different IRQ settings for different MPS settings?

different IRQ settings for different MPS settings?

[John R Lenton]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 1337 bytes --]

With a MPS setting of 1.4 USB doesn't work on me; it timeouts,
constantly.  With MPS setting of 1.1 everything is OK.

a dirty diff of lspci -vvvxx gives

     62c62
     <       Interrupt: pin D routed to IRQ 5
     ---
     >       Interrupt: pin D routed to IRQ 19
     77c77
     <       Interrupt: pin D routed to IRQ 5
     ---
     >       Interrupt: pin D routed to IRQ 19
     103c103
     <       Interrupt: pin A routed to IRQ 11
     ---
     >       Interrupt: pin A routed to IRQ 18
     124c124
     <       Interrupt: pin A routed to IRQ 5
     ---
     >       Interrupt: pin A routed to IRQ 19
     140c140
     <       Interrupt: pin A routed to IRQ 10
     ---
     >       Interrupt: pin A routed to IRQ 16
     157c157
     <       Interrupt: pin A routed to IRQ 12
     ---
     >       Interrupt: pin A routed to IRQ 17
     186c186
     <       Interrupt: pin A routed to IRQ 10
     ---
     >       Interrupt: pin A routed to IRQ 16

(I've attached both lspci -vvvxx's, so you can diff -u that if
you want).

My question(s) is(are) is this a known bug, is this correct
behaviour, am I missing something, and why is USB the only
subsystem affected.

Phew.

-- 
John Lenton (john@grulic.org.ar) -- Random fortune:
Se Deus é amor e o amor é cego, Ray Charles é Deus?

[-- Attachment #2: lspci_mps1.1.gz --]
[-- Type: application/octet-stream, Size: 2091 bytes --]

[-- Attachment #3: lspci_mps1.4.gz --]
[-- Type: application/octet-stream, Size: 2091 bytes --]

RE: different IRQ settings for different MPS settings?

[Dunlap, Randy]

Hi,

> From: John R Lenton [mailto:john@grulic.org.ar]
> 
> With a MPS setting of 1.4 USB doesn't work on me; it timeouts,
> constantly.  With MPS setting of 1.1 everything is OK.
> 
...
> 
> My question(s) is(are) is this a known bug, is this correct
> behaviour, am I missing something, and why is USB the only
> subsystem affected.

It is known that several systems have problems with USB if
MPS is set to 1.4, and some systems work OK with that
setting.  Problem cause is still unknown.

I'd be interested to see you boot logs (kernel messages,
specifically MP table info) from Linux boots with
MPS = 1.1 and 1.4, to see if there are any differences
in them.  The big difference should be the addition
of some address space mappings, which Linux ignores
(doesn't parse, doesn't use) -- but I've never seen
a BIOS that implements address space mappings in the
MP table [and if one did, it looks like Linux would
just loop forever in smp_read_mpc(), so that's probably
not the issue here].

~Randy

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/