hello ,
I have Fedora Core 1 one in my home PC. It' reports a bug like this ...
atkbd.c : This is an XFree86 bug .It should not access hardware directly. atkbd.c: unknown key released (translated set 2,code 0x7a on isa0060/serio0).
Please tell me if possible what the error message indicates.
----------------------------------------- Rajagiri School of Engineering and Technology Kakkanad, Ernakulam http://rajagiritech.ac.in/
Am Mi, den 04.02.2004 schrieb jeffrin_jose@rajagiritech.ac.in um 16:59:
hello ,
I have Fedora Core 1 one in my home PC. It' reports a bug like this ...
atkbd.c : This is an XFree86 bug .It should not access hardware directly. atkbd.c: unknown key released (translated set 2,code 0x7a on isa0060/serio0).
Please tell me if possible what the error message indicates.
Quoting http://lwn.net/Articles/69107/
Problem:
Kernel reports: atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0). atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly. Solution:
Well, the kernel means what it says. XFree86 boldly goes and accesses the keyboard controller registers when it starts up. This is a bad thing to do, as it can conflict with the kernel using these registers at the same time. The kernel spots this and complains, and in most cases is not affected by the problem.
So, unless you are an XFree86 developer and can fix X, ignore this message.
HTH
CU thl
On Wed, 4 Feb 2004, Thorsten Leemhuis wrote:
Date: Wed, 04 Feb 2004 18:12:28 +0100 From: Thorsten Leemhuis fedora@leemhuis.info To: "fedora-devel-list@redhat.com" fedora-devel-list@redhat.com Content-Type: text/plain List-Id: Development discussions related to Fedora Core <fedora-devel-list.redhat.com> Subject: Re: Xfree86 bug related.
Am Mi, den 04.02.2004 schrieb jeffrin_jose@rajagiritech.ac.in um 16:59:
hello ,
I have Fedora Core 1 one in my home PC. It' reports a bug like this ...
atkbd.c : This is an XFree86 bug .It should not access hardware directly. atkbd.c: unknown key released (translated set 2,code 0x7a on isa0060/serio0).
Please tell me if possible what the error message indicates.
Quoting http://lwn.net/Articles/69107/
Ah good, that's a start. It's more or less what I figured though. Wether the "fix" is simple or not depends on why X exactly is accessing the keyboard controller, which I may have known at one point but don't recall.
The FAQ seems to state that the kernel detects this *cough* problem *cough* and handles it properly however, so there's no problem actually occuring (at least that I can see). Wouldn't be a high priority to look into, but low priority bugs such as this are good low hanging fruit for interested volunteers to help out OSS development by tracking the problem down and trying to fix it.
If anyone is interested in looking at this and needs any assistance figuring out X, etc. feel free to ping me in #freedesktop, or #xfree86 on irc.freenode.net.
Might want to drop it in bugzilla also as a low priority tracker bug perhaps.
On Wed, Feb 04, 2004 at 05:21:18PM -0500, Mike A. Harris wrote:
The FAQ seems to state that the kernel detects this *cough* problem *cough* and handles it properly however, so there's no problem actually occuring (at least that I can see). Wouldn't be
*MOST* of the time. There is a finite possibility every time X touches the keyboard controller that the timing rules are violated and some machines hang or lose the keyboard until power cycle.
Alan
On Wed, 4 Feb 2004 jeffrin_jose@rajagiritech.ac.in wrote:
I have Fedora Core 1 one in my home PC. It' reports a bug like this ...
atkbd.c : This is an XFree86 bug .It should not access hardware directly. atkbd.c: unknown key released (translated set 2,code 0x7a on isa0060/serio0).
Please tell me if possible what the error message indicates.
$ locate atkbd.c linux-2.6.0/drivers/input/keyboard/atkbd.c
Looks like a kernel warning. XFree86 accesses a lot of hardware directly, by design. I'm not sure it's fair for the kernel to say that is a bug, however I would certainly agree that it isn't the best possible design for the Linux usage case.
From the message above alone though, having never seen it before,
it's hard to conclude what the real problem is. Perhaps the kernel keyboard driver maintainer, or another kernel hacker could extrapolate more on what this problem is.
If it's a small bug or issue of some kind in XFree86 that's easily fixable, then there's no reason I can see not to fix it.
If the kernel error above implies half of XFree86's keyboard support should be thrown out and rewritten though to please the Linux kernel, then I welcome patches from kernel people (or anyone) to implement that. ;o)
In the mean time, lets see what google says.
On Wed, Feb 04, 2004 at 04:48:16PM -0500, Mike A. Harris wrote:
Looks like a kernel warning. XFree86 accesses a lot of hardware directly, by design. I'm not sure it's fair for the kernel to say that is a bug, however I would certainly agree that it isn't the best possible design for the Linux usage case.
Its a bug in XFree86 or maybe even the build
If the kernel error above implies half of XFree86's keyboard support should be thrown out and rewritten though to please the Linux kernel, then I welcome patches from kernel people (or anyone) to implement that. ;o)
If you build XFree86 properly then it should use the stuff already present for keyboard control via ioctl. X has the stuff, and since Linux 2.4.x added the keyboard rate reprogramming ioctl shouldnt be touching the ports directly.
The ports X touches may not even exist or may be reassigned on legacy free systems (thankfully nobody has reassigned them on production boards)
Alan