http://bugs.winehq.org/show_bug.cgi?id=23183
Summary: RCT3: Crash in fglrx if you change/disable a ride sign Product: Wine Version: 1.1.43 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: wine-bugs@digital-trauma.de
Created an attachment (id=28839) --> (http://bugs.winehq.org/attachment.cgi?id=28839) rct3 stdout/err with backtrace, wine 1.2-rc2, fglrx 10.5
Since 1.1.43 (the first ver to fix a unrelated crashing regression) I had random crashes in Roller Coaster Tycoon 3 Platinum when setting a running ride to testing and vice versa. I only recently narrowed it down to ride signs: these are dynamically labelled signs in front of the ride queue which per default are shown for every ride in your park (they marquee the name of the ride, or "Testing..." etc.), Now if you cause this sign to change after it has been on screen already, the game may crash with an NPE in fglrx.
Steps to reproduce: Start the sandbox, build one(!) ride and a queue tile in front of the entrance so the sign is visible. Now in the details tab, disable the queue sign. If it doesn't crash already, set the ride to Testing, show sign, disable sign. At least now I consistently get a crash. If you have two or more rides, one of them may survive the sign on/off roundtrip. Expected behaviour: Just draw or not draw the sign. Actual behaviour: Crash (access violation, null pointer with offset; see stdout of two such crashes) Workaround: when building a ride, disable the queue sign before building an entrance. Doesn't work for parks you load where there already are signs present - if they change because a ride brakes down or because you want to close / edit them, the game may crash.
First seen in wine: 1.1.43 with fglrx 9.11 Last seen in wine: 1.2-rc2 with fglrx 10.5 (see logs) Also seen in wine: 1.1.43 with fglrx 10.4 1.1.44 with fglrx 10.4 1.1.44 with fglrx 10.5
http://bugs.winehq.org/show_bug.cgi?id=23183
--- Comment #1 from Dmitry Timoshkov dmitry@codeweavers.com 2010-06-15 00:46:26 --- If you think that this is a regression please perform the regression test: http://wiki.winehq.org/RegressionTesting
http://bugs.winehq.org/show_bug.cgi?id=23183
--- Comment #2 from Thomas R. wine-bugs@digital-trauma.de 2010-07-06 11:00:43 --- I can't perform the regression test due to the reasons stated. The problem persists with wine 1.2-rc5 and fglrx 10.6.
http://bugs.winehq.org/show_bug.cgi?id=23183
JD jd@steffennet.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jd@steffennet.org
--- Comment #3 from JD jd@steffennet.org 2010-11-21 17:28:20 CST --- Hello, I can reproduce this bug consistently following the instruction provided by the bug reporter. I am using Arch Linux x86_64 with Multilib installed. I have a Radeon HD 4860.
Console output is very similiar to the reporters output.
I just wanted to provide more corroboration for this bug.
http://bugs.winehq.org/show_bug.cgi?id=23183
--- Comment #4 from Thomas R. wine-bugs@digital-trauma.de 2010-11-28 21:47:00 CST --- Still crashes as described in wine 1.3.7 (Catalyst 10.11)
http://bugs.winehq.org/show_bug.cgi?id=23183
Thomas R. wine-bugs@digital-trauma.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |FIXED
--- Comment #5 from Thomas R. wine-bugs@digital-trauma.de 2011-07-11 19:10:37 CDT --- Hooray! The latest ATI fglrx driver fixes this and a lot of other bugs. Now the signs don't have strange colored pixels on the lower mipmap levels, too, and are readable from afar. I tried the usual and couldn't get the game to crash. Seems I will spend this weekend building coasters :).
Wine 1.3.7, ATI drivers 11.6.
http://bugs.winehq.org/show_bug.cgi?id=23183
Dmitry Timoshkov dmitry@baikal.ru changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|FIXED |INVALID
--- Comment #6 from Dmitry Timoshkov dmitry@baikal.ru 2011-07-11 22:58:14 CDT --- Nothing has been fixed in Wine. Invalid.
http://bugs.winehq.org/show_bug.cgi?id=23183
André H. nerv@dawncrow.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED CC| |nerv@dawncrow.de
--- Comment #7 from André H. nerv@dawncrow.de 2011-07-23 08:37:43 CDT --- Closing invalid.