PS2 mouse/kbd problems

PS2 mouse/kbd problems

[J.A. Magallon]

Hi all...

I got time to track my ps2 problems. I run 2.6.9-rc2-mm[123] (that was
enough).

Results:
- mm1: mouse and kbd work ok, both in console and X
- mm2: mouse works, no kbd. I had to unplug/plug the keyboard to get it
  responding.
- mm3: kbd ok, but ps2 mouse is sluggish.

In latest -rc3-mm2, behavior is like mm3 and above.

If you need more info, I have dmesg's, or still can boot on those kernels.
Perhaps this gives some clue. I will scan the changelog for -mm3.

TIA

--
J.A. Magallon <jamagallon()able!es>     \               Software is like sex:
werewolf!able!es                         \         It's better when it's free
Mandrakelinux release 10.1 (Community) for i586
Linux 2.6.9-rc3-mm2 (gcc 3.4.1 (Mandrakelinux 10.1 3.4.1-4mdk)) #1

Re: PS2 mouse/kbd problems

[Dmitry Torokhov]

On Tuesday 05 October 2004 12:45 pm, J.A. Magallon wrote:
> Hi all...
> 
> I got time to track my ps2 problems. I run 2.6.9-rc2-mm[123] (that was
> enough).
> 
> Results:
> - mm1: mouse and kbd work ok, both in console and X
> - mm2: mouse works, no kbd. I had to unplug/plug the keyboard to get it
>   responding.
> - mm3: kbd ok, but ps2 mouse is sluggish.
> 
> In latest -rc3-mm2, behavior is like mm3 and above.
> 

What about vanilla -rc3 and vanilla -rc3 with bk-input patch applied (if you
have some time of course). Do they exibit the same symptoms as -mm tree?

-- 
Dmitry

Re: PS2 mouse/kbd problems (gremlins?)

[J.A. Magallon]

On 2004.10.06, Dmitry Torokhov wrote:
> On Tuesday 05 October 2004 12:45 pm, J.A. Magallon wrote:
> > Hi all...
> > 
> > I got time to track my ps2 problems. I run 2.6.9-rc2-mm[123] (that was
> > enough).
> > 
> > Results:
> > - mm1: mouse and kbd work ok, both in console and X
> > - mm2: mouse works, no kbd. I had to unplug/plug the keyboard to get it
> >   responding.
> > - mm3: kbd ok, but ps2 mouse is sluggish.
> > 
> > In latest -rc3-mm2, behavior is like mm3 and above.
> > 
> 
> What about vanilla -rc3 and vanilla -rc3 with bk-input patch applied (if you
> have some time of course). Do they exibit the same symptoms as -mm tree?
> 

Both rc3 and rc3-bk.input work. Even rc3-mm2 works, depending on how I boot ;).
This is getting really strange....look:

ll /boot:
lrwxrwxrwx  1 root root      21 2004.10.05 14:16 vmlinuz -> vmlinuz-2.6.9-rc3-mm2
-rw-r--r--  1 root root 1485335 2004.09.16 23:28 vmlinuz-2.6.9-rc2-mm1
-rw-r--r--  1 root root 1500553 2004.09.23 22:50 vmlinuz-2.6.9-rc2-mm2
-rw-r--r--  1 root root 1501867 2004.09.24 16:20 vmlinuz-2.6.9-rc2-mm3
-rw-r--r--  1 root root 1498557 2004.10.01 01:05 vmlinuz-2.6.9-rc2-mm4
-rw-r--r--  1 root root 1448740 2004.10.06 22:49 vmlinuz-2.6.9-rc3
-rw-r--r--  1 root root 1451900 2004.10.06 23:57 vmlinuz-2.6.9-rc3-bk.input
-rw-r--r--  1 root root 1506097 2004.10.04 00:32 vmlinuz-2.6.9-rc3-mm1
-rw-r--r--  1 root root 1505218 2004.10.05 20:12 vmlinuz-2.6.9-rc3-mm2

lilo.conf:
root=/dev/sda1
read-only
default="linux"
append="psmouse.proto=exps"
image=/boot/vmlinuz
    label="linux"
...
image=/boot/vmlinuz-2.6.9-rc3-mm2
    label="linux-2.6.9-rc3-mm2"

If I boot with the default entry, mouse does not work. If I boot with
the specific entry in lilo for rc3-mm2, it works.

dmesg diff:
--- dm-1	2004-10-07 00:22:10.616105982 +0200
+++ dm-2	2004-10-07 00:22:10.616105982 +0200
@@ -45,10 +45,10 @@
 Using ACPI (MADT) for SMP configuration information
 Built 1 zonelists
 Initializing CPU#0
-Kernel command line: BOOT_IMAGE=linux ro root=801 psmouse.proto=exps 3
+Kernel command line: BOOT_IMAGE=linux-2.6.9-rc3-mm2 ro root=801 psmouse.proto=exps 3
 CPU 0 irqstacks, hard=b03d2000 soft=b03ce000
 PID hash table entries: 4096 (order: 12, 65536 bytes)
-Detected 2405.595 MHz processor.
+Detected 2405.625 MHz processor.
 Using tsc for high-res timesource
 Console: colour VGA+ 80x60
 Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
@@ -225,17 +225,17 @@
 md: raid1 personality registered as nr 3
 md: raid5 personality registered as nr 4
 raid5: automatically using best checksumming function: pIII_sse
-   pIII_sse  :  3072.000 MB/sec
-raid5: using function: pIII_sse (3072.000 MB/sec)
+   pIII_sse  :  3044.000 MB/sec
+raid5: using function: pIII_sse (3044.000 MB/sec)
 raid6: int32x1    664 MB/s
 raid6: int32x2    882 MB/s
-raid6: int32x4    613 MB/s
+raid6: int32x4    617 MB/s
 raid6: int32x8    464 MB/s
 raid6: mmxx1     1949 MB/s
-raid6: mmxx2     2441 MB/s
-raid6: sse1x1    1164 MB/s
+raid6: mmxx2     2437 MB/s
+raid6: sse1x1    1160 MB/s
 raid6: sse1x2    2386 MB/s
-raid6: sse2x1    2605 MB/s
+raid6: sse2x1    2601 MB/s
 raid6: sse2x2    2707 MB/s
 raid6: using algorithm sse2x2 (2707 MB/s)
 md: raid6 personality registered as nr 8
@@ -296,14 +296,14 @@
 usb 2-1: new full speed USB device using address 2
 hub 5-0:1.0: USB hub found
 hub 5-0:1.0: 2 ports detected
+drivers/usb/class/usblp.c: usblp0: USB Bidirectional printer dev 2 if 0 alt 0 proto 2 vid 0x03F0 pid 0x3404
+usbcore: registered new driver usblp
+drivers/usb/class/usblp.c: v0.13: USB Printer Device Class driver
 usbcore: registered new driver usbhid
 drivers/usb/input/hid-core.c: v2.0:USB HID core driver
 Initializing USB Mass Storage driver...
 usbcore: registered new driver usb-storage
 USB Mass Storage support registered.
-drivers/usb/class/usblp.c: usblp0: USB Bidirectional printer dev 2 if 0 alt 0 proto 2 vid 0x03F0 pid 0x3404
-usbcore: registered new driver usblp
-drivers/usb/class/usblp.c: v0.13: USB Printer Device Class driver
 EXT3 FS on sda1, internal journal
 Adding 1052248k swap on /dev/sda3.  Priority:-1 extents:1
 Linux agpgart interface v0.100 (c) Dave Jones
@@ -357,5 +357,5 @@
 ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 17 (level, low) -> IRQ 17
 PCI: Setting latency timer of device 0000:00:1f.5 to 64
 AC'97 0 analog subsections not ready
-intel8x0_measure_ac97_clock: measured 49595 usecs
+intel8x0_measure_ac97_clock: measured 50582 usecs
 intel8x0: clocking to 48000

(NOTE: look at the diff in usblp position....)

Somebody understands this ? Are there gremlins in my box ?
Why the init order of usblp is dependent on lilo entry ? It does not depend even
on different builds, as if some link order was broken, it is the same
kernel image accessed with a different name...

???

--
J.A. Magallon <jamagallon()able!es>     \               Software is like sex:
werewolf!able!es                         \         It's better when it's free
Mandrakelinux release 10.1 (Community) for i586
Linux 2.6.9-rc3-mm2 (gcc 3.4.1 (Mandrakelinux 10.1 3.4.1-4mdk)) #2

Re: PS2 mouse/kbd problems (gremlins?)

[Andries Brouwer]

On Wed, Oct 06, 2004 at 10:30:22PM +0000, J.A. Magallon wrote:
> 
> On 2004.10.06, Dmitry Torokhov wrote:
> >On Tuesday 05 October 2004 12:45 pm, J.A. Magallon wrote:
> >> Hi all...
> >> 
> >> I got time to track my ps2 problems. I run 2.6.9-rc2-mm[123] (that was
> >> enough).
> >> 
> >> Results:
> >> - mm1: mouse and kbd work ok, both in console and X
> >> - mm2: mouse works, no kbd. I had to unplug/plug the keyboard to get it
> >>   responding.
> >> - mm3: kbd ok, but ps2 mouse is sluggish.
> >> 
> >> In latest -rc3-mm2, behavior is like mm3 and above.
> >> 
> >
> >What about vanilla -rc3 and vanilla -rc3 with bk-input patch applied (if 
> >you
> >have some time of course). Do they exibit the same symptoms as -mm tree?
> >
> 
> Both rc3 and rc3-bk.input work. Even rc3-mm2 works, depending on how I boot 
> ;).
> This is getting really strange....look:
> 
> lrwxrwxrwx  1 root root      21 2004.10.05 14:16 vmlinuz -> 
> vmlinuz-2.6.9-rc3-mm2
> 
> lilo.conf:
> default="linux"
> append="psmouse.proto=exps"
> image=/boot/vmlinuz
>    label="linux"
> ...
> image=/boot/vmlinuz-2.6.9-rc3-mm2
>    label="linux-2.6.9-rc3-mm2"
> 
> If I boot with the default entry, mouse does not work. If I boot with
> the specific entry in lilo for rc3-mm2, it works.
> 
> dmesg diff:
> -Kernel command line: BOOT_IMAGE=linux ro root=801 psmouse.proto=exps 3
> +Kernel command line: BOOT_IMAGE=linux-2.6.9-rc3-mm2 ro root=801 
> 
> Somebody understands this ? Are there gremlins in my box ?

A week or so ago I had a problem and couldnt see which changeset
caused it. After a binary search it turned out to be the changeset
that changed the (length of the) kernel version string.
A wild pointer was harmless at first, but after shifting everything
by a few bytes it caused crashes.

Probably (hopefully) your problem is something entirely different,
but it is not impossible that kernel behaviour depends on kernel name.

Re: PS2 mouse/kbd problems (gremlins?)

[J.A. Magallon]

On 2004.10.07, Andries Brouwer wrote:
> On Wed, Oct 06, 2004 at 10:30:22PM +0000, J.A. Magallon wrote:
> > 
> > On 2004.10.06, Dmitry Torokhov wrote:
> > >On Tuesday 05 October 2004 12:45 pm, J.A. Magallon wrote:
> > >> Hi all...
> > >> 
> > >> I got time to track my ps2 problems. I run 2.6.9-rc2-mm[123] (that was
> > >> enough).
> > >> 
> > >> Results:
> > >> - mm1: mouse and kbd work ok, both in console and X
> > >> - mm2: mouse works, no kbd. I had to unplug/plug the keyboard to get it
> > >>   responding.
> > >> - mm3: kbd ok, but ps2 mouse is sluggish.
> > >> 
> > >> In latest -rc3-mm2, behavior is like mm3 and above.
> > >> 
> > >
> > >What about vanilla -rc3 and vanilla -rc3 with bk-input patch applied (if 
> > >you
> > >have some time of course). Do they exibit the same symptoms as -mm tree?
> > >
> > 
> > Both rc3 and rc3-bk.input work. Even rc3-mm2 works, depending on how I boot 
> > ;).
> > This is getting really strange....look:
> > 
> > lrwxrwxrwx  1 root root      21 2004.10.05 14:16 vmlinuz -> 
> > vmlinuz-2.6.9-rc3-mm2
> > 
> > lilo.conf:
> > default="linux"
> > append="psmouse.proto=exps"
> > image=/boot/vmlinuz
> >    label="linux"
> > ...
> > image=/boot/vmlinuz-2.6.9-rc3-mm2
> >    label="linux-2.6.9-rc3-mm2"
> > 
> > If I boot with the default entry, mouse does not work. If I boot with
> > the specific entry in lilo for rc3-mm2, it works.
> > 
> > dmesg diff:
> > -Kernel command line: BOOT_IMAGE=linux ro root=801 psmouse.proto=exps 3
> > +Kernel command line: BOOT_IMAGE=linux-2.6.9-rc3-mm2 ro root=801 
> > 
> > Somebody understands this ? Are there gremlins in my box ?
> 
> A week or so ago I had a problem and couldnt see which changeset
> caused it. After a binary search it turned out to be the changeset
> that changed the (length of the) kernel version string.
> A wild pointer was harmless at first, but after shifting everything
> by a few bytes it caused crashes.
> 
> Probably (hopefully) your problem is something entirely different,
> but it is not impossible that kernel behaviour depends on kernel name.

Well, partially solved...

I had USB legacy emulation active in the BIOS.  Disabling it makes
everything work as it should.

I had alwasy thought that Linux ignored the BIOS completely, as an
ancient remain of cr*p.

Thanks everyoune.

--
J.A. Magallon <jamagallon()able!es>     \               Software is like sex:
werewolf!able!es                         \         It's better when it's free
Mandrakelinux release 10.1 (Community) for i586
Linux 2.6.9-rc3-mm3 (gcc 3.4.1 (Mandrakelinux 10.1 3.4.1-4mdk)) #1