<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
	<id>https://tech-pubs.net/wiki/index.php?action=history&amp;feed=atom&amp;title=Inst_file_formats</id>
	<title>Inst file formats - Revision history</title>
	<link rel="self" type="application/atom+xml" href="https://tech-pubs.net/wiki/index.php?action=history&amp;feed=atom&amp;title=Inst_file_formats"/>
	<link rel="alternate" type="text/html" href="https://tech-pubs.net/wiki/index.php?title=Inst_file_formats&amp;action=history"/>
	<updated>2026-04-20T11:51:34Z</updated>
	<subtitle>Revision history for this page on the wiki</subtitle>
	<generator>MediaWiki 1.44.2</generator>
	<entry>
		<id>https://tech-pubs.net/wiki/index.php?title=Inst_file_formats&amp;diff=400&amp;oldid=prev</id>
		<title>Raion: Created page with &quot;This is a page documenting various parts of the IRIX 6.5 inst file formats. The goal here is to build a formal specification through reading and analyzing files. We do not intend to use disassembly or other methods to RE these formats if possible.   == &lt;code&gt;/var/inst&lt;/code&gt; — per-product record file format ==  === Overview === A product record file stores metadata for a single installed product. It contains:  * A fixed ASCII signature and version * One or more TLV...&quot;</title>
		<link rel="alternate" type="text/html" href="https://tech-pubs.net/wiki/index.php?title=Inst_file_formats&amp;diff=400&amp;oldid=prev"/>
		<updated>2025-10-12T15:54:57Z</updated>

		<summary type="html">&lt;p&gt;Created page with &amp;quot;This is a page documenting various parts of the &lt;a href=&quot;/wiki/index.php/IRIX_6.5&quot; title=&quot;IRIX 6.5&quot;&gt;IRIX 6.5&lt;/a&gt; inst file formats. The goal here is to build a formal specification through reading and analyzing files. We do not intend to use disassembly or other methods to RE these formats if possible.   == &amp;lt;code&amp;gt;/var/inst&amp;lt;/code&amp;gt; — per-product record file format ==  === Overview === A product record file stores metadata for a single installed product. It contains:  * A fixed ASCII signature and version * One or more TLV...&amp;quot;&lt;/p&gt;
&lt;p&gt;&lt;b&gt;New page&lt;/b&gt;&lt;/p&gt;&lt;div&gt;This is a page documenting various parts of the [[IRIX 6.5]] inst file formats. The goal here is to build a formal specification through reading and analyzing files. We do not intend to use disassembly or other methods to RE these formats if possible. &lt;br /&gt;
&lt;br /&gt;
== &amp;lt;code&amp;gt;/var/inst&amp;lt;/code&amp;gt; — per-product record file format ==&lt;br /&gt;
&lt;br /&gt;
=== Overview ===&lt;br /&gt;
A product record file stores metadata for a single installed product. It contains:&lt;br /&gt;
&lt;br /&gt;
* A fixed ASCII signature and version&lt;br /&gt;
* One or more TLV-like record blocks (binary header fields + NUL-terminated ASCII strings)&lt;br /&gt;
* 4-byte tokens (timestamps/IDs/checksums) associated with records&lt;br /&gt;
&lt;br /&gt;
All multi-byte integer fields are &amp;#039;&amp;#039;&amp;#039;big-endian&amp;#039;&amp;#039;&amp;#039;.&lt;br /&gt;
----&lt;br /&gt;
&lt;br /&gt;
=== File header ===&lt;br /&gt;
Notes: &amp;lt;code&amp;gt;0x0d..0x0f&amp;lt;/code&amp;gt; are consistent across many samples; treat as header flags/format info.&lt;br /&gt;
----&lt;br /&gt;
&lt;br /&gt;
=== Record block (high level) ===&lt;br /&gt;
Records are variable-length. A common pattern:&lt;br /&gt;
&lt;br /&gt;
Record header often appears as compact fields (examples observed):&lt;br /&gt;
&lt;br /&gt;
Strings are stored as usual ASCII and terminated by 0x00. The set and order of string fields are consistent per record type (product name -&amp;gt; long name -&amp;gt; build id -&amp;gt; subsystem entries, etc.).&lt;br /&gt;
----&lt;br /&gt;
&lt;br /&gt;
=== Common tokens &amp;amp; patterns ===&lt;br /&gt;
&lt;br /&gt;
* &amp;lt;code&amp;gt;pd001&amp;lt;/code&amp;gt; — product DB magic.&lt;br /&gt;
* &amp;lt;code&amp;gt;V630&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;V650&amp;lt;/code&amp;gt;, etc. — format/release code.&lt;br /&gt;
* &amp;lt;code&amp;gt;P00&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;P01&amp;lt;/code&amp;gt; — file-type marker.&lt;br /&gt;
* Repeating 4-byte sequences near records typically represent UNIX timestamps or stable IDs.&lt;br /&gt;
* &amp;lt;code&amp;gt;\0&amp;lt;/code&amp;gt; is used widely as a delimiter between strings.&lt;br /&gt;
&lt;br /&gt;
== &amp;lt;code&amp;gt;hist&amp;lt;/code&amp;gt; — inst history file (draft) ==&lt;br /&gt;
&lt;br /&gt;
=== High-level layout ===&lt;br /&gt;
&lt;br /&gt;
=== String table ===&lt;br /&gt;
&lt;br /&gt;
* Appears as a contiguous region filled with many NUL-terminated printable strings (product/subsystem names).&lt;br /&gt;
* The parser should locate it by picking the file window that contains many NUL-separated printable ASCII runs (heuristic).&lt;br /&gt;
&lt;br /&gt;
=== Binary record format (canonical candidate) ===&lt;br /&gt;
Many history records follow this layout (big-endian):&lt;br /&gt;
&lt;br /&gt;
* &amp;lt;code&amp;gt;name_off_be&amp;lt;/code&amp;gt; is an &amp;#039;&amp;#039;&amp;#039;absolute&amp;#039;&amp;#039;&amp;#039; offset into the history file; reading from that offset to the next NUL yields the referenced product/subsystem string.&lt;br /&gt;
* Follow-on payload bytes (after the 12-byte header) vary by &amp;lt;code&amp;gt;action_be&amp;lt;/code&amp;gt; value and can include counts, other offsets, path strings (NUL terminated), and 4-byte timestamps/checksums.&lt;br /&gt;
&lt;br /&gt;
----&lt;br /&gt;
&lt;br /&gt;
=== Semantics (what to interpret) ===&lt;br /&gt;
&lt;br /&gt;
* &amp;lt;code&amp;gt;ts_be&amp;lt;/code&amp;gt; — time the event occurred (seconds since epoch).&lt;br /&gt;
* &amp;lt;code&amp;gt;action_be&amp;lt;/code&amp;gt; — action code: e.g. install, remove, update, patch, etc. (numerical mapping TBD).&lt;br /&gt;
* &amp;lt;code&amp;gt;flags_be&amp;lt;/code&amp;gt; — modifiers for action (e.g. success/failure bits, location flags).&lt;br /&gt;
* &amp;lt;code&amp;gt;name_off_be&amp;lt;/code&amp;gt; — link to string in the string table.&lt;br /&gt;
&lt;br /&gt;
----&lt;br /&gt;
&lt;br /&gt;
=== Unknowns / caveats ===&lt;br /&gt;
&lt;br /&gt;
* Exact semantics of &amp;lt;code&amp;gt;action_be&amp;lt;/code&amp;gt; and &amp;lt;code&amp;gt;flags_be&amp;lt;/code&amp;gt; are not yet decoded into human labels. They are likely bitpacked and depend on record class.&lt;br /&gt;
* Some records use slightly different header shapes (longer/shorter), so the parser should be resilient: find timestamp  test for nearby &amp;lt;code&amp;gt;name_off&amp;lt;/code&amp;gt; that points into the string table → treat as candidate record.&lt;br /&gt;
* The string table location and size are variable; rely on heuristics (max-count window) or header-specified offsets that may exist in other product files.&lt;/div&gt;</summary>
		<author><name>Raion</name></author>
	</entry>
</feed>