https://bugs.winehq.org/show_bug.cgi?id=43120
Bug ID: 43120 Summary: Battle.net fails to launch with Wine 2.9-staging or development builds Product: Wine Version: 2.9 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: shawn.starr@rogers.com Distribution: ---
Created attachment 58322 --> https://bugs.winehq.org/attachment.cgi?id=58322 wine-2.9-dev-defaultlog-battlenet.log
Using the WineHQ RPMs 2.9 staging or development, The Battle.net agent fails to start, with the development RPMs a blizzard report dialog comes up sometimes when it dies.
Attached are the default logs from both staging/development.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #1 from Shawn Starr shawn.starr@rogers.com --- Created attachment 58323 --> https://bugs.winehq.org/attachment.cgi?id=58323 wine-2.9-staging-defaultlog-battlenet.log
wine-2.9-staging-defaultlog-battlenet.log
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #2 from Shawn Starr shawn.starr@rogers.com --- Detailed logs for staging/development can be found here:
https://spstarr.fedorapeople.org/wine/wine-2.9-dev-battlenet.log https://spstarr.fedorapeople.org/wine/wine-2.9-staging-battlenet.log
The error seems to be:
err:seh:raise_exception Unhandled exception code c0000005
https://bugs.winehq.org/show_bug.cgi?id=43120
Sebastian Lackner sebastian@fds-team.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sebastian@fds-team.de
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #3 from Shawn Starr shawn.starr@rogers.com --- Here is the heap debug from game:
https://spstarr.fedorapeople.org/wine/wine-2.9-heap-staging-battlenet.log
https://bugs.winehq.org/show_bug.cgi?id=43120
Olivier F. R. Dierick o.dierick@piezo-forte.be changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |o.dierick@piezo-forte.be
--- Comment #4 from Olivier F. R. Dierick o.dierick@piezo-forte.be --- *** WARNING! *** WARNING! *** WARNING! *** WARNING! *** The external links to log files are several hundreds megabytes each. Opening them may use lots of memory/bandwidth and cause your browser to hang/crash. *** WARNING! *** WARNING! *** WARNING! *** WARNING! ***
Next time you put external links to such big files it would be nice to warn people about the size of the file so that they don't find out when it's too late.
I confirm the crash in win7 mode.
In WinXP mode, Battle.net doesn't crash, but there is a warning about XP support being removed in the near future.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #5 from Shawn Starr shawn.starr@rogers.com --- Yeah, you want to save the links, don't click them.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #6 from Bruno Jesus 00cpxxx@gmail.com --- https://spstarr.fedorapeople.org/wine/wine-2.9-heap-staging-battlenet.log is 238Mb, after "bzip2 -9" is 8.5Mb. Compression is a must when creating log files.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #7 from Shawn Starr shawn.starr@rogers.com --- Grab all the logs in xz compression here:
https://spstarr.fedorapeople.org/wine/
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #8 from Shawn Starr shawn.starr@rogers.com --- Wine 2.5 works, Wine 2.6 does not, it turns out to be this patch
commit e87ccb8b055dc846211967e46ee2f17fbabef7a1 Author: Bruno Jesus bjesus@codeweavers.com Date: Sun Apr 9 22:25:35 2017 -0300
dinput: Assume a 1-to-1 axes map when no axes match.
The wiimote is a well known problematic device, mainly because it is not a joystick. It is a USB device. But the classic controller is at least advertised as a joystick with 6 axes and 11 buttons (js driver only, NO event).
Signed-off-by: Bruno Jesus bjesus@codeweavers.com Signed-off-by: Alexandre Julliard julliard@winehq.org
Test case:
Run wine control w/ commit applied to development (master) throws exception error Revert e87ccb8b055dc846211967e46ee2f17fbabef7a1 in wine development (master)
Also, if the device is disabled I am able to run wine w/ the commit still kept but only if the device being detected as a 'joystick' is disabled.
Device: Corsair M65 RGB Gaming Mouse
USB information:
[61080.734640] usb 3-5.3: Product: Corsair M65 RGB Gaming Mouse [61080.736010] input: Corsair Corsair M65 RGB Gaming Mouse as /devices/pci0000:00/0000:00:14.0/usb3/3-5/3-5.3/3-5.3:1.0/0003:1B1C:1B12.00D5/input/input235 [61080.736138] hid-generic 0003:1B1C:1B12.00D5: input,hidraw6: USB HID v1.11 Mouse [Corsair Corsair M65 RGB Gaming Mouse] on usb-0000:00:14.0-5.3/input0 [61080.736987] input: Corsair Corsair M65 RGB Gaming Mouse as /devices/pci0000:00/0000:00:14.0/usb3/3-5/3-5.3/3-5.3:1.1/0003:1B1C:1B12.00D6/input/input236 [61080.788953] hid-generic 0003:1B1C:1B12.00D6: input,hiddev96,hidraw7: USB HID v1.11 Mouse [Corsair Corsair M65 RGB Gaming Mouse] on usb-0000:00:14.0-5.3/input1
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #9 from Shawn Starr shawn.starr@rogers.com --- Working test case:
Revert e87ccb8b055dc846211967e46ee2f17fbabef7a1 in wine development (master) then you can run wine control w/o crash, no SEH exception occurs.
https://bugs.winehq.org/show_bug.cgi?id=43120
Gijs Vermeulen acescopezz@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |00cpxxx@gmail.com, | |acescopezz@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=43120
Sebastian Lackner sebastian@fds-team.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression Regression SHA1| |e87ccb8b055dc846211967e46ee | |2f17fbabef7a1
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #10 from Shawn Starr shawn.starr@rogers.com --- Created attachment 58370 --> https://bugs.winehq.org/attachment.cgi?id=58370 Wine control with dinput debugging from WineHQ RPM
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #11 from Bruno Jesus 00cpxxx@gmail.com --- Let's see if I get this right. Your mouse is being recognized as a joystick inside wine control joy.cpl? Can you also reproduce the same crash with wine control joy.cpl?
Please attach a +dinput log from opening the joy.cpl with and without the commit.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #12 from Bruno Jesus 00cpxxx@gmail.com --- (In reply to Shawn Starr from comment #10)
Created attachment 58370 [details] Wine control with dinput debugging from WineHQ RPM
Oops, log already here. Thanks.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #13 from Bruno Jesus 00cpxxx@gmail.com --- 0009:err:dinput:find_joystick_devices Incoherent joystick data, advertised 24 axes, detected 0. Assuming 1-to-1.
This mouse should be very big to have 24 axes. I'll have to figure out a better way to discard bogus devices.
https://bugs.winehq.org/show_bug.cgi?id=43120
Shawn Starr shawn.starr@rogers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |directx-dinput
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #14 from Shawn Starr shawn.starr@rogers.com --- I should note:
0009:trace:dinput:joydev_enum_deviceW Enumerating the linux Joystick device: /dev/input/js0 (Corsair Corsair M65 RGB Gaming Mouse (js))
I do not see a /dev/input/js0 device in /dev/input directory.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #15 from Sebastian Lackner sebastian@fds-team.de --- Based on the log there is an attempt to write past the end of the df->rgodf memory block, most likely here:
https://source.winehq.org/git/wine.git/blob/HEAD:/dlls/dinput/joystick_linux...
This is not really surprising because dwAxes + dwPOVs = 8 + 4 = 12, but the code attempts to write device_axis_count = 24 entries.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #16 from Bruno Jesus 00cpxxx@gmail.com --- I cannot confirm the regression so far. Wine will always crash if axes count > 8 with or without the commit. This seems to always have been a problem that surfaced now with such high amount of axes in your device descriptor.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #17 from Bruno Jesus 00cpxxx@gmail.com --- Created attachment 58386 --> https://bugs.winehq.org/attachment.cgi?id=58386 test patch
Please test the attached patch.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #18 from Shawn Starr shawn.starr@rogers.com --- Bruno,
Your patch worked, feel free to push into git.
Thanks, Shawn
trace:dinput:joydev_enum_deviceW Enumerating the linux Joystick device: /dev/input/js0 (Corsair Corsair M65 RGB Gaming Mouse (js)) trace:dinput:IDirectInput7WImpl_CreateDeviceEx (0x12e1f0)->({9e573ed9-7734-0000-8d4a-23903fb6bdf7}, (null), 0x12e600, (nil)) trace:dinput:mousedev_create_device 0x12e1f0 {9e573ed9-7734-0000-8d4a-23903fb6bdf7} (null) 0x12e600 1 trace:dinput:keyboarddev_create_device 0x12e1f0 {9e573ed9-7734-0000-8d4a-23903fb6bdf7} (null) 0x12e600 1 trace:dinput:joydev_create_device 0x12e1f0 {9e573ed9-7734-0000-8d4a-23903fb6bdf7} (null) 0x12e600 1 trace:dinput:joydev_create_device 0x12e1f0 {9e573ed9-7734-0000-8d4a-23903fb6bdf7} (null) 0x12e600 1 trace:dinput:alloc_device {9e573ed9-7734-0000-8d4a-23903fb6bdf7} 0x12e1f0 0x33f118 0 warn:dinput:alloc_device Can't support 24 axes. Clamping down to 8 trace:dinput:IDirectInputAImpl_AddRef (0x12e1f0) incrementing from 1
https://bugs.winehq.org/show_bug.cgi?id=43120
Bruno Jesus 00cpxxx@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Ever confirmed|0 |1 Status|UNCONFIRMED |NEW Summary|Battle.net fails to launch |Joystick enabled |with Wine 2.9-staging or |applications crash when a |development builds |device that advertises more | |than 8 axes is attached Keywords| |hardware
--- Comment #19 from Bruno Jesus 00cpxxx@gmail.com --- Confirming in Wine 2.10. The problem happens with any application, the easiest way to test is wine control joy.cpl.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #20 from Bruno Jesus 00cpxxx@gmail.com --- Should be fixed by https://source.winehq.org/git/wine.git/commitdiff/b50d1fa3140b7bc9b4749a6bfb... Please retest in wine-git or apply the patch manually.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #21 from Shawn Starr shawn.starr@rogers.com --- Building master now
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #22 from Sebastian Lackner sebastian@fds-team.de --- (In reply to Shawn Starr from comment #21)
Building master now
Does it work again in current Wine git? If yes this bug report should be marked as FIXED.
https://bugs.winehq.org/show_bug.cgi?id=43120
--- Comment #23 from Shawn Starr shawn.starr@rogers.com --- Building wine git now, It should be since wine-staging 2.11 worked fine.
https://bugs.winehq.org/show_bug.cgi?id=43120
Shawn Starr shawn.starr@rogers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Status|NEW |RESOLVED
--- Comment #24 from Shawn Starr shawn.starr@rogers.com --- This is working in latest RPMs from Fedora.
https://bugs.winehq.org/show_bug.cgi?id=43120
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #25 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 2.15.
-
wine-bugs＠winehq.org