Linux on the ASUS N10 N10J

Would you happen to know if the ExpressGate/SplashTop is coexisting with your Linux? I'd like to install Linux (Arch, specifically) and had desired to wipe the hard drive in order to do so. Now I hear that ExpressGate is by default set in the restore partition, but can also be reinstalled to any other partition with a windows-installer? I know ExpressGate itself is linux based, so it should be able to function on Linux, but I don't see any way to install it, because I hear the N10 doesn't have it installed in some onboard memory, instead of the HDD?

 

Express Gate works fine

Just make sure not to delete its partition... it shouldn't be an issue as long as you work from the end of the hard disk back when resizing the existing partitions.

Edit: Sorry I'm a bit tired and just reread your post, if you blast away the entire disk, Express Gate will get wiped too... so you should resize the partitions instead. There is no way to get express gate back on without doing a restore, which sort of defeats the purpose of wiping the disk.

 

Would you, or anyone else perhaps be able to help a newbie get setup in ArchLinux as well? I have my base system installed, but I'm sure I don't have the same kernal you do (as evidenced by an Alsamixer that detects Intel HDA audio, but puts out no sound).

 

EDIT: There is a possibility that this quick fix might work for you. The drawbacks are that you don't get the goodies that go with my howto and the newer kernel, but it's worth a shot:

http://ubuntuforums.org/showpost.php?p=6174925&postcount=4

 

Updated the howto for 2.6.28-rc8, includes new config. Also changed MONITORING_INTERVAL from 200 to 400, much better wireless reception, less time required to associate with AP as well.

Edit: N10J BIOS rev 209 as of 12/2:

http://dlsvr04.asus.com/pub/ASUS/nb/N10J/N10J209as.zip

Edit 3: THIS IS FOR ENABLING HARDWARE ACCELERATION FOR HD VIDEO!

Mplayer patches for VDPAU:
ftp://download.nvidia.com/XFree86/vdpau/

You will have to build it yourself for now. I will write some directions for the howto on page 1 when I have more time. READ THE README!

If you run into problems compiling this is a useful thread:
http://www.nvnews.net/vbulletin/showthread.php?t=123095

 

http://dlsvr04.asus.com/pub/ASUS/nb/N10J/N10J209as.zip

no longer works error 404

 

Change dlsvr04 to 02 or 03. The servers go down sometimes.

N10J BIOS rev 209:

http://dlsvr02.asus.com/pub/ASUS/nb/N10J/N10J209as.zip

http://dlsvr03.asus.com/pub/ASUS/nb/N10J/N10J209as.zip

Edit: updated howto for 2.6.28-rc9 released 12/19/2008

 

any way we can install the patched mplayer as the default mplayer? It must have to do with symlinks or somthing I guess..

 

You would need a default application setting in GNOME for each video format:

Code:

Running MPlayer:
$ cd mplayer-vdpau
$ ./mplayer -vc <VDPAU-codec-name> -vo vdpau <filename>

'VDPAU-codec-name' can be one of:

            ffmpeg12vdpau
            ffh264vdpau
            ffwmv3vdpau
            ffvc1vdpau

        based on the type of video bitstream (ffmpeg12vdpau for MPEG-1
        or MPEG-2, ffh264vdpau for H.264, ffwmv3vdpau for WMV3, and
        ffvc1vdpau for VC-1)

Default applications can be set by right clicking on a file, selecting properties, clicking the 'Open With' tab, clicking 'Add', and clicking the arrow to expand 'Use a custom command' where you can type the location of the patched MPlayer with the corresponding flags for the file type you are associating. For example using MPEG, if MPlayer was on my desktop, the command would be:

Code:

~/Desktop/mplayer-vdpau/mplayer -vc ffmpeg12vdpau -vo vdpau

EDIT:

An Xmas gift from Linus! 2.6.28 is STABLE!!

Updated howto for 2.6.28 with new config, nvidia 180.18

MERRY CHRISTMAS!

 

still can't seem to get it to compile it always exits with error code 1....prob dependencies....god I hate C/C++ libs (-_-'')!!!

YAY linus no more compiling 10 000 times

 

Yes that is the problem, that's why that thread I linked is useful.. it's a headache getting all the right stuff, and why I haven't had time to write it up yet. There are some (mostly ubuntu) .debs floating around from unofficial sources but I haven't tried them, possibly worth a shot if you feel it's worth the risk.

There isn't a whole lot changed in the stable release from rc9, expect bugfixes soon as wider release of 2.6.28 happens

 

Thanks for the tips, were really helpful.
I managed to get both intel and bvidia cards with full opengl using this script:

Code:

aifiltr0@sylwer:~$ cat /etc/double_xorg/vid_switch
#!/bin/bash                                       

VIDEO=`/usr/bin/lspci|grep -c nVidia`
if [ "$VIDEO" = 1 ]; then
echo "Using nVidia video."
ln -sf /etc/double_xorg/xorg.nvidia /etc/X11/xorg.conf
ln -sf /usr/lib/xorg/modules/extensions/libglx.so.180.18 /usr/lib/xorg/modules/extensions/libglx.so
ln -sf /usr/lib/libGL.so.180.18 /usr/lib/libGL.so.1
else
echo "Utilising Intel video."
ln -sf /usr/lib/xorg/modules/extensions/libglx.so.intel /usr/lib/xorg/modules/extensions/libglx.so
ln -sf /usr/lib/libGL.so.1.2 /usr/lib/libGL.so.1
ln -sf /etc/double_xorg/xorg.intel /etc/X11/xorg.conf
fi
aifiltr0@sylwer:~$

EDIT: If you have installed the nvidia drivers before backing up libglx.so to libglx.so.intel you have to reinstall xserver-xorg-core and libgl1-mesa. (the package names are that in debian). Now everything works nice.

Sound was not working until I added

Code:

options snd-hda-intel position_fix=1 model=3stack-6ch-dig

to /etc/modprobe.d/alsa-base

The wireless however required me to compile atheros madwifi from source, but after that it worked nicely.

Still no idea how to get any keycodes with xev from screen magnification button, and upon switching video cards & bluetooth (I guess I'll have to dig the asus_laptop source. The wireless switch on the side of my note switches bluetooth on and off (not wifi), and the fn+f2 does nothing.

 

Ok, another couple of fixes. The webcam on my notebook was working out of the box, but the image was upside down. The fix was found as a couple of patches to the source of uvc-linux here: http://ubuntuforums.org/showthread.php?t=838210
The only thing that is different - the patches didn't work for me and the svn sources from berlios are outdated. Therefore here is the updated procedure:

Code:

sudo -s
mkdir /temp
cd /temp
apt-get install mercurial
hg clone http://linuxtv.org/hg/~pinchartl/uvcvideo/
cd trunk/uvcvideo/linux/drivers/media/video/uvc

Now place the following patch as say video_fix.patch

Code:

--- uvc_video.c.orig    2008-12-28 17:02:02.000000000 +0300                                              
+++ uvc_video.c 2008-12-28 17:21:24.000000000 +0300                                                      
@@ -421,6 +421,10 @@                                                                                     
        struct uvc_video_queue *queue = &video->queue;                                                   
        unsigned int maxlen, nbytes;                                                                     
        void *mem;                                                                                       
+      /* Patch variables */                                                                             
+    __u8 *mem_tmp, *ptr_tmp;                                                                            
+    int i, k, pixel_size;                                                                               
+                                                                                                        
                                                                                                         
        if (len <= 0)                                                                                    
                return;                                                                                  
@@ -431,6 +435,60 @@                                                                                     
        nbytes = min((unsigned int)len, maxlen);                                                         
        memcpy(mem, data, nbytes);                                                                       
        buf->buf.bytesused += nbytes;                                                                    
+    /* Have the last copied bytes completed the current frame? */                                       
+    if (nbytes == maxlen) {                                                                             
+        /* Area where to save the original frame before manipulation. */                                
+        mem_tmp = (__u8 *) kmalloc(buf->buf.bytesused / 2, GFP_ATOMIC);                                 
+        if (mem_tmp != NULL ) {                                                                         
+            /* Copy the original frame in a temporary buffer. */                                        
+            memcpy(mem_tmp, queue->mem + buf->buf.m.offset,                                             
+                   buf->buf.bytesused / 2);                                                             
+            /* "pixel_size" depens on the pixel color depth (bpp),                                      
+             * but in YUY2 image format is constant and equal to 2.                                     
+             */                                                                                         
+             pixel_size = video->streaming->format->bpp / 8;                                            
+            /* The following loop copy 2 pixels at a time (4 bytes                                      
+             * in YUY2 format) from the last two stored in "mem"                                        
+             * (and going back) to the first two (and going on)                                         
+             * stored in "mem" itself following a sort of YUY2                                          
+             * algorithm.                                                                               
+             */                                                                                         
+            ptr_tmp = queue->mem + buf->buf.m.offset                                                    
+                  + buf->buf.bytesused / 2;                                                             
+            /* When the top-half of the frame has been reversed,                                        
+             * rows are copied from the last one stored in "mem_tmp"                                    
+             * (and going up) into the bottom half part of "mem"                                        
+             * buffer.                                                                                  
+             */                                                                                         
+            for (i = 0, k = buf->buf.bytesused / 2 - 2 * pixel_size;                                    
+                 i < buf->buf.bytesused;                                                                
+                 i += 2 * pixel_size, k -= 2 * pixel_size){                                             
+                /* If the top-half of the frame has been                                                
+                 * revesed, then it is needed to split the                                              
+                 * source buffer from "mem" to "mem_tmp".                                               
+                 */                                                                                     
+                if (i == buf->buf.bytesused / 2) {                                                      
+                    ptr_tmp = mem_tmp;                                                                  
+                    k = buf->buf.bytesused / 2                                                          
+                        - 2 * pixel_size;                                                               
+                }                                                                                       
+                 /* The order of copied bytes is changed from                                           
+                  * (Y0 U0 Y1 V1) to (Y1 U0 Y0 V1), i.e. from                                           
+                  * (#0 #1 #2 #3) to (#2 #1 #0 #3).                                                     
+                  */                                                                                    
+                 ((__u8 *)(queue->mem+buf->buf.m.offset + i))[0] =                                      
+                 ((__u8 *)(ptr_tmp + k))[2];                                                            
+                 ((__u8 *)(queue->mem+buf->buf.m.offset + i))[1] =                                      
+                 ((__u8 *)(ptr_tmp + k))[1];
+                 ((__u8 *)(queue->mem+buf->buf.m.offset + i))[2] =
+                 ((__u8 *)(ptr_tmp + k))[0];
+                 ((__u8 *)(queue->mem+buf->buf.m.offset + i))[3] =
+                 ((__u8 *)(ptr_tmp + k))[3];
+            }
+            /* For this frame "mem_tmp" is not needed any more. */
+            kfree(mem_tmp);
+        }
+    }

        /* Complete the current frame if the buffer size was exceeded. */
        if (len > maxlen) {

This is the patch from ubuntuforums manually applied and recreated (no idea why it didn't work in it's initial form)

now

Code:

patch -p0 < video_fix.patch
cd /temp/trunk/uvcvideo
make all
make install
modprobe -r uvcvideo
modprobe uvcvideo

That's it. The image from the webcam is now as it should be. While modprobing I got:

Code:

aifiltr0@sylwer:/temp/trunk/uvcvideo$ sudo modprobe uvcvideo
WARNING: Error inserting v4l2_compat_ioctl32 (/lib/modules/2.6.28-sylwer1/kernel/drivers/media/video/v4l2-compat-ioctl32.ko): Invalid module format
aifiltr0@sylwer:/temp/trunk/uvcvideo$

However, that didn't seem to affect anything. The webcam works as it should.

 

config-2.6.28 in first post have

But N10j have Atheros Card and WiFi doesn't work if i build kernel with selected config
It's bug?

 

ath5k is for the earlier Atheros cards, the module is ath9k. Maybe try a modprobe ath9k and then try to use the wireless.. also make sure you deleted those two shell scripts /etc/acpi/asus-wireless*

Edit: Also try: iwconfig wlan0
If that works then at least you know the module is loaded and the wireless card is functional.

If that doesn't work, try modprobe ath9k
Then repeat iwconfig wlan0

If modprobe returns errors something maybe went wrong with kernel/headers build.

In that case try lspci | grep Atheros

To make sure your card is present

 

I don't know why I didn't think of this earlier but those getting compile errors due to dependencies can just run:

and it will install all the dependencies to build the program from the source code

 

Okay, and, hopefully the last bunch of fixes. The thing that cauht my attention was suspend and resume. Both suspend and hibernate work out of the box in 2.6.28 without any need of tuxonice (Although tuxonice has the compression feature and is swifter hence, I'm too lazy to try).
The problem is, that we need some thing to be done upon suspend and resume. Wifi seems to have troubles upon resume, until we unload and load
again ath_pci. As well as we need to reload asus_laptop. Here is the fix o automate that stuff. Create as root the following file
/etc/pm/sleep.d/20asus
(you may call it whatever you like, numbers indicate the execution order)

Code:

#!/bin/bash                                                   
case $1 in                                                    
    hibernate)                                                
    modprobe -r ath_pci                                       
    modprobe -r asus_laptop                                   
        ;;                                                    
    suspend)                                                  
    modprobe -r ath_pci                                       
    modprobe -r asus_laptop                                   
    ;;                                                        
    thaw)
    modprobe ath_pci
    modprobe asus_laptop
        ;;
    resume)
    modprobe ath_pci
    modprobe asus_laptop
    ;;
    *)  echo "??"
        ;;
esac

Now chmod +x 20asus and you are done. Double check that this file is owned by root:root, otherwise it won't work.
The only problems now remaining are:
1. wifi and bt led are always on upon asus_laptop load.
2. No idea how to get scancodes from bt&videocard switches magnify, splashtop buttons and some fn+... keycombos.
3. Some redraw issues and minor glitches using nvidia in kde4 (e.g. in yakuake) (while on the desktop, same debian and kde, nv7600GT everything is fine)
4. wifi seems always on.

P.S. From now on I will be filling the info as well into the linlap.com wiki.

 

Thanks for posting your fixes! I linked to your webcam patch from the howto.

I get some redraw issues too.

Wireless is not controlled with Fn + F2 or the switch, they both seem to control bluetooth for some reason.

 

Playing around fn+f1 gave some severe oopses once or twice ended up in a suspend. Resuming then revealed X were restarted.
fn+f1 produces keycode 223, currently with NoSymbol as xev reported, However it DOES lock the screen in kde4, but I'm not sure if it triggers the suspend. Sometimes suspend can happen randomly some time after keypress. I'm wondering what is actually processing that keycode...
After pressing FN+f2 once or twice fn+ combos are locked for some time, or until asus_laptop reload. Yet no actions seem to be performed. I'll be looking into asus_laptop source some day later, when I have the time.

Brightness regulation and screen disable keys work, but produce no keycode. fn+f8 == no action or keycode.

P.S. I summed up the thread at linlap wiki.
P.P.S. And i've just fixed the patch to webcam above.. Just mixed the order of files while creating it...

 

Redraw on nvidia seems to be pure compiz issue. I haven't seen it on kwin with effects enabled. Playing around compiz options and settings and adding force_independent_output_painting produced less glitches (nearly no of them). I also commented out all the extra options except NoLogo in xorg.conf and let the drivver to autodetect where the card is.

 

It is a compiz issue. I still get redraw problems with 'force independent output painting' set under general options in ccsm, especially in terminal windows.

In xorg.conf adding this under the "Device" section helps:

Code:

Option          "AllowSHMPixmaps"               "0"

and commenting out the line:

Code:

#        Option          "BackingStore"                  "True"

I also added this line to the bottom of my /etc/modprobe.d/nvidia-kernel-nkc

Code:

options nvidia NVreg_Mobile=1 NVreg_RegistryDwords="PerfLevelSrc=0x2222"

There is a patch, but there is still an unsolved race condition which is at the heart of the problem:
http://www.nvnews.net/vbulletin/showthread.php?t=120829&highlight=redraw&page=2#18

More info here:
http://www.nvnews.net/vbulletin/showthread.php?t=104822&page=7#94

The following packages are required for it to compile:

Code:

apt-get install compiz-dev libx11-xcb1 libx11-xcb-dev libglib2.0-dev libgconf2-dev

Source tarball is here:
http://freedesktop.org/wiki/Software/Compiz#head-bf8cf8f5418f2e6371e66c7153bf95849532e362

Also need to recompile any compiz plugins because in the sid repository only 0.7.6 is available.

 

Hey All I decided to try out debian on my other laptop
I have to install the iwlagn driver but I don't get how...

I've fallowed the instructions here:
http://wiki.debian.org/iwlagn

but when I get to iwconfig
it doesn't find the wlan0

I was just wondering because you seem to know a lot about debian

 

What kinda wireless card do you have? Or what's the make/model of the laptop?

 

Intel PRO/Wireless 5300 AGN
It's a Dell latitude E6500

 

Ah, ok. Try a 'modprobe iwlagn'

If that doesn't work, you may need to build a custom kernel with the iwlagn module. You could also just build the module itself. I think as of 2.6.27 iwlagn was rolled in so 2.6.28 stable should be fine. Just make sure you select the module in 'make menuconfig'

If you have any further problems you should start a new thread so we don't clutter this up with non-N10 related stuff. Just PM me also if it's something small. Good luck

 

yup when I did modprobe iwlagn it said module not found...and I know it's included in 2.6.27+.... I was already compiling 2.6.28 with your guide

Oh well thanks anyways ^_^

 

Well, some glitches are still around, so I guess the only choice here is to wait till damage, compiz and other stuff will be fixed. I'm too lazy to recompile compiz anyway, since I mostly use it on intel video, which works just as it should.

 

Hmm... Looking around nvidia I've spotted an interesting thing: nvidia-settings reports 512mb of video memory, against 256 as written on a sticker. Is it a bug of the driers, or am I lucky? (haven't looked into it цршду on necro$oft window$, since wiping it was one of the first thingы I've done)

 

I'm not positive, but I think the extra 256 is TurboCache (shared system memory). In any event using 180.18 on my ASUS G50V with compiz results in the same problems, so it's not N10J-specific.

 

Hmm.. I guess it may be so... Anyway, the performance of 2d rendering is not quite satisfactory on both cards. (btw. it is better on intel card!). I'm currently looking for solution towards intel's GEM that is said to improve that. Since I mostly use intel card and nvidia just for final testing my self-written 3d apps when they grow quite complex, I'm looking into any ways of fixing that. Has anybody tried GEM on n10j hardware? Is it worth going slackware-way and compile that?

 

GEM is rolled into 2.6.28 I'm not sure if it supports GMA950 though.

 

I have search on this thread but i didn't find info about vga out. Does vga out work with ubuntu?

Thanks

 

But you still have to update intel xorg drivers as well as mesagl. The have been a thread somewhere at ubuntu forums containing a script for building svn version of that stuff.

 

Yes, external VGA, HDMI and HDMI to DVI work with the nvidia driver. It may autodetect your display if it is plugged in before X starts. You can configure it manually, run nvidia-settings as root. Make sure to click 'Save to X Configuration File' before you exit. It is at the bottom so you may have to move the window by right clicking on its window list (taskbar) icon and selecting move since it exceeds the screen height.

This should do it if you want the latest:
http://intellinuxgraphics.org/download.html

Code:

git clone git://anongit.freedesktop.org/git/xorg/driver/xf86-video-intel
git clone git://anongit.freedesktop.org/git/mesa/mesa

etc..

 

I've just found that webcam fails to recover from hybernation. and modprobe -r unvcvideo && modprobe uvcvideo doesn't help there. Ill be playing around it soon, perhaps patching kernel with tuxonice will help.

 

Ok, Here goes the manual of getting tuxonice on your kernel:
why do we need this?
1). MUCH faster hibernation/thawing due to compression.
2). No problems with camera stopping working upon hibernation.

Code:

cd /usr/src/linux-2.6.28
wget http://www.tuxonice.net/downloads/all/current-tuxonice-for-2.6.28.patch.bz2
bunzip2 current-tuxonice-for-2.6.28.patch.bz2
patch -p1 < current-tuxonice-for-2.6.28.patch.bz2
make menuconfig

Have a look over some hibernation settings, perhaps add resume partition. As well add some string to your kernel version, etc.
Actually, that is it. Be sure to recompile the atheros & uvcvideo stuff after that.
EDIT: our script in /etc/pm/sleep.d now requires no asus_laptop, since fn keys and stuff seem to work fine, but ath_pci should still be around. uvcvideo also works fine.
The next things I will be playing with are: fbsplash (gensplash) and pohmelfs.

 

Another tip: On current debian EXA acceleration offers terrible 2d performance. Setting AccelMethod to XAA imade a huge performance boost.

 

Well, playing around I seem to have found something.
go to terminal and type acpi_listen
Now press the buttons that are quite dead and produce no scancode. e.g. zoom and others.
You will see that they _actually_ produce acpi events. Wireless enable|disable also produces the event, but after that everything hangs for a time. I guess it IS the glich. Some timeout or whatever.
Does anyonce know if there is any nice way to bind some keycodes to those acpi keys or does it require digging with the code?
While now I'm studying asus_laptop source. I shall post the patches soon here. I will try to do the following:
* make bt be on or off according to the switch upon module load (on by default)
* try to fix wlan button
* tinker with leds, make them crossfade if that is possible or act as an equaliser. (just4fun)
Anyone wiling to help along are welcome. Find me in jabber aifiltr0 [at] invyl (dot )ath (dot) cx

 

Okay. Fix for the missing keys arrived.
Since they trigger acpi events we have to bind keycodes for them.
Off we go.

Code:

sudo -s
cd /etc/acpi/events
nano n10j

copy and paste this into the file:

Code:

# asus n10j extra keys fix
# Written by ANdrew 'Necromant' Andrianov
# /etc/acpi/n10j.sh for further processing.

event=hotkey (ATKD|HOTK) (0000005c|000000ba|0000008a|00000082|0000005c|00000080|00000081|000000b2)
action=/etc/acpi/n10j.sh "%e"

now, cd ..
nano n10j.sh

Code:

#!/bin/bash                       
# asus n10j extra keys handler    
# written by Andrew 'Necromant' Andrianov
KEY=`echo $1|cut -b 13-20`              

test -f /usr/share/acpi-support/key-constants || exit 0
. /usr/share/acpi-support/key-constants               

case $KEY in
0000005c)  
        #splashtop key
        acpi_fakekey $KEY_NEXTSONG
        ;;                       
000000ba)                        
        #zoom                    
        acpi_fakekey $KEY_PREVIOUSSONG
        ;;                           
00000081)                            
        #bt ->off                    
        acpi_fakekey $KEY_F13        
        ;;                           
00000080)                            
        #bt -> on                    
        acpi_fakekey $KEY_F14        
        ;;                           
000000b2)                            
        #video_toggle                
        acpi_fakekey $KEY_MEDIA
        ;;
0000008a)
        #fn+c
        acpi_fakekey $KEY_F14
        ;;
00000082)
        #fn+v
        acpi_fakekey $KEY_F15
        ;;
*)
        echo ?
esac

chmod +x n10j.sh

There you go. just reload acpid (/etc/init.d/acpid reload as root)
and your keys will pop up in xev. Now just assign them in .Xmodmap if anything is missing and you are ready to go.
Here is how my .Xmodmap looks like

Code:

keycode 153 = XF86AudioNext
keycode 144 = XF86AudioPrev
keycode 182 = XF86Search
keycode 183 = XF86Mail
keycode 237 = XF86Go
keycode 174 = XF86AudioLowerVolume
keycode 176 = XF86AudioRaiseVolume
keycode 160 = XF86AudioMute
remove Lock = Caps_Lock
keysym Caps_Lock = XF86WWW

EDIT: This may not work for you. Check for ATKD events using acpi_listen command. and add the to the scripts. The output ATKD values depend on asus_laptop wapf parameter that I tried with no luck to use to enable|disable wifi.

 

180.22 Nvidia drivers seem to fix the redraw issue finally.

 

Uhg.. you had me excited for a minute. I just got a redraw glitch. It doesn't fix the problem.

Also your hotkey script doesn't work for me, but I'm fine with Fn+F2 working for bluetooth instead of wifi (I've gotten used to it). Nice attempt anyway... thanks for the posts

 

It requires a little tuning. Acpi events seem to have different codes from laptop to laptop. Use acpi_listen to figure them out, and then use the values to fix the scripts. Qt doesn't seem to handle all keysyms correctly so consider looking at /usr/share/X11/XKeysymDB and playing around Xmodmap and xev, once you make acpid to generate keycodes for hotkeys and switches. Remember to /etc/init.d/acpid restart after you edit the events. However editing the bash script require no acpid restart.
It takes some time to make them all fully working.
While now I'm studying the asus_laptop source between my exams. I guess I can make up a patch, that will provide pure software wifi and bluetooth switching say through /proc, thus freeing the switch for some more useful actions.
And about nvidia, well I also notice the glitches in terminal after some time, but there were significally less of them. 2d rendering is still worse that on intel (which is not fast as well) but better that on the old drivers, 3d is also a bit faster as well.

 

From the 180.22 release highlights:

That's probably why you thought it went away -- I did too when I set this (see my earlier post about tweaking the x config) but it still comes back every once in a while, worst in the terminal, but I've had full window redraw problems as well. I don't think the race condition is going to be fixed any time soon. It looks pretty dirty.

 

2.6.29rc1 released...

btw has anyone got the expresscard slot to work? I tried running:

with no avail

 

its weird because my card reader works in 2.6.27-11

P.S. WIRELESS ON 2.6.29rc1 IS CRAZY I HAVE 100% when I use to have 56% HIGHLY recommended, you will have to patch 2.6.28 source code and then patch /drivers/net/irda/donauboe.c

I've attached the patched file(patching it other than manually gave me errors)

You will have to remove the 2.6.28 kernel or it will say it has a conflict with a driver thats in 2.6.28

if you have questions I'll be here

 

Haven't tried any expresscards on my N10J... I don't own one so I haven't had the chance. If I get my hands near an expresscard I will try though!

Edit: Uploaded new config to the howto on page 1, no build problems, no patch needed. Wireless *does* seem better but I kept the 400Hz.

 

I wonder if anyone needs my prepatched kernel tree: 2.6.28 featuring fbcondecor (wallpaper for consoles), tuxonice (but still I disabled in .config, because there are some issues with it, patched uvcvideo, patched asus_laptop (bluetooth is now off by default). Shall I upload it somewhere?

 

I'd check out your patched asus_laptop module... that sounds cool

 

That's a really small patch. Open it with your favourite editor arond line 1020
You will see this

Code:

	/* WLED and BLED are on by default */
	write_status(bt_switch_handle, 1, BT_ON);
	write_status(wl_switch_handle, 1, WL_ON);

	/* If the h/w switch is off, we need to check the real status */
	write_status(NULL, read_status(BT_ON), BT_ON);
	write_status(NULL, read_status(WL_ON), WL_ON);

It seems that the hardware switch has a bit of a glitch (c)
So that it's status seems to get read incorrectly and the device remains enabled.
change
write_status(bt_switch_handle, 1, BT_ON);
to
write_status(bt_switch_handle, 0, BT_ON);
to make it off by default. I think that commenting out reading the status upon init is also safe here. Since the switch is done actually by the module I think of remastering it into /proc interface, thus allowing the switch can be used for something else and the actual job can be done from desktop.
btw, I really wonder why there are NULL's in the second part. Just noticed them. I think I'll try fixing them a bit...
Oh, and ALLurGroceries what have you changed in your new .config?

I'll send my whole kernel tree to net and the patchset separately once I get myself to fast internet connection instead of gprs and fix a thing or two.

 

Cool

In the new config all I updated it with the new options for the newer kernel, and stripped out the IrDA stack since James- reported problems compiling with it, and the N10J has no infrared.

You're on a GPRS connect? That is crazy!!