I have decided to turn over a new leaf by running the kernel with the native video driver, thus not tainting the kernel and hopefully being able to post kernel problem reports, should one be found, though I doubt I will be able to describe the problem properly, but you have to start somewhere. Here is my setup tonight: http://paste.opensuse.org/view/download/6950566
For me and using the nVIDIA video card GTX 275, all is working properly with the nouveau driver even including the new Plymouth startup added to openSUSE 12.2. I can attest the driver also works with a GT 450, but Plymouth does not work with it even as the nouveau driver does for everything else. As normal, I am installing the latest kernel using my SAKC bash script. You can find the latest triad of bash scripts here for your kernel testing pleasure.
With these bash scripts, your six year old daughter could install the latest kernel into openSUSE 12.2. So, what are you waiting for? Avoid future regret and test kernel 3.7 today while it is hot off of the presses.
Well I get to post the very first problem I have run into with kernel 3.7-rc1, but of course, its with a program that requires its own binary file be compiled against the kernel. And the problem child tonight is: VirtualBox 4.2. It fails to compile against the new kernel and so, its off back to kernel 3.6.2 until I find a fix for this issue. Oh well, if its not one thing then it is two I guess.
On 10/14/2012 08:26 PM, jdmcdaniel3 wrote:
>
> Well I get to post the very first problem I have run into with kernel
> 3.7-rc1, but of course, its with a program that requires its own binary
> file be compiled against the kernel. And the problem child tonight is:
> VirtualBox 4.2. It fails to compile against the new kernel and so,
> its off back to kernel 3.6.2 until I find a fix for this issue. Oh
> well, if its not one thing then it is two I guess.
I have built a 3.7-rc1 kernel, but I have not yet booted it. When I do, I will
fix the VB module problem and post a patch.
On 10/15/2012 10:17 AM, Larry Finger wrote:
> On 10/14/2012 08:26 PM, jdmcdaniel3 wrote:
>>
>> Well I get to post the very first problem I have run into with kernel
>> 3.7-rc1, but of course, its with a program that requires its own binary
>> file be compiled against the kernel. And the problem child tonight is:
>> VirtualBox 4.2. It fails to compile against the new kernel and so,
>> its off back to kernel 3.6.2 until I find a fix for this issue. Oh
>> well, if its not one thing then it is two I guess.
>
> I have built a 3.7-rc1 kernel, but I have not yet booted it. When I do, I will
> fix the VB module problem and post a patch.
The fix was pretty easy. Some of the VM_xxxx flags have been modified in 3.7.
The necessary patch is as follows:
So I manually applied this patch and it works. As instructions for our readers, I guess we would/could do the following. Copy the following text in kwrite or gedit:
Index: vboxhost/vboxdrv/r0drv/linux/memobj-r0drv-linux.c
===================================================================
--- vboxhost.orig/vboxdrv/r0drv/linux/memobj-r0drv-linux.c
+++ vboxhost/vboxdrv/r0drv/linux/memobj-r0drv-linux.c
@@ -50,6 +50,10 @@
# define PAGE_READONLY_EXEC PAGE_READONLY
#endif
+#ifndef VM_RESERVED
+#define VM_RESERVED (VM_DONTEXPAND | VM_DONTDUMP)
+#endif
+
/*
* 2.6.29+ kernels don't work with remap_pfn_range() anymore because
* track_pfn_vma_new() is apparently not defined for non-RAM pages.
Save it as the file $HOME/vbox42.patch ($HOME is translated to /home/yourname). Open up a terminal session and execute the following terminal command:
cd /usr/share/virtualbox/src/
patch -p1 -i $HOME/vbox42.patch
If you happen to use dkms to reinstall the VirtualBox driver, the wrong code is loaded into dkms. You can copy over the file with this command:
After making these changes and restarting into kernel 3.7-rc1, I found that VirtualBox 4.2 was again working properly. Thanks very much for your help Larry.
On 10/15/2012 01:26 PM, jdmcdaniel3 wrote:
>
> After making these changes and restarting into kernel 3.7-rc1, I found
> that VirtualBox 4.2 was again working properly. Thanks very much for
> your help Larry.
You are welcome. Thanks for filling in all the details.
I live and die by the details. And its the details that can kill us, as they say. If anyone has tried out the details of the patch and have problems, you got to let us know and we will help you fix them.
Linus Torvalds has issued the first release candidate of Linux 3.7. With this release, Torvalds has, as usual, closed the development cycle’s “merge window”, the phase in which the majority of a new version’s important new features are added to the main development branch. Until the final release of Linux 3.7, which is expected to arrive in December, the developers will now focus on bug fixes, discounting a few potential stragglers as well as smaller, harmless improvements.
Among the most important new features in Linux 3.7 is the addition of support for the 64-bit ARM instruction set. NAT (Network Address Translation) features for IPv6 and the server-side code for “TCP Fast Open” (TFO) are also new; the client-side code for this experimental TCP extension which is designed to speed up HTTP connections was added in Linux 3.6.
Thanks as always for your help with the nVIDIA driver working in kernel 3.7 Larry. Well, I have found tonight that kernel 3.7-rc2 has been released to test. You can find your copy at this link:
I have installed kernel 3.7-rc2 using SAKC with no problems. I am still using the open source nvidia driver and so it works just fine in kernel 3.7-rc2. I do have the VirtualBox binary being loaded using dkms and with the patches provided by Larry, VirtualBox still works just fine then with kernel 3.7-rc2. I have no complaints tonight thought I did have a run in with the local gas company after they pulled my gas meter. I had to make two phone calls to find out they claimed it was leaking, but they left me no word of the deed. I guess they saved my day, except it started out kind of cold, showers too. What will tomorrow bring I wonder. Humm …
i have opensuse 12.2 32bits with PAE support
when i installed the new kernel 3.7-rc2 using SAKC i found that i lost PAE support and now i can see only 3.1GB of my RAM
how can i re-activate it?
On 10/22/2012 06:56 AM, soufsouf wrote:
>
> Hello,
>
> i have opensuse 12.2 32bits with PAE support
> when i installed the new kernel 3.7-rc2 using SAKC i found that i lost
> PAE support and now i can see only 3.1GB of my RAM
> how can i re-activate it?
You need to regenerate the configuration using ‘make menuconfig’ or ‘make
xconfig’. The question concerns the following configuration variables that are
found in the “Processor type and features” section under the “High memory
support” heading:
# CONFIG_NOHIGHMEM is not set
# CONFIG_HIGHMEM4G is not set
CONFIG_HIGHMEM64G=y
If you have more than 4 GB of RAM and need a PAE kernel, then you must have
CONFIG_HIGHMEM64G set, and the others must not be set, as shown above.
Nothing particularly stands out here. Lots of small fixes, exemplified
by the series of memory leak fixes in usb serial drivers. Just a lot
of random stuff…
Most of it is drivers (all over: drm, wireless, staging, usb, sound),
but there’s a few filesystem updates (nfs, btrfs, ext4), arch updates
(arm, x86 and m68k) and just random stuff. Shortlog appended.
And talking about the shortlog: christ people, some of you need to
change your names. I’m used to there being multiple "David"s and
"Peter"s etc, but there are three different Linus’s in just this rc.
People, people, I want to feel like the unique snowflake I am, not
like just another anonymous guy in a crowd.
I’m getting myself a broadsword.
Linus "there can be only one" Torvalds
It has installed just fine and really nothing to report. I continue to use the nouveau Gallium (8.0.4) driver and love the fact I don’t have to reload it each time I update the kernel. I am using the VirtualBox 4.2.4 driver without modifications with the latest kernel. By the way if your first name is Linus and your last is not Torvalds, I would get my first name changed tonight if you don’t want to met that broadsword.
