Gentoo Archives: gentoo-dev

From: Stephane Dudzinski <stephane@×××××××××.com>
To: gentoo-dev@g.o
Subject: Re: [gentoo-dev] IRQs in 2.4.19-r3+
Date: Wed, 08 May 2002 16:44:24
Message-Id: 1020894347.2949.4.camel@tux.power.users
In Reply to: [gentoo-dev] IRQs in 2.4.19-r3+ by Alex Walker
1 Hello,
2
3 ACPI issues, switch off ACPI or rollback to a 2.4.18 kernel with a
4 proper ACPI patch, that was mentionned on the lists (the ACPI ml)
5 earlier on last month.
6
7 Steph
8
9 On Wed, 2002-05-08 at 21:39, Alex Walker wrote:
10 > (Just joined the list right now, so sorry if this repeats stuff)
11 >
12 > I've been reading of problems with IRQ assignment in
13 > gentoo-sources-2.4.19-r3 and r4.
14 >
15 > I experienced very similar problems when I tried the 2.5.x series (I
16 > think 2.5.12 was the last on I tried) So I assume some of the code has
17 > been taken from there, or from another tree. Having looked at it a bit,
18 > it may be something to do with ACPI doing IRQ routing instead of the
19 > on-board.
20 >
21 > Is there a known solution?
22 >
23 > Anyway, here's a description of my problem:
24 >
25 > USB and network card do not work under r3 or r4. with errors in dmesg
26 > thus:
27 >
28 > Linux Tulip driver version 0.9.15-pre10 (Mar 8, 2002)
29 > PCI: Could not assign IRQ 17 to device 02:0c.0
30 > eth0: Lite-On PNIC-II rev 37 at 0xd480ac00, 00:C0:F0:76:95:0A, IRQ 17.
31 >
32 > [SNIP]
33 >
34 > usb.c: registered new driver hub
35 > uhci.c: USB Universal Host Controller Interface driver v1.1
36 > PCI: Could not assign IRQ 19 to device 00:1f.2
37 > PCI: Setting latency timer of device 00:1f.2 to 64
38 > uhci.c: USB UHCI at I/O 0xef40, IRQ 19
39 > usb.c: new USB bus registered, assigned bus number 1
40 > hub.c: USB hub found
41 > hub.c: 2 ports detected
42 > PCI: Could not assign IRQ 23 to device 00:1f.4
43 > PCI: Setting latency timer of device 00:1f.4 to 64
44 > uhci.c: USB UHCI at I/O 0xef80, IRQ 23
45 > usb.c: new USB bus registered, assigned bus number 2
46 > hub.c: USB hub found
47 > hub.c: 2 ports detected
48 > usb.c: registered new driver hid
49 > hid-core.c: v1.8.1 Andreas Gal, Vojtech Pavlik <vojtech@××××.cz>
50 > hid-core.c: USB HID support drivers
51 >
52 > All seems to be working-ish from that, but USB nor networrking actually
53 > work.
54 >
55 > Then I noticed the following differences in my dmesg output:
56 >
57 > r1:
58 > PCI: PCI BIOS revision 2.10 entry at 0xfda95, last bus=2
59 > PCI: Using configuration type 1
60 > PCI: Probing PCI hardware
61 > Unknown bridge resource 0: assuming transparent
62 > PCI: Discovered primary peer bus 08 [IRQ]
63 > PCI: Using IRQ router PIIX [8086/2440] at 00:1f.0
64 > Linux NET4.0 for Linux 2.4
65 > Based upon Swansea University Computer Society NET3.039
66 > Initializing RT netlink socket
67 > Starting kswapd
68 > Journalled Block Device driver loaded
69 > devfs: v1.12 (20020219) Richard Gooch (rgooch@××××××××××.au)
70 > devfs: boot_options: 0x0
71 > Installing knfsd (copyright (C) 1996 okir@×××××××××.de).
72 > ACPI: Core Subsystem version [20011018]
73 > ACPI: Subsystem enabled
74 > Power Resource: found
75 > Power Resource: found
76 > Power Resource: found
77 > Power Resource: found
78 > ACPI: System firmware supports S0 S1 S4 S5
79 > Processor[0]: C0 C1
80 > ACPI: Power Button (FF) found
81 > ACPI: Sleep Button (CM) found
82 >
83 >
84 > r3/4:
85 > ACPI: Bus Driver revision 20020404
86 > ACPI: Core Subsystem revision 20020403
87 > PCI: PCI BIOS revision 2.10 entry at 0xfda95, last bus=2
88 > PCI: Using configuration type 1
89 > ACPI: Interpreter enabled
90 > ACPI: Using IOAPIC for interrupt routing
91 > ACPI: System [ACPI] (supports S0 S1 S4 S5)
92 > ACPI: PCI Root Bridge [PCI0] (00:00:00.00)
93 > Unknown bridge resource 0: assuming transparent
94 > ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
95 > ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PCI1._PRT]
96 > ACPI: Power Resource [URP1] (off)
97 > ACPI: Power Resource [URP2] (off)
98 > ACPI: Power Resource [FDDP] (off)
99 > ACPI: Power Resource [LPTP] (off)
100 > PCI: Probing PCI hardware
101 > PCI: Using ACPI for IRQ routing
102 >
103 >
104 > The key things being:
105 > i) In r1, the PCI hardware is probed first.
106 > ii) In r1, the PIIX IRQ router is used, not the ACPI one.
107 >
108 > HTH - should I be mailing this to lkml?
109 >
110 > Cheers.
111 >
112 > Alex.
113 >
114 > --
115 > [ All programmers are playwrights and all computers are lousy actors. ]
116 > My on-line log. http://www.x3ja.co.uk/
117 > _______________________________________________
118 > gentoo-dev mailing list
119 > gentoo-dev@g.o
120 > http://lists.gentoo.org/mailman/listinfo/gentoo-dev