Gentoo Archives: gentoo-user

From: Michael Mol <mikemol@×××××.com>
To: gentoo-user@l.g.o
Subject: Re: [gentoo-user] new machine : mouse trouble : solved by accident
Date: Fri, 14 Sep 2012 15:46:53
Message-Id: CA+czFiBp--1GDv2h+1o8gTqH7UxSNOw0ZO+qkCsHOvC2ZS-HUA@mail.gmail.com
In Reply to: Re: [gentoo-user] new machine : mouse trouble : solved by accident by Paul Hartman
1 On Fri, Sep 14, 2012 at 10:22 AM, Paul Hartman
2 <paul.hartman+gentoo@×××××.com> wrote:
3 > On Fri, Sep 14, 2012 at 6:21 AM, Michael Mol <mikemol@×××××.com> wrote:
4 >> On Fri, Sep 14, 2012 at 12:24 AM, Walter Dnes <waltdnes@××××××××.org> wrote:
5 >>>
6 >>> On Thu, Sep 13, 2012 at 10:06:01PM -0400, Philip Webb wrote
7 >>>
8 >>> > The effect is quite bizarre & it was the gods who saved me :
9 >>> > if I hadn't happened to plug the mouse into the neighbouring port,
10 >>> > I could have spent days struggling to find out what was wrong
11 >>> > & even taken the mobo back to the store as defective.
12 >>> >
13 >>> > Both Mageia -- installed from USB stick in a partition on the
14 >>> > HDD -- & System Rescue show /dev/input/mouse0 after booting ;
15 >>> > in the case of SR it does it before I enter the GUI via 'wizard'.
16 >>> > They have no problem with the 2.0/1.1 port, but the Gentoo system as
17 >>> > I've installed it don't show /dev/input/mouse0 from that port, but
18 >>> > only if I plug the mouse into the 3.0/2.0 port. Someone suggested
19 >>> > it is caused by a Kernel .config setting, which if enabled seems to
20 >>> > force the system to look in the 3.0 port. Why it should do that
21 >>> > doesn't make much sense : such upgrades are usually permissive,
22 >>> > not restrictive. BTW there's no difference between 3.4.9 & 3.5.3 .
23 >>>
24 >>> Is the cpu AMD? Intel machines require UHCI (USB 1.1) and AMD
25 >>> machines require OHCI (USB 1.0) for lowspeed USB devices like keyboards
26 >>> and mice. There's a root hub translator selection in .config that's
27 >>> *SUPPOSED* to work with keyboards+mice, using only the EHCI kernel
28 >>> driver, but I never could get it to work.
29 >>
30 >>
31 >> UHCI vs OHCI has nothing to do with the CPU, but with the chipset on the
32 >> system. I haven't seen an OHCI-supporting chip in over a decade, either, and
33 >> most of my systems have been AMD.
34 >>
35 >> Either way, there's no harm in enabling both.
36 >
37 > On my laptop (circa 2004) I have to load the USB modules (?HCI) in a
38 > specific order otherwise things don't work properly. I don't remember
39 > what that order is, exactly, as I'm not using it at the moment, but
40 > thought I'd mention it FWIW...
41
42 That may be true. I always kept them as built-ins, as having a USB
43 keyboard be unavailable during a failed boot sequence would be a PITA.
44
45 >
46
47
48
49 --
50 :wq