|Re: [hatari-devel] Lethal Xcess load state|
[ Thread Index |
| More lists.tuxfamily.org/hatari-devel Archives
- To: hatari-devel@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [hatari-devel] Lethal Xcess load state
- From: Vincent Rivière <vincent.riviere@xxxxxxxxxxx>
- Date: Wed, 20 Jun 2018 09:11:12 +0200
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=sender:subject:to:references:from:message-id:date:user-agent :mime-version:in-reply-to:content-language:content-transfer-encoding; bh=PIXOFepBfXug1mqtzZ7O7KGN7yCujNMWxCTRXuz6h8I=; b=m/36OYLxIeVepioZOG44j31oxqnrtaSbs0m7gsStc2Mdztli5QRencdPP4zx37w9pR v7I+y8MLLQ+W6NTDVDWKAvZh4IIspmH9QAVdDSiO6VAib4cCrfNUpzFLv2plccpENie5 r2hVeH6IUCxoSnM9fKlS3TpUYEZgxSSrty5xjbk69tYeWrtodn/HHC3EF3RWUZci/oOk h0+dd9jaZnvkBV1uVbMBhpKJC0ZnwzhN/HY4eH5JqlfBhbIsWjlx7RVwVcL1rYKlZZ4W rPg51/vNI5RlK04azPx+2FaOtCJiqvtVQVFE3MwFprk8gaKT9H/Gc2n5KhksUcJVhiNy QEqg==
Le 19/06/2018 à 10:34, Eero Tamminen a écrit :
There might be some difference in HW (interrupt) initialization between
I forgot to mention something very important.
I always start Hatari normally, then I load the snapshot using AlrGR+L.
So I'm the cause of the randomness. The time spent between TOS boot and
the moment I press AltGr+L is different on each boot, so the internal
machine state is different. Of course it should always work, but it is
what triggers the bug.
Note that I now use Hatari HG on Linux, the issue is the same as on
Cygwin. Still using TOS 1.62.
1) If I run "hatari -c lethal.cfg --memstate
~/.config/hatari/hatari.sav", it always works fine (and this will be the
right workaround for serious Hatari usage).
2) If I run "hatari -c lethal.cfg", wait less than 1 second on white
screen, and press AltGr+L, it usually works fine.
3) If I run "hatari -c lethal.cfg", wait about 1.5 second on white
screen, and press AltGr+L, it usually fails with bad palette
On each case, after initial boot and AltGr+L, I can press AltGr+L any
times I want, it never changes the status: always good or always bad. I
need to close Hatari and start it again to get a different status.
So really, loading snapshots fails to reinitialize some hardware, so the
result incorrectly depends on machine state before loading the snapshot.