Skip to contentSkip to conversation

2.6-test4 problems: suspend and touchpad

DISCUSSION Nicolae Mihalache 4 participants 12 messages Sep 12, 2003 – Sep 16, 2003

lore
Share & tools
lore
#0 lore raw

Hello!

I have a Acer Travelmate 800 laptop and I'm using SuSE Linux 8.2 on it.
I tried recenly to install linux 2.6 and I observe two main problems:
1. The touchpad(synaptics) does not work. In kernel 2.4/X11 4.3 it works very well both as a generic ps2 mouse or as a synaptics (using X11 driver for synaptics). The kernel 2.6 seems to have included a driver for the synaptics device, it is detected at boot, but it does not work in X (I guess it must be some kind of conflict between X11 driver and kernel driver?).

2. suspend/resume. With version 2.6test2+acpi patch both swsusp and "echo 3 >/proc/acpi/sleep" worked, being able to somehow successfully resume. In version 2.6test4 there is no /proc/acpi/sleep and swsusp hangs somwhere during an IDE call (I can hand-copy the trace if needed).

Any ideea?
mache

#1 lore raw
Quoted 5 lines

2. suspend/resume. With version 2.6test2+acpi patch both swsusp and "echo 3 >/proc/acpi/sleep" worked, being able to somehow successfully resume. In version 2.6test4 there is no /proc/acpi/sleep and swsusp hangs somwhere during an IDE call (I can hand-copy the trace if needed).

Would you please try the latest -mm patch (2.6.0-test5-mm1, I believe) and report your findings?

Thanks,

Pat

#2 lore raw
Quoted 7 lines · Patrick Mochel
Quoted 3 lines

2. suspend/resume. With version 2.6test2+acpi patch both swsusp and "echo 3 >/proc/acpi/sleep" worked, being able to somehow successfully resume. In version 2.6test4 there is no /proc/acpi/sleep and swsusp hangs somwhere during an IDE call (I can hand-copy the trace if needed).

Would you please try the latest -mm patch (2.6.0-test5-mm1, I believe) and report your findings?

Well, the 2.6.0-test5-mm1 does not compile on my system (SuSE 8.2, gcc version 3.3 20030226 (prerelease) ):

mm/slab.c: In function `ptrinfo':
mm/slab.c:2792: warning: comparison between signed and unsigned mm/slab.c:2798: warning: implicit declaration of function dbg_redzone1' mm/slab.c:2798: error: invalid type argument of unary *' mm/slab.c:2798: warning: implicit declaration of function dbg_redzone2' mm/slab.c:2798: error: invalid type argument of unary *' mm/slab.c:2801: warning: implicit declaration of function dbg_userword' mm/slab.c:2801: error: invalid type argument of unary *' make[1]: *** [mm/slab.o] Error 1

thanks,
mache

#3 lore raw
Quoted 10 lines · Nicolae Mihalache
Quoted 8 lines · Patrick Mochel
Quoted 6 lines

2. suspend/resume. With version 2.6test2+acpi patch both swsusp and "echo 3 >/proc/acpi/sleep" worked, being able to somehow successfully resume. In version 2.6test4 there is no /proc/acpi/sleep and swsusp hangs somwhere during an IDE call (I can hand-copy the trace if needed).

Quoted 6 lines
Quoted 3 lines

Would you please try the latest -mm patch (2.6.0-test5-mm1, I believe) and report your findings?

Well, the 2.6.0-test5-mm1 does not compile on my system (SuSE 8.2, gcc version 3.3 20030226 (prerelease) ):

Ok, I solved the compilation problems and with this kernel swsusp does not hang anymore. The resume works however the network adapter (Broadcom 4400) does not even when restarting the network. ifconfig eth0 shows very big counters:
eth0 Link encap:Ethernet HWaddr 00:C0:9F:26:C7:15
UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1
RX packets:819 errors:4294966560 dropped:0 overruns:0
frame:4294966836

         TX packets:865 errors:4294966836 dropped:0 overruns:0 
carrier:4294967118
         collisions:4294967204 txqueuelen:100
         RX bytes:956732 (934.3 Kb)  TX bytes:89228 (87.1 Kb)
         Interrupt:5

Any ideas? Maybe the driver for this network card does not (correctly) implement suspend/resume ? mache

#11 lore raw
In reply to Nicolae Mihalache
Quoted 8 lines · Nicolae Mihalache

I have a Acer Travelmate 800 laptop and I'm using SuSE Linux 8.2 on it. I tried recenly to install linux 2.6 and I observe two main problems:
1. The touchpad(synaptics) does not work. In kernel 2.4/X11 4.3 it works very well both as a generic ps2 mouse or as a synaptics (using X11 driver for synaptics). The kernel 2.6 seems to have included a driver for the synaptics device, it is detected at boot, but it does not work in X (I guess it must be some kind of conflict between X11 driver and kernel driver?).

You need a new version of the X11 driver:

        http://w1.894.telia.com/~u89404340/touchpad/index.html
Signature · 2 lines
Peter Osterlund - [email protected]
http://w1.894.telia.com/~u89404340
#4 lore raw
In reply to Nicolae Mihalache
Quoted 15 lines

The resume works however the network adapter (Broadcom 4400) does not even when restarting the network.
ifconfig eth0 shows very big counters:
eth0 Link encap:Ethernet HWaddr 00:C0:9F:26:C7:15
UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1 RX packets:819 errors:4294966560 dropped:0 overruns:0
frame:4294966836

          TX packets:865 errors:4294966836 dropped:0 overruns:0 
carrier:4294967118
          collisions:4294967204 txqueuelen:100
          RX bytes:956732 (934.3 Kb)  TX bytes:89228 (87.1 Kb)
          Interrupt:5
Quoted 2 lines

Any ideas? Maybe the driver for this network card does not (correctly) implement suspend/resume ?

It does not. Could you please try removing the module before you suspend?

Thanks,

Pat

#5 lore raw
Quoted 3 lines · Patrick Mochel

It does not. Could you please try removing the module before you suspend?

Yes, removing and readding the module does the trick.
Unfortunately I've seen that something else does not work after resume: my USB mouse. But for some reason I can not remove the usbcore module, the kernel says it's in use.

Now, how can I help to solve these problems? Is somebody working to solve these problems or should I try to solve them myself (at least with the Broadcom 4400 driver) ?

Thanks,
mache

#6 lore raw
Quoted 8 lines · Nicolae Mihalache
Quoted 2 lines · Patrick Mochel

It does not. Could you please try removing the module before you suspend?

Yes, removing and readding the module does the trick.
Unfortunately I've seen that something else does not work after resume: my USB mouse. But for some reason I can not remove the usbcore module, the kernel says it's in use.

I see that the mouse is not even powered. It's an optical mouse and the light stays switched off.

mache

#8 lore raw
In reply to Nicolae Mihalache

Hi!

Quoted 11 lines
Quoted 2 lines

It does not. Could you please try removing the module before you suspend?

Yes, removing and readding the module does the trick. Unfortunately I've seen that something else does not work after resume:
my USB mouse.
But for some reason I can not remove the usbcore module, the kernel says it's in use.

Now, how can I help to solve these problems? Is somebody working to solve these problems or should I try to solve them myself (at least with the Broadcom 4400 driver) ?

Network driver should be reasonably easy to do; so please help :-).
Pavel

Signature · 2 lines
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
#7 lore raw
In reply to Nicolae Mihalache

Hi!

Quoted 11 lines
Quoted 8 lines
Quoted 2 lines

It does not. Could you please try removing the module before you suspend?

Yes, removing and readding the module does the trick. Unfortunately I've seen that something else does not work after
resume: my USB mouse.
But for some reason I can not remove the usbcore module, the kernel says it's in use.

I see that the mouse is not even powered. It's an optical mouse and the light stays switched off.

USB suspend/resume does not currently work. It worked for uhci in -test3, but it never worked for ohci.
Pavel

Signature · 2 lines
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
#9 lore raw
In reply to Nicolae Mihalache
Quoted 8 lines

Yes, removing and readding the module does the trick. Unfortunately I've seen that something else does not work after resume:
my USB mouse.

Now, how can I help to solve these problems? Is somebody working to solve these problems or should I try to solve them myself (at least with the Broadcom 4400 driver) ?

AFAIK, there is no one working on either of these issues. I would recommende contacting the maintainers about what it would take to implement suspend/resume. They should hopefully have an idea of how to go about it, if not able to do it themselves.

Thanks,

Pat

#10 lore raw
In reply to Nicolae Mihalache

Hi!

Quoted 34 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line

2. suspend/resume. With version 2.6test2+acpi patch both swsusp

and

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

"echo 3 >/proc/acpi/sleep" worked, being able to somehow successfully resume. In version 2.6test4 there is no /proc/acpi/sleep and swsusp hangs somwhere during an IDE call (I

can

Quoted 9 lines
Quoted 7 lines
Quoted 2 lines

hand-copy the trace if needed).

Would you please try the latest -mm patch (2.6.0-test5-mm1, I believe) and report your findings?

Well, the 2.6.0-test5-mm1 does not compile on my system (SuSE 8.2,

gcc

Quoted 1 line

version 3.3 20030226 (prerelease) ):

Ok, I solved the compilation problems and with this kernel swsusp does not hang anymore.
The resume works however the network adapter (Broadcom 4400) does not even when restarting the network.
ifconfig eth0 shows very big counters:
eth0 Link encap:Ethernet HWaddr 00:C0:9F:26:C7:15
UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1 RX packets:819 errors:4294966560 dropped:0 overruns:0
frame:4294966836

         TX packets:865 errors:4294966836 dropped:0 overruns:0 
carrier:4294967118
         collisions:4294967204 txqueuelen:100
         RX bytes:956732 (934.3 Kb)  TX bytes:89228 (87.1 Kb)
         Interrupt:5
Quoted 2 lines

Any ideas? Maybe the driver for this network card does not (correctly) implement suspend/resume ?

Pretty probable. Look at the driver sources, and if there's no _suspend method, write one. If you ifconfig down before suspend, does it help?

Signature · 2 lines
				Pavel
Written on sharp zaurus, because my Velo1 broke. If you have Velo you don't need...