ia64/linux-2.6.18-xen.hg

view Documentation/i386/IO-APIC.txt @ 524:7f8b544237bf

netfront: Allow netfront in domain 0.

This is useful if your physical network device is in a utility domain.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
author Keir Fraser <keir.fraser@citrix.com>
date Tue Apr 15 15:18:58 2008 +0100 (2008-04-15)
parents 831230e53067
children
line source
1 Most (all) Intel-MP compliant SMP boards have the so-called 'IO-APIC',
2 which is an enhanced interrupt controller, it enables us to route
3 hardware interrupts to multiple CPUs, or to CPU groups.
5 Linux supports all variants of compliant SMP boards, including ones with
6 multiple IO-APICs. (multiple IO-APICs are used in high-end servers to
7 distribute IRQ load further).
9 There are (a few) known breakages in certain older boards, which bugs are
10 usually worked around by the kernel. If your MP-compliant SMP board does
11 not boot Linux, then consult the linux-smp mailing list archives first.
13 If your box boots fine with enabled IO-APIC IRQs, then your
14 /proc/interrupts will look like this one:
16 ---------------------------->
17 hell:~> cat /proc/interrupts
18 CPU0
19 0: 1360293 IO-APIC-edge timer
20 1: 4 IO-APIC-edge keyboard
21 2: 0 XT-PIC cascade
22 13: 1 XT-PIC fpu
23 14: 1448 IO-APIC-edge ide0
24 16: 28232 IO-APIC-level Intel EtherExpress Pro 10/100 Ethernet
25 17: 51304 IO-APIC-level eth0
26 NMI: 0
27 ERR: 0
28 hell:~>
29 <----------------------------
31 some interrupts are still listed as 'XT PIC', but this is not a problem,
32 none of those IRQ sources is performance-critical.
35 in the unlikely case that your board does not create a working mp-table,
36 you can use the pirq= boot parameter to 'hand-construct' IRQ entries. This
37 is nontrivial though and cannot be automated. One sample /etc/lilo.conf
38 entry:
40 append="pirq=15,11,10"
42 the actual numbers depend on your system, on your PCI cards and on their
43 PCI slot position. Usually PCI slots are 'daisy chained' before they are
44 connected to the PCI chipset IRQ routing facility (the incoming PIRQ1-4
45 lines):
47 ,-. ,-. ,-. ,-. ,-.
48 PIRQ4 ----| |-. ,-| |-. ,-| |-. ,-| |--------| |
49 |S| \ / |S| \ / |S| \ / |S| |S|
50 PIRQ3 ----|l|-. `/---|l|-. `/---|l|-. `/---|l|--------|l|
51 |o| \/ |o| \/ |o| \/ |o| |o|
52 PIRQ2 ----|t|-./`----|t|-./`----|t|-./`----|t|--------|t|
53 |1| /\ |2| /\ |3| /\ |4| |5|
54 PIRQ1 ----| |- `----| |- `----| |- `----| |--------| |
55 `-' `-' `-' `-' `-'
57 every PCI card emits a PCI IRQ, which can be INTA,INTB,INTC,INTD:
59 ,-.
60 INTD--| |
61 |S|
62 INTC--|l|
63 |o|
64 INTB--|t|
65 |x|
66 INTA--| |
67 `-'
69 These INTA-D PCI IRQs are always 'local to the card', their real meaning
70 depends on which slot they are in. If you look at the daisy chaining diagram,
71 a card in slot4, issuing INTA IRQ, it will end up as a signal on PIRQ2 of
72 the PCI chipset. Most cards issue INTA, this creates optimal distribution
73 between the PIRQ lines. (distributing IRQ sources properly is not a
74 necessity, PCI IRQs can be shared at will, but it's a good for performance
75 to have non shared interrupts). Slot5 should be used for videocards, they
76 do not use interrupts normally, thus they are not daisy chained either.
78 so if you have your SCSI card (IRQ11) in Slot1, Tulip card (IRQ9) in
79 Slot2, then you'll have to specify this pirq= line:
81 append="pirq=11,9"
83 the following script tries to figure out such a default pirq= line from
84 your PCI configuration:
86 echo -n pirq=; echo `scanpci | grep T_L | cut -c56-` | sed 's/ /,/g'
88 note that this script wont work if you have skipped a few slots or if your
89 board does not do default daisy-chaining. (or the IO-APIC has the PIRQ pins
90 connected in some strange way). E.g. if in the above case you have your SCSI
91 card (IRQ11) in Slot3, and have Slot1 empty:
93 append="pirq=0,9,11"
95 [value '0' is a generic 'placeholder', reserved for empty (or non-IRQ emitting)
96 slots.]
98 generally, it's always possible to find out the correct pirq= settings, just
99 permute all IRQ numbers properly ... it will take some time though. An
100 'incorrect' pirq line will cause the booting process to hang, or a device
101 won't function properly (if it's inserted as eg. a module).
103 If you have 2 PCI buses, then you can use up to 8 pirq values. Although such
104 boards tend to have a good configuration.
106 Be prepared that it might happen that you need some strange pirq line:
108 append="pirq=0,0,0,0,0,0,9,11"
110 use smart try-and-err techniques to find out the correct pirq line ...
112 good luck and mail to linux-smp@vger.kernel.org or
113 linux-kernel@vger.kernel.org if you have any problems that are not covered
114 by this document.
116 -- mingo