2000-12-13 17:38:55

by M Sweger

[permalink] [raw]
Subject: linux 2.2.18 inconsistencies/problems (minor)


Hello all/alan,

I've got linux 2.2.18 (the latest) compiled and running on a
Dell 333mhz optiplex GX1. It was compiled using egcs v1.1.2 with
libc6 v2.0.7. Now that USB support is being added, I'm trying to get
it working on this UMSDOS kernel version.

Here is a list of the problems I'm encountering/seen so
far with this kernel version. ITEMS A-D.



A). Problem with the "%d" in usb-serial. Note: I don't have
any devices connected to any USB port, but the controllers
are detected and powered on based on syslog. My kernel is compiled
without the USB disconnect plug-n-play option set - meaning I can't
remove USB devices on the fly.

Here is my /dev/usb/tts subdirectory stuff for my two USB ports
based on the /proc/tty/drivers listing
below which I setup based on the dir path
listed.

crw-r--r-- 1 root root 188, 1 Oct 12 21:22 1
crw-r--r-- 1 root root 188, 0 Oct 12 21:22 0

However /usr/src/linux/Documentation/devices.txt states that
the usb ports for device 188 is /dev/ttyUSB0 and /dev/ttyUSB1

Here is my /dev/ttyUSB0 and 1 listing which I also setup too!

crw-r--r-- 1 root root 188, 3 Oct 12 21:14 ttyUSB3
crw-r--r-- 1 root root 188, 2 Oct 12 21:14 ttyUSB2
crw-r--r-- 1 root root 188, 1 Oct 12 21:14 ttyUSB1
crw-r--r-- 1 root root 188, 0 Oct 12 21:13 ttyUSB0


Here is my /proc listing of the "tty" sub-directory showing
that the kernel thinks a /dev/usb/tts directoy should exist.

In addition, if the subdirectory "tts" is valid, then I'd
expect a /proc/tts/drivers file then.

As a side note, why is my console stuff below listed as "unkown"?

cat /proc/tty/drivers

usb-serial /dev/usb/tts/%d 188 0-254 serial
pty_slave /dev/pts 136 0-255 pty:slave
pty_master /dev/ptm 128 0-255 pty:master
pty_slave /dev/ttyp 3 0-255 pty:slave
pty_master /dev/pty 2 0-255 pty:master
serial /dev/cua 5 64-67 serial:callout
serial /dev/ttyS 4 64-67 serial
/dev/tty0 /dev/tty0 4 0 system:vtmaster
/dev/ptmx /dev/ptmx 5 2 system
/dev/console /dev/console 5 1 system:console
/dev/tty /dev/tty 5 0 system:/dev/tty
unknown /dev/tty 4 1-63 console


Here is my "lsmod" listing.

Module Size Used by
usb-storage 22040 0 (unused)
usbserial 17512 0 (unused)
usb-uhci 19996 0 (unused)
usbcore 48072 0 [usb-storage usbserial usb-uhci]
loop 7648 0 (unused)
dummy 720 0 (unused)
3c59x 20808 1
opl3 11208 0 (unused)
cs4232 2408 0 (unused)
uart401 6032 0 [cs4232]
ad1848 16144 0 [cs4232]
sound 58988 0 [opl3 cs4232 uart401 ad1848]
soundcore 2756 5 [sound]
ppp_deflate 40652 0 (unused)
bsd_comp 3656 0 (unused)
ppp 20748 0 [ppp_deflate bsd_comp]
slip 7316 0 (unused)
slhc 4352 0 [ppp slip]
parport_pc 7284 0
parport 7648 0 [parport_pc]


B). Need to update the devices.txt document in the Documentation
sub-directory. I checked the maintainers site and it is the same
one as delivered with this kernel version. However, it is out-of-date
since there are quite a few devices missing, i.e. cpuid major #
203 and others. I've emailed the author of the document around
October 2000, but haven't seen this document updated for the
linux v2.2.x kernels. I assume the v2.4 devices.txt document is
the same. If so, then the document would need to clarify what
what devices are valid for each kernel (2.0, 2.2, 2.4)
so that devices that don't exist in these older kernel versions
aren't created.

Moreover, the same applies to device name changes, or if they are
put in a different subdir location.


C). Problem cat'g the proc pci bus files.

cat /proc/bus/pci/00/00.0

Get non-printable characters

The same occurs for any file in this sub-directory or in sub-dir "01".



D). Problem not being able to poweroff on the Dell Workstation.
I've tried "shutdown", "halt -p" and even "poweroff" and I'll
get the message "powering off", but nothing happens. I must then
manually turn off the PC. This last worked in linux 2.2.16.
Shortly after this, changes were made to the poweroff logic to
handle powering off with SMP support and since then it has not
worked.

Things I've tried.

a). Compile in SMP although I have a single processor system.
b). Don't compile in SMP support.
c). Compile in PCI BIOS or the "Any" kernel config option.

Other things that are setup or may be special about Dell's.


a). In this machines BIOS the ACPI power option is enabled.
When enabled, and I manually hit the power switch, it's
supposed to go in standby and when the power switch is
hit again, it powers off. Perhaps the "halt -p" must
do a two step process instead of one? Perhaps poweroff
must send two commands instead of just one similar to
this two step process.

When the ACPI option is disabled, then hitting the
power switch will turn the PC off without going into
standby.

b). The BIOS option to power-on the PC via the LAN is disabled
in the PC setup screens.





2000-12-13 18:18:32

by Alan Cox

[permalink] [raw]
Subject: Re: linux 2.2.18 inconsistencies/problems (minor)

> A). Problem with the "%d" in usb-serial. Note: I don't have
> any devices connected to any USB port, but the controllers

Real bug - thanks. I'll squash that in 19pre1

> B). Need to update the devices.txt document in the Documentation
> sub-directory. I checked the maintainers site and it is the same
> one as delivered with this kernel version. However, it is out-of-date
> since there are quite a few devices missing, i.e. cpuid major #
> 203 and others. I've emailed the author of the document around
> October 2000, but haven't seen this document updated for the
> linux v2.2.x kernels. I assume the v2.4 devices.txt document is
> the same. If so, then the document would need to clarify what
> what devices are valid for each kernel (2.0, 2.2, 2.4)
> so that devices that don't exist in these older kernel versions
> aren't created.

I've been talking to HPA already about that

> C). Problem cat'g the proc pci bus files.
>
> cat /proc/bus/pci/00/00.0
>
> Get non-printable characters

Not a bug. That is the raw pci configuration data

> When the ACPI option is disabled, then hitting the
> power switch will turn the PC off without going into
> standby.

We dont support ACPI in 2.2 just 2.4test

> b). The BIOS option to power-on the PC via the LAN is disabled
> in the PC setup screens.

Probably your ethernet card doesnt support it.

Can you report the APM problem to the APM maintainer ?

2000-12-13 19:56:36

by Michael Rothwell

[permalink] [raw]
Subject: Re: linux 2.2.18 inconsistencies/problems (minor)

Alan Cox wrote:
> Real bug - thanks. I'll squash that in 19pre1

When will you be starting 19pre1?