#234 closed defect (unknown)
ACPI, Intel 945GCNL
Reported by: | blondeguy | Owned by: | pasha |
---|---|---|---|
Priority: | major | Milestone: | Release version 3.08 |
Component: | ACPI PSD | Version: | 3.08 |
Keywords: | Cc: |
Description
ACPI.PSD 3.08 3/28/2008
no APM
Intel 945GCNL
Computer gets trap at boot time right after boot logo. Without ACPI.PSD, computer boots normally, but sound loops for either of UNIAUD or C-MEDIA audio drivers. There are no switches used in psd=acpi.psd line in config.sys.
I removed os2logo, used debug version of ACPI.PSD, and computer stops after two logging lines. Digital photos of the screen with and without debug version included
This computer is mine and I will be available for more testing. Let me know how I can help.
Attachments (9)
Change History (42)
by , 17 years ago
by , 17 years ago
comment:1 by , 17 years ago
comment:2 by , 17 years ago
e-co
disable unnecessary devices (audio adapter, ..)
check the documentation, do you follow the installation instructions?
comment:4 by , 17 years ago
e-co, even the eCS 2.0 rc4 CD will not boot (no audio driver) I tried to follow installation instructions, but I am an imperfect user, so who really knows?
pasha, system boots fine with /SAP:0x20 (but if audio adapter present, it still has the looping sound)
by , 17 years ago
Attachment: | acpi_with_uniaud.log added |
---|
comment:7 by , 17 years ago
I have attached acpi.log. I tried with and without the UniAud drivers and the only difference in the log is that without UniAud, the last four lines of acpi.log are not there.
e-co, the motherboard has HD audio, so I started with a PCI sound card, but I have now removed that card to make testing as simple as possible.
comment:8 by , 17 years ago
With ACPI13042008.ZIP components ACPI.PSD and ACPI32.DLL, I can now boot this computer with no command line switches (PSD=ACPI.PSD)
comment:9 by , 17 years ago
Resolution: | → fixed |
---|---|
Status: | new → closed |
OK, I added the MB to ACPI-Hardware.doc
btw, test with official 3.9 release
by , 17 years ago
Attachment: | acpi-log-Model-of-your-PC.ziq added |
---|
comment:10 by , 17 years ago
Resolution: | fixed |
---|---|
Status: | closed → reopened |
With ACPI 3.09 installed, computer boots OK.
Now I'm using this computer full time, and I see a number of oddities. None of this happens every time. If I reboot a few times, I'll get good performance that works all day until I turn the machine off.
- Slowness -- when the computer boots, there is a chance that it will act "slow." In this case, many actions such as starting Firefox, or rendering a web page will be very slow. Other times the computer boots and it appears to be fine. I used the log collector on a boot that was slow. The log file is attached. (for 3.09)
- X-Workplace sometimes can't shut down, but gets stuck with a message similar to "Ugh. waiting for exception..." (only with 13042008)
- Sometimes boot stops at logo. (only with 13042008)
comment:11 by , 17 years ago
e-co
1), 2) Pasha recommends to try Panorama VESA.
SMP + slow boot -> maybe caused by SNAP
3) do you have USB stick attached?
comment:12 by , 17 years ago
e-co
Pasha continues:
- Better try boot to /APIC /SMP mode so no IRQ conflicts and you can use sound, USB.
- do you use HT? disable it.
- UNI: if USB controllers occupy one IRQ then goto BIOS setup, try assign different IRQs.
comment:13 by , 17 years ago
Milestone: | → Release version 3.10 |
---|
comment:14 by , 17 years ago
Thanks for the suggestions.
I have set up another boot volume with Panorama, most drivers disabled, and /APIC /SMP mode in use. I can use sound and USB, but even with every driver disabled, it sometimes hangs on boot. (using Alt-F2, it seems to stop on any BASEDEV line, not the same one each time.)
No USB stick is in use, no HT device (real Core 2 Duo Pentium chip), HD audio disabled in BIOS and using C-Media sound card.
But the system still starts sporadically.
For my original boot volume with SNAP and plain eCS 2.0 rc 4, I checked IRQ and found that I had 3 lines of USB UHCD driver instead of 4 lines. After adding one more line, the system still hangs sometimes, but is more likely to start OK. (This is OK for me to use, but not quite enough for me to sell to a customer.)
Unfortunately, I have found no way to reassign IRQ in the BIOS. I remain available for test or log collection.
comment:15 by , 17 years ago
e-co:
let's continue the experiments,
try 1) boot without USB,
2) try disable USB 1,1 controllers,
3) USB 2.0 controllers.
comment:16 by , 17 years ago
OK, I have made version of config.sys with all USB statements disabled Booting with /SMP /APIC GenMac is still enabled Works most times, but sometimes hangs at boot logo.
comment:17 by , 17 years ago
OK, we need the photo of the screen, here are the instructions how to get photo:
http://ecomstation.ru/projects/acpitools/?action=logs -> LOGS-1
make the photo of the screen when the computer hangs, attach the photo to the ticket.
by , 17 years ago
Attachment: | hang 1.JPG added |
---|
by , 17 years ago
Attachment: | Hang 2.JPG added |
---|
by , 17 years ago
Attachment: | Hang 3.JPG added |
---|
comment:18 by , 17 years ago
Using ACPI 20080423 debug version
I placed a reboot icon in the startup folder. Started computer, rebooted many, many times without fault. Power off, then on, computer hangs (Hang 0.jpg) power off. Several hours later, power on again (Hang 1.jpg), pressed reset (Hang 2.jpg), pressed reset (Hang 1.jpg again), pressed reset, computer boots 5 times OK, then hangs again (Hang 3.jpg).
I hope this is helpful.
comment:19 by , 17 years ago
e-co:
Hang 2 jpg - a snooper is loaded..
Did you read the documentation, the snoop.lst should be emptied (read ACPI FAQ -> Q47: eComStation 2.0: is it necessary remove all snoopers from snoop.lst?)
by , 17 years ago
Attachment: | APIC hang.jpg added |
---|
comment:20 by , 17 years ago
Thanks e-co, for note about snoop.lst. The attachment APIC hang.jpg now shows a hang with the following snoop.lst. While this computer still does hang, it appears to be less frequent with this reduced snoop.lst.
; alternative SNOOP.LST file for ACPI support IBMKBD.SNP PCIBUS.SNP
comment:22 by , 16 years ago
I removed PCIBUS.SNP from the snoop.lst. This looks like it worked for partition with Panorama. I will try adding back device drivers and see if I can make a usable computer. Look for report in a few more days.
Thanks to Eugene for netlabs registration page!
comment:23 by , 16 years ago
Milestone: | Release version 3.10 → Release version 3.11 |
---|
comment:24 by , 16 years ago
e-co:
OK, Intel motherboards are supported very well.
The ticket should be closed.
comment:25 by , 16 years ago
Resolution: | → fixed |
---|---|
Status: | reopened → closed |
comment:26 by , 16 years ago
OK, now I have completed testing. With /APIC enabled, the computer can boot reliably, but it cannot support either USB or audio. Must use Panorama Video.
Without /APIC enabled, computer does not boot reliably, but when it does boot, then it supports fully accelerated SNAP video, USB and audio.
e-co, I disagree that Intel motherboards are supported well. I cannot offer this machine to customers at this time.
Are there any other command line options to try?
comment:27 by , 16 years ago
- problems with USB => try /CD
- do you test ACPI 3.10?
- please always attach acpi logs
comment:28 by , 16 years ago
- try different version of UNIAUD. it wass improved for APIC mode in this year.
comment:29 by , 16 years ago
Please close this ticket.
I have abandoned the Intel motherboard. At Warpstock, I compared notes with Lewis Rosenthal, and we discovered that neither of us has had any joy with it.
I'll follow up on eCS.ru with a report that the motherboard does not work well with eCS.
Thanks for trying. I have switched out the motherboard for an Asus PL5-Mx motherboard, and I hope that works better. So far at least, it does.
I think you should not mark this bug as fixed. There must be a status for "Can't fix".
comment:30 by , 16 years ago
Milestone: | Release version 3.11 → eCS 2.x |
---|---|
Reporter: | changed from | to
There's really no such thing as can't fix. The term I prefer is can't (or won't fix) now. The future is hard to predict. I've moved to ticket to a future milestone so that the developers can ignore it for now.
comment:31 by , 16 years ago
Resolution: | fixed |
---|---|
Status: | closed → reopened |
comment:32 by , 16 years ago
Resolution: | → unknown |
---|---|
Status: | reopened → closed |
comment:33 by , 16 years ago
Milestone: | eCS 2.x → Release version 3.08 |
---|
Bug reported by Neil Waldhauer, neil@…