Inferencium/website
1
0
Fork 0
Code Issues Activity

Fix 1st-level heading and section indentation

Browse Source
This commit is contained in:
inference 2024-01-29 20:23:29 +00:00
parent 90911cc33c
commit 67b88c3af1
Signed by: inference
SSH Key Fingerprint: SHA256:FtEVfx1CmTKMy40VwZvF4k+3TC+QhCWy+EmPRg50Nnc
1 changed files with 83 additions and 79 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

56
documentation/hardened_malloc.xhtml
View File

@ -1,7 +1,7 @@
<!DOCTYPE html>
<!-- Inferencium - Website - Documentation - hardened_malloc -->
<!-- Version: 3.0.0-alpha.8 -->
<!-- Version: 3.0.0-alpha.9 -->
<!-- Copyright 2023 Jake Winters -->
<!-- SPDX-License-Identifier: BSD-3-Clause -->
@ -27,16 +27,18 @@
<div><a href="../directory.xhtml">Directory</a></div>
<div><a href="../key.xhtml">Key</a></div>
</nav>
<h1 id="hardened_malloc"><a href="#hardened_malloc">Documentation - hardened_malloc</a></h1>
<section id="introduction">
<h1 id="introduction"><a href="#introduction">Documentation - hardened_malloc</a></h1>
<p>This documentation contains instructions to use
<a href="https://github.com/GrapheneOS/hardened_malloc">hardened_malloc</a>
memory allocator as the system's default memory allocator. These instructions apply to
both musl and glibc C libraries on Unix-based and Unix-like systems.</p>
<p>hardened_malloc can also be used per-application and/or per-user, in which case root
permissions are not required; this documentation focuses on system-wide usage of
hardened_malloc, assumes root privileges, and assumes the compiled library will be
located in a path readable and executable by all users of the system.</p>
memory allocator as the system's default memory allocator. These instructions
apply to both musl and glibc C libraries on Unix-based and Unix-like
systems.</p>
<p>hardened_malloc can also be used per-application and/or per-user, in which
case root permissions are not required; this documentation focuses on
system-wide usage of hardened_malloc, assumes root privileges, and assumes the
compiled library will be located in a path readable and executable by all users
of the system.</p>
<p>For the complete hardened_malloc documentation, visit its
<a href="https://github.com/GrapheneOS/hardened_malloc#hardened_malloc">official documentation</a>.</p>
<p>This documentation is also available in portable AsciiDoc format in my
@ -56,8 +58,8 @@
<section id="memory_pages">
<h2 id="memory_pages"><a href="#memory_pages">Increase Permitted Amount of Memory Pages</a></h2>
<p>Add <code>vm.max_map_count = 1048576</code> to
<code>/etc/sysctl.conf</code> to accommodate hardened_malloc's large amount of guard
pages.</p>
<code>/etc/sysctl.conf</code> to accommodate hardened_malloc's large amount of
guard pages.</p>
</section>
<section id="clone_source_code">
<h2 id="clone_source_code"><a href="#clone_source_code">Clone hardened_malloc Source Code</a></h2>
@ -71,24 +73,26 @@
<h2 id="compile"><a href="#compile">Compile hardened_malloc</a></h2>
<p><code>$ make <var>&lt;arguments&gt;</var></code></p>
<p><code>CONFIG_N_ARENA=<var>n</var></code> can be adjusted to increase parallel
performance at the expense of memory usage, or decrease memory usage at the expense of
parallel performance, where <var>n</var> is an integer. Higher values prefer parallel
performance, whereas lower values prefer lower memory usage. Note that having too many
arenas may cause memory fragmentation and decrease system performance. The number of
arenas has no impact on the security properties of hardened_malloc.</p>
performance at the expense of memory usage, or decrease memory usage at the
expense of parallel performance, where <var>n</var> is an integer. Higher values
prefer parallel performance, whereas lower values prefer lower memory usage.
Note that having too many arenas may cause memory fragmentation and decrease
system performance. The number of arenas has no impact on the security
properties of hardened_malloc.</p>
<p><b>Minimum number of arenas:</b> 1</p>
<p><b>Maximum number of arenas:</b> 256</p>
<p>For extra security, <code>CONFIG_SEAL_METADATA=true</code> can be used in order to
control whether Memory Protection Keys are used to disable access to all writable
allocator state outside of the memory allocator code. It's currently disabled by default
due to a significant performance cost for this use case on current-generation hardware.
Whether or not this feature is enabled, the metadata is all contained within an isolated
memory region with high-entropy random guard regions around it.</p>
<p>For low-memory systems, <code>VARIANT=light</code> can be used to compile the light
variant of hardened_malloc, which sacrifices some security for much less memory
usage. This option still produces a more hardened memory allocator than both the
default musl and glibc allocators, despite the security sacrifices over the full
variant.</p>
<p>For extra security, <code>CONFIG_SEAL_METADATA=true</code> can be used in
order to control whether Memory Protection Keys are used to disable access to
all writable allocator state outside of the memory allocator code. It's
currently disabled by default due to a significant performance cost for this use
case on current-generation hardware. Whether or not this feature is enabled, the
metadata is all contained within an isolated memory region with high-entropy
random guard regions around it.</p>
<p>For low-memory systems, <code>VARIANT=light</code> can be used to compile the
light variant of hardened_malloc, which sacrifices some security for much less
memory usage. This option still produces a more hardened memory allocator than
both the default musl and glibc allocators, despite the security sacrifices over
the full variant.</p>
<p>For all compile-time options, see the
<a href="https://github.com/GrapheneOS/hardened_malloc#configuration">configuration section</a>
of hardened_malloc's extensive official documentation.</p>