From mint-bounce@lists.fishpool.fi Fri Apr 23 19:07:39 2010 X-Authentication-Warning: antyk.ibi.uw.edu.pl: draco owned process doing -bs Date: Sat, 24 Apr 2010 01:01:41 +0200 (CEST) From: Konrad Kokoszkiewicz To: Helmut Karlowski cc: MiNT mailing list Subject: Re: [MiNT] Keyboard handler In-Reply-To: Message-ID: References: <1272014847.11375.233.camel@jetpack.demon.co.uk> <908E1FFDD13344DDB773CD5AD73BED90@mercatus.local> User-Agent: Alpine 2.00 (BSF 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.0.1 (antyk.ibi.uw.edu.pl [127.0.0.1]); Sat, 24 Apr 2010 01:01:42 +0200 (CEST) X-ecartis-version: Ecartis v1.0.0 Sender: mint-bounce@lists.fishpool.fi Errors-to: mint-bounce@lists.fishpool.fi X-original-sender: draco@ibi.uw.edu.pl Precedence: bulk List-help: List-unsubscribe: List-Id: X-List-ID: List-subscribe: List-owner: List-post: >>> Anyway, Helmut has said he had fixed it, if I understood correctly. So >>> let's wait for the patch. >> >> That only applies to the Texel/Insert-bug/feature, which is easily fixed by >> removing INSERT from the list of modifiers. The stuck-key problem that >> people experience when using an Eiffel is yet to be fixed. > > Correct. > > Here's what I've found so far (on aranym): > > Key pressed: -> scancode to ikbd_scan > > Key released -> same scancode with MSB set to ikbd_scan > > When I quit XaAES and hold a key while it quits, the release of this key is > never submitted to ikbd_scan, no matter what key it is. > > So I get the same endless repeat as the Eiffel-users. But it stops when I > press a key. Which is the same phenomenon, I guess, as when you press a key when TOS is starting, at an early stage, like when the Atari loggo appears; at this point, when you hit a key, you'll hear constant key-repeat sound from the loudspeakers until you press another key. You know, I have sold my Falcon three years ago and have stopped coding on Atari ST even earlier. So it obviously has a _very_ good chance to be a complete rubbish what I say here. But: > The interrupt-routine in acia.S is not called until a new key is pressed, so > the error seems to come from the ikbd-chip directly. As this is not present > in aranym, I assume it is reset or similar by something. > > By what or whom can this happen? v_clswk? > > I've found nothing MiNT itself. I don't think it's the v_clswk(). At least, N.AES also probably has a v_clswk() and I remember no problems with the keyboard under it. Also - I don't recognize the code in acia.S anymore. I am sure that I wrote something like that once, but now I don't remember, what part of that is mine. But: maybe interrupts get disabled for a longer period of time, when XaAES is getting shutdown? An ACIA interrupt - caused by a keypress or release - would possibly not be caught then. There are many places in MiNT code where spl(7) - or something like that - is done. Maybe one of that places causes a problem with receiving keyboard interrupts. Well, anyway, I guess I can't help - I simply forgot everything. it's been a long time. Pozdrawiam KMK