The collected works of diegel - Page 5

TeamBlackFox wrote:

diegel wrote: What's wrong with neko_gcc47? It works without any problem here.

For some reason it doesn't want to generate binaries no matter how may times I install it from the tardist. Regardless,, I can build it as there's a wiki guide.

There must be something wrong with your installation. Why should gcc work everywhere but not an your systems? The problem can't be the gcc package itself and I am sure a rebuild will not solve your problem.

Before you start to build nekoware packages it is a good idea to rebuild existing ones. And please read nekoware release notes, there are lots of tips and workarounds in it and of course sample build environments.

Dillo 3.0.5 is available in beta now:

The firefox rebuild and the rebuild dependencies are in beta now. I only uploaded the gcc build glib since the mipspro based version causes a lot of crashes. Here is a list of packages:

If you want to see this packages in current please vote for it.

hamei wrote: btw, herr diegel - how come we are un-versioning the libraries ? What effect does that have ?

This was not my idea, it had been done in the current libpng package also. If this cause any problem, we have the problem already. I don't change much at the dependencies, I just rebuild them with better optimization. If I can't build a package with maximum optimization with MipsPro I used to gcc.

Axatax_ wrote: The zlib is not versioned properly.

Thanks for your feedback. Would you give some more details about this versioning issue.

foetz wrote: ksh is not such a good choice for interactive shells. tcsh which comes with 6.5.x as well usually is much nicer

I am using ksh in vi mode for decades now and for me it is the best choice.

dukzcry wrote: [ And dillo version from -beta seems to have broken fonts rendering somewhere in the path.

If this is a real problem, I can build dillo without xft.

hamei wrote:

diegel wrote: If this is a real problem, I can build dillo without xft.

Can you ? Did you know that, at least on Irix, X will serve up Type1 fonts ? And we have those ? And they are better than Truetype ?

You can build fltk without xft support. In this case there is no support for any antialiased fonts.

As I mentioned in a private message before, I am operating an Internet provider with data center in Germany for 20 years now and since this is a very good business I plan to do this some more years. I would be happy to provide a free virtual or a real hardware server for nekochan. I do this for an OS/2 related site http://www.netlabs.org for many years.

nekonoko wrote: Yep I know. From my perspective sticking with what I have is the path of least resistance.

I am afraid there is also a path of no resistance: shutting down the site. This happened to too many sites in the past. And we all don't want this happen.

If I have to pack up the site and move somewhere else and deal with a completely new infrastructure/support regimen, it makes a good stepping off point for me. Let someone else have some fun.

I'm not threatening anything - I really wouldn't mind that outcome after 14 years of keeping the site alive. Maybe it's just time to pass the torch.

Thank you for providing this site for such a long time. I am very happy if you do so in the future, but if you feel better if someone else is doing this foetz and I are ready to take the torch.

SAQ wrote: Don't have a SGI with me right now, but I'm pretty sure that in the "new new" System Manager with the "SGI font" (6.5.14+) the y in System has the tail pointing to the right.

I never noticed, but it looks like this:

Nekoware is still the most used freeware distribution for Irix 6.5, but there are very few users left. Currently we have two main issues: get firefox 3.5 or newer running, get a newer gtk2 than 2.12 working. Both packages compile without major problems, but are not working. Firefox 3.5 stops with the same problem you have seen many years ago and gtk2 stops working if you open the file chooser.

Since we are so less active users, the beta testing process isn't working anymore. We have a very good gcc 4.7 running, but many nekoware users still don't like to use gcc compiled mips code. You still can't mix gcc build c++ objects with mips pro c++ objects, but otherwise there is no problem with gcc code from my perspective.

I still have no idea what you want with an updated MipsPro. There is no such a thing like a miracle compiler that makes run current open source on an O2 as fast as on any up to date hardware. MipsPro builds in many situations faster code than gcc, but we are talking about 10% or maybe 20%. We have to face the reality: the current iphone is faster than any Tezro and there will never be any compiler that will change this.

If we want to port current open source to Irix we need a lot of programming skills, because this code is build for Linux and possibly BSD. The folks that wrote this code has no idea that Irix ever exists. And there will never be a miracle compiler, that will port this code for us.

That's great news. I am currently at work and don't have access to my build system. I will post my current ff3.5 build config later this day. I will also have a look at dbus. By the way, it is quite common that nekoware packages do not pass their test suite.

dexter1 wrote: My attempts with Firefox 3 series was with version 3.0.15.

You have been very close to a working ff3. The only change I had to do is to add the following line to .mozconfig to avoid the bus error:



If you want to see you ff3.0.15 running you can try this.

I am working with ff3.5.19. I build it with dbus disabled. Attached you find my current patch. This is my build environment:



Here comes the .mozconfig:



The ff3.0.19 build instructions are still working:

Build instructions:

Get the source code from:

ftp://ftp.mozilla.org/pub/firefox/relea ... ce.tar.bz2

extract it and apply the attached patch.

Get a current version of sqlite3:

http://www.sqlite.org/download.html

Copy sqlite3.c and sqlite3.h from current sqlite3 to mozilla/db/sqlite3/src

There may be a problem with signlib during build. Make sure there
are no shared libs in /usr/nekoware/lib/firefox-3.0.19 during build instead
you must have a firefox2 or seamonkey1 lib directory in your LIB search path.
Otherwise signlib may coredump.

set the environment and start the build:

gmake -f client.mk build

There is one assembler file in libfreebl you have to compile manually with gcc3.

cd security/nss/lib/freebl
gcc -x assembler-with-cpp -o /work/mozilla-1.9.1/ff-deb-static/nss/freebl/IRIX_SINGLE_SHLIB/mpi_mips.o -Wp,-P -Wp,-traditional-cpp -c mpi/mpi_mips.s

I think the current problem is still in xpcom. This is possibly a start to find out more: https://bugzilla.mozilla.org/show_bug.cgi?id=482759

Indyboy wrote: Is it possible effectively to kill the machine with interchanging the parts?

Yes it is. I own a set of memory for O2 that kills every O2 mainboard for example. I also had a Cisco chassis that kills memory of a second route processor.

rosehillbob wrote: I found the missing file in dexter1's xptcall patch.

Sorry, this file was indeed missing in my patch but it's the same file I use for ff3 and also ff3.5. I am glad to read that you are able to build it now. If you are starting an optimized build it will take at least twice the time to build. There was an attempt to port ff10 some time ago, he stopped this project after he was unable to link a huge libmoz because of Irix ld limitations. If you are interested in this work, I can provide this for you. If I can support your work please let me know.

The reason for this advice is that the scsi of the mezzanine board occupies one dma channel. There is only one dma channel left for the gio32 boards. The graphics board don't need the gio dma, so you can use two gio32 boards with dma. There was also a gio extender that does the same without graphics board.

This is possibly helpful for you: I tried the ff3.1 alpha versions and they also have the same problem like 3.5. The main difference between 3.0 and 3.1 is the new javascript engine. So I guess the problem is related with this.

Thank you very much. I try to test it during the weekend.

rosehillbob wrote: Well to get DBUS to work I have to get a proper working GLIB which may solve some other problems.

From my experience you have to go back to glib-2.20.5 to find a glib that passes all tests. If you do so, you have to rebuild all glib dependencies.I did this already for debugging, but I never tried to use dbus.

Many bus errors are results of MipsPro aggressive optimization, so first try to build a version without optimization and look, if you still have the errors. If yes build a version with debug info (-g) and go into it with dbx to find the function where to error occurs. If you are not able to find it that way the error may possibly be in a shared library. In this case you can start to debug all dependent shared libraries in the same way. If you find the function you have to go into the code and try to understand what happens there.

I am still running Irix systems on the Internet. This are private projects, like the nekoware mirror and I had never problems with it. Our company used around the year 2000 a Challenge S as a secondary nameserver. This server was located at another Internet service provider (for free) and we simply forget this server. When this company moved the location some years ago, they asked us if we still using this server. So we got it back and examined it, it was running Irix 6.2 and never get hacked after 10 years running without any administration.

vishnu wrote:

diegel wrote: So wait, by decree of the IETF every domain name has to have at least two nameservers, with different IP addresses, serving the domain. So did you guys forget about that nameserver or just not worry about it since it was working? Cuz otherwise I don't see how you could actually literally forget about one of your nameservers, unless you had dozens, which admittedly the big providers do...

First of all we are using more than one secondary and we are spreading this over more than one network provider. And of course we first moved the service to more modern hardware before we forget this server.

This weekend I had some time to have a look on gtk2 and ff3.5. I also build a ff3.5 depending on tgcware instead of nekoware. It crashes in exactly the same way Dexter experienced many years ago. Since this problem occurs with gcc and mipspro and also with nekoware and tgcware, we can be quite sure this is not a problem of the compiler or any of the dependencies. The earliest ff3.1 beta releases also crashes in the same way,

edit: I removed my wrong gtk2 information. gtk2.18 is not running correctly with firefox.

FF3 bus errors are fixed, we have a running ff3 version now and some of these patches are part of the current version. ff3.5 and later have a different problems: it terminates during startup with an assignment error. I had a look at the patches of Zhang Le, who get ff3.5 running on a linux version for Loongson CPUs. He transformed the n32 stuff we already using for Irix to linux .

jan-jaap wrote: [quote="mopar5150" you cannot upgrade the single CPU version to dual or quad CPU .

This information is not correct. I own an upgraded single processor Tezro. But I never owned the single processor node board., if someone have one, I am interested.

Sorry for the fuzzy picture, my Tezro is looking like this: