https://bugs.winehq.org/show_bug.cgi?id=57523
Bug ID: 57523 Summary: PokerTracker 4: cannot launch anymore Product: Wine Version: 10.0-rc1 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: blocker Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: nicoco@nicoco.fr Distribution: ---
Created attachment 77573 --> https://bugs.winehq.org/attachment.cgi?id=77573 Backtrace provided by wine's GUI
PokerTracker 4 crashes on startup.
It used to work, cf my report on AppDB. Hopefully the backtrace will help.
I have a "feeling" it might be related to edge webview: the app has a web view of their website on the welcome screen, which never worked (but it would not crash the app before). I don't really have any more than my gut feeling to support that though.
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #1 from Rafał Mużyło galtgendo@o2.pl ---
Hopefully the backtrace will help.
No backtrace available
...:sigh:...
Hopefully, you can attach the rest of the console output.
(also, a backtrace without debug symbols rarely helps, especially if the crash happens in the proprietary program instead of wine itself)
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #2 from nicoco@nicoco.fr --- Here's the console output:
``` 0058:fixme:mountmgr:harddisk_ioctl Unsupported ioctl 74080 (device=7 access=1 func=20 method=0) 0058:fixme:mountmgr:query_property Faking StorageDeviceProperty data wine: Unhandled exception 0xc0000017 in thread 158 at address 00006FFFFFC8F7F8 (thread 0158), starting debugger... 010c:err:winebth:bluez_dbus_loop Error getting object list from BlueZ: 'org.freedesktop.DBus.Error.TimedOut': 'Failed to activate service 'org.bluez': timed out (service_start_timeout=25000ms)' 0160:fixme:dbghelp:elf_search_auxv can't find symbol in module ```
FWIW, PT4 can be downloaded and tested from https://www.pokertracker.com/products/PT4/download.php
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #3 from Rafał Mużyło galtgendo@o2.pl --- ...FFS, the *whole* console output.
If the crash message is:
wine: Unhandled exception 0xc0000017 in thread 158 at address 00006FFFFFC8F7F8 (thread 0158), starting debugger...
then (in most of the cases) the interesting part are messages from that thread, not from the others.
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #4 from nicoco@nicoco.fr --- Created attachment 77582 --> https://bugs.winehq.org/attachment.cgi?id=77582 console log with WINEDEBUG=+all
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #5 from nicoco@nicoco.fr --- Sorry about my noobishness, but I don't know how to provide the messages from a specific thread. I attached the console log with WINEDEBUG=+all
I'm happy to provide more traces or logs, or do experiments if this helps. Just tell me how. ;)
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #6 from Rafał Mużyło galtgendo@o2.pl --- ...yeah, about that: a '+all' output is *rarely* asked for as in overwhelming majority of the cases too verbose to be useful; as the first step it's almost always wrong thing to do.
If you're trying to say that those few lines from comment 2 were the complete output, that's odd. Otherwise lets start with the *standard* full output.
As for the 'thread' part: for most of the wine output lines that start with a number, that number is the thread number.
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #7 from nicoco@nicoco.fr --- It was the complete output, yes. If I uninstall winbind now so it says a few more lines, but I don't think they matter (maybe they do?):
/opt/wine-devel/bin/wine ~/.wine/drive_c/Program\ Files\ (x86)/PokerTracker\ 4/PokerTracker4.exe
0158:err:winediag:ntlm_check_version ntlm_auth was not found. Make sure that ntlm_auth >= 3.0.25 is in your path. Usually, you can find it in the winbind package of your distribution. 0158:err:ntlm:ntlm_LsaApInitializePackage no NTLM support, expect problems 0058:fixme:mountmgr:harddisk_ioctl Unsupported ioctl 74080 (device=7 access=1 func=20 method=0) 0058:fixme:mountmgr:query_property Faking StorageDeviceProperty data wine: Unhandled exception 0xc0000017 in thread 158 at address 00006FFFFFC8F7F8 (thread 0158), starting debugger... 0160:fixme:dbghelp:elf_search_auxv can't find symbol in module
(crash report/trace dialog→close) (exit with code 23)
then this (unreliably) appears in the terminal but I don't think it's related:
010c:err:winebth:bluez_dbus_loop Error getting object list from BlueZ: 'org.freedesktop.DBus.Error.TimedOut': 'Failed to activate service 'org.bluez': timed out (service_start_timeout=25000ms)'
https://bugs.winehq.org/show_bug.cgi?id=57523
Fabian Maurer dark.shadow4@web.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Severity|blocker |normal CC| |dark.shadow4@web.de
--- Comment #8 from Fabian Maurer dark.shadow4@web.de --- Not a blocker.
0xc0000017 is STATUS_NO_MEMORY, could it be simply out of memory?
You could try with older wine versions to see what the last working version is. Preferably in a clean WINEPREFIX.
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #9 from nicoco@nicoco.fr ---
Not a blocker.
Sorry about that, since I could edit the field I figured it meant that this specific program is not usable.
could it be simply out of memory?
This is unlikely, there is 24GB of RAM on that machine. I also witnessed the same crash on another distro (debian vs fedora).
You could try with older wine versions to see what the last working version is. Preferably in a clean WINEPREFIX.
I tried that but for some reason apt refuses to install wine-devel now and I haven't managed to fix it yet. Will report back when I do.
https://bugs.winehq.org/show_bug.cgi?id=57523
Fabian Maurer dark.shadow4@web.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW URL| |https://www.pokertracker.co | |m/products/PT4/download.php | |?file=PT-Install-x64-v4.18. | |5.exe Ever confirmed|0 |1 Component|-unknown |ntdll Keywords| |download, regression
--- Comment #10 from Fabian Maurer dark.shadow4@web.de --- No worries, I got it - it first appears in wine-8.6. Bisected to commit a612ab6f2a45bc08f5b39e22edc58bb541b26ae1 Author: Tatsuyuki Ishi ishitatsuyuki@gmail.com Date: Mon Apr 10 17:43:28 2023 +0900
ntdll: Use log-linear bucketing for free lists.
https://bugs.winehq.org/show_bug.cgi?id=57523
Fabian Maurer dark.shadow4@web.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ishitatsuyuki@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #11 from Tatsuyuki Ishi ishitatsuyuki@gmail.com --- Is this a 32-bit app? Have you tried patching in the LAA flag in the binary header?
e.g. the 4GB patch tool linked from https://unix.stackexchange.com/questions/729197/how-does-large-address-aware....
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #12 from Fabian Maurer dark.shadow4@web.de --- (In reply to Tatsuyuki Ishi from comment #11)
Is this a 32-bit app? Have you tried patching in the LAA flag in the binary header?
Unfortunately it's a 64bit program.
https://bugs.winehq.org/show_bug.cgi?id=57523
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |a612ab6f2a45bc08f5b39e22edc | |58bb541b26ae1
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #13 from Tatsuyuki Ishi ishitatsuyuki@gmail.com --- The app creates a tightly bounded heap, then allocate a single object out of it:
13761.528:0148:014c:trace:heap:RtlCreateHeap flags 0x4, addr 0000000000000000, total_size 0xc00, commit_size 0xc00, lock 0000000000000000, params 0000000000000000 13761.528:0148:014c:Ret KERNEL32.HeapCreate() retval=01470000 ret=14035a75b 13761.528:0148:014c:Call ntdll.RtlAllocateHeap(01470000,0000000c,00000af0) ret=14035a76f
commit_size is rounded to page boundary, so we have 0x1000 bytes for the entire heap. The allocation needs 0xaf0, which leaves us with 0x510 bytes left.
With the commit in question sizeof(struct heap) is 0x8f0, which is too large to fit into the equation. I'll give you a patch that shrinks struct heap to see if it solves your issue. If this works, let me think if there's a way to address this without regressing the performance intensive use case the commit was trying to fix.
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #14 from Tatsuyuki Ishi ishitatsuyuki@gmail.com --- Created attachment 77603 --> https://bugs.winehq.org/attachment.cgi?id=77603 Patch to reduce heap header size for testing
https://bugs.winehq.org/show_bug.cgi?id=57523
--- Comment #15 from Fabian Maurer dark.shadow4@web.de --- Yes, the patch works. If I can help in any way, please let me know. Truth be told, I don't really understand the issue.
https://bugs.winehq.org/show_bug.cgi?id=57523
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Fixed by SHA1| |1e1c9a60767371b1121b64302c7 | |70810ca9902ce Status|NEW |RESOLVED
--- Comment #16 from Alexandre Julliard julliard@winehq.org --- Fixed by 1e1c9a60767371b1121b64302c770810ca9902ce.
https://bugs.winehq.org/show_bug.cgi?id=57523
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #17 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 10.0-rc3.
-
WineHQ Bugzilla