[i945g] freezes shortly after loading desktop

some one please help me! I can't do much at this point and don't know where to turn.

I have compiled three test kernels for finding the triggering commit that introduced these freezes on 845G into ubuntu. They can be downloaded from http://www.kvante.info/845Gfreeze/ . There is some more detailed information in the comments to bug 456902. If this bug is the same as bug 456902, the kernels with freezetest1 and freezetest8 should not cause the freeze, while the one with freezetest9 should cause a freeze. If this is not the case for you, we know that your bug is different from bug 456902 and we will have to do some additional tests.

I originally compiled the kernels on Jaunty, but tests by others seems to indicate that the tests can also be done in Karmic or Lucid. A problem with Karmic and Lucid is that the default kernel will cause a freeze while it doesn't in Jaunty.

Sorry, I pasted the previous comment into the wrong bug report. It was supposed to go to bug 496903.

daves111, look at https://wiki.ubuntu.com/X/Troubleshooting/Freeze for more information about this kind of bug. Especially, we would like batchbuffer dumps. Also, if you can try some older mainline kernels from [1] (see also [2]) and find the first with this problem, that would be useful.

[1]: http://kernel.ubuntu.com/~kernel-ppa/mainline/
[2]: https://wiki.ubuntu.com/KernelTeam/MainlineBuilds

This comment goes to bug 400934, bug 451518, bug 475429, bug 500686,
bug 501875, bug 342675, bug 4611163 and 493111.

We have had several reports of freezes (GPU hangs) on i945 chipsets
(945G, 945GM and 945GME, all of which uses the GMA950 graphics core)
in Karmic. I would like to get some testing going so that we may
identify the cause and give upstream all the information they need
(and possibly some they do not need) so that they can fix this before
Lucid is out. I have set up a wiki page to coordinate the effort:
https://wiki.ubuntu.com/X/Lucidi945Freezes . There is a list of test
that I think will be useful in order to isolate where the problem is
and a table where you can fill in your test results. If we're lucky
there is only one underlying problem, and we will probably get
consistent test results. Otherwise, different test results may group
the problems in different categories.

I realize that doing all the tests can be quite time consuming (it
calls for tests on both Karmic and Lucid), and if you are only able to
do some of the tests it would be nice if you could fill in those cells
in the test result table.

My initial guess is that the bug is in mesa, since it seems to be
related to 3D operations. It would be nice if we could also get some
testing of different mesa versions, but I'm not aware of any that are
pre-built yet.

I'll try to get to it this afternoon - thanks for your help. Dave

----- Original Message -----
From: "Geir Ove Myhr" <email address hidden>
To: <email address hidden>
Sent: Friday, January 01, 2010 12:12 PM
Subject: [Bug 500686] Re: [i945g] freezes shortly after loading desktop

This comment goes to bug 400934, bug 451518, bug 475429, bug 500686,
bug 501875, bug 342675, bug 4611163 and 493111.

We have had several reports of freezes (GPU hangs) on i945 chipsets
(945G, 945GM and 945GME, all of which uses the GMA950 graphics core)
in Karmic. I would like to get some testing going so that we may
identify the cause and give upstream all the information they need
(and possibly some they do not need) so that they can fix this before
Lucid is out. I have set up a wiki page to coordinate the effort:
https://wiki.ubuntu.com/X/Lucidi945Freezes . There is a list of test
that I think will be useful in order to isolate where the problem is
and a table where you can fill in your test results. If we're lucky
there is only one underlying problem, and we will probably get
consistent test results. Otherwise, different test results may group
the problems in different categories.

I realize that doing all the tests can be quite time consuming (it
calls for tests on both Karmic and Lucid), and if you are only able to
do some of the tests it would be nice if you could fill in those cells
in the test result table.

My initial guess is that the bug is in mesa, since it seems to be
related to 3D operations. It would be nice if we could also get some
testing of different mesa versions, but I'm not aware of any that are
pre-built yet.

--
[i945g] freezes shortly after loading desktop
https://bugs.launchpad.net/bugs/500686
You received this bug notification because you are a direct subscriber
of the bug.

Geir Ove Myhr, I tried kernel v2.6.33-rc and exactly the same thing
happened: desktop loaded and I could move the mouse around until I moved
it to the top over a hardrive error icon at the top (my second drive is
bad) . I will attempt to enter this on your table on the page you set up
Dave

On Fri, 2010-01-01 at 20:12 +0000, Geir Ove Myhr wrote:
> This comment goes to bug 400934, bug 451518, bug 475429, bug 500686,
> bug 501875, bug 342675, bug 4611163 and 493111.
>
> We have had several reports of freezes (GPU hangs) on i945 chipsets
> (945G, 945GM and 945GME, all of which uses the GMA950 graphics core)
> in Karmic. I would like to get some testing going so that we may
> identify the cause and give upstream all the information they need
> (and possibly some they do not need) so that they can fix this before
> Lucid is out. I have set up a wiki page to coordinate the effort:
> https://wiki.ubuntu.com/X/Lucidi945Freezes . There is a list of test
> that I think will be useful in order to isolate where the problem is
> and a table where you can fill in your test results. If we're lucky
> there is only one underlying problem, and we will probably get
> consistent test results. Otherwise, different test results may group
> the problems in different categories.
>
> I realize that doing all the tests can be quite time consuming (it
> calls for tests on both Karmic and Lucid), and if you are only able to
> do some of the tests it would be nice if you could fill in those cells
> in the test result table.
>
> My initial guess is that the bug is in mesa, since it seems to be
> related to 3D operations. It would be nice if we could also get some
> testing of different mesa versions, but I'm not aware of any that are
> pre-built yet.
>

I didn't choose the "recovery" choice of v2.6.33-rc2 - I tried to enter the info on the chart/table on the page you made, but couldn't see how. Dave

> Geir Ove Myhr, I tried kernel v2.6.33-rc and exactly the same thing
> happened: desktop loaded and I could move the mouse around until I moved

I assume you were trying to do the following test:
 * Does it help to use a recent drm-intel-next kernel from this
archive? (more info at KernelTeam/MainlineBuild)
I see that this could be misunderstood. drm-intel-next is actually a
subfolder of the mainline kernel build directory you get to by
following the link. So v2.6.33-rc2 wasn't really what I had in mind
(but it's still useful information). I will update the link to point
directly to the drm-intel-next folder.

> I didn't choose the "recovery" choice of v2.6.33-rc2 - I tried to enter
> the info on the chart/table on the page you made, but couldn't see how.

You need to log in with your Launchpad username and password (in the
top right corner). Then you can choose the Edit link in the top left
corner. The easiest way to add your info to the table is to copy one
of the example lines (they're long and therefore wrapped over several
text lines) and substitute the info for your own (removing what is
there if you want to leave it blank). You can use the preview button
to check that it worked the way you wanted it to.

Geir Ove Myhr, I think I am going to drop back two versions of ubuntu to
7.10 or 8.10 (I can't remember which one it was that the computer last
worked). I spent 1/2 a day trying to load kernels and do a batchbuffer dump
with ssh. I really need to be able to use this computer ... so, I'll just
wait until it gets figured out .... or maybe try lucid when it comes
available. Thanks, Dave

----- Original Message -----
From: "Geir Ove Myhr" <email address hidden>
To: <email address hidden>
Sent: Friday, January 01, 2010 5:37 PM
Subject: Re: [Bug 500686] Re: [i945g] freezes shortly after loading desktop

> Geir Ove Myhr, I tried kernel v2.6.33-rc and exactly the same thing
> happened: desktop loaded and I could move the mouse around until I moved

I assume you were trying to do the following test:
 * Does it help to use a recent drm-intel-next kernel from this
archive? (more info at KernelTeam/MainlineBuild)
I see that this could be misunderstood. drm-intel-next is actually a
subfolder of the mainline kernel build directory you get to by
following the link. So v2.6.33-rc2 wasn't really what I had in mind
(but it's still useful information). I will update the link to point
directly to the drm-intel-next folder.

> I didn't choose the "recovery" choice of v2.6.33-rc2 - I tried to enter
> the info on the chart/table on the page you made, but couldn't see how.

You need to log in with your Launchpad username and password (in the
top right corner). Then you can choose the Edit link in the top left
corner. The easiest way to add your info to the table is to copy one
of the example lines (they're long and therefore wrapped over several
text lines) and substitute the info for your own (removing what is
there if you want to leave it blank). You can use the preview button
to check that it worked the way you wanted it to.

--
[i945g] freezes shortly after loading desktop
https://bugs.launchpad.net/bugs/500686
You received this bug notification because you are a direct subscriber
of the bug.

Okay, thanks for letting us know. I will close this bug report, since you will not be able to follow up.

My experience is that Jaunty (9.04) is a quite good release and I would probably fall back to this if I were you. In production environment I prefer Hardy (8.04) which is a Long Term Support release.

In order to help us getting this sorted out for Lucid, maybe you can test that release by upgrading (with `update-manager -d` and using the upgrade button) and verifying that the problem is still present there before you wipe your computer and install an older release.

ok, I will test Lucid for you. Is there a chance that a problem is that
I don't have a real 64 bit amd? my documentation shows that I have a
"intel celeron d-360 3.46GHz 64-bit processor". "System Monitor" shows
Processor: Intel Celeron D cpu 3.46GHz. Should I be using the 64 bit
version of unbuntu? Dave

On Mon, 2010-01-04 at 16:41 +0000, Geir Ove Myhr wrote:
> Okay, thanks for letting us know. I will close this bug report, since
> you will not be able to follow up.
>
> My experience is that Jaunty (9.04) is a quite good release and I would
> probably fall back to this if I were you. In production environment I
> prefer Hardy (8.04) which is a Long Term Support release.
>
> In order to help us getting this sorted out for Lucid, maybe you can
> test that release by upgrading (with `update-manager -d` and using the
> upgrade button) and verifying that the problem is still present there
> before you wipe your computer and install an older release.
>

On Mon, Jan 4, 2010 at 7:13 PM, daves111 wrote:
> ok, I will test Lucid for you. Is there a chance that a problem is that
> I don't have a real 64 bit amd?

Nope, the name amd64 is there only for historical reasons. The
platform was invented by AMD while Intel had another 64-bit
architecture (lpia), but intel has since adopted AMDs platform. I must
admit I was also confused by the name at first.

On the other hand, it shouldn't be a problem to use i386 either, as
amd64 is backwards compatible. I think i386 is currently recommended
for desktop systems and amd64 for servers, but I'm no authority on
this.

If you feel that you have time and Lucid still freezes, you may try
with xorg-edgers. That amounts to adding two lines to
/etc/apt/sources.list and running `sudo apt-get dist-update; sudo
apt-get dist-upgrade`. That gives you a bleeding edge ubuntu (with the
most recent bug fixes, but also the most recent bugs). It is also a
very useful datapoint to know if this freezes (unless Lucid without it
does not). xorg-edgers is at
https://launchpad.net/~xorg-edgers/+archive/ppa .

Feel free to decline the last test if you think it sounds like it is
too much for you.

I upgraded to Lucid; I still had a list of kernels showing before I get to
login; I tried several including 2.6.32-9-generic but those that went on to
the desktop here is what happened: I could move the mouse around the screen
until I hovered it over something that had a hint (or whatever it is called)
that showed ... then it would freeze and have to be shut off. On the 64-bit,
then, ... am I right in using ubuntu 64 for my computer? Dave

----- Original Message -----
From: "Geir Ove Myhr" <email address hidden>
To: <email address hidden>
Sent: Monday, January 04, 2010 4:32 PM
Subject: Re: [Bug 500686] Re: [i945g] freezes shortly after loading desktop

On Mon, Jan 4, 2010 at 7:13 PM, daves111 wrote:
> ok, I will test Lucid for you. Is there a chance that a problem is that
> I don't have a real 64 bit amd?

Nope, the name amd64 is there only for historical reasons. The
platform was invented by AMD while Intel had another 64-bit
architecture (lpia), but intel has since adopted AMDs platform. I must
admit I was also confused by the name at first.

On the other hand, it shouldn't be a problem to use i386 either, as
amd64 is backwards compatible. I think i386 is currently recommended
for desktop systems and amd64 for servers, but I'm no authority on
this.

If you feel that you have time and Lucid still freezes, you may try
with xorg-edgers. That amounts to adding two lines to
/etc/apt/sources.list and running `sudo apt-get dist-update; sudo
apt-get dist-upgrade`. That gives you a bleeding edge ubuntu (with the
most recent bug fixes, but also the most recent bugs). It is also a
very useful datapoint to know if this freezes (unless Lucid without it
does not). xorg-edgers is at
https://launchpad.net/~xorg-edgers/+archive/ppa .

Feel free to decline the last test if you think it sounds like it is
too much for you.

--
[i945g] freezes shortly after loading desktop
https://bugs.launchpad.net/bugs/500686
You received this bug notification because you are a direct subscriber
of the bug.

On Tue, Jan 5, 2010 at 12:29 PM, daves111 wrote:
> I upgraded to Lucid; I still had a list of kernels showing before I get to
> login; I tried several including 2.6.32-9-generic but those that went on to
> the desktop here is what happened: I could move the mouse around the screen
> until I hovered it over something that had a hint (or whatever it is called)
> that showed ... then it would freeze and have to be shut off. On the 64-bit,
> then, ... am I right in using ubuntu 64 for my computer? Dave

Thank you for verifying that the problem is still present in Lucid. I
have tried to reproduce it on my wife's computer (which has 945GM)
with a LiveCD, but that particular model apparently doesn't have this
problem.

When you upgrade, the old kernels are not uninstalled. That is why you
had several kernels to choose from. 2.6.32-9 should be the standard
Lucid kernel. I think the hint is called a tooltip. It's okay to use
64-bit. I have that on my laptop as well (on the test/daily-use
partition which used to have Karmic and now has Lucid, I have a 32-bit
Jaunty one for work that I set up and customised and don't want to
change).

here is some more info that may help: installed 9.04and it worked fine
with kernel 2.6.28-11 generic; then I updated which gave me kernel
2.6.28-17 and it still works in gnome; then I tried kernel 2.6.31-15
generic and it froze like before as did 2.6.31-14 (I think). How do I
get rid of the other choices so it will just boot without stopping at
that menu? Is there some way to let me know if this bug gets fixed so I
can upgrade and so I can move up to Lucid in April? Dave

On Tue, 2010-01-05 at 18:25 +0000, Geir Ove Myhr wrote:
> On Tue, Jan 5, 2010 at 12:29 PM, daves111 wrote:
> > I upgraded to Lucid; I still had a list of kernels showing before I get to
> > login; I tried several including 2.6.32-9-generic but those that went on to
> > the desktop here is what happened: I could move the mouse around the screen
> > until I hovered it over something that had a hint (or whatever it is called)
> > that showed ... then it would freeze and have to be shut off. On the 64-bit,
> > then, ... am I right in using ubuntu 64 for my computer? Dave
>
> Thank you for verifying that the problem is still present in Lucid. I
> have tried to reproduce it on my wife's computer (which has 945GM)
> with a LiveCD, but that particular model apparently doesn't have this
> problem.
>
> When you upgrade, the old kernels are not uninstalled. That is why you
> had several kernels to choose from. 2.6.32-9 should be the standard
> Lucid kernel. I think the hint is called a tooltip. It's okay to use
> 64-bit. I have that on my laptop as well (on the test/daily-use
> partition which used to have Karmic and now has Lucid, I have a 32-bit
> Jaunty one for work that I set up and customised and don't want to
> change).
>

On Wed, Jan 6, 2010 at 6:39 PM, daves111 <email address hidden> wrote:
> here is some more info that may help: installed 9.04and it worked fine
> with kernel 2.6.28-11 generic; then I updated which gave me kernel
> 2.6.28-17 and it still works in gnome; then I tried kernel 2.6.31-15
> generic and it froze like before as did 2.6.31-14 (I think).

Actually, that's awesome! That means that there is a kernel commit
that triggers this bug (this is also the case for some freezes on 845G
and 855GM that I'm troubleshooting. And since you have your standard
Jaunty kernel that works well, you may be able to help identify the
bad commit by testing some other kernels. That would be a very big
step towards resolving this bug.

> How do I
> get rid of the other choices so it will just boot without stopping at
> that menu?

I assumed you install deb-packages with file names
linux-image-2.6.31-14-generic_2.6.31-14.48_i386.deb and similar to
install the extra kernels. To remove them completely from your
system, run `dpkg -P linux-image-2.6.31-14-generic` and similar. To
get a list of kernel packages installed on your system, run `dpkg -l |
grep linux-image`.

> Is there some way to let me know if this bug gets fixed so I
> can upgrade and so I can move up to Lucid in April? Dave

Write down the bug numbers of the bug reports on these links:
https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bugs?field.tag=945g%20freeze&field.tags_combinator=ALL
https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bugs?field.tag=945gm%20freeze&field.tags_combinator=ALL
(bug reports are removed from this list as bugs get closed). Check on
them in April and see if the have beeen fixed. At least some of them
will be the same problem that you have.

To check which kernel version introduces the freezes, there are
prebuilt mainline kernels at [1], see [2] for more info. We may be
lucky that the commit that introduced freezes on either 845G or 855GM
is the same one that introduces them for you. To test that, install
the kernels with freezetest8 and freezetest9 in the filename from [3],
if freezetest8 does not freeze but freezetest9 does, the problematic
kernel change is the same as for 845G. If not, install the one with
599 and 600 in the filename from [4]. If 599 does not freeze bug 600
freezes, it's the same commit as for 855GM.

The Karmic kernel 2.6.31-14-generic corresponds to mainline kernel
2.6.31.4. See [5] for a full list.

[1]: http://kernel.ubuntu.com/~kernel-ppa/mainline/
[2]: https://wiki.ubuntu.com/KernelTeam/MainlineBuilds
[3]: http://www.kvante.info/845Gfreeze/
[4]: http://www.kvante.info/855GMfreeze/
[5] http://kernel.ubuntu.com/~kernel-ppa/info/kernel-version-map.html

ok, here's what I've done, so far: 2.6.30 020630 of Jun 10 '09 worked
fine, but none of the 2.6.31's have worked so far; I tried a number of
them. I downloaded several of them, but only one 31 would show up labled
"-generic"; I deleted the 31 that showed up when I did "dpkg -l" then I
downloaded 31-rc1-fix1 but it wouldn't show. next I'm trying 31-rc2 of
25 Jun. ... I'll let you know how it goes. Dave

On Thu, 2010-01-07 at 00:35 +0000, Geir Ove Myhr wrote:
> On Wed, Jan 6, 2010 at 6:39 PM, daves111 <email address hidden> wrote:
> > here is some more info that may help: installed 9.04and it worked fine
> > with kernel 2.6.28-11 generic; then I updated which gave me kernel
> > 2.6.28-17 and it still works in gnome; then I tried kernel 2.6.31-15
> > generic and it froze like before as did 2.6.31-14 (I think).
>
> Actually, that's awesome! That means that there is a kernel commit
> that triggers this bug (this is also the case for some freezes on 845G
> and 855GM that I'm troubleshooting. And since you have your standard
> Jaunty kernel that works well, you may be able to help identify the
> bad commit by testing some other kernels. That would be a very big
> step towards resolving this bug.
>
> > How do I
> > get rid of the other choices so it will just boot without stopping at
> > that menu?
>
> I assumed you install deb-packages with file names
> linux-image-2.6.31-14-generic_2.6.31-14.48_i386.deb and similar to
> install the extra kernels. To remove them completely from your
> system, run `dpkg -P linux-image-2.6.31-14-generic` and similar. To
> get a list of kernel packages installed on your system, run `dpkg -l |
> grep linux-image`.
>
> > Is there some way to let me know if this bug gets fixed so I
> > can upgrade and so I can move up to Lucid in April? Dave
>
> Write down the bug numbers of the bug reports on these links:
> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bugs?field.tag=945g%20freeze&field.tags_combinator=ALL
> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bugs?field.tag=945gm%20freeze&field.tags_combinator=ALL
> (bug reports are removed from this list as bugs get closed). Check on
> them in April and see if the have beeen fixed. At least some of them
> will be the same problem that you have.
>
> To check which kernel version introduces the freezes, there are
> prebuilt mainline kernels at [1], see [2] for more info. We may be
> lucky that the commit that introduced freezes on either 845G or 855GM
> is the same one that introduces them for you. To test that, install
> the kernels with freezetest8 and freezetest9 in the filename from [3],
> if freezetest8 does not freeze but freezetest9 does, the problematic
> kernel change is the same as for 845G. If not, install the one with
> 599 and 600 in the filename from [4]. If 599 does not freeze bug 600
> freezes, it's the same commit as for 855GM.
>
> The Karmic kernel 2.6.31-14-generic corresponds to mainline kernel
> 2.6.31.4. See [5] for a full list.
>
> [1]: http://kernel.ubuntu.com/~kernel-ppa/mainline/
> [2]: https://wiki.ubuntu.com/KernelTeam/MainlineBuilds
> [3]: http://www.kvante.info/845Gfreeze/
> [4]: http://www.kvant...

Well ... the reason some of the kernels weren't showing up was I was
forgetting to install them - forgot a step.

After a number of hours of testing kernels, it didn't seem to make as
much difference the order shown on
http://kernel.ubuntu.com/~kernel-ppa/mainline/ but, rather the date that
it was last modified; everthing up to 14 Aug worked
these did not work:
v2.6.31-rc7/22-Aug-2009 11:41
v2.6.31-rc8/28-Aug-2009 11:41 v
2.6.31-rc9/06-Sep-2009 11:43
v2.6.31.1/29-Sep-2009 18:32

v2.6.31/11-Sep-2009 00:29

these did:
v2.6.31-rc6/14-Aug-2009 11:24
v2.6.30/10-Jun-2009 11:27 -
v2.6.31-rc1-fix1/06-Jul-2009 13:06 -
v2.6.31-rc2/05-Jul-2009 11:16

How do I get rid of all the list of kernel choices on the page that
stops - I deleted most of them as you said "To remove them completely
from your" but they keep showing up as a long list...

MORE INFO:
v2.6.31-rc6/14-Aug-2009 11:24 worked fine but rc 7froze with a blank
screen - ctrl/alt/delete did not do anything;
v2.6.31-rc8/28-Aug-2009 11:41 froze a little differently ... at the
screen with the ubuntu symbol with a red line under it, the line turned
yellow, as usual for the first 1 centimeter of 9; then froze; I did a
ctrl/alt/delete and a screen which said the following appeared:
"Couldn't start the x server (your graphical environment) due to some
internal error. Contact Administrator or check system log to diagnose.
In the Meantime this display will be disabled. Please restart GDM when
problem is corrected." it had a dos/terminal-like command line with
"ubuntu 9.04 presario tty1" then "presario login _" (preario is the name
of this computer). and allowed me to log in at this command line.

well, let me know what you think; hope this all helps. Dave

system, run `dpkg -P linux-image-2.6.31-14-generic` and similar.

On Thu, 2010-01-07 at 00:35 +0000, Geir Ove Myhr wrote:
> On Wed, Jan 6, 2010 at 6:39 PM, daves111 <email address hidden> wrote:
> > here is some more info that may help: installed 9.04and it worked fine
> > with kernel 2.6.28-11 generic; then I updated which gave me kernel
> > 2.6.28-17 and it still works in gnome; then I tried kernel 2.6.31-15
> > generic and it froze like before as did 2.6.31-14 (I think).
>
> Actually, that's awesome! That means that there is a kernel commit
> that triggers this bug (this is also the case for some freezes on 845G
> and 855GM that I'm troubleshooting. And since you have your standard
> Jaunty kernel that works well, you may be able to help identify the
> bad commit by testing some other kernels. That would be a very big
> step towards resolving this bug.
>
> > How do I
> > get rid of the other choices so it will just boot without stopping at
> > that menu?
>
> I assumed you install deb-packages with file names
> linux-image-2.6.31-14-generic_2.6.31-14.48_i386.deb and similar to
> install the extra kernels. To remove them completely from your
> system, run `dpkg -P linux-image-2.6.31-14-generic` and similar. To
> get a list of kernel packages installed on your system, run `dpkg -l |
> grep linux-image`.
>
> > Is there some way to let me know if this bug gets fixed so I
> > can upgrade and so I can move up to Lucid in April? Dave
>
> Wri...

> After a number of hours of testing kernels, it didn't seem to make as
> much difference the order shown on
> http://kernel.ubuntu.com/~kernel-ppa/mainline/ but, rather the date that
> it was last modified; everthing up to 14 Aug worked

Oh.. the order on that list is purely alphabetical. There is no total
order. Both 2.6.30.1 and 2.6.31-rc1 are based on 2.6.30. So a
2.6.30.X is hard to compare to a 2.6.31-rcY. Since the Karmic kernel
is based on 2.6.31.4 I guess the relavant order to use is 2.6..29,
2.6.30-rc1, ..., 2.6.30-rc8 , 2.6.30, 2.6.31-rc1, ..., 2.6.31-rc9,
2.6.31, 2.6.31.1, 2.6.31.4. You can sort them according to date by
using the URL: http://kernel.ubuntu.com/~kernel-ppa/mainline/?C=M;O=D

> these did not work:
> v2.6.31-rc7/22-Aug-2009 11:41
> v2.6.31-rc8/28-Aug-2009 11:41 v
> 2.6.31-rc9/06-Sep-2009 11:43
> v2.6.31.1/29-Sep-2009 18:32
>
> v2.6.31/11-Sep-2009 00:29
>
> these did:
> v2.6.31-rc6/14-Aug-2009 11:24
> v2.6.30/10-Jun-2009 11:27 -
> v2.6.31-rc1-fix1/06-Jul-2009 13:06 -
> v2.6.31-rc2/05-Jul-2009 11:16

This is great! That means that there is something between 2.6.31-rc6
and 2.6.31-rc7 that introduced the freezes. I will see if I can get
some test kernels built for you that has only some of those changes.
That way we may find the exact commit that introduced the freezes.

> How do I get rid of all the list of kernel choices on the page that
> stops - I deleted most of them as you said "To remove them completely
> from your" but they keep showing up as a long list...

Do the kernels that you tried to remove still show up in the output of
`dpkg -l | grep linux-image`? If not, try to run `sudo update-grub`. I
thought this was automatically run when you removed a kernel package,
but I know the behaviour has changed somewhat between the Jaunty and
Karmic versions, and maybe Jaunty doesn't do it. The actual entries
come from the file /boot/grub/menu.lst, but this should be poplulated
with only the installed kernels when you run `sudo update-grub`.

> MORE INFO:

Let's hope the difference in symptoms between rc7 and rc8 is not
relevant for now. I'll keep it in mind, though.

> This is great! That means that there is something between 2.6.31-rc6
> and 2.6.31-rc7 that introduced the freezes. I will see if I can get
> some test kernels built for you that has only some of those changes.
> That way we may find the exact commit that introduced the freezes.

I have compiled a few kernels and uploaded them to
http://www.kvante.info/945freeze/ . If you could tell me which ones
freeze, we should be able to identify the bad commit with one more
round of testing. The 3-digit number (299,300,499) in the file name
indicate the order. If you test 300 first, you only need to test 299
if it freezes and 499 if it doesn't freeze. I'm compiling a 799 now in
case none of them freeze, but I may not be able to upload it right
away.

Once we have found the bad commit, I will get someone on Karmic to
reproduce the findings there. Hopefully things do not change there.

Geir Ove

I am going to try this one, 799, next; 300 and 499 did not freeze. I
thought you might be interested in the details shown in terminal as I
did sudo dpkg -i *.deb on 799 Dave

daves111@presario:~/Desktop$ dir
10126.pdf
linux-image-2.6.31-rc6-945test799-gite3b2415_gomyhr1_amd64.deb
daves111@presario:~/Desktop$ sudo dpkg -i *.deb
[sudo] password for daves111:
Selecting previously deselected package
linux-image-2.6.31-rc6-945test799-gite3b2415.
(Reading database ... 327688 files and directories currently installed.)
Unpacking linux-image-2.6.31-rc6-945test799-gite3b2415 (from
linux-image-2.6.31-rc6-945test799-gite3b2415_gomyhr1_amd64.deb) ...
Done.
Setting up linux-image-2.6.31-rc6-945test799-gite3b2415 (gomyhr1) ...

 Hmm. There is a symbolic
link /lib/modules/2.6.31-rc6-945test799-gite3b2415/build
 However, I can not read it: No such file or directory
 Therefore, I am
deleting /lib/modules/2.6.31-rc6-945test799-gite3b2415/build

 Hmm. The package shipped with a symbolic
link /lib/modules/2.6.31-rc6-945test799-gite3b2415/source
 However, I can not read the target: No such file or directory
 Therefore, I am
deleting /lib/modules/2.6.31-rc6-945test799-gite3b2415/source

Running depmod.
Finding valid ramdisk creators.
Using mkinitramfs-kpkg to build the ramdisk.
Running postinst hook script update-grub.
Searching for GRUB installation directory ... found: /boot/grub
Searching for default file ... found: /boot/grub/default
Testing for an existing GRUB menu.lst file ...
found: /boot/grub/menu.lst
Searching for splash image ... none found, skipping ...
Found kernel: /boot/vmlinuz-2.6.31-rc6-945test799-gite3b2415
Found kernel: /boot/vmlinuz-2.6.31-rc6-945test499-git6a719e0
Found kernel: /boot/vmlinuz-2.6.31-rc6-945test300-gitebb177d
Found kernel: /boot/vmlinuz-2.6.28-17-generic
Found kernel: /boot/vmlinuz-2.6.28-11-generic
Found kernel: /boot/memtest86+.bin
Replacing config file /var/run/grub/menu.lst with new version
Updating /boot/grub/menu.lst ... done

Examining /etc/kernel/postinst.d.
run-parts: executing /etc/kernel/postinst.d/nvidia-common

Well, 799 didn't freeze either; it said it was 799 when it gave a choice
of which kernel to use. Take a look at my previous email to see if it
did something strange when it installed the 799 kernel. I hope this
doesn't confuse things; it looked like we were narrowing it down. Dave

On Sat, 2010-01-09 at 09:38 +0000, Geir Ove Myhr wrote:
> > This is great! That means that there is something between 2.6.31-rc6
> > and 2.6.31-rc7 that introduced the freezes. I will see if I can get
> > some test kernels built for you that has only some of those changes.
> > That way we may find the exact commit that introduced the freezes.
>
> I have compiled a few kernels and uploaded them to
> http://www.kvante.info/945freeze/ . If you could tell me which ones
> freeze, we should be able to identify the bad commit with one more
> round of testing. The 3-digit number (299,300,499) in the file name
> indicate the order. If you test 300 first, you only need to test 299
> if it freezes and 499 if it doesn't freeze. I'm compiling a 799 now in
> case none of them freeze, but I may not be able to upload it right
> away.
>
> Once we have found the bad commit, I will get someone on Karmic to
> reproduce the findings there. Hopefully things do not change there.
>
> Geir Ove
>

Don't know if this is of any interest to you; this is what it said when
I went to delete 300. I am copying these off my terminal program. Dave

daves111@presario:~$ sudo dpkg -l |
> grep linux-image
[sudo] password for daves111:
pi linux-image-2.6.28-11-generic 2.6.28-11.42
Linux kernel image for version 2.6.28 on x86
ii linux-image-2.6.28-17-generic 2.6.28-17.58
Linux kernel image for version 2.6.28 on x86
iHR linux-image-2.6.31-rc6-945test300-gitebb177d gomyhr1
Linux kernel binary image for version 2.6.31
ii linux-image-2.6.31-rc6-945test499-git6a719e0 gomyhr1
Linux kernel binary image for version 2.6.31
ii linux-image-2.6.31-rc6-945test799-gite3b2415 gomyhr1
Linux kernel binary image for version 2.6.31
ii linux-image-generic 2.6.28.17.22
Generic Linux kernel image
daves111@presario:~$ sudo dpkg -P
linux-image-2.6.31-rc6-945test300-gitebb177d
dpkg: error processing linux-image-2.6.31-rc6-945test300-gitebb177d
(--purge):
 Package is in a very bad inconsistent state - you should
 reinstall it before attempting a removal.
Errors were encountered while processing:
 linux-image-2.6.31-rc6-945test300-gitebb177d

when I deleted 499 here are the terminal details (don't know if this is
of interest to you, but ...):

daves111@presario:~$ sudo dpkg -P
linux-image-2.6.31-rc6-945test499-git6a719e0
[sudo] password for daves111:
(Reading database ... 329968 files and directories currently installed.)
Removing linux-image-2.6.31-rc6-945test499-git6a719e0 ...
Examining /etc/kernel/prerm.d.
run-parts: executing /etc/kernel/prerm.d/last-good-boot
Running postrm hook script /sbin/update-grub.
Searching for GRUB installation directory ... found: /boot/grub
Searching for default file ... found: /boot/grub/default
Testing for an existing GRUB menu.lst file ...
found: /boot/grub/menu.lst
Searching for splash image ... none found, skipping ...
Found kernel: /boot/vmlinuz-2.6.31-rc6-945test799-gite3b2415
Found kernel: /boot/vmlinuz-2.6.31-rc6-945test300-gitebb177d
Found kernel: /boot/vmlinuz-2.6.28-17-generic
Found kernel: /boot/vmlinuz-2.6.28-11-generic
Found kernel: /boot/memtest86+.bin
Replacing config file /var/run/grub/menu.lst with new version
Updating /boot/grub/menu.lst ... done

The link /vmlinuz.old is a damaged link
Removing symbolic link vmlinuz.old
Unless you used the optional flag in lilo,
 you may need to re-run your boot loader[lilo]
The link /initrd.img.old is a damaged link
Removing symbolic link initrd.img.old
Unless you used the optional flag in lilo,
 you may need to re-run your boot loader[lilo]
Purging configuration files for
linux-image-2.6.31-rc6-945test499-git6a719e0 ...
Running postrm hook script /sbin/update-grub.
Searching for GRUB installation directory ... found: /boot/grub
Searching for default file ... found: /boot/grub/default
Testing for an existing GRUB menu.lst file ...
found: /boot/grub/menu.lst
Searching for splash image ... none found, skipping ...
Found kernel: /boot/vmlinuz-2.6.31-rc6-945test799-gite3b2415
Found kernel: /boot/vmlinuz-2.6.31-rc6-945test300-gitebb177d
Found kernel: /boot/vmlinuz-2.6.28-17-generic
Found kernel: /boot/vmlinuz-2.6.28-11-generic
Found kernel: /boot/memtest86+.bin
Updating /boot/grub/menu.lst ... done

daves111@presario:~$

It seems that there is something wrong with the kernel package or the
installation of them. Have you tried the suggested workaround which is
to install it again (with dpkg -i) before removing it? If you are
unsure which kernel is running at any point, the command `uname -a`
will tell you which one it is.

Since the 799 kernel did not freeze, it is most likely the commit
after that which triggers the freezes. I'll get an 800 kernel built
to verify that.

800 kernel is now available.

I downloaded and installed 800 ... and it did not freeze. Dave

On Mon, 2010-01-11 at 22:49 +0000, Geir Ove Myhr wrote:
> 800 kernel is now available.
>

I used the 800 for a while and noticed that in firefox, my screen kept
dimming and there was more wait time while loading new pages. seems like
the 800 isn't working as well.

On Mon, 2010-01-11 at 22:49 +0000, Geir Ove Myhr wrote:
> 800 kernel is now available.
>

I'm puzzled... There doesn't seem to be any relevant changes between 800 and 2.6.31-rc7. I have started a build of 2.6.31-rc7 in my environment (it will be 999) to make sure. I'm beginning to wonder if we're on a wild goose chase here.

I suppose I could try the 3 or 4 kernels that went from working to freezing.

----- Original Message -----
From: "Geir Ove Myhr" <email address hidden>
To: <email address hidden>
Sent: Thursday, January 14, 2010 3:31 PM
Subject: [Bug 500686] Re: [i945g] freezes shortly after loading desktop

I'm puzzled... There doesn't seem to be any relevant changes between 800
and 2.6.31-rc7. I have started a build of 2.6.31-rc7 in my environment
(it will be 999) to make sure. I'm beginning to wonder if we're on a
wild goose chase here.

--
[i945g] freezes shortly after loading desktop
https://bugs.launchpad.net/bugs/500686
You received this bug notification because you are a direct subscriber
of the bug.