lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Wed, 21 Feb 2007 01:03:25 +0100
From:	Krzysztof Halasa <khc@...waw.pl>
To:	Udo van den Heuvel <udovdh@...all.nl>
Cc:	linux-kernel@...r.kernel.org,
	Lennart Sorensen <lsorense@...lub.uwaterloo.ca>
Subject: Re: PCI riser cards and PCI irq routing, etc

Udo van den Heuvel <udovdh@...all.nl> writes:

> saa7146: found saa7146 @ mem f896a000 (revision 1, irq 145) (0x153b,0x1157).
> saa7146: found saa7146 @ mem f89e6000 (revision 1, irq 153) (0x153b,0x1155).

IO-APICs can do such things...

Ok, I have experimented a bit with my old unused EPIA-M 600 MHz MB.

INT A, B, C, D - as seen at the MB PCI connector (using PCI-PCI
bridge or 4-function device).

Device#	IDSEL	INT (first)
0x08    A19	n/a
0x09	A20	n/a
0x0A	A21	INT C
0x0B	A22	n/a
0x0C	A23	n/a
0x0D	A24	IEEE1394 chip (INT A)
0x0E	A25	n/a
0x0F	A26	n/a
0x10	A27	USB (INT D, A, B, C)
0x11	A28	VT823x (11.5 uses INT B so it means INT A, B, C, D)
0x12	A29	onboard Ethernet (INT D)
0x13	A30	INT D
0x14	A31	INT A (the MB PCI slot is wired this way)

That's from BIOS summary screen, I haven't bothered to run Linux
and check IO-APIC stuff (there may be more than 1 set of 4 INTs).

I haven't tested devices 1-7, you can't probably use them anyway.

It means that (assuming your MB can use the same riser card as this
one), you need the following mapping on the riser:
- first slot, device 0x14 (=20), INT lines 1:1
  (the same INT and IDSEL wiring as at the motherboard PCI slot)
- second slot, device 0x13 (=19),
  INT lines rotated (device) ABCD -> DABC (MB) (i.e., line INT A as
  seen at the MB PCI slot becomes INT B at the device on the riser card
  and INT A as seen at the riser slot becomes INT D at the motherboard).

Chances are that you could probably use device 0x0A (=10) as well,
but it would require another INT rotation (= double rotation).

I bet your riser card have the following mapping:
device X   INTs 1:1
device X+1 INTs (device X+1) ABCD -> BCDA (MB = device X) (INT A
at the device slot becomes INT B at the MB connector and so on).

That means (unless INT rotations are configurable) you have to make
some (quite simple, in fact) modifications to the riser card :-(
-- 
Krzysztof Halasa
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ