[llvm-commits] [parallel] CVS: llvm/docs/CommandGuide/analyze.html bugpoint.html extract.html gccld.html llc.html llvm-prof.html opt.html

Misha Brukman brukman at cs.uiuc.edu
Mon Mar 1 17:57:02 PST 2004


Changes in directory llvm/docs/CommandGuide:

analyze.html updated: 1.8 -> 1.8.6.1
bugpoint.html updated: 1.19 -> 1.19.4.1
extract.html updated: 1.5 -> 1.5.6.1
gccld.html updated: 1.8 -> 1.8.4.1
llc.html updated: 1.6 -> 1.6.6.1
llvm-prof.html updated: 1.3 -> 1.3.4.1
opt.html updated: 1.8 -> 1.8.6.1

---
Log message:

Merge from trunk

---
Diffs of the changes:  (+103 -58)

Index: llvm/docs/CommandGuide/analyze.html
diff -u llvm/docs/CommandGuide/analyze.html:1.8 llvm/docs/CommandGuide/analyze.html:1.8.6.1
--- llvm/docs/CommandGuide/analyze.html:1.8	Tue Oct  7 15:36:00 2003
+++ llvm/docs/CommandGuide/analyze.html	Mon Mar  1 17:56:01 2004
@@ -51,16 +51,16 @@
 	Quiet mode.  With this option, analysis pass names are not printed.
 	<p>
 
-	<li> -load <plugin.so>
+	<li> -load <plugin>
 	<br>
-	Load the specified dynamic object with name plugin.so.  This file
+	Load the specified dynamic object with name <tt>plugin</tt>.  This file
 	should contain additional analysis passes that register themselves with
 	the <tt>analyze</tt> program after being loaded.
 	<p>
 
 	After being loaded, additional command line options are made available
-	for running the passes made available by plugin.so.  Use
-	'<tt><tt>analyze</tt> -load <plugin.so> -help</tt>' to see the new
+	for running the passes made available by <tt>plugin</tt>.  Use
+	'<tt><tt>analyze</tt> -load <plugin> -help</tt>' to see the new
 	list of available analysis passes.
 	<p>
 </ul>


Index: llvm/docs/CommandGuide/bugpoint.html
diff -u llvm/docs/CommandGuide/bugpoint.html:1.19 llvm/docs/CommandGuide/bugpoint.html:1.19.4.1
--- llvm/docs/CommandGuide/bugpoint.html:1.19	Wed Nov  5 15:46:13 2003
+++ llvm/docs/CommandGuide/bugpoint.html	Mon Mar  1 17:56:01 2004
@@ -47,43 +47,43 @@
 <tt>bugpoint</tt> starts the <a href="#crashdebug">crash debugger</a>.<p>
 
 Otherwise, if the <a href="#opt_output"><tt>-output</tt></a> option was not
-specified, <tt>bugpoint</tt> runs the test program with the C backend (which
-is assumed to generate good code) to generate a reference output.  Once
+specified, <tt>bugpoint</tt> runs the test program with the C backend (which is
+assumed to generate good code) to generate a reference output.  Once
 <tt>bugpoint</tt> has a reference output for the test program, it tries
-executing it
-with the <a href="#opt_run-">selected</a> code generator.  If
-the resulting output differs from the reference output, it assumes the
-difference resulted from a code generator failure, and starts the
-<a href="#codegendebug">code generator debugger</a>.<p>
-
-Otherwise, <tt>bugpoint</tt> runs the test program after all of the LLVM passes
-have been applied to it.  If its output differs from the reference output,
-it assumes the difference resulted from a failure in one of the LLVM passes,
-and enters the
-<a href="#miscompilationdebug">miscompilation debugger</a>. Otherwise,
-there is no problem <tt>bugpoint</tt> can debug.<p>
+executing it with the <a href="#opt_run-">selected</a> code generator.  If the
+selected code generator crashes, <tt>bugpoint</tt> starts the <a
+href="#crashdebug">crash debugger</a> on the code generator.  Otherwise, if the
+resulting output differs from the reference output, it assumes the difference
+resulted from a code generator failure, and starts the <a
+href="#codegendebug">code generator debugger</a>.<p>
+
+Finally, if the output of the selected code generator matches the reference
+output, <tt>bugpoint</tt> runs the test program after all of the LLVM passes
+have been applied to it.  If its output differs from the reference output, it
+assumes the difference resulted from a failure in one of the LLVM passes, and
+enters the <a href="#miscompilationdebug">miscompilation
+debugger</a>. Otherwise, there is no problem <tt>bugpoint</tt> can debug.<p>
 
 <a name="crashdebug">
 <h4>Crash debugger</h4>
 
-If an optimizer crashes, <tt>bugpoint</tt> will try as hard as it can to
-reduce the list of passes and the size of the test program.  First,
-<tt>bugpoint</tt> figures out which combination of passes triggers the bug. This
-is useful when debugging a problem exposed by <tt>gccas</tt>, for example,
-because it runs over 25 optimizations.<p>
+If an optimizer or code generator crashes, <tt>bugpoint</tt> will try as hard as
+it can to reduce the list of passes (for optimizer crashes) and the size of the
+test program.  First, <tt>bugpoint</tt> figures out which combination of
+optimizer passes triggers the bug. This is useful when debugging a problem
+exposed by <tt>gccas</tt>, for example, because it runs over 25
+optimizations.<p>
 
 Next, <tt>bugpoint</tt> tries removing functions from the test program, to
-reduce its
-size.  Usually it is able to reduce a test program
-to a single function, when debugging intraprocedural optimizations.  Once the
-number of
+reduce its size.  Usually it is able to reduce a test program to a single
+function, when debugging intraprocedural optimizations.  Once the number of
 functions has been reduced, it attempts to delete various edges in the control
 flow graph, to reduce the size of the function as much as possible.  Finally,
 <tt>bugpoint</tt> deletes any individual LLVM instructions whose absence does
 not eliminate the failure.  At the end, <tt>bugpoint</tt> should tell you what
 passes crash, give you a bytecode file, and give you instructions on how to
-reproduce the failure with <tt><a href="opt.html">opt</a></tt> or
-<tt><a href="analyze.html">analyze</a></tt>.<p>
+reproduce the failure with <tt><a href="opt.html">opt</a></tt>, <tt><a
+href="analyze.html">analyze</a></tt>, or <tt><a href="llc.html">llc</a></tt>.<p>
 
 <a name="codegendebug">
 <h4>Code generator debugger</h4>
@@ -152,14 +152,19 @@
     <tt>bugpoint</tt> crashes before you see its "All input ok" message,
     you might try <tt>llvm-link -v</tt> on the same set of input files. If
     that also crashes, you may be experiencing a linker bug.
+
+<li>If your program is <b>supposed</b> to crash, <tt>bugpoint</tt> will be
+    confused. One way to deal with this is to cause bugpoint to ignore the exit
+    code from your program, by giving it the <tt>-check-exit-code=false</tt>
+    option.
     
 </ol>
 
 <h3>OPTIONS</h3>
 
 <ul>
-	<li><tt>-additional-so <library.so></tt><br>
-    Load <tt><library.so></tt> into the test program whenever it is run.
+	<li><tt>-additional-so <library></tt><br>
+    Load <tt><library></tt> into the test program whenever it is run.
     This is useful if you are debugging programs which depend on non-LLVM
     libraries (such as the X or curses libraries) to run.<p>
 
@@ -175,6 +180,10 @@
         part of the <tt>-args</tt> option, not as options to <tt>bugpoint</tt>
         itself.<p>
 
+	<li><tt>-check-exit-code={true,false}</tt><br>
+    Assume a non-zero exit code or core dump from the test program is
+    a failure. Defaults to true.<p>
+
 	<li><tt>-disable-{adce,dce,simplifycfg}</tt><br>
     Do not run the specified passes to clean up and reduce the size of the
     test program. By default, <tt>bugpoint</tt> uses these passes internally
@@ -189,14 +198,14 @@
     test program, whenever it runs, to come from that file.
 	<p>
 
-	<a name="opt_load"><li> <tt>-load <plugin.so></tt><br>
-	Load the dynamic object <tt><plugin.so></tt> into <tt>bugpoint</tt>
+	<a name="opt_load"><li> <tt>-load <plugin></tt><br>
+	Load the dynamic object <tt><plugin></tt> into <tt>bugpoint</tt>
     itself.  This object should register new
 	optimization passes.  Once loaded, the object will add new command line
 	options to enable various optimizations.  To see the new complete list
 	of optimizations, use the -help and -load options together:
 	<p>
-	<tt>bugpoint -load <plugin.so> -help</tt>
+	<tt>bugpoint -load <plugin> -help</tt>
 	<p>
 
 	<a name="opt_output"><li><tt>-output <filename></tt><br>


Index: llvm/docs/CommandGuide/extract.html
diff -u llvm/docs/CommandGuide/extract.html:1.5 llvm/docs/CommandGuide/extract.html:1.5.6.1
--- llvm/docs/CommandGuide/extract.html:1.5	Tue Oct  7 15:12:03 2003
+++ llvm/docs/CommandGuide/extract.html	Mon Mar  1 17:56:01 2004
@@ -35,11 +35,17 @@
 The <tt>extract</tt> command reads its input from standard input if filename is
 omitted or if filename is -.  The output is always written to standard output.
 
-<h3>
-OPTIONS
-</h3>
+<h3>OPTIONS</h3>
 
 <ul>
+<ul>
+	<li> -f
+	<br>
+	Force overwrite.  Normally, <tt>extract</tt> will refuse to overwrite an
+	output file that already exists.  With this option, <tt>extract</tt>
+	will overwrite the output file and replace it with new bytecode.
+	<p>
+
 	<li>-func <function>
 	<br>
 	Extract the specified function from the LLVM bytecode.
@@ -48,6 +54,12 @@
 	<li> -help
 	<br>
 	Print a summary of command line options.
+	<p>
+
+	<li> -o <filename>
+	<br>
+	Specify the output filename.  If filename is "-" (the default), then
+	<tt>extract</tt> sends its output to standard output.
 	<p>
 </ul>
 


Index: llvm/docs/CommandGuide/gccld.html
diff -u llvm/docs/CommandGuide/gccld.html:1.8 llvm/docs/CommandGuide/gccld.html:1.8.4.1
--- llvm/docs/CommandGuide/gccld.html:1.8	Sun Nov 16 17:39:11 2003
+++ llvm/docs/CommandGuide/gccld.html	Mon Mar  1 17:56:01 2004
@@ -32,24 +32,35 @@
 
 <h4>Search Order</h4>
 
+<p>
 When looking for objects specified on the command line, <tt>gccld</tt> will
 search for the object first in the current directory and then in the directory
 specified by the <tt>LLVM_LIB_SEARCH_PATH</tt> environment variable.  If it
 cannot find the object, it fails.
-<p>
+</p>
 
+<p>
 When looking for a library specified with the -l option, <tt>gccld</tt> first
 attempts to load a file with that name from the current directory.  If that
 fails, it looks for lib<library>.bc, lib<library>.a, or
-lib<library>.so, in that order, in each directory added to the library
-search path with the -L option.  These directories are searched in order they
+lib<library>.<shared library extension>, in that order, in each
+directory added to the library search path with the -L option.  These
+directories are searched in the order they
 were specified.  If the library cannot be located, then <tt>gccld</tt> looks in
 the directory specified by the <tt>LLVM_LIB_SEARCH_PATH</tt> environment
-variable.  If it does not find lib<library>.[bc | a | so] there, it fails.
+variable.  If it does not find a library there, it fails.
+</p>
 
+<p>
+The shared library extension is usually <tt>.so</tt>, but it may differ
+depending upon the system.
+</p>
+
+<p>
 The -L option is global.  It does not matter where it is specified in the list
 of command line arguments; the directory is simply added to the search path and
 is applied to all libraries, preceding or succeeding, in the command line.
+</p>
 
 <h4>Link order</h4>
 
@@ -131,9 +142,9 @@
 	<br>
 	Specify libraries to include when linking the output file.  When
 	linking, <tt>gccld</tt> will first attempt to load a file with the
-	pathname library.  If that fails, it will then attempt to load
-	lib<library>.bc, lib<library>.a, and lib<library>.so,
-	in that order.
+	pathname <tt>library</tt>.  If that fails, it will then attempt to load
+	lib<library>.bc, lib<library>.a, and
+  lib<library>.<shared library extension>, in that order.
 	<p>
 
 	<li> -link-as-library


Index: llvm/docs/CommandGuide/llc.html
diff -u llvm/docs/CommandGuide/llc.html:1.6 llvm/docs/CommandGuide/llc.html:1.6.6.1
--- llvm/docs/CommandGuide/llc.html:1.6	Tue Oct  7 15:12:04 2003
+++ llvm/docs/CommandGuide/llc.html	Mon Mar  1 17:56:01 2004
@@ -95,6 +95,9 @@
 
 		<di> sparc
 		<dd>SPARC V9</dd>
+
+		<di> c
+		<dd>Emit C code</dd>
 	</dl>
 	<p>
 
@@ -147,6 +150,20 @@
 
 		<di> local
 		<dd>Local register allocator</dd>
+
+		<di> linearscan
+		<dd>Linear scan global register allocator (experimental)</dd>
+        <p>
+
+	<li>-spiller=<sp>
+	<br>
+        Specify the spiller to use for register allocators that support it.
+        Currently this option is used by the linear scan register
+        allocator. The default is <i>local</i>.
+        Valid spillers are:
+	<dl compact>
+		<di> local
+		<dd>Local spiller</dd>
 	</dl>
 	<p>
 
@@ -157,11 +174,6 @@
 	<li>-disable-peephole
 	<br>
 	Disable peephole optimization pass.
-	<p>
-
-	<li>-disable-preopt
-	<br>
-	Disable optimizations prior to instruction selection.
 	<p>
 
 	<li>-disable-sched


Index: llvm/docs/CommandGuide/llvm-prof.html
diff -u llvm/docs/CommandGuide/llvm-prof.html:1.3 llvm/docs/CommandGuide/llvm-prof.html:1.3.4.1
--- llvm/docs/CommandGuide/llvm-prof.html:1.3	Thu Nov  6 14:29:33 2003
+++ llvm/docs/CommandGuide/llvm-prof.html	Mon Mar  1 17:56:01 2004
@@ -10,18 +10,19 @@
 <tt>llvm-prof</tt>
 
 <h3>SYNOPSIS</h3>
-<tt>llvm-prof [options] [bytecode file] [LLVM passes]</tt>
+<tt>llvm-prof [options] [bytecode file] [llvmprof.out]</tt>
 
 <h3>DESCRIPTION</h3>
 
-The <tt>llvm-prof</tt> tool reads in an '<tt>llvmprof.out</tt>' file, a bytecode
-file for the program, and produces a human readable report, suitable for
-determining where the program hotspots are.<p>
+The <tt>llvm-prof</tt> tool reads in an '<tt>llvmprof.out</tt>' file (which can
+optionally use a specific file with the third program argument), a bytecode file
+for the program, and produces a human readable report, suitable for determining
+where the program hotspots are.<p>
 
-This program is often used in conjunction with the <tt>utils/profile.sh</tt>
+This program is often used in conjunction with the <tt>utils/profile.pl</tt>
 script.  This script automatically instruments a program, runs it with the JIT,
 then runs <tt>llvm-prof</tt> to format a report.  To get more information about
-<tt>utils/profile.sh</tt>, execute it with the <tt>--help</tt> option.<p>
+<tt>utils/profile.pl</tt>, execute it with the <tt>--help</tt> option.<p>
 
 <h3>OPTIONS</h3>
 


Index: llvm/docs/CommandGuide/opt.html
diff -u llvm/docs/CommandGuide/opt.html:1.8 llvm/docs/CommandGuide/opt.html:1.8.6.1
--- llvm/docs/CommandGuide/opt.html:1.8	Sun Oct 19 13:08:57 2003
+++ llvm/docs/CommandGuide/opt.html	Mon Mar  1 17:56:01 2004
@@ -88,14 +88,14 @@
 	Quiet mode.  Do not print messages on whether the program was modified.
 	<p>
 
-	<li> -load <plugin.so>
+	<li> -load <plugin>
 	<br>
-	Load the dynamic object <plugin.so>.  This object should register new
+	Load the dynamic object <plugin>.  This object should register new
 	optimization passes.  Once loaded, the object will add new command line
 	options to enable various optimizations.  To see the new complete list
 	of optimizations, use the -help and -load options together:
 	<p>
-	<tt>opt -load  <plugin.so> -help</tt>
+	<tt>opt -load  <plugin> -help</tt>
 	<p>
 
 	<li> -p





More information about the llvm-commits mailing list