acpi breaking synaptics mouse support on Acer Travelmate 260

acpi breaking synaptics mouse support on Acer Travelmate 260

[Santiago Garcia Mantinan]

Hi!

I recently (the 11th) posted on lkml a message with subject "2.6.0test3
problems on Acer TravelMate 260 (ALSA,ACPIvsSynaptics,yenta)" on which I
commented all the major problems I was seing on this machine.

I have not received any replies there about the ACPI problem, so I have done
further testing and I'm posting here hoping this is a better place to give
this info, if this is not the right place please tell me where could I post
this (I'm not on the list but given the description my guess was that this
was the right place).

The machine is a PIII mobile at 1GHz with intel chipset (i82801+i830M) 256
Megs of RAM the mouse is a Synaptics touchpad. The problem is that the mouse
gets crazy (if using the normal mouse driver) or looses sync (if using the
synaptics driver). I have looked at the interrupts happening when one moves
the mouse and the number is much higher (from 3 to 10 times) in 2.6 compared
with 2.4, also when one stops moving the mouse in 2.6 the interrupts keep on
happening for half a second or so, while it stops inmediately in 2.4.

I have tested this in 2.6.0test3 and test2 with and without the latest patch
available on the acpi.sf.net site, the acpi-20030730-2.6.0-test2.diff.gz
patch, I have also tested with and without APIC, having always the same
result. The only way I found to get the mouse working ok was to fully remove
ACPI, removing only some parts of it was of no use.

Well, hope this helps to put a little bit of light on this problems, if you
need more info to track this down, you want this as a bug on the bts, or
whatever, just tell me.

Regards...
-- 
Manty/BestiaTester -> http://manty.net


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01

Re: acpi breaking synaptics mouse support on Acer Travelmate 260

[Nate Lawson]

On Wed, 13 Aug 2003, Santiago Garcia Mantinan wrote:
> I recently (the 11th) posted on lkml a message with subject "2.6.0test3
> problems on Acer TravelMate 260 (ALSA,ACPIvsSynaptics,yenta)" on which I
> commented all the major problems I was seing on this machine.
>
> The machine is a PIII mobile at 1GHz with intel chipset (i82801+i830M) 256
> Megs of RAM the mouse is a Synaptics touchpad. The problem is that the mouse
> gets crazy (if using the normal mouse driver) or looses sync (if using the
> synaptics driver). I have looked at the interrupts happening when one moves
> the mouse and the number is much higher (from 3 to 10 times) in 2.6 compared
> with 2.4, also when one stops moving the mouse in 2.6 the interrupts keep on
> happening for half a second or so, while it stops inmediately in 2.4.

I don't use linux but perhaps there is an interrupt sharing problem that
appears when interrupts are routed differently (as they probably are
between 2.4 and 2.6).

-Nate


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01

Re: acpi breaking synaptics mouse support on Acer Travelmate 260

[Peter Osterlund]

Santiago Garcia Mantinan <manty-gaW6/AuhO2xeoWH0uzbU5w@public.gmane.org> writes:

> The machine is a PIII mobile at 1GHz with intel chipset (i82801+i830M) 256
> Megs of RAM the mouse is a Synaptics touchpad. The problem is that the mouse
> gets crazy (if using the normal mouse driver) or looses sync (if using the
> synaptics driver). I have looked at the interrupts happening when one moves
> the mouse and the number is much higher (from 3 to 10 times) in 2.6 compared
> with 2.4, also when one stops moving the mouse in 2.6 the interrupts keep on
> happening for half a second or so, while it stops inmediately in 2.4.

Actually it is expected to see interrupts for about one second after
you stop touching the touchpad, if you are using the synaptics driver.
There is a feature in the synaptics hardware that keeps generating
packets for one extra second, to make it easier for drivers to
implement gesture decoding without having to rely on additional
software timers. Maybe you didn't use the synaptics driver in 2.4, in
which case the touchpad would emulate a standard mouse and not
generate the extra packets.

-- 
Peter Osterlund - petero2-zq6IREYz3ykAvxtiuMwx3w@public.gmane.org
http://w1.894.telia.com/~u89404340


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01

Re: acpi breaking synaptics mouse support on Acer Travelmate 260

[Santiago Garcia Mantinan]

> Actually it is expected to see interrupts for about one second after
> you stop touching the touchpad, if you are using the synaptics driver.

After reading this I went and rechecked the 2.4 behaviour, in fact the
interrupts seem to work more or less the same on 2.6 and 2.4 producing
aproximately the same number of interrupts and stopping both 1 second after.
You are right, when I reported the bug I had tested 2.4 with the normal
driver and not the synaptics driver for X, sorry, I didn't know the X driver
could change the interrupt behaviour.

Well, to sum up, for what I see looking at /proc/interrupts both 2.4 using
the X synaptics driver, and 2.6 using the synaptics driver, produce
aproximately the same interrupt behaviour, however 2.6 looses sync when acpi
is compiled and works ok when acpi is not compiled :-(

Hope this clarifies the thing a little bit, sorry about the inexactitudes in
the first report.

If you need any other info or want me to make any tests just tell me.

Regards...
-- 
Manty/BestiaTester -> http://manty.net


-------------------------------------------------------
This SF.net email is sponsored by Dice.com.
Did you know that Dice has over 25,000 tech jobs available today? From
careers in IT to Engineering to Tech Sales, Dice has tech jobs from the
best hiring companies. http://www.dice.com/index.epl?rel_code=104

Re: acpi breaking synaptics mouse support on Acer Travelmate 260

[hgfelger-9nAOAgdJVo4b1SvskN2V4Q]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Salut Garcia,
maybe your interrupts are routed differently depending on acpi compiled in
or not.
What does
 cat /proc/interrupts
differ for both setups? What are the respectively acpi-versions?
 cat /proc/acpi/info |grep version
Did you check the /var/log/boot.msg for error-messages, especially related
to ACPI?

On Wed, 20 Aug 2003, Santiago Garcia Mantinan wrote:
> Well, to sum up, for what I see looking at /proc/interrupts both 2.4 using
> the X synaptics driver, and 2.6 using the synaptics driver, produce
> aproximately the same interrupt behaviour, however 2.6 looses sync when acpi
> is compiled and works ok when acpi is not compiled :-(

cheers
  hartwig felger

Hartwig Felger informatics

- -- 
1024D/339FD693 Hartwig Felger <hgfelger-9nAOAgdJVo4b1SvskN2V4Q@public.gmane.org>
Key fingerprint = FB2F 3EE9 345A D55B 6FF2  0EC1 F5B0 684F 339F D693
For the pulic keys, please visit my page.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQE/RIXp9bBoTzOf1pMRAkIGAJ9aSdR8AcAMDi2NfwVBUIWeZmVF8ACg3AtT
DaRAlqB8mSWUuK42dCXs2Qk=
=ha6j
-----END PGP SIGNATURE-----



-------------------------------------------------------
This SF.net email is sponsored by: VM Ware
With VMware you can run multiple operating systems on a single machine.
WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines
at the same time. Free trial click here:http://www.vmware.com/wl/offer/358/0

Re: acpi breaking synaptics mouse support on Acer Travelmate 260

[Santiago Garcia Mantinan]

Hi!

I've been testing today with 2.6.0test4 to see if things were going any
better, the problem is still there but the mouse seems to recover faster of
the problem, which makes me think... could it be that this very same problem
was also happening on 2.4 but I wasn't noticing it because no errors were
appearing on dmesg as the driver is no in the kernel?

> maybe your interrupts are routed differently depending on acpi compiled in
> or not.

Yes, seems so.

> What does
>  cat /proc/interrupts
> differ for both setups?

The only diff I can see on the /proc/interrupts is that acpi, wich was
sharing irq 9 with the sound card and the cardbus adapter is not shown on
no-acpi system and the sound and carbus adapter are on irq 10 now, also eth0
is on irq 5 using acpi and on irq 11 without acpi.

However, if I look at lspci changing from acpi to noacpi I can see that:
vga & usb go from irq 9 to 11
the other usb port goes from 5 to 11
the ide goes from 255 to 11 (there is a message on dmesg about the 255 thing)
SMBus goes from 9 to 10
sound card/modem go from 9 to 10
ethernet goes from 5 to 11
cardbus goes from 9 to 10

Some things like the modem,usb,smbus, ... were not shown on interrupts
because I don't have a driver loaded for them. BTW, ide does appear on
/proc/interrupts at irqs 14 and 15 as usual.

> What are the respectively acpi-versions?
>  cat /proc/acpi/info |grep version

On 2.4.22, which works ok, we have 20030619 and on 2.6.0 all of them fail,
and I have tried up to test4 (20030813).

> Did you check the /var/log/boot.msg for error-messages, especially related
> to ACPI?

I suppose you mean the kernel messages log, if I look at dmesg or the kernel
messages log the only weird thing I can see is the message I said before
about the ide irq:

ACPI: No IRQ known for interrupt pin A of device 0000:00:1f.1 - using IRQ 255

I have tried running with pci=noacpi and the only change was that this
message about the ide irq is not there anymore, the irqs I get assigned this
way are the same I get without acpi, of course, but the problem is still
there. I guess this means that the problem is not the irq assignement.

I don't know what else I can say or try. If you need any more info just ask
for it.

Thanks for your help.

Regards...
-- 
Manty/BestiaTester -> http://manty.net


-------------------------------------------------------
This SF.net email is sponsored by: VM Ware
With VMware you can run multiple operating systems on a single machine.
WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines
at the same time. Free trial click here:http://www.vmware.com/wl/offer/358/0

RE: acpi breaking synaptics mouse support on Acer Travelmate 260

[Yu, Luming]

It's interesting. Please file it in ACPI category on http://bugme.osdl.org 
Thanks,
Luming

-----Original Message-----
From: Santiago Garcia Mantinan [mailto:manty-gaW6/AuhO2xeoWH0uzbU5w@public.gmane.org]
Sent: 2003?8?13? 6:18
To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Cc: petero2-zq6IREYz3ykAvxtiuMwx3w@public.gmane.org
Subject: [ACPI] acpi breaking synaptics mouse support on Acer Travelmate 260


Hi!

I recently (the 11th) posted on lkml a message with subject "2.6.0test3
problems on Acer TravelMate 260 (ALSA,ACPIvsSynaptics,yenta)" on which I
commented all the major problems I was seing on this machine.

I have not received any replies there about the ACPI problem, so I have done
further testing and I'm posting here hoping this is a better place to give
this info, if this is not the right place please tell me where could I post
this (I'm not on the list but given the description my guess was that this
was the right place).

The machine is a PIII mobile at 1GHz with intel chipset (i82801+i830M) 256
Megs of RAM the mouse is a Synaptics touchpad. The problem is that the mouse
gets crazy (if using the normal mouse driver) or looses sync (if using the
synaptics driver). I have looked at the interrupts happening when one moves
the mouse and the number is much higher (from 3 to 10 times) in 2.6 compared
with 2.4, also when one stops moving the mouse in 2.6 the interrupts keep on
happening for half a second or so, while it stops inmediately in 2.4.

I have tested this in 2.6.0test3 and test2 with and without the latest patch
available on the acpi.sf.net site, the acpi-20030730-2.6.0-test2.diff.gz
patch, I have also tested with and without APIC, having always the same
result. The only way I found to get the mouse working ok was to fully remove
ACPI, removing only some parts of it was of no use.

Well, hope this helps to put a little bit of light on this problems, if you
need more info to track this down, you want this as a bug on the bts, or
whatever, just tell me.

Regards...
-- 
Manty/BestiaTester -> http://manty.net


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01
_______________________________________________
Acpi-devel mailing list
Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/acpi-devel


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01

Re: acpi breaking synaptics mouse support on Acer Travelmate 260

[Santiago Garcia Mantinan]

> It's interesting. Please file it in ACPI category on http://bugme.osdl.org

Filed, http://bugme.osdl.org/show_bug.cgi?id=1093

You can contact me for more info or testing whenever you want.

Regards...
-- 
Manty/BestiaTester -> http://manty.net


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01