[Linux-ia64] ski and gcc-3.x

[Linux-ia64] ski and gcc-3.x

[Randolph Chung]

hi,

two questions really....

1. is linux-2.5.x on ski supposed to work?
2. is there a version of ski that works with gcc-3.2?

the latest ski i can find (0.94311) doesn't work with gcc-3.2 built
kernels. On startup it gives:
BFD: Dwarf Error: Invalid or unhandled FORM value: 14.

iirc this is a DWARF debug directive that was introduced in gcc
around gcc-3.1 or so....

randolph
-- 
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/

Re: [Linux-ia64] ski and gcc-3.x

[David Mosberger]

>>>>> On Thu, 20 Feb 2003 22:53:34 -0800, Randolph Chung <randolph@tausq.org> said:

  Randolph> hi, two questions really....

  Randolph> 1. is linux-2.5.x on ski supposed to work?

Yes, I use it all the time.  2.5.60 certainly works.

  Randolph> 2. is there a version of ski that works with gcc-3.2?

Yes.

  Randolph> the latest ski i can find (0.94311) doesn't work with
  Randolph> gcc-3.2 built kernels. On startup it gives: BFD: Dwarf
  Randolph> Error: Invalid or unhandled FORM value: 14.

That's only a warning though, right?  It should be harmless.

In any case: I currently have v0.9.81l1 of Ski in alpha testing and
beta testing should start next week.  If you can wait till then, that
would be easiest.  Otherwise, let me know and I'll see about sending
you the current test-binary.

	--david

Re: [Linux-ia64] ski and gcc-3.x

[Randolph Chung]

> That's only a warning though, right?  It should be harmless.

no, ski segfaults right after that.

I recompiled 2.5.60 without -g and it loads, but when i run it, it says:
* run
Unable to create console window
21 insts (0 faults), 0.01 sec, 4086 i/s

(i'm running this remotely, so no X ...)

> In any case: I currently have v0.9.81l1 of Ski in alpha testing and
> beta testing should start next week.  If you can wait till then, that
> would be easiest.  Otherwise, let me know and I'll see about sending
> you the current test-binary.

ok, this can wait :)

thanks,
randolph
-- 
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/

Re: [Linux-ia64] ski and gcc-3.x

[David Mosberger]

>>>>> On Thu, 20 Feb 2003 23:13:35 -0800, Randolph Chung <randolph@tausq.org> said:

  >> That's only a warning though, right?  It should be harmless.

  Randolph> no, ski segfaults right after that.

Hmmh, that's odd.  In any case, the new version should fix that.

  Randolph> * run Unable to create console window

  Randolph> (i'm running this remotely, so no X ...)

I'm afraid you'll be out of luck without X: you _can_ run in
batch-mode without X, but then you don't get the debugger.  If you
want the debugger, there must be a way to create a simulated console
and that's done by starting a slave xterm.  Without X, that's hard to
do.  If I work remotely, I use ssh X forwarding to prevent this
problem.

	--david

Re: [Linux-ia64] ski and gcc-3.x

[Randolph Chung]

> I'm afraid you'll be out of luck without X: you _can_ run in
> batch-mode without X, but then you don't get the debugger.  If you
> want the debugger, there must be a way to create a simulated console
> and that's done by starting a slave xterm.  Without X, that's hard to
> do.  If I work remotely, I use ssh X forwarding to prevent this
> problem.

ah, ok... Matt's fake-xterm program was very useful too...

now i just have to figure out what i'm doing wrong and why it doesn't
see the simulated scsi disk... :(

I'm running ski with:
ski linux-2.5.60/bootloader linux-2.5.60/vmlinux simscsi=/var/ski-disks/sd

and all i see is:

scsi0 : simulated SCSI host adapter
NET4: Linux TCP/IP 1.0 for NET4.0
IP: routing cache hash table of 1024 buckets, 8Kbytes
TCP: Hash tables configured (established 16384 bind 16384)
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
VFS: Cannot open root device "" or 08:01
Please append a correct "root=" boot option
Kernel panic: VFS: Unable to mount root fs on 08:01

confused,
randolph
-- 
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/

Re: [Linux-ia64] ski and gcc-3.x

[David Mosberger]

>>>>> On Fri, 21 Feb 2003 00:14:22 -0800, Randolph Chung <randolph@tausq.org> said:

  Randolph> now i just have to figure out what i'm doing wrong and why
  Randolph> it doesn't see the simulated scsi disk... :(

  Randolph> I'm running ski with: ski linux-2.5.60/bootloader
  Randolph> linux-2.5.60/vmlinux simscsi=/var/ski-disks/sd

  Randolph> and all i see is:

  Randolph> VFS: Cannot open root device "" or 08:01

Did you verify that you have write access on /var/ski-disks/sda?

	--david

Re: [Linux-ia64] ski and gcc-3.x

[Randolph Chung]

> Did you verify that you have write access on /var/ski-disks/sda?

yes, but i didn't read the instructions carefully enough so i thought if
i had simscsi=/var/ski-disks/sd my disk file is also called
/var/ski-disks/sd (instead of .../sda)

it's booting now with 1.30 bogomips :-)

thanks,
randolph
-- 
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/

Re: [Linux-ia64] ski and gcc-3.x

[David Mosberger]

>>>>> On Fri, 21 Feb 2003 08:10:25 -0800, Randolph Chung <randolph@tausq.org> said:

  >> Did you verify that you have write access on /var/ski-disks/sda?

  Randolph> yes, but i didn't read the instructions carefully enough
  Randolph> so i thought if i had simscsi=/var/ski-disks/sd my disk
  Randolph> file is also called /var/ski-disks/sd (instead of .../sda)

  Randolph> it's booting now with 1.30 bogomips :-)

Great!  1.30 bogomips is the right number (it's deterministic and not
dependent on the simulation speed).

	--david