Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 15 Jan 2009 09:36:27 +0300
From:      pluknet <pluknet@gmail.com>
To:        "Paul B. Mahol" <onemda@gmail.com>
Cc:        Alexey Shuvaev <shuvaev@physik.uni-wuerzburg.de>, freebsd-current <freebsd-current@freebsd.org>
Subject:   Re: kernel doesn't boot when is built with ukbd/ums
Message-ID:  <a31046fc0901142236h522b1b02mbff4e0614ba7aacf@mail.gmail.com>
In-Reply-To: <3a142e750901141500i3445a0c1j1f07c24df547bfab@mail.gmail.com>
References:  <a31046fc0901091428j7871e6d9h97ab00776026aadf@mail.gmail.com> <20090114223714.GA2573@wep4059.physik.uni-wuerzburg.de> <3a142e750901141500i3445a0c1j1f07c24df547bfab@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
2009/1/15 Paul B. Mahol <onemda@gmail.com>:
> On 1/14/09, Alexey Shuvaev <shuvaev@physik.uni-wuerzburg.de> wrote:
>> On Sat, Jan 10, 2009 at 01:28:19AM +0300, pluknet wrote:
>>> Hi.
>>>
>>> Today I noticed that when kernel has build-in ums and ukbd support,
>>> it stops booting with the last seen messages (transcribed):
>>> ...
>>> uart0: [FILTER]
>>> atkbdc0: <Keyboard controller (i8042)> port 0x60,0x64 irq1 on acpi0
>>> [stops here]
>>> ...
>>>
>>> It boots fine if kernel is built without ums and ukbd devices (and
>>> they loaded as modules). And I see in this case.
>>> ...
>>> atrtc0: <AT realtime clock> port 0x70-0x71 irq 8 on acpi0
>>> uart0: <16550 or compatible> port 0x3f8-0x3ff irq 4 flags 0x10 on acpi0
>>> uart0: [FILTER]
>>> atkbdc0: <Keyboard controller (i8042)> port 0x60,0x64 irq 1 on acpi0
>>> atkbd0: <AT Keyboard> irq 1 on atkbdc0
>>> kbd0 at atkbd0
>>> atkbd0: [GIANT-LOCKED]
>>> atkbd0: [ITHREAD]
>>> cpu0: <ACPI CPU> on acpi0
>>> ...
>>>
>>> In both cases I have in loader.conf:
>>>
>>> ums_load="YES"
>>> ukbd_load="YES"
>>>
>> Maybe this is the problem? I have some feeling that it is illegal to load
>> modules already compiled in the kernel with the loader.
>> On the running system kldload (or what is responsible for it) detects
>> that the module is already in the kernel, but loader can't detect this.
>> Try commenting out these lines while booting kernel with ums/ukbd modules.
>
> +1. Loader cant detect this, but kernel can (at same rate) and in such
> cases it will survive boot.
> Old incompatible modules can cause other problems ...
>

Ok, guys. I'll try this next time :)
Just wanted to mention that this is first time when I encounter this minor issue
(i.e. problem with boot when driver is built in kernel and loaded as a module).

-- 
wbr,
pluknet



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?a31046fc0901142236h522b1b02mbff4e0614ba7aacf>