[ira-merge] A branch to track IRA regressions

[ira-merge] A branch to track IRA regressions

[H.J. Lu]

Hi,

I will create an IRA merge branch:

svn://gcc.gnu.org/svn/gcc/branches/ira-merge

It will be branched from trunk at revision 139590.
It will be used to track IRA related regressions.
Only IRA fixes from trunk will be applied to this
branch. Its goal is there should no "make check"
regressions against trunk at revision 139589 on
Linux/ia32, Linux/x86-64 and Linux/ia64. We will
also test SPEC CPU 2000/2006 on Linux/ia32 and
Linux/x86-64.

This branch will be maintained by me and Vladimir.

Thanks.

-- 
H.J.

Update wwwdocs for ira-merge and AVX

[H.J. Lu]

On Sat, Sep 06, 2008 at 07:07:47PM +0200, Gerald Pfeifer wrote:
> Hi HJ,
> 
> On Tue, 2 Sep 2008, H.J. Lu wrote:
> > I will create an IRA merge branch:
> > 
> > svn://gcc.gnu.org/svn/gcc/branches/ira-merge
> 
> even if it's not going to live for an extended period, would you
> mind documenting this at http://gcc.gnu.org/svn.html ?
> 

Here is a patch for ira-merge as well as AVX and stack alignment.

H.J.
---
Index: svn.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/svn.html,v
retrieving revision 1.97
diff -u -p -r1.97 svn.html
--- svn.html	20 Aug 2008 05:08:24 -0000	1.97
+++ svn.html	6 Sep 2008 17:30:15 -0000
@@ -245,6 +245,17 @@ list therefore provides only some repres
   will be merged with mainline from time to time.  Patches will be
   marked with the tag <code>[ira]</code> in the subject line.</dd>
 
+  <dt>ira-merge</dt>
+  <dd>This branch contains bug fixes for the Integrated Register Allocator
+  (IRA).  It is branched from trunk at revision 139590 when IRA was
+  merged into trunk.  It is used to track IRA related regressions.
+  Only IRA fixes from trunk will be applied to this branch. Its goal is
+  there should be no "make check"  and performance regressions against
+  trunk at revision 139589.  The branch is maintained by H.J. Lu &lt;<a
+  href="mailto:hjl.tools@gmail.com">hjl.tools@gmail.com</a>&gt; and
+  Vladimir Makarov &lt;
+  <a href="mailto:vmakarov@redhat.com">vmakarov@redhat.com</a>&gt;.</dd>
+
   <dt>sel-sched-branch</dt>
   <dd>This branch contains the implementation of the selective scheduling
   approach.  The goal of the branch is to provide more aggressive scheduler 
Index: gcc-4.4/changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-4.4/changes.html,v
retrieving revision 1.24
diff -u -p -r1.24 changes.html
--- gcc-4.4/changes.html	6 Sep 2008 08:15:46 -0000	1.24
+++ gcc-4.4/changes.html	6 Sep 2008 17:30:16 -0000
@@ -165,6 +165,10 @@
 	available via <code>-maes</code>.</li>
     <li>Support for Intel PCLMUL built-in function and code generation are
 	available via <code>-mpclmul</code>.</li>
+    <li>Support for Intel AVX built-in functions and code generation are
+	available via <code>-mavx</code>.</li>
+    <li>Automatically align stack for local variables with alignment
+	requirement. </li>
     <li>GCC can now utilize the SVML library for vectorizing calls to
 	a set of C99 functions if <code>-mveclibabi=svml</code> is specified
 	and you link to an SVML ABI compatible library.</li>

Re: Update wwwdocs for ira-merge and AVX

[Gerald Pfeifer]

On Sat, 6 Sep 2008, H.J. Lu wrote:
> Here is a patch for ira-merge as well as AVX and stack alignment.

Thanks.  Please go ahead and commit after addressing to two suggestions
for the second part.

> +    <li>Support for Intel AVX built-in functions and code generation are
> +	available via <code>-mavx</code>.</li>

are -> is

> +    <li>Automatically align stack for local variables with alignment
> +	requirement. </li>

stack -> the stack, I believe.

Gerald

Re: Update wwwdocs for ira-merge and AVX

[H.J. Lu]

On Sun, Sep 07, 2008 at 12:13:11AM +0200, Gerald Pfeifer wrote:
> On Sat, 6 Sep 2008, H.J. Lu wrote:
> > Here is a patch for ira-merge as well as AVX and stack alignment.
> 
> Thanks.  Please go ahead and commit after addressing to two suggestions
> for the second part.
> 
> > +    <li>Support for Intel AVX built-in functions and code generation are
> > +	available via <code>-mavx</code>.</li>
> 
> are -> is
> 
> > +    <li>Automatically align stack for local variables with alignment
> > +	requirement. </li>
> 
> stack -> the stack, I believe.
> 

This is the patch I checked in.

Thanks.


H.J.
----
Index: svn.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/svn.html,v
retrieving revision 1.97
diff -u -p -r1.97 svn.html
--- svn.html	20 Aug 2008 05:08:24 -0000	1.97
+++ svn.html	6 Sep 2008 23:04:33 -0000
@@ -245,6 +245,17 @@ list therefore provides only some repres
   will be merged with mainline from time to time.  Patches will be
   marked with the tag <code>[ira]</code> in the subject line.</dd>
 
+  <dt>ira-merge</dt>
+  <dd>This branch contains bug fixes for the Integrated Register Allocator
+  (IRA).  It is branched from trunk at revision 139590 when IRA was
+  merged into trunk.  It is used to track IRA related regressions.
+  Only IRA fixes from trunk will be applied to this branch. Its goal is
+  there should be no "make check"  and performance regressions against
+  trunk at revision 139589.  The branch is maintained by H.J. Lu &lt;<a
+  href="mailto:hjl.tools@gmail.com">hjl.tools@gmail.com</a>&gt; and
+  Vladimir Makarov &lt;
+  <a href="mailto:vmakarov@redhat.com">vmakarov@redhat.com</a>&gt;.</dd>
+
   <dt>sel-sched-branch</dt>
   <dd>This branch contains the implementation of the selective scheduling
   approach.  The goal of the branch is to provide more aggressive scheduler 
Index: gcc-4.4/changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-4.4/changes.html,v
retrieving revision 1.24
diff -u -p -r1.24 changes.html
--- gcc-4.4/changes.html	6 Sep 2008 08:15:46 -0000	1.24
+++ gcc-4.4/changes.html	6 Sep 2008 23:04:33 -0000
@@ -161,10 +161,14 @@
 
 <h3>IA-32/x86-64</h3>
   <ul>
-    <li>Support for Intel AES built-in functions and code generation are
+    <li>Support for Intel AES built-in functions and code generation is
 	available via <code>-maes</code>.</li>
-    <li>Support for Intel PCLMUL built-in function and code generation are
+    <li>Support for Intel PCLMUL built-in function and code generation is
 	available via <code>-mpclmul</code>.</li>
+    <li>Support for Intel AVX built-in functions and code generation is
+	available via <code>-mavx</code>.</li>
+    <li>Automatically align the stack for local variables with alignment
+	requirement. </li>
     <li>GCC can now utilize the SVML library for vectorizing calls to
 	a set of C99 functions if <code>-mveclibabi=svml</code> is specified
 	and you link to an SVML ABI compatible library.</li>