project-base/server-skynet-source-3rd-jemalloc
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
3c2c5a5071
server-skynet-source-3rd-je.../doc/jemalloc.xml.in

2868 lines
126 KiB
XML
Raw Normal View History

Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<?xml version='1.0' encoding='UTF-8'?>
<?xml-stylesheet type="text/xsl"
href="http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl"?>
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.4//EN"
"http://www.oasis-open.org/docbook/xml/4.4/docbookx.dtd" [
]>
<refentry>
<refentryinfo>
<title>User Manual</title>
<productname>jemalloc</productname>
<releaseinfo role="version">@jemalloc_version@</releaseinfo>
<authorgroup>
<author>
<firstname>Jason</firstname>
<surname>Evans</surname>
<personblurb>Author</personblurb>
</author>
</authorgroup>
</refentryinfo>
<refmeta>
<refentrytitle>JEMALLOC</refentrytitle>
<manvolnum>3</manvolnum>
</refmeta>
<refnamediv>
<refdescriptor>jemalloc</refdescriptor>
<refname>jemalloc</refname>
<!-- Each refname causes a man page file to be created. Only if this were
the system malloc(3) implementation would these files be appropriate.
<refname>malloc</refname>
<refname>calloc</refname>
<refname>posix_memalign</refname>
Implement aligned_alloc(). Implement aligned_alloc(), which was added in the C11 standard. The function is weakly specified to the point that a minimally compliant implementation would be painful to use (size must be an integral multiple of alignment!), which in practice makes posix_memalign() a safer choice.
2012-03-14 03:55:21 +08:00
<refname>aligned_alloc</refname>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<refname>realloc</refname>
<refname>free</refname>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<refname>mallocx</refname>
<refname>rallocx</refname>
<refname>xallocx</refname>
<refname>sallocx</refname>
<refname>dallocx</refname>
Add support for sized deallocation. This adds a new `sdallocx` function to the external API, allowing the size to be passed by the caller. It avoids some extra reads in the thread cache fast path. In the case where stats are enabled, this avoids the work of calculating the size from the pointer. An assertion validates the size that's passed in, so enabling debugging will allow users of the API to debug cases where an incorrect size is passed in. The performance win for a contrived microbenchmark doing an allocation and immediately freeing it is ~10%. It may have a different impact on a real workload. Closes #28
2014-08-29 03:41:48 +08:00
<refname>sdallocx</refname>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<refname>nallocx</refname>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<refname>mallctl</refname>
<refname>mallctlnametomib</refname>
<refname>mallctlbymib</refname>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<refname>malloc_stats_print</refname>
<refname>malloc_usable_size</refname>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
-->
<refpurpose>general purpose memory allocation functions</refpurpose>
</refnamediv>
<refsect1 id="library">
<title>LIBRARY</title>
<para>This manual describes jemalloc @jemalloc_version@. More information
can be found at the <ulink
url="http://www.canonware.com/jemalloc/">jemalloc website</ulink>.</para>
</refsect1>
<refsynopsisdiv>
<title>SYNOPSIS</title>
<funcsynopsis>
Add missing header includes in jemalloc/jemalloc.h . Add stdlib.h, stdbool.h, and stdint.h to jemalloc/jemalloc.h so that applications only have to #include <jemalloc/jemalloc.h>. This resolves #132.
2014-10-06 03:05:37 +08:00
<funcsynopsisinfo>#include &lt;<filename class="headerfile">jemalloc/jemalloc.h</filename>&gt;</funcsynopsisinfo>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<refsect2>
<title>Standard API</title>
<funcprototype>
<funcdef>void *<function>malloc</function></funcdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>void *<function>calloc</function></funcdef>
<paramdef>size_t <parameter>number</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>int <function>posix_memalign</function></funcdef>
<paramdef>void **<parameter>ptr</parameter></paramdef>
<paramdef>size_t <parameter>alignment</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
</funcprototype>
Implement aligned_alloc(). Implement aligned_alloc(), which was added in the C11 standard. The function is weakly specified to the point that a minimally compliant implementation would be painful to use (size must be an integral multiple of alignment!), which in practice makes posix_memalign() a safer choice.
2012-03-14 03:55:21 +08:00
<funcprototype>
<funcdef>void *<function>aligned_alloc</function></funcdef>
<paramdef>size_t <parameter>alignment</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
</funcprototype>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<funcprototype>
<funcdef>void *<function>realloc</function></funcdef>
<paramdef>void *<parameter>ptr</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>void <function>free</function></funcdef>
<paramdef>void *<parameter>ptr</parameter></paramdef>
</funcprototype>
</refsect2>
<refsect2>
<title>Non-standard API</title>
<funcprototype>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<funcdef>void *<function>mallocx</function></funcdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
<paramdef>int <parameter>flags</parameter></paramdef>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</funcprototype>
<funcprototype>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<funcdef>void *<function>rallocx</function></funcdef>
<paramdef>void *<parameter>ptr</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
<paramdef>int <parameter>flags</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>size_t <function>xallocx</function></funcdef>
<paramdef>void *<parameter>ptr</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
<paramdef>size_t <parameter>extra</parameter></paramdef>
<paramdef>int <parameter>flags</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>size_t <function>sallocx</function></funcdef>
<paramdef>void *<parameter>ptr</parameter></paramdef>
<paramdef>int <parameter>flags</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>void <function>dallocx</function></funcdef>
<paramdef>void *<parameter>ptr</parameter></paramdef>
<paramdef>int <parameter>flags</parameter></paramdef>
</funcprototype>
Add support for sized deallocation. This adds a new `sdallocx` function to the external API, allowing the size to be passed by the caller. It avoids some extra reads in the thread cache fast path. In the case where stats are enabled, this avoids the work of calculating the size from the pointer. An assertion validates the size that's passed in, so enabling debugging will allow users of the API to debug cases where an incorrect size is passed in. The performance win for a contrived microbenchmark doing an allocation and immediately freeing it is ~10%. It may have a different impact on a real workload. Closes #28
2014-08-29 03:41:48 +08:00
<funcprototype>
<funcdef>void <function>sdallocx</function></funcdef>
<paramdef>void *<parameter>ptr</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
<paramdef>int <parameter>flags</parameter></paramdef>
</funcprototype>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<funcprototype>
<funcdef>size_t <function>nallocx</function></funcdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
<paramdef>int <parameter>flags</parameter></paramdef>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</funcprototype>
<funcprototype>
<funcdef>int <function>mallctl</function></funcdef>
<paramdef>const char *<parameter>name</parameter></paramdef>
<paramdef>void *<parameter>oldp</parameter></paramdef>
<paramdef>size_t *<parameter>oldlenp</parameter></paramdef>
<paramdef>void *<parameter>newp</parameter></paramdef>
<paramdef>size_t <parameter>newlen</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>int <function>mallctlnametomib</function></funcdef>
<paramdef>const char *<parameter>name</parameter></paramdef>
<paramdef>size_t *<parameter>mibp</parameter></paramdef>
<paramdef>size_t *<parameter>miblenp</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>int <function>mallctlbymib</function></funcdef>
<paramdef>const size_t *<parameter>mib</parameter></paramdef>
<paramdef>size_t <parameter>miblen</parameter></paramdef>
<paramdef>void *<parameter>oldp</parameter></paramdef>
<paramdef>size_t *<parameter>oldlenp</parameter></paramdef>
<paramdef>void *<parameter>newp</parameter></paramdef>
<paramdef>size_t <parameter>newlen</parameter></paramdef>
</funcprototype>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<funcprototype>
<funcdef>void <function>malloc_stats_print</function></funcdef>
<paramdef>void <parameter>(*write_cb)</parameter>
<funcparams>void *, const char *</funcparams>
</paramdef>
<paramdef>void *<parameter>cbopaque</parameter></paramdef>
<paramdef>const char *<parameter>opts</parameter></paramdef>
</funcprototype>
<funcprototype>
<funcdef>size_t <function>malloc_usable_size</function></funcdef>
<paramdef>const void *<parameter>ptr</parameter></paramdef>
</funcprototype>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<funcprototype>
<funcdef>void <function>(*malloc_message)</function></funcdef>
<paramdef>void *<parameter>cbopaque</parameter></paramdef>
<paramdef>const char *<parameter>s</parameter></paramdef>
</funcprototype>
<para><type>const char *</type><varname>malloc_conf</varname>;</para>
</refsect2>
</funcsynopsis>
</refsynopsisdiv>
<refsect1 id="description">
<title>DESCRIPTION</title>
<refsect2>
<title>Standard API</title>
<para>The <function>malloc<parameter/></function> function allocates
<parameter>size</parameter> bytes of uninitialized memory. The allocated
space is suitably aligned (after possible pointer coercion) for storage
of any type of object.</para>
<para>The <function>calloc<parameter/></function> function allocates
space for <parameter>number</parameter> objects, each
<parameter>size</parameter> bytes in length. The result is identical to
calling <function>malloc<parameter/></function> with an argument of
<parameter>number</parameter> * <parameter>size</parameter>, with the
exception that the allocated memory is explicitly initialized to zero
bytes.</para>
<para>The <function>posix_memalign<parameter/></function> function
allocates <parameter>size</parameter> bytes of memory such that the
Minor doc edit.
2014-05-16 13:46:24 +08:00
allocation's base address is a multiple of
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<parameter>alignment</parameter>, and returns the allocation in the value
pointed to by <parameter>ptr</parameter>. The requested
Minor doc edit.
2014-05-16 13:46:24 +08:00
<parameter>alignment</parameter> must be a power of 2 at least as large as
<code language="C">sizeof(<type>void *</type>)</code>.</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
Implement aligned_alloc(). Implement aligned_alloc(), which was added in the C11 standard. The function is weakly specified to the point that a minimally compliant implementation would be painful to use (size must be an integral multiple of alignment!), which in practice makes posix_memalign() a safer choice.
2012-03-14 03:55:21 +08:00
<para>The <function>aligned_alloc<parameter/></function> function
allocates <parameter>size</parameter> bytes of memory such that the
Minor doc edit.
2014-05-16 13:46:24 +08:00
allocation's base address is a multiple of
Implement aligned_alloc(). Implement aligned_alloc(), which was added in the C11 standard. The function is weakly specified to the point that a minimally compliant implementation would be painful to use (size must be an integral multiple of alignment!), which in practice makes posix_memalign() a safer choice.
2012-03-14 03:55:21 +08:00
<parameter>alignment</parameter>. The requested
<parameter>alignment</parameter> must be a power of 2. Behavior is
undefined if <parameter>size</parameter> is not an integral multiple of
<parameter>alignment</parameter>.</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<para>The <function>realloc<parameter/></function> function changes the
size of the previously allocated memory referenced by
<parameter>ptr</parameter> to <parameter>size</parameter> bytes. The
contents of the memory are unchanged up to the lesser of the new and old
sizes. If the new size is larger, the contents of the newly allocated
portion of the memory are undefined. Upon success, the memory referenced
by <parameter>ptr</parameter> is freed and a pointer to the newly
allocated memory is returned. Note that
<function>realloc<parameter/></function> may move the memory allocation,
resulting in a different return value than <parameter>ptr</parameter>.
If <parameter>ptr</parameter> is <constant>NULL</constant>, the
<function>realloc<parameter/></function> function behaves identically to
<function>malloc<parameter/></function> for the specified size.</para>
<para>The <function>free<parameter/></function> function causes the
allocated memory referenced by <parameter>ptr</parameter> to be made
available for future allocations. If <parameter>ptr</parameter> is
<constant>NULL</constant>, no action occurs.</para>
</refsect2>
<refsect2>
<title>Non-standard API</title>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>mallocx<parameter/></function>,
<function>rallocx<parameter/></function>,
<function>xallocx<parameter/></function>,
<function>sallocx<parameter/></function>,
Add support for sized deallocation. This adds a new `sdallocx` function to the external API, allowing the size to be passed by the caller. It avoids some extra reads in the thread cache fast path. In the case where stats are enabled, this avoids the work of calculating the size from the pointer. An assertion validates the size that's passed in, so enabling debugging will allow users of the API to debug cases where an incorrect size is passed in. The performance win for a contrived microbenchmark doing an allocation and immediately freeing it is ~10%. It may have a different impact on a real workload. Closes #28
2014-08-29 03:41:48 +08:00
<function>dallocx<parameter/></function>,
<function>sdallocx<parameter/></function>, and
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<function>nallocx<parameter/></function> functions all have a
<parameter>flags</parameter> argument that can be used to specify
options. The functions only check the options that are contextually
relevant. Use bitwise or (<code language="C">|</code>) operations to
specify one or more of the following:
<variablelist>
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<varlistentry id="MALLOCX_LG_ALIGN">
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<term><constant>MALLOCX_LG_ALIGN(<parameter>la</parameter>)
</constant></term>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<listitem><para>Align the memory allocation to start at an address
that is a multiple of <code language="C">(1 &lt;&lt;
<parameter>la</parameter>)</code>. This macro does not validate
that <parameter>la</parameter> is within the valid
range.</para></listitem>
</varlistentry>
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<varlistentry id="MALLOCX_ALIGN">
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<term><constant>MALLOCX_ALIGN(<parameter>a</parameter>)
</constant></term>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<listitem><para>Align the memory allocation to start at an address
that is a multiple of <parameter>a</parameter>, where
<parameter>a</parameter> is a power of two. This macro does not
validate that <parameter>a</parameter> is a power of 2.
</para></listitem>
</varlistentry>
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<varlistentry id="MALLOCX_ZERO">
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<term><constant>MALLOCX_ZERO</constant></term>
<listitem><para>Initialize newly allocated memory to contain zero
bytes. In the growing reallocation case, the real size prior to
reallocation defines the boundary between untouched bytes and those
that are initialized to contain zero bytes. If this macro is
absent, newly allocated memory is uninitialized.</para></listitem>
</varlistentry>
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<varlistentry id="MALLOCX_TCACHE">
<term><constant>MALLOCX_TCACHE(<parameter>tc</parameter>)
</constant></term>
<listitem><para>Use the thread-specific cache (tcache) specified by
the identifier <parameter>tc</parameter>, which must have been
acquired via the <link
linkend="tcache.create"><mallctl>tcache.create</mallctl></link>
mallctl. This macro does not validate that
<parameter>tc</parameter> specifies a valid
identifier.</para></listitem>
</varlistentry>
<varlistentry id="MALLOC_TCACHE_NONE">
<term><constant>MALLOCX_TCACHE_NONE</constant></term>
<listitem><para>Do not use a thread-specific cache (tcache). Unless
<constant>MALLOCX_TCACHE(<parameter>tc</parameter>)</constant> or
<constant>MALLOCX_TCACHE_NONE</constant> is specified, an
automatically managed tcache will be used under many circumstances.
This macro cannot be used in the same <parameter>flags</parameter>
argument as
<constant>MALLOCX_TCACHE(<parameter>tc</parameter>)</constant>.</para></listitem>
</varlistentry>
<varlistentry id="MALLOCX_ARENA">
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<term><constant>MALLOCX_ARENA(<parameter>a</parameter>)
</constant></term>
<listitem><para>Use the arena specified by the index
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<parameter>a</parameter>. This macro has no effect for regions that
were allocated via an arena other than the one specified. This
macro does not validate that <parameter>a</parameter> specifies an
arena index in the valid range.</para></listitem>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
</varlistentry>
</variablelist>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</para>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>mallocx<parameter/></function> function allocates at
least <parameter>size</parameter> bytes of memory, and returns a pointer
to the base address of the allocation. Behavior is undefined if
Extract profiling code from [re]allocation functions. Extract profiling code from malloc(), imemalign(), calloc(), realloc(), mallocx(), rallocx(), and xallocx(). This slightly reduces the amount of code compiled into the fast paths, but the primary benefit is the combinatorial complexity reduction. Simplify iralloc[t]() by creating a separate ixalloc() that handles the no-move cases. Further simplify [mrxn]allocx() (and by implication [mrn]allocm()) to make request size overflows due to size class and/or alignment constraints trigger undefined behavior (detected by debug-only assertions). Report ENOMEM rather than EINVAL if an OOM occurs during heap profiling backtrace creation in imemalign(). This bug impacted posix_memalign() and aligned_alloc().
2014-01-13 07:05:44 +08:00
<parameter>size</parameter> is <constant>0</constant>, or if request size
overflows due to size class and/or alignment constraints.</para>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>rallocx<parameter/></function> function resizes the
allocation at <parameter>ptr</parameter> to be at least
<parameter>size</parameter> bytes, and returns a pointer to the base
address of the resulting allocation, which may or may not have moved from
its original location. Behavior is undefined if
Extract profiling code from [re]allocation functions. Extract profiling code from malloc(), imemalign(), calloc(), realloc(), mallocx(), rallocx(), and xallocx(). This slightly reduces the amount of code compiled into the fast paths, but the primary benefit is the combinatorial complexity reduction. Simplify iralloc[t]() by creating a separate ixalloc() that handles the no-move cases. Further simplify [mrxn]allocx() (and by implication [mrn]allocm()) to make request size overflows due to size class and/or alignment constraints trigger undefined behavior (detected by debug-only assertions). Report ENOMEM rather than EINVAL if an OOM occurs during heap profiling backtrace creation in imemalign(). This bug impacted posix_memalign() and aligned_alloc().
2014-01-13 07:05:44 +08:00
<parameter>size</parameter> is <constant>0</constant>, or if request size
overflows due to size class and/or alignment constraints.</para>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>xallocx<parameter/></function> function resizes the
allocation at <parameter>ptr</parameter> in place to be at least
<parameter>size</parameter> bytes, and returns the real size of the
allocation. If <parameter>extra</parameter> is non-zero, an attempt is
made to resize the allocation to be at least <code
language="C">(<parameter>size</parameter> +
<parameter>extra</parameter>)</code> bytes, though inability to allocate
the extra byte(s) will not by itself result in failure to resize.
Behavior is undefined if <parameter>size</parameter> is
<constant>0</constant>, or if <code
language="C">(<parameter>size</parameter> + <parameter>extra</parameter>
&gt; <constant>SIZE_T_MAX</constant>)</code>.</para>
<para>The <function>sallocx<parameter/></function> function returns the
real size of the allocation at <parameter>ptr</parameter>.</para>
<para>The <function>dallocx<parameter/></function> function causes the
memory referenced by <parameter>ptr</parameter> to be made available for
future allocations.</para>
Add support for sized deallocation. This adds a new `sdallocx` function to the external API, allowing the size to be passed by the caller. It avoids some extra reads in the thread cache fast path. In the case where stats are enabled, this avoids the work of calculating the size from the pointer. An assertion validates the size that's passed in, so enabling debugging will allow users of the API to debug cases where an incorrect size is passed in. The performance win for a contrived microbenchmark doing an allocation and immediately freeing it is ~10%. It may have a different impact on a real workload. Closes #28
2014-08-29 03:41:48 +08:00
<para>The <function>sdallocx<parameter/></function> function is an
extension of <function>dallocx<parameter/></function> with a
<parameter>size</parameter> parameter to allow the caller to pass in the
allocation size as an optimization. The minimum valid input size is the
original requested size of the allocation, and the maximum valid input
size is the corresponding value returned by
<function>nallocx<parameter/></function> or
<function>sallocx<parameter/></function>.</para>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>nallocx<parameter/></function> function allocates no
memory, but it performs the same size computation as the
<function>mallocx<parameter/></function> function, and returns the real
size of the allocation that would result from the equivalent
<function>mallocx<parameter/></function> function call. Behavior is
Extract profiling code from [re]allocation functions. Extract profiling code from malloc(), imemalign(), calloc(), realloc(), mallocx(), rallocx(), and xallocx(). This slightly reduces the amount of code compiled into the fast paths, but the primary benefit is the combinatorial complexity reduction. Simplify iralloc[t]() by creating a separate ixalloc() that handles the no-move cases. Further simplify [mrxn]allocx() (and by implication [mrn]allocm()) to make request size overflows due to size class and/or alignment constraints trigger undefined behavior (detected by debug-only assertions). Report ENOMEM rather than EINVAL if an OOM occurs during heap profiling backtrace creation in imemalign(). This bug impacted posix_memalign() and aligned_alloc().
2014-01-13 07:05:44 +08:00
undefined if <parameter>size</parameter> is <constant>0</constant>, or if
request size overflows due to size class and/or alignment
constraints.</para>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<para>The <function>mallctl<parameter/></function> function provides a
general interface for introspecting the memory allocator, as well as
setting modifiable parameters and triggering actions. The
period-separated <parameter>name</parameter> argument specifies a
location in a tree-structured namespace; see the <xref
linkend="mallctl_namespace" xrefstyle="template:%t"/> section for
documentation on the tree contents. To read a value, pass a pointer via
<parameter>oldp</parameter> to adequate space to contain the value, and a
pointer to its length via <parameter>oldlenp</parameter>; otherwise pass
<constant>NULL</constant> and <constant>NULL</constant>. Similarly, to
write a value, pass a pointer to the value via
<parameter>newp</parameter>, and its length via
<parameter>newlen</parameter>; otherwise pass <constant>NULL</constant>
and <constant>0</constant>.</para>
<para>The <function>mallctlnametomib<parameter/></function> function
provides a way to avoid repeated name lookups for applications that
repeatedly query the same portion of the namespace, by translating a name
to a &ldquo;Management Information Base&rdquo; (MIB) that can be passed
repeatedly to <function>mallctlbymib<parameter/></function>. Upon
successful return from <function>mallctlnametomib<parameter/></function>,
<parameter>mibp</parameter> contains an array of
<parameter>*miblenp</parameter> integers, where
<parameter>*miblenp</parameter> is the lesser of the number of components
in <parameter>name</parameter> and the input value of
<parameter>*miblenp</parameter>. Thus it is possible to pass a
<parameter>*miblenp</parameter> that is smaller than the number of
period-separated name components, which results in a partial MIB that can
be used as the basis for constructing a complete MIB. For name
components that are integers (e.g. the 2 in
<link
linkend="arenas.bin.i.size"><mallctl>arenas.bin.2.size</mallctl></link>),
the corresponding MIB component will always be that integer. Therefore,
it is legitimate to construct code like the following: <programlisting
language="C"><![CDATA[
unsigned nbins, i;
Add mallctl*() unit tests.
2013-12-19 15:21:42 +08:00
size_t mib[4];
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
size_t len, miblen;
len = sizeof(nbins);
mallctl("arenas.nbins", &nbins, &len, NULL, 0);
miblen = 4;
Add mallctl*() unit tests.
2013-12-19 15:21:42 +08:00
mallctlnametomib("arenas.bin.0.size", mib, &miblen);
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
for (i = 0; i < nbins; i++) {
size_t bin_size;
mib[2] = i;
len = sizeof(bin_size);
mallctlbymib(mib, miblen, &bin_size, &len, NULL, 0);
/* Do something with bin_size... */
}]]></programlisting></para>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>malloc_stats_print<parameter/></function> function
writes human-readable summary statistics via the
<parameter>write_cb</parameter> callback function pointer and
<parameter>cbopaque</parameter> data passed to
<parameter>write_cb</parameter>, or
<function>malloc_message<parameter/></function> if
<parameter>write_cb</parameter> is <constant>NULL</constant>. This
function can be called repeatedly. General information that never
changes during execution can be omitted by specifying "g" as a character
within the <parameter>opts</parameter> string. Note that
<function>malloc_message<parameter/></function> uses the
<function>mallctl*<parameter/></function> functions internally, so
inconsistent statistics can be reported if multiple threads use these
functions simultaneously. If <option>--enable-stats</option> is
specified during configuration, &ldquo;m&rdquo; and &ldquo;a&rdquo; can
be specified to omit merged arena and per arena statistics, respectively;
Add per size class huge allocation statistics. Add per size class huge allocation statistics, and normalize various stats: - Change the arenas.nlruns type from size_t to unsigned. - Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's. - Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with stats.arenas.<i>.bins.<j>.curregs . - Add the stats.arenas.<i>.hchunks.<j>.nmalloc, stats.arenas.<i>.hchunks.<j>.ndalloc, stats.arenas.<i>.hchunks.<j>.nrequests, and stats.arenas.<i>.hchunks.<j>.curhchunks mallctl's.
2014-10-13 13:53:59 +08:00
&ldquo;b&rdquo;, &ldquo;l&rdquo;, and &ldquo;h&rdquo; can be specified to
omit per size class statistics for bins, large objects, and huge objects,
respectively. Unrecognized characters are silently ignored. Note that
thread caching may prevent some statistics from being completely up to
date, since extra locking would be required to merge counters that track
thread cache operations.
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
</para>
<para>The <function>malloc_usable_size<parameter/></function> function
returns the usable size of the allocation pointed to by
<parameter>ptr</parameter>. The return value may be larger than the size
that was requested during allocation. The
<function>malloc_usable_size<parameter/></function> function is not a
mechanism for in-place <function>realloc<parameter/></function>; rather
it is provided solely as a tool for introspection purposes. Any
discrepancy between the requested allocation size and the size reported
by <function>malloc_usable_size<parameter/></function> should not be
depended on, since such behavior is entirely implementation-dependent.
</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</refsect2>
</refsect1>
<refsect1 id="tuning">
<title>TUNING</title>
<para>Once, when the first call is made to one of the memory allocation
routines, the allocator initializes its internals based in part on various
options that can be specified at compile- or run-time.</para>
Add --with-malloc-conf. Add --with-malloc-conf, which makes it possible to embed a default options string during configuration.
2016-02-08 06:23:22 +08:00
<para>The string specified via <option>--with-malloc-conf</option>, the
string pointed to by the global variable <varname>malloc_conf</varname>, the
&ldquo;name&rdquo; of the file referenced by the symbolic link named
<filename class="symlink">/etc/malloc.conf</filename>, and the value of the
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
environment variable <envar>MALLOC_CONF</envar>, will be interpreted, in
Clarify how to use malloc_conf. Clarify that malloc_conf is intended only for compile-time configuration, since jemalloc may be initialized before main() is entered.
2013-03-20 07:28:41 +08:00
that order, from left to right as options. Note that
<varname>malloc_conf</varname> may be read before
<function>main<parameter/></function> is entered, so the declaration of
<varname>malloc_conf</varname> should specify an initializer that contains
Add --with-malloc-conf. Add --with-malloc-conf, which makes it possible to embed a default options string during configuration.
2016-02-08 06:23:22 +08:00
the final value to be read by jemalloc. <option>--with-malloc-conf</option>
and <varname>malloc_conf</varname> are compile-time mechanisms, whereas
<filename class="symlink">/etc/malloc.conf</filename> and
<envar>MALLOC_CONF</envar> can be safely set any time prior to program
invocation.</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<para>An options string is a comma-separated list of option:value pairs.
There is one key corresponding to each <link
linkend="opt.abort"><mallctl>opt.*</mallctl></link> mallctl (see the <xref
linkend="mallctl_namespace" xrefstyle="template:%t"/> section for options
documentation). For example, <literal>abort:true,narenas:1</literal> sets
the <link linkend="opt.abort"><mallctl>opt.abort</mallctl></link> and <link
linkend="opt.narenas"><mallctl>opt.narenas</mallctl></link> options. Some
options have boolean values (true/false), others have integer values (base
8, 10, or 16, depending on prefix), and yet others have raw string
values.</para>
</refsect1>
<refsect1 id="implementation_notes">
<title>IMPLEMENTATION NOTES</title>
<para>Traditionally, allocators have used
<citerefentry><refentrytitle>sbrk</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> to obtain memory, which is
suboptimal for several reasons, including race conditions, increased
fragmentation, and artificial limitations on maximum usable memory. If
Make dss non-optional, and fix an "arena.<i>.dss" mallctl bug. Make dss non-optional on all platforms which support sbrk(2). Fix the "arena.<i>.dss" mallctl to return an error if "primary" or "secondary" precedence is specified, but sbrk(2) is not supported.
2014-04-16 03:09:48 +08:00
<citerefentry><refentrytitle>sbrk</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> is supported by the operating
system, this allocator uses both
<citerefentry><refentrytitle>mmap</refentrytitle>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<manvolnum>2</manvolnum></citerefentry> and
Fix a memory corruption bug in chunk_alloc_dss(). Fix a memory corruption bug in chunk_alloc_dss() that was due to claiming newly allocated memory is zeroed. Reverse order of preference between mmap() and sbrk() to prefer mmap(). Clean up management of 'zero' parameter in chunk_alloc*().
2012-04-22 04:33:48 +08:00
<citerefentry><refentrytitle>sbrk</refentrytitle>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<manvolnum>2</manvolnum></citerefentry>, in that order of preference;
otherwise only <citerefentry><refentrytitle>mmap</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> is used.</para>
<para>This allocator uses multiple arenas in order to reduce lock
contention for threaded programs on multi-processor systems. This works
well with regard to threading scalability, but incurs some costs. There is
a small fixed per-arena overhead, and additionally, arenas manage memory
completely independently of each other, which means a small fixed increase
in overall memory fragmentation. These overheads are not generally an
issue, given the number of arenas normally used. Note that using
substantially more arenas than the default is not likely to improve
performance, mainly due to reduced cache performance. However, it may make
sense to reduce the number of arenas if an application does not make much
use of the allocation functions.</para>
<para>In addition to multiple arenas, unless
<option>--disable-tcache</option> is specified during configuration, this
allocator supports thread-specific caching for small and large objects, in
order to make it possible to completely avoid synchronization for most
allocation requests. Such caching allows very fast allocation in the
common case, but it increases memory usage and fragmentation, since a
bounded number of objects can remain allocated in each thread cache.</para>
<para>Memory is conceptually broken into equal-sized chunks, where the
chunk size is a power of two that is greater than the page size. Chunks
are always aligned to multiples of the chunk size. This alignment makes it
possible to find metadata for user objects very quickly.</para>
<para>User objects are broken into three categories according to size:
Update size class documentation.
2014-10-15 13:19:21 +08:00
small, large, and huge. Small and large objects are managed entirely by
arenas; huge objects are additionally aggregated in a single data structure
that is shared by all threads. Huge objects are typically used by
applications infrequently enough that this single data structure is not a
scalability issue.</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<para>Each chunk that is managed by an arena tracks its contents as runs of
contiguous pages (unused, backing a set of small objects, or backing one
large object). The combination of chunk alignment and chunk page maps
makes it possible to determine all metadata regarding small and large
allocations in constant time.</para>
<para>Small objects are managed in groups by page runs. Each run maintains
Update size class documentation.
2014-10-15 13:19:21 +08:00
a bitmap to track which regions are in use. Allocation requests that are no
more than half the quantum (8 or 16, depending on architecture) are rounded
up to the nearest power of two that is at least <code
language="C">sizeof(<type>double</type>)</code>. All other object size
classes are multiples of the quantum, spaced such that there are four size
classes for each doubling in size, which limits internal fragmentation to
approximately 20% for all but the smallest size classes. Small size classes
are smaller than four times the page size, large size classes are smaller
than the chunk size (see the <link
linkend="opt.lg_chunk"><mallctl>opt.lg_chunk</mallctl></link> option), and
huge size classes extend from the chunk size up to one size class less than
the full address space size.</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<para>Allocations are packed tightly together, which can be an issue for
multi-threaded applications. If you need to assure that allocations do not
suffer from cacheline sharing, round your allocation requests up to the
nearest multiple of the cacheline size, or specify cacheline alignment when
allocating.</para>
Document under what circumstances in-place resizing succeeds. This resolves #100.
2015-01-23 07:28:25 +08:00
<para>The <function>realloc<parameter/></function>,
<function>rallocx<parameter/></function>, and
<function>xallocx<parameter/></function> functions may resize allocations
without moving them under limited circumstances. Unlike the
<function>*allocx<parameter/></function> API, the standard API does not
officially round up the usable size of an allocation to the nearest size
class, so technically it is necessary to call
<function>realloc<parameter/></function> to grow e.g. a 9-byte allocation to
16 bytes, or shrink a 16-byte allocation to 9 bytes. Growth and shrinkage
trivially succeeds in place as long as the pre-size and post-size both round
up to the same size class. No other API guarantees are made regarding
in-place resizing, but the current implementation also tries to resize large
and huge allocations in place, as long as the pre-size and post-size are
Update in-place reallocation documentation.
2015-08-14 15:55:44 +08:00
both large or both huge. In such cases shrinkage always succeeds for large
size classes, but for huge size classes the chunk allocator must support
splitting (see <link
linkend="arena.i.chunk_hooks"><mallctl>arena.&lt;i&gt;.chunk_hooks</mallctl></link>).
Growth only succeeds if the trailing memory is currently available, and
additionally for huge size classes the chunk allocator must support
merging.</para>
Document under what circumstances in-place resizing succeeds. This resolves #100.
2015-01-23 07:28:25 +08:00
Update large/huge size class cutoff documentation.
2015-08-14 15:35:11 +08:00
<para>Assuming 2 MiB chunks, 4 KiB pages, and a 16-byte quantum on a
Change default chunk size from 4 MiB to 256 KiB. Recent changes have improved huge allocation scalability, which removes upward pressure to set the chunk size so large that huge allocations are rare. Smaller chunks are more likely to completely drain, so set the default to the smallest size that doesn't leave excessive unusable trailing space in chunk headers.
2015-03-07 12:05:16 +08:00
64-bit system, the size classes in each category are as shown in <xref
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
linkend="size_classes" xrefstyle="template:Table %n"/>.</para>
<table xml:id="size_classes" frame="all">
<title>Size classes</title>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<tgroup cols="3" colsep="1" rowsep="1">
<colspec colname="c1" align="left"/>
<colspec colname="c2" align="right"/>
<colspec colname="c3" align="left"/>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<thead>
<row>
<entry>Category</entry>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>Spacing</entry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<entry>Size</entry>
</row>
</thead>
<tbody>
<row>
Update size class documentation.
2014-10-15 13:19:21 +08:00
<entry morerows="8">Small</entry>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>lg</entry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<entry>[8]</entry>
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>16</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
<entry>[16, 32, 48, 64, 80, 96, 112, 128]</entry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>32</entry>
Simplify small size class infrastructure. Program-generate small size class tables for all valid combinations of LG_TINY_MIN, LG_QUANTUM, and PAGE_SHIFT. Use the appropriate table to generate all relevant data structures, and remove the distinction between tiny/quantum/cacheline/subpage bins. Remove --enable-dynamic-page-shift. This option didn't prove useful in practice, and it prevented optimizations. Add Tilera architecture support.
2012-02-29 08:50:47 +08:00
<entry>[160, 192, 224, 256]</entry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>64</entry>
Simplify small size class infrastructure. Program-generate small size class tables for all valid combinations of LG_TINY_MIN, LG_QUANTUM, and PAGE_SHIFT. Use the appropriate table to generate all relevant data structures, and remove the distinction between tiny/quantum/cacheline/subpage bins. Remove --enable-dynamic-page-shift. This option didn't prove useful in practice, and it prevented optimizations. Add Tilera architecture support.
2012-02-29 08:50:47 +08:00
<entry>[320, 384, 448, 512]</entry>
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>128</entry>
Simplify small size class infrastructure. Program-generate small size class tables for all valid combinations of LG_TINY_MIN, LG_QUANTUM, and PAGE_SHIFT. Use the appropriate table to generate all relevant data structures, and remove the distinction between tiny/quantum/cacheline/subpage bins. Remove --enable-dynamic-page-shift. This option didn't prove useful in practice, and it prevented optimizations. Add Tilera architecture support.
2012-02-29 08:50:47 +08:00
<entry>[640, 768, 896, 1024]</entry>
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>256</entry>
Simplify small size class infrastructure. Program-generate small size class tables for all valid combinations of LG_TINY_MIN, LG_QUANTUM, and PAGE_SHIFT. Use the appropriate table to generate all relevant data structures, and remove the distinction between tiny/quantum/cacheline/subpage bins. Remove --enable-dynamic-page-shift. This option didn't prove useful in practice, and it prevented optimizations. Add Tilera architecture support.
2012-02-29 08:50:47 +08:00
<entry>[1280, 1536, 1792, 2048]</entry>
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>512</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
<entry>[2560, 3072, 3584, 4096]</entry>
</row>
<row>
<entry>1 KiB</entry>
<entry>[5 KiB, 6 KiB, 7 KiB, 8 KiB]</entry>
</row>
<row>
<entry>2 KiB</entry>
<entry>[10 KiB, 12 KiB, 14 KiB]</entry>
</row>
<row>
Update large/huge size class cutoff documentation.
2015-08-14 15:35:11 +08:00
<entry morerows="7">Large</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
<entry>2 KiB</entry>
<entry>[16 KiB]</entry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>4 KiB</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
<entry>[20 KiB, 24 KiB, 28 KiB, 32 KiB]</entry>
</row>
<row>
<entry>8 KiB</entry>
<entry>[40 KiB, 48 KiB, 54 KiB, 64 KiB]</entry>
</row>
<row>
<entry>16 KiB</entry>
<entry>[80 KiB, 96 KiB, 112 KiB, 128 KiB]</entry>
</row>
<row>
<entry>32 KiB</entry>
Update large/huge size class cutoff documentation.
2015-08-14 15:35:11 +08:00
<entry>[160 KiB, 192 KiB, 224 KiB, 256 KiB]</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
</row>
<row>
<entry>64 KiB</entry>
<entry>[320 KiB, 384 KiB, 448 KiB, 512 KiB]</entry>
</row>
<row>
<entry>128 KiB</entry>
Change default chunk size from 4 MiB to 256 KiB. Recent changes have improved huge allocation scalability, which removes upward pressure to set the chunk size so large that huge allocations are rare. Smaller chunks are more likely to completely drain, so set the default to the smallest size that doesn't leave excessive unusable trailing space in chunk headers.
2015-03-07 12:05:16 +08:00
<entry>[640 KiB, 768 KiB, 896 KiB, 1 MiB]</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
</row>
<row>
<entry>256 KiB</entry>
Update large/huge size class cutoff documentation.
2015-08-14 15:35:11 +08:00
<entry>[1280 KiB, 1536 KiB, 1792 KiB]</entry>
</row>
<row>
<entry morerows="6">Huge</entry>
<entry>256 KiB</entry>
<entry>[2 MiB]</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
</row>
<row>
<entry>512 KiB</entry>
Change default chunk size from 4 MiB to 256 KiB. Recent changes have improved huge allocation scalability, which removes upward pressure to set the chunk size so large that huge allocations are rare. Smaller chunks are more likely to completely drain, so set the default to the smallest size that doesn't leave excessive unusable trailing space in chunk headers.
2015-03-07 12:05:16 +08:00
<entry>[2560 KiB, 3 MiB, 3584 KiB, 4 MiB]</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
</row>
<row>
<entry>1 MiB</entry>
<entry>[5 MiB, 6 MiB, 7 MiB, 8 MiB]</entry>
</row>
<row>
<entry>2 MiB</entry>
<entry>[10 MiB, 12 MiB, 14 MiB, 16 MiB]</entry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</row>
<row>
Fix various documentation formatting regressions.
2012-03-20 00:36:44 +08:00
<entry>4 MiB</entry>
Update size class documentation.
2014-10-15 13:19:21 +08:00
<entry>[20 MiB, 24 MiB, 28 MiB, 32 MiB]</entry>
</row>
<row>
<entry>8 MiB</entry>
<entry>[40 MiB, 48 MiB, 56 MiB, 64 MiB]</entry>
</row>
<row>
<entry>...</entry>
<entry>...</entry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</row>
</tbody>
</tgroup>
</table>
</refsect1>
<refsect1 id="mallctl_namespace">
<title>MALLCTL NAMESPACE</title>
<para>The following names are defined in the namespace accessible via the
<function>mallctl*<parameter/></function> functions. Value types are
specified in parentheses, their readable/writable statuses are encoded as
<literal>rw</literal>, <literal>r-</literal>, <literal>-w</literal>, or
<literal>--</literal>, and required build configuration flags follow, if
any. A name element encoded as <literal>&lt;i&gt;</literal> or
<literal>&lt;j&gt;</literal> indicates an integer component, where the
integer varies from 0 to some upper value that must be determined via
introspection. In the case of <mallctl>stats.arenas.&lt;i&gt;.*</mallctl>,
<literal>&lt;i&gt;</literal> equal to <link
linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link> can be
used to access the summation of statistics from all arenas. Take special
note of the <link linkend="epoch"><mallctl>epoch</mallctl></link> mallctl,
which controls refreshing of cached dynamic statistics.</para>
<variablelist>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="version">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>version</mallctl>
(<type>const char *</type>)
<literal>r-</literal>
</term>
<listitem><para>Return the jemalloc version string.</para></listitem>
</varlistentry>
<varlistentry id="epoch">
<term>
<mallctl>epoch</mallctl>
(<type>uint64_t</type>)
<literal>rw</literal>
</term>
<listitem><para>If a value is passed in, refresh the data from which
the <function>mallctl*<parameter/></function> functions report values,
and increment the epoch. Return the current epoch. This is useful for
detecting whether another thread caused a refresh.</para></listitem>
</varlistentry>
Add the config.cache_oblivious mallctl.
2015-07-18 07:38:25 +08:00
<varlistentry id="config.cache_oblivious">
<term>
<mallctl>config.cache_oblivious</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-cache-oblivious</option> was specified
during build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.debug">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.debug</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-debug</option> was specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.fill">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.fill</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-fill</option> was specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.lazy_lock">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.lazy_lock</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-lazy-lock</option> was specified
during build configuration.</para></listitem>
</varlistentry>
Add --with-malloc-conf. Add --with-malloc-conf, which makes it possible to embed a default options string during configuration.
2016-02-08 06:23:22 +08:00
<varlistentry id="config.malloc_conf">
<term>
<mallctl>config.malloc_conf</mallctl>
(<type>const char *</type>)
<literal>r-</literal>
</term>
<listitem><para>Embedded configure-time-specified run-time options
string, empty unless <option>--with-malloc-conf</option> was specified
during build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.munmap">
Add the --disable-munmap option. Add the --disable-munmap option, remove the configure test that attempted to detect the VM allocation quirk known to exist on Linux x86[_64], and make --disable-munmap implicit on Linux.
2012-04-17 08:52:27 +08:00
<term>
<mallctl>config.munmap</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-munmap</option> was specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.prof">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.prof</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-prof</option> was specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.prof_libgcc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.prof_libgcc</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--disable-prof-libgcc</option> was not
specified during build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.prof_libunwind">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.prof_libunwind</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-prof-libunwind</option> was specified
during build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.stats">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.stats</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-stats</option> was specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.tcache">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.tcache</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--disable-tcache</option> was not specified
during build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.tls">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.tls</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--disable-tls</option> was not specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.utrace">
Add utrace(2)-based tracing (--enable-utrace).
2012-04-06 04:36:17 +08:00
<term>
<mallctl>config.utrace</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-utrace</option> was specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.valgrind">
Implement Valgrind support, redzones, and quarantine. Implement Valgrind support, as well as the redzone and quarantine features, which help Valgrind detect memory errors. Redzones are only implemented for small objects because the changes necessary to support redzones around large and huge objects are complicated by in-place reallocation, to the point that it isn't clear that the maintenance burden is worth the incremental improvement to Valgrind support. Merge arena_salloc() and arena_salloc_demote(). Refactor i[v]salloc() to expose the 'demote' option.
2012-04-06 15:35:09 +08:00
<term>
<mallctl>config.valgrind</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-valgrind</option> was specified during
build configuration.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="config.xmalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>config.xmalloc</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para><option>--enable-xmalloc</option> was specified during
build configuration.</para></listitem>
</varlistentry>
<varlistentry id="opt.abort">
<term>
<mallctl>opt.abort</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para>Abort-on-warning enabled/disabled. If true, most
warnings are fatal. The process will call
<citerefentry><refentrytitle>abort</refentrytitle>
<manvolnum>3</manvolnum></citerefentry> in these cases. This option is
disabled by default unless <option>--enable-debug</option> is
specified during configuration, in which case it is enabled by default.
</para></listitem>
</varlistentry>
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<varlistentry id="opt.dss">
<term>
<mallctl>opt.dss</mallctl>
(<type>const char *</type>)
<literal>r-</literal>
</term>
<listitem><para>dss (<citerefentry><refentrytitle>sbrk</refentrytitle>
<manvolnum>2</manvolnum></citerefentry>) allocation precedence as
related to <citerefentry><refentrytitle>mmap</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> allocation. The following
Make dss non-optional, and fix an "arena.<i>.dss" mallctl bug. Make dss non-optional on all platforms which support sbrk(2). Fix the "arena.<i>.dss" mallctl to return an error if "primary" or "secondary" precedence is specified, but sbrk(2) is not supported.
2014-04-16 03:09:48 +08:00
settings are supported if
<citerefentry><refentrytitle>sbrk</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> is supported by the operating
system: &ldquo;disabled&rdquo;, &ldquo;primary&rdquo;, and
&ldquo;secondary&rdquo;; otherwise only &ldquo;disabled&rdquo; is
supported. The default is &ldquo;secondary&rdquo; if
<citerefentry><refentrytitle>sbrk</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> is supported by the operating
system; &ldquo;disabled&rdquo; otherwise.
Fix a few mallctl() documentation errors. Normalize mallctl() order (code and documentation).
2013-12-20 13:40:41 +08:00
</para></listitem>
</varlistentry>
<varlistentry id="opt.lg_chunk">
<term>
<mallctl>opt.lg_chunk</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Virtual memory chunk size (log base 2). If a chunk
size outside the supported size range is specified, the size is
silently clipped to the minimum/maximum supported size. The default
Change default chunk size from 256 KiB to 2 MiB. This change improves interaction with transparent huge pages, e.g. reduced page faults (at least in the absence of unused dirty page purging).
2015-07-16 08:15:26 +08:00
chunk size is 2 MiB (2^21).
Fix a few mallctl() documentation errors. Normalize mallctl() order (code and documentation).
2013-12-20 13:40:41 +08:00
</para></listitem>
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="opt.narenas">
<term>
<mallctl>opt.narenas</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<listitem><para>Maximum number of arenas to use for automatic
multiplexing of threads and arenas. The default is four times the
number of CPUs, or one if there is a single CPU.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Implement decay-based unused dirty page purging. This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
2016-02-20 12:09:31 +08:00
<varlistentry id="opt.purge">
<term>
<mallctl>opt.purge</mallctl>
(<type>const char *</type>)
<literal>r-</literal>
</term>
<listitem><para>Purge mode is &ldquo;ratio&rdquo; (default) or
&ldquo;decay&rdquo;. See <link
linkend="opt.lg_dirty_mult"><mallctl>opt.lg_dirty_mult</mallctl></link>
for details of the ratio mode. See <link
linkend="opt.decay_time"><mallctl>opt.decay_time</mallctl></link> for
details of the decay mode.</para></listitem>
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="opt.lg_dirty_mult">
<term>
<mallctl>opt.lg_dirty_mult</mallctl>
(<type>ssize_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Per-arena minimum ratio (log base 2) of active to dirty
pages. Some dirty unused pages may be allowed to accumulate, within
the limit set by the ratio (or one chunk worth of dirty pages,
whichever is greater), before informing the kernel about some of those
pages via <citerefentry><refentrytitle>madvise</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> or a similar system call. This
provides the kernel with sufficient information to recycle dirty pages
if physical memory becomes scarce and the pages remain unused. The
Purge unused dirty pages in a fragmentation-reducing order. Purge unused dirty pages in an order that first performs clean/dirty run defragmentation, in order to mitigate available run fragmentation. Remove the limitation that prevented purging unless at least one chunk worth of dirty pages had accumulated in an arena. This limitation was intended to avoid excessive purging for small applications, but the threshold was arbitrary, and the effect of questionable utility. Relax opt_lg_dirty_mult from 5 to 3. This compensates for increased likelihood of allocating clean runs, given the same ratio of clean:dirty runs, and reduces the potential for repeated purging in pathological large malloc/free loops that push the active:dirty page ratio just over the purge threshold.
2012-10-31 06:42:37 +08:00
default minimum ratio is 8:1 (2^3:1); an option value of -1 will
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
disable dirty page purging. See <link
linkend="arenas.lg_dirty_mult"><mallctl>arenas.lg_dirty_mult</mallctl></link>
and <link
linkend="arena.i.lg_dirty_mult"><mallctl>arena.&lt;i&gt;.lg_dirty_mult</mallctl></link>
for related dynamic control options.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Implement decay-based unused dirty page purging. This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
2016-02-20 12:09:31 +08:00
<varlistentry id="opt.decay_time">
<term>
<mallctl>opt.decay_time</mallctl>
(<type>ssize_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Approximate time in seconds from the creation of a set
of unused dirty pages until an equivalent set of unused dirty pages is
purged and/or reused. The pages are incrementally purged according to a
sigmoidal decay curve that starts and ends with zero purge rate. A
decay time of 0 causes all unused dirty pages to be purged immediately
upon creation. A decay time of -1 disables purging. The default decay
time is 10 seconds. See <link
linkend="arenas.decay_time"><mallctl>arenas.decay_time</mallctl></link>
and <link
linkend="arena.i.decay_time"><mallctl>arena.&lt;i&gt;.decay_time</mallctl></link>
for related dynamic control options.
</para></listitem>
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="opt.stats_print">
<term>
<mallctl>opt.stats_print</mallctl>
(<type>bool</type>)
<literal>r-</literal>
</term>
<listitem><para>Enable/disable statistics printing at exit. If
enabled, the <function>malloc_stats_print<parameter/></function>
function is called at program exit via an
<citerefentry><refentrytitle>atexit</refentrytitle>
<manvolnum>3</manvolnum></citerefentry> function. If
<option>--enable-stats</option> is specified during configuration, this
has the potential to cause deadlock for a multi-threaded process that
exits while one or more threads are executing in the memory allocation
Avoid atexit(3) when possible, disable prof_final by default. atexit(3) can deadlock internally during its own initialization if jemalloc calls atexit() during jemalloc initialization. Mitigate the impact by restructuring prof initialization to avoid calling atexit() unless the registered function will actually dump a final heap profile. Additionally, disable prof_final by default so that this land mine is opt-in rather than opt-out. This resolves #144.
2014-10-09 08:57:19 +08:00
functions. Furthermore, <function>atexit<parameter/></function> may
allocate memory during application initialization and then deadlock
internally when jemalloc in turn calls
<function>atexit<parameter/></function>, so this option is not
univerally usable (though the application can register its own
<function>atexit<parameter/></function> function with equivalent
functionality). Therefore, this option should only be used with care;
it is primarily intended as a performance tuning aid during application
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
development. This option is disabled by default.</para></listitem>
</varlistentry>
<varlistentry id="opt.junk">
<term>
<mallctl>opt.junk</mallctl>
Introduce two new modes of junk filling: "alloc" and "free". In addition to true/false, opt.junk can now be either "alloc" or "free", giving applications the possibility of junking memory only on allocation or deallocation. This resolves #172.
2014-12-09 05:12:41 +08:00
(<type>const char *</type>)
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<literal>r-</literal>
[<option>--enable-fill</option>]
</term>
Introduce two new modes of junk filling: "alloc" and "free". In addition to true/false, opt.junk can now be either "alloc" or "free", giving applications the possibility of junking memory only on allocation or deallocation. This resolves #172.
2014-12-09 05:12:41 +08:00
<listitem><para>Junk filling. If set to "alloc", each byte of
uninitialized allocated memory will be initialized to
<literal>0xa5</literal>. If set to "free", all deallocated memory will
be initialized to <literal>0x5a</literal>. If set to "true", both
allocated and deallocated memory will be initialized, and if set to
"false", junk filling be disabled entirely. This is intended for
debugging and will impact performance negatively. This option is
"false" by default unless <option>--enable-debug</option> is specified
during configuration, in which case it is "true" by default unless
running inside <ulink
Auto-detect whether running inside Valgrind. Auto-detect whether running inside Valgrind, thus removing the need to manually specify MALLOC_CONF=valgrind:true.
2012-05-16 05:48:14 +08:00
url="http://valgrind.org/">Valgrind</ulink>.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Implement Valgrind support, redzones, and quarantine. Implement Valgrind support, as well as the redzone and quarantine features, which help Valgrind detect memory errors. Redzones are only implemented for small objects because the changes necessary to support redzones around large and huge objects are complicated by in-place reallocation, to the point that it isn't clear that the maintenance burden is worth the incremental improvement to Valgrind support. Merge arena_salloc() and arena_salloc_demote(). Refactor i[v]salloc() to expose the 'demote' option.
2012-04-06 15:35:09 +08:00
<varlistentry id="opt.quarantine">
<term>
<mallctl>opt.quarantine</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-fill</option>]
</term>
<listitem><para>Per thread quarantine size in bytes. If non-zero, each
thread maintains a FIFO object quarantine that stores up to the
specified number of bytes of memory. The quarantined memory is not
freed until it is released from quarantine, though it is immediately
junk-filled if the <link
linkend="opt.junk"><mallctl>opt.junk</mallctl></link> option is
enabled. This feature is of particular use in combination with <ulink
Fix Valgrind URL in documentation. Reported by Daichi GOTO.
2012-04-26 14:17:57 +08:00
url="http://valgrind.org/">Valgrind</ulink>, which can detect attempts
to access quarantined objects. This is intended for debugging and will
Auto-detect whether running inside Valgrind. Auto-detect whether running inside Valgrind, thus removing the need to manually specify MALLOC_CONF=valgrind:true.
2012-05-16 05:48:14 +08:00
impact performance negatively. The default quarantine size is 0 unless
running inside Valgrind, in which case the default is 16
MiB.</para></listitem>
Implement Valgrind support, redzones, and quarantine. Implement Valgrind support, as well as the redzone and quarantine features, which help Valgrind detect memory errors. Redzones are only implemented for small objects because the changes necessary to support redzones around large and huge objects are complicated by in-place reallocation, to the point that it isn't clear that the maintenance burden is worth the incremental improvement to Valgrind support. Merge arena_salloc() and arena_salloc_demote(). Refactor i[v]salloc() to expose the 'demote' option.
2012-04-06 15:35:09 +08:00
</varlistentry>
<varlistentry id="opt.redzone">
<term>
<mallctl>opt.redzone</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-fill</option>]
</term>
<listitem><para>Redzones enabled/disabled. If enabled, small
allocations have redzones before and after them. Furthermore, if the
<link linkend="opt.junk"><mallctl>opt.junk</mallctl></link> option is
enabled, the redzones are checked for corruption during deallocation.
However, the primary intended purpose of this feature is to be used in
Fix Valgrind URL in documentation. Reported by Daichi GOTO.
2012-04-26 14:17:57 +08:00
combination with <ulink url="http://valgrind.org/">Valgrind</ulink>,
which needs redzones in order to do effective buffer overflow/underflow
detection. This option is intended for debugging and will impact
performance negatively. This option is disabled by
Auto-detect whether running inside Valgrind. Auto-detect whether running inside Valgrind, thus removing the need to manually specify MALLOC_CONF=valgrind:true.
2012-05-16 05:48:14 +08:00
default unless running inside Valgrind.</para></listitem>
Implement Valgrind support, redzones, and quarantine. Implement Valgrind support, as well as the redzone and quarantine features, which help Valgrind detect memory errors. Redzones are only implemented for small objects because the changes necessary to support redzones around large and huge objects are complicated by in-place reallocation, to the point that it isn't clear that the maintenance burden is worth the incremental improvement to Valgrind support. Merge arena_salloc() and arena_salloc_demote(). Refactor i[v]salloc() to expose the 'demote' option.
2012-04-06 15:35:09 +08:00
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="opt.zero">
<term>
<mallctl>opt.zero</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-fill</option>]
</term>
<listitem><para>Zero filling enabled/disabled. If enabled, each byte
of uninitialized allocated memory will be initialized to 0. Note that
this initialization only happens once for each byte, so
Remove the *allocm() API, which is superceded by the *allocx() API.
2014-04-15 13:32:31 +08:00
<function>realloc<parameter/></function> and
<function>rallocx<parameter/></function> calls do not zero memory that
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
was previously allocated. This is intended for debugging and will
impact performance negatively. This option is disabled by default.
</para></listitem>
</varlistentry>
Add utrace(2)-based tracing (--enable-utrace).
2012-04-06 04:36:17 +08:00
<varlistentry id="opt.utrace">
<term>
<mallctl>opt.utrace</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-utrace</option>]
</term>
<listitem><para>Allocation tracing based on
<citerefentry><refentrytitle>utrace</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> enabled/disabled. This option
is disabled by default.</para></listitem>
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="opt.xmalloc">
<term>
<mallctl>opt.xmalloc</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-xmalloc</option>]
</term>
<listitem><para>Abort-on-out-of-memory enabled/disabled. If enabled,
rather than returning failure for any allocation function, display a
diagnostic message on <constant>STDERR_FILENO</constant> and cause the
program to drop core (using
<citerefentry><refentrytitle>abort</refentrytitle>
<manvolnum>3</manvolnum></citerefentry>). If an application is
designed to depend on this behavior, set the option at compile time by
including the following in the source code:
<programlisting language="C"><![CDATA[
malloc_conf = "xmalloc:true";]]></programlisting>
This option is disabled by default.</para></listitem>
</varlistentry>
<varlistentry id="opt.tcache">
<term>
<mallctl>opt.tcache</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-tcache</option>]
</term>
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<listitem><para>Thread-specific caching (tcache) enabled/disabled. When
there are multiple threads, each thread uses a tcache for objects up to
a certain size. Thread-specific caching allows many allocations to be
satisfied without performing any thread synchronization, at the cost of
increased memory use. See the <link
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
linkend="opt.lg_tcache_max"><mallctl>opt.lg_tcache_max</mallctl></link>
Remove the lg_tcache_gc_sweep option. Remove the lg_tcache_gc_sweep option, because it is no longer very useful. Prior to the addition of dynamic adjustment of tcache fill count, it was possible for fill/flush overhead to be a problem, but this problem no longer occurs.
2012-03-06 06:34:37 +08:00
option for related tuning information. This option is enabled by
Disable tcache by default if running inside Valgrind. Disable tcache by default if running inside Valgrind, in order to avoid making unallocated objects appear reachable to Valgrind.
2012-05-16 14:31:53 +08:00
default unless running inside <ulink
Optimize Valgrind integration. Forcefully disable tcache if running inside Valgrind, and remove Valgrind calls in tcache-specific code. Restructure Valgrind-related code to move most Valgrind calls out of the fast path functions. Take advantage of static knowledge to elide some branches in JEMALLOC_VALGRIND_REALLOC().
2014-04-16 07:35:08 +08:00
url="http://valgrind.org/">Valgrind</ulink>, in which case it is
forcefully disabled.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
<varlistentry id="opt.lg_tcache_max">
<term>
<mallctl>opt.lg_tcache_max</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-tcache</option>]
</term>
<listitem><para>Maximum size class (log base 2) to cache in the
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
thread-specific cache (tcache). At a minimum, all small size classes
are cached, and at a maximum all large size classes are cached. The
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
default maximum is 32 KiB (2^15).</para></listitem>
</varlistentry>
<varlistentry id="opt.prof">
<term>
<mallctl>opt.prof</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Memory profiling enabled/disabled. If enabled, profile
Update prof defaults to match common usage. Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB). Change the "opt.prof_accum" default from true to false. Add the "opt.prof_final" mallctl, so that "opt.prof_prefix" need not be abused to disable final profile dumping.
2012-04-18 07:39:33 +08:00
memory allocation activity. See the <link
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
linkend="opt.prof_active"><mallctl>opt.prof_active</mallctl></link>
option for on-the-fly activation/deactivation. See the <link
linkend="opt.lg_prof_sample"><mallctl>opt.lg_prof_sample</mallctl></link>
option for probabilistic sampling control. See the <link
linkend="opt.prof_accum"><mallctl>opt.prof_accum</mallctl></link>
option for control of cumulative sample reporting. See the <link
linkend="opt.lg_prof_interval"><mallctl>opt.lg_prof_interval</mallctl></link>
Update prof defaults to match common usage. Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB). Change the "opt.prof_accum" default from true to false. Add the "opt.prof_final" mallctl, so that "opt.prof_prefix" need not be abused to disable final profile dumping.
2012-04-18 07:39:33 +08:00
option for information on interval-triggered profile dumping, the <link
linkend="opt.prof_gdump"><mallctl>opt.prof_gdump</mallctl></link>
option for information on high-water-triggered profile dumping, and the
<link linkend="opt.prof_final"><mallctl>opt.prof_final</mallctl></link>
option for final profile dumping. Profile output is compatible with
Rename pprof to jeprof. This rename avoids installation collisions with the upstream gperftools. Additionally, jemalloc's per thread heap profile functionality introduced an incompatible file format, so it's now worthwhile to clearly distinguish jemalloc's version of this script from the upstream version. This resolves #229.
2015-05-02 03:31:12 +08:00
the <command>jeprof</command> command, which is based on the
<command>pprof</command> that is developed as part of the <ulink
url="http://code.google.com/p/gperftools/">gperftools
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
package</ulink>.</para></listitem>
</varlistentry>
<varlistentry id="opt.prof_prefix">
<term>
<mallctl>opt.prof_prefix</mallctl>
(<type>const char *</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Filename prefix for profile dumps. If the prefix is
set to the empty string, no automatic dumps will occur; this is
primarily useful for disabling the automatic final heap dump (which
also disables leak reporting, if enabled). The default prefix is
<filename>jeprof</filename>.</para></listitem>
</varlistentry>
<varlistentry id="opt.prof_active">
<term>
<mallctl>opt.prof_active</mallctl>
(<type>bool</type>)
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
<literal>r-</literal>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
[<option>--enable-prof</option>]
</term>
<listitem><para>Profiling activated/deactivated. This is a secondary
control mechanism that makes it possible to start the application with
profiling enabled (see the <link
linkend="opt.prof"><mallctl>opt.prof</mallctl></link> option) but
inactive, then toggle profiling at any time during program execution
with the <link
linkend="prof.active"><mallctl>prof.active</mallctl></link> mallctl.
This option is enabled by default.</para></listitem>
</varlistentry>
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
<varlistentry id="opt.prof_thread_active_init">
<term>
<mallctl>opt.prof_thread_active_init</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Initial setting for <link
linkend="thread.prof.active"><mallctl>thread.prof.active</mallctl></link>
in newly created threads. The initial setting for newly created threads
can also be changed during execution via the <link
linkend="prof.thread_active_init"><mallctl>prof.thread_active_init</mallctl></link>
mallctl. This option is enabled by default.</para></listitem>
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="opt.lg_prof_sample">
<term>
<mallctl>opt.lg_prof_sample</mallctl>
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
(<type>size_t</type>)
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Average interval (log base 2) between allocation
samples, as measured in bytes of allocation activity. Increasing the
sampling interval decreases profile fidelity, but also decreases the
Update prof defaults to match common usage. Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB). Change the "opt.prof_accum" default from true to false. Add the "opt.prof_final" mallctl, so that "opt.prof_prefix" need not be abused to disable final profile dumping.
2012-04-18 07:39:33 +08:00
computational overhead. The default sample interval is 512 KiB (2^19
B).</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
<varlistentry id="opt.prof_accum">
<term>
<mallctl>opt.prof_accum</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Reporting of cumulative object/byte counts in profile
dumps enabled/disabled. If this option is enabled, every unique
backtrace must be stored for the duration of execution. Depending on
the application, this can impose a large memory overhead, and the
Update prof defaults to match common usage. Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB). Change the "opt.prof_accum" default from true to false. Add the "opt.prof_final" mallctl, so that "opt.prof_prefix" need not be abused to disable final profile dumping.
2012-04-18 07:39:33 +08:00
cumulative counts are not always of interest. This option is disabled
Remove the opt.lg_prof_tcmax option. Remove the opt.lg_prof_tcmax option and hard-code a cache size of 1024. This setting is something that users just shouldn't have to worry about. If lock contention actually ends up being a problem, the simple solution available to the user is to reduce sampling frequency.
2012-02-14 10:04:26 +08:00
by default.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
<varlistentry id="opt.lg_prof_interval">
<term>
<mallctl>opt.lg_prof_interval</mallctl>
(<type>ssize_t</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Average interval (log base 2) between memory profile
dumps, as measured in bytes of allocation activity. The actual
interval between dumps may be sporadic because decentralized allocation
counters are used to avoid synchronization bottlenecks. Profiles are
dumped to files named according to the pattern
<filename>&lt;prefix&gt;.&lt;pid&gt;.&lt;seq&gt;.i&lt;iseq&gt;.heap</filename>,
where <literal>&lt;prefix&gt;</literal> is controlled by the
<link
linkend="opt.prof_prefix"><mallctl>opt.prof_prefix</mallctl></link>
option. By default, interval-triggered profile dumping is disabled
(encoded as -1).
</para></listitem>
</varlistentry>
<varlistentry id="opt.prof_gdump">
<term>
<mallctl>opt.prof_gdump</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
Implement the prof.gdump mallctl. This feature makes it possible to toggle the gdump feature on/off during program execution, whereas the the opt.prof_dump mallctl value can only be set during program startup. This resolves #72.
2015-01-26 13:16:57 +08:00
<listitem><para>Set the initial state of <link
linkend="prof.gdump"><mallctl>prof.gdump</mallctl></link>, which when
enabled triggers a memory profile dump every time the total virtual
memory exceeds the previous maximum. This option is disabled by
default.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Update prof defaults to match common usage. Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB). Change the "opt.prof_accum" default from true to false. Add the "opt.prof_final" mallctl, so that "opt.prof_prefix" need not be abused to disable final profile dumping.
2012-04-18 07:39:33 +08:00
<varlistentry id="opt.prof_final">
<term>
<mallctl>opt.prof_final</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Use an
<citerefentry><refentrytitle>atexit</refentrytitle>
<manvolnum>3</manvolnum></citerefentry> function to dump final memory
usage to a file named according to the pattern
<filename>&lt;prefix&gt;.&lt;pid&gt;.&lt;seq&gt;.f.heap</filename>,
where <literal>&lt;prefix&gt;</literal> is controlled by the <link
linkend="opt.prof_prefix"><mallctl>opt.prof_prefix</mallctl></link>
Avoid atexit(3) when possible, disable prof_final by default. atexit(3) can deadlock internally during its own initialization if jemalloc calls atexit() during jemalloc initialization. Mitigate the impact by restructuring prof initialization to avoid calling atexit() unless the registered function will actually dump a final heap profile. Additionally, disable prof_final by default so that this land mine is opt-in rather than opt-out. This resolves #144.
2014-10-09 08:57:19 +08:00
option. Note that <function>atexit<parameter/></function> may allocate
memory during application initialization and then deadlock internally
when jemalloc in turn calls <function>atexit<parameter/></function>, so
this option is not univerally usable (though the application can
register its own <function>atexit<parameter/></function> function with
equivalent functionality). This option is disabled by
default.</para></listitem>
Update prof defaults to match common usage. Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB). Change the "opt.prof_accum" default from true to false. Add the "opt.prof_final" mallctl, so that "opt.prof_prefix" need not be abused to disable final profile dumping.
2012-04-18 07:39:33 +08:00
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="opt.prof_leak">
<term>
<mallctl>opt.prof_leak</mallctl>
(<type>bool</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Leak reporting enabled/disabled. If enabled, use an
<citerefentry><refentrytitle>atexit</refentrytitle>
<manvolnum>3</manvolnum></citerefentry> function to report memory leaks
detected by allocation sampling. See the
<link linkend="opt.prof"><mallctl>opt.prof</mallctl></link> option for
information on analyzing heap profile output. This option is disabled
by default.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="thread.arena">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>thread.arena</mallctl>
(<type>unsigned</type>)
<literal>rw</literal>
</term>
<listitem><para>Get or set the arena associated with the calling
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
thread. If the specified arena was not initialized beforehand (see the
<link
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
linkend="arenas.initialized"><mallctl>arenas.initialized</mallctl></link>
mallctl), it will be automatically initialized as a side effect of
calling this interface.</para></listitem>
</varlistentry>
Add the "thread.[de]allocatedp" mallctl's.
2010-12-04 07:55:47 +08:00
<varlistentry id="thread.allocated">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>thread.allocated</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Get the total number of bytes ever allocated by the
calling thread. This counter has the potential to wrap around; it is
up to the application to appropriately interpret the counter in such
cases.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="thread.allocatedp">
Add the "thread.[de]allocatedp" mallctl's.
2010-12-04 07:55:47 +08:00
<term>
<mallctl>thread.allocatedp</mallctl>
(<type>uint64_t *</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Get a pointer to the the value that is returned by the
<link
linkend="thread.allocated"><mallctl>thread.allocated</mallctl></link>
mallctl. This is useful for avoiding the overhead of repeated
<function>mallctl*<parameter/></function> calls.</para></listitem>
</varlistentry>
<varlistentry id="thread.deallocated">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>thread.deallocated</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Get the total number of bytes ever deallocated by the
calling thread. This counter has the potential to wrap around; it is
up to the application to appropriately interpret the counter in such
cases.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="thread.deallocatedp">
Add the "thread.[de]allocatedp" mallctl's.
2010-12-04 07:55:47 +08:00
<term>
<mallctl>thread.deallocatedp</mallctl>
(<type>uint64_t *</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Get a pointer to the the value that is returned by the
<link
linkend="thread.deallocated"><mallctl>thread.deallocated</mallctl></link>
mallctl. This is useful for avoiding the overhead of repeated
<function>mallctl*<parameter/></function> calls.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="thread.tcache.enabled">
Add the "thread.tcache.enabled" mallctl.
2012-03-27 09:54:44 +08:00
<term>
<mallctl>thread.tcache.enabled</mallctl>
(<type>bool</type>)
<literal>rw</literal>
[<option>--enable-tcache</option>]
</term>
<listitem><para>Enable/disable calling thread's tcache. The tcache is
implicitly flushed as a side effect of becoming
disabled (see <link
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
linkend="thread.tcache.flush"><mallctl>thread.tcache.flush</mallctl></link>).
Add the "thread.tcache.enabled" mallctl.
2012-03-27 09:54:44 +08:00
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="thread.tcache.flush">
Rename the "tcache.flush" mallctl to "thread.tcache.flush".
2012-03-17 08:09:32 +08:00
<term>
<mallctl>thread.tcache.flush</mallctl>
(<type>void</type>)
<literal>--</literal>
[<option>--enable-tcache</option>]
</term>
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<listitem><para>Flush calling thread's thread-specific cache (tcache).
This interface releases all cached objects and internal data structures
associated with the calling thread's tcache. Ordinarily, this interface
Rename the "tcache.flush" mallctl to "thread.tcache.flush".
2012-03-17 08:09:32 +08:00
need not be called, since automatic periodic incremental garbage
collection occurs, and the thread cache is automatically discarded when
a thread exits. However, garbage collection is triggered by allocation
activity, so it is possible for a thread that stops
allocating/deallocating to retain its cache indefinitely, in which case
the developer may find manual flushing useful.</para></listitem>
</varlistentry>
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
<varlistentry id="thread.prof.name">
<term>
<mallctl>thread.prof.name</mallctl>
(<type>const char *</type>)
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
<literal>r-</literal> or
<literal>-w</literal>
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
[<option>--enable-prof</option>]
</term>
<listitem><para>Get/set the descriptive name associated with the calling
thread in memory profile dumps. An internal copy of the name string is
created, so the input string need not be maintained after this interface
completes execution. The output string of this interface should be
copied for non-ephemeral uses, because multiple implementation details
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
can cause asynchronous string deallocation. Furthermore, each
invocation of this interface can only read or write; simultaneous
read/write is not supported due to string lifetime limitations. The
Fix a manual editing error.
2015-10-20 04:56:05 +08:00
name string must be nil-terminated and comprised only of characters in
the sets recognized
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
by <citerefentry><refentrytitle>isgraph</refentrytitle>
<manvolnum>3</manvolnum></citerefentry> and
<citerefentry><refentrytitle>isblank</refentrytitle>
<manvolnum>3</manvolnum></citerefentry>.</para></listitem>
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
</varlistentry>
<varlistentry id="thread.prof.active">
<term>
<mallctl>thread.prof.active</mallctl>
(<type>bool</type>)
<literal>rw</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Control whether sampling is currently active for the
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
calling thread. This is an activation mechanism in addition to <link
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
linkend="prof.active"><mallctl>prof.active</mallctl></link>; both must
be active for the calling thread to sample. This flag is enabled by
default.</para></listitem>
</varlistentry>
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
<varlistentry id="tcache.create">
<term>
<mallctl>tcache.create</mallctl>
(<type>unsigned</type>)
<literal>r-</literal>
[<option>--enable-tcache</option>]
</term>
<listitem><para>Create an explicit thread-specific cache (tcache) and
return an identifier that can be passed to the <link
linkend="MALLOCX_TCACHE"><constant>MALLOCX_TCACHE(<parameter>tc</parameter>)</constant></link>
macro to explicitly use the specified cache rather than the
automatically managed one that is used by default. Each explicit cache
can be used by only one thread at a time; the application must assure
that this constraint holds.
</para></listitem>
</varlistentry>
<varlistentry id="tcache.flush">
<term>
<mallctl>tcache.flush</mallctl>
(<type>unsigned</type>)
<literal>-w</literal>
[<option>--enable-tcache</option>]
</term>
<listitem><para>Flush the specified thread-specific cache (tcache). The
same considerations apply to this interface as to <link
linkend="thread.tcache.flush"><mallctl>thread.tcache.flush</mallctl></link>,
Fix a documentation editing error.
2016-02-20 04:08:14 +08:00
except that the tcache will never be automatically discarded.
Implement explicit tcache support. Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be used in conjunction with the *allocx() API. Add the tcache.create, tcache.flush, and tcache.destroy mallctls. This resolves #145.
2015-01-30 07:30:47 +08:00
</para></listitem>
</varlistentry>
<varlistentry id="tcache.destroy">
<term>
<mallctl>tcache.destroy</mallctl>
(<type>unsigned</type>)
<literal>-w</literal>
[<option>--enable-tcache</option>]
</term>
<listitem><para>Flush the specified thread-specific cache (tcache) and
make the identifier available for use during a future tcache creation.
</para></listitem>
</varlistentry>
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<varlistentry id="arena.i.purge">
<term>
<mallctl>arena.&lt;i&gt;.purge</mallctl>
Remove the "arenas.purge" mallctl. Remove the "arenas.purge" mallctl, which was obsoleted by the "arena.<i>.purge" mallctl in 3.1.0.
2014-04-16 03:46:28 +08:00
(<type>void</type>)
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<literal>--</literal>
</term>
Implement decay-based unused dirty page purging. This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
2016-02-20 12:09:31 +08:00
<listitem><para>Purge all unused dirty pages for arena &lt;i&gt;, or for
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
all arenas if &lt;i&gt; equals <link
linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link>.
</para></listitem>
</varlistentry>
Implement decay-based unused dirty page purging. This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
2016-02-20 12:09:31 +08:00
<varlistentry id="arena.i.decay">
<term>
<mallctl>arena.&lt;i&gt;.decay</mallctl>
(<type>void</type>)
<literal>--</literal>
</term>
<listitem><para>Trigger decay-based purging of unused dirty pages for
arena &lt;i&gt;, or for all arenas if &lt;i&gt; equals <link
linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link>.
The proportion of unused dirty pages to be purged depends on the current
time; see <link
linkend="opt.decay_time"><mallctl>opt.decay_time</mallctl></link> for
details.</para></listitem>
</varlistentry>
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<varlistentry id="arena.i.dss">
<term>
<mallctl>arena.&lt;i&gt;.dss</mallctl>
(<type>const char *</type>)
<literal>rw</literal>
</term>
<listitem><para>Set the precedence of dss allocation as related to mmap
allocation for arena &lt;i&gt;, or for all arenas if &lt;i&gt; equals
<link
Refactor huge allocation to be managed by arenas. Refactor huge allocation to be managed by arenas (though the global red-black tree of huge allocations remains for lookup during deallocation). This is the logical conclusion of recent changes that 1) made per arena dss precedence apply to huge allocation, and 2) made it possible to replace the per arena chunk allocation/deallocation functions. Remove the top level huge stats, and replace them with per arena huge stats. Normalize function names and types to *dalloc* (some were *dealloc*). Remove the --enable-mremap option. As jemalloc currently operates, this is a performace regression for some applications, but planned work to logarithmically space huge size classes should provide similar amortized performance. The motivation for this change was that mremap-based huge reallocation forced leaky abstractions that prevented refactoring.
2014-05-16 13:22:27 +08:00
linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link>. See
<link linkend="opt.dss"><mallctl>opt.dss</mallctl></link> for supported
settings.</para></listitem>
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
</varlistentry>
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
<varlistentry id="arena.i.lg_dirty_mult">
<term>
<mallctl>arena.&lt;i&gt;.lg_dirty_mult</mallctl>
(<type>ssize_t</type>)
<literal>rw</literal>
</term>
<listitem><para>Current per-arena minimum ratio (log base 2) of active
to dirty pages for arena &lt;i&gt;. Each time this interface is set and
the ratio is increased, pages are synchronously purged as necessary to
impose the new ratio. See <link
linkend="opt.lg_dirty_mult"><mallctl>opt.lg_dirty_mult</mallctl></link>
for additional information.</para></listitem>
</varlistentry>
Implement decay-based unused dirty page purging. This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
2016-02-20 12:09:31 +08:00
<varlistentry id="arena.i.decay_time">
<term>
<mallctl>arena.&lt;i&gt;.decay_time</mallctl>
(<type>ssize_t</type>)
<literal>rw</literal>
</term>
<listitem><para>Current per-arena approximate time in seconds from the
creation of a set of unused dirty pages until an equivalent set of
unused dirty pages is purged and/or reused. Each time this interface is
set, all currently unused dirty pages are considered to have fully
decayed, which causes immediate purging of all unused dirty pages unless
the decay time is set to -1 (i.e. purging disabled). See <link
linkend="opt.decay_time"><mallctl>opt.decay_time</mallctl></link> for
additional information.</para></listitem>
</varlistentry>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<varlistentry id="arena.i.chunk_hooks">
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<term>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<mallctl>arena.&lt;i&gt;.chunk_hooks</mallctl>
(<type>chunk_hooks_t</type>)
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<literal>rw</literal>
</term>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<listitem><para>Get or set the chunk management hook functions for arena
&lt;i&gt;. The functions must be capable of operating on all extant
chunks associated with arena &lt;i&gt;, usually by passing unknown
chunks to the replaced functions. In practice, it is feasible to
control allocation for arenas created via <link
linkend="arenas.extend"><mallctl>arenas.extend</mallctl></link> such
that all chunks originate from an application-supplied chunk allocator
(by setting custom chunk hook functions just after arena creation), but
the automatically created arenas may have already created chunks prior
to the application having an opportunity to take over chunk
allocation.</para>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<programlisting language="C"><![CDATA[
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
typedef struct {
chunk_alloc_t *alloc;
chunk_dalloc_t *dalloc;
chunk_commit_t *commit;
chunk_decommit_t *decommit;
chunk_purge_t *purge;
chunk_split_t *split;
chunk_merge_t *merge;
} chunk_hooks_t;]]></programlisting>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<para>The <type>chunk_hooks_t</type> structure comprises function
pointers which are described individually below. jemalloc uses these
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
functions to manage chunk lifetime, which starts off with allocation of
mapped committed memory, in the simplest case followed by deallocation.
However, there are performance and platform reasons to retain chunks for
later reuse. Cleanup attempts cascade from deallocation to decommit to
purging, which gives the chunk management functions opportunities to
reject the most permanent cleanup operations in favor of less permanent
(and often less costly) operations. The chunk splitting and merging
operations can also be opted out of, but this is mainly intended to
support platforms on which virtual memory mappings provided by the
Update in-place reallocation documentation.
2015-08-14 15:55:44 +08:00
operating system kernel do not automatically coalesce and split, e.g.
Windows.</para>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<funcsynopsis><funcprototype>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<funcdef>typedef void *<function>(chunk_alloc_t)</function></funcdef>
Attempt to expand huge allocations in-place. This adds support for expanding huge allocations in-place by requesting memory at a specific address from the chunk allocator. It's currently only implemented for the chunk recycling path, although in theory it could also be done by optimistically allocating new chunks. On Linux, it could attempt an in-place mremap. However, that won't work in practice since the heap is grown downwards and memory is not unmapped (in a normal build, at least). Repeated vector reallocation micro-benchmark: #include <string.h> #include <stdlib.h> int main(void) { for (size_t i = 0; i < 100; i++) { void *ptr = NULL; size_t old_size = 0; for (size_t size = 4; size < (1 << 30); size *= 2) { ptr = realloc(ptr, size); if (!ptr) return 1; memset(ptr + old_size, 0xff, size - old_size); old_size = size; } free(ptr); } } The glibc allocator fails to do any in-place reallocations on this benchmark once it passes the M_MMAP_THRESHOLD (default 128k) but it elides the cost of copies via mremap, which is currently not something that jemalloc can use. With this improvement, jemalloc still fails to do any in-place huge reallocations for the first outer loop, but then succeeds 100% of the time for the remaining 99 iterations. The time spent doing allocations and copies drops down to under 5%, with nearly all of it spent doing purging + faulting (when huge pages are disabled) and the array memset. An improved mremap API (MREMAP_RETAIN - #138) would be far more general but this is a portable optimization and would still be useful on Linux for xallocx. Numbers with transparent huge pages enabled: glibc (copies elided via MREMAP_MAYMOVE): 8.471s jemalloc: 17.816s jemalloc + no-op madvise: 13.236s jemalloc + this commit: 6.787s jemalloc + this commit + no-op madvise: 6.144s Numbers with transparent huge pages disabled: glibc (copies elided via MREMAP_MAYMOVE): 15.403s jemalloc: 39.456s jemalloc + no-op madvise: 12.768s jemalloc + this commit: 15.534s jemalloc + this commit + no-op madvise: 6.354s Closes #137
2014-10-04 13:39:32 +08:00
<paramdef>void *<parameter>chunk</parameter></paramdef>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<paramdef>size_t <parameter>size</parameter></paramdef>
<paramdef>size_t <parameter>alignment</parameter></paramdef>
<paramdef>bool *<parameter>zero</parameter></paramdef>
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<paramdef>bool *<parameter>commit</parameter></paramdef>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<paramdef>unsigned <parameter>arena_ind</parameter></paramdef>
Fix a docbook element nesting nit. According to the docbook documentation for <funcprototype>, its parent must be <funcsynopsis>; fix accordingly. Nonetheless, the man page processor fails badly when this construct is embedded in a <para> (which is documented to be legal), although the html processor does fine.
2014-10-06 05:48:44 +08:00
</funcprototype></funcsynopsis>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<literallayout></literallayout>
<para>A chunk allocation function conforms to the
<type>chunk_alloc_t</type> type and upon success returns a pointer to
<parameter>size</parameter> bytes of mapped memory on behalf of arena
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<parameter>arena_ind</parameter> such that the chunk's base address is a
multiple of <parameter>alignment</parameter>, as well as setting
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<parameter>*zero</parameter> to indicate whether the chunk is zeroed and
<parameter>*commit</parameter> to indicate whether the chunk is
committed. Upon error the function returns <constant>NULL</constant>
and leaves <parameter>*zero</parameter> and
<parameter>*commit</parameter> unmodified. The
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<parameter>size</parameter> parameter is always a multiple of the chunk
size. The <parameter>alignment</parameter> parameter is always a power
of two at least as large as the chunk size. Zeroing is mandatory if
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<parameter>*zero</parameter> is true upon function entry. Committing is
mandatory if <parameter>*commit</parameter> is true upon function entry.
If <parameter>chunk</parameter> is not <constant>NULL</constant>, the
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
returned pointer must be <parameter>chunk</parameter> on success or
<constant>NULL</constant> on error. Committed memory may be committed
in absolute terms as on a system that does not overcommit, or in
implicit terms as on a system that overcommits and satisfies physical
memory needs on demand via soft page faults. Note that replacing the
default chunk allocation function makes the arena's <link
Refactor huge allocation to be managed by arenas. Refactor huge allocation to be managed by arenas (though the global red-black tree of huge allocations remains for lookup during deallocation). This is the logical conclusion of recent changes that 1) made per arena dss precedence apply to huge allocation, and 2) made it possible to replace the per arena chunk allocation/deallocation functions. Remove the top level huge stats, and replace them with per arena huge stats. Normalize function names and types to *dalloc* (some were *dealloc*). Remove the --enable-mremap option. As jemalloc currently operates, this is a performace regression for some applications, but planned work to logarithmically space huge size classes should provide similar amortized performance. The motivation for this change was that mremap-based huge reallocation forced leaky abstractions that prevented refactoring.
2014-05-16 13:22:27 +08:00
linkend="arena.i.dss"><mallctl>arena.&lt;i&gt;.dss</mallctl></link>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
setting irrelevant.</para>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<funcsynopsis><funcprototype>
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
<funcdef>typedef bool <function>(chunk_dalloc_t)</function></funcdef>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<paramdef>void *<parameter>chunk</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<paramdef>bool <parameter>committed</parameter></paramdef>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
<paramdef>unsigned <parameter>arena_ind</parameter></paramdef>
Fix a docbook element nesting nit. According to the docbook documentation for <funcprototype>, its parent must be <funcsynopsis>; fix accordingly. Nonetheless, the man page processor fails badly when this construct is embedded in a <para> (which is documented to be legal), although the html processor does fine.
2014-10-06 05:48:44 +08:00
</funcprototype></funcsynopsis>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<literallayout></literallayout>
<para>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
A chunk deallocation function conforms to the
Refactor huge allocation to be managed by arenas. Refactor huge allocation to be managed by arenas (though the global red-black tree of huge allocations remains for lookup during deallocation). This is the logical conclusion of recent changes that 1) made per arena dss precedence apply to huge allocation, and 2) made it possible to replace the per arena chunk allocation/deallocation functions. Remove the top level huge stats, and replace them with per arena huge stats. Normalize function names and types to *dalloc* (some were *dealloc*). Remove the --enable-mremap option. As jemalloc currently operates, this is a performace regression for some applications, but planned work to logarithmically space huge size classes should provide similar amortized performance. The motivation for this change was that mremap-based huge reallocation forced leaky abstractions that prevented refactoring.
2014-05-16 13:22:27 +08:00
<type>chunk_dalloc_t</type> type and deallocates a
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<parameter>chunk</parameter> of given <parameter>size</parameter> with
<parameter>committed</parameter>/decommited memory as indicated, on
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
behalf of arena <parameter>arena_ind</parameter>, returning false upon
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
success. If the function returns true, this indicates opt-out from
deallocation; the virtual memory mapping associated with the chunk
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
remains mapped, in the same commit state, and available for future use,
in which case it will be automatically retained for later reuse.</para>
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<funcsynopsis><funcprototype>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<funcdef>typedef bool <function>(chunk_commit_t)</function></funcdef>
<paramdef>void *<parameter>chunk</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<paramdef>size_t <parameter>offset</parameter></paramdef>
<paramdef>size_t <parameter>length</parameter></paramdef>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<paramdef>unsigned <parameter>arena_ind</parameter></paramdef>
</funcprototype></funcsynopsis>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<literallayout></literallayout>
<para>A chunk commit function conforms to the
<type>chunk_commit_t</type> type and commits zeroed physical memory to
back pages within a <parameter>chunk</parameter> of given
<parameter>size</parameter> at <parameter>offset</parameter> bytes,
extending for <parameter>length</parameter> on behalf of arena
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<parameter>arena_ind</parameter>, returning false upon success.
Committed memory may be committed in absolute terms as on a system that
does not overcommit, or in implicit terms as on a system that
overcommits and satisfies physical memory needs on demand via soft page
faults. If the function returns true, this indicates insufficient
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
physical memory to satisfy the request.</para>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<funcsynopsis><funcprototype>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<funcdef>typedef bool <function>(chunk_decommit_t)</function></funcdef>
<paramdef>void *<parameter>chunk</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<paramdef>size_t <parameter>offset</parameter></paramdef>
<paramdef>size_t <parameter>length</parameter></paramdef>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<paramdef>unsigned <parameter>arena_ind</parameter></paramdef>
</funcprototype></funcsynopsis>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<literallayout></literallayout>
<para>A chunk decommit function conforms to the
<type>chunk_decommit_t</type> type and decommits any physical memory
that is backing pages within a <parameter>chunk</parameter> of given
<parameter>size</parameter> at <parameter>offset</parameter> bytes,
extending for <parameter>length</parameter> on behalf of arena
Implement chunk hook support for page run commit/decommit. Cascade from decommit to purge when purging unused dirty pages, so that it is possible to decommit cleaned memory rather than just purging. For non-Windows debug builds, decommit runs rather than purging them, since this causes access of deallocated runs to segfault. This resolves #251.
2015-08-05 01:49:46 +08:00
<parameter>arena_ind</parameter>, returning false upon success, in which
case the pages will be committed via the chunk commit function before
being reused. If the function returns true, this indicates opt-out from
decommit; the memory remains committed and available for future use, in
which case it will be automatically retained for later reuse.</para>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<funcsynopsis><funcprototype>
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
<funcdef>typedef bool <function>(chunk_purge_t)</function></funcdef>
<paramdef>void *<parameter>chunk</parameter></paramdef>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<paramdef>size_t<parameter>size</parameter></paramdef>
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
<paramdef>size_t <parameter>offset</parameter></paramdef>
<paramdef>size_t <parameter>length</parameter></paramdef>
<paramdef>unsigned <parameter>arena_ind</parameter></paramdef>
</funcprototype></funcsynopsis>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<literallayout></literallayout>
<para>A chunk purge function conforms to the <type>chunk_purge_t</type>
type and optionally discards physical pages within the virtual memory
mapping associated with <parameter>chunk</parameter> of given
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<parameter>size</parameter> at <parameter>offset</parameter> bytes,
extending for <parameter>length</parameter> on behalf of arena
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
<parameter>arena_ind</parameter>, returning false if pages within the
purged virtual memory range will be zero-filled the next time they are
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
accessed.</para>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<funcsynopsis><funcprototype>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<funcdef>typedef bool <function>(chunk_split_t)</function></funcdef>
<paramdef>void *<parameter>chunk</parameter></paramdef>
<paramdef>size_t <parameter>size</parameter></paramdef>
<paramdef>size_t <parameter>size_a</parameter></paramdef>
<paramdef>size_t <parameter>size_b</parameter></paramdef>
<paramdef>bool <parameter>committed</parameter></paramdef>
<paramdef>unsigned <parameter>arena_ind</parameter></paramdef>
</funcprototype></funcsynopsis>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<literallayout></literallayout>
<para>A chunk split function conforms to the <type>chunk_split_t</type>
type and optionally splits <parameter>chunk</parameter> of given
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<parameter>size</parameter> into two adjacent chunks, the first of
<parameter>size_a</parameter> bytes, and the second of
<parameter>size_b</parameter> bytes, operating on
<parameter>committed</parameter>/decommitted memory as indicated, on
behalf of arena <parameter>arena_ind</parameter>, returning false upon
success. If the function returns true, this indicates that the chunk
remains unsplit and therefore should continue to be operated on as a
whole.</para>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<funcsynopsis><funcprototype>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
<funcdef>typedef bool <function>(chunk_merge_t)</function></funcdef>
<paramdef>void *<parameter>chunk_a</parameter></paramdef>
<paramdef>size_t <parameter>size_a</parameter></paramdef>
<paramdef>void *<parameter>chunk_b</parameter></paramdef>
<paramdef>size_t <parameter>size_b</parameter></paramdef>
<paramdef>bool <parameter>committed</parameter></paramdef>
<paramdef>unsigned <parameter>arena_ind</parameter></paramdef>
</funcprototype></funcsynopsis>
Improve arena.<i>.chunk_hooks documentation formatting.
2015-08-15 04:46:08 +08:00
<literallayout></literallayout>
<para>A chunk merge function conforms to the <type>chunk_merge_t</type>
type and optionally merges adjacent chunks,
<parameter>chunk_a</parameter> of given <parameter>size_a</parameter>
and <parameter>chunk_b</parameter> of given
<parameter>size_b</parameter> into one contiguous chunk, operating on
<parameter>committed</parameter>/decommitted memory as indicated, on
behalf of arena <parameter>arena_ind</parameter>, returning false upon
success. If the function returns true, this indicates that the chunks
remain distinct mappings and therefore should continue to be operated on
independently.</para>
Generalize chunk management hooks. Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls. The chunk hooks allow control over chunk allocation/deallocation, decommit/commit, purging, and splitting/merging, such that the application can rely on jemalloc's internal chunk caching and retaining functionality, yet implement a variety of chunk management mechanisms and policies. Merge the chunks_[sz]ad_{mmap,dss} red-black trees into chunks_[sz]ad_retained. This slightly reduces how hard jemalloc tries to honor the dss precedence setting; prior to this change the precedence setting was also consulted when recycling chunks. Fix chunk purging. Don't purge chunks in arena_purge_stashed(); instead deallocate them in arena_unstash_purged(), so that the dirty memory linkage remains valid until after the last time it is used. This resolves #176 and #201.
2015-07-28 23:28:19 +08:00
</listitem>
Add support for user-specified chunk allocators/deallocators. Add new mallctl endpoints "arena<i>.chunk.alloc" and "arena<i>.chunk.dealloc" to allow userspace to configure jemalloc's chunk allocator and deallocator on a per-arena basis.
2014-05-06 06:16:56 +08:00
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="arenas.narenas">
<term>
<mallctl>arenas.narenas</mallctl>
(<type>unsigned</type>)
<literal>r-</literal>
</term>
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<listitem><para>Current limit on number of arenas.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
<varlistentry id="arenas.initialized">
<term>
<mallctl>arenas.initialized</mallctl>
(<type>bool *</type>)
<literal>r-</literal>
</term>
<listitem><para>An array of <link
linkend="arenas.narenas"><mallctl>arenas.narenas</mallctl></link>
booleans. Each boolean indicates whether the corresponding arena is
initialized.</para></listitem>
</varlistentry>
Implement dynamic per arena control over dirty page purging. Add mallctls: - arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be modified to change the initial lg_dirty_mult setting for newly created arenas. - arena.<i>.lg_dirty_mult controls an individual arena's dirty page purging threshold, and synchronously triggers any purging that may be necessary to maintain the constraint. - arena.<i>.chunk.purge allows the per arena dirty page purging function to be replaced. This resolves #93.
2015-03-19 09:55:33 +08:00
<varlistentry id="arenas.lg_dirty_mult">
<term>
<mallctl>arenas.lg_dirty_mult</mallctl>
(<type>ssize_t</type>)
<literal>rw</literal>
</term>
<listitem><para>Current default per-arena minimum ratio (log base 2) of
active to dirty pages, used to initialize <link
linkend="arena.i.lg_dirty_mult"><mallctl>arena.&lt;i&gt;.lg_dirty_mult</mallctl></link>
during arena creation. See <link
linkend="opt.lg_dirty_mult"><mallctl>opt.lg_dirty_mult</mallctl></link>
for additional information.</para></listitem>
</varlistentry>
Implement decay-based unused dirty page purging. This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
2016-02-20 12:09:31 +08:00
<varlistentry id="arenas.decay_time">
<term>
<mallctl>arenas.decay_time</mallctl>
(<type>ssize_t</type>)
<literal>rw</literal>
</term>
<listitem><para>Current default per-arena approximate time in seconds
from the creation of a set of unused dirty pages until an equivalent set
of unused dirty pages is purged and/or reused, used to initialize <link
linkend="arena.i.decay_time"><mallctl>arena.&lt;i&gt;.decay_time</mallctl></link>
during arena creation. See <link
linkend="opt.decay_time"><mallctl>opt.decay_time</mallctl></link> for
additional information.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.quantum">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.quantum</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Quantum size.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.page">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
Clean up *PAGE* macros. s/PAGE_SHIFT/LG_PAGE/g and s/PAGE_SIZE/PAGE/g. Remove remnants of the dynamic-page-shift code. Rename the "arenas.pagesize" mallctl to "arenas.page". Remove the "arenas.chunksize" mallctl, which is redundant with "opt.lg_chunk".
2012-04-02 22:04:34 +08:00
<mallctl>arenas.page</mallctl>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Page size.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.tcache_max">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.tcache_max</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-tcache</option>]
</term>
<listitem><para>Maximum thread-cached size class.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.nbins">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.nbins</mallctl>
(<type>unsigned</type>)
<literal>r-</literal>
</term>
Simplify small size class infrastructure. Program-generate small size class tables for all valid combinations of LG_TINY_MIN, LG_QUANTUM, and PAGE_SHIFT. Use the appropriate table to generate all relevant data structures, and remove the distinction between tiny/quantum/cacheline/subpage bins. Remove --enable-dynamic-page-shift. This option didn't prove useful in practice, and it prevented optimizations. Add Tilera architecture support.
2012-02-29 08:50:47 +08:00
<listitem><para>Number of bin size classes.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.nhbins">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.nhbins</mallctl>
(<type>unsigned</type>)
<literal>r-</literal>
[<option>--enable-tcache</option>]
</term>
<listitem><para>Total number of thread cache bin size
classes.</para></listitem>
</varlistentry>
<varlistentry id="arenas.bin.i.size">
<term>
<mallctl>arenas.bin.&lt;i&gt;.size</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Maximum size supported by size class.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.bin.i.nregs">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.bin.&lt;i&gt;.nregs</mallctl>
(<type>uint32_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Number of regions per page run.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.bin.i.run_size">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.bin.&lt;i&gt;.run_size</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Number of bytes per page run.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.nlruns">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.nlruns</mallctl>
Add per size class huge allocation statistics. Add per size class huge allocation statistics, and normalize various stats: - Change the arenas.nlruns type from size_t to unsigned. - Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's. - Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with stats.arenas.<i>.bins.<j>.curregs . - Add the stats.arenas.<i>.hchunks.<j>.nmalloc, stats.arenas.<i>.hchunks.<j>.ndalloc, stats.arenas.<i>.hchunks.<j>.nrequests, and stats.arenas.<i>.hchunks.<j>.curhchunks mallctl's.
2014-10-13 13:53:59 +08:00
(<type>unsigned</type>)
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<literal>r-</literal>
</term>
<listitem><para>Total number of large size classes.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.lrun.i.size">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>arenas.lrun.&lt;i&gt;.size</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Maximum size supported by this large size
class.</para></listitem>
</varlistentry>
Add per size class huge allocation statistics. Add per size class huge allocation statistics, and normalize various stats: - Change the arenas.nlruns type from size_t to unsigned. - Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's. - Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with stats.arenas.<i>.bins.<j>.curregs . - Add the stats.arenas.<i>.hchunks.<j>.nmalloc, stats.arenas.<i>.hchunks.<j>.ndalloc, stats.arenas.<i>.hchunks.<j>.nrequests, and stats.arenas.<i>.hchunks.<j>.curhchunks mallctl's.
2014-10-13 13:53:59 +08:00
<varlistentry id="arenas.nhchunks">
<term>
<mallctl>arenas.nhchunks</mallctl>
(<type>unsigned</type>)
<literal>r-</literal>
</term>
<listitem><para>Total number of huge size classes.</para></listitem>
</varlistentry>
Fix mallctl doc: arenas.hchunk.<i>.size
2015-04-22 07:57:42 +08:00
<varlistentry id="arenas.hchunk.i.size">
Add per size class huge allocation statistics. Add per size class huge allocation statistics, and normalize various stats: - Change the arenas.nlruns type from size_t to unsigned. - Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's. - Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with stats.arenas.<i>.bins.<j>.curregs . - Add the stats.arenas.<i>.hchunks.<j>.nmalloc, stats.arenas.<i>.hchunks.<j>.ndalloc, stats.arenas.<i>.hchunks.<j>.nrequests, and stats.arenas.<i>.hchunks.<j>.curhchunks mallctl's.
2014-10-13 13:53:59 +08:00
<term>
Fix mallctl doc: arenas.hchunk.<i>.size
2015-04-22 07:57:42 +08:00
<mallctl>arenas.hchunk.&lt;i&gt;.size</mallctl>
Add per size class huge allocation statistics. Add per size class huge allocation statistics, and normalize various stats: - Change the arenas.nlruns type from size_t to unsigned. - Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's. - Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with stats.arenas.<i>.bins.<j>.curregs . - Add the stats.arenas.<i>.hchunks.<j>.nmalloc, stats.arenas.<i>.hchunks.<j>.ndalloc, stats.arenas.<i>.hchunks.<j>.nrequests, and stats.arenas.<i>.hchunks.<j>.curhchunks mallctl's.
2014-10-13 13:53:59 +08:00
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Maximum size supported by this huge size
class.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="arenas.extend">
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<term>
<mallctl>arenas.extend</mallctl>
(<type>unsigned</type>)
<literal>r-</literal>
</term>
<listitem><para>Extend the array of arenas by appending a new arena,
and returning the new arena index.</para></listitem>
</varlistentry>
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
<varlistentry id="prof.thread_active_init">
<term>
<mallctl>prof.thread_active_init</mallctl>
(<type>bool</type>)
<literal>rw</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Control the initial setting for <link
linkend="thread.prof.active"><mallctl>thread.prof.active</mallctl></link>
in newly created threads. See the <link
linkend="opt.prof_thread_active_init"><mallctl>opt.prof_thread_active_init</mallctl></link>
option for additional information.</para></listitem>
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<varlistentry id="prof.active">
<term>
<mallctl>prof.active</mallctl>
(<type>bool</type>)
<literal>rw</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Control whether sampling is currently active. See the
<link
linkend="opt.prof_active"><mallctl>opt.prof_active</mallctl></link>
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
option for additional information, as well as the interrelated <link
linkend="thread.prof.active"><mallctl>thread.prof.active</mallctl></link>
mallctl.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="prof.dump">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>prof.dump</mallctl>
(<type>const char *</type>)
<literal>-w</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Dump a memory profile to the specified file, or if NULL
is specified, to a file according to the pattern
<filename>&lt;prefix&gt;.&lt;pid&gt;.&lt;seq&gt;.m&lt;mseq&gt;.heap</filename>,
where <literal>&lt;prefix&gt;</literal> is controlled by the
<link
linkend="opt.prof_prefix"><mallctl>opt.prof_prefix</mallctl></link>
option.</para></listitem>
</varlistentry>
Implement the prof.gdump mallctl. This feature makes it possible to toggle the gdump feature on/off during program execution, whereas the the opt.prof_dump mallctl value can only be set during program startup. This resolves #72.
2015-01-26 13:16:57 +08:00
<varlistentry id="prof.gdump">
<term>
<mallctl>prof.gdump</mallctl>
(<type>bool</type>)
<literal>rw</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>When enabled, trigger a memory profile dump every time
the total virtual memory exceeds the previous maximum. Profiles are
dumped to files named according to the pattern
<filename>&lt;prefix&gt;.&lt;pid&gt;.&lt;seq&gt;.u&lt;useq&gt;.heap</filename>,
where <literal>&lt;prefix&gt;</literal> is controlled by the <link
linkend="opt.prof_prefix"><mallctl>opt.prof_prefix</mallctl></link>
option.</para></listitem>
</varlistentry>
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
<varlistentry id="prof.reset">
<term>
<mallctl>prof.reset</mallctl>
(<type>size_t</type>)
<literal>-w</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Reset all memory profile statistics, and optionally
update the sample rate (see <link
Test prof.reset mallctl and fix numerous discovered bugs.
2014-10-03 14:01:10 +08:00
linkend="opt.lg_prof_sample"><mallctl>opt.lg_prof_sample</mallctl></link>
Implement/test/fix prof-related mallctl's. Implement/test/fix the opt.prof_thread_active_init, prof.thread_active_init, and thread.prof.active mallctl's. Test/fix the thread.prof.name mallctl. Refactor opt_prof_active to be read-only and move mutable state into the prof_active variable. Stop leaning on ctl-related locking for protection.
2014-10-04 14:25:30 +08:00
and <link
linkend="prof.lg_sample"><mallctl>prof.lg_sample</mallctl></link>).
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
</para></listitem>
</varlistentry>
<varlistentry id="prof.lg_sample">
<term>
<mallctl>prof.lg_sample</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
Test prof.reset mallctl and fix numerous discovered bugs.
2014-10-03 14:01:10 +08:00
<listitem><para>Get the current sample rate (see <link
Implement per thread heap profiling. Rename data structures (prof_thr_cnt_t-->prof_tctx_t, prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for sampled objects. Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace depth within jemalloc functions is no longer an issue (pprof prunes irrelevant frames). Implement mallctl's: - prof.reset implements full sample data reset, and optional change of sample interval. - prof.lg_sample reads the current sample interval (opt.lg_prof_sample was the permanent source of truth prior to prof.reset). - thread.prof.name provides naming capability for threads within heap profile dumps. - thread.prof.active makes it possible to activate/deactivate heap profiling for individual threads. Modify the heap dump files to contain per thread heap profile data. This change is incompatible with the existing pprof, which will require enhancements to read and process the enriched data.
2014-08-19 07:22:13 +08:00
linkend="opt.lg_prof_sample"><mallctl>opt.lg_prof_sample</mallctl></link>).
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="prof.interval">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>prof.interval</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-prof</option>]
</term>
<listitem><para>Average number of bytes allocated between
inverval-based profile dumps. See the
<link
linkend="opt.lg_prof_interval"><mallctl>opt.lg_prof_interval</mallctl></link>
option for additional information.</para></listitem>
Add the "stats.cactive" mallctl. Add the "stats.cactive" mallctl, which can be used to efficiently and repeatedly query approximately how much active memory the application is utilizing.
2011-03-19 08:56:14 +08:00
</varlistentry>
<varlistentry id="stats.cactive">
<term>
<mallctl>stats.cactive</mallctl>
(<type>size_t *</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Pointer to a counter that contains an approximate count
of the current number of bytes in active pages. The estimate may be
Fix huge allocation statistics.
2014-10-15 13:20:00 +08:00
high, but never low, because each arena rounds up when computing its
contribution to the counter. Note that the <link
Add the "stats.cactive" mallctl. Add the "stats.cactive" mallctl, which can be used to efficiently and repeatedly query approximately how much active memory the application is utilizing.
2011-03-19 08:56:14 +08:00
linkend="epoch"><mallctl>epoch</mallctl></link> mallctl has no bearing
on this counter. Furthermore, counter consistency is maintained via
atomic operations, so it is necessary to use an atomic operation in
order to guarantee a consistent read when dereferencing the pointer.
</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
<varlistentry id="stats.allocated">
<term>
<mallctl>stats.allocated</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Total number of bytes allocated by the
application.</para></listitem>
</varlistentry>
<varlistentry id="stats.active">
<term>
<mallctl>stats.active</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Total number of bytes in active pages allocated by the
application. This is a multiple of the page size, and greater than or
equal to <link
linkend="stats.allocated"><mallctl>stats.allocated</mallctl></link>.
document what stats.active does not track Based on http://www.canonware.com/pipermail/jemalloc-discuss/2012-March/000164.html
2012-10-18 04:06:32 +08:00
This does not include <link linkend="stats.arenas.i.pdirty">
Move centralized chunk management into arenas. Migrate all centralized data structures related to huge allocations and recyclable chunks into arena_t, so that each arena can manage huge allocations and recyclable virtual memory completely independently of other arenas. Add chunk node caching to arenas, in order to avoid contention on the base allocator. Use chunks_rtree to look up huge allocations rather than a red-black tree. Maintain a per arena unsorted list of huge allocations (which will be needed to enumerate huge allocations during arena reset). Remove the --enable-ivsalloc option, make ivsalloc() always available, and use it for size queries if --enable-debug is enabled. The only practical implications to this removal are that 1) ivsalloc() is now always available during live debugging (and the underlying radix tree is available during core-based debugging), and 2) size query validation can no longer be enabled independent of --enable-debug. Remove the stats.chunks.{current,total,high} mallctls, and replace their underlying statistics with simpler atomically updated counters used exclusively for gdump triggering. These statistics are no longer very useful because each arena manages chunks independently, and per arena statistics provide similar information. Simplify chunk synchronization code, now that base chunk allocation cannot cause recursive lock acquisition.
2015-02-12 04:24:27 +08:00
<mallctl>stats.arenas.&lt;i&gt;.pdirty</mallctl></link>, nor pages
document what stats.active does not track Based on http://www.canonware.com/pipermail/jemalloc-discuss/2012-March/000164.html
2012-10-18 04:06:32 +08:00
entirely devoted to allocator metadata.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Implement metadata statistics. There are three categories of metadata: - Base allocations are used for bootstrap-sensitive internal allocator data structures. - Arena chunk headers comprise pages which track the states of the non-metadata pages. - Internal allocations differ from application-originated allocations in that they are for internal use, and that they are omitted from heap profiles. The metadata statistics comprise the metadata categories as follows: - stats.metadata: All metadata -- base + arena chunk headers + internal allocations. - stats.arenas.<i>.metadata.mapped: Arena chunk headers. - stats.arenas.<i>.metadata.allocated: Internal allocations. This is reported separately from the other metadata statistics because it overlaps with the allocated and active statistics, whereas the other metadata statistics do not. Base allocations are not reported separately, though their magnitude can be computed by subtracting the arena-specific metadata. This resolves #163.
2014-11-28 03:22:36 +08:00
<varlistentry id="stats.metadata">
<term>
<mallctl>stats.metadata</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Total number of bytes dedicated to metadata, which
comprise base allocations used for bootstrap-sensitive internal
allocator data structures, arena chunk headers (see <link
linkend="stats.arenas.i.metadata.mapped"><mallctl>stats.arenas.&lt;i&gt;.metadata.mapped</mallctl></link>),
and internal allocations (see <link
linkend="stats.arenas.i.metadata.allocated"><mallctl>stats.arenas.&lt;i&gt;.metadata.allocated</mallctl></link>).</para></listitem>
</varlistentry>
Add the "stats.allocated" mallctl.
2015-03-24 08:25:57 +08:00
<varlistentry id="stats.resident">
<term>
<mallctl>stats.resident</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Maximum number of bytes in physically resident data
pages mapped by the allocator, comprising all pages dedicated to
allocator metadata, pages backing active allocations, and unused dirty
pages. This is a maximum rather than precise because pages may not
actually be physically resident if they correspond to demand-zeroed
virtual memory that has not yet been touched. This is a multiple of the
page size, and is larger than <link
linkend="stats.active"><mallctl>stats.active</mallctl></link>.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.mapped">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.mapped</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
Add the "stats.allocated" mallctl.
2015-03-24 08:25:57 +08:00
<listitem><para>Total number of bytes in active chunks mapped by the
Clarify relationship between stats.resident and stats.mapped.
2015-05-29 06:03:58 +08:00
allocator. This is a multiple of the chunk size, and is larger than
<link linkend="stats.active"><mallctl>stats.active</mallctl></link>.
This does not include inactive chunks, even those that contain unused
dirty pages, which means that there is no strict ordering between this
and <link
linkend="stats.resident"><mallctl>stats.resident</mallctl></link>.</para></listitem>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.dss">
Add arena-specific and selective dss allocation. Add the "arenas.extend" mallctl, so that it is possible to create new arenas that are outside the set that jemalloc automatically multiplexes threads onto. Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible to explicitly allocate from a particular arena. Add the "opt.dss" mallctl, which controls the default precedence of dss allocation relative to mmap allocation. Add the "arena.<i>.dss" mallctl, which makes it possible to set the default dss precedence on a per arena or global basis. Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge". Add the "stats.arenas.<i>.dss" mallctl.
2012-10-12 04:53:15 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.dss</mallctl>
(<type>const char *</type>)
<literal>r-</literal>
</term>
<listitem><para>dss (<citerefentry><refentrytitle>sbrk</refentrytitle>
<manvolnum>2</manvolnum></citerefentry>) allocation precedence as
related to <citerefentry><refentrytitle>mmap</refentrytitle>
<manvolnum>2</manvolnum></citerefentry> allocation. See <link
linkend="opt.dss"><mallctl>opt.dss</mallctl></link> for details.
</para></listitem>
</varlistentry>
Add the "stats.arenas.<i>.lg_dirty_mult" mallctl.
2015-03-25 07:36:12 +08:00
<varlistentry id="stats.arenas.i.lg_dirty_mult">
<term>
<mallctl>stats.arenas.&lt;i&gt;.lg_dirty_mult</mallctl>
(<type>ssize_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Minimum ratio (log base 2) of active to dirty pages.
See <link
linkend="opt.lg_dirty_mult"><mallctl>opt.lg_dirty_mult</mallctl></link>
for details.</para></listitem>
</varlistentry>
Implement decay-based unused dirty page purging. This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
2016-02-20 12:09:31 +08:00
<varlistentry id="stats.arenas.i.decay_time">
<term>
<mallctl>stats.arenas.&lt;i&gt;.decay_time</mallctl>
(<type>ssize_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Approximate time in seconds from the creation of a set
of unused dirty pages until an equivalent set of unused dirty pages is
purged and/or reused. See <link
linkend="opt.decay_time"><mallctl>opt.decay_time</mallctl></link>
for details.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.nthreads">
Improve thread-->arena assignment. Rather than blindly assigning threads to arenas in round-robin fashion, choose the lowest-numbered arena that currently has the smallest number of threads assigned to it. Add the "stats.arenas.<i>.nthreads" mallctl.
2011-03-19 04:41:33 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.nthreads</mallctl>
(<type>unsigned</type>)
<literal>r-</literal>
</term>
<listitem><para>Number of threads currently assigned to
arena.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.pactive">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.pactive</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Number of pages in active runs.</para></listitem>
</varlistentry>
document what stats.active does not track Based on http://www.canonware.com/pipermail/jemalloc-discuss/2012-March/000164.html
2012-10-18 04:06:32 +08:00
<varlistentry id="stats.arenas.i.pdirty">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.pdirty</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
</term>
<listitem><para>Number of pages within unused runs that are potentially
dirty, and for which <function>madvise<parameter>...</parameter>
<parameter><constant>MADV_DONTNEED</constant></parameter></function> or
similar has not been called.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.mapped">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.mapped</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of mapped bytes.</para></listitem>
</varlistentry>
Implement metadata statistics. There are three categories of metadata: - Base allocations are used for bootstrap-sensitive internal allocator data structures. - Arena chunk headers comprise pages which track the states of the non-metadata pages. - Internal allocations differ from application-originated allocations in that they are for internal use, and that they are omitted from heap profiles. The metadata statistics comprise the metadata categories as follows: - stats.metadata: All metadata -- base + arena chunk headers + internal allocations. - stats.arenas.<i>.metadata.mapped: Arena chunk headers. - stats.arenas.<i>.metadata.allocated: Internal allocations. This is reported separately from the other metadata statistics because it overlaps with the allocated and active statistics, whereas the other metadata statistics do not. Base allocations are not reported separately, though their magnitude can be computed by subtracting the arena-specific metadata. This resolves #163.
2014-11-28 03:22:36 +08:00
<varlistentry id="stats.arenas.i.metadata.mapped">
<term>
<mallctl>stats.arenas.&lt;i&gt;.metadata.mapped</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of mapped bytes in arena chunk headers, which
track the states of the non-metadata pages.</para></listitem>
</varlistentry>
<varlistentry id="stats.arenas.i.metadata.allocated">
<term>
<mallctl>stats.arenas.&lt;i&gt;.metadata.allocated</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of bytes dedicated to internal allocations.
Internal allocations differ from application-originated allocations in
that they are for internal use, and that they are omitted from heap
profiles. This statistic is reported separately from <link
linkend="stats.metadata"><mallctl>stats.metadata</mallctl></link> and
<link
linkend="stats.arenas.i.metadata.mapped"><mallctl>stats.arenas.&lt;i&gt;.metadata.mapped</mallctl></link>
because it overlaps with e.g. the <link
linkend="stats.allocated"><mallctl>stats.allocated</mallctl></link> and
<link linkend="stats.active"><mallctl>stats.active</mallctl></link>
statistics, whereas the other metadata statistics do
not.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.npurge">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.npurge</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of dirty page purge sweeps performed.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.nmadvise">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.nmadvise</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of <function>madvise<parameter>...</parameter>
<parameter><constant>MADV_DONTNEED</constant></parameter></function> or
similar calls made to purge dirty pages.</para></listitem>
</varlistentry>
Fix a few mallctl() documentation errors. Normalize mallctl() order (code and documentation).
2013-12-20 13:40:41 +08:00
<varlistentry id="stats.arenas.i.purged">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
Fix a few mallctl() documentation errors. Normalize mallctl() order (code and documentation).
2013-12-20 13:40:41 +08:00
<mallctl>stats.arenas.&lt;i&gt;.purged</mallctl>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of pages purged.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.small.allocated">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.small.allocated</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of bytes currently allocated by small objects.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.small.nmalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.small.nmalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocation requests served by
small bins.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.small.ndalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.small.ndalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of small objects returned to bins.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.small.nrequests">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.small.nrequests</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of small allocation requests.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.large.allocated">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.large.allocated</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of bytes currently allocated by large objects.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.large.nmalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.large.nmalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of large allocation requests served
directly by the arena.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.large.ndalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.large.ndalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of large deallocation requests served
directly by the arena.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.large.nrequests">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.large.nrequests</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of large allocation requests.
</para></listitem>
</varlistentry>
Refactor huge allocation to be managed by arenas. Refactor huge allocation to be managed by arenas (though the global red-black tree of huge allocations remains for lookup during deallocation). This is the logical conclusion of recent changes that 1) made per arena dss precedence apply to huge allocation, and 2) made it possible to replace the per arena chunk allocation/deallocation functions. Remove the top level huge stats, and replace them with per arena huge stats. Normalize function names and types to *dalloc* (some were *dealloc*). Remove the --enable-mremap option. As jemalloc currently operates, this is a performace regression for some applications, but planned work to logarithmically space huge size classes should provide similar amortized performance. The motivation for this change was that mremap-based huge reallocation forced leaky abstractions that prevented refactoring.
2014-05-16 13:22:27 +08:00
<varlistentry id="stats.arenas.i.huge.allocated">
<term>
<mallctl>stats.arenas.&lt;i&gt;.huge.allocated</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Number of bytes currently allocated by huge objects.
</para></listitem>
</varlistentry>
<varlistentry id="stats.arenas.i.huge.nmalloc">
<term>
<mallctl>stats.arenas.&lt;i&gt;.huge.nmalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of huge allocation requests served
directly by the arena.</para></listitem>
</varlistentry>
<varlistentry id="stats.arenas.i.huge.ndalloc">
<term>
<mallctl>stats.arenas.&lt;i&gt;.huge.ndalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of huge deallocation requests served
directly by the arena.</para></listitem>
</varlistentry>
<varlistentry id="stats.arenas.i.huge.nrequests">
<term>
<mallctl>stats.arenas.&lt;i&gt;.huge.nrequests</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of huge allocation requests.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.nmalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.nmalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocations served by bin.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.ndalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.ndalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocations returned to bin.
</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.nrequests">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.nrequests</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocation
requests.</para></listitem>
</varlistentry>
Add per size class huge allocation statistics. Add per size class huge allocation statistics, and normalize various stats: - Change the arenas.nlruns type from size_t to unsigned. - Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's. - Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with stats.arenas.<i>.bins.<j>.curregs . - Add the stats.arenas.<i>.hchunks.<j>.nmalloc, stats.arenas.<i>.hchunks.<j>.ndalloc, stats.arenas.<i>.hchunks.<j>.nrequests, and stats.arenas.<i>.hchunks.<j>.curhchunks mallctl's.
2014-10-13 13:53:59 +08:00
<varlistentry id="stats.arenas.i.bins.j.curregs">
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.curregs</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Current number of regions for this size
class.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.nfills">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.nfills</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option> <option>--enable-tcache</option>]
</term>
<listitem><para>Cumulative number of tcache fills.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.nflushes">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.nflushes</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option> <option>--enable-tcache</option>]
</term>
<listitem><para>Cumulative number of tcache flushes.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.nruns">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.nruns</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of runs created.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.nreruns">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.nreruns</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of times the current run from which
to allocate changed.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.bins.j.curruns">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.bins.&lt;j&gt;.curruns</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Current number of runs.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.lruns.j.nmalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.lruns.&lt;j&gt;.nmalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocation requests for this size
class served directly by the arena.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.lruns.j.ndalloc">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.lruns.&lt;j&gt;.ndalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of deallocation requests for this
size class served directly by the arena.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.lruns.j.nrequests">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.lruns.&lt;j&gt;.nrequests</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocation requests for this size
class.</para></listitem>
</varlistentry>
Add ids for all mallctl entries. Add ids for all mallctl entries, so that external documents can link to arbitrary mallctl entries.
2013-10-31 05:52:09 +08:00
<varlistentry id="stats.arenas.i.lruns.j.curruns">
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<term>
<mallctl>stats.arenas.&lt;i&gt;.lruns.&lt;j&gt;.curruns</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Current number of runs for this size class.
</para></listitem>
</varlistentry>
Add per size class huge allocation statistics. Add per size class huge allocation statistics, and normalize various stats: - Change the arenas.nlruns type from size_t to unsigned. - Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's. - Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with stats.arenas.<i>.bins.<j>.curregs . - Add the stats.arenas.<i>.hchunks.<j>.nmalloc, stats.arenas.<i>.hchunks.<j>.ndalloc, stats.arenas.<i>.hchunks.<j>.nrequests, and stats.arenas.<i>.hchunks.<j>.curhchunks mallctl's.
2014-10-13 13:53:59 +08:00
<varlistentry id="stats.arenas.i.hchunks.j.nmalloc">
<term>
<mallctl>stats.arenas.&lt;i&gt;.hchunks.&lt;j&gt;.nmalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocation requests for this size
class served directly by the arena.</para></listitem>
</varlistentry>
<varlistentry id="stats.arenas.i.hchunks.j.ndalloc">
<term>
<mallctl>stats.arenas.&lt;i&gt;.hchunks.&lt;j&gt;.ndalloc</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of deallocation requests for this
size class served directly by the arena.</para></listitem>
</varlistentry>
<varlistentry id="stats.arenas.i.hchunks.j.nrequests">
<term>
<mallctl>stats.arenas.&lt;i&gt;.hchunks.&lt;j&gt;.nrequests</mallctl>
(<type>uint64_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Cumulative number of allocation requests for this size
class.</para></listitem>
</varlistentry>
<varlistentry id="stats.arenas.i.hchunks.j.curhchunks">
<term>
<mallctl>stats.arenas.&lt;i&gt;.hchunks.&lt;j&gt;.curhchunks</mallctl>
(<type>size_t</type>)
<literal>r-</literal>
[<option>--enable-stats</option>]
</term>
<listitem><para>Current number of huge allocations for this size class.
</para></listitem>
</varlistentry>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</variablelist>
</refsect1>
<refsect1 id="debugging_malloc_problems">
<title>DEBUGGING MALLOC PROBLEMS</title>
<para>When debugging, it is a good idea to configure/build jemalloc with
the <option>--enable-debug</option> and <option>--enable-fill</option>
options, and recompile the program with suitable options and symbols for
debugger support. When so configured, jemalloc incorporates a wide variety
of run-time assertions that catch application errors such as double-free,
write-after-free, etc.</para>
<para>Programs often accidentally depend on &ldquo;uninitialized&rdquo;
memory actually being filled with zero bytes. Junk filling
(see the <link linkend="opt.junk"><mallctl>opt.junk</mallctl></link>
option) tends to expose such bugs in the form of obviously incorrect
results and/or coredumps. Conversely, zero
filling (see the <link
linkend="opt.zero"><mallctl>opt.zero</mallctl></link> option) eliminates
the symptoms of such bugs. Between these two options, it is usually
possible to quickly detect, diagnose, and eliminate such bugs.</para>
<para>This implementation does not provide much detail about the problems
it detects, because the performance impact for storing such information
Implement Valgrind support, redzones, and quarantine. Implement Valgrind support, as well as the redzone and quarantine features, which help Valgrind detect memory errors. Redzones are only implemented for small objects because the changes necessary to support redzones around large and huge objects are complicated by in-place reallocation, to the point that it isn't clear that the maintenance burden is worth the incremental improvement to Valgrind support. Merge arena_salloc() and arena_salloc_demote(). Refactor i[v]salloc() to expose the 'demote' option.
2012-04-06 15:35:09 +08:00
would be prohibitive. However, jemalloc does integrate with the most
Fix Valgrind URL in documentation. Reported by Daichi GOTO.
2012-04-26 14:17:57 +08:00
excellent <ulink url="http://valgrind.org/">Valgrind</ulink> tool if the
Auto-detect whether running inside Valgrind. Auto-detect whether running inside Valgrind, thus removing the need to manually specify MALLOC_CONF=valgrind:true.
2012-05-16 05:48:14 +08:00
<option>--enable-valgrind</option> configuration option is enabled.</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</refsect1>
<refsect1 id="diagnostic_messages">
<title>DIAGNOSTIC MESSAGES</title>
<para>If any of the memory allocation/deallocation functions detect an
error or warning condition, a message will be printed to file descriptor
<constant>STDERR_FILENO</constant>. Errors will result in the process
dumping core. If the <link
linkend="opt.abort"><mallctl>opt.abort</mallctl></link> option is set, most
warnings are treated as errors.</para>
<para>The <varname>malloc_message</varname> variable allows the programmer
to override the function which emits the text strings forming the errors
and warnings if for some reason the <constant>STDERR_FILENO</constant> file
descriptor is not suitable for this.
<function>malloc_message<parameter/></function> takes the
<parameter>cbopaque</parameter> pointer argument that is
<constant>NULL</constant> unless overridden by the arguments in a call to
<function>malloc_stats_print<parameter/></function>, followed by a string
pointer. Please note that doing anything which tries to allocate memory in
this function is likely to result in a crash or deadlock.</para>
<para>All messages are prefixed by
&ldquo;<computeroutput>&lt;jemalloc&gt;: </computeroutput>&rdquo;.</para>
</refsect1>
<refsect1 id="return_values">
<title>RETURN VALUES</title>
<refsect2>
<title>Standard API</title>
<para>The <function>malloc<parameter/></function> and
<function>calloc<parameter/></function> functions return a pointer to the
allocated memory if successful; otherwise a <constant>NULL</constant>
pointer is returned and <varname>errno</varname> is set to
<errorname>ENOMEM</errorname>.</para>
<para>The <function>posix_memalign<parameter/></function> function
returns the value 0 if successful; otherwise it returns an error value.
The <function>posix_memalign<parameter/></function> function will fail
if:
<variablelist>
<varlistentry>
<term><errorname>EINVAL</errorname></term>
<listitem><para>The <parameter>alignment</parameter> parameter is
not a power of 2 at least as large as
<code language="C">sizeof(<type>void *</type>)</code>.
</para></listitem>
</varlistentry>
<varlistentry>
<term><errorname>ENOMEM</errorname></term>
<listitem><para>Memory allocation error.</para></listitem>
</varlistentry>
</variablelist>
</para>
Implement aligned_alloc(). Implement aligned_alloc(), which was added in the C11 standard. The function is weakly specified to the point that a minimally compliant implementation would be painful to use (size must be an integral multiple of alignment!), which in practice makes posix_memalign() a safer choice.
2012-03-14 03:55:21 +08:00
<para>The <function>aligned_alloc<parameter/></function> function returns
a pointer to the allocated memory if successful; otherwise a
<constant>NULL</constant> pointer is returned and
<varname>errno</varname> is set. The
<function>aligned_alloc<parameter/></function> function will fail if:
<variablelist>
<varlistentry>
<term><errorname>EINVAL</errorname></term>
<listitem><para>The <parameter>alignment</parameter> parameter is
not a power of 2.
</para></listitem>
</varlistentry>
<varlistentry>
<term><errorname>ENOMEM</errorname></term>
<listitem><para>Memory allocation error.</para></listitem>
</varlistentry>
</variablelist>
</para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<para>The <function>realloc<parameter/></function> function returns a
pointer, possibly identical to <parameter>ptr</parameter>, to the
allocated memory if successful; otherwise a <constant>NULL</constant>
pointer is returned, and <varname>errno</varname> is set to
<errorname>ENOMEM</errorname> if the error was the result of an
allocation failure. The <function>realloc<parameter/></function>
function always leaves the original buffer intact when an error occurs.
</para>
<para>The <function>free<parameter/></function> function returns no
value.</para>
</refsect2>
<refsect2>
<title>Non-standard API</title>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>mallocx<parameter/></function> and
<function>rallocx<parameter/></function> functions return a pointer to
the allocated memory if successful; otherwise a <constant>NULL</constant>
pointer is returned to indicate insufficient contiguous memory was
available to service the allocation request. </para>
<para>The <function>xallocx<parameter/></function> function returns the
real size of the resulting resized allocation pointed to by
<parameter>ptr</parameter>, which is a value less than
<parameter>size</parameter> if the allocation could not be adequately
grown in place. </para>
<para>The <function>sallocx<parameter/></function> function returns the
real size of the allocation pointed to by <parameter>ptr</parameter>.
</para>
<para>The <function>nallocx<parameter/></function> returns the real size
that would result from a successful equivalent
<function>mallocx<parameter/></function> function call, or zero if
insufficient memory is available to perform the size computation. </para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<para>The <function>mallctl<parameter/></function>,
<function>mallctlnametomib<parameter/></function>, and
<function>mallctlbymib<parameter/></function> functions return 0 on
success; otherwise they return an error value. The functions will fail
if:
<variablelist>
<varlistentry>
<term><errorname>EINVAL</errorname></term>
<listitem><para><parameter>newp</parameter> is not
<constant>NULL</constant>, and <parameter>newlen</parameter> is too
large or too small. Alternatively, <parameter>*oldlenp</parameter>
is too large or too small; in this case as much data as possible
are read despite the error.</para></listitem>
</varlistentry>
<varlistentry>
<term><errorname>ENOENT</errorname></term>
<listitem><para><parameter>name</parameter> or
<parameter>mib</parameter> specifies an unknown/invalid
value.</para></listitem>
</varlistentry>
<varlistentry>
<term><errorname>EPERM</errorname></term>
<listitem><para>Attempt to read or write void value, or attempt to
write read-only value.</para></listitem>
</varlistentry>
<varlistentry>
<term><errorname>EAGAIN</errorname></term>
<listitem><para>A memory allocation failure
occurred.</para></listitem>
</varlistentry>
<varlistentry>
<term><errorname>EFAULT</errorname></term>
<listitem><para>An interface with side effects failed in some way
not directly related to <function>mallctl*<parameter/></function>
read/write processing.</para></listitem>
</varlistentry>
</variablelist>
</para>
Implement the *allocx() API. Implement the *allocx() API, which is a successor to the *allocm() API. The *allocx() functions are slightly simpler to use because they have fewer parameters, they directly return the results of primary interest, and mallocx()/rallocx() avoid the strict aliasing pitfall that allocm()/rallocx() share with posix_memalign(). The following code violates strict aliasing rules: foo_t *foo; allocm((void **)&foo, NULL, 42, 0); whereas the following is safe: foo_t *foo; void *p; allocm(&p, NULL, 42, 0); foo = (foo_t *)p; mallocx() does not have this problem: foo_t *foo = (foo_t *)mallocx(42, 0);
2013-12-13 14:35:52 +08:00
<para>The <function>malloc_usable_size<parameter/></function> function
returns the usable size of the allocation pointed to by
<parameter>ptr</parameter>. </para>
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
</refsect2>
</refsect1>
<refsect1 id="environment">
<title>ENVIRONMENT</title>
<para>The following environment variable affects the execution of the
allocation functions:
<variablelist>
<varlistentry>
<term><envar>MALLOC_CONF</envar></term>
<listitem><para>If the environment variable
<envar>MALLOC_CONF</envar> is set, the characters it contains
will be interpreted as options.</para></listitem>
</varlistentry>
</variablelist>
</para>
</refsect1>
<refsect1 id="examples">
<title>EXAMPLES</title>
<para>To dump core whenever a problem occurs:
<screen>ln -s 'abort:true' /etc/malloc.conf</screen>
</para>
<para>To specify in the source a chunk size that is 16 MiB:
<programlisting language="C"><![CDATA[
malloc_conf = "lg_chunk:24";]]></programlisting></para>
</refsect1>
<refsect1 id="see_also">
<title>SEE ALSO</title>
<para><citerefentry><refentrytitle>madvise</refentrytitle>
<manvolnum>2</manvolnum></citerefentry>,
<citerefentry><refentrytitle>mmap</refentrytitle>
<manvolnum>2</manvolnum></citerefentry>,
<citerefentry><refentrytitle>sbrk</refentrytitle>
<manvolnum>2</manvolnum></citerefentry>,
Add utrace(2)-based tracing (--enable-utrace).
2012-04-06 04:36:17 +08:00
<citerefentry><refentrytitle>utrace</refentrytitle>
<manvolnum>2</manvolnum></citerefentry>,
Convert man page from roff to DocBook. Convert the man page source from roff to DocBook, and generate html and roff output. Modify the build system such that the documentation can be built as part of the release process, so that users need not have DocBook tools installed.
2010-11-25 14:00:02 +08:00
<citerefentry><refentrytitle>alloca</refentrytitle>
<manvolnum>3</manvolnum></citerefentry>,
<citerefentry><refentrytitle>atexit</refentrytitle>
<manvolnum>3</manvolnum></citerefentry>,
<citerefentry><refentrytitle>getpagesize</refentrytitle>
<manvolnum>3</manvolnum></citerefentry></para>
</refsect1>
<refsect1 id="standards">
<title>STANDARDS</title>
<para>The <function>malloc<parameter/></function>,
<function>calloc<parameter/></function>,
<function>realloc<parameter/></function>, and
<function>free<parameter/></function> functions conform to ISO/IEC
9899:1990 (&ldquo;ISO C90&rdquo;).</para>
<para>The <function>posix_memalign<parameter/></function> function conforms
to IEEE Std 1003.1-2001 (&ldquo;POSIX.1&rdquo;).</para>
</refsect1>
</refentry>
Reference in New Issue Copy Permalink