Re: [osol-discuss] osol-11 repair boot of new vmware guest install

[Aubrey Li]

On Tue, Feb 17, 2009 at 7:31 AM, Harry Putnam  wrote:
> Summary:
> I'm having trouble finding advice or a documentation of how to get
> inside and repair and install that cannot be booted.
>
> Details:
>
> This is a fairly new install of osol-11 from a recent iso image.
> Host is winXP using VMware (VirtualBox freezes the OS repeatedly)
>
> I'd been running the new install for a few days and attempted to
> install the vmware tools.  When I got to the part where you run the
> installation script I apparently either overwrote the config file for
> starting X or installed a new one with a different name
> `/etc/X11/xorg.conf'.  When that question came up it asked me if I
> wanted to install the conf file..  I said yes thinking I would be
> given a chance to say where.
>
> Instead the vmware tool  install routine apparently installed
> something and finished.. On reboot, after the grub prompt the gui
> desktop never comes up.
>
> Selecting the `text' boot option seemed to allow me to login at a
> console but then it goes on to starting the xserver with no action on
> my part... and that never completes.
>
> It appears I am effectively locked out.  I can edit the grub prompt
> but really could not find by googling, what (exactly) to do there.
>
> A screen shot of the grub offerings after typing `e' to edit,  here:
>
>  http://home.jtan.com/~reader/vu/disp.cgi
>
> I haven't tried them all but the ones I have tried led to a console
> prompt but then after logging in as my user... in a minute or so it
> goes on trying for the gui login I guess... and then all I see is
> black.  No further action seems possible
>
> I do have the iso but not really sure how to gain entry with it
> either.
>

It looks like xorg.conf messed your box up.
Now edit grub menu.lst to the following(remove the first 4 line and
remove console = graphics):

bootfs rpool/ROOT/opensolaris
kernel$ /platform/i86pc/kernel/$ISADIR/unix -B $ZFS-BOOTFS -s
module$ /platform/i86pc/$ISADIR/boot_archive

This(-s kernel option) will give a chance to enter the system and
remove xorg.conf or disable gdm(svcadm disable gdm).

Thanks,
-Aubrey
Intel OpenSolaris Team
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Rich Reynolds]

this IS what IPS has shown as the potential...  the implementation is as 
usual slow to be manifested, and from the outside we see little(as we 
should for a distro) of the release process..   that is the secret sauce 
that makes a disto work...


I wait with baited breath,

rich

Shawn Walker wrote:

Rich Reynolds wrote:

shawn -

while things have changed...   the fact that a IPS upgrade from b106 
to b107 involved 600+MB of download tells me that the model is still 
substantially a WOS. I can understand that in a substantial upgrade 
like 2008.05 to .11, but a basic two week development effort results 
in 600+ MB of "new" code???   so if you hadn't guessed im still not a 
huge fan of the IPS repository model... my base system is SXCE 101, 
with LU's back to 88, and I keep hoping I can migrate to IPS and BE, 
because it has the potential for substantially smaller bandwidth 
requirements, but...

its potentially better than SXCE, but nowhere near as atomic as BFU...


You're confusing shortcomings in the build and upgrade process with a 
perceived lack of change when it comes to download size.


Those download sizes will be significantly reduced in the future.

Cheers,


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Rich Reynolds]

sorry ---  i mis-typed that upgrade was from 105 to 106, I have been 
trying to justify a base system upgrade to IPS based 107, but still 
can't..

I just keep compiling my own
it's just like SunOS 3.5 days..
build Unix from Source!!!

rich

 Rich Reynolds wrote:

shawn -

while things have changed...   the fact that a IPS upgrade from b106 to 
b107 involved 600+MB of download tells me that the model is still 
substantially a WOS. I can understand that in a substantial upgrade like 
2008.05 to .11, but a basic two week development effort results in 600+ 
MB of "new" code???   so if you hadn't guessed im still not a huge fan 
of the IPS repository model... my base system is SXCE 101, with LU's 
back to 88, and I keep hoping I can migrate to IPS and BE, because it 
has the potential for substantially smaller bandwidth requirements, but...

its potentially better than SXCE, but nowhere near as atomic as BFU...

just a view from the dark,

rich

Shawn Walker wrote:

Rich Reynolds wrote:

Greg -

thanx for the clarification...   my intent was only to suggest that 
the old BWOS "train" model that Sun has used for MANY years many no 
longer hold them in good stead and that should a better distro model 
come about, which finally views a distro as nothing more than an 
integrated base set of code, a simple and granular repository, and an 
easy to use install/upgrade method, OpenSolaris.org based systems 
could truly compete for the desktop with the likes of a Ubuntu 
"Intrepid Ibex". Dont get me wrong, I am a dyed in the wool SunOS 
guy, from SunOS3.2 on, but the release engineering world and the 
nature of the customer base has changed and we as a community need to 
help all the opensolaris.org based distros compete more effectively 
or perish under our own weight...


I have to ask, haven't you noticed how significantly the model has 
already changed and is continuing to change with the OpenSolaris 200x 
releases?







___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Shawn Walker]

Rich Reynolds wrote:

shawn -

while things have changed...   the fact that a IPS upgrade from b106 to 
b107 involved 600+MB of download tells me that the model is still 
substantially a WOS. I can understand that in a substantial upgrade like 
2008.05 to .11, but a basic two week development effort results in 600+ 
MB of "new" code???   so if you hadn't guessed im still not a huge fan 
of the IPS repository model... my base system is SXCE 101, with LU's 
back to 88, and I keep hoping I can migrate to IPS and BE, because it 
has the potential for substantially smaller bandwidth requirements, but...

its potentially better than SXCE, but nowhere near as atomic as BFU...


You're confusing shortcomings in the build and upgrade process with a 
perceived lack of change when it comes to download size.


Those download sizes will be significantly reduced in the future.

Cheers,
--
Shawn Walker
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Rich Reynolds]

shawn -

while things have changed...   the fact that a IPS upgrade from b106 to 
b107 involved 600+MB of download tells me that the model is still 
substantially a WOS. I can understand that in a substantial upgrade like 
2008.05 to .11, but a basic two week development effort results in 600+ 
MB of "new" code???   so if you hadn't guessed im still not a huge fan 
of the IPS repository model... my base system is SXCE 101, with LU's 
back to 88, and I keep hoping I can migrate to IPS and BE, because it 
has the potential for substantially smaller bandwidth requirements, but...

its potentially better than SXCE, but nowhere near as atomic as BFU...

just a view from the dark,

rich

Shawn Walker wrote:

Rich Reynolds wrote:

Greg -

thanx for the clarification...   my intent was only to suggest that 
the old BWOS "train" model that Sun has used for MANY years many no 
longer hold them in good stead and that should a better distro model 
come about, which finally views a distro as nothing more than an 
integrated base set of code, a simple and granular repository, and an 
easy to use install/upgrade method, OpenSolaris.org based systems 
could truly compete for the desktop with the likes of a Ubuntu 
"Intrepid Ibex". Dont get me wrong, I am a dyed in the wool SunOS guy, 
from SunOS3.2 on, but the release engineering world and the nature of 
the customer base has changed and we as a community need to help all 
the opensolaris.org based distros compete more effectively or perish 
under our own weight...


I have to ask, haven't you noticed how significantly the model has 
already changed and is continuing to change with the OpenSolaris 200x 
releases?




___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Shawn Walker]

Rich Reynolds wrote:

Greg -

thanx for the clarification...   my intent was only to suggest that the 
old BWOS "train" model that Sun has used for MANY years many no longer 
hold them in good stead and that should a better distro model come 
about, which finally views a distro as nothing more than an integrated 
base set of code, a simple and granular repository, and an easy to use 
install/upgrade method, OpenSolaris.org based systems could truly 
compete for the desktop with the likes of a Ubuntu "Intrepid Ibex". Dont 
get me wrong, I am a dyed in the wool SunOS guy, from SunOS3.2 on, but 
the release engineering world and the nature of the customer base has 
changed and we as a community need to help all the opensolaris.org based 
distros compete more effectively or perish under our own weight...


I have to ask, haven't you noticed how significantly the model has 
already changed and is continuing to change with the OpenSolaris 200x 
releases?


--
Shawn Walker
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Rich Reynolds]

Greg -

thanx for the clarification...   my intent was only to suggest that the 
old BWOS "train" model that Sun has used for MANY years many no longer 
hold them in good stead and that should a better distro model come 
about, which finally views a distro as nothing more than an integrated 
base set of code, a simple and granular repository, and an easy to use 
install/upgrade method, OpenSolaris.org based systems could truly 
compete for the desktop with the likes of a Ubuntu "Intrepid Ibex". 
Dont get me wrong, I am a dyed in the wool SunOS guy, from SunOS3.2 on, 
but the release engineering world and the nature of the customer base 
has changed and we as a community need to help all the opensolaris.org 
based distros compete more effectively or perish under our own weight...


rich


Greg Palmer wrote:

Martin Bochnig wrote:
On Mon, Feb 16, 2009 at 4:47 PM, Rich Reynolds 
 wrote:
 

Martin -

I firmly beg to differ...

the tests were done with out of the box released versions in that 
there is

no mention of doing ubuntu package upgrades. this is just an artifact of
Sun's release engineering schedule.  should another distribution of the
opensolaris.org code base opt to include all the additional features 
that
are mentioned then not including that distro would be damning on the 
part of

the testers.
...

Rich, are you then even suggesting, that - in case the graphics chip
on that box would have been too modern for native Xorg-driver support
in Xorg server-version 1.3 as shipped in 2008.11 with all the older
matching ddx module versions (pre-libpciaccess era) - that the
benchmark results from OpenSolaris running Xorg 1.3 with vesa-fallback
driver should """directly""" be compared""" to the results from other
OS'es and being sold as / propagated as (with no specific comment)
"The Solaris results" (so damn poor) versus "The Linux xyz vers. N.2 "
results (Verry fast)?
  


Hi Martin,

 I think that it's fair to do the comparison they did, even if I hate 
the results. The fact is people getting OpenSolaris or Linux get a 
particular distribution. This compares two of those distributions.  Sure 
there are times when it hurts one side or the other more based on where 
they are in the release schedule. In this case, the Linux was ahead of 
OpenSolaris in incorporating new video enhancements and so they got much 
better benchmark results. That's fair.


 What we need to do is understand how these performance disparities can 
be addressed. Video enhancements not yet on the desktop for OpenSolaris 
means taking a look at if there is a problem or if it is just because 
Linux had a head start. If they weren't there because the process isn't 
moving quickly enough on OpenSolaris then we need to figure out how to 
improve that. On the other hand, if the performance disparities are 
because there are system functions which might not be as optimized as 
they could be we need to understand and address that.


 I believe that is all Rich is saying...

Regards,
 Greg
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] osol-11 repair boot of new vmware guest install

[Harry Putnam]

Summary:
I'm having trouble finding advice or a documentation of how to get
inside and repair and install that cannot be booted.

Details:

This is a fairly new install of osol-11 from a recent iso image.
Host is winXP using VMware (VirtualBox freezes the OS repeatedly)

I'd been running the new install for a few days and attempted to
install the vmware tools.  When I got to the part where you run the
installation script I apparently either overwrote the config file for
starting X or installed a new one with a different name 
`/etc/X11/xorg.conf'.  When that question came up it asked me if I
wanted to install the conf file..  I said yes thinking I would be
given a chance to say where.

Instead the vmware tool  install routine apparently installed
something and finished.. On reboot, after the grub prompt the gui
desktop never comes up.

Selecting the `text' boot option seemed to allow me to login at a
console but then it goes on to starting the xserver with no action on
my part... and that never completes.

It appears I am effectively locked out.  I can edit the grub prompt
but really could not find by googling, what (exactly) to do there.

A screen shot of the grub offerings after typing `e' to edit,  here:

  http://home.jtan.com/~reader/vu/disp.cgi

I haven't tried them all but the ones I have tried led to a console
prompt but then after logging in as my user... in a minute or so it
goes on trying for the gui login I guess... and then all I see is
black.  No further action seems possible

I do have the iso but not really sure how to gain entry with it
either. 

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OS laptop operates with lid closed?

[Chris Mahan]

thanks all for your answers!

Chris Mahan
chris.ma...@gmail.com
grandcentral (818) 671-1709


On Mon, Feb 16, 2009 at 3:45 AM, Joerg Schilling <
joerg.schill...@fokus.fraunhofer.de> wrote:

> Chris Mahan  wrote:
>
> > Question: Does anyone know if it would be possible to run opensolaris on
> a
> > laptop (sucha as an Acer Aspire One) and have it continue to run normally
> > when the lid was closed?
>
> if a laptop does an auto-shutdown when closing the screen, there is usually
> a BIOS menue to swoth off this feature.
>
> Jörg
>
> --
>  
> EMail:jo...@schily.isdn.cs.tu-berlin.de(home)
>  Jörg Schilling D-13353 Berlin
>   j...@cs.tu-berlin.de(uni)
>   joerg.schill...@fokus.fraunhofer.de (work) Blog:
> http://schily.blogspot.com/
>  URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily
>
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OpenSolaris install on x4150?

[Rich Reynolds]

Hi all -

I had a similar problem on different hardware, and found that hand 
plumbing the interface and re-enable nwam served as a sufficient work 
around...


hth,

rich
Alexander Vlasov wrote:

Hello,

this is known P2 bug, see 
http://defect.opensolaris.org/bz/show_bug.cgi?id=6331

Actually nwam fails and machin doesn't boots up into normal state.


Kristian Rink wrote:

Folks;

trying to boot (or even install) a recent OpenSolaris build (off
genunix.org) on an x4150 machine, but so far failed in a rather simple 
way:
Booting up the machine using the live medium works well, but in the 
end, a
text based console is all I eventually get on this machine, also no 
messages

whatsoever stating what eventually could prevent the GUI from coming up.
Unfortunately so far I haven't figured out how to do a GUI-less 
OpenSolaris
installation... So, can some kind soul point me where to peek to 
either get
the OpenSolaris UI up and working on a Sun x4150 or (which I would 
prefer)

how to do an OpenSolaris install from a text based console?

Thanks in advance and all the best,
Kristian

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
  

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Greg Palmer]

Martin Bochnig wrote:

On Mon, Feb 16, 2009 at 4:47 PM, Rich Reynolds  wrote:
  

Martin -

I firmly beg to differ...

the tests were done with out of the box released versions in that there is
no mention of doing ubuntu package upgrades. this is just an artifact of
Sun's release engineering schedule.  should another distribution of the
opensolaris.org code base opt to include all the additional features that
are mentioned then not including that distro would be damning on the part of
the testers.
...

Rich, are you then even suggesting, that - in case the graphics chip
on that box would have been too modern for native Xorg-driver support
in Xorg server-version 1.3 as shipped in 2008.11 with all the older
matching ddx module versions (pre-libpciaccess era) - that the
benchmark results from OpenSolaris running Xorg 1.3 with vesa-fallback
driver should """directly""" be compared""" to the results from other
OS'es and being sold as / propagated as (with no specific comment)
"The Solaris results" (so damn poor) versus "The Linux xyz vers. N.2 "
results (Verry fast)?
  


Hi Martin,

 I think that it's fair to do the comparison they did, even if I hate 
the results. The fact is people getting OpenSolaris or Linux get a 
particular distribution. This compares two of those distributions.  Sure 
there are times when it hurts one side or the other more based on where 
they are in the release schedule. In this case, the Linux was ahead of 
OpenSolaris in incorporating new video enhancements and so they got much 
better benchmark results. That's fair.


 What we need to do is understand how these performance disparities can 
be addressed. Video enhancements not yet on the desktop for OpenSolaris 
means taking a look at if there is a problem or if it is just because 
Linux had a head start. If they weren't there because the process isn't 
moving quickly enough on OpenSolaris then we need to figure out how to 
improve that. On the other hand, if the performance disparities are 
because there are system functions which might not be as optimized as 
they could be we need to understand and address that.


 I believe that is all Rich is saying...

Regards,
 Greg
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Martin Bochnig]

On Mon, Feb 16, 2009 at 4:47 PM, Rich Reynolds  wrote:
> Martin -
>
> I firmly beg to differ...
>
> the tests were done with out of the box released versions in that there is
> no mention of doing ubuntu package upgrades. this is just an artifact of
> Sun's release engineering schedule.  should another distribution of the
> opensolaris.org code base opt to include all the additional features that
> are mentioned then not including that distro would be damning on the part of
> the testers.
>
> this article simply points out the need for aggressive development of
> targeted distributions or far better coordination of project release
> cycles...
>
> just a thought,
>
> rich
>


Rich, are you then even suggesting, that - in case the graphics chip
on that box would have been too modern for native Xorg-driver support
in Xorg server-version 1.3 as shipped in 2008.11 with all the older
matching ddx module versions (pre-libpciaccess era) - that the
benchmark results from OpenSolaris running Xorg 1.3 with vesa-fallback
driver should """directly""" be compared""" to the results from other
OS'es and being sold as / propagated as (with no specific comment)
"The Solaris results" (so damn poor) versus "The Linux xyz vers. N.2 "
results (Verry fast)?

I cannot agree with this style of making "benchmarks".
I do see your point, but I would only agree with you, if the
benchmarker's would have employed the required care and documented it
in the appropriate verbosity.

For some reason they have done neither: Either they didn't know /
think about it, or they simply didn't want.
If I check a product and make much noise about it, then I also have
some responsibility for doing it in a fair and unbiased way.
In above case I did not get this impression.

Imagine such a thing would happen in real sports, like during
Olympics? One team gets scored from the results by a defective stop
watch, the other by a proper one, that's neither "fair", nor delivers
actual meaningful results.


Regards,
Martin


>
> Martin Bochnig wrote:
>>
>> Also take into consideration, that 2008.11 still used Xorg server 1.3
>> (rather than 1.5.x), and that DRM is only available for a limited set
>> of cards on opensolaris, and that - even on supported hw - direct
>> rendering might not been have enabled by default.
>> (Whereas on modern Linux distros all those things are the other way
>> around.)
>> The X11 system is half an OS itself.
>>
>> That's what some professional benchmarking experts do not even mention
>> in theirs misleading summary of results.
>> They should at least give Indiana 2008.11 another try after having
>> installed the latest upgrades from the Sun X11-group and fox-gate,
>> which brings the X-server to version 1.5.3 and also updates most
>> graphics drivers to their newest versions.
>> Also Mesa OpenGL has been upgraded in the interim (since 2008.11 was
>> released).
>> Where in that "benchmark" is being talked about this?
>>
>> %martin
>> ___
>> opensolaris-discuss mailing list
>> opensolaris-discuss@opensolaris.org
>
>
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Rich Reynolds]

Martin -

I firmly beg to differ...

the tests were done with out of the box released versions in that there 
is no mention of doing ubuntu package upgrades. this is just an artifact 
of Sun's release engineering schedule.  should another distribution of 
the opensolaris.org code base opt to include all the additional features 
that are mentioned then not including that distro would be damning on 
the part of the testers.


this article simply points out the need for aggressive development of 
targeted distributions or far better coordination of project release 
cycles...


just a thought,

rich


Martin Bochnig wrote:

Also take into consideration, that 2008.11 still used Xorg server 1.3
(rather than 1.5.x), and that DRM is only available for a limited set
of cards on opensolaris, and that - even on supported hw - direct
rendering might not been have enabled by default.
(Whereas on modern Linux distros all those things are the other way around.)
The X11 system is half an OS itself.

That's what some professional benchmarking experts do not even mention
in theirs misleading summary of results.
They should at least give Indiana 2008.11 another try after having
installed the latest upgrades from the Sun X11-group and fox-gate,
which brings the X-server to version 1.5.3 and also updates most
graphics drivers to their newest versions.
Also Mesa OpenGL has been upgraded in the interim (since 2008.11 was released).
Where in that "benchmark" is being talked about this?

%martin
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] emacs under solaris suggested keybindings please

[Dieter Kluenter]

Hi,

Harry Putnam  writes:

> I'm hoping some fellow emacs users will be willing to share the
> keybindings they use under solaris.
>
> I'm new to solaris but a long time emacs user (although not
> particuilarly skilled).  Certain keybindings
> that are completely integral to my usage such a Alt+x to get the M-x
> prompt do not work the same under solaris as they do under linux.
>
> I wondered what other emacs users do to make emacs more usable under
> solaris.

I first created a /etc/X11/org.conf file with proper keyboad settings
Section "InputDevice"
 Identifier  "Keyboard1"
 Driver  "kbd"
 Option  "AutoRepeat" "500 30"
 Option  "XkbRules" "xorg"
 Option  "XkbModel" "pc102"
 Option  "XkbLayout" "us"
 Option  "XkbVariant" "nodeadkeys"
EndSection

and a ~/.xmodmaprc file
keycode 64 = Meta_L Alt_L
this allows Alt+x operations

-Dieter


-- 
Dieter Klünter | Systemberatung
http://dkluenter.de
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OpenSolaris install on x4150?

[Alexander Vlasov]

Hello,

this is known P2 bug, see 
http://defect.opensolaris.org/bz/show_bug.cgi?id=6331

Actually nwam fails and machin doesn't boots up into normal state.


Kristian Rink wrote:

Folks;

trying to boot (or even install) a recent OpenSolaris build (off
genunix.org) on an x4150 machine, but so far failed in a rather simple way:
Booting up the machine using the live medium works well, but in the end, a
text based console is all I eventually get on this machine, also no messages
whatsoever stating what eventually could prevent the GUI from coming up.
Unfortunately so far I haven't figured out how to do a GUI-less OpenSolaris
installation... So, can some kind soul point me where to peek to either get
the OpenSolaris UI up and working on a Sun x4150 or (which I would prefer)
how to do an OpenSolaris install from a text based console?

Thanks in advance and all the best,
Kristian

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
  

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[ken mays]

Nice review and gives us a perspective on how things are performing on 
OpenSolaris at this time. 

As you know, there are inconsistencies in the testing. OpenSolaris binaries are 
not all built against GCC (or Sun Studio 11/12) and the installed dependency 
packages differ greatly in versions. You may look at
the test and go "it was GCC that caused the failure" but this is not true.

The concern is why so slow 'here and here' and how do we fix it?

~ Ken Mays





--- On Mon, 2/16/09, Tomasz Kloczko  wrote:

> From: Tomasz Kloczko 
> Subject: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark
> To: opensolaris-discuss@opensolaris.org
> Date: Monday, February 16, 2009, 4:20 AM
> http://www.phoronix.com/scan.php?page=article&item=amd_shanghai_opensolaris
> 
> Any comments?
> 
> Tomasz
> -- 
> Tomasz Kloczko
> Senior System Administrator, Tiscali UK Ltd
> 
> ___
> opensolaris-discuss mailing list
> opensolaris-discuss@opensolaris.org


  

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OpenSolaris install on x4150?

[Kristian Rink]

Hi Martin;

and first off thanks a bunch for your hint.

Martin Bochnig schrieb:
[...]
> Hello, as far as I know Indiana still lacks a text/console mode installer.
> However, in case your box has whatever graphics card you can simply
> use the vesa driver (as replacement for a native driver, which appears
> to be missing for you gfx chipset).
> Then you can perform a caiman GUI install.


Hmm, sad seeing the Sun operating system not nicely operating out-of-the-box
on Sun hardware. :/ Anyhow, is there an easy way of how to choose the
graphics driver used during bootup, or will I have to go the Xorg -configure
after booting up in text mode?

TIA and all the best,
Kristian

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OpenSolaris install on x4150?

[Martin Bochnig]

On Mon, Feb 16, 2009 at 1:24 PM, Kristian Rink  wrote:
> Folks;
>
> trying to boot (or even install) a recent OpenSolaris build (off
> genunix.org) on an x4150 machine, but so far failed in a rather simple way:
> Booting up the machine using the live medium works well, but in the end, a
> text based console is all I eventually get on this machine, also no messages
> whatsoever stating what eventually could prevent the GUI from coming up.
> Unfortunately so far I haven't figured out how to do a GUI-less OpenSolaris
> installation... So, can some kind soul point me where to peek to either get
> the OpenSolaris UI up and working on a Sun x4150 or (which I would prefer)
> how to do an OpenSolaris install from a text based console?
>
> Thanks in advance and all the best,
> Kristian


Hello, as far as I know Indiana still lacks a text/console mode installer.
However, in case your box has whatever graphics card you can simply
use the vesa driver (as replacement for a native driver, which appears
to be missing for you gfx chipset).
Then you can perform a caiman GUI install.

%m
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] OpenSolaris install on x4150?

[Kristian Rink]

Folks;

trying to boot (or even install) a recent OpenSolaris build (off
genunix.org) on an x4150 machine, but so far failed in a rather simple way:
Booting up the machine using the live medium works well, but in the end, a
text based console is all I eventually get on this machine, also no messages
whatsoever stating what eventually could prevent the GUI from coming up.
Unfortunately so far I haven't figured out how to do a GUI-less OpenSolaris
installation... So, can some kind soul point me where to peek to either get
the OpenSolaris UI up and working on a Sun x4150 or (which I would prefer)
how to do an OpenSolaris install from a text based console?

Thanks in advance and all the best,
Kristian

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OS laptop operates with lid closed?

[Martin Bochnig]

On Mon, Feb 16, 2009 at 11:05 AM, Chris Mahan  wrote:
> Question: Does anyone know if it would be possible to run opensolaris on a
> laptop (sucha as an Acer Aspire One) and have it continue to run normally
> when the lid was closed?
>
>
>
>
> Chris Mahan
> chris.ma...@gmail.com
> grandcentral (818) 671-1709


If power management happens to interfere in an unwanted manner, then
first you should try to change its behavior by means of dtpower. As
this utility frequently fails to work (as intended), you should first
have a look at power.conf(4) and then edit /etc/power.conf by hand.
Another easy way to get rid of power management would be to  #
"rem_drv pm" it driver. But there is more to your question, than just
power management, as you can see in a thread from last week:
http://mail.opensolaris.org/pipermail/xwin-discuss/2009-February/004419.html

%m
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OS laptop operates with lid closed?

[Joerg Schilling]

Chris Mahan  wrote:

> Question: Does anyone know if it would be possible to run opensolaris on a
> laptop (sucha as an Acer Aspire One) and have it continue to run normally
> when the lid was closed?

if a laptop does an auto-shutdown when closing the screen, there is usually
a BIOS menue to swoth off this feature.

Jörg

-- 
 EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
   j...@cs.tu-berlin.de(uni)  
   joerg.schill...@fokus.fraunhofer.de (work) Blog: 
http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OS laptop operates with lid closed?

[Alfred Peng]

If the laptop doesn't support power suspend/resume yet or the action for 
"When laptop lid is closed" in power management preference isn't set to 
suspend/shutdown, the laptop should run continuously in this case.


-Alfred

Chris Mahan wrote:
Question: Does anyone know if it would be possible to run opensolaris 
on a laptop (sucha as an Acer Aspire One) and have it continue to run 
normally when the lid was closed?





Chris Mahan
chris.ma...@gmail.com 
grandcentral (818) 671-1709


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] OS laptop operates with lid closed?

[Martin Bochnig]

On Mon, Feb 16, 2009 at 11:05 AM, Chris Mahan  wrote:
> Question: Does anyone know if it would be possible to run opensolaris on a
> laptop (sucha as an Acer Aspire One) and have it continue to run normally
> when the lid was closed?
>
>
>
>
> Chris Mahan
> chris.ma...@gmail.com
> grandcentral (818) 671-1709


lid, what's that??
the display / keyboard?

then yes, i do this with my 2 x64 amilo laptops.
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[przemolicc]

On Mon, Feb 16, 2009 at 09:20:34AM +, Tomasz Kloczko wrote:
> 
> http://www.phoronix.com/scan.php?page=article&item=amd_shanghai_opensolaris
> 
> Any comments?

You have many comments and discussion at
http://www.phoronix.com/forums/showthread.php?t=15321#post61968



Regards
Przemyslaw Bak (przemol)
--
http://przemol.blogspot.com/






















--
Zostan mistrzem parkowania w Bombaju!
Zagraj >> http://link.interia.pl/f204e 

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] OS laptop operates with lid closed?

[Chris Mahan]

Question: Does anyone know if it would be possible to run opensolaris on a
laptop (sucha as an Acer Aspire One) and have it continue to run normally
when the lid was closed?




Chris Mahan
chris.ma...@gmail.com
grandcentral (818) 671-1709
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Casper . Dik]

It's true on SPARC (not a whole lot of extra registers, except for a few fps),
not true on Amd64.  (Twice the registers, using the registers to call
functions and not the stack)

(Open)Solaris still runs on 32 bit machines; so we need to have 32 bit
binaries for all commands.

I don't know exactly how all the benchmarks where compiled, but:

- which compiler?
- by default, our compilers (as installed) generate 32 bit binaries

Casper
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Martin Bochnig]

Also take into consideration, that 2008.11 still used Xorg server 1.3
(rather than 1.5.x), and that DRM is only available for a limited set
of cards on opensolaris, and that - even on supported hw - direct
rendering might not been have enabled by default.
(Whereas on modern Linux distros all those things are the other way around.)
The X11 system is half an OS itself.

That's what some professional benchmarking experts do not even mention
in theirs misleading summary of results.
They should at least give Indiana 2008.11 another try after having
installed the latest upgrades from the Sun X11-group and fox-gate,
which brings the X-server to version 1.5.3 and also updates most
graphics drivers to their newest versions.
Also Mesa OpenGL has been upgraded in the interim (since 2008.11 was released).
Where in that "benchmark" is being talked about this?

%martin
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Tomasz Kloczko]

On Mon, 2009-02-16 at 10:37 +0100, casper@sun.com wrote:
> There are too many variables; e.g., how is the GraphicsMagick compiled
> and run?
> 
> Also, if they are running the standard Solaris software, OpenSolaris may 
> hurt because most of the software is 32 bit.

Interesting .. IIRC one of the argument still using 32 bit version in
user space was 32 bit version is faster. I know that it is not always
true because compiler can generate code for use more registers available
in this mode. So .. which binaries are faster ? 64 or 32 bit in case
standard CMDL tools? :)
If 32 bit binaries are slower why not switch user space all to default
64 bit and keep 32 libraries for backward compatibility?

Never the less .. this can´t explain this difference like ~two times
slower sqlite.

Tomasz
-- 
Tomasz Kloczko
Senior System Administrator, Tiscali UK Ltd

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Casper . Dik]

There are too many variables; e.g., how is the GraphicsMagick compiled
and run?

Also, if they are running the standard Solaris software, OpenSolaris may 
hurt because most of the software is 32 bit.

Casper

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Tomasz Kloczko]

On Mon, 2009-02-16 at 09:20 +, Tomasz Kloczko wrote:
> http://www.phoronix.com/scan.php?page=article&item=amd_shanghai_opensolaris
> 
> Any comments?

Mine comments is: I know that some differences are result of using
different versions of gcc (gcc 4.x on Linux and 3.x on OpenSolaris).
Question: why and/or when OpenSoloaris will be switched to new gcc?
(if not why?)

More interesting are results like GraphicsMagics benchmarks and other
similar where OS is few times slower than Linux.

Tomasz
-- 
Tomasz Kloczko
Senior System Administrator, Tiscali UK Ltd

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Damian Wojslaw]

Tomasz Kloczko pisze:

http://www.phoronix.com/scan.php?page=article&item=amd_shanghai_opensolaris

Any comments?

Tomasz


	Of course. In my opinion OpenSolaris makes up in scalability and 
stability, what it "looses" in speed in those tests. Also, in everyday 
work I can't tell speed difference between Ubuntu and OpenSolaris, 
desktop use included.
	The geek and Solaris fan inside me would like those numbers to be 
reversed, but honestly I see them as quite artificial.


Regards

--
Damian Wojsław
http://pl.opensolaris.org/ | SCA OS0073
http://www.opensolaris.org/os/community/on/
http://opensolaris.org/os/project/szc-osug/
http://www.linkedin.com/in/trochej
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] Phoronix Linux vs OpenSolaris benchmark

[Tomasz Kloczko]

http://www.phoronix.com/scan.php?page=article&item=amd_shanghai_opensolaris

Any comments?

Tomasz
-- 
Tomasz Kloczko
Senior System Administrator, Tiscali UK Ltd

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org