Re: [9fans] software ports

Re: [9fans] software ports

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 689 bytes --]

Vita Nuova will be making available the portable version of the plan 9 compiler suites, and mk
and a few other supporting tools that compile native Inferno on all platforms,
for more general use, separately from Inferno.
also included is the set of libraries used to support a limited plan 9 environment
sufficient to compile the compilers, mk and the others.

producing binaries, even statically linked, for other operating systems is trickier than
it once was (in my experience), though not impossible.  i have enough trouble working out
what ELF is on each platform that uses it to generate bootable binaries.
there might have been good intentions, but it is not well done.


[-- Attachment #2: Type: message/rfc822, Size: 3870 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 229 bytes --]

Hi Russ

Was just looking at your home page @Harvard. I'm interested in
the 'cross compiling for other operating systems' for various
reasons.

Have you done any work on it so far/ is anyone else working on it ?
-
pip


[-- Attachment #2.1.2: Type: message/rfc822, Size: 1689 bytes --]

From: "Russ Cox" <rsc@plan9.bell-labs.com>
To: 9fans@cse.psu.edu
Subject: [9fans] software ports
Date: Sat, 15 Jul 2000 16:56:21 -0400
Message-ID: <200007152056.QAA04972@smtp1.fas.harvard.edu>

Plan 9 ports of Norman Ramsey's noweb,
CVS 1.10.8, and Moscow ML 2.0 are all
at http://www.eecs.harvard.edu/~rsc/plan9.html.

(Use wrap/inst to install them.)

Russ

[9fans] fs patch for K6

[arisawa]

Hello,

This is a private mail.

You said about fs patch for K6.
I have waited that to be officially released but I cannot wait no  
longer. Please send me your patch.

By the way, I will continue to use your ykhhtpd.
I added some changes to be suite for new version.
That was: getmfields and varargs.
Are there anything else?

Kenji Arisawa
E-mail: arisawa@aichi-u.ac.jp

Re: [9fans] software ports

[Michael Joosten]

Mike Haertel wrote:
> 

> there.  Anyway the stabs format has probably evolved somewhat since
> the 4.3bsd man page was written, moreover I think gdb has its own
> slightly idiosyncratic variation that fixes some problems with the
> original dbx format.  So if you want a well-documented format you're
> probably better off with DWARF.

There was a stabs.texinfo either in GDB or in GCC some time...
And, SunSoft has also a separate document as part of their SunWork
development suite (compilers, dbx, dbxtool ,etc) documenting 'their' 
extensions to plain BSD stabs.

-- 
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| Michael Joosten   |       Tel.  : (+49) (+) 5251-60-6127     |
| C-LAB             |       Fax   : (+49) (+) 5251-60-6065     |
| Fuerstenalle 11   |       E-Mail: joost@c-lab.de             |
| 33094 Paderborn   |      C-LAB is a cooperation between      |
| Germany           |    University Paderborn & SIEMENS AG     |
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: [9fans] software ports

[pip]

[-- Attachment #1: Type: text/plain, Size: 229 bytes --]

Hi Russ

Was just looking at your home page @Harvard. I'm interested in
the 'cross compiling for other operating systems' for various
reasons.

Have you done any work on it so far/ is anyone else working on it ?
-
pip


[-- Attachment #2: Type: message/rfc822, Size: 1689 bytes --]

From: "Russ Cox" <rsc@plan9.bell-labs.com>
To: 9fans@cse.psu.edu
Subject: [9fans] software ports
Date: Sat, 15 Jul 2000 16:56:21 -0400
Message-ID: <200007152056.QAA04972@smtp1.fas.harvard.edu>

Plan 9 ports of Norman Ramsey's noweb,
CVS 1.10.8, and Moscow ML 2.0 are all
at http://www.eecs.harvard.edu/~rsc/plan9.html.

(Use wrap/inst to install them.)

Russ

Re: [9fans] software ports

[Mike Haertel]

In comp.os.plan9, you wrote:
>Without that, I'm sure we'll just hit a debugging wall.
>Anyone know what gdb expects?

With Elf binaries, gdb can work with two different kinds of debugging
symbols: Berkeley dbx-style "stabs" (which gdb expects to find in
Elf sections named ".stab" and ".stabstr"), or AT&T-style DWARF
debugging symbols, which have a different set of ELF section names.

ftp://tsx-11.mit.edu/pub/linux/packages/GCC/ELF.doc.tar.gz
(describes basic Elf file format, but not debugging symbols)

There is some documentation for the DWARF2 debugging format
in ftp://sgigate.sgi.com/pub/dwarf.

The "stabs" debugging format is described in the dbx(5) man page
under 4.3BSD.  Unfortunately, this documentation does not seem to
exist in any of the recent free *BSD's; this is a valuable man page
and I don't know what became of it.  If you have a printed copy of
the old Usenix 4.3BSD manuals you can find it there (that's what I
have), or if you have a 4.3bsd tape sitting around you'll find it
there.  Anyway the stabs format has probably evolved somewhat since
the 4.3bsd man page was written, moreover I think gdb has its own
slightly idiosyncratic variation that fixes some problems with the
original dbx format.  So if you want a well-documented format you're
probably better off with DWARF.

Re: [9fans] software ports

[Russ Cox]

William Josephson and I have been discussing it, but
haven't done anything.  Generating the raw ELF binaries
should be easy, but I haven't found any documents
that say what the debugging info -- the symbol/pc table
and the pc/source line table, not to mention the actual
variables -- looks like.  

Without that, I'm sure we'll just hit a debugging wall.
Anyone know what gdb expects?

Russ

[9fans] software ports

[Russ Cox]

Plan 9 ports of Norman Ramsey's noweb,
CVS 1.10.8, and Moscow ML 2.0 are all
at http://www.eecs.harvard.edu/~rsc/plan9.html.

(Use wrap/inst to install them.)

Russ