Bug 3928

Summary:	S3: irda fails - Asus M2400N
Product:	ACPI	Reporter:	Pedro C (pedro)
Component:	Power-Sleep-Wake	Assignee:	Shaohua (shaohua.li)
Status:	REJECTED UNREPRODUCIBLE
Severity:	normal	CC:	acpi-bugzilla, drivers_input-devices
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.9-ac13	Subsystem:
Regression:	---	Bisected commit-id:

Description Pedro C 2004-12-21 17:46:54 UTC

Distribution: Debian Sid
Hardware Environment: Asus M2400N Centrino Laptop with nsc-ircc irda port
Software Environment: 
Problem Description:

Irda stops working after the computer returns from S3 suspend (using ACPI).

Steps to reproduce:

1. Cold boot the computer.
2. Run irdadump. Output is:

01:35:36.097770 xid:cmd 32832722 > ffffffff S=6 s=0 (14)
01:35:36.187656 xid:cmd 32832722 > ffffffff S=6 s=1 (14)
01:35:36.277637 xid:cmd 32832722 > ffffffff S=6 s=2 (14)
01:35:36.367619 xid:cmd 32832722 > ffffffff S=6 s=3 (14)
01:35:36.457604 xid:cmd 32832722 > ffffffff S=6 s=4 (14)
01:35:36.547583 xid:cmd 32832722 > ffffffff S=6 s=5 (14)
01:35:36.637566 xid:cmd 32832722 > ffffffff S=6 s=* dijkstra hint=0400 [
Computer ] (24)

(forever)

3. Suspend
4. Resume
5. Run irdadump. There is no output at all.

This is an nsc-ircc irda chipset (from dmesg):

nsc-ircc, Found chip at base=0x02e
nsc-ircc, driver loaded (Dag Brattli)
IrDA: Registered device irda0
nsc-ircc, Found dongle: HP HSDL-1100/HSDL-2100

Unloading an reloading all the irda modules makes no difference.

Comment 1 Len Brown 2005-01-03 21:13:01 UTC

assigning to Luming b/c I think he's got one of these 2400's.

Comment 2 Pedro C 2005-03-13 16:41:10 UTC

This is still a bug in 2.6.11.

Comment 3 Shaohua 2005-08-25 00:13:05 UTC

Could you please try latest kernel?
If you unload the irda driver before suspend and reload it after resume, 
what's the result?

Comment 4 Pedro C 2005-09-02 09:52:39 UTC

Just tried it with 2.6.13 and unloaded/loaded the driver before/after the
resume. Nothing changes. irdadump still shows nothing after the resume. Maybe
the hardware isn't being correctly reprogrammed.

Comment 5 Len Brown 2007-08-18 14:26:05 UTC

Looks like Dmitry added suspend/resume methods to nsc-ircc.c
for linux-2.6.17.  3b99b93baba4cbf4fd3d206e65e81a070b21b560
Can you try a more recent kernel -- say linux-2.6.22.stable?

I'm a little skeptical that this would solve the problem at hand, however,
since rmmod before suspend and modprobe after resume didn't help --
and presumably that would totally re-initialize the device.

Comment 6 Len Brown 2007-09-06 14:43:14 UTC

please re-open if this is still a problem in linux-2.6.22.stable or later