Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / c23b933d5f8be9b51a1d22e717c0311f65f87dcd / . / docs / Passes.html
blob: 85292e37412247b8237fd565e5ddd57809af98c7 [file] [log] [blame]
NAKAMURA Takumib5f74912011-04-21 01:32:19 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]2 "http://www.w3.org/TR/html4/strict.dtd">
3<html>
4<head>
5 <title>LLVM's Analysis and Transform Passes</title>
Daniel Dunbaradea4972012-04-19 20:20:34 +0000[diff] [blame]6 <link rel="stylesheet" href="_static/llvm.css" type="text/css">
Gordon Henriksenddaa61d2007-10-25 08:58:56 +0000[diff] [blame]7 <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]8</head>
9<body>
10
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]11<!--
12
13If Passes.html is up to date, the following "one-liner" should print
14an empty diff.
15
16egrep -e '^<tr><td><a href="#.*">-.*</a></td><td>.*</td></tr>$' \
17 -e '^ <a name=".*">.*</a>$' < Passes.html >html; \
18perl >help <<'EOT' && diff -u help html; rm -f help html
19open HTML, "<Passes.html" or die "open: Passes.html: $!\n";
20while (<HTML>) {
21 m:^<tr><td><a href="#(.*)">-.*</a></td><td>.*</td></tr>$: or next;
22 $order{$1} = sprintf("%03d", 1 + int %order);
23}
Gordon Henriksenddaa61d2007-10-25 08:58:56 +0000[diff] [blame]24open HELP, "../Release/bin/opt -help|" or die "open: opt -help: $!\n";
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]25while (<HELP>) {
26 m:^ -([^ ]+) +- (.*)$: or next;
27 my $o = $order{$1};
28 $o = "000" unless defined $o;
29 push @x, "$o<tr><td><a href=\"#$1\">-$1</a></td><td>$2</td></tr>\n";
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]30 push @y, "$o <a name=\"$1\">-$1: $2</a>\n";
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]31}
32@x = map { s/^\d\d\d//; $_ } sort @x;
33@y = map { s/^\d\d\d//; $_ } sort @y;
34print @x, @y;
35EOT
36
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]37This (real) one-liner can also be helpful when converting comments to HTML:
38
39perl -e '$/ = undef; for (split(/\n/, <>)) { s:^ *///? ?::; print " <p>\n" if !$on && $_ =~ /\S/; print " </p>\n" if $on && $_ =~ /^\s*$/; print " $_\n"; $on = ($_ =~ /\S/); } print " </p>\n" if $on'
40
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]41 -->
42
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]43<h1>LLVM's Analysis and Transform Passes</h1>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]44
45<ol>
46 <li><a href="#intro">Introduction</a></li>
47 <li><a href="#analyses">Analysis Passes</a>
48 <li><a href="#transforms">Transform Passes</a></li>
49 <li><a href="#utilities">Utility Passes</a></li>
50</ol>
51
52<div class="doc_author">
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]53 <p>Written by <a href="mailto:rspencer@x10sys.com">Reid Spencer</a>
54 and Gordon Henriksen</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]55</div>
56
57<!-- ======================================================================= -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]58<h2><a name="intro">Introduction</a></h2>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]59<div>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]60 <p>This document serves as a high level summary of the optimization features
61 that LLVM provides. Optimizations are implemented as Passes that traverse some
62 portion of a program to either collect information or transform the program.
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]63 The table below divides the passes that LLVM provides into three categories.
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]64 Analysis passes compute information that other passes can use or for debugging
65 or program visualization purposes. Transform passes can use (or invalidate)
66 the analysis passes. Transform passes all mutate the program in some way.
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]67 Utility passes provides some utility but don't otherwise fit categorization.
Gabor Greif04367bf2007-07-06 22:07:22 +0000[diff] [blame]68 For example passes to extract functions to bitcode or write a module to
69 bitcode are neither analysis nor transform passes.
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]70 <p>The table below provides a quick summary of each pass and links to the more
71 complete pass description later in the document.</p>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]72
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]73<table>
Gordon Henriksenddaa61d2007-10-25 08:58:56 +0000[diff] [blame]74<tr><th colspan="2"><b>ANALYSIS PASSES</b></th></tr>
75<tr><th>Option</th><th>Name</th></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]76<tr><td><a href="#aa-eval">-aa-eval</a></td><td>Exhaustive Alias Analysis Precision Evaluator</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]77<tr><td><a href="#basicaa">-basicaa</a></td><td>Basic Alias Analysis (stateless AA impl)</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]78<tr><td><a href="#basiccg">-basiccg</a></td><td>Basic CallGraph Construction</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]79<tr><td><a href="#count-aa">-count-aa</a></td><td>Count Alias Analysis Query Responses</td></tr>
80<tr><td><a href="#debug-aa">-debug-aa</a></td><td>AA use debugger</td></tr>
81<tr><td><a href="#domfrontier">-domfrontier</a></td><td>Dominance Frontier Construction</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]82<tr><td><a href="#domtree">-domtree</a></td><td>Dominator Tree Construction</td></tr>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]83<tr><td><a href="#dot-callgraph">-dot-callgraph</a></td><td>Print Call Graph to 'dot' file</td></tr>
84<tr><td><a href="#dot-cfg">-dot-cfg</a></td><td>Print CFG of function to 'dot' file</td></tr>
85<tr><td><a href="#dot-cfg-only">-dot-cfg-only</a></td><td>Print CFG of function to 'dot' file (with no function bodies)</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]86<tr><td><a href="#dot-dom">-dot-dom</a></td><td>Print dominance tree of function to 'dot' file</td></tr>
87<tr><td><a href="#dot-dom-only">-dot-dom-only</a></td><td>Print dominance tree of function to 'dot' file (with no function bodies)</td></tr>
88<tr><td><a href="#dot-postdom">-dot-postdom</a></td><td>Print postdominance tree of function to 'dot' file</td></tr>
89<tr><td><a href="#dot-postdom-only">-dot-postdom-only</a></td><td>Print postdominance tree of function to 'dot' file (with no function bodies)</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]90<tr><td><a href="#globalsmodref-aa">-globalsmodref-aa</a></td><td>Simple mod/ref analysis for globals</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]91<tr><td><a href="#instcount">-instcount</a></td><td>Counts the various types of Instructions</td></tr>
92<tr><td><a href="#intervals">-intervals</a></td><td>Interval Partition Construction</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]93<tr><td><a href="#iv-users">-iv-users</a></td><td>Induction Variable Users</td></tr>
94<tr><td><a href="#lazy-value-info">-lazy-value-info</a></td><td>Lazy Value Information Analysis</td></tr>
95<tr><td><a href="#lda">-lda</a></td><td>Loop Dependence Analysis</td></tr>
96<tr><td><a href="#libcall-aa">-libcall-aa</a></td><td>LibCall Alias Analysis</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]97<tr><td><a href="#lint">-lint</a></td><td>Statically lint-checks LLVM IR</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]98<tr><td><a href="#loops">-loops</a></td><td>Natural Loop Information</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]99<tr><td><a href="#memdep">-memdep</a></td><td>Memory Dependence Analysis</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]100<tr><td><a href="#module-debuginfo">-module-debuginfo</a></td><td>Decodes module-level debug info</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]101<tr><td><a href="#no-aa">-no-aa</a></td><td>No Alias Analysis (always returns 'may' alias)</td></tr>
102<tr><td><a href="#no-profile">-no-profile</a></td><td>No Profile Information</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]103<tr><td><a href="#postdomtree">-postdomtree</a></td><td>Post-Dominator Tree Construction</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]104<tr><td><a href="#print-alias-sets">-print-alias-sets</a></td><td>Alias Set Printer</td></tr>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]105<tr><td><a href="#print-callgraph">-print-callgraph</a></td><td>Print a call graph</td></tr>
106<tr><td><a href="#print-callgraph-sccs">-print-callgraph-sccs</a></td><td>Print SCCs of the Call Graph</td></tr>
107<tr><td><a href="#print-cfg-sccs">-print-cfg-sccs</a></td><td>Print SCCs of each function CFG</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]108<tr><td><a href="#print-dbginfo">-print-dbginfo</a></td><td>Print debug info in human readable form</td></tr>
109<tr><td><a href="#print-dom-info">-print-dom-info</a></td><td>Dominator Info Printer</td></tr>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]110<tr><td><a href="#print-externalfnconstants">-print-externalfnconstants</a></td><td>Print external fn callsites passed constants</td></tr>
111<tr><td><a href="#print-function">-print-function</a></td><td>Print function to stderr</td></tr>
112<tr><td><a href="#print-module">-print-module</a></td><td>Print module to stderr</td></tr>
113<tr><td><a href="#print-used-types">-print-used-types</a></td><td>Find Used Types</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]114<tr><td><a href="#profile-estimator">-profile-estimator</a></td><td>Estimate profiling information</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]115<tr><td><a href="#profile-loader">-profile-loader</a></td><td>Load profile information from llvmprof.out</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]116<tr><td><a href="#profile-verifier">-profile-verifier</a></td><td>Verify profiling information</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]117<tr><td><a href="#regions">-regions</a></td><td>Detect single entry single exit regions</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]118<tr><td><a href="#scalar-evolution">-scalar-evolution</a></td><td>Scalar Evolution Analysis</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]119<tr><td><a href="#scev-aa">-scev-aa</a></td><td>ScalarEvolution-based Alias Analysis</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]120<tr><td><a href="#targetdata">-targetdata</a></td><td>Target Data Layout</td></tr>
121
122
Gordon Henriksenddaa61d2007-10-25 08:58:56 +0000[diff] [blame]123<tr><th colspan="2"><b>TRANSFORM PASSES</b></th></tr>
124<tr><th>Option</th><th>Name</th></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]125<tr><td><a href="#adce">-adce</a></td><td>Aggressive Dead Code Elimination</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]126<tr><td><a href="#always-inline">-always-inline</a></td><td>Inliner for always_inline functions</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]127<tr><td><a href="#argpromotion">-argpromotion</a></td><td>Promote 'by reference' arguments to scalars</td></tr>
Hal Finkelde5e5ec2012-02-01 03:51:43 +0000[diff] [blame]128<tr><td><a href="#bb-vectorize">-bb-vectorize</a></td><td>Combine instructions to form vector instructions within basic blocks</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]129<tr><td><a href="#block-placement">-block-placement</a></td><td>Profile Guided Basic Block Placement</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]130<tr><td><a href="#break-crit-edges">-break-crit-edges</a></td><td>Break critical edges in CFG</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]131<tr><td><a href="#codegenprepare">-codegenprepare</a></td><td>Optimize for code generation</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]132<tr><td><a href="#constmerge">-constmerge</a></td><td>Merge Duplicate Global Constants</td></tr>
133<tr><td><a href="#constprop">-constprop</a></td><td>Simple constant propagation</td></tr>
134<tr><td><a href="#dce">-dce</a></td><td>Dead Code Elimination</td></tr>
135<tr><td><a href="#deadargelim">-deadargelim</a></td><td>Dead Argument Elimination</td></tr>
136<tr><td><a href="#deadtypeelim">-deadtypeelim</a></td><td>Dead Type Elimination</td></tr>
137<tr><td><a href="#die">-die</a></td><td>Dead Instruction Elimination</td></tr>
138<tr><td><a href="#dse">-dse</a></td><td>Dead Store Elimination</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]139<tr><td><a href="#functionattrs">-functionattrs</a></td><td>Deduce function attributes</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]140<tr><td><a href="#globaldce">-globaldce</a></td><td>Dead Global Elimination</td></tr>
141<tr><td><a href="#globalopt">-globalopt</a></td><td>Global Variable Optimizer</td></tr>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]142<tr><td><a href="#gvn">-gvn</a></td><td>Global Value Numbering</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]143<tr><td><a href="#indvars">-indvars</a></td><td>Canonicalize Induction Variables</td></tr>
144<tr><td><a href="#inline">-inline</a></td><td>Function Integration/Inlining</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]145<tr><td><a href="#insert-edge-profiling">-insert-edge-profiling</a></td><td>Insert instrumentation for edge profiling</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]146<tr><td><a href="#insert-optimal-edge-profiling">-insert-optimal-edge-profiling</a></td><td>Insert optimal instrumentation for edge profiling</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]147<tr><td><a href="#instcombine">-instcombine</a></td><td>Combine redundant instructions</td></tr>
148<tr><td><a href="#internalize">-internalize</a></td><td>Internalize Global Symbols</td></tr>
149<tr><td><a href="#ipconstprop">-ipconstprop</a></td><td>Interprocedural constant propagation</td></tr>
150<tr><td><a href="#ipsccp">-ipsccp</a></td><td>Interprocedural Sparse Conditional Constant Propagation</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]151<tr><td><a href="#jump-threading">-jump-threading</a></td><td>Jump Threading</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]152<tr><td><a href="#lcssa">-lcssa</a></td><td>Loop-Closed SSA Form Pass</td></tr>
153<tr><td><a href="#licm">-licm</a></td><td>Loop Invariant Code Motion</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]154<tr><td><a href="#loop-deletion">-loop-deletion</a></td><td>Delete dead loops</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]155<tr><td><a href="#loop-extract">-loop-extract</a></td><td>Extract loops into new functions</td></tr>
156<tr><td><a href="#loop-extract-single">-loop-extract-single</a></td><td>Extract at most one loop into a new function</td></tr>
157<tr><td><a href="#loop-reduce">-loop-reduce</a></td><td>Loop Strength Reduction</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]158<tr><td><a href="#loop-rotate">-loop-rotate</a></td><td>Rotate Loops</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]159<tr><td><a href="#loop-simplify">-loop-simplify</a></td><td>Canonicalize natural loops</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]160<tr><td><a href="#loop-unroll">-loop-unroll</a></td><td>Unroll loops</td></tr>
161<tr><td><a href="#loop-unswitch">-loop-unswitch</a></td><td>Unswitch loops</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]162<tr><td><a href="#loweratomic">-loweratomic</a></td><td>Lower atomic intrinsics to non-atomic form</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]163<tr><td><a href="#lowerinvoke">-lowerinvoke</a></td><td>Lower invoke and unwind, for unwindless code generators</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]164<tr><td><a href="#lowerswitch">-lowerswitch</a></td><td>Lower SwitchInst's to branches</td></tr>
165<tr><td><a href="#mem2reg">-mem2reg</a></td><td>Promote Memory to Register</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]166<tr><td><a href="#memcpyopt">-memcpyopt</a></td><td>MemCpy Optimization</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]167<tr><td><a href="#mergefunc">-mergefunc</a></td><td>Merge Functions</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]168<tr><td><a href="#mergereturn">-mergereturn</a></td><td>Unify function exit nodes</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]169<tr><td><a href="#partial-inliner">-partial-inliner</a></td><td>Partial Inliner</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]170<tr><td><a href="#prune-eh">-prune-eh</a></td><td>Remove unused exception handling info</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]171<tr><td><a href="#reassociate">-reassociate</a></td><td>Reassociate expressions</td></tr>
172<tr><td><a href="#reg2mem">-reg2mem</a></td><td>Demote all values to stack slots</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]173<tr><td><a href="#scalarrepl">-scalarrepl</a></td><td>Scalar Replacement of Aggregates (DT)</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]174<tr><td><a href="#sccp">-sccp</a></td><td>Sparse Conditional Constant Propagation</td></tr>
175<tr><td><a href="#simplify-libcalls">-simplify-libcalls</a></td><td>Simplify well-known library calls</td></tr>
176<tr><td><a href="#simplifycfg">-simplifycfg</a></td><td>Simplify the CFG</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]177<tr><td><a href="#sink">-sink</a></td><td>Code sinking</td></tr>
178<tr><td><a href="#sretpromotion">-sretpromotion</a></td><td>Promote sret arguments to multiple ret values</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]179<tr><td><a href="#strip">-strip</a></td><td>Strip all symbols from a module</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]180<tr><td><a href="#strip-dead-debug-info">-strip-dead-debug-info</a></td><td>Strip debug info for unused symbols</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]181<tr><td><a href="#strip-dead-prototypes">-strip-dead-prototypes</a></td><td>Strip Unused Function Prototypes</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]182<tr><td><a href="#strip-debug-declare">-strip-debug-declare</a></td><td>Strip all llvm.dbg.declare intrinsics</td></tr>
183<tr><td><a href="#strip-nondebug">-strip-nondebug</a></td><td>Strip all symbols, except dbg symbols, from a module</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]184<tr><td><a href="#tailcallelim">-tailcallelim</a></td><td>Tail Call Elimination</td></tr>
185<tr><td><a href="#tailduplicate">-tailduplicate</a></td><td>Tail Duplication</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]186
187
Gordon Henriksenddaa61d2007-10-25 08:58:56 +0000[diff] [blame]188<tr><th colspan="2"><b>UTILITY PASSES</b></th></tr>
189<tr><th>Option</th><th>Name</th></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]190<tr><td><a href="#deadarghaX0r">-deadarghaX0r</a></td><td>Dead Argument Hacking (BUGPOINT USE ONLY; DO NOT USE)</td></tr>
191<tr><td><a href="#extract-blocks">-extract-blocks</a></td><td>Extract Basic Blocks From Module (for bugpoint use)</td></tr>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]192<tr><td><a href="#instnamer">-instnamer</a></td><td>Assign names to anonymous instructions</td></tr>
Gordon Henriksen90a52142007-11-05 02:05:35 +0000[diff] [blame]193<tr><td><a href="#preverify">-preverify</a></td><td>Preliminary module verification</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]194<tr><td><a href="#verify">-verify</a></td><td>Module Verifier</td></tr>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]195<tr><td><a href="#view-cfg">-view-cfg</a></td><td>View CFG of function</td></tr>
196<tr><td><a href="#view-cfg-only">-view-cfg-only</a></td><td>View CFG of function (with no function bodies)</td></tr>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]197<tr><td><a href="#view-dom">-view-dom</a></td><td>View dominance tree of function</td></tr>
198<tr><td><a href="#view-dom-only">-view-dom-only</a></td><td>View dominance tree of function (with no function bodies)</td></tr>
199<tr><td><a href="#view-postdom">-view-postdom</a></td><td>View postdominance tree of function</td></tr>
200<tr><td><a href="#view-postdom-only">-view-postdom-only</a></td><td>View postdominance tree of function (with no function bodies)</td></tr>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]201</table>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]202
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]203</div>
204
205<!-- ======================================================================= -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]206<h2><a name="analyses">Analysis Passes</a></h2>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]207<div>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]208 <p>This section describes the LLVM Analysis Passes.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]209
210<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]211<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]212 <a name="aa-eval">-aa-eval: Exhaustive Alias Analysis Precision Evaluator</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]213</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]214<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]215 <p>This is a simple N^2 alias analysis accuracy evaluator.
216 Basically, for each function in the program, it simply queries to see how the
217 alias analysis implementation answers alias queries between each pair of
218 pointers in the function.</p>
219
220 <p>This is inspired and adapted from code by: Naveen Neelakantam, Francesco
221 Spadini, and Wojciech Stryjewski.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]222</div>
223
224<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]225<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]226 <a name="basicaa">-basicaa: Basic Alias Analysis (stateless AA impl)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]227</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]228<div>
Eli Friedman146af5a2011-10-27 22:32:13 +0000[diff] [blame]229 <p>A basic alias analysis pass that implements identities (two different
230 globals cannot alias, etc), but does no stateful analysis.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]231</div>
232
233<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]234<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]235 <a name="basiccg">-basiccg: Basic CallGraph Construction</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]236</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]237<div>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]238 <p>Yet to be written.</p>
239</div>
240
241<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]242<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]243 <a name="count-aa">-count-aa: Count Alias Analysis Query Responses</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]244</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]245<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]246 <p>
247 A pass which can be used to count how many alias queries
248 are being made and how the alias analysis implementation being used responds.
249 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]250</div>
251
252<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]253<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]254 <a name="debug-aa">-debug-aa: AA use debugger</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]255</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]256<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]257 <p>
258 This simple pass checks alias analysis users to ensure that if they
259 create a new value, they do not query AA without informing it of the value.
260 It acts as a shim over any other AA pass you want.
261 </p>
262
263 <p>
264 Yes keeping track of every value in the program is expensive, but this is
265 a debugging pass.
266 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]267</div>
268
269<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]270<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]271 <a name="domfrontier">-domfrontier: Dominance Frontier Construction</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]272</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]273<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]274 <p>
275 This pass is a simple dominator construction algorithm for finding forward
276 dominator frontiers.
277 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]278</div>
279
280<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]281<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]282 <a name="domtree">-domtree: Dominator Tree Construction</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]283</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]284<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]285 <p>
286 This pass is a simple dominator construction algorithm for finding forward
287 dominators.
288 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]289</div>
290
291<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]292<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]293 <a name="dot-callgraph">-dot-callgraph: Print Call Graph to 'dot' file</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]294</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]295<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]296 <p>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]297 This pass, only available in <code>opt</code>, prints the call graph into a
298 <code>.dot</code> graph. This graph can then be processed with the "dot" tool
299 to convert it to postscript or some other suitable format.
300 </p>
301</div>
302
303<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]304<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]305 <a name="dot-cfg">-dot-cfg: Print CFG of function to 'dot' file</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]306</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]307<div>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]308 <p>
309 This pass, only available in <code>opt</code>, prints the control flow graph
310 into a <code>.dot</code> graph. This graph can then be processed with the
311 "dot" tool to convert it to postscript or some other suitable format.
312 </p>
313</div>
314
315<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]316<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]317 <a name="dot-cfg-only">-dot-cfg-only: Print CFG of function to 'dot' file (with no function bodies)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]318</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]319<div>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]320 <p>
321 This pass, only available in <code>opt</code>, prints the control flow graph
322 into a <code>.dot</code> graph, omitting the function bodies. This graph can
323 then be processed with the "dot" tool to convert it to postscript or some
324 other suitable format.
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]325 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]326</div>
327
328<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]329<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]330 <a name="dot-dom">-dot-dom: Print dominance tree of function to 'dot' file</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]331</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]332<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]333 <p>
334 This pass, only available in <code>opt</code>, prints the dominator tree
335 into a <code>.dot</code> graph. This graph can then be processed with the
336 "dot" tool to convert it to postscript or some other suitable format.
337 </p>
338</div>
339
340<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]341<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]342 <a name="dot-dom-only">-dot-dom-only: Print dominance tree of function to 'dot' file (with no function bodies)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]343</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]344<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]345 <p>
346 This pass, only available in <code>opt</code>, prints the dominator tree
347 into a <code>.dot</code> graph, omitting the function bodies. This graph can
348 then be processed with the "dot" tool to convert it to postscript or some
349 other suitable format.
350 </p>
351</div>
352
353<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]354<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]355 <a name="dot-postdom">-dot-postdom: Print postdominance tree of function to 'dot' file</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]356</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]357<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]358 <p>
359 This pass, only available in <code>opt</code>, prints the post dominator tree
360 into a <code>.dot</code> graph. This graph can then be processed with the
361 "dot" tool to convert it to postscript or some other suitable format.
362 </p>
363</div>
364
365<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]366<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]367 <a name="dot-postdom-only">-dot-postdom-only: Print postdominance tree of function to 'dot' file (with no function bodies)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]368</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]369<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]370 <p>
371 This pass, only available in <code>opt</code>, prints the post dominator tree
372 into a <code>.dot</code> graph, omitting the function bodies. This graph can
373 then be processed with the "dot" tool to convert it to postscript or some
374 other suitable format.
375 </p>
376</div>
377
378<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]379<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]380 <a name="globalsmodref-aa">-globalsmodref-aa: Simple mod/ref analysis for globals</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]381</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]382<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]383 <p>
384 This simple pass provides alias and mod/ref information for global values
385 that do not have their address taken, and keeps track of whether functions
386 read or write memory (are "pure"). For this simple (but very common) case,
387 we can provide pretty accurate and useful information.
388 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]389</div>
390
391<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]392<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]393 <a name="instcount">-instcount: Counts the various types of Instructions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]394</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]395<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]396 <p>
397 This pass collects the count of all instructions and reports them
398 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]399</div>
400
401<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]402<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]403 <a name="intervals">-intervals: Interval Partition Construction</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]404</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]405<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]406 <p>
407 This analysis calculates and represents the interval partition of a function,
408 or a preexisting interval partition.
409 </p>
410
411 <p>
412 In this way, the interval partition may be used to reduce a flow graph down
413 to its degenerate single node interval partition (unless it is irreducible).
414 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]415</div>
416
417<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]418<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]419 <a name="iv-users">-iv-users: Induction Variable Users</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]420</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]421<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]422 <p>Bookkeeping for "interesting" users of expressions computed from
423 induction variables.</p>
424</div>
425
426<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]427<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]428 <a name="lazy-value-info">-lazy-value-info: Lazy Value Information Analysis</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]429</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]430<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]431 <p>Interface for lazy computation of value constraint information.</p>
432</div>
433
434<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]435<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]436 <a name="lda">-lda: Loop Dependence Analysis</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]437</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]438<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]439 <p>Loop dependence analysis framework, which is used to detect dependences in
440 memory accesses in loops.</p>
441</div>
442
443<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]444<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]445 <a name="libcall-aa">-libcall-aa: LibCall Alias Analysis</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]446</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]447<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]448 <p>LibCall Alias Analysis.</p>
449</div>
450
451<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]452<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]453 <a name="lint">-lint: Statically lint-checks LLVM IR</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]454</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]455<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]456 <p>This pass statically checks for common and easily-identified constructs
457 which produce undefined or likely unintended behavior in LLVM IR.</p>
458
459 <p>It is not a guarantee of correctness, in two ways. First, it isn't
460 comprehensive. There are checks which could be done statically which are
461 not yet implemented. Some of these are indicated by TODO comments, but
462 those aren't comprehensive either. Second, many conditions cannot be
463 checked statically. This pass does no dynamic instrumentation, so it
464 can't check for all possible problems.</p>
465
466 <p>Another limitation is that it assumes all code will be executed. A store
467 through a null pointer in a basic block which is never reached is harmless,
468 but this pass will warn about it anyway.</p>
469
470 <p>Optimization passes may make conditions that this pass checks for more or
471 less obvious. If an optimization pass appears to be introducing a warning,
472 it may be that the optimization pass is merely exposing an existing
473 condition in the code.</p>
474
475 <p>This code may be run before instcombine. In many cases, instcombine checks
476 for the same kinds of things and turns instructions with undefined behavior
477 into unreachable (or equivalent). Because of this, this pass makes some
478 effort to look through bitcasts and so on.
479 </p>
480</div>
481
482<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]483<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]484 <a name="loops">-loops: Natural Loop Information</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]485</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]486<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]487 <p>
488 This analysis is used to identify natural loops and determine the loop depth
489 of various nodes of the CFG. Note that the loops identified may actually be
490 several natural loops that share the same header node... not just a single
491 natural loop.
492 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]493</div>
494
495<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]496<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]497 <a name="memdep">-memdep: Memory Dependence Analysis</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]498</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]499<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]500 <p>
501 An analysis that determines, for a given memory operation, what preceding
502 memory operations it depends on. It builds on alias analysis information, and
503 tries to provide a lazy, caching interface to a common kind of alias
504 information query.
505 </p>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]506</div>
507
508<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]509<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]510 <a name="module-debuginfo">-module-debuginfo: Decodes module-level debug info</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]511</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]512<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]513 <p>This pass decodes the debug info metadata in a module and prints in a
514 (sufficiently-prepared-) human-readable form.
515
516 For example, run this pass from opt along with the -analyze option, and
517 it'll print to standard output.
518 </p>
519</div>
520
521<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]522<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]523 <a name="no-aa">-no-aa: No Alias Analysis (always returns 'may' alias)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]524</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]525<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]526 <p>
Eli Friedman146af5a2011-10-27 22:32:13 +0000[diff] [blame]527 This is the default implementation of the Alias Analysis interface. It always
528 returns "I don't know" for alias queries. NoAA is unlike other alias analysis
529 implementations, in that it does not chain to a previous analysis. As such it
530 doesn't follow many of the rules that other alias analyses must.
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]531 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]532</div>
533
534<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]535<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]536 <a name="no-profile">-no-profile: No Profile Information</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]537</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]538<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]539 <p>
540 The default "no profile" implementation of the abstract
541 <code>ProfileInfo</code> interface.
542 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]543</div>
544
545<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]546<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]547 <a name="postdomfrontier">-postdomfrontier: Post-Dominance Frontier Construction</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]548</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]549<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]550 <p>
551 This pass is a simple post-dominator construction algorithm for finding
552 post-dominator frontiers.
553 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]554</div>
555
556<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]557<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]558 <a name="postdomtree">-postdomtree: Post-Dominator Tree Construction</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]559</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]560<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]561 <p>
562 This pass is a simple post-dominator construction algorithm for finding
563 post-dominators.
564 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]565</div>
566
567<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]568<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]569 <a name="print-alias-sets">-print-alias-sets: Alias Set Printer</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]570</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]571<div>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]572 <p>Yet to be written.</p>
573</div>
574
575<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]576<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]577 <a name="print-callgraph">-print-callgraph: Print a call graph</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]578</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]579<div>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]580 <p>
581 This pass, only available in <code>opt</code>, prints the call graph to
Dan Gohman52fdaed2010-08-20 01:03:44 +0000[diff] [blame]582 standard error in a human-readable form.
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]583 </p>
584</div>
585
586<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]587<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]588 <a name="print-callgraph-sccs">-print-callgraph-sccs: Print SCCs of the Call Graph</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]589</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]590<div>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]591 <p>
592 This pass, only available in <code>opt</code>, prints the SCCs of the call
Dan Gohman52fdaed2010-08-20 01:03:44 +0000[diff] [blame]593 graph to standard error in a human-readable form.
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]594 </p>
595</div>
596
597<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]598<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]599 <a name="print-cfg-sccs">-print-cfg-sccs: Print SCCs of each function CFG</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]600</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]601<div>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]602 <p>
603 This pass, only available in <code>opt</code>, prints the SCCs of each
Dan Gohman52fdaed2010-08-20 01:03:44 +0000[diff] [blame]604 function CFG to standard error in a human-readable form.
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]605 </p>
606</div>
607
608<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]609<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]610 <a name="print-dbginfo">-print-dbginfo: Print debug info in human readable form</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]611</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]612<div>
Peter Collingbournec3086ba2010-08-06 02:13:25 +0000[diff] [blame]613 <p>Pass that prints instructions, and associated debug info:</p>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]614 <ul>
615
616 <li>source/line/col information</li>
617 <li>original variable name</li>
618 <li>original type name</li>
619 </ul>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]620</div>
621
622<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]623<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]624 <a name="print-dom-info">-print-dom-info: Dominator Info Printer</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]625</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]626<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]627 <p>Dominator Info Printer.</p>
628</div>
629
630<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]631<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]632 <a name="print-externalfnconstants">-print-externalfnconstants: Print external fn callsites passed constants</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]633</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]634<div>
Duncan Sands3ee8fc92008-09-23 12:47:39 +0000[diff] [blame]635 <p>
636 This pass, only available in <code>opt</code>, prints out call sites to
637 external functions that are called with constant arguments. This can be
638 useful when looking for standard library functions we should constant fold
639 or handle in alias analyses.
640 </p>
641</div>
642
643<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]644<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]645 <a name="print-function">-print-function: Print function to stderr</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]646</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]647<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]648 <p>
649 The <code>PrintFunctionPass</code> class is designed to be pipelined with
650 other <code>FunctionPass</code>es, and prints out the functions of the module
651 as they are processed.
652 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]653</div>
654
655<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]656<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]657 <a name="print-module">-print-module: Print module to stderr</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]658</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]659<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]660 <p>
661 This pass simply prints out the entire module when it is executed.
662 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]663</div>
664
665<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]666<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]667 <a name="print-used-types">-print-used-types: Find Used Types</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]668</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]669<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]670 <p>
671 This pass is used to seek out all of the types in use by the program. Note
672 that this analysis explicitly does not include types only used by the symbol
673 table.
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]674</div>
675
676<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]677<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]678 <a name="profile-estimator">-profile-estimator: Estimate profiling information</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]679</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]680<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]681 <p>Profiling information that estimates the profiling information
682 in a very crude and unimaginative way.
683 </p>
684</div>
685
686<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]687<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]688 <a name="profile-loader">-profile-loader: Load profile information from llvmprof.out</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]689</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]690<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]691 <p>
692 A concrete implementation of profiling information that loads the information
693 from a profile dump file.
694 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]695</div>
696
697<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]698<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]699 <a name="profile-verifier">-profile-verifier: Verify profiling information</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]700</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]701<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]702 <p>Pass that checks profiling information for plausibility.</p>
703</div>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]704<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]705 <a name="regions">-regions: Detect single entry single exit regions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]706</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]707<div>
Tobias Grosserf96b0062010-07-22 07:46:31 +0000[diff] [blame]708 <p>
709 The <code>RegionInfo</code> pass detects single entry single exit regions in a
710 function, where a region is defined as any subgraph that is connected to the
711 remaining graph at only two spots. Furthermore, an hierarchical region tree is
712 built.
713 </p>
714</div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]715
716<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]717<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]718 <a name="scalar-evolution">-scalar-evolution: Scalar Evolution Analysis</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]719</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]720<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]721 <p>
722 The <code>ScalarEvolution</code> analysis can be used to analyze and
723 catagorize scalar expressions in loops. It specializes in recognizing general
724 induction variables, representing them with the abstract and opaque
725 <code>SCEV</code> class. Given this analysis, trip counts of loops and other
726 important properties can be obtained.
727 </p>
728
729 <p>
730 This analysis is primarily useful for induction variable substitution and
731 strength reduction.
732 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]733</div>
734
735<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]736<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]737 <a name="scev-aa">-scev-aa: ScalarEvolution-based Alias Analysis</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]738</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]739<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]740 <p>Simple alias analysis implemented in terms of ScalarEvolution queries.
741
742 This differs from traditional loop dependence analysis in that it tests
743 for dependencies within a single iteration of a loop, rather than
744 dependencies between different iterations.
745
746 ScalarEvolution has a more complete understanding of pointer arithmetic
747 than BasicAliasAnalysis' collection of ad-hoc analyses.
748 </p>
749</div>
750
751<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]752<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]753 <a name="targetdata">-targetdata: Target Data Layout</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]754</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]755<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]756 <p>Provides other passes access to information on how the size and alignment
Sylvestre Ledruc8e41c52012-07-23 08:51:15 +0000[diff] [blame]757 required by the target ABI for various data types.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]758</div>
759
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]760</div>
761
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]762<!-- ======================================================================= -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]763<h2><a name="transforms">Transform Passes</a></h2>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]764<div>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]765 <p>This section describes the LLVM Transform Passes.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]766
767<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]768<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]769 <a name="adce">-adce: Aggressive Dead Code Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]770</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]771<div>
Reid Spenceraf4af3a2007-03-27 02:49:31 +0000[diff] [blame]772 <p>ADCE aggressively tries to eliminate code. This pass is similar to
773 <a href="#dce">DCE</a> but it assumes that values are dead until proven
774 otherwise. This is similar to <a href="#sccp">SCCP</a>, except applied to
775 the liveness of values.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]776</div>
777
778<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]779<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]780 <a name="always-inline">-always-inline: Inliner for always_inline functions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]781</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]782<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]783 <p>A custom inliner that handles only functions that are marked as
784 "always inline".</p>
785</div>
786
787<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]788<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]789 <a name="argpromotion">-argpromotion: Promote 'by reference' arguments to scalars</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]790</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]791<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]792 <p>
793 This pass promotes "by reference" arguments to be "by value" arguments. In
794 practice, this means looking for internal functions that have pointer
795 arguments. If it can prove, through the use of alias analysis, that an
796 argument is *only* loaded, then it can pass the value into the function
797 instead of the address of the value. This can cause recursive simplification
798 of code and lead to the elimination of allocas (especially in C++ template
799 code like the STL).
800 </p>
801
802 <p>
803 This pass also handles aggregate arguments that are passed into a function,
804 scalarizing them if the elements of the aggregate are only loaded. Note that
805 it refuses to scalarize aggregates which would require passing in more than
806 three operands to the function, because passing thousands of operands for a
807 large array or structure is unprofitable!
808 </p>
809
810 <p>
811 Note that this transformation could also be done for arguments that are only
812 stored to (returning the value instead), but does not currently. This case
813 would be best handled when and if LLVM starts supporting multiple return
814 values from functions.
815 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]816</div>
817
818<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]819<h3>
Hal Finkelde5e5ec2012-02-01 03:51:43 +0000[diff] [blame]820 <a name="bb-vectorize">-bb-vectorize: Basic-Block Vectorization</a>
821</h3>
822<div>
823 <p>This pass combines instructions inside basic blocks to form vector
824 instructions. It iterates over each basic block, attempting to pair
825 compatible instructions, repeating this process until no additional
826 pairs are selected for vectorization. When the outputs of some pair
827 of compatible instructions are used as inputs by some other pair of
828 compatible instructions, those pairs are part of a potential
829 vectorization chain. Instruction pairs are only fused into vector
830 instructions when they are part of a chain longer than some
831 threshold length. Moreover, the pass attempts to find the best
832 possible chain for each pair of compatible instructions. These
833 heuristics are intended to prevent vectorization in cases where
834 it would not yield a performance increase of the resulting code.
835 </p>
836</div>
837
838<!-------------------------------------------------------------------------- -->
839<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]840 <a name="block-placement">-block-placement: Profile Guided Basic Block Placement</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]841</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]842<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]843 <p>This pass is a very simple profile guided basic block placement algorithm.
844 The idea is to put frequently executed blocks together at the start of the
845 function and hopefully increase the number of fall-through conditional
846 branches. If there is no profile information for a particular function, this
847 pass basically orders blocks in depth-first order.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]848</div>
849
850<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]851<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]852 <a name="break-crit-edges">-break-crit-edges: Break critical edges in CFG</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]853</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]854<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]855 <p>
856 Break all of the critical edges in the CFG by inserting a dummy basic block.
857 It may be "required" by passes that cannot deal with critical edges. This
858 transformation obviously invalidates the CFG, but can update forward dominator
859 (set, immediate dominators, tree, and frontier) information.
860 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]861</div>
862
863<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]864<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]865 <a name="codegenprepare">-codegenprepare: Optimize for code generation</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]866</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]867<div>
Gordon Henriksena8a118b2008-05-08 17:46:35 +0000[diff] [blame]868 This pass munges the code in the input function to better prepare it for
869 SelectionDAG-based code generation. This works around limitations in it's
870 basic-block-at-a-time approach. It should eventually be removed.
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]871</div>
872
873<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]874<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]875 <a name="constmerge">-constmerge: Merge Duplicate Global Constants</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]876</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]877<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]878 <p>
879 Merges duplicate global constants together into a single constant that is
880 shared. This is useful because some passes (ie TraceValues) insert a lot of
881 string constants into the program, regardless of whether or not an existing
882 string is available.
883 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]884</div>
885
886<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]887<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]888 <a name="constprop">-constprop: Simple constant propagation</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]889</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]890<div>
Reid Spenceraf4af3a2007-03-27 02:49:31 +0000[diff] [blame]891 <p>This file implements constant propagation and merging. It looks for
892 instructions involving only constant operands and replaces them with a
Gordon Henriksenddaa61d2007-10-25 08:58:56 +0000[diff] [blame]893 constant value instead of an instruction. For example:</p>
894 <blockquote><pre>add i32 1, 2</pre></blockquote>
895 <p>becomes</p>
896 <blockquote><pre>i32 3</pre></blockquote>
Reid Spenceraf4af3a2007-03-27 02:49:31 +0000[diff] [blame]897 <p>NOTE: this pass has a habit of making definitions be dead. It is a good
898 idea to to run a <a href="#die">DIE</a> (Dead Instruction Elimination) pass
899 sometime after running this pass.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]900</div>
901
902<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]903<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]904 <a name="dce">-dce: Dead Code Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]905</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]906<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]907 <p>
908 Dead code elimination is similar to <a href="#die">dead instruction
909 elimination</a>, but it rechecks instructions that were used by removed
910 instructions to see if they are newly dead.
911 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]912</div>
913
914<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]915<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]916 <a name="deadargelim">-deadargelim: Dead Argument Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]917</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]918<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]919 <p>
920 This pass deletes dead arguments from internal functions. Dead argument
921 elimination removes arguments which are directly dead, as well as arguments
922 only passed into function calls as dead arguments of other functions. This
923 pass also deletes dead arguments in a similar way.
924 </p>
925
926 <p>
927 This pass is often useful as a cleanup pass to run after aggressive
928 interprocedural passes, which add possibly-dead arguments.
929 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]930</div>
931
932<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]933<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]934 <a name="deadtypeelim">-deadtypeelim: Dead Type Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]935</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]936<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]937 <p>
938 This pass is used to cleanup the output of GCC. It eliminate names for types
939 that are unused in the entire translation unit, using the <a
940 href="#findusedtypes">find used types</a> pass.
941 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]942</div>
943
944<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]945<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]946 <a name="die">-die: Dead Instruction Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]947</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]948<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]949 <p>
950 Dead instruction elimination performs a single pass over the function,
951 removing instructions that are obviously dead.
952 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]953</div>
954
955<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]956<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]957 <a name="dse">-dse: Dead Store Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]958</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]959<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]960 <p>
961 A trivial dead store elimination that only considers basic-block local
962 redundant stores.
963 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]964</div>
965
966<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]967<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]968 <a name="functionattrs">-functionattrs: Deduce function attributes</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]969</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]970<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]971 <p>A simple interprocedural pass which walks the call-graph, looking for
972 functions which do not access or only read non-local memory, and marking them
973 readnone/readonly. In addition, it marks function arguments (of pointer type)
974 'nocapture' if a call to the function does not create any copies of the pointer
975 value that outlive the call. This more or less means that the pointer is only
976 dereferenced, and not returned from the function or stored in a global.
977 This pass is implemented as a bottom-up traversal of the call-graph.
978 </p>
979</div>
980
981<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]982<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]983 <a name="globaldce">-globaldce: Dead Global Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]984</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]985<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]986 <p>
987 This transform is designed to eliminate unreachable internal globals from the
988 program. It uses an aggressive algorithm, searching out globals that are
989 known to be alive. After it finds all of the globals which are needed, it
990 deletes whatever is left over. This allows it to delete recursive chunks of
991 the program which are unreachable.
992 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]993</div>
994
995<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]996<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]997 <a name="globalopt">-globalopt: Global Variable Optimizer</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]998</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]999<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]1000 <p>
1001 This pass transforms simple global variables that never have their address
1002 taken. If obviously true, it marks read/write globals as constant, deletes
1003 variables only stored to, etc.
1004 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1005</div>
1006
1007<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1008<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1009 <a name="gvn">-gvn: Global Value Numbering</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1010</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1011<div>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]1012 <p>
Chris Lattner60f03402009-10-10 18:40:48 +0000[diff] [blame]1013 This pass performs global value numbering to eliminate fully and partially
1014 redundant instructions. It also performs redundant load elimination.
Matthijs Kooijman845f5242008-06-05 07:55:49 +0000[diff] [blame]1015 </p>
Gordon Henriksen0e15dc22007-10-25 10:18:27 +0000[diff] [blame]1016</div>
1017
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]1018<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1019<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1020 <a name="indvars">-indvars: Canonicalize Induction Variables</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1021</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1022<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]1023 <p>
1024 This transformation analyzes and transforms the induction variables (and
1025 computations derived from them) into simpler forms suitable for subsequent
1026 analysis and transformation.
1027 </p>
1028
1029 <p>
1030 This transformation makes the following changes to each loop with an
1031 identifiable induction variable:
1032 </p>
1033
1034 <ol>
1035 <li>All loops are transformed to have a <em>single</em> canonical
1036 induction variable which starts at zero and steps by one.</li>
1037 <li>The canonical induction variable is guaranteed to be the first PHI node
1038 in the loop header block.</li>
1039 <li>Any pointer arithmetic recurrences are raised to use array
1040 subscripts.</li>
1041 </ol>
1042
1043 <p>
1044 If the trip count of a loop is computable, this pass also makes the following
1045 changes:
1046 </p>
1047
1048 <ol>
1049 <li>The exit condition for the loop is canonicalized to compare the
1050 induction value against the exit value. This turns loops like:
1051 <blockquote><pre>for (i = 7; i*i < 1000; ++i)</pre></blockquote>
1052 into
1053 <blockquote><pre>for (i = 0; i != 25; ++i)</pre></blockquote></li>
1054 <li>Any use outside of the loop of an expression derived from the indvar
1055 is changed to compute the derived value outside of the loop, eliminating
1056 the dependence on the exit value of the induction variable. If the only
1057 purpose of the loop is to compute the exit value of some derived
1058 expression, this transformation will make the loop dead.</li>
Gordon Henriksene626bbe2007-11-04 16:17:00 +0000[diff] [blame]1059 </ol>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]1060
1061 <p>
1062 This transformation should be followed by strength reduction after all of the
1063 desired loop transformations have been performed. Additionally, on targets
1064 where it is profitable, the loop could be transformed to count down to zero
1065 (the "do loop" optimization).
1066 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1067</div>
1068
1069<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1070<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1071 <a name="inline">-inline: Function Integration/Inlining</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1072</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1073<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]1074 <p>
1075 Bottom-up inlining of functions into callees.
1076 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1077</div>
1078
1079<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1080<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1081 <a name="insert-edge-profiling">-insert-edge-profiling: Insert instrumentation for edge profiling</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1082</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1083<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]1084 <p>
1085 This pass instruments the specified program with counters for edge profiling.
1086 Edge profiling can give a reasonable approximation of the hot paths through a
1087 program, and is used for a wide variety of program transformations.
1088 </p>
1089
1090 <p>
1091 Note that this implementation is very naïve. It inserts a counter for
1092 <em>every</em> edge in the program, instead of using control flow information
1093 to prune the number of counters inserted.
1094 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1095</div>
1096
1097<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1098<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1099 <a name="insert-optimal-edge-profiling">-insert-optimal-edge-profiling: Insert optimal instrumentation for edge profiling</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1100</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1101<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1102 <p>This pass instruments the specified program with counters for edge profiling.
1103 Edge profiling can give a reasonable approximation of the hot paths through a
1104 program, and is used for a wide variety of program transformations.
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]1105 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1106</div>
1107
1108<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1109<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1110 <a name="instcombine">-instcombine: Combine redundant instructions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1111</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1112<div>
Gordon Henriksen55cbec32007-10-26 03:03:51 +0000[diff] [blame]1113 <p>
1114 Combine instructions to form fewer, simple
1115 instructions. This pass does not modify the CFG This pass is where algebraic
1116 simplification happens.
1117 </p>
1118
1119 <p>
1120 This pass combines things like:
1121 </p>
1122
1123<blockquote><pre
1124>%Y = add i32 %X, 1
1125%Z = add i32 %Y, 1</pre></blockquote>
1126
1127 <p>
1128 into:
1129 </p>
1130
1131<blockquote><pre
1132>%Z = add i32 %X, 2</pre></blockquote>
1133
1134 <p>
1135 This is a simple worklist driven algorithm.
1136 </p>
1137
1138 <p>
1139 This pass guarantees that the following canonicalizations are performed on
1140 the program:
1141 </p>
1142
1143 <ul>
1144 <li>If a binary operator has a constant operand, it is moved to the right-
1145 hand side.</li>
1146 <li>Bitwise operators with constant operands are always grouped so that
1147 shifts are performed first, then <code>or</code>s, then
1148 <code>and</code>s, then <code>xor</code>s.</li>
1149 <li>Compare instructions are converted from <code>&lt;</code>,
1150 <code>&gt;</code>, <code>≤</code>, or <code>≥</code> to
1151 <code>=</code> or <code>≠</code> if possible.</li>
1152 <li>All <code>cmp</code> instructions on boolean values are replaced with
1153 logical operations.</li>
1154 <li><code>add <var>X</var>, <var>X</var></code> is represented as
1155 <code>mul <var>X</var>, 2</code> ⇒ <code>shl <var>X</var>, 1</code></li>
1156 <li>Multiplies with a constant power-of-two argument are transformed into
1157 shifts.</li>
1158 <li>… etc.</li>
1159 </ul>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1160</div>
1161
1162<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1163<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1164 <a name="internalize">-internalize: Internalize Global Symbols</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1165</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1166<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1167 <p>
1168 This pass loops over all of the functions in the input module, looking for a
1169 main function. If a main function is found, all other functions and all
1170 global variables with initializers are marked as internal.
1171 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1172</div>
1173
1174<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1175<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1176 <a name="ipconstprop">-ipconstprop: Interprocedural constant propagation</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1177</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1178<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1179 <p>
1180 This pass implements an <em>extremely</em> simple interprocedural constant
1181 propagation pass. It could certainly be improved in many different ways,
1182 like using a worklist. This pass makes arguments dead, but does not remove
1183 them. The existing dead argument elimination pass should be run after this
1184 to clean up the mess.
1185 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1186</div>
1187
1188<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1189<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1190 <a name="ipsccp">-ipsccp: Interprocedural Sparse Conditional Constant Propagation</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1191</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1192<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1193 <p>
1194 An interprocedural variant of <a href="#sccp">Sparse Conditional Constant
1195 Propagation</a>.
1196 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1197</div>
1198
1199<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1200<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]1201 <a name="jump-threading">-jump-threading: Jump Threading</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1202</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1203<div>
Gordon Henriksena8a118b2008-05-08 17:46:35 +0000[diff] [blame]1204 <p>
1205 Jump threading tries to find distinct threads of control flow running through
1206 a basic block. This pass looks at blocks that have multiple predecessors and
1207 multiple successors. If one or more of the predecessors of the block can be
1208 proven to always cause a jump to one of the successors, we forward the edge
1209 from the predecessor to the successor by duplicating the contents of this
1210 block.
1211 </p>
1212 <p>
1213 An example of when this can occur is code like this:
1214 </p>
1215
1216 <pre
1217>if () { ...
1218 X = 4;
1219}
1220if (X &lt; 3) {</pre>
1221
1222 <p>
1223 In this case, the unconditional branch at the end of the first if can be
1224 revectored to the false side of the second if.
1225 </p>
1226</div>
1227
1228<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1229<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1230 <a name="lcssa">-lcssa: Loop-Closed SSA Form Pass</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1231</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1232<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1233 <p>
1234 This pass transforms loops by placing phi nodes at the end of the loops for
1235 all values that are live across the loop boundary. For example, it turns
1236 the left into the right code:
1237 </p>
1238
1239 <pre
1240>for (...) for (...)
1241 if (c) if (c)
1242 X1 = ... X1 = ...
1243 else else
1244 X2 = ... X2 = ...
1245 X3 = phi(X1, X2) X3 = phi(X1, X2)
1246... = X3 + 4 X4 = phi(X3)
1247 ... = X4 + 4</pre>
1248
1249 <p>
1250 This is still valid LLVM; the extra phi nodes are purely redundant, and will
1251 be trivially eliminated by <code>InstCombine</code>. The major benefit of
1252 this transformation is that it makes many other loop optimizations, such as
1253 LoopUnswitching, simpler.
1254 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1255</div>
1256
1257<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1258<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1259 <a name="licm">-licm: Loop Invariant Code Motion</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1260</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1261<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1262 <p>
1263 This pass performs loop invariant code motion, attempting to remove as much
1264 code from the body of a loop as possible. It does this by either hoisting
1265 code into the preheader block, or by sinking code to the exit blocks if it is
1266 safe. This pass also promotes must-aliased memory locations in the loop to
1267 live in registers, thus hoisting and sinking "invariant" loads and stores.
1268 </p>
1269
1270 <p>
1271 This pass uses alias analysis for two purposes:
1272 </p>
1273
1274 <ul>
1275 <li>Moving loop invariant loads and calls out of loops. If we can determine
1276 that a load or call inside of a loop never aliases anything stored to,
1277 we can hoist it or sink it like any other instruction.</li>
1278 <li>Scalar Promotion of Memory - If there is a store instruction inside of
1279 the loop, we try to move the store to happen AFTER the loop instead of
1280 inside of the loop. This can only happen if a few conditions are true:
1281 <ul>
1282 <li>The pointer stored through is loop invariant.</li>
1283 <li>There are no stores or loads in the loop which <em>may</em> alias
1284 the pointer. There are no calls in the loop which mod/ref the
1285 pointer.</li>
1286 </ul>
1287 If these conditions are true, we can promote the loads and stores in the
1288 loop of the pointer to use a temporary alloca'd variable. We then use
1289 the mem2reg functionality to construct the appropriate SSA form for the
1290 variable.</li>
1291 </ul>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1292</div>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1293
Gordon Henriksena8a118b2008-05-08 17:46:35 +0000[diff] [blame]1294<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1295<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]1296 <a name="loop-deletion">-loop-deletion: Delete dead loops</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1297</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1298<div>
Gordon Henriksena8a118b2008-05-08 17:46:35 +0000[diff] [blame]1299 <p>
1300 This file implements the Dead Loop Deletion Pass. This pass is responsible
1301 for eliminating loops with non-infinite computable trip counts that have no
1302 side effects or volatile instructions, and do not contribute to the
1303 computation of the function's return value.
1304 </p>
1305</div>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1306
1307<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1308<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1309 <a name="loop-extract">-loop-extract: Extract loops into new functions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1310</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1311<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1312 <p>
1313 A pass wrapper around the <code>ExtractLoop()</code> scalar transformation to
1314 extract each top-level loop into its own new function. If the loop is the
1315 <em>only</em> loop in a given function, it is not touched. This is a pass most
1316 useful for debugging via bugpoint.
1317 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1318</div>
1319
1320<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1321<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1322 <a name="loop-extract-single">-loop-extract-single: Extract at most one loop into a new function</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1323</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1324<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1325 <p>
1326 Similar to <a href="#loop-extract">Extract loops into new functions</a>,
1327 this pass extracts one natural loop from the program into a function if it
1328 can. This is used by bugpoint.
1329 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1330</div>
1331
1332<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1333<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1334 <a name="loop-reduce">-loop-reduce: Loop Strength Reduction</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1335</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1336<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1337 <p>
1338 This pass performs a strength reduction on array references inside loops that
1339 have as one or more of their components the loop induction variable. This is
1340 accomplished by creating a new value to hold the initial value of the array
1341 access for the first iteration, and then creating a new GEP instruction in
1342 the loop to increment the value by the appropriate amount.
1343 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1344</div>
1345
1346<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1347<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1348 <a name="loop-rotate">-loop-rotate: Rotate Loops</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1349</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1350<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1351 <p>A simple loop rotation transformation.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1352</div>
1353
1354<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1355<h3>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1356 <a name="loop-simplify">-loop-simplify: Canonicalize natural loops</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1357</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1358<div>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1359 <p>
1360 This pass performs several transformations to transform natural loops into a
1361 simpler form, which makes subsequent analyses and transformations simpler and
1362 more effective.
1363 </p>
1364
1365 <p>
1366 Loop pre-header insertion guarantees that there is a single, non-critical
1367 entry edge from outside of the loop to the loop header. This simplifies a
1368 number of analyses and transformations, such as LICM.
1369 </p>
1370
1371 <p>
1372 Loop exit-block insertion guarantees that all exit blocks from the loop
1373 (blocks which are outside of the loop that have predecessors inside of the
1374 loop) only have predecessors from inside of the loop (and are thus dominated
1375 by the loop header). This simplifies transformations such as store-sinking
1376 that are built into LICM.
1377 </p>
1378
1379 <p>
1380 This pass also guarantees that loops will have exactly one backedge.
1381 </p>
1382
1383 <p>
1384 Note that the simplifycfg pass will clean up blocks which are split out but
1385 end up being unnecessary, so usage of this pass should not pessimize
1386 generated code.
1387 </p>
1388
1389 <p>
1390 This pass obviously modifies the CFG, but updates loop information and
1391 dominator information.
1392 </p>
1393</div>
1394
1395<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1396<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1397 <a name="loop-unroll">-loop-unroll: Unroll loops</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1398</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1399<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1400 <p>
1401 This pass implements a simple loop unroller. It works best when loops have
1402 been canonicalized by the <a href="#indvars"><tt>-indvars</tt></a> pass,
1403 allowing it to determine the trip counts of loops easily.
1404 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1405</div>
1406
1407<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1408<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1409 <a name="loop-unswitch">-loop-unswitch: Unswitch loops</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1410</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1411<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1412 <p>
1413 This pass transforms loops that contain branches on loop-invariant conditions
1414 to have multiple loops. For example, it turns the left into the right code:
1415 </p>
1416
1417 <pre
1418>for (...) if (lic)
1419 A for (...)
1420 if (lic) A; B; C
1421 B else
1422 C for (...)
1423 A; C</pre>
1424
1425 <p>
1426 This can increase the size of the code exponentially (doubling it every time
1427 a loop is unswitched) so we only unswitch if the resultant code will be
1428 smaller than a threshold.
1429 </p>
1430
1431 <p>
1432 This pass expects LICM to be run before it to hoist invariant conditions out
1433 of the loop, to make the unswitching opportunity obvious.
1434 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1435</div>
1436
1437<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1438<h3>
Eli Friedmande8ec5b2011-03-19 04:55:29 +0000[diff] [blame]1439 <a name="loweratomic">-loweratomic: Lower atomic intrinsics to non-atomic form</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1440</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1441<div>
Peter Collingbourne3bababf2010-08-03 16:19:16 +0000[diff] [blame]1442 <p>
1443 This pass lowers atomic intrinsics to non-atomic form for use in a known
1444 non-preemptible environment.
1445 </p>
1446
1447 <p>
1448 The pass does not verify that the environment is non-preemptible (in
1449 general this would require knowledge of the entire call graph of the
1450 program including any libraries which may not be available in bitcode form);
1451 it simply lowers every atomic intrinsic.
1452 </p>
1453</div>
1454
1455<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1456<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1457 <a name="lowerinvoke">-lowerinvoke: Lower invoke and unwind, for unwindless code generators</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1458</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1459<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1460 <p>
1461 This transformation is designed for use by code generators which do not yet
1462 support stack unwinding. This pass supports two models of exception handling
1463 lowering, the 'cheap' support and the 'expensive' support.
1464 </p>
1465
1466 <p>
1467 'Cheap' exception handling support gives the program the ability to execute
1468 any program which does not "throw an exception", by turning 'invoke'
1469 instructions into calls and by turning 'unwind' instructions into calls to
1470 abort(). If the program does dynamically use the unwind instruction, the
1471 program will print a message then abort.
1472 </p>
1473
1474 <p>
1475 'Expensive' exception handling support gives the full exception handling
1476 support to the program at the cost of making the 'invoke' instruction
1477 really expensive. It basically inserts setjmp/longjmp calls to emulate the
1478 exception handling as necessary.
1479 </p>
1480
1481 <p>
1482 Because the 'expensive' support slows down programs a lot, and EH is only
1483 used for a subset of the programs, it must be specifically enabled by the
1484 <tt>-enable-correct-eh-support</tt> option.
1485 </p>
1486
1487 <p>
1488 Note that after this pass runs the CFG is not entirely accurate (exceptional
1489 control flow edges are not correct anymore) so only very simple things should
1490 be done after the lowerinvoke pass has run (like generation of native code).
1491 This should not be used as a general purpose "my LLVM-to-LLVM pass doesn't
1492 support the invoke instruction yet" lowering pass.
1493 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1494</div>
1495
1496<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1497<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1498 <a name="lowerswitch">-lowerswitch: Lower SwitchInst's to branches</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1499</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1500<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1501 <p>
1502 Rewrites <tt>switch</tt> instructions with a sequence of branches, which
1503 allows targets to get away with not implementing the switch instruction until
1504 it is convenient.
1505 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1506</div>
1507
1508<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1509<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1510 <a name="mem2reg">-mem2reg: Promote Memory to Register</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1511</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1512<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1513 <p>
1514 This file promotes memory references to be register references. It promotes
1515 <tt>alloca</tt> instructions which only have <tt>load</tt>s and
1516 <tt>store</tt>s as uses. An <tt>alloca</tt> is transformed by using dominator
1517 frontiers to place <tt>phi</tt> nodes, then traversing the function in
1518 depth-first order to rewrite <tt>load</tt>s and <tt>store</tt>s as
1519 appropriate. This is just the standard SSA construction algorithm to construct
1520 "pruned" SSA form.
1521 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1522</div>
1523
1524<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1525<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]1526 <a name="memcpyopt">-memcpyopt: MemCpy Optimization</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1527</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1528<div>
Gordon Henriksena8a118b2008-05-08 17:46:35 +0000[diff] [blame]1529 <p>
1530 This pass performs various transformations related to eliminating memcpy
1531 calls, or transforming sets of stores into memset's.
1532 </p>
1533</div>
1534
1535<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1536<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1537 <a name="mergefunc">-mergefunc: Merge Functions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1538</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1539<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1540 <p>This pass looks for equivalent functions that are mergable and folds them.
1541
1542 A hash is computed from the function, based on its type and number of
1543 basic blocks.
1544
1545 Once all hashes are computed, we perform an expensive equality comparison
1546 on each function pair. This takes n^2/2 comparisons per bucket, so it's
1547 important that the hash function be high quality. The equality comparison
1548 iterates through each instruction in each basic block.
1549
1550 When a match is found the functions are folded. If both functions are
1551 overridable, we move the functionality into a new internal function and
1552 leave two overridable thunks to it.
1553 </p>
1554</div>
1555
1556<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1557<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1558 <a name="mergereturn">-mergereturn: Unify function exit nodes</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1559</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1560<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1561 <p>
1562 Ensure that functions have at most one <tt>ret</tt> instruction in them.
1563 Additionally, it keeps track of which node is the new exit node of the CFG.
1564 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1565</div>
1566
1567<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1568<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1569 <a name="partial-inliner">-partial-inliner: Partial Inliner</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1570</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1571<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1572 <p>This pass performs partial inlining, typically by inlining an if
1573 statement that surrounds the body of the function.
1574 </p>
1575</div>
1576
1577<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1578<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1579 <a name="prune-eh">-prune-eh: Remove unused exception handling info</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1580</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1581<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1582 <p>
1583 This file implements a simple interprocedural pass which walks the call-graph,
1584 turning <tt>invoke</tt> instructions into <tt>call</tt> instructions if and
1585 only if the callee cannot throw an exception. It implements this as a
1586 bottom-up traversal of the call-graph.
1587 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1588</div>
1589
1590<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1591<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1592 <a name="reassociate">-reassociate: Reassociate expressions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1593</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1594<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1595 <p>
1596 This pass reassociates commutative expressions in an order that is designed
1597 to promote better constant propagation, GCSE, LICM, PRE, etc.
1598 </p>
1599
1600 <p>
1601 For example: 4 + (<var>x</var> + 5) ⇒ <var>x</var> + (4 + 5)
1602 </p>
1603
1604 <p>
1605 In the implementation of this algorithm, constants are assigned rank = 0,
1606 function arguments are rank = 1, and other values are assigned ranks
1607 corresponding to the reverse post order traversal of current function
1608 (starting at 2), which effectively gives values in deep loops higher rank
1609 than values not in loops.
1610 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1611</div>
1612
1613<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1614<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1615 <a name="reg2mem">-reg2mem: Demote all values to stack slots</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1616</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1617<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1618 <p>
Benjamin Kramerd9b0b022012-06-02 10:20:22 +0000[diff] [blame]1619 This file demotes all registers to memory references. It is intended to be
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1620 the inverse of <a href="#mem2reg"><tt>-mem2reg</tt></a>. By converting to
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]1621 <tt>load</tt> instructions, the only values live across basic blocks are
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1622 <tt>alloca</tt> instructions and <tt>load</tt> instructions before
1623 <tt>phi</tt> nodes. It is intended that this should make CFG hacking much
1624 easier. To make later hacking easier, the entry block is split into two, such
1625 that all introduced <tt>alloca</tt> instructions (and nothing else) are in the
1626 entry block.
1627 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1628</div>
1629
1630<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1631<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]1632 <a name="scalarrepl">-scalarrepl: Scalar Replacement of Aggregates (DT)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1633</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1634<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1635 <p>
1636 The well-known scalar replacement of aggregates transformation. This
1637 transform breaks up <tt>alloca</tt> instructions of aggregate type (structure
1638 or array) into individual <tt>alloca</tt> instructions for each member if
1639 possible. Then, if possible, it transforms the individual <tt>alloca</tt>
1640 instructions into nice clean scalar SSA form.
1641 </p>
1642
1643 <p>
1644 This combines a simple scalar replacement of aggregates algorithm with the <a
1645 href="#mem2reg"><tt>mem2reg</tt></a> algorithm because often interact,
1646 especially for C++ programs. As such, iterating between <tt>scalarrepl</tt>,
1647 then <a href="#mem2reg"><tt>mem2reg</tt></a> until we run out of things to
1648 promote works well.
1649 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1650</div>
1651
1652<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1653<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1654 <a name="sccp">-sccp: Sparse Conditional Constant Propagation</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1655</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1656<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1657 <p>
1658 Sparse conditional constant propagation and merging, which can be summarized
1659 as:
1660 </p>
1661
1662 <ol>
1663 <li>Assumes values are constant unless proven otherwise</li>
1664 <li>Assumes BasicBlocks are dead unless proven otherwise</li>
1665 <li>Proves values to be constant, and replaces them with constants</li>
1666 <li>Proves conditional branches to be unconditional</li>
1667 </ol>
1668
1669 <p>
1670 Note that this pass has a habit of making definitions be dead. It is a good
1671 idea to to run a DCE pass sometime after running this pass.
1672 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1673</div>
1674
1675<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1676<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1677 <a name="simplify-libcalls">-simplify-libcalls: Simplify well-known library calls</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1678</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1679<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1680 <p>
1681 Applies a variety of small optimizations for calls to specific well-known
1682 function calls (e.g. runtime library functions). For example, a call
1683 <tt>exit(3)</tt> that occurs within the <tt>main()</tt> function can be
1684 transformed into simply <tt>return 3</tt>.
1685 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1686</div>
1687
1688<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1689<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1690 <a name="simplifycfg">-simplifycfg: Simplify the CFG</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1691</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1692<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1693 <p>
1694 Performs dead code elimination and basic block merging. Specifically:
1695 </p>
1696
1697 <ol>
1698 <li>Removes basic blocks with no predecessors.</li>
1699 <li>Merges a basic block into its predecessor if there is only one and the
1700 predecessor only has one successor.</li>
1701 <li>Eliminates PHI nodes for basic blocks with a single predecessor.</li>
1702 <li>Eliminates a basic block that only contains an unconditional
1703 branch.</li>
1704 </ol>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1705</div>
1706
1707<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1708<h3>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1709 <a name="sink">-sink: Code sinking</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1710</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1711<div>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1712 <p>This pass moves instructions into successor blocks, when possible, so that
1713 they aren't executed on paths where their results aren't needed.
1714 </p>
1715</div>
1716
1717<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1718<h3>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1719 <a name="sretpromotion">-sretpromotion: Promote sret arguments to multiple ret values</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1720</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1721<div>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1722 <p>
1723 This pass finds functions that return a struct (using a pointer to the struct
1724 as the first argument of the function, marked with the '<tt>sret</tt>' attribute) and
1725 replaces them with a new function that simply returns each of the elements of
1726 that struct (using multiple return values).
1727 </p>
1728
1729 <p>
1730 This pass works under a number of conditions:
1731 </p>
1732
1733 <ul>
1734 <li>The returned struct must not contain other structs</li>
1735 <li>The returned struct must only be used to load values from</li>
1736 <li>The placeholder struct passed in is the result of an <tt>alloca</tt></li>
1737 </ul>
1738</div>
1739
1740<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1741<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1742 <a name="strip">-strip: Strip all symbols from a module</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1743</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1744<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1745 <p>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1746 performs code stripping. this transformation can delete:
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1747 </p>
1748
1749 <ol>
1750 <li>names for virtual registers</li>
1751 <li>symbols for internal globals and functions</li>
1752 <li>debug information</li>
1753 </ol>
1754
1755 <p>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1756 note that this transformation makes code much less readable, so it should
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1757 only be used in situations where the <tt>strip</tt> utility would be used,
1758 such as reducing code size or making it harder to reverse engineer code.
1759 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1760</div>
1761
1762<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1763<h3>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1764 <a name="strip-dead-debug-info">-strip-dead-debug-info: Strip debug info for unused symbols</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1765</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1766<div>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]1767 <p>
1768 performs code stripping. this transformation can delete:
1769 </p>
1770
1771 <ol>
1772 <li>names for virtual registers</li>
1773 <li>symbols for internal globals and functions</li>
1774 <li>debug information</li>
1775 </ol>
1776
1777 <p>
1778 note that this transformation makes code much less readable, so it should
1779 only be used in situations where the <tt>strip</tt> utility would be used,
1780 such as reducing code size or making it harder to reverse engineer code.
1781 </p>
1782</div>
1783
1784<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1785<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]1786 <a name="strip-dead-prototypes">-strip-dead-prototypes: Strip Unused Function Prototypes</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1787</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1788<div>
Gordon Henriksena8a118b2008-05-08 17:46:35 +0000[diff] [blame]1789 <p>
1790 This pass loops over all of the functions in the input module, looking for
1791 dead declarations and removes them. Dead declarations are declarations of
1792 functions for which no implementation is available (i.e., declarations for
1793 unused library functions).
1794 </p>
1795</div>
1796
1797<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1798<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1799 <a name="strip-debug-declare">-strip-debug-declare: Strip all llvm.dbg.declare intrinsics</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1800</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1801<div>
Peter Collingbournec3086ba2010-08-06 02:13:25 +0000[diff] [blame]1802 <p>This pass implements code stripping. Specifically, it can delete:</p>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1803 <ul>
1804 <li>names for virtual registers</li>
1805 <li>symbols for internal globals and functions</li>
1806 <li>debug information</li>
1807 </ul>
Peter Collingbournec3086ba2010-08-06 02:13:25 +0000[diff] [blame]1808 <p>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1809 Note that this transformation makes code much less readable, so it should
1810 only be used in situations where the 'strip' utility would be used, such as
1811 reducing code size or making it harder to reverse engineer code.
1812 </p>
1813</div>
1814
1815<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1816<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1817 <a name="strip-nondebug">-strip-nondebug: Strip all symbols, except dbg symbols, from a module</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1818</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1819<div>
Peter Collingbournec3086ba2010-08-06 02:13:25 +0000[diff] [blame]1820 <p>This pass implements code stripping. Specifically, it can delete:</p>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1821 <ul>
1822 <li>names for virtual registers</li>
1823 <li>symbols for internal globals and functions</li>
1824 <li>debug information</li>
1825 </ul>
Peter Collingbournec3086ba2010-08-06 02:13:25 +0000[diff] [blame]1826 <p>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1827 Note that this transformation makes code much less readable, so it should
1828 only be used in situations where the 'strip' utility would be used, such as
1829 reducing code size or making it harder to reverse engineer code.
1830 </p>
1831</div>
1832
1833<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1834<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1835 <a name="tailcallelim">-tailcallelim: Tail Call Elimination</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1836</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1837<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1838 <p>
1839 This file transforms calls of the current function (self recursion) followed
1840 by a return instruction with a branch to the entry of the function, creating
1841 a loop. This pass also implements the following extensions to the basic
1842 algorithm:
1843 </p>
1844
1845 <ul>
1846 <li>Trivial instructions between the call and return do not prevent the
1847 transformation from taking place, though currently the analysis cannot
1848 support moving any really useful instructions (only dead ones).
1849 <li>This pass transforms functions that are prevented from being tail
1850 recursive by an associative expression to use an accumulator variable,
1851 thus compiling the typical naive factorial or <tt>fib</tt> implementation
1852 into efficient code.
1853 <li>TRE is performed if the function returns void, if the return
1854 returns the result returned by the call, or if the function returns a
1855 run-time constant on all exits from the function. It is possible, though
1856 unlikely, that the return returns something else (like constant 0), and
1857 can still be TRE'd. It can be TRE'd if <em>all other</em> return
1858 instructions in the function return the exact same value.
1859 <li>If it can prove that callees do not access theier caller stack frame,
1860 they are marked as eligible for tail call elimination (by the code
1861 generator).
1862 </ul>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1863</div>
1864
1865<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1866<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1867 <a name="tailduplicate">-tailduplicate: Tail Duplication</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1868</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1869<div>
Gordon Henriksenc86b6772007-11-04 16:15:04 +0000[diff] [blame]1870 <p>
1871 This pass performs a limited form of tail duplication, intended to simplify
1872 CFGs by removing some unconditional branches. This pass is necessary to
1873 straighten out loops created by the C front-end, but also is capable of
1874 making other code nicer. After this pass is run, the CFG simplify pass
1875 should be run to clean up the mess.
1876 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1877</div>
1878
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1879</div>
1880
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1881<!-- ======================================================================= -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1882<h2><a name="utilities">Utility Passes</a></h2>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1883<div>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1884 <p>This section describes the LLVM Utility Passes.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1885
1886<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1887<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1888 <a name="deadarghaX0r">-deadarghaX0r: Dead Argument Hacking (BUGPOINT USE ONLY; DO NOT USE)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1889</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1890<div>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1891 <p>
1892 Same as dead argument elimination, but deletes arguments to functions which
1893 are external. This is only for use by <a
1894 href="Bugpoint.html">bugpoint</a>.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1895</div>
1896
1897<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1898<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1899 <a name="extract-blocks">-extract-blocks: Extract Basic Blocks From Module (for bugpoint use)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1900</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1901<div>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1902 <p>
1903 This pass is used by bugpoint to extract all blocks from the module into their
1904 own functions.</p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1905</div>
1906
1907<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1908<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1909 <a name="instnamer">-instnamer: Assign names to anonymous instructions</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1910</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1911<div>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1912 <p>This is a little utility pass that gives instructions names, this is mostly
1913 useful when diffing the effect of an optimization because deleting an
1914 unnamed instruction can change all other instruction numbering, making the
1915 diff very noisy.
1916 </p>
1917</div>
1918
1919<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1920<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1921 <a name="preverify">-preverify: Preliminary module verification</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1922</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1923<div>
Gordon Henriksen90a52142007-11-05 02:05:35 +0000[diff] [blame]1924 <p>
1925 Ensures that the module is in the form required by the <a
1926 href="#verifier">Module Verifier</a> pass.
1927 </p>
1928
1929 <p>
1930 Running the verifier runs this pass automatically, so there should be no need
1931 to use it directly.
1932 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1933</div>
1934
1935<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1936<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1937 <a name="verify">-verify: Module Verifier</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1938</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1939<div>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1940 <p>
1941 Verifies an LLVM IR code. This is useful to run after an optimization which is
1942 undergoing testing. Note that <tt>llvm-as</tt> verifies its input before
1943 emitting bitcode, and also that malformed bitcode is likely to make LLVM
1944 crash. All language front-ends are therefore encouraged to verify their output
1945 before performing optimizing transformations.
1946 </p>
1947
Gordon Henriksen23a8ce52007-11-04 18:14:08 +0000[diff] [blame]1948 <ul>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1949 <li>Both of a binary operator's parameters are of the same type.</li>
1950 <li>Verify that the indices of mem access instructions match other
1951 operands.</li>
1952 <li>Verify that arithmetic and other things are only performed on
1953 first-class types. Verify that shifts and logicals only happen on
1954 integrals f.e.</li>
1955 <li>All of the constants in a switch statement are of the correct type.</li>
1956 <li>The code is in valid SSA form.</li>
Chris Lattner46b3abc2009-10-28 04:47:06 +0000[diff] [blame]1957 <li>It is illegal to put a label into any other type (like a structure) or
1958 to return one.</li>
Nick Lewycky0c78ac12008-03-28 06:46:51 +0000[diff] [blame]1959 <li>Only phi nodes can be self referential: <tt>%x = add i32 %x, %x</tt> is
Gordon Henriksen873390e2007-11-04 18:17:58 +0000[diff] [blame]1960 invalid.</li>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1961 <li>PHI nodes must have an entry for each predecessor, with no extras.</li>
1962 <li>PHI nodes must be the first thing in a basic block, all grouped
1963 together.</li>
1964 <li>PHI nodes must have at least one entry.</li>
1965 <li>All basic blocks should only end with terminator insts, not contain
1966 them.</li>
1967 <li>The entry node to a function must not have predecessors.</li>
1968 <li>All Instructions must be embedded into a basic block.</li>
1969 <li>Functions cannot take a void-typed parameter.</li>
1970 <li>Verify that a function's argument list agrees with its declared
1971 type.</li>
1972 <li>It is illegal to specify a name for a void value.</li>
Sylvestre Ledru7f7390e2012-07-25 22:01:31 +0000[diff] [blame]1973 <li>It is illegal to have an internal global value with no initializer.</li>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1974 <li>It is illegal to have a ret instruction that returns a value that does
1975 not agree with the function return value type.</li>
1976 <li>Function call argument types match the function prototype.</li>
1977 <li>All other things that are tested by asserts spread about the code.</li>
Gordon Henriksen23a8ce52007-11-04 18:14:08 +0000[diff] [blame]1978 </ul>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1979
1980 <p>
1981 Note that this does not provide full security verification (like Java), but
1982 instead just tries to ensure that code is well-formed.
1983 </p>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]1984</div>
1985
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]1986<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1987<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1988 <a name="view-cfg">-view-cfg: View CFG of function</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1989</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]1990<div>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]1991 <p>
1992 Displays the control flow graph using the GraphViz tool.
1993 </p>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]1994</div>
1995
1996<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1997<h3>
Duncan Sands5c603862010-07-06 15:52:15 +0000[diff] [blame]1998 <a name="view-cfg-only">-view-cfg-only: View CFG of function (with no function bodies)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]1999</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]2000<div>
Gordon Henriksen75ff18e2007-11-04 18:10:18 +0000[diff] [blame]2001 <p>
2002 Displays the control flow graph using the GraphViz tool, but omitting function
2003 bodies.
2004 </p>
Gordon Henriksen1f5cce02007-10-25 08:46:12 +0000[diff] [blame]2005</div>
2006
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]2007<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2008<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]2009 <a name="view-dom">-view-dom: View dominance tree of function</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2010</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]2011<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]2012 <p>
2013 Displays the dominator tree using the GraphViz tool.
2014 </p>
2015</div>
2016
2017<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2018<h3>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]2019 <a name="view-dom-only">-view-dom-only: View dominance tree of function (with no function bodies)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2020</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]2021<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]2022 <p>
2023 Displays the dominator tree using the GraphViz tool, but omitting function
2024 bodies.
2025 </p>
2026</div>
2027
2028<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2029<h3>
Eli Friedmane6ed15b2011-03-19 04:47:52 +0000[diff] [blame]2030 <a name="view-postdom">-view-postdom: View postdominance tree of function</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2031</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]2032<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]2033 <p>
2034 Displays the post dominator tree using the GraphViz tool.
2035 </p>
2036</div>
2037
2038<!-------------------------------------------------------------------------- -->
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2039<h3>
Eli Friedman415247d2011-03-19 05:02:14 +0000[diff] [blame]2040 <a name="view-postdom-only">-view-postdom-only: View postdominance tree of function (with no function bodies)</a>
NAKAMURA Takumi05d02652011-04-18 23:59:50 +0000[diff] [blame]2041</h3>
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]2042<div>
Tobias Grosser733783b2010-05-07 09:33:18 +0000[diff] [blame]2043 <p>
2044 Displays the post dominator tree using the GraphViz tool, but omitting
2045 function bodies.
2046 </p>
2047</div>
2048
NAKAMURA Takumif5af6ad2011-04-23 00:30:22 +0000[diff] [blame]2049</div>
2050
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]2051<!-- *********************************************************************** -->
2052
2053<hr>
2054<address>
2055 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]2056 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]2057 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]2058 src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]2059
2060 <a href="mailto:rspencer@x10sys.com">Reid Spencer</a><br>
NAKAMURA Takumib9a33632011-04-09 02:13:37 +0000[diff] [blame]2061 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
Reid Spencerd9aac122007-03-26 09:32:31 +0000[diff] [blame]2062 Last modified: $Date$
2063</address>
2064
2065</body>
2066</html>