Disable touchpad tapping in HP G60 on Linux

I scoured all over the internet for a good hour and a half before finding the solution. To disable touchpad tapping for the G60, you need to: (1) enable synaptics from /etc/X11/xorg.conf, and (2) use a very specific, hard-coded device option to point to your touchpad device.

Here is the relevant portion of my xorg.conf for the HP G60 (G60t-200, to be exact — at least according to the sticker on the bottom; NOTE: I’m using the latest Xorg 1.6, but that shouldn’t really make any difference for our purposes here):

Section "ServerLayout"
    Identifier     "X.org Configured"
    Screen      0  "Screen0" 0 0
    InputDevice    "Mouse0" "CorePointer"
    InputDevice    "Keyboard0" "CoreKeyboard"
    InputDevice     "SynapticsTouchpad" "SendCoreEvents"
EndSection

Section "Module"
    Load  "glx"
    Load  "dri"
    Load  "dbe"
    Load  "xtrap"
    Load  "extmod"
    Load  "freetype"
    Load  "synaptics"
EndSection

Section "InputDevice"
   Identifier  "SynapticsTouchpad"
    Driver      "synaptics"
    Option        "SendCoreEvents" "true"
    Option        "Device" "/dev/input/by-path/platform-i8042-serio-2-mouse"
    Option        "Protocol" "auto-dev"
    Option        "HorizEdgeScroll" "0"
    Option      "SHMConfig" "true"
EndSection

Notice the italicized lines. The real key here, from what I can tell (I tried copying and pasting many different configs from various forums), is the line

    Option        "Device" "/dev/input/by-path/platform-i8042-serio-2-mouse"

I first got the idea to add the above weird-looking line from a Gentoo wiki page on another HP laptop model. I bet the above format will work for any recent HP laptop, not just the G60. The only difference between the G60 and the Pavilion DV5000 is that the DV5000 uses “…serio-4-mouse”, whereas the G60 uses “…serio-2-mouse.” To figure out this critical “serio” number, use the following command:

less -eX /proc/bus/input/devices

The output for my G60 looked like this (other irrelevant devices omitted):

I: Bus=0011 Vendor=0002 Product=0007 Version=01b1
N: Name="SynPS/2 Synaptics TouchPad"
P: Phys=isa0060/serio2/input0
S: Sysfs=/devices/platform/i8042/serio2/input/input10
U: Uniq=
H: Handlers=mouse1 event10
B: EV=b
B: KEY=420 70000 0 0 0 0
B: ABS=11000003

Notice the italicized line. It says “serio2”. Anyway, the minimal xorg.conf settings work really well — senstivity and acceleration are handled quite nicely by default. Also, tapping is NOT enabled by default, so that’s why my settings do not say anything about it.’

NOTE: The touchpad by default works very well without synaptics (vertical scrolling and all) using the Linux Kernel 2.6.28 and the newer 2.6.29, with Xorg 1.5.3 and 1.6 — however, tapping is enabled by default. The above solution is the only way to disable (the rather annoying and extremely sensitive) touchpad tapping for the HP G60.

Sources:

http://en.gentoo-wiki.com/wiki/Hp_Pavilion_DV5000

http://www.linuxquestions.org/questions/slackware-14/synaptics-touchpad-in-slackware-12.2-on-hp-g60-702809/ (post#4)

Q6600, Zalman CNPS 9500, and eVGA 680i

My PC has a Q6600 processor overclocked to 2.8 GHz (312.5 FSB, or 1250 as it shows up in the BIOS) on a eVGA 680i TR motherboard. Although the 680i was quite excellent for overclocking the e6600, it isn’t so nice with the Q6600. I have the Q6600 paired up with the Zalman 9500 cooler from my e6600 days.

For quite some time, I tried overclocking my Q6600 beyond 2.8 GHz, without any success. I tried voltages on the RAM, other motherboard features, RAM timings, etc. Today, I installed Coretemp 0.99 and realized that my cooler was the bottleneck all along. If I run Prime 95, the core temperatures on all four cores (with a few degrees difference) reach the high 60’s! No wonder I couldn’t overclock beyond 2.8 GHz. And this is with the Q6600 on 1.2 volts in BIOS (CPU-Z says 1.184 volts). It idles at 45, 40, 40, and 44 degrees for each core respectively.

The moral of the story: make sure you keep a close eye on the CPU temperature when overclocking!

EVGA 8800 GTS + EVGA 680i Issues Resolved

My system has 1 SATA drive and 1 IDE drive. The SATA holds my Windows XP, along with several other partitions for pure data purposes. My IDE holds another partition for storage purposes, but also has my Mint installation. (Yes, I dual-boot. It’s the best of both worlds, with games on XP and real work on my solid Linux environment.) I had some random crashings (no blue screens), and some strange behavior. Here were the symptoms of my system:

1. BIOS would not detect my SATA drive, and would change the HDD (hard drive) boot priority from my IDE (which holds the GRUB bootloader, I think) to my SATA, thus making me automatically go into Windows without loading GRUB a choice. (The temporary solution was to change the HDD priority in BIOS to load my IDE drive first, before my SATA, and this made GRUB show up.)
2. Even after changing the HDD priority back, GRUB would stop detecting my SATA drive, and thus would not boot Windows.
3. Within Linux, sometimes one of the partitions on my SATA drive would not be auto-mounted by Mint like they always were before.
4. I noticed strange characters (data reading corruption) for some of my mp3 files that were on the SATA drive in my Rhythmbox music player.
5. My dual LCD screens would suddenly stop receiving signals from my video card.
6. When I ran Memtest86+ v1.70 from the GRUB menu, it would run for a while, but then all of a sudden, my computer screen would go blank/black!
7. I got error code 26 from my motherboard during POST. (EVGA 122-CK-NF63-TR LGA 775 NVIDIA nForce 680i SLI ATX Intel Motherboard – Retail)
8. Suspecting bad video ram, I checked my video card (EVGA 320-P2-N811-AR GeForce 8800GTS 320MB 320-bit GDDR3 PCI Express x16 HDCP Ready SLI Supported Video Card – Retail) by going into NVIDIA Control Panel in Windows XP just to see if it would crash my system. It did. Clicking on NVIDIA Control Panel immediately crashed my system (blank/black screen, but no automatic reboot).
9. When I booted up Mint once, one of the Screenlets started out with messed up graphics.
10. My system gave a blank/black screen randomly while in both Windows and Mint.

The solution? I changed the power connector to my video card, from the default one that came with the video card, to a 6-pin connector that my PSU came with. The one that came with my card actually had only 5 pins (if you look inside, there is only metal for 5 of the pins, since 1 pin out of the 6 doesn’t have any conductive metal in it), although the NewEgg images on the product denote a 6-pin (PCIe) connection (see the other images in the link)! It now all makes sense, since I had attached both the 5-pin connector and my SATA drive’s power connection with the same power cable. My SATA hard drive was receiving not enough power, which explains the strange data reading corruption I pointed out (and why BIOS failed to recognize it sometimes — since the drive was not starting up at all). It also explains why my screens were going blank – my video card was not getting enough power! It’s strange, because I had not had any problems for several months with my old configuration. Either my PSU is slowly generating less power (it’s a 530 watt power supply), or the old 5-pin connector was bad.

Anyway, I just wanted to share my experience. The lesson? Strange symptoms do not always mean that you need a BIOS update, or your PSU is dying, or your SATA drive is dying, or your motherboard killed your SATA drive, or your video card is dying, or that your RAM is bad. All of these ideas were brought up in the many forums I googled up to see any similar troubles for owners of the same hardware as myself. No one suggested a faulty power connection (or that the power connection on 1 power cable (“rail” as they call it?) had been overloaded with too many connections). So, next time, be sure to check your power cables, and make sure to distribute power as evenly as possible across all the cables! (My system has since been up without a hitch running through graphics benchmarks, rendering tests, and what have you.)