[WEB][PATCH] Describe -pg and LTO changes

[WEB][PATCH] Describe -pg and LTO changes

[Andi Kleen]

This patch describes some user visible changes that were 
added to gcc 5.

Ok to commit? 

-Andi


Index: htdocs/gcc-5/changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-5/changes.html,v
retrieving revision 1.25
diff -u -r1.25 changes.html
--- htdocs/gcc-5/changes.html	14 Nov 2014 21:32:32 -0000	1.25
+++ htdocs/gcc-5/changes.html	16 Nov 2014 20:03:07 -0000
@@ -154,6 +154,11 @@
 	about qualifiers on pointers being discarded via a new warning option
 	<code>-Wno-discarded-qualifiers</code>.</li>
     <li>The C front end now generates more precise caret diagnostics.</li>
+    <li>The -pg option now only affects the current file in a LTO build.</li>
+    <li>A new no_reorder attribute has been added, that prevents reordering
+	of a specific symbol against other such symbols or inline assembler.
+	This is a more focussed alternative to 
+	<code>-fno-toplevel-reorder</code>.
   </ul>
 
 <h3 id="cxx">C++</h3>
@@ -295,6 +300,17 @@
 
    </ul>
 
+<h3 id="x86">IA-32/x86-64</h3>
+  <ul>
+	<li>The new <code>-mrecord-mcount</code> option for <code>-pg</code>
+	generates a Linux kernel style table of pointers to mcount or
+	__fentry__ calls at the beginning of functions. The new
+	<code>-mnop-mcount</code> option in addition also generates nops in
+	place of the __fentry__ or mcount call, so that a call per function
+	can be later patched in. This can be used for low overhead tracing or
+	hot code patching.</li>
+  </ul>
+
 <h2 id="os">Operating Systems</h2>
 
   <h3 id="dragonfly">DragonFly BSD</h3>
@@ -304,7 +320,11 @@
   </ul>
 
 <!-- h2>Documentation improvements</h2-->
-<!-- h2>Other significant improvements</h2 -->
-
+<h2>Other significant improvements</h2>
+  <h3 id="gcc-ar"></h3>
+  <ul>
+    <li>The <code>gcc-ar, gcc-nm, gcc-ranlib</code> wrappers now
+	understand a <code>-B</code> option to set the compiler to use.</li>
+  </ul>
 </body>
 </html>

Re: [WEB][PATCH] Describe -pg and LTO changes

[Jeff Law]

On 11/16/14 13:02, Andi Kleen wrote:
>
> This patch describes some user visible changes that were
> added to gcc 5.
>
> Ok to commit?
Yes, this is fine.  Thanks,
Jeff

Re: [WEB][PATCH] Describe -pg and LTO changes

[Jan Hubicka]

> On 11/16/14 13:02, Andi Kleen wrote:
> >
> >This patch describes some user visible changes that were
> >added to gcc 5.
> >
> >Ok to commit?
> Yes, this is fine.  Thanks,
... which remind me that I also wrote some update that apparently was never approved

Index: changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-5/changes.html,v
retrieving revision 1.26
diff -u -r1.26 changes.html
--- changes.html	17 Nov 2014 19:14:47 -0000	1.26
+++ changes.html	17 Nov 2014 19:51:15 -0000
@@ -21,6 +21,51 @@
 <h2 id="general">General Optimizer Improvements</h2>
 
   <ul>
+    <li>Inter-procedural optimization improvements:
+    <ul>
+     <li>Devirtualization pass was significantly improved by adding
+	 better support for speculative devirtualization and dynamic type
+	 detection. About 50% of virtual calls in Firefox are speculatively
+	 devirtualized during link-time optimization.
+     <li>New comdat localization pass lets linker to eliminate more dead code
+	 in presence of C++ inline functions.</li>
+     <li>Virtual tables are now optimized. Local aliases are used to reduce
+	 dynamic linking time of C++ virtual tables on ELF targets and
+	 data alignment has been reduced to limit data segment bloat.</li>
+     <li>New <code>-fno-semantic-interposition</code> flag can be used
+	 to improve code quality of shared libraries where interposition of
+	 exported symbols is not allowed.</li>
+     <li>Write-only variables are now detected and optimized out.</li>
+     <li>With profile feedback the function inliner can now bypass
+	 <code>--param inline-insns-auto</code> and <code>--param
+	 inline-insns-single</code> limits for hot calls.</li>
+     <li>IPA reference pass was significantly sped up making it feasible
+	 to enable <code>-fipa-reference</code> with
+	 <code>-fprofile-generage</code>. This also solve bottleneck
+	 seen when optimizing Chromium with link time optimization.</li>
+     <li>Symbol table and call-graph API was reworked to C++ and
+	 simplified.</li>
+    </ul></li>
+    <li>Link-time optimization improvements:
+    <ul>
+      <li>New One Definition Rule based merging of C++ types implemented.
+	  Type merging enables better devirtualization and alias analysis.
+	  Streaming extra information needed to merge types adds about 2-6% of
+	  memory size and object size increase. This can be controlled by
+	  <code>-flto-odr-type-merging</code>.</li>
+      <li>GCC bootstrap now use slim LTO object files.</li>
+      <li>Memory usage and link times was improved.  Tree merging was sped up,
+	  memory usage of GIMPLE declarations and types was reduced, and,
+	  support for on-demand streaming of variable constructors was added.</li>
+    </ul></li>
+    <li>Feedback directed optimization improvements:
+    <ul>
+      <li>Profile precision was improved in presence of C++ inline and extern
+	  inline functions.</li>
+      <li>New <code>gcov-tool</code> to manipulate profiles.</li>
+      <li>Profile is now more tolerant to source file changes (this can be
+	  controlled by <code>--param profile-func-internal-id</code>).</li>
+    </ul></li>
     <li>UndefinedBehaviorSanitizer gained a few new sanitization options:
     <ul>
       <li><code>-fsanitize=float-divide-by-zero</code>: detect floating-point
@@ -77,6 +122,9 @@
     <li>Full support for <a href="https://www.cilkplus.org/">Cilk Plus</a>
 	has been added to the GCC compiler. Cilk Plus is an extension to
 	the C and C++ languages to support data and task parallelism.</li>
+    <li>New attribute <code>no_reorder</code> prevents reordering of selected symbols.
+	This enables to link-time optimize Linux kernel without need to use
+	<code>-fno-toplevel-reorder</code> that disable several optimizations.</li>
     <li>New preprocessor constructs, <code>__has_include</code>
         and <code>__has_include_next</code>, to test the availability of headers
         have been added.<br/>
@@ -236,6 +284,15 @@
     </li>
     <li>New random number distributions <code>logistic_distribution</code> and
       <code>uniform_on_sphere_distribution</code> as extensions.</li>
+    <li>New One Definition Rule violation warning (controlled by <code>-Wodr</code)
+        detects mismatches in type definitions and virtual table contents
+	during link-time optimization.</li>
+    <li>New warnings <code>-Wsuggest-final-types</code> and
+	<code>-Wsuggest-final-methods</code> helps developers
+	to annotate programs by <code>final</code> specifiers (or anonymous
+	namespaces) in the cases where code generation improves.
+	These warnings can be used at compile time, but they are more
+	useful in combination with link-time optimization.</li>
     <li><a href="https://sourceware.org/gdb/current/onlinedocs/gdb/Xmethods-In-Python.html">GDB
       Xmethods</a> for Sequence Containers and <code>std::unique_ptr</code>;</li>
   </ul>

Re: [WEB][PATCH] Describe -pg and LTO changes

[Jeff Law]

On 11/17/14 12:52, Jan Hubicka wrote:
>> On 11/16/14 13:02, Andi Kleen wrote:
>>>
>>> This patch describes some user visible changes that were
>>> added to gcc 5.
>>>
>>> Ok to commit?
>> Yes, this is fine.  Thanks,
> ... which remind me that I also wrote some update that apparently was never approved
Looks good to me.

jeff

Re: [WEB][PATCH] Describe -pg and LTO changes

[Andi Kleen]

Jan Hubicka <hubicka@ucw.cz> writes:
>  	the C and C++ languages to support data and task parallelism.</li>
> +    <li>New attribute <code>no_reorder</code> prevents reordering of selected symbols.
> +	This enables to link-time optimize Linux kernel without need to use
> +	<code>-fno-toplevel-reorder</code> that disable several
> optimizations.</li>

I already had that one. One of us needs to drop.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only

Re: [WEB][PATCH] Describe -pg and LTO changes

[Gerald Pfeifer]

On Sun, 16 Nov 2014, Andi Kleen wrote:
> This patch describes some user visible changes that were 
> added to gcc 5.

Thanks, Andi!

I added some <code> environments, broke up a long one, changed
"a LTO build" to "an LTO build",

Index: changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-5/changes.html,v
retrieving revision 1.97
diff -u -r1.97 changes.html
--- changes.html	8 Apr 2015 07:26:23 -0000	1.97
+++ changes.html	8 Apr 2015 16:39:59 -0000
@@ -330,7 +330,8 @@
 	about qualifiers on pointers being discarded via a new warning option
 	<code>-Wno-discarded-qualifiers</code>.</li>
     <li>The C front end now generates more precise caret diagnostics.</li>
-    <li>The -pg option now only affects the current file in a LTO build.</li>
+    <li>The <code>-pg</code> command-line option now only affects the current
+        file in an LTO build.</li>
   </ul>
 
 <h3 id="cxx">C++</h3>
@@ -684,12 +685,13 @@
 	<code>-mavx512ifma</code> and for AVX-512 Vector Bit Manipulation
 	Instructions: <code>-mavx512vbmi</code>.</li>
 	<li>The new <code>-mrecord-mcount</code> option for <code>-pg</code>
-	generates a Linux kernel style table of pointers to mcount or
-	__fentry__ calls at the beginning of functions. The new
-	<code>-mnop-mcount</code> option in addition also generates nops in
-	place of the __fentry__ or mcount call, so that a call per function
-	can be later patched in. This can be used for low overhead tracing or
-	hot code patching.</li>
+	generates a Linux kernel style table of pointers to
+	<code>mcount</code> or <code>__fentry__</code> calls at the beginning
+	of functions. The new <code>-mnop-mcount</code> option in addition
+	also generates nops in place of the <code>__fentry__</code> or
+	<code>mcount</code> call, so that a call per function can be later
+	patched in. This can be used for low overhead tracing or hot code
+	patching.</li>
 	<li>The new <code>-malign-data</code> option controls how
 	GCC aligns variables.  <code>-malign-data=compat</code> uses
 	increased alignment compatible with GCC 4.8 and earlier,
@@ -838,8 +840,9 @@
 <h2>Other significant improvements</h2>
   <h3 id="gcc-ar"></h3>
   <ul>
-    <li>The <code>gcc-ar, gcc-nm, gcc-ranlib</code> wrappers now
-	understand a <code>-B</code> option to set the compiler to use.</li>
+    <li>The <code>gcc-ar</code>, <code>gcc-nm</code>, <code>gcc-ranlib</code>
+      wrappers now understand a <code>-B</code> option to set the compiler
+      to use.</li>
   </ul>
   <h3 id="driver"></h3>
   <ul>