How effect the OpenSource EKOPath the GCC ?
Hi! Currently I have nothing about it found in the mailinglist. So I try to ask it: How effect the OpenSource EKOPath the GCC ? Have a look at the latest press news of PathScale: http://www.pathscale.com/taxonomy/term/27 Have additional a look at this articls of phronix: http://www.phoronix.com/scan.php?page=article&item=pathscale_ekopath4_open&num=1 http://www.phoronix.com/scan.php?page=news_item&px=OTU2OA The important things: EKOPath will be step by step opend. First published opend code you can find at https://github.com/path64 At http://www.pathscale.com/ekopath-compiler-suite you can download nightly builds. To compile a helloworld.c program type $ ~/ekopath-4.0.10/bin/pathcc helloworld.c -o helloworld You can use different benchmarks. For me it seems, that with EKOPath compiled programs are a lot faster then with GCC. The only disadvantage is. that currently EKOPath exists only for 64bit systems with x86_64 CPUs. But how will the OpenSource EKOPath effect GCC? Can GCC make use of some EKOPath code? (EKOPath itself uses parts of GCC) Or can GCC learn from it by studying it? Greatings theuserbl
viewcvs: Python error
http://gcc.gnu.org/viewcvs/trunk/gcc/testsuite/gcc.dg/unroll_1.c?view=markup&pathrev=174762 Since several days now I observe Python exceptions in viewcvs ike that: An Exception Has Occurred Python Traceback Traceback (most recent call last): File "/usr/lib/python2.3/site-packages/viewvc/lib/viewvc.py", line 4318, in main request.run_viewvc() File "/usr/lib/python2.3/site-packages/viewvc/lib/viewvc.py", line 397, in run_viewvc self.view_func(self) File "/usr/lib/python2.3/site-packages/viewvc/lib/viewvc.py", line 1763, in view_markup markup_or_annotate(request, 0) File "/usr/lib/python2.3/site-packages/viewvc/lib/viewvc.py", line 1696, in markup_or_annotate path[-1], mime_type) File "/usr/lib/python2.3/site-packages/viewvc/lib/viewvc.py", line 1509, in markup_stream_pygments from pygments.formatters import HtmlFormatter File "/usr/lib/python2.3/site-packages/pygments/formatters/__init__.py", line 14, in ? from pygments.formatters._mapping import FORMATTERS File "/usr/lib/python2.3/site-packages/pygments/formatters/_mapping.py", line 20, in ? from pygments.formatters.html import HtmlFormatter File "/usr/lib/python2.3/site-packages/pygments/formatters/html.py", line 17, in ? from pygments.token import Token, Text, STANDARD_TYPES File "/usr/lib/python2.3/site-packages/pygments/token.py", line 47, in ? Token = _TokenType() File "/usr/lib/python2.3/site-packages/pygments/token.py", line 26, in __init__ self.subtypes = set() NameError: global name 'set' is not defined
Our Publication Proposal: Your dissertation in book form
Dear Qiong Cai, Some time ago I offered you the possibility of making your academic paper entitled «Profile-guided redundancy elimination» available as printed book. Since I did not hear back from you, I am now wondering if you received my first email. I would appreciate if you could confirm your interest in our publishing house and I will be glad to provide you with detailed information about our services. I am looking forward to receiving a positive response from you. Best Regards, Sarah Lynch Acquisition Editor LAP LAMBERT Academic Publishing GmbH& Co. KG Dudweiler Landstraße 99, 66123 Saarbrücken, Germany Fon +49 681 3720-310 Fax +49 681 3720-3109 s.lynch(at)lappublishing.com / www.lappublishing.com Handelsregister Amtsgericht Saarbrücken HRA 10752 General unlimited partner VDM Management GmbH Handelsregister Amtsgericht Saarbrücken HRB 16777 Board of Directors: Dr. Wolfgang Müller (CEO), Christoph Schulligen, Esther von Krosigk
RE: How effect the OpenSource EKOPath the GCC ?
> You can use different benchmarks. For me it seems, that with EKOPath compiled > programs are a lot faster then with GCC. Hmm.. have tested it again. It seems, that this is only true without GCC-optimization. With the option -O I no longer see any difference between the performance of GCC and EKOPath. :-/ Greatings theuserbl
Re: How effect the OpenSource EKOPath the GCC ?
On 06/18/2011 11:24 AM, theUser BL wrote: Hi! Currently I have nothing about it found in the mailinglist. So I try to ask it: How effect the OpenSource EKOPath the GCC ? Perhaps in this way ? http://gcc.gnu.org/ml/fortran/2011-06/msg00141.html -- Toon Moene - e-mail: t...@moene.org - phone: +31 346 214290 Saturnushof 14, 3738 XG Maartensdijk, The Netherlands At home: http://moene.org/~toon/; weather: http://moene.org/~hirlam/ Progress of GNU Fortran: http://gcc.gnu.org/wiki/GFortran#news
Re: Original commit history for gfortran
On 06/18/2011 12:12 PM, Toon Moene wrote: On 06/18/2011 05:05 AM, Christopher Bergström wrote: Hi We're in the process of considering contributing to gfortran for a special project, but when we started to vet the codebase we hit a bump in lack of commit history. Additional information is here: http://sourceforge.net/projects/gcc-g95 The above gives you the history after the split from the g95 project: http://sourceforge.net/projects/g95 in January 2003. The original commit by Paul Brook of the gcc-g95 repository contents to the GCC repository is here: http://gcc.gnu.org/ml/gcc-cvs/2003-07/msg01087.html -- Toon Moene - e-mail: t...@moene.org - phone: +31 346 214290 Saturnushof 14, 3738 XG Maartensdijk, The Netherlands At home: http://moene.org/~toon/; weather: http://moene.org/~hirlam/ Progress of GNU Fortran: http://gcc.gnu.org/wiki/GFortran#news
Re: Original commit history for gfortran
On 06/18/11 05:16 PM, Toon Moene wrote: On 06/18/2011 12:12 PM, Toon Moene wrote: On 06/18/2011 05:05 AM, Christopher Bergström wrote: Hi We're in the process of considering contributing to gfortran for a special project, but when we started to vet the codebase we hit a bump in lack of commit history. Additional information is here: http://sourceforge.net/projects/gcc-g95 The above gives you the history after the split from the g95 project: http://sourceforge.net/projects/g95 in January 2003. The original commit by Paul Brook of the gcc-g95 repository contents to the GCC repository is here: http://gcc.gnu.org/ml/gcc-cvs/2003-07/msg01087.html Ok I pulled the gcc-g95 and g95 projects gcc-g95 starts with this.. revision 1.1.1.1 date: 2003/01/06 21:04:20; author: paul_brook; state: Exp; lines: +0 -0 Initial import of all files into CVS. - So I think I still have the same question - Import from where? (I think I'm looking for the exact point which it was forked from sourceforge g95 repo)
Re: Original commit history for gfortran
On 06/18/2011 01:02 PM, "C. Bergström" wrote: On 06/18/11 05:16 PM, Toon Moene wrote: On 06/18/2011 12:12 PM, Toon Moene wrote: On 06/18/2011 05:05 AM, Christopher Bergström wrote: Hi We're in the process of considering contributing to gfortran for a special project, but when we started to vet the codebase we hit a bump in lack of commit history. Additional information is here: http://sourceforge.net/projects/gcc-g95 The above gives you the history after the split from the g95 project: http://sourceforge.net/projects/g95 in January 2003. The original commit by Paul Brook of the gcc-g95 repository contents to the GCC repository is here: http://gcc.gnu.org/ml/gcc-cvs/2003-07/msg01087.html Ok I pulled the gcc-g95 and g95 projects gcc-g95 starts with this.. revision 1.1.1.1 date: 2003/01/06 21:04:20; author: paul_brook; state: Exp; lines: +0 -0 Initial import of all files into CVS. - So I think I still have the same question - Import from where? (I think I'm looking for the exact point which it was forked from sourceforge g95 repo) Well, that question is very simple to answer - the point at which Andy Vaught turned the g95 repository in a read-only one for the other contributors. That was the whole reason to create a new repository. -- Toon Moene - e-mail: t...@moene.org - phone: +31 346 214290 Saturnushof 14, 3738 XG Maartensdijk, The Netherlands At home: http://moene.org/~toon/; weather: http://moene.org/~hirlam/ Progress of GNU Fortran: http://gcc.gnu.org/wiki/GFortran#news
Re: Original commit history for gfortran
2011/6/18 Toon Moene : > On 06/18/2011 01:02 PM, "C. Bergström" wrote: > >> On 06/18/11 05:16 PM, Toon Moene wrote: > >>> On 06/18/2011 12:12 PM, Toon Moene wrote: > >>> On 06/18/2011 05:05 AM, Christopher Bergström wrote: > Hi > > We're in the process of considering contributing to gfortran for a > special project, but when we started to vet the codebase we hit a bump > in lack of commit history. Additional information is here: http://sourceforge.net/projects/gcc-g95 The above gives you the history after the split from the g95 project: http://sourceforge.net/projects/g95 in January 2003. >>> The original commit by Paul Brook of the gcc-g95 repository contents >>> to the GCC repository is here: >>> >>> http://gcc.gnu.org/ml/gcc-cvs/2003-07/msg01087.html >>> >> Ok I pulled the gcc-g95 and g95 projects >> >> gcc-g95 starts with this.. >> >> revision 1.1.1.1 >> date: 2003/01/06 21:04:20; author: paul_brook; state: Exp; lines: +0 -0 >> Initial import of all files into CVS. >> - >> >> So I think I still have the same question - Import from where? (I think >> I'm looking for the exact point which it was forked from sourceforge g95 >> repo) > > Well, that question is very simple to answer - the point at which Andy > Vaught turned the g95 repository in a read-only one for the other > contributors. > > That was the whole reason to create a new repository. *cough* I feel the pain in that reply and it's not what I'm trying to stir up here. I'll see if we can directly match something against the 2 trees. (Actually some of the point to all my questions are trying to make amends to some of this past history)
Re: How effect the OpenSource EKOPath the GCC ?
2011/6/18 theUser BL : > > Hi! > > Currently I have nothing about it found in the mailinglist. So I try to ask > it: How effect the OpenSource EKOPath the GCC ? > > Have a look at the latest press news of PathScale: > http://www.pathscale.com/taxonomy/term/27 > > Have additional a look at this articls of phronix: > http://www.phoronix.com/scan.php?page=article&item=pathscale_ekopath4_open&num=1 > http://www.phoronix.com/scan.php?page=news_item&px=OTU2OA > > > The important things: > EKOPath will be step by step opend. First published opend code you can find at > https://github.com/path64 > > At > http://www.pathscale.com/ekopath-compiler-suite > you can download nightly builds. > To compile a helloworld.c program type > $ ~/ekopath-4.0.10/bin/pathcc helloworld.c -o helloworld > > You can use different benchmarks. For me it seems, that with EKOPath compiled > programs are a lot faster then with GCC. > > The only disadvantage is. that currently EKOPath exists only for 64bit > systems with x86_64 CPUs. > > But how will the OpenSource EKOPath effect GCC? > Can GCC make use of some EKOPath code? (EKOPath itself uses parts of GCC) > Or can GCC learn from it by studying it? > > Greatings > theuserbl > > > Hi theuserbl Chris is your boss? I know that EKOPath is much more better than open64, but open64 can compile nothing but spec2000. So, you want something about Effect, still? GCC is the real compiler that can work! GCC can bootstrap, can compile Linux Kernel, can compile GNU LibC, can compile almost everything. Will open64? Does open64 really can be used? or just a tony with a really suck codes? GCC is the only choice in the real world except LLVM, but never open64. --Liu
Re: Original commit history for gfortran
On 06/18/11 05:16 PM, Toon Moene wrote: On 06/18/2011 12:12 PM, Toon Moene wrote: On 06/18/2011 05:05 AM, Christopher Bergström wrote: Hi We're in the process of considering contributing to gfortran for a special project, but when we started to vet the codebase we hit a bump in lack of commit history. Additional information is here: http://sourceforge.net/projects/gcc-g95 The above gives you the history after the split from the g95 project: http://sourceforge.net/projects/g95 in January 2003. The original commit by Paul Brook of the gcc-g95 repository contents to the GCC repository is here: http://gcc.gnu.org/ml/gcc-cvs/2003-07/msg01087.html So I converted the cvs repo to git so I could actually dig and compare a little better.. Here's an example of what we're trying to understand This file wasn't in g95, but then magically appears in Paul's initial commit. gcc/f95/arith.h # Unless I've messed up somewhere along my path.. # Was this file in gcc the whole time and just an external dep? Here's the header at the time of commit --- /* Compiler arithmetic header. Copyright (C) 2000, 2001. 2002 Free Software Foundation, Inc. Contributed by Steven Bosscher This file is part of GNU G95. Seems like it's been around for a while, but certainly wasn't part of the sf.net g95 I cloned... If anyone spots any errors in this conversion process # Yes I realize beyond the scope of this ML git-cvsimport -v -a -i -k -d:pserver:anonym...@gcc-g95.cvs.sourceforge.net:/cvsroot/gcc-g95 -C gcc-g95-cvs.git gcc-g95 git-cvsimport -v -a -i -k -d:pserver:anonym...@g95.cvs.sourceforge.net:/cvsroot/g95 -C g95-cvs.git g95 I'll maybe have a more complete set of questions if everyone doesn't mind.. Thanks ./C
slides "GCC plugins & MELT extensions"
Hello All,
For those interested, I uploaded this morning 130 slides on a tutorial
I gave last week on GCC plugins & MELT extensions at Archi11 summer school.
(slides are CC by SA)
See http://gcc-melt.org/ where you can download the PDF file
Regards.
--
Basile STARYNKEVITCH http://starynkevitch.net/Basile/
email: basilestarynkevitchnet mobile: +33 6 8501 2359
8, rue de la Faiencerie, 92340 Bourg La Reine, France
*** opinions {are only mines, sont seulement les miennes} ***
Re: Original commit history for gfortran
On Sat, Jun 18, 2011 at 06:22:21PM +0700, Christopher Bergstr?m wrote: > 2011/6/18 Toon Moene : > > On 06/18/2011 01:02 PM, "C. Bergstr?m" wrote: > >> On 06/18/11 05:16 PM, Toon Moene wrote: > >>> On 06/18/2011 12:12 PM, Toon Moene wrote: > On 06/18/2011 05:05 AM, Christopher Bergstr?m wrote: > > > > We're in the process of considering contributing to gfortran for a > > special project, but when we started to vet the codebase we hit a bump > > in lack of commit history. > > Additional information is here: > > http://sourceforge.net/projects/gcc-g95 > > The above gives you the history after the split from the g95 project: > > http://sourceforge.net/projects/g95 > > in January 2003. > > >>> The original commit by Paul Brook of the gcc-g95 repository contents > >>> to the GCC repository is here: > >>> > >>> http://gcc.gnu.org/ml/gcc-cvs/2003-07/msg01087.html > >>> > >> Ok I pulled the gcc-g95 and g95 projects > >> > >> gcc-g95 starts with this.. > >> > >> revision 1.1.1.1 > >> date: 2003/01/06 21:04:20; author: paul_brook; state: Exp; lines: +0 -0 > >> Initial import of all files into CVS. > >> - > >> > >> So I think I still have the same question - Import from where? (I think > >> I'm looking for the exact point which it was forked from sourceforge g95 > >> repo) > > > > Well, that question is very simple to answer - the point at which Andy > > Vaught turned the g95 repository in a read-only one for the other > > contributors. > > > > That was the whole reason to create a new repository. > > *cough* > > I feel the pain in that reply and it's not what I'm trying to stir up > here. I'll see if we can directly match something against the 2 > trees. Good luck with that endeavor. After the gfortran fork, a certain individual would routinely obfusicate the code in one of the repositories via gratuitious code motion, varaible renaming, and whitespace munging. This was an attempt to neutered diff. This same individual did not include a ChangeLog along with the code, which again made it difficult to understand what and why code was changed. Finally, note I use the word repository here rather loosely because one repository became hidden and only snapshot tarballs were released. -- steve
RE: How effect the OpenSource EKOPath the GCC ?
Hi Liu > Chris is your boss? No. Who is Chris? > I know that EKOPath is much more better than open64, And could code of it useful for GCC or not? > but open64 can compile nothing but spec2000. Open64? I have googled at it. Do you mean that at http://www.open64.net/ I talked about Path64 https://github.com/path64 But you are right, at http://www.open64.net/about-open64.html there stand, that PathScale using it for EKOPath. And interestingly Open64 is a port of the old SGI-compiler. So Irix was written with MIPSPro and Open64 is a derivated work of it? Thats nice. > So, you want something about Effect, still? Yes. And I am looking, if there existing other OpenSource Compiler then GCC, which could be better in some areas. > GCC is the real compiler that can work! GCC can bootstrap, can compile > Linux Kernel, can compile GNU LibC, can compile almost everything. > Will open64? Does open64 really can be used? or just a tony with a > really suck codes? Why do you talk everytime about open64, when I talk about Path64? Open64 is the old one. I think if there existing nice code in it, GCC already have using it. Path64 is the new one, which opens. > GCC is the only choice in the real world except LLVM, but never open64. Can you descripe it a little bit more clear? What are the disadvantages of Open64 and what are the disadvantages of EKOPath/Path64? And if it have only disadvantages over GCC, why are people still working on Open64 and EKOPath/Path64? As I see, the newest version of Open64 was released April 13th, 2011. And EKOPath will now the times opened and becomes lot of updates. Open64 is under the GPL (similar to the GCC). But if GCC is the better one, why using PathScale for EKOPath and ENZO Open64 as base? But as I said before, it looks that PathScale using additional lot of parts of GCC: https://github.com/path64/compiler/tree/master/GCC https://github.com/path64/compiler/tree/master/gcc_incl Greatings theuserbl
gengtype & C++ [Gcc gathering london, session Victoria room]
Hello All
(I tried to send a PDF attached to a mail here but it don't work)
For information, I attached under http://gcc.gnu.org/wiki/Memory_management
the slides I showed today in Victoria room about gengtype & C++
Regards.
--
Basile STARYNKEVITCH http://starynkevitch.net/Basile/
email: basilestarynkevitchnet mobile: +33 6 8501 2359
8, rue de la Faiencerie, 92340 Bourg La Reine, France
*** opinions {are only mines, sont seulement les miennes} ***
Re: Original commit history for gfortran
On 06/18/11 11:41 PM, Steve Kargl wrote: Good luck with that endeavor. After the gfortran fork, a certain individual would routinely obfusicate the code in one of the repositories via gratuitious code motion, varaible renaming, and whitespace munging. This was an attempt to neutered diff. This same individual did not include a ChangeLog along with the code, which again made it difficult to understand what and why code was changed. Finally, note I use the word repository here rather loosely because one repository became hidden and only snapshot tarballs were released. Lets not beat up Andy on this at all, ok? I want to put all this shit behind us and possibly work together. - So what I'm looking at and comparing has nothing to with him at all in fact. It's the last commit from Paul in the g95 tree to the 1st commit from Paul in the gcc-g95 tree. Those should imho be equivalent, but they aren't. If the additional files were from Paul it wouldn't be interesting at all, but they aren't and I want to know who/where/what on them. "We" (PathScale) have to be *very* careful with the code we're working with. If we can't trace the history back on a file it may mean we need to write it from scratch unfortunately. (This is worst case of course) I'm not trying to be a troll here for the record. I'm honestly (against my will) trying to vet every line and where it came from. After this I'd like to open a discussion about how to work together and see if we can build a really strong Fortran community. (By working together I don't mean just PathScale, but a broader audience of possible contributors/testers/etc)
Re: How effect the OpenSource EKOPath the GCC ?
2011/6/19 theUser BL : > > Hi Liu > > >> Chris is your boss? > > > > No. Who is Chris? > > > > >> I know that EKOPath is much more better than open64, > > > > And could code of it useful for GCC or not? > > > > > >> but open64 can compile nothing but spec2000. > > > > Open64? I have googled at it. Do you mean that at > > http://www.open64.net/ > > > > I talked about Path64 > > https://github.com/path64 > > > > But you are right, at > > http://www.open64.net/about-open64.html > > there stand, that PathScale using it for EKOPath. > > > > And interestingly Open64 is a port of the old SGI-compiler. > > So Irix was written with MIPSPro and Open64 is a derivated work of it? Thats > nice. > > > > >> So, you want something about Effect, still? > > > > Yes. And I am looking, if there existing other OpenSource Compiler then GCC, > which could be better in some areas. > > > > >> GCC is the real compiler that can work! GCC can bootstrap, can compile >> Linux Kernel, can compile GNU LibC, can compile almost everything. > >> Will open64? Does open64 really can be used? or just a tony with a >> really suck codes? > > > > Why do you talk everytime about open64, when I talk about Path64? > > Open64 is the old one. I think if there existing nice code in it, GCC already > have using it. > > Path64 is the new one, which opens. > > > > >> GCC is the only choice in the real world except LLVM, but never open64. > > > Can you descripe it a little bit more clear? > > What are the disadvantages of Open64 and what are the disadvantages of > EKOPath/Path64? > > > > And if it have only disadvantages over GCC, why are people still working on > Open64 and EKOPath/Path64? > > As I see, the newest version of Open64 was released April 13th, 2011. > > And EKOPath will now the times opened and becomes lot of updates. > > > > Open64 is under the GPL (similar to the GCC). > > But if GCC is the better one, why using PathScale for EKOPath and ENZO Open64 > as base? > > > > But as I said before, it looks that PathScale using additional lot of parts > of GCC: > > https://github.com/path64/compiler/tree/master/GCC > > https://github.com/path64/compiler/tree/master/gcc_incl > > > > Greatings > > theuserbl > > > > Hi theuserbl What I mean is pathcc is much more better than open64, but not enough to gcc. For pathcc and open64 have the same code-base! Why Pathscale chose it? Fred Chow! Open64 always say "we will replace gcc" for almost 20 years... it still play with spec2000, only! --Liu
Re: Original commit history for gfortran
On Sun, Jun 19, 2011 at 12:13:25AM +0700, "C. Bergstr?m" wrote: > On 06/18/11 11:41 PM, Steve Kargl wrote: > >Good luck with that endeavor. After the gfortran fork, > >a certain individual would routinely obfusicate the code > >in one of the repositories via gratuitious code motion, > >varaible renaming, and whitespace munging. This was an > >attempt to neutered diff. This same individual did not > >include a ChangeLog along with the code, which again made > >it difficult to understand what and why code was changed. > >Finally, note I use the word repository here rather loosely > >because one repository became hidden and only snapshot > >tarballs were released. > Lets not beat up Andy on this at all, ok? I want to put all this shit > behind us and possibly work together. I'm not beating up on Andy. I'm just reporting history and why it is nearly impossible to compare g95 code to gfortran code. > So what I'm looking at and comparing has nothing to with him at all in > fact. It's the last commit from Paul in the g95 tree to the 1st commit > from Paul in the gcc-g95 tree. Those should imho be equivalent, but > they aren't. If the additional files were from Paul it wouldn't be > interesting at all, but they aren't and I want to know who/where/what on > them. You'll probably need to talk with Paul Brook. I suspect the timeline goes something like Andy closes g95 repository. pbrook and stevenb fork g95 to gfortran, and asks the FSF/GCC steering committee for permission to import code into GCC. 3 months later FSF/GCC steering committee gives the OK. Meanwhile, pbrook and stevenb continued to work on code in their private trees. -- Steve
gcc-4.7-20110618 is now available
Snapshot gcc-4.7-20110618 is now available on ftp://gcc.gnu.org/pub/gcc/snapshots/4.7-20110618/ and on various mirrors, see http://gcc.gnu.org/mirrors.html for details. This snapshot has been generated from the GCC 4.7 SVN branch with the following options: svn://gcc.gnu.org/svn/gcc/trunk revision 175183 You'll find: gcc-4.7-20110618.tar.bz2 Complete GCC MD5=704f377a7dc2f93d180116fc0635a354 SHA1=a2cf3fb72f91ffd730d7631c9982328e7441a802 Diffs from 4.7-20110611 are available in the diffs/ subdirectory. When a particular snapshot is ready for public consumption the LATEST-4.7 link is updated and a message is sent to the gcc list. Please do not use a snapshot before it has been announced that way.
