http://bugs.winehq.org/show_bug.cgi?id=25206
Summary: Anarchy Online v18.4, Ingame browser crashes (Awesomium.dll) Product: Wine Version: 1.1.35 Platform: x86 OS/Version: Mac OS X 10.6 Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: ao_main@gmx.net
Created an attachment (id=32006) --> (http://bugs.winehq.org/attachment.cgi?id=32006) output of the wine command
As of patch 18.4 funcom integrated an ingame browser based on the Awesomium.dll (which itself bases on the chromium framework).
Starting the game (which worked flawless until the patch), the game client crashes with the following in game error:
---------------------------- Exception code: 80000003 BREAKPOINT Fault address: 00F8AFF0 01:008A9FF0 C:\Program Files\Anarchy Online - Clean 18.2.2\Awesomium.dll
Registers: EAX:00000000 EBX:00000010 ECX:079DF434 EDX:00000001 ESI:079DF49C EDI:079DF4A4 CS:EIP:0017:00F8AFF0 SS:ESP:001F:079DF434 EBP:079DF6B8 DS:001F ES:001F FS:103F GS:0037 Flags:00000246 Fpu control: 0x9001f
Stack: From adr 079DF434 009FAA9C 00000000 02E6A700 02E6B000 02E6A6E0 006F8E41 079DF4A0 079DF49C 079DF6B8 00000000 00000000 0000005F 009FB080 02E49600 02E6A6A0 02E6A700 02E6B000 0000005F 0000005F 9F4B3D04 013446B3 02E6B000 000003E5 00000000 02E6A6A0 00000032 00000004 019FBE58 019FBE1C 02E6A6E0
Call stack: Address Frame Logical addr Module 00F8AFF0 079DF6B8 0001:008A9FF0 C:\Program Files\Anarchy Online - Clean 18.2.2\Awesomium.dll 7BC64FEA 079DF738 0001:00053FEA C:\windows\system32\ntdll.dll 7BC6BF07 079DFF88 0001:0005AF07 C:\windows\system32\ntdll.dll GetLogicalAddress failed for 951DF85D GetLogicalAddress failed for 951DF6E2 <end> ---------------------------- Anarchy Online has been started with a set $WINEPREFIX and "wine Anarchy.exe".
http://bugs.winehq.org/show_bug.cgi?id=25206
Nikolay Sivov bunglehead@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #32006|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=25206
ao_main@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- See Also| |http://bugs.winehq.org/show | |_bug.cgi?id=25159
http://bugs.winehq.org/show_bug.cgi?id=25206
ao_main@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- See Also|http://bugs.winehq.org/show |http://bugs.winehq.org/show |_bug.cgi?id=25159 |_bug.cgi?id=25169
http://bugs.winehq.org/show_bug.cgi?id=25206
Luke Bratch l_bratch@yahoo.co.uk changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |l_bratch@yahoo.co.uk
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #1 from Luke Bratch l_bratch@yahoo.co.uk 2010-11-17 17:10:06 CST --- I have this problem too. I have already commented in bug 25169 which I assumed was the problem I was having, but perhaps that is some other problem. I'm not sure if this is a dupe.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #2 from Luke Bratch l_bratch@yahoo.co.uk 2010-11-17 17:49:23 CST --- Applying attachment 30775 from bug 24607 changes the behaviour. It no longer crashes completely (rather it hangs), and you are able to see the game window outline be drawn briefly. It no longer gives any error, nor does it write anything to awesomium.log.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #3 from Luke Bratch l_bratch@yahoo.co.uk 2010-11-17 17:52:39 CST --- Without that patch, awesomium.log is written to, and contains the following line only:
[41:51:10049:FATAL:network_change_notifier_win.cc(54)] Check failed: 997L == ret (997 vs. 50)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #4 from Erin fiant.tenebrae@gmail.com 2010-11-17 18:12:05 CST --- I have a very similar problem; since today's patch for Anarchy Online (18.4), I haven't been able to run the client, apparently due to problems with Awesomium.dll failing to initialize. I've submitted a report to the game's support, too, if only to let them know what's happening.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #5 from Juan Lang juan_lang@yahoo.com 2010-11-17 19:20:56 CST --- Created an attachment (id=32010) --> (http://bugs.winehq.org/attachment.cgi?id=32010) Patch: Improve NotifyAddrChange stub
Does this patch help?
http://bugs.winehq.org/show_bug.cgi?id=25206
Juan Lang juan_lang@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |juan_lang@yahoo.com
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #6 from Luke Bratch l_bratch@yahoo.co.uk 2010-11-18 04:39:49 CST --- Created an attachment (id=32016) --> (http://bugs.winehq.org/attachment.cgi?id=32016) Output with attachment 32010
Hi Juan
That patch gives the same terminal output as with attachment 30775, except that it now exits and doesn't just hang. Output attached.
http://bugs.winehq.org/show_bug.cgi?id=25206
krop kropx77@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #7 from krop kropx77@gmail.com 2010-11-18 05:31:28 CST --- *** This bug has been confirmed by popular vote. ***
http://bugs.winehq.org/show_bug.cgi?id=25206
krop kropx77@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |kropx77@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=25206
rawfox rawfox@freenet.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |rawfox@freenet.de
http://bugs.winehq.org/show_bug.cgi?id=25206
ao_main@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Version|1.1.35 |1.3.7
http://bugs.winehq.org/show_bug.cgi?id=25206
perestroj perestroj1@email.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |perestroj1@email.cz
http://bugs.winehq.org/show_bug.cgi?id=25206
Eskild Hustvedt user@zerodogg.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |user@zerodogg.org
http://bugs.winehq.org/show_bug.cgi?id=25206
ao_main@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ao_main@gmx.net
http://bugs.winehq.org/show_bug.cgi?id=25206
Kaeli me@kaeli.us changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |me@kaeli.us
http://bugs.winehq.org/show_bug.cgi?id=25206
Xyphos TheGreatXyphos@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |TheGreatXyphos@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=25206
Dmitry Timoshkov dmitry@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Version|1.3.7 |1.1.35
http://bugs.winehq.org/show_bug.cgi?id=25206
Tyler Fowler tfowler6@kent.edu changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |tfowler6@kent.edu
--- Comment #8 from Tyler Fowler tfowler6@kent.edu 2010-11-21 01:47:57 CST --- same with me awaiting a response from wine
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #9 from ao_main@gmx.net 2010-11-21 15:53:21 CST --- I did some inestigations: - loading anarchy.exe using $WINEDEBUG=+loaddll yields the following: ---------- ... fixme:win:RegisterRawInputDevices (pRawInputDevices=0x33fd68, uiNumDevices=1, cbSize=12) stub! fixme:win:EnumDisplayDevicesW ((null),0,0x33ed80,0x00000000), stub! fixme:ddraw:DirectDrawEnumerateExA flags 0x00000007 not handled fixme:win:EnumDisplayDevicesW ((null),0,0x33e070,0x00000000), stub! fixme:win:EnumDisplayDevicesW ((null),0,0x33ea80,0x00000000), stub! fixme:x11drv:X11DRV_desktop_SetCurrentMode Cannot change screen BPP from 32 to 16 fixme:d3d:swapchain_init Add OpenGL context recreation support to context_validate_onscreen_formats fixme:d3d_surface:IWineD3DBaseSurfaceImpl_Blt Can't handle WINEDDBLT_ASYNC flag right now. fixme:heap:HeapSetInformation 0x0 1 0x0 0 fixme:iphlpapi:NotifyAddrChange (Handle 0x79df490, overlapped 0x2e6b010): stub ----------- And there it hangs. As stated, the Awesomium.dll throws only one line in its log file: ----------- [37:47:25651:FATAL:network_change_notifier_win.cc(54)] Check failed: 997L == ret (997 vs. 50) ----------- network_change_notifier_win.cc is a chromium source file, the calling code seems to be: ----------- ... void NetworkChangeNotifierWin::Impl::WatchForAddressChange() { HANDLE handle = NULL; DWORD ret = NotifyAddrChange(&handle, &addr_overlapped_); CHECK_EQ(ERROR_IO_PENDING, ret); addr_watcher_.StartWatching(addr_overlapped_.hEvent, this); } ... ----------- The process anarchy.exe seems to hang at .StartWatching (its still in the processlist. If i read it correct, the chromium code is installing a watcher for changes in the address (DHCP). The log file entry in awesomium.dll comes from a failing CHECK_EQ (it expects ERROR_IO_PENDING (997), but gets the handle created from the patched iphlpapi_main.c: ----------- DWORD WINAPI NotifyAddrChange(PHANDLE Handle, LPOVERLAPPED overlapped) { FIXME("(Handle %p, overlapped %p): stub\n", Handle, overlapped); *Handle = CreateEventW(NULL, FALSE, FALSE, NULL); return *Handle ? ERROR_IO_PENDING : ERROR_NOT_ENOUGH_MEMORY; } ----------- The header documentation for NotifyAddrChange states that if the handle could be returned, NO_ERROR shall be returned. The Patch returns either IO_PENDING or NOT_ENOUGH_MEMORY. Can anybody have a look at this? I am just compiling with ----------- return *Handle ? NO_ERROR : ERROR_NOT_ENOUGH_MEMORY; ----------- to test it out, but i am really a sorcerer's apprentice here.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #10 from ao_main@gmx.net 2010-11-22 13:47:35 CST --- Ok, this didn't helped. Seems i get the hanging process either with ERROR_IO_PENDING/NO_ERROR.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #11 from Dmitry Timoshkov dmitry@codeweavers.com 2010-11-22 17:12:32 CST --- Please don't change an originally reported Wine version, adding a comment stating the status with the tested version is enough.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #12 from ao_main@gmx.net 2010-11-23 02:20:52 CST --- Ok, sorry if i messed something up. But the original reported version number was simply wrong :(
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #13 from ao_main@gmx.net 2010-11-23 02:30:22 CST --- Tested versions: @ 1.3.4 --> Crashing with exception listed in the first post @ 1.3.7 patched/unpatched: --> hangs at NotifyAddrChange as described
http://bugs.winehq.org/show_bug.cgi?id=25206
Martijn Schmidt martijn.schmidt@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |martijn.schmidt@gmail.com
--- Comment #14 from Martijn Schmidt martijn.schmidt@gmail.com 2010-11-23 17:22:33 CST --- We have to consider that it might not be Awesomium that breaks the game, but rather the new input system that was introduced with the same patch?
Interesting parts of the patch notes:
Anarchy Online no longer supports versions of Windows earlier than Windows XP.
The Anarchy Online in-game store is now available.
New input system is now available. All keys should now be bindable/customizable through the "binds" tab in the F10 options menu.
Enabling/disabling sound options no longer requires the character to zone or relog to notice applied changes.
---
It also seems the current dev team for Anarchy Online is unwilling to give any details about their version of Awesomium (see DocJones' post about a PM conversation with Means). It saddens me, considering the Game Director's earlier post saying "I'm also certain that a workaround will be found for this by the communities that made it possible in the first place. It might not be today and it might not be tomorrow...but someone will figure it out." sort of gave the feeling he would at least be willing to give some answers - while not actively working on a solution (which is fine).
References: Complete patch notes: http://forums.anarchyonline.com/showthread.php?t=583943 Game Director post: http://forums.anarchyonline.com/showpost.php?p=5898198&postcount=62 DocJones post: http://forums.anarchyonline.com/showthread.php?t=584405
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #15 from ao_main@gmx.net 2010-11-24 03:09:12 CST --- While you are possibly right with the additional changes, the following investigations cocnlude me to "blame" the new dll: - v1.3.4 with a direct reference to Awesomium.dll - the NotifyAddrChange (which is the last call in the call stack) is being called from the chromium-code, which itself is part of the Awesomium-Framework
I am not saying you are wrong. All i want to say, i am totally clueless on what the issue could be here.
Thanks fur bumping and helping :)
http://bugs.winehq.org/show_bug.cgi?id=25206
maxwelldekock@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |maxwelldekock@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=25206
Neo-Vortex neovortex@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |neovortex@gmail.com
--- Comment #16 from Neo-Vortex neovortex@gmail.com 2010-11-24 16:01:43 CST --- Replacing Awesomium.dll with a stub DLL that exports all required symbols used by GUI.dll (the only DLL in Anarchy Online that utilizes Awesomium) allows the game to load fine with Wine 1.3.3, however Wine 1.3.7 causes a crash seemingly unrelated to Awesomium, so I suspect there is a regression in 1.3.7 for another bug.
Please see http://aodevs.com/index.php/topic,852.0.html for a link to the replacement DLL (including source) and information on how to use it for testing. Note that this is a dirty hack that removes functionality from the game (Item Shop), and Funcom have announced that they will be using Awesomium.dll to further extend the games user interface, so this workaround is merely for testing and to help fix the issue, not a permanent solution.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #17 from Kaeli me@kaeli.us 2010-11-24 16:22:53 CST --- The fix posted by Neo-Vortex is confirmed working for AO under Wine 1.0.1 - however, when I logged in I couldn't move the camera around using the mouse - this includes both LMB panning and RMB steering. The mouse buttons were working, because I was able to click on and interact with things and PCs in-game, just no mouse camera control. I'll investigate further when I have more time.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #18 from ao_main@gmx.net 2010-11-24 16:30:25 CST --- Awesome (You, not the dll :D): It works with wine-1.3.7 on my Mac. I am running with the NotifyAddrChange patch in my comment 9.
From here on, what can be done further? Can i test something to help you to
understand the problem better or shall i switch wine versions around (i think i have v1.3.0, 1.3.4 and 1.3.7 installed)
http://bugs.winehq.org/show_bug.cgi?id=25206
Laryllan laryllan@web.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |laryllan@web.de
--- Comment #19 from Laryllan laryllan@web.de 2010-11-29 11:21:11 CST --- I don't know if it's important, but this bug affects Linux x64 too.
http://bugs.winehq.org/show_bug.cgi?id=25206
Pugtheblack pugtheblack@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |pugtheblack@gmail.com
--- Comment #20 from Pugtheblack pugtheblack@gmail.com 2010-12-14 15:24:49 CST --- Can confirm that the Awesomium.dll trick that Neo-Vortex posted works on Ubuntu 10.10 x64 (2.6.35-23 ) and wine 1.2.2, but like Kaeli mentioned you can not reposition the camera using the mouse
cheers
http://bugs.winehq.org/show_bug.cgi?id=25206
Janne Hakonen joyer83@live.fi changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |joyer83@live.fi
--- Comment #21 from Janne Hakonen joyer83@live.fi 2010-12-18 07:54:41 CST --- (In reply to comment #2)
Applying attachment 30775 [details] from bug 24607 changes the behaviour. It no longer crashes completely (rather it hangs), and you are able to see the game window outline be drawn briefly. It no longer gives any error, nor does it write anything to awesomium.log.
I can confirm that this patch works with wine 1.3.4. I can log in to AO and play just fine.
However, trying to open item store's window causes AwesomiumProcess.exe to crash. AO just shows item store's window with white empty content, the crash doesn't crash AO's client and I can still continue to play (Screenshot: http://www.jhakonen.com/kuvat/awecrash.png).
-------------------- Ubuntu 10.10 x64 Wine 1.3.4 compiled from sources + attachment 30775 Using original Awesomium.dll that comes with AO
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #22 from Janne Hakonen joyer83@live.fi 2010-12-18 09:16:52 CST --- Created an attachment (id=32536) --> (http://bugs.winehq.org/attachment.cgi?id=32536) Backtrace of a stack overflow crash with Wine 1.3.7
Tested also with Wine versions 1.3.5, 1.3.6 and 1.3.7. 1.3.5 and 1.3.6 works fine with attachment 30775.
Wine 1.3.7 + attachment 30775 crashes with stack overflow exception while client.exe is starting up, a moment later after calling NotifyAddrChange. Attached is backtrace of the crash taken with Winedbg. Looks like D3D's texture loading is stuck in infinitive recursion.
http://bugs.winehq.org/show_bug.cgi?id=25206
ao_main@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Version|1.1.35 |1.3.7
http://bugs.winehq.org/show_bug.cgi?id=25206
llie@nano.exofire.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |llie@nano.exofire.net
--- Comment #23 from llie@nano.exofire.net 2010-12-25 15:36:32 CST --- Patch also seems to allow the Awesomium.dll to load with stable wine 1.2.2 but AwesomiumProcess.exe also crashes when attempting to open the in-game store.
The crash upon start-up affecting versions 1.3.7-1.3.10 does not seem to be related to this patch, this bug, or the Awesomium.dll but rather something related to wine D3D in 1.3.7+ on some platforms.
It is also important to mention that the inability to reposition the camera or steer using the mouse is unrelated to this bug. It is a problem within Anarchy Online itself, as Windows users playing the game natively are reporting the same problem.
How does one debug the AwesomiumProcess.exe crash? I'm not sure how to troubleshoot the secondary AwesomiumProcess.exe process.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #24 from Janne Hakonen joyer83@live.fi 2010-12-26 07:17:53 CST --- Do you think it would be wise to make different bug reports for each of these problems? * Client.exe process crashes with wine 1.3.7+ (D3D related problem) * Mouse look doesn't work * Hang of client.exe process at log off * AwesomiumProcess.exe crashes when in-game store is opened
And leave this report only for the NotifyAddrChange problem discussion?
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #25 from Neo-Vortex neovortex@gmail.com 2010-12-26 07:31:00 CST --- Making a new bug for the separate issue is a good idea however the mouse look issue affects native windows users as well, so not a wine-specific bug.
I'm guessing the Awesomium crash on item store use and hang on logout are both related to NotifyAddrChange however, so might be worth keeping those in the one ticket until proven otherwise.
The D3D bug is definitely unrelated however, and a new bug report should be made for it.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #26 from Martijn Schmidt martijn.schmidt@gmail.com 2010-12-26 07:40:37 CST --- (In reply to comment #24)
Do you think it would be wise to make different bug reports for each of these problems?
- Client.exe process crashes with wine 1.3.7+ (D3D related problem)
- Mouse look doesn't work
- Hang of client.exe process at log off
- AwesomiumProcess.exe crashes when in-game store is opened
And leave this report only for the NotifyAddrChange problem discussion?
Hang of client.exe process at log off is a non-Wine bug as well. Several Windows users have been complaining about this over at the official AO forums, and a guy who's doing an internship at the AO development project - Vhab - has pretty much acknowledged the bug.
http://forums.anarchy-online.com/showthread.php?t=584501
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #27 from Janne Hakonen joyer83@live.fi 2010-12-26 12:35:21 CST --- (In reply to comment #23)
Patch also seems to allow the Awesomium.dll to load with stable wine 1.2.2 but AwesomiumProcess.exe also crashes when attempting to open the in-game store.
The crash upon start-up affecting versions 1.3.7-1.3.10 does not seem to be related to this patch, this bug, or the Awesomium.dll but rather something related to wine D3D in 1.3.7+ on some platforms.
It is also important to mention that the inability to reposition the camera or steer using the mouse is unrelated to this bug. It is a problem within Anarchy Online itself, as Windows users playing the game natively are reporting the same problem.
I searched AO's forums and noticed only now how much there has been problems with the mouse. Although, I didn't seen mention of a problem where the mouse look or steering with mouse didn't work at all.
On windows 7, pressing RMB button down, the mouse cursor disappears and moving the mouse rotates camera.
On Ubuntu 10.10: pressing RMB button down, nothing happens, cursor is still visible and moving mouse just moves the cursor around instead of the camera.
Same problem with steering using LMB.
How does one debug the AwesomiumProcess.exe crash? I'm not sure how to troubleshoot the secondary AwesomiumProcess.exe process.
Umm, I tried to debug the crash with winedbg, but for some reason the debugger couldn't print backtrace from it. I tried this by printing backtraces of all running wine processes (and their threads). You can try this by running winedbg when the crash dialog is open and in the following command prompt give command: bt all
Another solution could be to rename the original AwesomimumProcess.exe to different name and replace the original with a new executable that runs the real AwesomimumProcess inside Winedbg.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #28 from llie@nano.exofire.net 2010-12-30 11:13:33 CST --- (In reply to comment #24)
Do you think it would be wise to make different bug reports for each of these problems?
- Client.exe process crashes with wine 1.3.7+ (D3D related problem)
- Mouse look doesn't work
- Hang of client.exe process at log off
- AwesomiumProcess.exe crashes when in-game store is opened
And leave this report only for the NotifyAddrChange problem discussion?
I have opened bugs for Client.exe crash immediately upon start up in wine 1.3.7+ here (http://bugs.winehq.org/show_bug.cgi?id=25655) and based on a recent response from a developer at FunCom (http://forums.anarchy-online.com/showpost.php?p=5917231&postcount=10) it appears the inability to steer or look with the mouse buttons is a wine issue so I also opened a bug for that issue (http://bugs.winehq.org/show_bug.cgi?id=25654).
Hanging of client.exe process at log off, at this point, does not appear to limited to Wine.
I am still unable to get any meaningful debugging information for the AwesomiumProcess.exe crash when the in-game store is opened, but I would propose one of two things. 1) Either this goal of this bug report is to track down and fix the cause AwesomiumProcess.exe crash, or 2) a new bug report should be opened for the AwesomiumProcess.exe crash and this bug declared fixed by the attached patch for NotifyAddrChange, because with this patch, wine version 1.2.2 can run Anarchy Online without the need for the swapping the Awesomium.dll.
I'm not sure what else needs to be done for NotifyAddrChange.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #29 from llie@nano.exofire.net 2011-01-02 20:29:37 CST --- Created an attachment (id=32709) --> (http://bugs.winehq.org/attachment.cgi?id=32709) Screenshot of in-game store
I'm still trying to get debugging information from the AwesomiumProcess.exe crash, but I'm unable to get anything useful, but whatever it is that the AwesomiumProcess.exe is doing, it appears that it almost manages to accomplish it. I've switched to working with Wine 1.2.2 instead of the 1.3.x series, and I was able to obtain the attached partial rendering of the in-game store when AwesomiumProcess.exe crashed.
I attempted to attach to the crashed process while the crash window was still open with winedbg and do a bt all, but the AwesomiumProcess.exe did not appear at all in the process list. Instead I got this:
Can't attach process 003a: error 5
warning: could not attach to 003a Can't attach process 003a: error 5
warning: could not attach to 003a Can't attach process 003a: error 5
warning: could not attach to 003a Can't attach process 003a: error 5
warning: could not attach to 003a 0x95ef4459: call 0x95ee93d8
The console message was:
wine: Unhandled page fault on read access to 0xfffffffd at address 0x7bc70d1e (thread 003b), starting debugger...
After I got the crash, I attempted to move the AO window, and then client.exe crashed with this message on the console:
ERROR: the memory pool "Spell_t" contains unfreed slots! ERROR: - deleting pool (slotsize=56): 1528 of 10000 slots not freed! ERROR: the memory pool "n3Zone_t" contains unfreed slots! ERROR: - deleting pool (slotsize=56): 225 of 1000 slots not freed!
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #30 from Janne Hakonen joyer83@live.fi 2011-01-05 14:13:25 CST --- Wine 1.3.10 fixes the crash on logout. CancelIPChangeNotify function was missing from iphlpapi.dll and AO was trying to call it. A stub for it was added in commit bf6a20fad39e788638f1c7c9ebc9865a9e0cdf66.
http://bugs.winehq.org/show_bug.cgi?id=25206
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #31 from Austin English austinenglish@gmail.com 2011-01-05 16:36:49 CST --- (In reply to comment #30)
Wine 1.3.10 fixes the crash on logout. CancelIPChangeNotify function was missing from iphlpapi.dll and AO was trying to call it. A stub for it was added in commit bf6a20fad39e788638f1c7c9ebc9865a9e0cdf66.
Fixed then.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #32 from Martijn Schmidt martijn.schmidt@gmail.com 2011-01-05 16:57:39 CST --- (In reply to comment #31)
(In reply to comment #30)
Wine 1.3.10 fixes the crash on logout. CancelIPChangeNotify function was missing from iphlpapi.dll and AO was trying to call it. A stub for it was added in commit bf6a20fad39e788638f1c7c9ebc9865a9e0cdf66.
Fixed then.
I'm not sure if you've bothered to read the bug description - the actual major problem, the crash of the Awesomium browser, has not been solved at all.
The bug marked as fixed in #30 was a minor, separate issue, which should have had its own Bugzilla entry because it was completely unrelated to Awesomium.
Please reopen it.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #33 from krop kropx77@gmail.com 2011-01-05 17:09:01 CST --- (In reply to comment #31)
(In reply to comment #30)
Wine 1.3.10 fixes the crash on logout. CancelIPChangeNotify function was missing from iphlpapi.dll and AO was trying to call it. A stub for it was added in commit bf6a20fad39e788638f1c7c9ebc9865a9e0cdf66.
Fixed then.
Well, no. The main issue is not fixed.
with Wine 1.3.10, Anarchy Online still crashes due to awesomium.dll.
Please reopen.
http://bugs.winehq.org/show_bug.cgi?id=25206
ao_main@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED |
--- Comment #34 from ao_main@gmx.net 2011-01-05 17:12:55 CST --- Reopened, since the original problem still persists. Patch added in comment #30 has nothing to do with the OP.
http://bugs.winehq.org/show_bug.cgi?id=25206
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |NEW
--- Comment #35 from Austin English austinenglish@gmail.com 2011-01-05 17:19:46 CST --- (In reply to comment #32)
(In reply to comment #31)
(In reply to comment #30)
Wine 1.3.10 fixes the crash on logout. CancelIPChangeNotify function was missing from iphlpapi.dll and AO was trying to call it. A stub for it was added in commit bf6a20fad39e788638f1c7c9ebc9865a9e0cdf66.
Fixed then.
I'm not sure if you've bothered to read the bug description - the actual major problem, the crash of the Awesomium browser, has not been solved at all.
The bug marked as fixed in #30 was a minor, separate issue, which should have had its own Bugzilla entry because it was completely unrelated to Awesomium.
Please reopen it.
Sorry, itchy fingers.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #36 from Janne Hakonen joyer83@live.fi 2011-01-06 01:32:55 CST --- Sorry, couldn't think of better place to inform about the fix for logout crash. The crash was considered in comment #26 to be a bug related to AO so we didn't create another bugzilla entry for it.
I have tried to figure out why AwesomiumProcess.exe crashes by running samples from Awesomium's SDK 1.6 RC2, but I can't replicate crash that would happen at startup of any sample. However, when running AwesomiumGL sample, the AwesomiumProcess.exe that comes with SDK crashes after a while if you go to some animation heavy website or move scroll bars of the web window a lot. Exception is the same page fault as with AO.
AwesomiumGL wine: Unhandled page fault on read access to 0xffffffff at address 0x7bc70d02 (thread 0029)
Anarchy Online, when trying to open Item Store wine: Unhandled page fault on read access to 0xffffffff at address 0x7bc70d02 (thread 0037)
I tried also NaviDemo (based on Awesomium 1.5) that can be downloaded from Awesomium's website, but I can't that one to crash at all. It doesn't have AwesomiumProcess.exe though, so I guess it renders the website in same the process.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #37 from Janne Hakonen joyer83@live.fi 2011-01-06 14:11:01 CST --- I think I found a way to stop the crash with AwesomiumProcess.exe when you open Item Store.
Open dlls/ntll/signal_i386.c. Look for RtlCaptureStackBackTrace function, it's on line 2421 (at least with Wine version 1.3.10). Remove the function's content and replace it with: return 0;
Does this work for you?
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #38 from Luke Bratch l_bratch@yahoo.co.uk 2011-01-06 14:25:01 CST --- Created an attachment (id=32750) --> (http://bugs.winehq.org/attachment.cgi?id=32750) Screenshot of working in game store
Yes, that makes it work fine for me! Screenshot attached... :)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #39 from Martijn Schmidt martijn.schmidt@gmail.com 2011-01-07 15:26:49 CST --- Created an attachment (id=32761) --> (http://bugs.winehq.org/attachment.cgi?id=32761) RtlCaptureStackBackTrace changes from comment #37
(In reply to comment #37)
I think I found a way to stop the crash with AwesomiumProcess.exe when you open Item Store.
Open dlls/ntll/signal_i386.c. Look for RtlCaptureStackBackTrace function, it's on line 2421 (at least with Wine version 1.3.10). Remove the function's content and replace it with: return 0;
Does this work for you?
Weird, using wine-1.3.10 distributed from Gentoo this change doesn't work. I had already applied 3 other patches:
- To fix >wine-1.3.7 crash on startup: http://bugs2.winehq.org/attachment.cgi?id=32720 - To avoid having to replace Awesomium.dll with a stub every time you start up: http://bugs2.winehq.org/attachment.cgi?id=30775 - To fix mouselook: http://bugs2.winehq.org/attachment.cgi?id=32688
My method of working, and yes I know I should have used git for an official patch:
- Copy sources, open one folder and "nano -w dlls/ntdll/signal_i386.c", make the change you described and save. - CD back to the folder just above both wine source folders, and execute "diff -crB wine-1.3.10/ wine-1.3.10-patch/ > awesomium-anarchy-online.patch". - Move awesomium-anarchy-online.patch to /etc/portage/patches/app-emulation/wine, this means it'll be automatically applied every time you install a new wine version. - Install wine (verified that the patch was applied to the used sources).
Opening the ingame store results in:
- White background where the browser should've been. - AwesomiumProcess.exe crashing.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #40 from Janne Hakonen joyer83@live.fi 2011-01-07 16:15:36 CST --- Hmm, maybe it's crashing somewhere else then. I tried your patch and no crash for me. Try running 'wine anarchy.exe' from terminal, it should print crash report to the terminal at the moment of crash. Can you take grab it and attach it here?
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #41 from Martijn Schmidt martijn.schmidt@gmail.com 2011-01-08 19:10:14 CST --- Created an attachment (id=32780) --> (http://bugs.winehq.org/attachment.cgi?id=32780) Winedbg terminal output from an attempt to run AwesomiumGL.exe
Followed these instructions for the debugging: http://forums.anarchy-online.com/showpost.php?p=5921351&postcount=41
It seems you need to have vcrun2008 installed (from winetricks) to get the browser window to start at all, but even then it's only a blank window.
The Windows-styled command prompt opened along with the browser window just shows a couple lines of "Backtrace:" without anything else.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #42 from krop kropx77@gmail.com 2011-01-08 19:16:28 CST ---
(In reply to comment #39) [cut]
- To avoid having to replace Awesomium.dll with a stub every time you start up:
http://bugs2.winehq.org/attachment.cgi?id=30775
- To fix mouselook: http://bugs2.winehq.org/attachment.cgi?id=32688
With these two patches applied to wine master, the two issues mentioned are gone. - AO starts without the stub workaround from comment #16 - The mouselook works.
I didn't need the 3rd one since AO doesn't crash on startup with master (rev fe6735b184d0067424647fdb69eba4a2a67a6165)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #43 from Martijn Schmidt martijn.schmidt@gmail.com 2011-01-13 21:53:38 CST --- (In reply to comment #42)
(In reply to comment #39) [cut]
- To avoid having to replace Awesomium.dll with a stub every time you start up:
http://bugs2.winehq.org/attachment.cgi?id=30775
- To fix mouselook: http://bugs2.winehq.org/attachment.cgi?id=32688
With these two patches applied to wine master, the two issues mentioned are gone.
- AO starts without the stub workaround from comment #16
- The mouselook works.
I didn't need the 3rd one since AO doesn't crash on startup with master (rev fe6735b184d0067424647fdb69eba4a2a67a6165)
Confirming that with Wine-1.3.11 the crash on start-up is fixed.
The following patch - http://bugs2.winehq.org/attachment.cgi?id=32720 - is NOT needed anymore.
Mouselook and anti-stub-workaround patches are still required and work properly in Wine-1.3.11.
http://bugs.winehq.org/show_bug.cgi?id=25206
jonathan-vola@hotmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jonathan-vola@hotmail.com
--- Comment #44 from jonathan-vola@hotmail.com 2011-01-18 09:27:07 CST --- I have patched 1.3.11 with the two patches (30775 and 32688) and emptied the RtlCaptureStackBackTrace function as advised but I am still getting the Awesomium.dll error.
I have uploaded this custom wine build if you would like to test it: http://www.2shared.com/file/7R7BuZ10/1311-ao.html
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #45 from Martijn Schmidt martijn.schmidt@gmail.com 2011-01-18 09:46:27 CST --- (In reply to comment #44)
I have patched 1.3.11 with the two patches (30775 and 32688) and emptied the RtlCaptureStackBackTrace function as advised but I am still getting the Awesomium.dll error.
I have uploaded this custom wine build if you would like to test it: http://www.2shared.com/file/7R7BuZ10/1311-ao.html
Does AO start up though?
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #46 from jonathan-vola@hotmail.com 2011-01-18 09:54:07 CST --- (In reply to comment #45)
(In reply to comment #44)
I have patched 1.3.11 with the two patches (30775 and 32688) and emptied the RtlCaptureStackBackTrace function as advised but I am still getting the Awesomium.dll error.
I have uploaded this custom wine build if you would like to test it: http://www.2shared.com/file/7R7BuZ10/1311-ao.html
Does AO start up though?
Anarchy.exe does yes but after logging in Client.exe crashes with the awesomium bug (So basically the same as before patching)
I also no longer see the outline of the window (I see it in the workspace switcher on the panel but not on screen) while I did before applying the patches.
Does manually compiling wine change a setting somewhere as opposed to the default that could be causing this? (I'm on Ubuntu Lucid btw)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #47 from Martijn Schmidt martijn.schmidt@gmail.com 2011-01-18 11:39:31 CST --- (In reply to comment #46)
(In reply to comment #45)
(In reply to comment #44)
I have patched 1.3.11 with the two patches (30775 and 32688) and emptied the RtlCaptureStackBackTrace function as advised but I am still getting the Awesomium.dll error.
I have uploaded this custom wine build if you would like to test it: http://www.2shared.com/file/7R7BuZ10/1311-ao.html
Does AO start up though?
Anarchy.exe does yes but after logging in Client.exe crashes with the awesomium bug (So basically the same as before patching)
I also no longer see the outline of the window (I see it in the workspace switcher on the panel but not on screen) while I did before applying the patches.
Does manually compiling wine change a setting somewhere as opposed to the default that could be causing this? (I'm on Ubuntu Lucid btw)
1: Ubuntu is a staggered release distro. This means they only throw in new packages every 6 months and keep other changes down to exclusively bugfixes or security patches. Lucid is a few months old which means they don't have the latest wine in the repositories.
2: Ubuntu is a binary distro. This means they precompile their packages before throwing them into the repositories.
Since you need the latest Wine version, and need to apply the patches before compiling, you won't get the right Wine through apt-get.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #48 from jonathan-vola@hotmail.com 2011-01-18 11:55:35 CST ---
I have patched 1.3.11 with the two patches (30775 and 32688) [...]
Usually I use playonlinux to get specific versions, but because this needs the specific patches I manually downloaded the source from SF, patched it compiled it and I still can't run AO.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #49 from Martijn Schmidt martijn.schmidt@gmail.com 2011-01-18 12:50:14 CST --- (In reply to comment #48)
I have patched 1.3.11 with the two patches (30775 and 32688) [...]
Usually I use playonlinux to get specific versions, but because this needs the specific patches I manually downloaded the source from SF, patched it compiled it and I still can't run AO.
Does it work with http://bugs2.winehq.org/attachment.cgi?id=32720 ?
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #50 from jonathan-vola@hotmail.com 2011-01-18 12:55:54 CST --- (In reply to comment #49)
(In reply to comment #48)
I have patched 1.3.11 with the two patches (30775 and 32688) [...]
Usually I use playonlinux to get specific versions, but because this needs the specific patches I manually downloaded the source from SF, patched it compiled it and I still can't run AO.
Does it work with http://bugs2.winehq.org/attachment.cgi?id=32720 ?
I can only reverse the patch (It is already applied with 1.3.11)
I will try this patch and the others with 1.3.7 and report back
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #51 from jonathan-vola@hotmail.com 2011-01-18 13:39:59 CST ---
I will try this patch and the others with 1.3.7 and report back
Nope, 1.3.7 doesn't work even with all 3 patches. Wasn't there a registry key to make wine display output from client.exe even though it's called from Anarchy.exe?
That would help with debugging.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #52 from Janne Hakonen joyer83@live.fi 2011-01-18 15:07:58 CST --- (In reply to comment #51)
I will try this patch and the others with 1.3.7 and report back
Nope, 1.3.7 doesn't work even with all 3 patches. Wasn't there a registry key to make wine display output from client.exe even though it's called from Anarchy.exe?
That would help with debugging.
The patch that Martijn offered is there in 1.3.11 already so that's not needed, unless you're using earlier version.
Wine should also print debug output from Anarchy.exe and from all other processes that it starts. Try running it from terminal and pipe what it prints to log file and attach the log file here.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #53 from jonathan-vola@hotmail.com 2011-01-19 03:49:12 CST --- Created an attachment (id=32901) --> (http://bugs.winehq.org/attachment.cgi?id=32901) Awesomium errors with patched wine
It crashes as expected at "NotifyAddrChange", but wine also crashes (Won't shut down without Ctrl-C) and doesn't write anything to the file.
winedbg hits an error before anything pops up and stops responding, further comments in file
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #54 from jonathan-vola@hotmail.com 2011-01-19 06:26:02 CST --- If anyone else on ubuntu 64-bit could download the version I compiled at http://www.2shared.com/file/7R7BuZ10/1311-ao.html and try it out we can at least discover if it's a problem with my system or config or a problem with my compile.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #55 from Janne Hakonen joyer83@live.fi 2011-01-19 07:30:38 CST --- (In reply to comment #53)
Created an attachment (id=32901)
--> (http://bugs.winehq.org/attachment.cgi?id=32901) [details]
Awesomium errors with patched wine
It crashes as expected at "NotifyAddrChange", but wine also crashes (Won't shut down without Ctrl-C) and doesn't write anything to the file.
winedbg hits an error before anything pops up and stops responding, further comments in file
I've never used PlayOnLinux, and don't get me wrong, but I think you should try to find source of the crash with vanilla Wine, that way there is less stuff that can cause problems.
So: $ git clone git://source.winehq.org/git/wine.git ~/wine-git $ cd ~/wine-git
Then get http://bugs2.winehq.org/attachment.cgi?id=30775 patch and apply it: $ patch -p1 < path/to/patch/file
$ ./configure $ make -j4 $ sudo make install
Then cd to AO's folder and set WINEPREFIX to point to a fresh location before running it through wine.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #56 from Janne Hakonen joyer83@live.fi 2011-01-19 07:34:09 CST --- (In reply to comment #54)
If anyone else on ubuntu 64-bit could download the version I compiled at http://www.2shared.com/file/7R7BuZ10/1311-ao.html and try it out we can at least discover if it's a problem with my system or config or a problem with my compile.
I can give it a spin in a few hours, although, I'm a bit sceptical that it will work since I'm running Ubuntu Maverick, might stop to some dependency problems...
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #57 from jonathan-vola@hotmail.com 2011-01-19 07:52:03 CST --- Created an attachment (id=32905) --> (http://bugs.winehq.org/attachment.cgi?id=32905) Steps taken and errors shown with brand new wine build and prefix
but I think you should try to find source of the crash with vanilla Wine
You wouldn't believe how often I hear that :)
All PoL does is download specific versions of wine in folders, basically the basic vanilla wine compiled with "./configure --prefix=...", create bash scripts to make starting the programs easier, and automatically create wine prefixes.
I used PoL to setup the scripts for AO because it's quicker that way but I run them manually, 100% bash and wine.
Therefor, all these compilations are vanilla wine (AFAIK PoL doesn't even change wine)
In the attached file I added the steps I take to patch and compile wine (After download/unpack via GUI) and the steps and output of me making a brand new prefix just in case. Still doesn't work.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #58 from Janne Hakonen joyer83@live.fi 2011-01-19 08:13:40 CST --- I see that you are running wine without specifying path where the wine binary is located. However, you do set the PATH env variable, but don't you need to export it too?
What does command: $ wine --version give you?
Try running Anarchy.exe by giving full path to the wine binary, something like this: /home/j/winehaxing/compiled/usr/bin/wine Anarchy.exe
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #59 from jonathan-vola@hotmail.com 2011-01-19 08:23:03 CST --- (In reply to comment #58)
I see that you are running wine without specifying path where the wine binary is located. However, you do set the PATH env variable, but don't you need to export it too?
What does command: $ wine --version give you?
Try running Anarchy.exe by giving full path to the wine binary, something like this: /home/j/winehaxing/compiled/usr/bin/wine Anarchy.exe
Oh wow, you got me there! Apparently I should have been setting my PATH to /home/j/winehaxing/compiled/bin/ not /usr/bin.
I'm betting this fixes the problem.
Oh wow, 2 days wasted over a freaking spelling error. Sorry about this guys, very embarrassed logging off (-_-)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #60 from Steffen Rilk steffen.rilk@gmail.com 2011-01-22 04:58:49 CST --- Is it possible to convert this patch to x86_64 : http://bugs2.winehq.org/attachment.cgi?id=30775 ? Or is there another reason why it wont patch for me ?
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #61 from Steffen Rilk steffen.rilk@gmail.com 2011-01-22 05:26:15 CST --- (In reply to comment #60)
Is it possible to convert this patch to x86_64 : http://bugs2.winehq.org/attachment.cgi?id=30775 ? Or is there another reason why it wont patch for me ?
went for the two patches in appdb, sorry to bother , all good
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #62 from rawfox rawfox@freenet.de 2011-01-22 20:31:48 CST --- Its getting closer. Fedora14_64 here on ADM 5000+, Nvidia9800GT, 6GB ram
Got me the 1.3.12 sources today and applied the following patches: http://bugs2.winehq.org/attachment.cgi?id=30775 for the startup awesomium.dll problem and http://bugs2.winehq.org/attachment.cgi?id=30775 for the mouselook. both patches are not applied to wine master anyhow, as it hasnt worked without these patches before, but just compiling the actual source from the wine git.
AO starts without problems now and the mouse is working again.
opening the ingame shop results in a wine crash and this error, shown in the console:
fixme:gdi:GdiInitializeLanguagePack stub wine: Unhandled page fault on read access to 0x00000000 at address 0x108a8f64 (thread 002f), starting debugger... fixme:gdi:GdiInitializeLanguagePack stub
Ingame opens a little window, messaging "the webview has crashed and will be restarted now" what results in a white shopwindow.
AO is finally playable, even if the webview thingy does not work for now.
Carry on the good work people :)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #63 from rawfox rawfox@freenet.de 2011-01-22 20:36:36 CST --- (In reply to comment #62) The shop window wine crash does not affect the running Anarchy.exe, the game is still playable.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #64 from jonathan-vola@hotmail.com 2011-01-23 04:43:00 CST --- To fix the ingame store: Open dlls/ntll/signal_i386.c, change the RtlCaptureStackBackTrace funtion so all it does is: return 0;
Install vcrun2008 into the prefix: Voila, in game store works fine
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #65 from rawfox rawfox@freenet.de 2011-01-23 20:23:49 CST --- (In reply to comment #64)
To fix the ingame store: Open dlls/ntll/signal_i386.c, change the RtlCaptureStackBackTrace funtion so all it does is: return 0;
Install vcrun2008 into the prefix: Voila, in game store works fine
Just returning 0 from RtlCaptureStackBackTrace does not fix it here. In fact, it changes nothing, the crash on the shop and the symptoms are all the same, returning 0 or i, AwesomiumProcess.exe crashes:
wine: Unhandled page fault on read access to 0x00000000 at address 0x10a1ce22 (thread 0036), starting debugger...
The threadnumber is completely different every time, so the source of the crashes is random fault adressed threads. Janne Harkonen fell also over this and tryed already a solution, what sadly doesnt work here, read about: http://permalink.gmane.org/gmane.comp.emulators.wine.devel/82820
Janne described it as a crash situation where application crashes if RtlCaptureStackBackTrace() is called and one of the frame pointers of call stack is invalid, pointing usually to 0xffffffff, sometimes to other invalid locations.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #66 from Janne Hakonen joyer83@live.fi 2011-01-23 22:21:15 CST --- Rawfox, your problem seems similar to what Lupusceleri has at AO's forums. Can you check my answers to that at here: http://forums.anarchy-online.com/showpost.php?p=5921532&postcount=43
And, see if those help at all?
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #67 from rawfox rawfox@freenet.de 2011-01-24 17:50:37 CST --- (In reply to comment #66)
Rawfox, your problem seems similar to what Lupusceleri has at AO's forums. Can you check my answers to that at here: http://forums.anarchy-online.com/showpost.php?p=5921532&postcount=43
And, see if those help at all?
You like to get debugger output from AwesomiumGL, no sorry, i havnt done that yet.
Some google'n on wine page fault problems made me guess problems in the multithreading code, in our case, the problem exists already -before- it enteres RtlCaptureStackBackTrace.
RtlCaptureStackBackTrace works so far, honestly i dont belive, that returning 0; fixes anything at all and i like to have that explained, kinda "pics or it didnt happen" ;)
My investigations made me learn, that while the shop window is loading, RtlCaptureStackBackTrace (modified to return 0;) is touched multible times before it chrashes and even -after- the page fault crash, it is touched again 2 times.
The function looks like this:
USHORT WINAPI RtlCaptureStackBackTrace( ULONG skip, ULONG count, PVOID *buffer, ULONG *hash ) { FIXME( "(%d, %d, %p, %p) MY_i386_1_stub!\n", skip, count, buffer, hash ); return 0;
}
Terminaloutput while pressing the shop button:
fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33f0c4, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (5, 30, 0x29c0118, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33f070, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (5, 30, 0x3e01a0, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e900, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x7bfda50, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x7bfd850, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x7bfd4d8, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (5, 30, 0x29c0228, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (5, 30, 0x3e0228, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e39c, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e474, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (5, 30, 0x3e0338, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (5, 30, 0x3e0448, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e544, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (5, 30, 0x3e04d0, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e39c, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e5a4, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e474, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e474, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33f1bc, (nil)) MY_i386_1_stub! fixme:seh:RtlCaptureStackBackTrace (3, 31, 0x33e4bc, (nil)) MY_i386_1_stub! wine: Unhandled page fault on read access to 0x00000000 at address 0x108a8f64 (thread 002a), starting debugger...
At this point, the wine error window pops up.
You can see the multible functioncalls before it crashes.
Changeing the function back to its original state, has the exact same output, when i place the fixme: in there. This should be proof, that RtlCaptureStackBackTrace is -not- the problem.
I'd be happy if someone would correct me, this multithreadig stuff is far over my head ^^
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #68 from Janne Hakonen joyer83@live.fi 2011-01-24 18:50:55 CST --- I also have multiple successful calls to RtlCaptureStackBackTrace before it crashes, in that case the debugger shows clearly that AwesomiumProcess.exe crashed inside RtlCaptureStackBackTrace, it's frame is visible at the back trace.
I have tried multiple times to find some way to get it crash to the same location as you have, but no luck so far, Awesomium just keeps on working to me :(
Only thing I can think of is that there is something that you have different from my wine setup.
Could you try this: move Anarchy Online outside of the prefix, e.g. /home/username/Anarchy $ cd /home/username/Anarchy run Anarchy.exe with clean prefix: $ WINEPREFIX=~/.wine-foo123 wine Anarchy.exe Install Gecko if wine asks you to Then log in and try to open the ingame store
If the crash still comes then, try to get more debug messages with these WINEDEBUG setting: WINEDEBUG=+seh,+tid,+relay
the +tid adds thread id to each debug message, that, hopefully could be checked against the thread id which crashed. Using that it "might" be possible to check what WINAPI functions were called before stuff broke.
RtlCaptureStackBackTrace works so far, honestly i dont belive, that returning 0; fixes anything at all and i like to have that explained, kinda "pics or it didnt happen" ;)
http://www.jhakonen.com/kuvat/wine-ao.jpg :)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #69 from rawfox rawfox@freenet.de 2011-01-25 15:11:32 CST --- Created an attachment (id=32991) --> (http://bugs.winehq.org/attachment.cgi?id=32991) AO sessionlog until wine error window comes up
WINEDEBUG=+seg,+tid,+process
Loading AO, logging in, pressing the shop button, shop crashes then.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #70 from Janne Hakonen joyer83@live.fi 2011-01-25 16:13:30 CST --- (In reply to comment #69)
Created an attachment (id=32991)
--> (http://bugs.winehq.org/attachment.cgi?id=32991) [details]
AO sessionlog until wine error window comes up
WINEDEBUG=+seg,+tid,+process
Loading AO, logging in, pressing the shop button, shop crashes then.
Looking at the log, thread 1c, which was the one that had crashed.
At first, there were quite a lot access violation (code=c0000005) exceptions at address 0x6ff3602b, but those were handled OK.
There is also code=406d1388 exception which basically just gives the thread a name. A debugger can catch it if needed. That was handled OK too.
Then the exception that crashed the application was again an access violation at different address 0x108a8f64, trying to read from memory address 0x00000000, this time the exception was not caught and it killed the app.
Next course of actions could be to add +relay to the WINEDEBUG to get more debug information. Another thing to try, could be to attach the debugger to AwesomiumProcess.exe when it crashes and using disassembler to check what that address 0x108a8f64 contains.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #71 from rawfox rawfox@freenet.de 2011-01-26 16:29:20 CST --- (In reply to comment #70)
Next course of actions could be to add +relay to the WINEDEBUG to get more debug information. Another thing to try, could be to attach the debugger to AwesomiumProcess.exe when it crashes and using disassembler to check what that address 0x108a8f64 contains.
With the +relay option im getting a timeout while the char is loading. The messages, typed to console need too mutch time, the program executes too slow. I piped that overhelming message salad to a file, bigger then 1GB. Now i havnt made it yet, to copy some pages before and after the crash because the file is so damn huge, im still working on that.
I was not able to bind any debugging to the AwesomiumProcess.exe, how can i do that ?
Btw, i used a clean wineprefix while trying this.
Do you use your exception solution in your RtlCaptureStackackTrace ? Or do you also just return 0; ?
Ill make some futher tests with AwesomiumGL
cheers ^^
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #72 from Janne Hakonen joyer83@live.fi 2011-01-26 18:11:25 CST ---
With the +relay option im getting a timeout while the char is loading. The messages, typed to console need too mutch time, the program executes too slow. I piped that overhelming message salad to a file, bigger then 1GB. Now i havnt made it yet, to copy some pages before and after the crash because the file is so damn huge, im still working on that.
Oh yeah, just remembered that I have been using the +relay mainly with testing of AwesomiumGL, but I tried it now to see what happens. I was able to log in, open the browser and then log off properly (client was kinda [very] sluggish though). Log file was something like 650MB, huge that too, but not as huge as yours.
Try logging a toon without much of prefs data, in backyard, or someplace that is small and void of other players and monsters / npcs, maybe that would help.
If that's not enough, you could use some kind of faster storage device for the log file, maybe a ram disk, like: http://en.wikipedia.org/wiki/Tmpfs
I was not able to bind any debugging to the AwesomiumProcess.exe, how can i do that ?
See http://wiki.winehq.org/UsefulRegistryKeys, look for AeDebug -key, value is "winedbg --auto %ld %ld", remove the --auto part so that it says "winedbg %ld %ld", now the debugger stay open when the crash happens.
Do you use your exception solution in your RtlCaptureStackackTrace ? Or do you also just return 0; ?
I use the exception solution.
Ill make some futher tests with AwesomiumGL
Great, lets hope that the crash comes with that one too. :)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #73 from rawfox rawfox@freenet.de 2011-01-26 20:39:41 CST --- http://i51.tinypic.com/x1aybn.jpg
Dont ask yet please, i did a lot to make it work ... Now i do a lot to make it crash coz it doesnt anymore ;)
The "from scratch experience:
- new wine-git clone - 2 patches, mouse and errorfix - return 0; - vc2008 - orm=backbuffer, multisampling=on - corefonts - d3dx9
At this point the shop opens and is working !
Ok, im actually at that point, where i can crash it while restoring RtlCaptureStackBackTrace to original.
It seems i finally reached a working __TRY exception and so far it seems a good solution for now, to finally play the game completely like before.
Now to hunt down the crashing frame ..
Thanks for your support so far :)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #74 from Janne Hakonen joyer83@live.fi 2011-01-27 02:17:50 CST --- (In reply to comment #73)
http://i51.tinypic.com/x1aybn.jpg
Dont ask yet please, i did a lot to make it work ... Now i do a lot to make it crash coz it doesnt anymore ;)
That's great! congrats! :)
Here's some comparison:
The "from scratch experience:
- new wine-git clone
Same as me.
- 2 patches, mouse and errorfix
Same.
- return 0;
Yup, that works too.
- vc2008
Don't have this one, installing it removes some errors from output, but has no other effect.
- orm=backbuffer, multisampling=on
I have tried these previously, using backbuffer caused insane flickering of the client's window, multisampling... didn't see anything different with that one enabled. I'll try these again to be sure.
- corefonts
Haven't installed that, but... I do have this installed: http://packages.ubuntu.com/maverick/ttf-mscorefonts-installer Maybe it is the same thing? I'll have to remove it and see what happens.
- d3dx9
Never tried to install that one.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #75 from rawfox rawfox@freenet.de 2011-01-27 06:13:14 CST --- (In reply to comment #73)
- return 0;
Using the original function with your __TRY exeption example and noticed, i have no negative skip values anymore.
Is a negative skip value a problem ? (-1 i.e.)
I tweaked that out by a if-statement before to prevent skip going below 0, but it crashed anyways ^^ /shrug, Illegal read adress is just illegal ..
Do we have a downloadable patch for the __TRY exception implementation already ? Or is http://permalink.gmane.org/gmane.comp.emulators.wine.devel/82820 the only source yet ?
With these 3 patches the game is playable and i think about how we can give the com a easy solution for this game, without let them go thru all these steps. A rpm, well .. hmm ..
- vc2008
- d3dx9
reduced errors
- corefonts
without them, my ingame fonts go MickyMouse
- orm=backbuffer, multisampling=on
Blackscreen without backbuffer, multisampling coz i like candy (AA) :)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #76 from Janne Hakonen joyer83@live.fi 2011-01-27 17:40:08 CST --- (In reply to comment #75)
Using the original function with your __TRY exeption example and noticed, i have no negative skip values anymore.
Is a negative skip value a problem ? (-1 i.e.)
Well, the return 0; was a makeshift solution. I didn't expect it to work very well. The __TRY solution is what I have proposed as a real correction.
Do we have a downloadable patch for the __TRY exception implementation already ? Or is http://permalink.gmane.org/gmane.comp.emulators.wine.devel/82820 the only source yet ?
When I manage to iron out remaining problems with the correction's conformance tests (http://www.winehq.org/pipermail/wine-devel/2011-January/088479.html), I will post the correction+tests to wine-patches@winehq.org, for final code review and, hopefully, inclusion to official repository. Then the correction can be downloaded at least from the archives.
With these 3 patches the game is playable and i think about how we can give the com a easy solution for this game, without let them go thru all these steps. A rpm, well .. hmm ..
Well, the ready build binaries always have the problem that they will work (usually) only on the specific distro & version where they are build and packaged. So, if I were to build and package wine, it would work only on Ubuntu 10.10.
Another way could be to try to get all the problems that AO has, corrected to official repository as fast as possible so that the stuff would be part of the unstable devel version. Then just install ready build binaries, like in Ubuntu's case from here: https://launchpad.net/~ubuntu-wine/+archive/ppa
- corefonts
without them, my ingame fonts go MickyMouse
I removed that ttf-mscorefonts-installer package with Ubuntu's package manager and ran AO. When I tried to open the in-game store, AwesomiumProcess crashed to the exact same place as it did for you. So, I think this could be it!! :)
And, yeah, the text in-game was kinda funky without the fonts.
http://bugs.winehq.org/show_bug.cgi?id=25206
Neo-Vortex neovortex@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|neovortex@gmail.com |
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #77 from rawfox rawfox@freenet.de 2011-01-28 19:34:13 CST --- (In reply to comment #76)
When I tried to open the in-game store, AwesomiumProcess crashed to the exact same place as it did for you. So, I think this could be it!! :)
I just used a clean WINEPREFIX and loaded my valid running AO and guess what ? The shop crashed. It was fixed after i installed the corefonts via winetricks.
Is this the illegal addressed threads, caused by non existing registry (font-) entries so it returns something corrupt ? Or is the fonts thingy a secondary problem, other then the one related to your finding of a crashed frame inside the RtlCaptureStackBackTrace function ??
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #78 from Janne Hakonen joyer83@live.fi 2011-01-29 04:56:01 CST --- (In reply to comment #77)
I just used a clean WINEPREFIX and loaded my valid running AO and guess what ? The shop crashed. It was fixed after i installed the corefonts via winetricks.
Is this the illegal addressed threads, caused by non existing registry (font-) entries so it returns something corrupt ?
I'm sorry, but what is this about registry entries?
Or is the fonts thingy a secondary problem, other then the one related to your finding of a crashed frame inside the RtlCaptureStackBackTrace function ??
Looks like AwesomiumProcess requires both, the RtlCaptureStackBackTrace correction and corefonts installed. It will not work without both.
The corefonts, looks like, can be installed either via winetricks to a specific WINEPREFIX or via your distro's packet manager to your linux OS. Both cases seem to work. I guess, Wine can find the fonts both from $WINEPREFIX/drive_c/windows/Fonts/ and from /usr/share/fonts/truetype/ (or where ever your distro saves its fonts).
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #79 from rawfox rawfox@freenet.de 2011-01-30 06:31:35 CST --- (In reply to comment #78)
(In reply to comment #77)
I just used a clean WINEPREFIX and loaded my valid running AO and guess what ? The shop crashed. It was fixed after i installed the corefonts via winetricks.
Is this the illegal addressed threads, caused by non existing registry (font-) entries so it returns something corrupt ?
I'm sorry, but what is this about registry entries?
My bad, i thought the registry would be asked where the fonts are stored, but i installed corefonts (clean wineprefix) and copied the registry without the font additions over. Yeah, something is looking for the file (the font) instead of looking in the registry.
So its just a "file not found" what causes the browser to crash ?
Btw, i actually played all night yesterday and it was stable at all :)
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #80 from llie@nano.exofire.net 2011-02-05 09:14:03 CST --- Does anyone know if the NotifyAddrChange return ERROR_IO_PENDING will ever get committed or if not what the "correct fix" would be if the return ERROR_IO_PENDING is just a hack?
The RtlCaptureStackBackTrace problem has been fixed by a recent commit, and so the NotifyAddrChange is the only thing standing in the way of closing this bug.
Also, could rawfox or anyone else with the black screen please drop a vote on http://bugs.winehq.org/show_bug.cgi?id=25821 so that it can work with the default orm? Because it works in wine 1.2.2, so I don't see why it shouldn't work in 1.3.xx.
Thanks
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #81 from Janne Hakonen joyer83@live.fi 2011-02-05 15:33:11 CST --- Created an attachment (id=33157) --> (http://bugs.winehq.org/attachment.cgi?id=33157) Attached are some test cases for NotifyAddrChange + CancelIPChangeNotify.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #82 from Janne Hakonen joyer83@live.fi 2011-02-05 15:55:30 CST --- (In reply to comment #81)
Created an attachment (id=33157)
--> (http://bugs.winehq.org/attachment.cgi?id=33157) [details]
Attached are some test cases for NotifyAddrChange + CancelIPChangeNotify.
Ah, forgot to mention that I have ran non-interactive tests successfully against following OSes: Win NT 4 SP6 Win 2000 Pro SP4 Win XP Pro SP3 Win 2003 SP2 Win Vista Win 2008 Win 7 Pro x86 Win 7 Pro x64
Interactive tests, which requires user to change IP address of a network interface, I have successfully ran on Win 7 Pro x64.
To turn on interactivity set following: set WINETEST_INTERACTIVE=1
To change address: Start --> Search: "View Network Connections" --> double click one of the connections --> Networking tab --> double click Internet Protocol Version 4 --> change IP address or Subnet mask --> click Ok --> click Ok
The tests will fail on Wine since NotifyAddrChange and CancelIPChangeNotify are still just stubs.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #83 from Juan Lang juan_lang@yahoo.com 2011-02-07 11:06:07 CST --- (In reply to comment #81)
Created an attachment (id=33157)
--> (http://bugs.winehq.org/attachment.cgi?id=33157) [details]
Attached are some test cases for NotifyAddrChange + CancelIPChangeNotify.
Yanna, can you make the tests succeed on Wine (using todo_wine) and send them in, along with the change to NotifyAddrChange? That one-line change is obvious enough you can just include it with your test patch, no need to credit me with it.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #84 from Juan Lang juan_lang@yahoo.com 2011-02-07 11:07:07 CST --- (In reply to comment #83)
Yanna,
I meant, Janne.. Anteksi, still drinking my coffee.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #85 from Janne Hakonen joyer83@live.fi 2011-02-07 15:12:57 CST --- (In reply to comment #83)
(In reply to comment #81)
Created an attachment (id=33157)
--> (http://bugs.winehq.org/attachment.cgi?id=33157) [details] [details]
Attached are some test cases for NotifyAddrChange + CancelIPChangeNotify.
Yanna, can you make the tests succeed on Wine (using todo_wine) and send them in, along with the change to NotifyAddrChange?
Alright, I just sent a patch with that NotifyAddrChange's return value changed to ERROR_IO_PENDING + test cases with todo_wine wrappings to patches list.
For some reason, Hotmail keeps changing the patch's line endings to CRLF, no matter what I do. I hope Wine project accepts the patch even with incorrect line endings.
Yanna,
I meant, Janne.. Anteksi, still drinking my coffee.
Np. My work colleagues from China keeps calling me Jani, dunno why ;)
http://bugs.winehq.org/show_bug.cgi?id=25206
Juan Lang juan_lang@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #86 from Juan Lang juan_lang@yahoo.com 2011-02-16 09:41:21 CST --- According to comment 80, this should have been fixed by commit 82a2ec13f01b48e78d28d630914db311b55f4948. Reopen if I'm mistaken.
http://bugs.winehq.org/show_bug.cgi?id=25206
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #87 from Alexandre Julliard julliard@winehq.org 2011-02-18 12:37:28 CST --- Closing bugs fixed in 1.3.14.
http://bugs.winehq.org/show_bug.cgi?id=25206
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Target Milestone|--- |1.2.x
http://bugs.winehq.org/show_bug.cgi?id=25206
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Target Milestone|1.2.x |---
http://bugs.winehq.org/show_bug.cgi?id=25206
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- OS/Version|Mac OS X 10.6 |Mac OS X
http://bugs.winehq.org/show_bug.cgi?id=25206
David Martin azjor@hotmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |azjor@hotmail.com
--- Comment #88 from David Martin azjor@hotmail.com 2012-04-30 22:40:51 CDT --- Hi. I am trying to get Ao running on Ubuntu 12.04, in wine 1.4. I see that there are patches to be made, but I'm oblivious as to where I need to copy/paste them. Could someone tell me what file it is I'm editing. I understand I'm editing something in wine, but I don't know where.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #89 from Laryllan laryllan@web.de 2012-05-01 04:31:21 CDT --- (In reply to comment #88)
Hi. I am trying to get Ao running on Ubuntu 12.04, in wine 1.4. I see that there are patches to be made, but I'm oblivious as to where I need to copy/paste them. Could someone tell me what file it is I'm editing. I understand I'm editing something in wine, but I don't know where.
You may try this guide: http://insomnia-gravis.org/forum/viewtopic.php?f=10&t=1431
There is only one patch needed, raw3. The guide doesn't work on 12.04 if you're on a 64bit system. On 32bit, you need to dpkg-commit the changes. Read the output, it tells you when.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #90 from David Martin azjor@hotmail.com 2012-05-01 10:01:37 CDT --- Hi. I am trying to get Ao running on Ubuntu 12.04, in wine 1.4. I see that there are patches to be made, but I'm oblivious as to where I need to copy/paste them. Could someone tell me what file it is I'm editing. I understand I'm editing something in wine, but I don't know where.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #91 from David Martin azjor@hotmail.com 2012-05-01 10:03:25 CDT --- I follow the instructions up to this point
Now it's time to apply the raw input patch needed for mouse look in AO. You can get the patch here or at the bottom of this post. You need to extract first.
Apply the patch. Code: patch -p1 < raw3.patch ./tools/make_requests
at which time the terminal responds with says there's no such file in the directory.
And I am unsure what dpkg-commit is.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #92 from llie@nano.exofire.net 2012-05-01 23:53:11 CDT --- This bug is closed.
Wine 1.4 does not require the patch described in this bug, as the fix is already incorporated in Wine 1.4. You will, however, need to apply a patch in order to enable mouse look/steer in AO in Wine. That is not related to this bug.
raw3.patch is described in a different bug related to raw input for all games. The patch is reported as being not 100% working with AO, but there is a different patch available for AO.
You can consult this link:
http://forums.anarchy-online.com/showpost.php?p=5957906&postcount=88
for specific instructions regarding AO + Wine on Ubuntu. Try to bear in mind that this is a bug tracking database, not a user support forum.
http://bugs.winehq.org/show_bug.cgi?id=25206
--- Comment #93 from David Martin azjor@hotmail.com 2012-05-02 00:37:10 CDT --- Okay well. AO still crashes in the same manner as this bug in wine 1.4
Apparently its a different bug then.
-
wine-bugs＠winehq.org