constraints.</p><p>The <code class="function">rallocx</code>(<em class="parameter"><code></code></em>) function resizes the 77 allocation at <em class="parameter"><code>ptr</code></em> to be at least 78 <em class="parameter"><code>size</code></em> bytes, and returns a pointer to the base 79 address of the resulting allocation, which may or may not have moved from 80 its original location. Behavior is undefined if 81 <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, or if request size 82 overflows due to size class and/or alignment constraints.</p><p>The <code class="function">xallocx</code>(<em class="parameter"><code></code></em>) function resizes the 83 allocation at <em class="parameter"><code>ptr</code></em> in place to be at least 84 <em class="parameter"><code>size</code></em> bytes, and returns the real size of the 85 allocation. If <em class="parameter"><code>extra</code></em> is non-zero, an attempt is 86 made to resize the allocation to be at least <code class="code">(<em class="parameter"><code>size</code></em> + 87 <em class="parameter"><code>extra</code></em>)</code> bytes, though inability to allocate 88 the extra byte(s) will not by itself result in failure to resize. 89 Behavior is undefined if <em class="parameter"><code>size</code></em> is 90 <code class="constant">0</code>, or if <code class="code">(<em class="parameter"><code>size</code></em> + <em class="parameter"><code>extra</code></em> 91 > <code class="constant">SIZE_T_MAX</code>)</code>.</p><p>The <code class="function">sallocx</code>(<em class="parameter"><code></code></em>) function returns the 92 real size of the allocation at <em class="parameter"><code>ptr</code></em>.</p><p>The <code class="function">dallocx</code>(<em class="parameter"><code></code></em>) function causes the 93 memory referenced by <em class="parameter"><code>ptr</code></em> to be made available for 94 future allocations.</p><p>The <code class="function">nallocx</code>(<em class="parameter"><code></code></em>) function allocates no 95 memory, but it performs the same size computation as the 96 <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function, and returns the real 97 size of the allocation that would result from the equivalent 98 <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function call. Behavior is 99 undefined if <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, or if 100 request size overflows due to size class and/or alignment 101 constraints.</p><p>The <code class="function">mallctl</code>(<em class="parameter"><code></code></em>) function provides a 102 general interface for introspecting the memory allocator, as well as 103 setting modifiable parameters and triggering actions. The 104 period-separated <em class="parameter"><code>name</code></em> argument specifies a 105 location in a tree-structured namespace; see the <a class="xref" href="#mallctl_namespace" title="MALLCTL NAMESPACE">MALLCTL NAMESPACE</a> section for 106 documentation on the tree contents. To read a value, pass a pointer via 107 <em class="parameter"><code>oldp</code></em> to adequate space to contain the value, and a 108 pointer to its length via <em class="parameter"><code>oldlenp</code></em>; otherwise pass 109 <code class="constant">NULL</code> and <code class="constant">NULL</code>. Similarly, to 110 write a value, pass a pointer to the value via 111 <em class="parameter"><code>newp</code></em>, and its length via 112 <em class="parameter"><code>newlen</code></em>; otherwise pass <code class="constant">NULL</code> 113 and <code class="constant">0</code>.</p><p>The <code class="function">mallctlnametomib</code>(<em class="parameter"><code></code></em>) function 114 provides a way to avoid repeated name lookups for applications that 115 repeatedly query the same portion of the namespace, by translating a name 116 to a “Management Information Base” (MIB) that can be passed 117 repeatedly to <code class="function">mallctlbymib</code>(<em class="parameter"><code></code></em>). Upon 118 successful return from <code class="function">mallctlnametomib</code>(<em class="parameter"><code></code></em>), 119 <em class="parameter"><code>mibp</code></em> contains an array of 120 <em class="parameter"><code>*miblenp</code></em> integers, where 121 <em class="parameter"><code>*miblenp</code></em> is the lesser of the number of components 122 in <em class="parameter"><code>name</code></em> and the input value of 123 <em class="parameter"><code>*miblenp</code></em>. Thus it is possible to pass a 124 <em class="parameter"><code>*miblenp</code></em> that is smaller than the number of 125 period-separated name components, which results in a partial MIB that can 126 be used as the basis for constructing a complete MIB. For name 127 components that are integers (e.g. the 2 in 128 <a class="link" href="#arenas.bin.i.size"> 129 "<code class="mallctl">arenas.bin.2.size</code>" 130 </a>), 131 the corresponding MIB component will always be that integer. Therefore, 132 it is legitimate to construct code like the following: </p><pre class="programlisting"> 133unsigned nbins, i; 134size_t mib[4]; 135size_t len, miblen; 136 137len = sizeof(nbins); 138mallctl("arenas.nbins", &nbins, &len, NULL, 0); 139 140miblen = 4; 141mallctlnametomib("arenas.bin.0.size", mib, &miblen); 142for (i = 0; i < nbins; i++) { 143 size_t bin_size; 144 145 mib[2] = i; 146 len = sizeof(bin_size); 147 mallctlbymib(mib, miblen, &bin_size, &len, NULL, 0); 148 /* Do something with bin_size... */ 149}</pre><p>The <code class="function">malloc_stats_print</code>(<em class="parameter"><code></code></em>) function 150 writes human-readable summary statistics via the 151 <em class="parameter"><code>write_cb</code></em> callback function pointer and 152 <em class="parameter"><code>cbopaque</code></em> data passed to 153 <em class="parameter"><code>write_cb</code></em>, or 154 <code class="function">malloc_message</code>(<em class="parameter"><code></code></em>) if 155 <em class="parameter"><code>write_cb</code></em> is <code class="constant">NULL</code>. This 156 function can be called repeatedly. General information that never 157 changes during execution can be omitted by specifying "g" as a character 158 within the <em class="parameter"><code>opts</code></em> string. Note that 159 <code class="function">malloc_message</code>(<em class="parameter"><code></code></em>) uses the 160 <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) functions internally, so 161 inconsistent statistics can be reported if multiple threads use these 162 functions simultaneously. If <code class="option">--enable-stats</code> is 163 specified during configuration, “m” and “a” can 164 be specified to omit merged arena and per arena statistics, respectively; 165 “b” and “l” can be specified to omit per size 166 class statistics for bins and large objects, respectively. Unrecognized 167 characters are silently ignored. Note that thread caching may prevent 168 some statistics from being completely up to date, since extra locking 169 would be required to merge counters that track thread cache operations. 170 </p><p>The <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) function 171 returns the usable size of the allocation pointed to by 172 <em class="parameter"><code>ptr</code></em>. The return value may be larger than the size 173 that was requested during allocation. The 174 <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) function is not a 175 mechanism for in-place <code class="function">realloc</code>(<em class="parameter"><code></code></em>); rather 176 it is provided solely as a tool for introspection purposes. Any 177 discrepancy between the requested allocation size and the size reported 178 by <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) should not be 179 depended on, since such behavior is entirely implementation-dependent. 180 </p></div><div class="refsect2"><a name="idm316388574208"></a><h3>Experimental API</h3><p>The experimental API is subject to change or removal without regard 181 for backward compatibility. If <code class="option">--disable-experimental</code> 182 is specified during configuration, the experimental API is 183 omitted.</p><p>The <code class="function">allocm</code>(<em class="parameter"><code></code></em>), 184 <code class="function">rallocm</code>(<em class="parameter"><code></code></em>), 185 <code class="function">sallocm</code>(<em class="parameter"><code></code></em>), 186 <code class="function">dallocm</code>(<em class="parameter"><code></code></em>), and 187 <code class="function">nallocm</code>(<em class="parameter"><code></code></em>) functions all have a 188 <em class="parameter"><code>flags</code></em> argument that can be used to specify 189 options. The functions only check the options that are contextually 190 relevant. Use bitwise or (<code class="code">|</code>) operations to 191 specify one or more of the following: 192 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="constant">ALLOCM_LG_ALIGN(<em class="parameter"><code>la</code></em>) 193 </code></span></dt><dd><p>Align the memory allocation to start at an address 194 that is a multiple of <code class="code">(1 << 195 <em class="parameter"><code>la</code></em>)</code>. This macro does not validate 196 that <em class="parameter"><code>la</code></em> is within the valid 197 range.</p></dd><dt><span class="term"><code class="constant">ALLOCM_ALIGN(<em class="parameter"><code>a</code></em>) 198 </code></span></dt><dd><p>Align the memory allocation to start at an address 199 that is a multiple of <em class="parameter"><code>a</code></em>, where 200 <em class="parameter"><code>a</code></em> is a power of two. This macro does not 201 validate that <em class="parameter"><code>a</code></em> is a power of 2. 202 </p></dd><dt><span class="term"><code class="constant">ALLOCM_ZERO</code></span></dt><dd><p>Initialize newly allocated memory to contain zero 203 bytes. In the growing reallocation case, the real size prior to 204 reallocation defines the boundary between untouched bytes and those 205 that are initialized to contain zero bytes. If this macro is 206 absent, newly allocated memory is uninitialized.</p></dd><dt><span class="term"><code class="constant">ALLOCM_NO_MOVE</code></span></dt><dd><p>For reallocation, fail rather than moving the 207 object. This constraint can apply to both growth and 208 shrinkage.</p></dd><dt><span class="term"><code class="constant">ALLOCM_ARENA(<em class="parameter"><code>a</code></em>) 209 </code></span></dt><dd><p>Use the arena specified by the index 210 <em class="parameter"><code>a</code></em> (and by necessity bypass the thread 211 cache). This macro has no effect for huge regions, nor for regions 212 that were allocated via an arena other than the one specified. 213 This macro does not validate that <em class="parameter"><code>a</code></em> 214 specifies an arena index in the valid range.</p></dd></dl></div><p> 215 </p><p>The <code class="function">allocm</code>(<em class="parameter"><code></code></em>) function allocates at 216 least <em class="parameter"><code>size</code></em> bytes of memory, sets 217 <em class="parameter"><code>*ptr</code></em> to the base address of the allocation, and 218 sets <em class="parameter"><code>*rsize</code></em> to the real size of the allocation if 219 <em class="parameter"><code>rsize</code></em> is not <code class="constant">NULL</code>. Behavior 220 is undefined if <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, or 221 if request size overflows due to size class and/or alignment 222 constraints.</p><p>The <code class="function">rallocm</code>(<em class="parameter"><code></code></em>) function resizes the 223 allocation at <em class="parameter"><code>*ptr</code></em> to be at least 224 <em class="parameter"><code>size</code></em> bytes, sets <em class="parameter"><code>*ptr</code></em> to 225 the base address of the allocation if it moved, and sets 226 <em class="parameter"><code>*rsize</code></em> to the real size of the allocation if 227 <em class="parameter"><code>rsize</code></em> is not <code class="constant">NULL</code>. If 228 <em class="parameter"><code>extra</code></em> is non-zero, an attempt is made to resize 229 the allocation to be at least <code class="code">(<em class="parameter"><code>size</code></em> + 230 <em class="parameter"><code>extra</code></em>)</code> bytes, though inability to allocate 231 the extra byte(s) will not by itself result in failure. Behavior is 232 undefined if <em class="parameter"><code>size</code></em> is <code class="constant">0</code>, if 233 request size overflows due to size class and/or alignment constraints, or 234 if <code class="code">(<em class="parameter"><code>size</code></em> + 235 <em class="parameter"><code>extra</code></em> > 236 <code class="constant">SIZE_T_MAX</code>)</code>.</p><p>The <code class="function">sallocm</code>(<em class="parameter"><code></code></em>) function sets 237 <em class="parameter"><code>*rsize</code></em> to the real size of the allocation.</p><p>The <code class="function">dallocm</code>(<em class="parameter"><code></code></em>) function causes the 238 memory referenced by <em class="parameter"><code>ptr</code></em> to be made available for 239 future allocations.</p><p>The <code class="function">nallocm</code>(<em class="parameter"><code></code></em>) function allocates no 240 memory, but it performs the same size computation as the 241 <code class="function">allocm</code>(<em class="parameter"><code></code></em>) function, and if 242 <em class="parameter"><code>rsize</code></em> is not <code class="constant">NULL</code> it sets 243 <em class="parameter"><code>*rsize</code></em> to the real size of the allocation that 244 would result from the equivalent <code class="function">allocm</code>(<em class="parameter"><code></code></em>) 245 function call. Behavior is undefined if <em class="parameter"><code>size</code></em> is 246 <code class="constant">0</code>, or if request size overflows due to size class 247 and/or alignment constraints.</p></div></div><div class="refsect1"><a name="tuning"></a><h2>TUNING</h2><p>Once, when the first call is made to one of the memory allocation 248 routines, the allocator initializes its internals based in part on various 249 options that can be specified at compile- or run-time.</p><p>The string pointed to by the global variable 250 <code class="varname">malloc_conf</code>, the “name” of the file 251 referenced by the symbolic link named <code class="filename">/etc/malloc.conf</code>, and the value of the 252 environment variable <code class="envar">MALLOC_CONF</code>, will be interpreted, in 253 that order, from left to right as options. Note that 254 <code class="varname">malloc_conf</code> may be read before 255 <code class="function">main</code>(<em class="parameter"><code></code></em>) is entered, so the declaration of 256 <code class="varname">malloc_conf</code> should specify an initializer that contains 257 the final value to be read by jemalloc. <code class="varname">malloc_conf</code> is 258 a compile-time setting, whereas <code class="filename">/etc/malloc.conf</code> and <code class="envar">MALLOC_CONF</code> 259 can be safely set any time prior to program invocation.</p><p>An options string is a comma-separated list of option:value pairs. 260 There is one key corresponding to each <a class="link" href="#opt.abort"> 261 "<code class="mallctl">opt.*</code>" 262 </a> mallctl (see the <a class="xref" href="#mallctl_namespace" title="MALLCTL NAMESPACE">MALLCTL NAMESPACE</a> section for options 263 documentation). For example, <code class="literal">abort:true,narenas:1</code> sets 264 the <a class="link" href="#opt.abort"> 265 "<code class="mallctl">opt.abort</code>" 266 </a> and <a class="link" href="#opt.narenas"> 267 "<code class="mallctl">opt.narenas</code>" 268 </a> options. Some 269 options have boolean values (true/false), others have integer values (base 270 8, 10, or 16, depending on prefix), and yet others have raw string 271 values.</p></div><div class="refsect1"><a name="implementation_notes"></a><h2>IMPLEMENTATION NOTES</h2><p>Traditionally, allocators have used 272 <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span> to obtain memory, which is 273 suboptimal for several reasons, including race conditions, increased 274 fragmentation, and artificial limitations on maximum usable memory. If 275 <code class="option">--enable-dss</code> is specified during configuration, this 276 allocator uses both <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> and 277 <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>, in that order of preference; 278 otherwise only <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> is used.</p><p>This allocator uses multiple arenas in order to reduce lock 279 contention for threaded programs on multi-processor systems. This works 280 well with regard to threading scalability, but incurs some costs. There is 281 a small fixed per-arena overhead, and additionally, arenas manage memory 282 completely independently of each other, which means a small fixed increase 283 in overall memory fragmentation. These overheads are not generally an 284 issue, given the number of arenas normally used. Note that using 285 substantially more arenas than the default is not likely to improve 286 performance, mainly due to reduced cache performance. However, it may make 287 sense to reduce the number of arenas if an application does not make much 288 use of the allocation functions.</p><p>In addition to multiple arenas, unless 289 <code class="option">--disable-tcache</code> is specified during configuration, this 290 allocator supports thread-specific caching for small and large objects, in 291 order to make it possible to completely avoid synchronization for most 292 allocation requests. Such caching allows very fast allocation in the 293 common case, but it increases memory usage and fragmentation, since a 294 bounded number of objects can remain allocated in each thread cache.</p><p>Memory is conceptually broken into equal-sized chunks, where the 295 chunk size is a power of two that is greater than the page size. Chunks 296 are always aligned to multiples of the chunk size. This alignment makes it 297 possible to find metadata for user objects very quickly.</p><p>User objects are broken into three categories according to size: 298 small, large, and huge. Small objects are smaller than one page. Large 299 objects are smaller than the chunk size. Huge objects are a multiple of 300 the chunk size. Small and large objects are managed by arenas; huge 301 objects are managed separately in a single data structure that is shared by 302 all threads. Huge objects are used by applications infrequently enough 303 that this single data structure is not a scalability issue.</p><p>Each chunk that is managed by an arena tracks its contents as runs of 304 contiguous pages (unused, backing a set of small objects, or backing one 305 large object). The combination of chunk alignment and chunk page maps 306 makes it possible to determine all metadata regarding small and large 307 allocations in constant time.</p><p>Small objects are managed in groups by page runs. Each run maintains 308 a frontier and free list to track which regions are in use. Allocation 309 requests that are no more than half the quantum (8 or 16, depending on 310 architecture) are rounded up to the nearest power of two that is at least 311 <code class="code">sizeof(<span class="type">double</span>)</code>. All other small 312 object size classes are multiples of the quantum, spaced such that internal 313 fragmentation is limited to approximately 25% for all but the smallest size 314 classes. Allocation requests that are larger than the maximum small size 315 class, but small enough to fit in an arena-managed chunk (see the <a class="link" href="#opt.lg_chunk"> 316 "<code class="mallctl">opt.lg_chunk</code>" 317 </a> option), are 318 rounded up to the nearest run size. Allocation requests that are too large 319 to fit in an arena-managed chunk are rounded up to the nearest multiple of 320 the chunk size.</p><p>Allocations are packed tightly together, which can be an issue for 321 multi-threaded applications. If you need to assure that allocations do not 322 suffer from cacheline sharing, round your allocation requests up to the 323 nearest multiple of the cacheline size, or specify cacheline alignment when 324 allocating.</p><p>Assuming 4 MiB chunks, 4 KiB pages, and a 16-byte quantum on a 64-bit 325 system, the size classes in each category are as shown in <a class="xref" href="#size_classes" title="Table�1.�Size classes">Table 1</a>.</p><div class="table"><a name="size_classes"></a><p class="title"><b>Table�1.�Size classes</b></p><div class="table-contents"><table summary="Size classes" border="1"><colgroup><col align="left" class="c1"><col align="right" class="c2"><col align="left" class="c3"></colgroup><thead><tr><th align="left">Category</th><th align="right">Spacing</th><th align="left">Size</th></tr></thead><tbody><tr><td rowspan="7" align="left">Small</td><td align="right">lg</td><td align="left">[8]</td></tr><tr><td align="right">16</td><td align="left">[16, 32, 48, ..., 128]</td></tr><tr><td align="right">32</td><td align="left">[160, 192, 224, 256]</td></tr><tr><td align="right">64</td><td align="left">[320, 384, 448, 512]</td></tr><tr><td align="right">128</td><td align="left">[640, 768, 896, 1024]</td></tr><tr><td align="right">256</td><td align="left">[1280, 1536, 1792, 2048]</td></tr><tr><td align="right">512</td><td align="left">[2560, 3072, 3584]</td></tr><tr><td align="left">Large</td><td align="right">4 KiB</td><td align="left">[4 KiB, 8 KiB, 12 KiB, ..., 4072 KiB]</td></tr><tr><td align="left">Huge</td><td align="right">4 MiB</td><td align="left">[4 MiB, 8 MiB, 12 MiB, ...]</td></tr></tbody></table></div></div><br class="table-break"></div><div class="refsect1"><a name="mallctl_namespace"></a><h2>MALLCTL NAMESPACE</h2><p>The following names are defined in the namespace accessible via the 326 <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) functions. Value types are 327 specified in parentheses, their readable/writable statuses are encoded as 328 <code class="literal">rw</code>, <code class="literal">r-</code>, <code class="literal">-w</code>, or 329 <code class="literal">--</code>, and required build configuration flags follow, if 330 any. A name element encoded as <code class="literal"><i></code> or 331 <code class="literal"><j></code> indicates an integer component, where the 332 integer varies from 0 to some upper value that must be determined via 333 introspection. In the case of 334 "<code class="mallctl">stats.arenas.<i>.*</code>" 335 , 336 <code class="literal"><i></code> equal to <a class="link" href="#arenas.narenas"> 337 "<code class="mallctl">arenas.narenas</code>" 338 </a> can be 339 used to access the summation of statistics from all arenas. Take special 340 note of the <a class="link" href="#epoch"> 341 "<code class="mallctl">epoch</code>" 342 </a> mallctl, 343 which controls refreshing of cached dynamic statistics.</p><div class="variablelist"><dl class="variablelist"><dt><a name="version"></a><span class="term"> 344 345 "<code class="mallctl">version</code>" 346 347 (<span class="type">const char *</span>) 348 <code class="literal">r-</code> 349 </span></dt><dd><p>Return the jemalloc version string.</p></dd><dt><a name="epoch"></a><span class="term"> 350 351 "<code class="mallctl">epoch</code>" 352 353 (<span class="type">uint64_t</span>) 354 <code class="literal">rw</code> 355 </span></dt><dd><p>If a value is passed in, refresh the data from which 356 the <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) functions report values, 357 and increment the epoch. Return the current epoch. This is useful for 358 detecting whether another thread caused a refresh.</p></dd><dt><a name="config.debug"></a><span class="term"> 359 360 "<code class="mallctl">config.debug</code>" 361 362 (<span class="type">bool</span>) 363 <code class="literal">r-</code> 364 </span></dt><dd><p><code class="option">--enable-debug</code> was specified during 365 build configuration.</p></dd><dt><a name="config.dss"></a><span class="term"> 366 367 "<code class="mallctl">config.dss</code>" 368 369 (<span class="type">bool</span>) 370 <code class="literal">r-</code> 371 </span></dt><dd><p><code class="option">--enable-dss</code> was specified during 372 build configuration.</p></dd><dt><a name="config.fill"></a><span class="term"> 373 374 "<code class="mallctl">config.fill</code>" 375 376 (<span class="type">bool</span>) 377 <code class="literal">r-</code> 378 </span></dt><dd><p><code class="option">--enable-fill</code> was specified during 379 build configuration.</p></dd><dt><a name="config.lazy_lock"></a><span class="term"> 380 381 "<code class="mallctl">config.lazy_lock</code>" 382 383 (<span class="type">bool</span>) 384 <code class="literal">r-</code> 385 </span></dt><dd><p><code class="option">--enable-lazy-lock</code> was specified 386 during build configuration.</p></dd><dt><a name="config.mremap"></a><span class="term"> 387 388 "<code class="mallctl">config.mremap</code>" 389 390 (<span class="type">bool</span>) 391 <code class="literal">r-</code> 392 </span></dt><dd><p><code class="option">--enable-mremap</code> was specified during 393 build configuration.</p></dd><dt><a name="config.munmap"></a><span class="term"> 394 395 "<code class="mallctl">config.munmap</code>" 396 397 (<span class="type">bool</span>) 398 <code class="literal">r-</code> 399 </span></dt><dd><p><code class="option">--enable-munmap</code> was specified during 400 build configuration.</p></dd><dt><a name="config.prof"></a><span class="term"> 401 402 "<code class="mallctl">config.prof</code>" 403 404 (<span class="type">bool</span>) 405 <code class="literal">r-</code> 406 </span></dt><dd><p><code class="option">--enable-prof</code> was specified during 407 build configuration.</p></dd><dt><a name="config.prof_libgcc"></a><span class="term"> 408 409 "<code class="mallctl">config.prof_libgcc</code>" 410 411 (<span class="type">bool</span>) 412 <code class="literal">r-</code> 413 </span></dt><dd><p><code class="option">--disable-prof-libgcc</code> was not 414 specified during build configuration.</p></dd><dt><a name="config.prof_libunwind"></a><span class="term"> 415 416 "<code class="mallctl">config.prof_libunwind</code>" 417 418 (<span class="type">bool</span>) 419 <code class="literal">r-</code> 420 </span></dt><dd><p><code class="option">--enable-prof-libunwind</code> was specified 421 during build configuration.</p></dd><dt><a name="config.stats"></a><span class="term"> 422 423 "<code class="mallctl">config.stats</code>" 424 425 (<span class="type">bool</span>) 426 <code class="literal">r-</code> 427 </span></dt><dd><p><code class="option">--enable-stats</code> was specified during 428 build configuration.</p></dd><dt><a name="config.tcache"></a><span class="term"> 429 430 "<code class="mallctl">config.tcache</code>" 431 432 (<span class="type">bool</span>) 433 <code class="literal">r-</code> 434 </span></dt><dd><p><code class="option">--disable-tcache</code> was not specified 435 during build configuration.</p></dd><dt><a name="config.tls"></a><span class="term"> 436 437 "<code class="mallctl">config.tls</code>" 438 439 (<span class="type">bool</span>) 440 <code class="literal">r-</code> 441 </span></dt><dd><p><code class="option">--disable-tls</code> was not specified during 442 build configuration.</p></dd><dt><a name="config.utrace"></a><span class="term"> 443 444 "<code class="mallctl">config.utrace</code>" 445 446 (<span class="type">bool</span>) 447 <code class="literal">r-</code> 448 </span></dt><dd><p><code class="option">--enable-utrace</code> was specified during 449 build configuration.</p></dd><dt><a name="config.valgrind"></a><span class="term"> 450 451 "<code class="mallctl">config.valgrind</code>" 452 453 (<span class="type">bool</span>) 454 <code class="literal">r-</code> 455 </span></dt><dd><p><code class="option">--enable-valgrind</code> was specified during 456 build configuration.</p></dd><dt><a name="config.xmalloc"></a><span class="term"> 457 458 "<code class="mallctl">config.xmalloc</code>" 459 460 (<span class="type">bool</span>) 461 <code class="literal">r-</code> 462 </span></dt><dd><p><code class="option">--enable-xmalloc</code> was specified during 463 build configuration.</p></dd><dt><a name="opt.abort"></a><span class="term"> 464 465 "<code class="mallctl">opt.abort</code>" 466 467 (<span class="type">bool</span>) 468 <code class="literal">r-</code> 469 </span></dt><dd><p>Abort-on-warning enabled/disabled. If true, most 470 warnings are fatal. The process will call 471 <span class="citerefentry"><span class="refentrytitle">abort</span>(3)</span> in these cases. This option is 472 disabled by default unless <code class="option">--enable-debug</code> is 473 specified during configuration, in which case it is enabled by default. 474 </p></dd><dt><a name="opt.dss"></a><span class="term"> 475 476 "<code class="mallctl">opt.dss</code>" 477 478 (<span class="type">const char *</span>) 479 <code class="literal">r-</code> 480 </span></dt><dd><p>dss (<span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>) allocation precedence as 481 related to <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> allocation. The following 482 settings are supported: “disabled”, “primary”, 483 and “secondary”. The default is “secondary” if 484 <a class="link" href="#config.dss"> 485 "<code class="mallctl">config.dss</code>" 486 </a> is 487 true, “disabled” otherwise. 488 </p></dd><dt><a name="opt.lg_chunk"></a><span class="term"> 489 490 "<code class="mallctl">opt.lg_chunk</code>" 491 492 (<span class="type">size_t</span>) 493 <code class="literal">r-</code> 494 </span></dt><dd><p>Virtual memory chunk size (log base 2). If a chunk 495 size outside the supported size range is specified, the size is 496 silently clipped to the minimum/maximum supported size. The default 497 chunk size is 4 MiB (2^22). 498 </p></dd><dt><a name="opt.narenas"></a><span class="term"> 499 500 "<code class="mallctl">opt.narenas</code>" 501 502 (<span class="type">size_t</span>) 503 <code class="literal">r-</code> 504 </span></dt><dd><p>Maximum number of arenas to use for automatic 505 multiplexing of threads and arenas. The default is four times the 506 number of CPUs, or one if there is a single CPU.</p></dd><dt><a name="opt.lg_dirty_mult"></a><span class="term"> 507 508 "<code class="mallctl">opt.lg_dirty_mult</code>" 509 510 (<span class="type">ssize_t</span>) 511 <code class="literal">r-</code> 512 </span></dt><dd><p>Per-arena minimum ratio (log base 2) of active to dirty 513 pages. Some dirty unused pages may be allowed to accumulate, within 514 the limit set by the ratio (or one chunk worth of dirty pages, 515 whichever is greater), before informing the kernel about some of those 516 pages via <span class="citerefentry"><span class="refentrytitle">madvise</span>(2)</span> or a similar system call. This 517 provides the kernel with sufficient information to recycle dirty pages 518 if physical memory becomes scarce and the pages remain unused. The 519 default minimum ratio is 8:1 (2^3:1); an option value of -1 will 520 disable dirty page purging.</p></dd><dt><a name="opt.stats_print"></a><span class="term"> 521 522 "<code class="mallctl">opt.stats_print</code>" 523 524 (<span class="type">bool</span>) 525 <code class="literal">r-</code> 526 </span></dt><dd><p>Enable/disable statistics printing at exit. If 527 enabled, the <code class="function">malloc_stats_print</code>(<em class="parameter"><code></code></em>) 528 function is called at program exit via an 529 <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span> function. If 530 <code class="option">--enable-stats</code> is specified during configuration, this 531 has the potential to cause deadlock for a multi-threaded process that 532 exits while one or more threads are executing in the memory allocation 533 functions. Therefore, this option should only be used with care; it is 534 primarily intended as a performance tuning aid during application 535 development. This option is disabled by default.</p></dd><dt><a name="opt.junk"></a><span class="term"> 536 537 "<code class="mallctl">opt.junk</code>" 538 539 (<span class="type">bool</span>) 540 <code class="literal">r-</code> 541 [<code class="option">--enable-fill</code>] 542 </span></dt><dd><p>Junk filling enabled/disabled. If enabled, each byte 543 of uninitialized allocated memory will be initialized to 544 <code class="literal">0xa5</code>. All deallocated memory will be initialized to 545 <code class="literal">0x5a</code>. This is intended for debugging and will 546 impact performance negatively. This option is disabled by default 547 unless <code class="option">--enable-debug</code> is specified during 548 configuration, in which case it is enabled by default unless running 549 inside <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>.</p></dd><dt><a name="opt.quarantine"></a><span class="term"> 550 551 "<code class="mallctl">opt.quarantine</code>" 552 553 (<span class="type">size_t</span>) 554 <code class="literal">r-</code> 555 [<code class="option">--enable-fill</code>] 556 </span></dt><dd><p>Per thread quarantine size in bytes. If non-zero, each 557 thread maintains a FIFO object quarantine that stores up to the 558 specified number of bytes of memory. The quarantined memory is not 559 freed until it is released from quarantine, though it is immediately 560 junk-filled if the <a class="link" href="#opt.junk"> 561 "<code class="mallctl">opt.junk</code>" 562 </a> option is 563 enabled. This feature is of particular use in combination with <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>, which can detect attempts 564 to access quarantined objects. This is intended for debugging and will 565 impact performance negatively. The default quarantine size is 0 unless 566 running inside Valgrind, in which case the default is 16 567 MiB.</p></dd><dt><a name="opt.redzone"></a><span class="term"> 568 569 "<code class="mallctl">opt.redzone</code>" 570 571 (<span class="type">bool</span>) 572 <code class="literal">r-</code> 573 [<code class="option">--enable-fill</code>] 574 </span></dt><dd><p>Redzones enabled/disabled. If enabled, small 575 allocations have redzones before and after them. Furthermore, if the 576 <a class="link" href="#opt.junk"> 577 "<code class="mallctl">opt.junk</code>" 578 </a> option is 579 enabled, the redzones are checked for corruption during deallocation. 580 However, the primary intended purpose of this feature is to be used in 581 combination with <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>, 582 which needs redzones in order to do effective buffer overflow/underflow 583 detection. This option is intended for debugging and will impact 584 performance negatively. This option is disabled by 585 default unless running inside Valgrind.</p></dd><dt><a name="opt.zero"></a><span class="term"> 586 587 "<code class="mallctl">opt.zero</code>" 588 589 (<span class="type">bool</span>) 590 <code class="literal">r-</code> 591 [<code class="option">--enable-fill</code>] 592 </span></dt><dd><p>Zero filling enabled/disabled. If enabled, each byte 593 of uninitialized allocated memory will be initialized to 0. Note that 594 this initialization only happens once for each byte, so 595 <code class="function">realloc</code>(<em class="parameter"><code></code></em>), 596 <code class="function">rallocx</code>(<em class="parameter"><code></code></em>) and 597 <code class="function">rallocm</code>(<em class="parameter"><code></code></em>) calls do not zero memory that 598 was previously allocated. This is intended for debugging and will 599 impact performance negatively. This option is disabled by default. 600 </p></dd><dt><a name="opt.utrace"></a><span class="term"> 601 602 "<code class="mallctl">opt.utrace</code>" 603 604 (<span class="type">bool</span>) 605 <code class="literal">r-</code> 606 [<code class="option">--enable-utrace</code>] 607 </span></dt><dd><p>Allocation tracing based on 608 <span class="citerefentry"><span class="refentrytitle">utrace</span>(2)</span> enabled/disabled. This option 609 is disabled by default.</p></dd><dt><a name="opt.valgrind"></a><span class="term"> 610 611 "<code class="mallctl">opt.valgrind</code>" 612 613 (<span class="type">bool</span>) 614 <code class="literal">r-</code> 615 [<code class="option">--enable-valgrind</code>] 616 </span></dt><dd><p><a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a> 617 support enabled/disabled. This option is vestigal because jemalloc 618 auto-detects whether it is running inside Valgrind. This option is 619 disabled by default, unless running inside Valgrind.</p></dd><dt><a name="opt.xmalloc"></a><span class="term"> 620 621 "<code class="mallctl">opt.xmalloc</code>" 622 623 (<span class="type">bool</span>) 624 <code class="literal">r-</code> 625 [<code class="option">--enable-xmalloc</code>] 626 </span></dt><dd><p>Abort-on-out-of-memory enabled/disabled. If enabled, 627 rather than returning failure for any allocation function, display a 628 diagnostic message on <code class="constant">STDERR_FILENO</code> and cause the 629 program to drop core (using 630 <span class="citerefentry"><span class="refentrytitle">abort</span>(3)</span>). If an application is 631 designed to depend on this behavior, set the option at compile time by 632 including the following in the source code: 633 </p><pre class="programlisting"> 634malloc_conf = "xmalloc:true";</pre><p> 635 This option is disabled by default.</p></dd><dt><a name="opt.tcache"></a><span class="term"> 636 637 "<code class="mallctl">opt.tcache</code>" 638 639 (<span class="type">bool</span>) 640 <code class="literal">r-</code> 641 [<code class="option">--enable-tcache</code>] 642 </span></dt><dd><p>Thread-specific caching enabled/disabled. When there 643 are multiple threads, each thread uses a thread-specific cache for 644 objects up to a certain size. Thread-specific caching allows many 645 allocations to be satisfied without performing any thread 646 synchronization, at the cost of increased memory use. See the 647 <a class="link" href="#opt.lg_tcache_max"> 648 "<code class="mallctl">opt.lg_tcache_max</code>" 649 </a> 650 option for related tuning information. This option is enabled by 651 default unless running inside <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a>.</p></dd><dt><a name="opt.lg_tcache_max"></a><span class="term"> 652 653 "<code class="mallctl">opt.lg_tcache_max</code>" 654 655 (<span class="type">size_t</span>) 656 <code class="literal">r-</code> 657 [<code class="option">--enable-tcache</code>] 658 </span></dt><dd><p>Maximum size class (log base 2) to cache in the 659 thread-specific cache. At a minimum, all small size classes are 660 cached, and at a maximum all large size classes are cached. The 661 default maximum is 32 KiB (2^15).</p></dd><dt><a name="opt.prof"></a><span class="term"> 662 663 "<code class="mallctl">opt.prof</code>" 664 665 (<span class="type">bool</span>) 666 <code class="literal">r-</code> 667 [<code class="option">--enable-prof</code>] 668 </span></dt><dd><p>Memory profiling enabled/disabled. If enabled, profile 669 memory allocation activity. See the <a class="link" href="#opt.prof_active"> 670 "<code class="mallctl">opt.prof_active</code>" 671 </a> 672 option for on-the-fly activation/deactivation. See the <a class="link" href="#opt.lg_prof_sample"> 673 "<code class="mallctl">opt.lg_prof_sample</code>" 674 </a> 675 option for probabilistic sampling control. See the <a class="link" href="#opt.prof_accum"> 676 "<code class="mallctl">opt.prof_accum</code>" 677 </a> 678 option for control of cumulative sample reporting. See the <a class="link" href="#opt.lg_prof_interval"> 679 "<code class="mallctl">opt.lg_prof_interval</code>" 680 </a> 681 option for information on interval-triggered profile dumping, the <a class="link" href="#opt.prof_gdump"> 682 "<code class="mallctl">opt.prof_gdump</code>" 683 </a> 684 option for information on high-water-triggered profile dumping, and the 685 <a class="link" href="#opt.prof_final"> 686 "<code class="mallctl">opt.prof_final</code>" 687 </a> 688 option for final profile dumping. Profile output is compatible with 689 the included <span class="command"><strong>pprof</strong></span> Perl script, which originates 690 from the <a class="ulink" href="http://code.google.com/p/gperftools/" target="_top">gperftools 691 package</a>.</p></dd><dt><a name="opt.prof_prefix"></a><span class="term"> 692 693 "<code class="mallctl">opt.prof_prefix</code>" 694 695 (<span class="type">const char *</span>) 696 <code class="literal">r-</code> 697 [<code class="option">--enable-prof</code>] 698 </span></dt><dd><p>Filename prefix for profile dumps. If the prefix is 699 set to the empty string, no automatic dumps will occur; this is 700 primarily useful for disabling the automatic final heap dump (which 701 also disables leak reporting, if enabled). The default prefix is 702 <code class="filename">jeprof</code>.</p></dd><dt><a name="opt.prof_active"></a><span class="term"> 703 704 "<code class="mallctl">opt.prof_active</code>" 705 706 (<span class="type">bool</span>) 707 <code class="literal">rw</code> 708 [<code class="option">--enable-prof</code>] 709 </span></dt><dd><p>Profiling activated/deactivated. This is a secondary 710 control mechanism that makes it possible to start the application with 711 profiling enabled (see the <a class="link" href="#opt.prof"> 712 "<code class="mallctl">opt.prof</code>" 713 </a> option) but 714 inactive, then toggle profiling at any time during program execution 715 with the <a class="link" href="#prof.active"> 716 "<code class="mallctl">prof.active</code>" 717 </a> mallctl. 718 This option is enabled by default.</p></dd><dt><a name="opt.lg_prof_sample"></a><span class="term"> 719 720 "<code class="mallctl">opt.lg_prof_sample</code>" 721 722 (<span class="type">ssize_t</span>) 723 <code class="literal">r-</code> 724 [<code class="option">--enable-prof</code>] 725 </span></dt><dd><p>Average interval (log base 2) between allocation 726 samples, as measured in bytes of allocation activity. Increasing the 727 sampling interval decreases profile fidelity, but also decreases the 728 computational overhead. The default sample interval is 512 KiB (2^19 729 B).</p></dd><dt><a name="opt.prof_accum"></a><span class="term"> 730 731 "<code class="mallctl">opt.prof_accum</code>" 732 733 (<span class="type">bool</span>) 734 <code class="literal">r-</code> 735 [<code class="option">--enable-prof</code>] 736 </span></dt><dd><p>Reporting of cumulative object/byte counts in profile 737 dumps enabled/disabled. If this option is enabled, every unique 738 backtrace must be stored for the duration of execution. Depending on 739 the application, this can impose a large memory overhead, and the 740 cumulative counts are not always of interest. This option is disabled 741 by default.</p></dd><dt><a name="opt.lg_prof_interval"></a><span class="term"> 742 743 "<code class="mallctl">opt.lg_prof_interval</code>" 744 745 (<span class="type">ssize_t</span>) 746 <code class="literal">r-</code> 747 [<code class="option">--enable-prof</code>] 748 </span></dt><dd><p>Average interval (log base 2) between memory profile 749 dumps, as measured in bytes of allocation activity. The actual 750 interval between dumps may be sporadic because decentralized allocation 751 counters are used to avoid synchronization bottlenecks. Profiles are 752 dumped to files named according to the pattern 753 <code class="filename"><prefix>.<pid>.<seq>.i<iseq>.heap</code>, 754 where <code class="literal"><prefix></code> is controlled by the 755 <a class="link" href="#opt.prof_prefix"> 756 "<code class="mallctl">opt.prof_prefix</code>" 757 </a> 758 option. By default, interval-triggered profile dumping is disabled 759 (encoded as -1). 760 </p></dd><dt><a name="opt.prof_gdump"></a><span class="term"> 761 762 "<code class="mallctl">opt.prof_gdump</code>" 763 764 (<span class="type">bool</span>) 765 <code class="literal">r-</code> 766 [<code class="option">--enable-prof</code>] 767 </span></dt><dd><p>Trigger a memory profile dump every time the total 768 virtual memory exceeds the previous maximum. Profiles are dumped to 769 files named according to the pattern 770 <code class="filename"><prefix>.<pid>.<seq>.u<useq>.heap</code>, 771 where <code class="literal"><prefix></code> is controlled by the <a class="link" href="#opt.prof_prefix"> 772 "<code class="mallctl">opt.prof_prefix</code>" 773 </a> 774 option. This option is disabled by default.</p></dd><dt><a name="opt.prof_final"></a><span class="term"> 775 776 "<code class="mallctl">opt.prof_final</code>" 777 778 (<span class="type">bool</span>) 779 <code class="literal">r-</code> 780 [<code class="option">--enable-prof</code>] 781 </span></dt><dd><p>Use an 782 <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span> function to dump final memory 783 usage to a file named according to the pattern 784 <code class="filename"><prefix>.<pid>.<seq>.f.heap</code>, 785 where <code class="literal"><prefix></code> is controlled by the <a class="link" href="#opt.prof_prefix"> 786 "<code class="mallctl">opt.prof_prefix</code>" 787 </a> 788 option. This option is enabled by default.</p></dd><dt><a name="opt.prof_leak"></a><span class="term"> 789 790 "<code class="mallctl">opt.prof_leak</code>" 791 792 (<span class="type">bool</span>) 793 <code class="literal">r-</code> 794 [<code class="option">--enable-prof</code>] 795 </span></dt><dd><p>Leak reporting enabled/disabled. If enabled, use an 796 <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span> function to report memory leaks 797 detected by allocation sampling. See the 798 <a class="link" href="#opt.prof"> 799 "<code class="mallctl">opt.prof</code>" 800 </a> option for 801 information on analyzing heap profile output. This option is disabled 802 by default.</p></dd><dt><a name="thread.arena"></a><span class="term"> 803 804 "<code class="mallctl">thread.arena</code>" 805 806 (<span class="type">unsigned</span>) 807 <code class="literal">rw</code> 808 </span></dt><dd><p>Get or set the arena associated with the calling 809 thread. If the specified arena was not initialized beforehand (see the 810 <a class="link" href="#arenas.initialized"> 811 "<code class="mallctl">arenas.initialized</code>" 812 </a> 813 mallctl), it will be automatically initialized as a side effect of 814 calling this interface.</p></dd><dt><a name="thread.allocated"></a><span class="term"> 815 816 "<code class="mallctl">thread.allocated</code>" 817 818 (<span class="type">uint64_t</span>) 819 <code class="literal">r-</code> 820 [<code class="option">--enable-stats</code>] 821 </span></dt><dd><p>Get the total number of bytes ever allocated by the 822 calling thread. This counter has the potential to wrap around; it is 823 up to the application to appropriately interpret the counter in such 824 cases.</p></dd><dt><a name="thread.allocatedp"></a><span class="term"> 825 826 "<code class="mallctl">thread.allocatedp</code>" 827 828 (<span class="type">uint64_t *</span>) 829 <code class="literal">r-</code> 830 [<code class="option">--enable-stats</code>] 831 </span></dt><dd><p>Get a pointer to the the value that is returned by the 832 <a class="link" href="#thread.allocated"> 833 "<code class="mallctl">thread.allocated</code>" 834 </a> 835 mallctl. This is useful for avoiding the overhead of repeated 836 <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) calls.</p></dd><dt><a name="thread.deallocated"></a><span class="term"> 837 838 "<code class="mallctl">thread.deallocated</code>" 839 840 (<span class="type">uint64_t</span>) 841 <code class="literal">r-</code> 842 [<code class="option">--enable-stats</code>] 843 </span></dt><dd><p>Get the total number of bytes ever deallocated by the 844 calling thread. This counter has the potential to wrap around; it is 845 up to the application to appropriately interpret the counter in such 846 cases.</p></dd><dt><a name="thread.deallocatedp"></a><span class="term"> 847 848 "<code class="mallctl">thread.deallocatedp</code>" 849 850 (<span class="type">uint64_t *</span>) 851 <code class="literal">r-</code> 852 [<code class="option">--enable-stats</code>] 853 </span></dt><dd><p>Get a pointer to the the value that is returned by the 854 <a class="link" href="#thread.deallocated"> 855 "<code class="mallctl">thread.deallocated</code>" 856 </a> 857 mallctl. This is useful for avoiding the overhead of repeated 858 <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) calls.</p></dd><dt><a name="thread.tcache.enabled"></a><span class="term"> 859 860 "<code class="mallctl">thread.tcache.enabled</code>" 861 862 (<span class="type">bool</span>) 863 <code class="literal">rw</code> 864 [<code class="option">--enable-tcache</code>] 865 </span></dt><dd><p>Enable/disable calling thread's tcache. The tcache is 866 implicitly flushed as a side effect of becoming 867 disabled (see 868 "<code class="mallctl">thread.tcache.flush</code>" 869 ). 870 </p></dd><dt><a name="thread.tcache.flush"></a><span class="term"> 871 872 "<code class="mallctl">thread.tcache.flush</code>" 873 874 (<span class="type">void</span>) 875 <code class="literal">--</code> 876 [<code class="option">--enable-tcache</code>] 877 </span></dt><dd><p>Flush calling thread's tcache. This interface releases 878 all cached objects and internal data structures associated with the 879 calling thread's thread-specific cache. Ordinarily, this interface 880 need not be called, since automatic periodic incremental garbage 881 collection occurs, and the thread cache is automatically discarded when 882 a thread exits. However, garbage collection is triggered by allocation 883 activity, so it is possible for a thread that stops 884 allocating/deallocating to retain its cache indefinitely, in which case 885 the developer may find manual flushing useful.</p></dd><dt><a name="arena.i.purge"></a><span class="term"> 886 887 "<code class="mallctl">arena.<i>.purge</code>" 888 889 (<span class="type">unsigned</span>) 890 <code class="literal">--</code> 891 </span></dt><dd><p>Purge unused dirty pages for arena <i>, or for 892 all arenas if <i> equals <a class="link" href="#arenas.narenas"> 893 "<code class="mallctl">arenas.narenas</code>" 894 </a>. 895 </p></dd><dt><a name="arena.i.dss"></a><span class="term"> 896 897 "<code class="mallctl">arena.<i>.dss</code>" 898 899 (<span class="type">const char *</span>) 900 <code class="literal">rw</code> 901 </span></dt><dd><p>Set the precedence of dss allocation as related to mmap 902 allocation for arena <i>, or for all arenas if <i> equals 903 <a class="link" href="#arenas.narenas"> 904 "<code class="mallctl">arenas.narenas</code>" 905 </a>. Note 906 that even during huge allocation this setting is read from the arena 907 that would be chosen for small or large allocation so that applications 908 can depend on consistent dss versus mmap allocation regardless of 909 allocation size. See <a class="link" href="#opt.dss"> 910 "<code class="mallctl">opt.dss</code>" 911 </a> for supported 912 settings. 913 </p></dd><dt><a name="arenas.narenas"></a><span class="term"> 914 915 "<code class="mallctl">arenas.narenas</code>" 916 917 (<span class="type">unsigned</span>) 918 <code class="literal">r-</code> 919 </span></dt><dd><p>Current limit on number of arenas.</p></dd><dt><a name="arenas.initialized"></a><span class="term"> 920 921 "<code class="mallctl">arenas.initialized</code>" 922 923 (<span class="type">bool *</span>) 924 <code class="literal">r-</code> 925 </span></dt><dd><p>An array of <a class="link" href="#arenas.narenas"> 926 "<code class="mallctl">arenas.narenas</code>" 927 </a> 928 booleans. Each boolean indicates whether the corresponding arena is 929 initialized.</p></dd><dt><a name="arenas.quantum"></a><span class="term"> 930 931 "<code class="mallctl">arenas.quantum</code>" 932 933 (<span class="type">size_t</span>) 934 <code class="literal">r-</code> 935 </span></dt><dd><p>Quantum size.</p></dd><dt><a name="arenas.page"></a><span class="term"> 936 937 "<code class="mallctl">arenas.page</code>" 938 939 (<span class="type">size_t</span>) 940 <code class="literal">r-</code> 941 </span></dt><dd><p>Page size.</p></dd><dt><a name="arenas.tcache_max"></a><span class="term"> 942 943 "<code class="mallctl">arenas.tcache_max</code>" 944 945 (<span class="type">size_t</span>) 946 <code class="literal">r-</code> 947 [<code class="option">--enable-tcache</code>] 948 </span></dt><dd><p>Maximum thread-cached size class.</p></dd><dt><a name="arenas.nbins"></a><span class="term"> 949 950 "<code class="mallctl">arenas.nbins</code>" 951 952 (<span class="type">unsigned</span>) 953 <code class="literal">r-</code> 954 </span></dt><dd><p>Number of bin size classes.</p></dd><dt><a name="arenas.nhbins"></a><span class="term"> 955 956 "<code class="mallctl">arenas.nhbins</code>" 957 958 (<span class="type">unsigned</span>) 959 <code class="literal">r-</code> 960 [<code class="option">--enable-tcache</code>] 961 </span></dt><dd><p>Total number of thread cache bin size 962 classes.</p></dd><dt><a name="arenas.bin.i.size"></a><span class="term"> 963 964 "<code class="mallctl">arenas.bin.<i>.size</code>" 965 966 (<span class="type">size_t</span>) 967 <code class="literal">r-</code> 968 </span></dt><dd><p>Maximum size supported by size class.</p></dd><dt><a name="arenas.bin.i.nregs"></a><span class="term"> 969 970 "<code class="mallctl">arenas.bin.<i>.nregs</code>" 971 972 (<span class="type">uint32_t</span>) 973 <code class="literal">r-</code> 974 </span></dt><dd><p>Number of regions per page run.</p></dd><dt><a name="arenas.bin.i.run_size"></a><span class="term"> 975 976 "<code class="mallctl">arenas.bin.<i>.run_size</code>" 977 978 (<span class="type">size_t</span>) 979 <code class="literal">r-</code> 980 </span></dt><dd><p>Number of bytes per page run.</p></dd><dt><a name="arenas.nlruns"></a><span class="term"> 981 982 "<code class="mallctl">arenas.nlruns</code>" 983 984 (<span class="type">size_t</span>) 985 <code class="literal">r-</code> 986 </span></dt><dd><p>Total number of large size classes.</p></dd><dt><a name="arenas.lrun.i.size"></a><span class="term"> 987 988 "<code class="mallctl">arenas.lrun.<i>.size</code>" 989 990 (<span class="type">size_t</span>) 991 <code class="literal">r-</code> 992 </span></dt><dd><p>Maximum size supported by this large size 993 class.</p></dd><dt><a name="arenas.purge"></a><span class="term"> 994 995 "<code class="mallctl">arenas.purge</code>" 996 997 (<span class="type">unsigned</span>) 998 <code class="literal">-w</code> 999 </span></dt><dd><p>Purge unused dirty pages for the specified arena, or 1000 for all arenas if none is specified.</p></dd><dt><a name="arenas.extend"></a><span class="term"> 1001 1002 "<code class="mallctl">arenas.extend</code>" 1003 1004 (<span class="type">unsigned</span>) 1005 <code class="literal">r-</code> 1006 </span></dt><dd><p>Extend the array of arenas by appending a new arena, 1007 and returning the new arena index.</p></dd><dt><a name="prof.active"></a><span class="term"> 1008 1009 "<code class="mallctl">prof.active</code>" 1010 1011 (<span class="type">bool</span>) 1012 <code class="literal">rw</code> 1013 [<code class="option">--enable-prof</code>] 1014 </span></dt><dd><p>Control whether sampling is currently active. See the 1015 <a class="link" href="#opt.prof_active"> 1016 "<code class="mallctl">opt.prof_active</code>" 1017 </a> 1018 option for additional information. 1019 </p></dd><dt><a name="prof.dump"></a><span class="term"> 1020 1021 "<code class="mallctl">prof.dump</code>" 1022 1023 (<span class="type">const char *</span>) 1024 <code class="literal">-w</code> 1025 [<code class="option">--enable-prof</code>] 1026 </span></dt><dd><p>Dump a memory profile to the specified file, or if NULL 1027 is specified, to a file according to the pattern 1028 <code class="filename"><prefix>.<pid>.<seq>.m<mseq>.heap</code>, 1029 where <code class="literal"><prefix></code> is controlled by the 1030 <a class="link" href="#opt.prof_prefix"> 1031 "<code class="mallctl">opt.prof_prefix</code>" 1032 </a> 1033 option.</p></dd><dt><a name="prof.interval"></a><span class="term"> 1034 1035 "<code class="mallctl">prof.interval</code>" 1036 1037 (<span class="type">uint64_t</span>) 1038 <code class="literal">r-</code> 1039 [<code class="option">--enable-prof</code>] 1040 </span></dt><dd><p>Average number of bytes allocated between 1041 inverval-based profile dumps. See the 1042 <a class="link" href="#opt.lg_prof_interval"> 1043 "<code class="mallctl">opt.lg_prof_interval</code>" 1044 </a> 1045 option for additional information.</p></dd><dt><a name="stats.cactive"></a><span class="term"> 1046 1047 "<code class="mallctl">stats.cactive</code>" 1048 1049 (<span class="type">size_t *</span>) 1050 <code class="literal">r-</code> 1051 [<code class="option">--enable-stats</code>] 1052 </span></dt><dd><p>Pointer to a counter that contains an approximate count 1053 of the current number of bytes in active pages. The estimate may be 1054 high, but never low, because each arena rounds up to the nearest 1055 multiple of the chunk size when computing its contribution to the 1056 counter. Note that the <a class="link" href="#epoch"> 1057 "<code class="mallctl">epoch</code>" 1058 </a> mallctl has no bearing 1059 on this counter. Furthermore, counter consistency is maintained via 1060 atomic operations, so it is necessary to use an atomic operation in 1061 order to guarantee a consistent read when dereferencing the pointer. 1062 </p></dd><dt><a name="stats.allocated"></a><span class="term"> 1063 1064 "<code class="mallctl">stats.allocated</code>" 1065 1066 (<span class="type">size_t</span>) 1067 <code class="literal">r-</code> 1068 [<code class="option">--enable-stats</code>] 1069 </span></dt><dd><p>Total number of bytes allocated by the 1070 application.</p></dd><dt><a name="stats.active"></a><span class="term"> 1071 1072 "<code class="mallctl">stats.active</code>" 1073 1074 (<span class="type">size_t</span>) 1075 <code class="literal">r-</code> 1076 [<code class="option">--enable-stats</code>] 1077 </span></dt><dd><p>Total number of bytes in active pages allocated by the 1078 application. This is a multiple of the page size, and greater than or 1079 equal to <a class="link" href="#stats.allocated"> 1080 "<code class="mallctl">stats.allocated</code>" 1081 </a>. 1082 This does not include <a class="link" href="#stats.arenas.i.pdirty"> 1083 1084 "<code class="mallctl">stats.arenas.<i>.pdirty</code>" 1085 </a> and pages 1086 entirely devoted to allocator metadata.</p></dd><dt><a name="stats.mapped"></a><span class="term"> 1087 1088 "<code class="mallctl">stats.mapped</code>" 1089 1090 (<span class="type">size_t</span>) 1091 <code class="literal">r-</code> 1092 [<code class="option">--enable-stats</code>] 1093 </span></dt><dd><p>Total number of bytes in chunks mapped on behalf of the 1094 application. This is a multiple of the chunk size, and is at least as 1095 large as <a class="link" href="#stats.active"> 1096 "<code class="mallctl">stats.active</code>" 1097 </a>. This 1098 does not include inactive chunks.</p></dd><dt><a name="stats.chunks.current"></a><span class="term"> 1099 1100 "<code class="mallctl">stats.chunks.current</code>" 1101 1102 (<span class="type">size_t</span>) 1103 <code class="literal">r-</code> 1104 [<code class="option">--enable-stats</code>] 1105 </span></dt><dd><p>Total number of chunks actively mapped on behalf of the 1106 application. This does not include inactive chunks. 1107 </p></dd><dt><a name="stats.chunks.total"></a><span class="term"> 1108 1109 "<code class="mallctl">stats.chunks.total</code>" 1110 1111 (<span class="type">uint64_t</span>) 1112 <code class="literal">r-</code> 1113 [<code class="option">--enable-stats</code>] 1114 </span></dt><dd><p>Cumulative number of chunks allocated.</p></dd><dt><a name="stats.chunks.high"></a><span class="term"> 1115 1116 "<code class="mallctl">stats.chunks.high</code>" 1117 1118 (<span class="type">size_t</span>) 1119 <code class="literal">r-</code> 1120 [<code class="option">--enable-stats</code>] 1121 </span></dt><dd><p>Maximum number of active chunks at any time thus far. 1122 </p></dd><dt><a name="stats.huge.allocated"></a><span class="term"> 1123 1124 "<code class="mallctl">stats.huge.allocated</code>" 1125 1126 (<span class="type">size_t</span>) 1127 <code class="literal">r-</code> 1128 [<code class="option">--enable-stats</code>] 1129 </span></dt><dd><p>Number of bytes currently allocated by huge objects. 1130 </p></dd><dt><a name="stats.huge.nmalloc"></a><span class="term"> 1131 1132 "<code class="mallctl">stats.huge.nmalloc</code>" 1133 1134 (<span class="type">uint64_t</span>) 1135 <code class="literal">r-</code> 1136 [<code class="option">--enable-stats</code>] 1137 </span></dt><dd><p>Cumulative number of huge allocation requests. 1138 </p></dd><dt><a name="stats.huge.ndalloc"></a><span class="term"> 1139 1140 "<code class="mallctl">stats.huge.ndalloc</code>" 1141 1142 (<span class="type">uint64_t</span>) 1143 <code class="literal">r-</code> 1144 [<code class="option">--enable-stats</code>] 1145 </span></dt><dd><p>Cumulative number of huge deallocation requests. 1146 </p></dd><dt><a name="stats.arenas.i.dss"></a><span class="term"> 1147 1148 "<code class="mallctl">stats.arenas.<i>.dss</code>" 1149 1150 (<span class="type">const char *</span>) 1151 <code class="literal">r-</code> 1152 </span></dt><dd><p>dss (<span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>) allocation precedence as 1153 related to <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span> allocation. See <a class="link" href="#opt.dss"> 1154 "<code class="mallctl">opt.dss</code>" 1155 </a> for details. 1156 </p></dd><dt><a name="stats.arenas.i.nthreads"></a><span class="term"> 1157 1158 "<code class="mallctl">stats.arenas.<i>.nthreads</code>" 1159 1160 (<span class="type">unsigned</span>) 1161 <code class="literal">r-</code> 1162 </span></dt><dd><p>Number of threads currently assigned to 1163 arena.</p></dd><dt><a name="stats.arenas.i.pactive"></a><span class="term"> 1164 1165 "<code class="mallctl">stats.arenas.<i>.pactive</code>" 1166 1167 (<span class="type">size_t</span>) 1168 <code class="literal">r-</code> 1169 </span></dt><dd><p>Number of pages in active runs.</p></dd><dt><a name="stats.arenas.i.pdirty"></a><span class="term"> 1170 1171 "<code class="mallctl">stats.arenas.<i>.pdirty</code>" 1172 1173 (<span class="type">size_t</span>) 1174 <code class="literal">r-</code> 1175 </span></dt><dd><p>Number of pages within unused runs that are potentially 1176 dirty, and for which <code class="function">madvise</code>(<em class="parameter"><code>...</code></em>, 1177 <em class="parameter"><code><code class="constant">MADV_DONTNEED</code></code></em>) or 1178 similar has not been called.</p></dd><dt><a name="stats.arenas.i.mapped"></a><span class="term"> 1179 1180 "<code class="mallctl">stats.arenas.<i>.mapped</code>" 1181 1182 (<span class="type">size_t</span>) 1183 <code class="literal">r-</code> 1184 [<code class="option">--enable-stats</code>] 1185 </span></dt><dd><p>Number of mapped bytes.</p></dd><dt><a name="stats.arenas.i.npurge"></a><span class="term"> 1186 1187 "<code class="mallctl">stats.arenas.<i>.npurge</code>" 1188 1189 (<span class="type">uint64_t</span>) 1190 <code class="literal">r-</code> 1191 [<code class="option">--enable-stats</code>] 1192 </span></dt><dd><p>Number of dirty page purge sweeps performed. 1193 </p></dd><dt><a name="stats.arenas.i.nmadvise"></a><span class="term"> 1194 1195 "<code class="mallctl">stats.arenas.<i>.nmadvise</code>" 1196 1197 (<span class="type">uint64_t</span>) 1198 <code class="literal">r-</code> 1199 [<code class="option">--enable-stats</code>] 1200 </span></dt><dd><p>Number of <code class="function">madvise</code>(<em class="parameter"><code>...</code></em>, 1201 <em class="parameter"><code><code class="constant">MADV_DONTNEED</code></code></em>) or 1202 similar calls made to purge dirty pages.</p></dd><dt><a name="stats.arenas.i.purged"></a><span class="term"> 1203 1204 "<code class="mallctl">stats.arenas.<i>.purged</code>" 1205 1206 (<span class="type">uint64_t</span>) 1207 <code class="literal">r-</code> 1208 [<code class="option">--enable-stats</code>] 1209 </span></dt><dd><p>Number of pages purged.</p></dd><dt><a name="stats.arenas.i.small.allocated"></a><span class="term"> 1210 1211 "<code class="mallctl">stats.arenas.<i>.small.allocated</code>" 1212 1213 (<span class="type">size_t</span>) 1214 <code class="literal">r-</code> 1215 [<code class="option">--enable-stats</code>] 1216 </span></dt><dd><p>Number of bytes currently allocated by small objects. 1217 </p></dd><dt><a name="stats.arenas.i.small.nmalloc"></a><span class="term"> 1218 1219 "<code class="mallctl">stats.arenas.<i>.small.nmalloc</code>" 1220 1221 (<span class="type">uint64_t</span>) 1222 <code class="literal">r-</code> 1223 [<code class="option">--enable-stats</code>] 1224 </span></dt><dd><p>Cumulative number of allocation requests served by 1225 small bins.</p></dd><dt><a name="stats.arenas.i.small.ndalloc"></a><span class="term"> 1226 1227 "<code class="mallctl">stats.arenas.<i>.small.ndalloc</code>" 1228 1229 (<span class="type">uint64_t</span>) 1230 <code class="literal">r-</code> 1231 [<code class="option">--enable-stats</code>] 1232 </span></dt><dd><p>Cumulative number of small objects returned to bins. 1233 </p></dd><dt><a name="stats.arenas.i.small.nrequests"></a><span class="term"> 1234 1235 "<code class="mallctl">stats.arenas.<i>.small.nrequests</code>" 1236 1237 (<span class="type">uint64_t</span>) 1238 <code class="literal">r-</code> 1239 [<code class="option">--enable-stats</code>] 1240 </span></dt><dd><p>Cumulative number of small allocation requests. 1241 </p></dd><dt><a name="stats.arenas.i.large.allocated"></a><span class="term"> 1242 1243 "<code class="mallctl">stats.arenas.<i>.large.allocated</code>" 1244 1245 (<span class="type">size_t</span>) 1246 <code class="literal">r-</code> 1247 [<code class="option">--enable-stats</code>] 1248 </span></dt><dd><p>Number of bytes currently allocated by large objects. 1249 </p></dd><dt><a name="stats.arenas.i.large.nmalloc"></a><span class="term"> 1250 1251 "<code class="mallctl">stats.arenas.<i>.large.nmalloc</code>" 1252 1253 (<span class="type">uint64_t</span>) 1254 <code class="literal">r-</code> 1255 [<code class="option">--enable-stats</code>] 1256 </span></dt><dd><p>Cumulative number of large allocation requests served 1257 directly by the arena.</p></dd><dt><a name="stats.arenas.i.large.ndalloc"></a><span class="term"> 1258 1259 "<code class="mallctl">stats.arenas.<i>.large.ndalloc</code>" 1260 1261 (<span class="type">uint64_t</span>) 1262 <code class="literal">r-</code> 1263 [<code class="option">--enable-stats</code>] 1264 </span></dt><dd><p>Cumulative number of large deallocation requests served 1265 directly by the arena.</p></dd><dt><a name="stats.arenas.i.large.nrequests"></a><span class="term"> 1266 1267 "<code class="mallctl">stats.arenas.<i>.large.nrequests</code>" 1268 1269 (<span class="type">uint64_t</span>) 1270 <code class="literal">r-</code> 1271 [<code class="option">--enable-stats</code>] 1272 </span></dt><dd><p>Cumulative number of large allocation requests. 1273 </p></dd><dt><a name="stats.arenas.i.bins.j.allocated"></a><span class="term"> 1274 1275 "<code class="mallctl">stats.arenas.<i>.bins.<j>.allocated</code>" 1276 1277 (<span class="type">size_t</span>) 1278 <code class="literal">r-</code> 1279 [<code class="option">--enable-stats</code>] 1280 </span></dt><dd><p>Current number of bytes allocated by 1281 bin.</p></dd><dt><a name="stats.arenas.i.bins.j.nmalloc"></a><span class="term"> 1282 1283 "<code class="mallctl">stats.arenas.<i>.bins.<j>.nmalloc</code>" 1284 1285 (<span class="type">uint64_t</span>) 1286 <code class="literal">r-</code> 1287 [<code class="option">--enable-stats</code>] 1288 </span></dt><dd><p>Cumulative number of allocations served by bin. 1289 </p></dd><dt><a name="stats.arenas.i.bins.j.ndalloc"></a><span class="term"> 1290 1291 "<code class="mallctl">stats.arenas.<i>.bins.<j>.ndalloc</code>" 1292 1293 (<span class="type">uint64_t</span>) 1294 <code class="literal">r-</code> 1295 [<code class="option">--enable-stats</code>] 1296 </span></dt><dd><p>Cumulative number of allocations returned to bin. 1297 </p></dd><dt><a name="stats.arenas.i.bins.j.nrequests"></a><span class="term"> 1298 1299 "<code class="mallctl">stats.arenas.<i>.bins.<j>.nrequests</code>" 1300 1301 (<span class="type">uint64_t</span>) 1302 <code class="literal">r-</code> 1303 [<code class="option">--enable-stats</code>] 1304 </span></dt><dd><p>Cumulative number of allocation 1305 requests.</p></dd><dt><a name="stats.arenas.i.bins.j.nfills"></a><span class="term"> 1306 1307 "<code class="mallctl">stats.arenas.<i>.bins.<j>.nfills</code>" 1308 1309 (<span class="type">uint64_t</span>) 1310 <code class="literal">r-</code> 1311 [<code class="option">--enable-stats</code> <code class="option">--enable-tcache</code>] 1312 </span></dt><dd><p>Cumulative number of tcache fills.</p></dd><dt><a name="stats.arenas.i.bins.j.nflushes"></a><span class="term"> 1313 1314 "<code class="mallctl">stats.arenas.<i>.bins.<j>.nflushes</code>" 1315 1316 (<span class="type">uint64_t</span>) 1317 <code class="literal">r-</code> 1318 [<code class="option">--enable-stats</code> <code class="option">--enable-tcache</code>] 1319 </span></dt><dd><p>Cumulative number of tcache flushes.</p></dd><dt><a name="stats.arenas.i.bins.j.nruns"></a><span class="term"> 1320 1321 "<code class="mallctl">stats.arenas.<i>.bins.<j>.nruns</code>" 1322 1323 (<span class="type">uint64_t</span>) 1324 <code class="literal">r-</code> 1325 [<code class="option">--enable-stats</code>] 1326 </span></dt><dd><p>Cumulative number of runs created.</p></dd><dt><a name="stats.arenas.i.bins.j.nreruns"></a><span class="term"> 1327 1328 "<code class="mallctl">stats.arenas.<i>.bins.<j>.nreruns</code>" 1329 1330 (<span class="type">uint64_t</span>) 1331 <code class="literal">r-</code> 1332 [<code class="option">--enable-stats</code>] 1333 </span></dt><dd><p>Cumulative number of times the current run from which 1334 to allocate changed.</p></dd><dt><a name="stats.arenas.i.bins.j.curruns"></a><span class="term"> 1335 1336 "<code class="mallctl">stats.arenas.<i>.bins.<j>.curruns</code>" 1337 1338 (<span class="type">size_t</span>) 1339 <code class="literal">r-</code> 1340 [<code class="option">--enable-stats</code>] 1341 </span></dt><dd><p>Current number of runs.</p></dd><dt><a name="stats.arenas.i.lruns.j.nmalloc"></a><span class="term"> 1342 1343 "<code class="mallctl">stats.arenas.<i>.lruns.<j>.nmalloc</code>" 1344 1345 (<span class="type">uint64_t</span>) 1346 <code class="literal">r-</code> 1347 [<code class="option">--enable-stats</code>] 1348 </span></dt><dd><p>Cumulative number of allocation requests for this size 1349 class served directly by the arena.</p></dd><dt><a name="stats.arenas.i.lruns.j.ndalloc"></a><span class="term"> 1350 1351 "<code class="mallctl">stats.arenas.<i>.lruns.<j>.ndalloc</code>" 1352 1353 (<span class="type">uint64_t</span>) 1354 <code class="literal">r-</code> 1355 [<code class="option">--enable-stats</code>] 1356 </span></dt><dd><p>Cumulative number of deallocation requests for this 1357 size class served directly by the arena.</p></dd><dt><a name="stats.arenas.i.lruns.j.nrequests"></a><span class="term"> 1358 1359 "<code class="mallctl">stats.arenas.<i>.lruns.<j>.nrequests</code>" 1360 1361 (<span class="type">uint64_t</span>) 1362 <code class="literal">r-</code> 1363 [<code class="option">--enable-stats</code>] 1364 </span></dt><dd><p>Cumulative number of allocation requests for this size 1365 class.</p></dd><dt><a name="stats.arenas.i.lruns.j.curruns"></a><span class="term"> 1366 1367 "<code class="mallctl">stats.arenas.<i>.lruns.<j>.curruns</code>" 1368 1369 (<span class="type">size_t</span>) 1370 <code class="literal">r-</code> 1371 [<code class="option">--enable-stats</code>] 1372 </span></dt><dd><p>Current number of runs for this size class. 1373 </p></dd></dl></div></div><div class="refsect1"><a name="debugging_malloc_problems"></a><h2>DEBUGGING MALLOC PROBLEMS</h2><p>When debugging, it is a good idea to configure/build jemalloc with 1374 the <code class="option">--enable-debug</code> and <code class="option">--enable-fill</code> 1375 options, and recompile the program with suitable options and symbols for 1376 debugger support. When so configured, jemalloc incorporates a wide variety 1377 of run-time assertions that catch application errors such as double-free, 1378 write-after-free, etc.</p><p>Programs often accidentally depend on “uninitialized” 1379 memory actually being filled with zero bytes. Junk filling 1380 (see the <a class="link" href="#opt.junk"> 1381 "<code class="mallctl">opt.junk</code>" 1382 </a> 1383 option) tends to expose such bugs in the form of obviously incorrect 1384 results and/or coredumps. Conversely, zero 1385 filling (see the <a class="link" href="#opt.zero"> 1386 "<code class="mallctl">opt.zero</code>" 1387 </a> option) eliminates 1388 the symptoms of such bugs. Between these two options, it is usually 1389 possible to quickly detect, diagnose, and eliminate such bugs.</p><p>This implementation does not provide much detail about the problems 1390 it detects, because the performance impact for storing such information 1391 would be prohibitive. However, jemalloc does integrate with the most 1392 excellent <a class="ulink" href="http://valgrind.org/" target="_top">Valgrind</a> tool if the 1393 <code class="option">--enable-valgrind</code> configuration option is enabled.</p></div><div class="refsect1"><a name="diagnostic_messages"></a><h2>DIAGNOSTIC MESSAGES</h2><p>If any of the memory allocation/deallocation functions detect an 1394 error or warning condition, a message will be printed to file descriptor 1395 <code class="constant">STDERR_FILENO</code>. Errors will result in the process 1396 dumping core. If the <a class="link" href="#opt.abort"> 1397 "<code class="mallctl">opt.abort</code>" 1398 </a> option is set, most 1399 warnings are treated as errors.</p><p>The <code class="varname">malloc_message</code> variable allows the programmer 1400 to override the function which emits the text strings forming the errors 1401 and warnings if for some reason the <code class="constant">STDERR_FILENO</code> file 1402 descriptor is not suitable for this. 1403 <code class="function">malloc_message</code>(<em class="parameter"><code></code></em>) takes the 1404 <em class="parameter"><code>cbopaque</code></em> pointer argument that is 1405 <code class="constant">NULL</code> unless overridden by the arguments in a call to 1406 <code class="function">malloc_stats_print</code>(<em class="parameter"><code></code></em>), followed by a string 1407 pointer. Please note that doing anything which tries to allocate memory in 1408 this function is likely to result in a crash or deadlock.</p><p>All messages are prefixed by 1409 “<code class="computeroutput"><jemalloc>: </code>”.</p></div><div class="refsect1"><a name="return_values"></a><h2>RETURN VALUES</h2><div class="refsect2"><a name="idm316388028784"></a><h3>Standard API</h3><p>The <code class="function">malloc</code>(<em class="parameter"><code></code></em>) and 1410 <code class="function">calloc</code>(<em class="parameter"><code></code></em>) functions return a pointer to the 1411 allocated memory if successful; otherwise a <code class="constant">NULL</code> 1412 pointer is returned and <code class="varname">errno</code> is set to 1413 <span class="errorname">ENOMEM</span>.</p><p>The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function 1414 returns the value 0 if successful; otherwise it returns an error value. 1415 The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function will fail 1416 if: 1417 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">EINVAL</span></span></dt><dd><p>The <em class="parameter"><code>alignment</code></em> parameter is 1418 not a power of 2 at least as large as 1419 <code class="code">sizeof(<span class="type">void *</span>)</code>. 1420 </p></dd><dt><span class="term"><span class="errorname">ENOMEM</span></span></dt><dd><p>Memory allocation error.</p></dd></dl></div><p> 1421 </p><p>The <code class="function">aligned_alloc</code>(<em class="parameter"><code></code></em>) function returns 1422 a pointer to the allocated memory if successful; otherwise a 1423 <code class="constant">NULL</code> pointer is returned and 1424 <code class="varname">errno</code> is set. The 1425 <code class="function">aligned_alloc</code>(<em class="parameter"><code></code></em>) function will fail if: 1426 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">EINVAL</span></span></dt><dd><p>The <em class="parameter"><code>alignment</code></em> parameter is 1427 not a power of 2. 1428 </p></dd><dt><span class="term"><span class="errorname">ENOMEM</span></span></dt><dd><p>Memory allocation error.</p></dd></dl></div><p> 1429 </p><p>The <code class="function">realloc</code>(<em class="parameter"><code></code></em>) function returns a 1430 pointer, possibly identical to <em class="parameter"><code>ptr</code></em>, to the 1431 allocated memory if successful; otherwise a <code class="constant">NULL</code> 1432 pointer is returned, and <code class="varname">errno</code> is set to 1433 <span class="errorname">ENOMEM</span> if the error was the result of an 1434 allocation failure. The <code class="function">realloc</code>(<em class="parameter"><code></code></em>) 1435 function always leaves the original buffer intact when an error occurs. 1436 </p><p>The <code class="function">free</code>(<em class="parameter"><code></code></em>) function returns no 1437 value.</p></div><div class="refsect2"><a name="idm316388003104"></a><h3>Non-standard API</h3><p>The <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) and 1438 <code class="function">rallocx</code>(<em class="parameter"><code></code></em>) functions return a pointer to 1439 the allocated memory if successful; otherwise a <code class="constant">NULL</code> 1440 pointer is returned to indicate insufficient contiguous memory was 1441 available to service the allocation request. </p><p>The <code class="function">xallocx</code>(<em class="parameter"><code></code></em>) function returns the 1442 real size of the resulting resized allocation pointed to by 1443 <em class="parameter"><code>ptr</code></em>, which is a value less than 1444 <em class="parameter"><code>size</code></em> if the allocation could not be adequately 1445 grown in place. </p><p>The <code class="function">sallocx</code>(<em class="parameter"><code></code></em>) function returns the 1446 real size of the allocation pointed to by <em class="parameter"><code>ptr</code></em>. 1447 </p><p>The <code class="function">nallocx</code>(<em class="parameter"><code></code></em>) returns the real size 1448 that would result from a successful equivalent 1449 <code class="function">mallocx</code>(<em class="parameter"><code></code></em>) function call, or zero if 1450 insufficient memory is available to perform the size computation. </p><p>The <code class="function">mallctl</code>(<em class="parameter"><code></code></em>), 1451 <code class="function">mallctlnametomib</code>(<em class="parameter"><code></code></em>), and 1452 <code class="function">mallctlbymib</code>(<em class="parameter"><code></code></em>) functions return 0 on 1453 success; otherwise they return an error value. The functions will fail 1454 if: 1455 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">EINVAL</span></span></dt><dd><p><em class="parameter"><code>newp</code></em> is not 1456 <code class="constant">NULL</code>, and <em class="parameter"><code>newlen</code></em> is too 1457 large or too small. Alternatively, <em class="parameter"><code>*oldlenp</code></em> 1458 is too large or too small; in this case as much data as possible 1459 are read despite the error.</p></dd><dt><span class="term"><span class="errorname">ENOENT</span></span></dt><dd><p><em class="parameter"><code>name</code></em> or 1460 <em class="parameter"><code>mib</code></em> specifies an unknown/invalid 1461 value.</p></dd><dt><span class="term"><span class="errorname">EPERM</span></span></dt><dd><p>Attempt to read or write void value, or attempt to 1462 write read-only value.</p></dd><dt><span class="term"><span class="errorname">EAGAIN</span></span></dt><dd><p>A memory allocation failure 1463 occurred.</p></dd><dt><span class="term"><span class="errorname">EFAULT</span></span></dt><dd><p>An interface with side effects failed in some way 1464 not directly related to <code class="function">mallctl*</code>(<em class="parameter"><code></code></em>) 1465 read/write processing.</p></dd></dl></div><p> 1466 </p><p>The <code class="function">malloc_usable_size</code>(<em class="parameter"><code></code></em>) function 1467 returns the usable size of the allocation pointed to by 1468 <em class="parameter"><code>ptr</code></em>. </p></div><div class="refsect2"><a name="idm316387973360"></a><h3>Experimental API</h3><p>The <code class="function">allocm</code>(<em class="parameter"><code></code></em>), 1469 <code class="function">rallocm</code>(<em class="parameter"><code></code></em>), 1470 <code class="function">sallocm</code>(<em class="parameter"><code></code></em>), 1471 <code class="function">dallocm</code>(<em class="parameter"><code></code></em>), and 1472 <code class="function">nallocm</code>(<em class="parameter"><code></code></em>) functions return 1473 <code class="constant">ALLOCM_SUCCESS</code> on success; otherwise they return an 1474 error value. The <code class="function">allocm</code>(<em class="parameter"><code></code></em>), 1475 <code class="function">rallocm</code>(<em class="parameter"><code></code></em>), and 1476 <code class="function">nallocm</code>(<em class="parameter"><code></code></em>) functions will fail if: 1477 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">ALLOCM_ERR_OOM</span></span></dt><dd><p>Out of memory. Insufficient contiguous memory was 1478 available to service the allocation request. The 1479 <code class="function">allocm</code>(<em class="parameter"><code></code></em>) function additionally sets 1480 <em class="parameter"><code>*ptr</code></em> to <code class="constant">NULL</code>, whereas 1481 the <code class="function">rallocm</code>(<em class="parameter"><code></code></em>) function leaves 1482 <code class="constant">*ptr</code> unmodified.</p></dd></dl></div><p> 1483 The <code class="function">rallocm</code>(<em class="parameter"><code></code></em>) function will also 1484 fail if: 1485 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><span class="errorname">ALLOCM_ERR_NOT_MOVED</span></span></dt><dd><p><code class="constant">ALLOCM_NO_MOVE</code> was specified, 1486 but the reallocation request could not be serviced without moving 1487 the object.</p></dd></dl></div><p> 1488 </p></div></div><div class="refsect1"><a name="environment"></a><h2>ENVIRONMENT</h2><p>The following environment variable affects the execution of the 1489 allocation functions: 1490 </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="envar">MALLOC_CONF</code></span></dt><dd><p>If the environment variable 1491 <code class="envar">MALLOC_CONF</code> is set, the characters it contains 1492 will be interpreted as options.</p></dd></dl></div><p> 1493 </p></div><div class="refsect1"><a name="examples"></a><h2>EXAMPLES</h2><p>To dump core whenever a problem occurs: 1494 </p><pre class="screen">ln -s 'abort:true' /etc/malloc.conf</pre><p> 1495 </p><p>To specify in the source a chunk size that is 16 MiB: 1496 </p><pre class="programlisting"> 1497malloc_conf = "lg_chunk:24";</pre></div><div class="refsect1"><a name="see_also"></a><h2>SEE ALSO</h2><p><span class="citerefentry"><span class="refentrytitle">madvise</span>(2)</span>, 1498 <span class="citerefentry"><span class="refentrytitle">mmap</span>(2)</span>, 1499 <span class="citerefentry"><span class="refentrytitle">sbrk</span>(2)</span>, 1500 <span class="citerefentry"><span class="refentrytitle">utrace</span>(2)</span>, 1501 <span class="citerefentry"><span class="refentrytitle">alloca</span>(3)</span>, 1502 <span class="citerefentry"><span class="refentrytitle">atexit</span>(3)</span>, 1503 <span class="citerefentry"><span class="refentrytitle">getpagesize</span>(3)</span></p></div><div class="refsect1"><a name="standards"></a><h2>STANDARDS</h2><p>The <code class="function">malloc</code>(<em class="parameter"><code></code></em>), 1504 <code class="function">calloc</code>(<em class="parameter"><code></code></em>), 1505 <code class="function">realloc</code>(<em class="parameter"><code></code></em>), and 1506 <code class="function">free</code>(<em class="parameter"><code></code></em>) functions conform to ISO/IEC 1507 9899:1990 (“ISO C90”).</p><p>The <code class="function">posix_memalign</code>(<em class="parameter"><code></code></em>) function conforms 1508 to IEEE Std 1003.1-2001 (“POSIX.1”).</p></div></div></body></html> 1509