qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Bochnig, Martin" <mb1x@gmx.com>
To: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] sparc?
Date: Sun, 29 Aug 2004 16:26:11 +0200	[thread overview]
Message-ID: <4131E783.5000307@gmx.com> (raw)
In-Reply-To: <4130B905.5090706@gmx.com>

Bochnig, Martin wrote:

> But I'm a *newbie* in that field.
> 
> 
> Under Solaris9 32bit mode:
> I looks pretty promising: The build process nearly completes now!!!
> Just 3 "Sybol not found" errors when linking 'qemu'.
> 

Hi,

QEMU 0.5.5 does completely build on my SPARCengine Ultra AXi running 
Solaris 10 Beta build 60 (64 bit kernel only).

Current situation:
Good news first: The recent linking errors (unreferenced symbols lrint, 
llrint and fesetround not found) were caused by the lack of these 
symbols in Solaris9's libc.
Fortunately Sun managed to integrate these GLIBC functions into Sun's 
Solaris10 Beta libc ("via Solaris Express" freely downloadable by anyone 
at sun.com ).
GLIBC is *not* available on any Solaris. Nobody did the port.

The QEMU user interface is fully functional (QEMU monitor; QEMU SDL gui).
'qemu' w/o options gives usage information - as well as '?' in QEMU Monitor.

Now the bad rest - unfortunately the main thing, the cpu, does *not* work:
When the build completes I get the following warning:
gmake[1]: warning:  Clock skew detected.  Your build may be incomplete.
And indeed - while the SDL QEMU window pops up, the screen remains 
black. Nothing happens. Not even the VGA Bios screen is displayed.
Eventually the whole qemu process either segfaults or hangs.
'-nographic' does the same.

I need help.
*Please* HELP!

Please have a look at the following compiling output:
gmake[1]: Entering directory `/tmp_qemu0.5.5_COPLETE_BUILD/i386-softmmu'
gmake[1]: Warning: File `../config-host.mak' has modification time 33 s 
in the future
gcc -Wall -O2 -g -fno-strict-aliasing -m32 -ffixed-g1 -ffixed-g2 
-ffixed-g3 -ffixed-g6 -I. -I/tmp_qemu0.5.5_COPLETE_BUILD/target-i386 
-I/tmp_qemu0.5.5_COPLETE_BUILD -D_GNU_SOURCE -c -o vl.o 
/tmp_qemu0.5.5_COPLETE_BUILD/vl.c
/tmp_qemu0.5.5_COPLETE_BUILD/vl.c: In function `serial_open_device':
/tmp_qemu0.5.5_COPLETE_BUILD/vl.c:953: warning: unused variable `slave_name'
/tmp_qemu0.5.5_COPLETE_BUILD/vl.c:954: warning: unused variable `master_fd'
/tmp_qemu0.5.5_COPLETE_BUILD/vl.c:954: warning: unused variable `slave_fd'
/tmp_qemu0.5.5_COPLETE_BUILD/vl.c: At top level:
/tmp_qemu0.5.5_COPLETE_BUILD/vl.c:846: warning: `start_rtc_timer' 
defined but not used
gcc -Wall -O2 -g -fno-strict-aliasing -m32 -ffixed-g1 -ffixed-g2 
-ffixed-g3 -ffixed-g6 -I. -I/tmp_qemu0.5.5_COPLETE_BUILD/target-i386 
-I/tmp_qemu0.5.5_COPLETE_BUILD -D_GNU_SOURCE -c -o osdep.o 
/tmp_qemu0.5.5_COPLETE_BUILD/osdep.c
gcc -Wall -O2 -g -fno-strict-aliasing -m32 -ffixed-g1 -ffixed-g2 
-ffixed-g3 -ffixed-g6 -I. -I/tmp_qemu0.5.5_COPLETE_BUILD/target-i386 
-I/tmp_qemu0.5.5_COPLETE_BUILD -D_GNU_SOURCE -c -o block.o 
/tmp_qemu0.5.5_COPLETE_BUILD/block.c

I think, especially the fact, that `start_rtc_timer' is not used is the 
(or at least one essential) root of all the problems.

Please help me in implementing the SPARC host code for 'vl.c'.
For every existing cpu, everything looks completely different and I 
cannot see how to reproduce parts of it for sparc.

All I was/am able to handle was just a copy&paste of Richard Zidlicky's 
patch at 
http://lists.gnu.org/archive/html/qemu-devel/2004-03/msg00078.html and 
changing its name to sparc.
I browsed some header files having 'time' in their name, but I cannot 
realize how to get that for a SPARC host working.

*Many* thanks in advance
Martin

  reply	other threads:[~2004-08-29 14:33 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-08-26 11:06 [Qemu-devel] sparc? Bochnig, Martin
2004-08-26 22:11 ` Bochnig, Martin
2004-08-27 23:29 ` Bochnig, Martin
2004-08-27 23:34   ` Lennert Buytenhek
2004-08-27 23:58     ` Bochnig, Martin
2004-08-28  0:05       ` Lennert Buytenhek
2004-08-28  0:14         ` Bochnig, Martin
     [not found]   ` <413083B1.2040102@bellard.org>
2004-08-28 16:55     ` Bochnig, Martin
2004-08-29 14:26       ` Bochnig, Martin [this message]
2004-08-29 14:37         ` Johannes Schindelin
2004-08-29 14:53           ` Bochnig, Martin
2004-08-29 15:12           ` Bochnig, Martin
2004-08-31 22:22         ` Richard Zidlicky
  -- strict thread matches above, loose matches on Subject: below --
2004-08-26 10:46 Bochnig, Martin
2004-01-02 16:57 Yelich, Scott D.
2004-01-02 18:16 ` Fabrice Bellard
2004-01-02 20:48   ` NunO fELICIO
2004-01-02 21:49   ` Raymond W. Lucke IV
2004-01-02 23:01     ` Satadru Pramanik

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4131E783.5000307@gmx.com \
    --to=mb1x@gmx.com \
    --cc=bochnig@pool.math.tu-berlin.de \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).