Hi all,

I currently use the FC6 2.6.18-1.2798 distribution, on a x86 platform

I'd like to use two PCI cards (for details, an I/O card and a RS422 one)
These two cards work fine on an embedded system, based on the same hardware plateform but using a very reduced OS (only kernel 2.6 and minimal necessary file system)

Well, I need to use these cards on this richer system based on the FC6 distribution stated above (to take advantage of a graphical environnement etc..)
Unfortunatly, while rebooting after having plugged any of the cards (or both), the system fails to start up udev.

For now, I do not paste the error messages which appear before the system freezes, because there would be at least two full screens..

I first decided to stop the udev activation while system starts (by commenting
# start_udev in /etc/rc.d/rc/init script)
Then, the system startup fails to lunch the HAL daemon.

So I retry, using this time the interactive mode (-I) to avoid lunching this daemon.
Of course, it is worst... the system fails to lunch the X server because of /dev/mem which does not exist.
Yet, if I go on, without the X server, the two cards seems to (almost) work well with the relevant drivers... there is at least a good point.


Can anybody help me fixing that ?

1 ) More accurately, is it possible to start the X server while the udev service hasnot been lunched ?

2 ) Better, instead of not lunching udev, and then not to create necessary stuffs (such as start start X), is there any setting so that udev allows the system to start without taking care of these PCI cards?
Maybe there is a way to disable new plugged devices detection on the PCI bus ?

By the way, the /etc/udev/rules.d directory contains the following ones:
05-udev-early.rules
40--multipath.rules
51-hotplug.rules
60-libsane.rules
60-net.rules
60-pcmcia.rules
60-wacom.rules
90-alsa.rules
90-hal.rules
95-pam-console.rules

I am afraid there is not any udev settings which can fix the problem because udev rules seems to run on already created device nodes...

Here is the output of lspci if needed:
00:00.0 Host bridge: Silicon Integrated Systems [SiS] 651 Host (rev 02)
00:01.0 PCI bridge: Silicon Integrated Systems [SiS] Virtual PCI-to-PCI bridge (AGP)
00:02.0 ISA bridge: Silicon Integrated Systems [SiS] SiS962 [MuTIOL Media IO] (rev 14)
00:02.5 IDE interface: Silicon Integrated Systems [SiS] Unknown device 5518
00:03.0 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 0f)
00:03.1 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 0f)
00:03.2 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 0f)
00:03.3 USB Controller: Silicon Integrated Systems [SiS] USB 2.0 Controller
00:04.0 Ethernet controller: Silicon Integrated Systems [SiS] SiS900 PCI Fast Ethernet (rev 90)
00:08.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge (transparent mode) (rev 04)
00:09.0 Ethernet controller: Intel Corporation 82540EM Gigabit Ethernet Controller (rev 02)
01:00.0 VGA compatible controller: Silicon Integrated Systems [SiS] 65x/M650/740 PCI/AGP VGA Display Adapter
02:0b.0 Class ffff: Datum Inc. Bancomm-Timing Division Unknown device 4013 (rev 03)

Any idea, xp or piece of advice would be very precious !

Thanks,

Julian

My advice would be to undo any changes you've made and apply maintenance. Since you are running the original kernel, you've probably not applied maintenance so you're now 8 months behind. There have been hundreds (maybe even thousands) of bugs fixed since then.

yum -y update yum* rpm
yum -y update

Actually, I am not really running the original kernel. I recompile FC6 2.6.18-1.2798 sources just to add few features ( KERN_PREEMPT and TIMER_FREQ=1khz...) Sorry for having forgotten to state that point.

I do not have a web access directly on this platefom, do you know a way to maintain the system without using yum?
Maybe download the last version of every rpms and update them one by one ?


Thanks

Whatever distribution you choose, the first step after installation is always to apply maintenance. Good luck.