http://gcc.gnu.org/gcc-3.1/changes.html

http://gcc.gnu.org/gcc-3.1/changes.html

[keith.briggs]

http://gcc.gnu.org/gcc-3.1/changes.html:
> A new __builtin_prefetch instrics is available 

What's an "instrics"?

Keith

	Dr. Keith M. Briggs
	Senior Mathematician, Complexity Research, BTexact Technologies

	email: Keith.Briggs@bt.com
	phone: +44(0)1473  work: 641 911 home: 610 517  fax: 647 410
        web:  (work)     http://more.btexact.com/people/briggsk2/ 
        web:  (private) http://members.lycos.co.uk/keithmbriggs/
	mail: Keith Briggs, Antares 2pp5, Adastral Park, Martlesham, Suffolk
IP5 3RE, UK

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Kurt Wall]

Scribbling feverishly on May 16, keith.briggs@bt.com managed to emit:
> http://gcc.gnu.org/gcc-3.1/changes.html:
> > A new __builtin_prefetch instrics is available 
> 
> What's an "instrics"?

Perchance a distant relation of "intrinsics"?

Kurt
-- 
Fortune's Office Door Sign of the Week:

	Incorrigible punster -- Do not incorrige.

PATCH for Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Gerald Pfeifer]

[ Reply-To: gcc-patches ]

On Thu, 16 May 2002, Kurt Wall wrote:
>>> A new __builtin_prefetch instrics is available
>> What's an "instrics"?
> Perchance a distant relation of "intrinsics"?

I committed the patch below.

Index: changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-3.1/changes.html,v
retrieving revision 1.41
diff -u -3 -p -r1.41 changes.html
--- changes.html	16 May 2002 02:13:10 -0000	1.41
+++ changes.html	17 May 2002 09:12:52 -0000
@@ -62,7 +62,7 @@

 <li>Support for data prefetching instructions has been added to the GCC
     back end and several targets.  A new <code>__builtin_prefetch</code>
-    instrics is available to explicitly insert prefetch instructions and
+    intrinsics is available to explicitly insert prefetch instructions and
     experimental support for loop array prefetching has been added (see
     <code>-fprefetch-loop-array</code> documentation).
 </li>

RE: http://gcc.gnu.org/gcc-3.1/changes.html

[keith.briggs]

Maybe... that page also has "intrics" twice.
Keith

	Dr. Keith M. Briggs
	Senior Mathematician, Complexity Research, BTexact Technologies

	email: Keith.Briggs@bt.com
	phone: +44(0)1473  work: 641 911 home: 610 517  fax: 647 410
        web:  (work)     http://more.btexact.com/people/briggsk2/ 
        web:  (private) http://members.lycos.co.uk/keithmbriggs/
	mail: Keith Briggs, Antares 2pp5, Adastral Park, Martlesham, Suffolk
IP5 3RE, UK




> -----Original Message-----
> From:	Kurt Wall [SMTP:kwall@kurtwerks.com]
> Sent:	Thursday, May 16, 2002 12:36 PM
> To:	keith.briggs@bt.com
> Cc:	gcc@gnu.org; mark@codesourcery.com
> Subject:	Re: http://gcc.gnu.org/gcc-3.1/changes.html
> 
> Scribbling feverishly on May 16, keith.briggs@bt.com managed to emit:
> > http://gcc.gnu.org/gcc-3.1/changes.html:
> > > A new __builtin_prefetch instrics is available 
> > 
> > What's an "instrics"?
> 
> Perchance a distant relation of "intrinsics"?
> 
> Kurt

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Bryce McKinlay]

Aldy Hernandez wrote:

>Hmm, i'm wondering whether we should add altivec to the list of new
>things, or just wait till ti settles down in 3.2?
>
>Comments?
>

You could add it with a note saying that its still experimental.

regards

Bryce.

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Aldy Hernandez]

>>>>> "Gerald" == Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at> writes:

 > I have been reminded that http://gcc.gnu.org/gcc-3.1/changes.html "is in
 > a pretty sorry state", and there's not much there apart from CPP and GCJ,
 > especially platform specific fixes, optimizations and C++ issues are
 > scarce.

Hmm, i'm wondering whether we should add altivec to the list of new
things, or just wait till ti settles down in 3.2?

Comments?

Aldy

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Gerald Pfeifer]

On 19 Apr 2002, Tom Tromey wrote:
> The Fortran link now appears in the Java section.

Oops. Fixes thusly.

Thanks for pointing this out.

Gerald

Index: changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-3.1/changes.html,v
retrieving revision 1.13
diff -u -3 -p -r1.13 changes.html
--- changes.html	19 Apr 2002 16:02:22 -0000	1.13
+++ changes.html	20 Apr 2002 10:49:47 -0000
@@ -65,10 +65,13 @@
 <li><code>java.lang.Character</code> was rewritten to comply with
     the Unicode 3.0 standard.</li>
 <li>Partial support for many more locales was added to libgcj.</li>
-<li>Fortran improvements are listed in
-    <a href="http://gcc.gnu.org/onlinedocs/g77_news.html">the Fortran
-    documentation</a>.</li>
 </ul>
+
+<h3>Fortran</h3>
+
+<p>Fortran improvements are listed in
+<a href="http://gcc.gnu.org/onlinedocs/g77_news.html">the Fortran
+documentation</a>.</p>


 <h2>New Targets and Target Specific Improvements</h2>

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Tom Tromey]

>>>>> "Gerald" == Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at> writes:

Gerald> Based on that, I applied the patch below, which also adds a
Gerald> new section for C/C++ and one for Java, and has a couple of
Gerald> other minor fixes.

The Fortran link now appears in the Java section.

Tom

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Gerald Pfeifer]

[ gcc -> gcc-patches ]

On Fri, 19 Apr 2002 kelley.r.cook@gm.com wrote:
> Gerald,
> This is off the top of my head.  It also corrects one small typo.

Thanks for your patch!

Based on that, I applied the patch below, which also adds a new section
for C/C++ and one for Java, and has a couple of other minor fixes.

(For example, it's UltraSPARC, not Ultrasparc, and there must not be
a line break directly after "<a href=...>", or we will have an extra
blank.)

Gerald

PS: Andreas, Jan, please have a look at the x86-related changes, and
feel free to update them as you see fit.


Index: changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-3.1/changes.html,v
retrieving revision 1.12
diff -u -3 -p -w -r1.12 changes.html
--- changes.html	17 Apr 2002 13:41:43 -0000	1.12
+++ changes.html	19 Apr 2002 15:57:24 -0000
@@ -29,6 +29,8 @@

 <h2>New Languages and Language specific improvements</h2>

+<h3>C/C++</h3>
+
   <ul>
     <li>A few more <a href="c99status.html">ISO C99 features</a>.</li>
     <li>The preprocessor is 10-50% faster than the preprocessor in GCC
@@ -39,7 +41,11 @@
         GCC 3.0, often significantly so.  On normal input files, it
         typically consumes less memory than pre-3.0 cccp-based GCC,
         too.</li>
+</ul>

+<h3>Java</h3>
+
+<ul>
     <li>libgcj now includes RMI, <code>java.lang.ref.*</code>,
         <code>javax.naming</code>, and <code>javax.transaction</code>.</li>
     <li>Property files and other system resources can be compiled into
@@ -50,7 +56,7 @@
     <li>gcj can now use builtins functions for certain known methods,
         for instance <code>Math.cos</code>.  gcj also has a new
         <code>--no-store-checks</code> optimization option.</li>
-    <li>The following thrd party interface standards were added to
+<li>The following third party interface standards were added to
         libgcj: <code>org.w3c.dom</code> and <code>org.xml.sax</code>.</li>
     <li><code>java.security</code> has been merged with GNU
         Classpath. The new package is now JDK 1.2 compliant, and much
@@ -68,19 +74,47 @@
 <h2>New Targets and Target Specific Improvements</h2>

   <ul>
-    <li><a href="http://www.suse.com">SuSE Labs</a> developers Jan
-    Hubicka, Bo Thorsen and Andreas Jaeger have contributed a port to
-    the AMD x86-64 architecture.  For more information on x86-64 see
-    <a href="http://www.x86-64.org">http://www.x86-64.org</a>.</li>
     <li>Hans-Peter Nilsson has contributed a port to <a
     href="http://www-cs-faculty.stanford.edu/~knuth/mmix.html">MMIX</a>,
     the CPU architecture used in new editions of Donald E. Knuth's The Art
     of Computer Programming.</li>
+
     <li><a href="http://www.axis.com/">Axis Communications</a> has
     contributed its port to the CRIS CPU architecture, used in the ETRAX
     system-on-a-chip series.  See <a
-    href="http://developer.axis.com/">developer.axis.com</a> for technical
-    information.</li>
+    href="http://developer.axis.com/">Axis' developer site</a> for
+    technical information.</li>
+
+<li>UltraSPARC is fully supported in 64-bit mode.  The option
+    <code>-m64</code> enables it.</li>
+
+<li>The x86 back-end has had some noticeable work done to it.
+    <ul>
+      <li><a href="http://www.suse.com/">SuSE Labs</a> developers Jan
+      Hubicka, Bo Thorsen and Andreas Jaeger have contributed a port to
+      the AMD x86-64 architecture.  For more information on x86-64 see
+      <a href="http://www.x86-64.org/">http://www.x86-64.org</a>.</li>
+
+      <li>The compiler now supports MMX, 3DNow!, SSE, and SSE2 instructions.
+      Options <code>-mmmx</code>, <code>-m3dnow</code>, <code>-msse</code>,
+      and <code>-msse2</code> will enable the respective instruction sets.</li>
+
+      <li>Following those improvements, targets for Pentium MMX, K6-2, K6-3,
+      Pentium III, Pentium 4, and Athlon 4 Mobile/XP/MP were added.
+      Refer to the documentation on <code>-march=</code> and
+      <code>-mcpu=</code> options for details.</li>
+
+      <li>For those targets that support it, <code>-mfpmath=sse</code> will
+      cause the compiler to generate SSE/SSE2 instructions for floating point
+      math instead of x87 instructions.  Usually, this will lead to quicker
+      code &mdash; especially on the Pentium 4.</li>
+
+      <li>Prefetch support has been added to the Pentium III, Pentium 4, and
+      Athlon series.</li>
+   </ul></li>
+
+<li>Altivec support has been added for the PowerPC variants that
+    support it.  See <code>-maltivec</code>.</li>
   </ul>

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[kelley.r.cook]

[-- Attachment #1: Type: text/plain, Size: 114 bytes --]

Gerald,
This is off the top of my head.  It also corrects one small typo.

(See attached file: changes.html.diff)

[-- Attachment #2: changes.html.diff --]
[-- Type: application/octet-stream, Size: 3495 bytes --]

--- changes.html.orig	Fri Apr 19 10:54:21 2002
+++ changes.html	Fri Apr 19 11:06:51 2002
@@ -82,7 +82,7 @@
     <li>gcj can now use builtins functions for certain known methods,
         for instance <code>Math.cos</code>.  gcj also has a new
         <code>--no-store-checks</code> optimization option.</li>
-    <li>The following thrd party interface standards were added to
+    <li>The following third party interface standards were added to
         libgcj: <code>org.w3c.dom</code> and <code>org.xml.sax</code>.</li>
     <li><code>java.security</code> has been merged with GNU
         Classpath. The new package is now JDK 1.2 compliant, and much
@@ -100,19 +100,46 @@
 <h2>New Targets and Target Specific Improvements</h2>
 
   <ul>
-    <li><a href="http://www.suse.com">SuSE Labs</a> developers Jan
-    Hubicka, Bo Thorsen and Andreas Jaeger have contributed a port to
-    the AMD x86-64 architecture.  For more information on x86-64 see
-    <a href="http://www.x86-64.org">http://www.x86-64.org</a>.</li>
-    <li>Hans-Peter Nilsson has contributed a port to <a
-    href="http://www-cs-faculty.stanford.edu/~knuth/mmix.html">MMIX</a>,
+    <li>Hans-Peter Nilsson has contributed a port to 
+    <a href="http://www-cs-faculty.stanford.edu/~knuth/mmix.html">MMIX</a>,
     the CPU architecture used in new editions of Donald E. Knuth's The Art
     of Computer Programming.</li>
+    
     <li><a href="http://www.axis.com/">Axis Communications</a> has
     contributed its port to the CRIS CPU architecture, used in the ETRAX
-    system-on-a-chip series.  See <a
-    href="http://developer.axis.com/">developer.axis.com</a> for technical
-    information.</li>
+    system-on-a-chip series.  See <a href="http://developer.axis.com/">
+    Axis' developer site</a> for technical information.</li>
+    
+    <li>Ultrasparc is fully supported in 64-bit mode.  The option 
+    <code>-m64</code> enables it.</li>
+    
+    <li>The x86 back-end has had some noticeable work done to it.
+     <ul>
+      <li><a href="http://www.suse.com/">SuSE Labs</a> developers Jan
+      Hubicka, Bo Thorsen and Andreas Jaeger have contributed a port to
+      the AMD x86-64 architecture.  For more information on x86-64 see
+      <a href="http://www.x86-64.org/">http://www.x86-64.org</a>.</li>
+      
+      <li>The compiler now supports MMX, 3DNow!, SSE, and SSE2 instructions.
+      Options <code>-mmmx</code>, <code>-m3dnow</code>, <code>-msse</code>, and 
+      <code>-msse2</code> will enable the respective instruction sets.</li>
+
+      <li>Following those improvements, the following new targets were added
+      Pentium MMX, K6-2, K6-3, Pentium III, Pentium 4, Athlon 4 Mobile/XP/MP.
+      Refer to the documentation on <code>-march=</code> &amp; <code>-mcpu=</code> 
+      options for details.</li>
+
+      <li>For those targets that support it, <code>-mfpmath=sse</code> will
+      cause the compiler to generate SSE/SSE2 instructions for floating point
+      math instead of x87 instructions.  Usually, this will lead to quicker
+      code &mdash; especially on the Pentium 4.</li>
+
+      <li>Prefetch support has been added to the Pentium III, Pentium 4, and
+      Athlon series.</li>
+     </ul></li>
+     
+    <li>Altivec support has been added for the PowerPC variants that support it.
+    See <code>-maltivec</code>.</li>
   </ul>
 
 
@@ -155,5 +182,6 @@
 
 </body>
 </html>
+
 
 

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Joseph S. Myers]

On Fri, 19 Apr 2002, Robert Dewar wrote:

> > We *do* point to the criteria.html file (twice) but say nothing about
> > the changes, improvements, fixes, etc. in GCC 3.1.
> 
> The approach we use with GNAT is to maintain a features file into which
> entries are made for all improvements as they are made. If you don't have
> something like this which you maintain as you go along, it is very hard
> work to reconstruct later.

We do have a list, the news entries on index.html rotated into news.html,
which should be checked for anything to go in changes.html.  (Though many
relevant changes never get on the home page.)

branching.html should probably say to create the changes.html file for the
*next* release (3.2 when branching for 3.1) when the previous release
branch is created.

-- 
Joseph S. Myers
jsm28@cam.ac.uk

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Robert Dewar]

> We *do* point to the criteria.html file (twice) but say nothing about
> the changes, improvements, fixes, etc. in GCC 3.1.

The approach we use with GNAT is to maintain a features file into which
entries are made for all improvements as they are made. If you don't have
something like this which you maintain as you go along, it is very hard
work to reconstruct later.

Re: http://gcc.gnu.org/gcc-3.1/changes.html

[Toon Moene]

Gerald,

> I have been reminded that http://gcc.gnu.org/gcc-3.1/changes.html 
> "is in a pretty sorry state", and there's not much there apart 
> from CPP and GCJ, especially platform specific fixes, 
> optimizations and C++ issues are scarce.

> We should really do a better job advertising our work!

"Advertising" would also mean: Pointing to this document from the front
page ...

We *do* point to the criteria.html file (twice) but say nothing about
the changes, improvements, fixes, etc. in GCC 3.1.

--
Toon Moene, KNMI, PO Box 201, 3730 AE De Bilt, The Netherlands.
Tel. +31302206443, Fax +31302210407,  e-mail moene@knmi.nl
URL: http://www.knmi.nl/hirlam

http://gcc.gnu.org/gcc-3.1/changes.html

[Gerald Pfeifer]

I have been reminded that http://gcc.gnu.org/gcc-3.1/changes.html "is in
a pretty sorry state", and there's not much there apart from CPP and GCJ,
especially platform specific fixes, optimizations and C++ issues are
scarce.

We should really do a better job advertising our work!

Anyone, please feel free to update this page mentioning your changes (or
other changes you are aware of).

Thanks,
Gerald
-- 
Gerald "Jerry" pfeifer@dbai.tuwien.ac.at http://www.dbai.tuwien.ac.at/~pfeifer/