https://bugs.winehq.org/show_bug.cgi?id=49919
Bug ID: 49919 Summary: Regression: Logos Bible Software wineserver crash entering text in passage guide search field Product: Wine Version: 5.18 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: johnpgoodman@gmail.com Distribution: ---
Created attachment 68295 --> https://bugs.winehq.org/attachment.cgi?id=68295 Crash from entering text in passage guide.
Using wine 5.18 staging I opened a new passage guide from the guides menu. Entering references such as jn 1, rm 2 etc and pressing enter should result in the guide populating. This happens about half the time. However many times it crashes. Output attached. I've tried entering text in other guides but it appears to be only the passage guide affected. The output is different in the 2 attachments. I believe this to be a regression as it hasn't happened to me before today when I first used 5.18 staging.
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #1 from m0rvj johnpgoodman@gmail.com --- Created attachment 68296 --> https://bugs.winehq.org/attachment.cgi?id=68296 same passage guide crash
This time the terminal output differs. Same experience.
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #2 from m0rvj johnpgoodman@gmail.com --- Gone back to 5.17 and can't cause this so assuming the regression is in 5.18.
https://bugs.winehq.org/show_bug.cgi?id=49919
Daniel Ribeiro da Silva ferion11@hotmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ferion11@hotmail.com
--- Comment #3 from Daniel Ribeiro da Silva ferion11@hotmail.com --- I was able to reproduce this only in the 64-bit wine version.
https://bugs.winehq.org/show_bug.cgi?id=49919
Nick nick.andrewes@phonecoop.coop changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |nick.andrewes@phonecoop.coo | |p
--- Comment #4 from Nick nick.andrewes@phonecoop.coop --- For me, the first time, it worked. The second time it crashed. With 64bit and 5.18-staging. Then it worked the first two times and crashed the third. For some reason it won't let me run Logos in a terminal, or I would attach logs. Many thanks.
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #5 from Daniel Ribeiro da Silva ferion11@hotmail.com --- git bisect start # good: [01bacebba418bf6f58a644b1aab41be215bd200a] Release 5.17. git bisect good 01bacebba418bf6f58a644b1aab41be215bd200a # bad: [9a6e5b23293fbad3bbdcd52007402a3b9a1cb99d] Release 5.18. git bisect bad 9a6e5b23293fbad3bbdcd52007402a3b9a1cb99d # skip: [c460d4ff35e69052a917a8d75f97f39d36747e52] sxs: Return correct path from IAssemblyCache::QueryAssemblyInfo(). git bisect skip c460d4ff35e69052a917a8d75f97f39d36747e52 # skip: [b936df428dc3f30f03c854abac8860888a983d67] wmc: Add Dxgi facility. git bisect skip b936df428dc3f30f03c854abac8860888a983d67 # skip: [158471d676e799d3d23a356b442ecc4afd2ade57] conhost: Delay creating input thread in Unix mode. git bisect skip 158471d676e799d3d23a356b442ecc4afd2ade57 # good: [da4dc79db93309ede68ca8baa82716f51f55fa0d] oleacc/tests: Use wide string constants. git bisect good da4dc79db93309ede68ca8baa82716f51f55fa0d # bad: [96dcee64307b63ed3c07d13832fee15c3dc21ff1] server: Implement object name information for registry keys. git bisect bad 96dcee64307b63ed3c07d13832fee15c3dc21ff1 # skip: [a9722145a86ce3392630d469308a9dd731e62e79] ntdll: Fix the spelling of the arch_context_parameters variable. git bisect skip a9722145a86ce3392630d469308a9dd731e62e79 # bad: [d28c7938c704fe4dcfc8920db494a7175a46fe6b] dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif. git bisect bad d28c7938c704fe4dcfc8920db494a7175a46fe6b # good: [43115a74f94293994904c9857996e9576dc3a8eb] setupapi/tests: Add explicit field for inf contents size. git bisect good 43115a74f94293994904c9857996e9576dc3a8eb # good: [d26ad6db7d5a70c49d051af403d9befe2be7c8f8] ole32/clipboard: Use wide string constants. git bisect good d26ad6db7d5a70c49d051af403d9befe2be7c8f8 # good: [4e11e6e045edd514d553d32ef797cecf98d9a203] crypt32: Set key context if PKCS12_NO_PERSIST_KEY is passed, otherwise set key provider info. git bisect good 4e11e6e045edd514d553d32ef797cecf98d9a203 # first bad commit: [d28c7938c704fe4dcfc8920db494a7175a46fe6b] dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif.
d28c7938c704fe4dcfc8920db494a7175a46fe6b is the first bad commit commit d28c7938c704fe4dcfc8920db494a7175a46fe6b Author: Myah Caron qsniyg@protonmail.com Date: Thu Sep 17 09:04:02 2020 +0000
dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=49850 Signed-off-by: Myah Caron qsniyg@protonmail.com Signed-off-by: Michael Stefaniuc mstefani@winehq.org Signed-off-by: Alexandre Julliard julliard@winehq.org
dlls/dmstyle/style.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #6 from Daniel Ribeiro da Silva ferion11@hotmail.com --- The d28c7938c704fe4dcfc8920db494a7175a46fe6b definitely causes the LogosBible to crash when doing this operation. But when I tried on the last git with: git revert -n d28c7938c704fe4dcfc8920db494a7175a46fe6b
I'm still getting a crash with the error: 0100:err:eventlog:ReportEventW L"Application: Logos.exe\nFramework Version: v4.0.30319\nDescription: The process was terminated due to an internal error in the .NET Runtime at IP 0000000000DEADE4 (0000000000DE0000) with exit code 80131506.\n"
Maybe, I will have to do the bisect again including the revert patch.
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #7 from Daniel Ribeiro da Silva ferion11@hotmail.com --- Making the second bisect using the patch from the first regression, I fall into the block that wine doesn't run:
git bisect start # good: [d28c7938c704fe4dcfc8920db494a7175a46fe6b] dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif. git bisect good d28c7938c704fe4dcfc8920db494a7175a46fe6b # bad: [9a6e5b23293fbad3bbdcd52007402a3b9a1cb99d] Release 5.18. git bisect bad 9a6e5b23293fbad3bbdcd52007402a3b9a1cb99d # bad: [810873eb7561e147dff871aa20315e930d1f01bb] kernel32: Move Beep implementation to conhost. git bisect bad 810873eb7561e147dff871aa20315e930d1f01bb # skip: [52f6a56a1aa450c893e19a0e63a90c60626d3dcf] server: Introduce IOCTL_CONDRV_SETUP_INPUT ioctl. git bisect skip 52f6a56a1aa450c893e19a0e63a90c60626d3dcf # skip: [b936df428dc3f30f03c854abac8860888a983d67] wmc: Add Dxgi facility. git bisect skip b936df428dc3f30f03c854abac8860888a983d67 # good: [305aaec8773a7cbd0fdd4d12c5e0210e9cb5f33d] evr/mixer: Fix error handling in output types collection helper. git bisect good 305aaec8773a7cbd0fdd4d12c5e0210e9cb5f33d # skip: [9dcdd89a9241cd0066b67c9133e9c1fa5e5c1d0f] kernel32: Add string for DXGI_INVALID_CALL error. git bisect skip 9dcdd89a9241cd0066b67c9133e9c1fa5e5c1d0f # skip: [5db533d39d0a924a562586bc82a12cec7f9b8884] kernel32: Add string for DXGI_ERROR_NOT_CURRENTLY_AVAILABLE error. git bisect skip 5db533d39d0a924a562586bc82a12cec7f9b8884 # bad: [bbae35f0fb04ea7efb8e1d6e5535e42715ae7766] winex11.drv: Support finding a full display mode with specified orientation. git bisect bad bbae35f0fb04ea7efb8e1d6e5535e42715ae7766 # skip: [aecf53d0f745a65c15e0f1abbaf27af0d7c0a486] conhost: Use CP_UNIXCP in Unix mode. git bisect skip aecf53d0f745a65c15e0f1abbaf27af0d7c0a486 # skip: [c6909e52543b62dfcd5b183014e6eae27b69f3fc] conhost: Terminate current read request when input when reading tty input fails. git bisect skip c6909e52543b62dfcd5b183014e6eae27b69f3fc # skip: [05d9df73d12d202d51b784f1d5eadc55f8448f16] winex11.drv: If the end of the sequence is '\r\n', there is no need to add extra '\r'. git bisect skip 05d9df73d12d202d51b784f1d5eadc55f8448f16 # skip: [158471d676e799d3d23a356b442ecc4afd2ade57] conhost: Delay creating input thread in Unix mode. git bisect skip 158471d676e799d3d23a356b442ecc4afd2ade57 # good: [3541deb0c3a64d3474a96728d259907180fd7cf6] server: Don't overflow if timeout doesn't fit into int range in get_next_timeout. git bisect good 3541deb0c3a64d3474a96728d259907180fd7cf6 # skip: [6bcc1b9225efe028319bbc64ce6b42a03e4d7ec7] ntdll: Pass RTL_USER_PROCESS_PARAMETERS to get_initial_console. git bisect skip 6bcc1b9225efe028319bbc64ce6b42a03e4d7ec7 # good: [6d7037c9ff1219ae959e365d8a95a3fceb57ca87] crypt32: Move GnuTLS-specific helpers into the #ifdef. git bisect good 6d7037c9ff1219ae959e365d8a95a3fceb57ca87 # skip: [0957f9302ca40bd85e092ed335d3fd10ab9bf4da] ntdll: Fill console size in get_initial_console. git bisect skip 0957f9302ca40bd85e092ed335d3fd10ab9bf4da # good: [148d3aa461126ab6e77e36e33493f87d539cd6ec] user32: Move PNG support to a new Unix library. git bisect good 148d3aa461126ab6e77e36e33493f87d539cd6ec # skip: [5ad25c293b6a7b5cd5dea43f29f6fa94d8302722] user32: Use wide character string literals. git bisect skip 5ad25c293b6a7b5cd5dea43f29f6fa94d8302722 # skip: [fc408f3d4376722dbe173495be10ef2ce34874d1] conhost: Force using relative cursor positioning in update_read_output. git bisect skip fc408f3d4376722dbe173495be10ef2ce34874d1 # skip: [6456973f0a64d326bb54da4675310caffc2588f1] user32: Use wide character string literals for system parameter entries. git bisect skip 6456973f0a64d326bb54da4675310caffc2588f1 # skip: [4eb3ab95c75eb4eb33c01d22affdfd303590d87a] conhost: Add missing LeaveCriticalSection. git bisect skip 4eb3ab95c75eb4eb33c01d22affdfd303590d87a # bad: [de508759c67543e3bd21ee6a0873b06aa4014418] winevulkan: Update to VK spec version 1.2.154. git bisect bad de508759c67543e3bd21ee6a0873b06aa4014418 # skip: [38c91043f626bd6755185bc62ae85251c9e88207] vbscript: Fix the spelling of the delimiter* variables. git bisect skip 38c91043f626bd6755185bc62ae85251c9e88207 # skip: [54e117018cd4cc58c258da92686bfad13946bde2] kernelbase: Use conhost to handle Unix consoles. git bisect skip 54e117018cd4cc58c258da92686bfad13946bde2
https://bugs.winehq.org/show_bug.cgi?id=49919
qsniyg qsniyg@mail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |qsniyg@mail.com
--- Comment #8 from qsniyg qsniyg@mail.com --- (In reply to Daniel Ribeiro da Silva from comment #5)
# first bad commit: [d28c7938c704fe4dcfc8920db494a7175a46fe6b] dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif.
d28c7938c704fe4dcfc8920db494a7175a46fe6b is the first bad commit commit d28c7938c704fe4dcfc8920db494a7175a46fe6b Author: Myah Caron qsniyg@protonmail.com Date: Thu Sep 17 09:04:02 2020 +0000
dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=49850 Signed-off-by: Myah Caron <qsniyg@protonmail.com> Signed-off-by: Michael Stefaniuc <mstefani@winehq.org> Signed-off-by: Alexandre Julliard <julliard@winehq.org>
dlls/dmstyle/style.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
At least from the log provided by m0rvj, I have literally no idea how this would break it, as IDirectMusicStyle8Impl_GetMotif isn't even called (it should show a fixme).
The only possible explanation I can think of is that there's some kind of issue with the debugstr_w thing, but I highly doubt it. Why would logos even use directmusic?
Considering m0rvj and Nick mentioned that the error occurs inconsistently, perhaps the bisect was incorrect?
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #9 from Daniel Ribeiro da Silva ferion11@hotmail.com --- (In reply to qsniyg from comment #8)
(In reply to Daniel Ribeiro da Silva from comment #5)
# first bad commit: [d28c7938c704fe4dcfc8920db494a7175a46fe6b] dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif.
d28c7938c704fe4dcfc8920db494a7175a46fe6b is the first bad commit commit d28c7938c704fe4dcfc8920db494a7175a46fe6b Author: Myah Caron qsniyg@protonmail.com Date: Thu Sep 17 09:04:02 2020 +0000
dmstyle: Return S_FALSE for IDirectMusicStyle8::GetMotif. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=49850 Signed-off-by: Myah Caron <qsniyg@protonmail.com> Signed-off-by: Michael Stefaniuc <mstefani@winehq.org> Signed-off-by: Alexandre Julliard <julliard@winehq.org>
dlls/dmstyle/style.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
At least from the log provided by m0rvj, I have literally no idea how this would break it, as IDirectMusicStyle8Impl_GetMotif isn't even called (it should show a fixme).
The only possible explanation I can think of is that there's some kind of issue with the debugstr_w thing, but I highly doubt it. Why would logos even use directmusic?
Considering m0rvj and Nick mentioned that the error occurs inconsistently, perhaps the bisect was incorrect?
I'm sorry, I didn't have updated the report on this subject, but I can reproduce the bug consistently, and after applying the regression patch, the bug is no longer present. The git bisect is correct.
LogosBible uses audio to reproduce the pronunciation in the original languages of ancient texts, and other similar uses. Which isn't a serious issue if it doesn't work for some time, but preventing the use of the "Passage Guide" is something that must be solved.
I don't think that your game without sound, as you yourself reported, can be considerably playable. And that you should look for a real implementation, that really solves your problem, and propose only a patch that does this and not one that do only part of it. Therefore, I believe that LogosBible would have priority over this fix return value. But if you have made progress on your side, and your game is playable with sound, then I will look for an implementation or stub for this method.
https://bugs.winehq.org/show_bug.cgi?id=49919
Gijs Vermeulen gijsvrm@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression Ever confirmed|0 |1 Status|UNCONFIRMED |NEW Summary|Regression: Logos Bible |Logos Bible Software |Software wineserver crash |crashes when entering text |entering text in passage |in passage guide search |guide search field |field
--- Comment #10 from Gijs Vermeulen gijsvrm@gmail.com --- Confirming since multiple users can confirm. Rewording title a little and adding 'regression' keyword.
(In reply to Daniel Ribeiro da Silva from comment #9)
-snip-
What you could do is attach (please don't paste) the full console output from a run where the issue was reproduced.
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #11 from qsniyg qsniyg@mail.com --- (In reply to Daniel Ribeiro da Silva from comment #9)
I'm sorry, I didn't have updated the report on this subject, but I can reproduce the bug consistently, and after applying the regression patch, the bug is no longer present. The git bisect is correct.
LogosBible uses audio to reproduce the pronunciation in the original languages of ancient texts, and other similar uses. Which isn't a serious issue if it doesn't work for some time, but preventing the use of the "Passage Guide" is something that must be solved.
I don't think that your game without sound, as you yourself reported, can be considerably playable. And that you should look for a real implementation, that really solves your problem, and propose only a patch that does this and not one that do only part of it. Therefore, I believe that LogosBible would have priority over this fix return value. But if you have made progress on your side, and your game is playable with sound, then I will look for an implementation or stub for this method.
Could you attach a WINEDEBUG=+module,+initdll,+seh debug trace then?
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #12 from Daniel Ribeiro da Silva ferion11@hotmail.com --- (In reply to qsniyg from comment #11)
(In reply to Daniel Ribeiro da Silva from comment #9)
...
Could you attach a WINEDEBUG=+module,+initdll,+seh debug trace then?
The LogosBible 8.17.0.0011 come out the last weak, so I took the opportunity to update and test everything again.
The new LogosBible 8.17.0.0011 don't have issues with d28c7938c704fe4dcfc8920db494a7175a46fe6b anymore, only the last 8.16.0.0002 have it. So qsniyg, you don't have to worry about it anymore, thanks. And for this part, the users should just update.
For the second regression, the bug is still there, testing with vanilla-wine-5.19. I will post the log with the same WINEDEBUG variable set (maybe useful).
I'm still continuing the git bisect for the second regression (now the only one), but doing it manually because it's in the commits that make WINE not even be able to run winecfg.
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #13 from Daniel Ribeiro da Silva ferion11@hotmail.com --- Created attachment 68479 --> https://bugs.winehq.org/attachment.cgi?id=68479 Logos 8.17 with wine-vanilla-5.19
Logos 8.17 with wine-vanilla-5.19 using WINEDEBUG=+module,+initdll,+seh and crashing when accessing the "Passage Guide"
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #14 from Daniel Ribeiro da Silva ferion11@hotmail.com --- ok! Manual bisect is over, i guess:
The first commit with the issue is: https://source.winehq.org/git/wine.git/commit/9cc92365560f19c2fd2b9796f79aa7... 2020-09-21 Alexandre Julliard user32: Build with msvcrt.
But i'm also using the 2 patches of the commits: https://source.winehq.org/git/wine.git/commit/b16f552e195048f434d486a084fad0... 2020-09-22 Rémi Bernon gdi32: Use explicit CDECL in window_surface_funcs.
and
https://source.winehq.org/git/wine.git/commit/a404eb7272db53883cc6d74db044a5... 2020-09-22 Rémi Bernon gdi32: Use explicit CDECL in gdi_image_bits.
so the 9cc92365560f19c2fd2b9796f79aa75e02381bb1 can execute applications (otherwise it fails to run even winecfg).
Then it, the issue, can be on the other 2 commits too.
https://bugs.winehq.org/show_bug.cgi?id=49919
--- Comment #15 from Daniel Ribeiro da Silva ferion11@hotmail.com --- ok! I was able to test reverting only the a404eb7, and despite being able to run, the bug was still present, so it's one of these 2 commits that introduced the issue: https://source.winehq.org/git/wine.git/commit/9cc92365560f19c2fd2b9796f79aa7... 2020-09-21 Alexandre Julliard user32: Build with msvcrt.
and the (to make WINE run after 9cc9236): https://source.winehq.org/git/wine.git/commit/b16f552e195048f434d486a084fad0... 2020-09-22 Rémi Bernon gdi32: Use explicit CDECL in window_surface_funcs.
https://bugs.winehq.org/show_bug.cgi?id=49919
Anastasius Focht focht@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download URL| |https://web.archive.org/web | |/20210209172851/https://dow | |nloads.logoscdn.com/LBS8/In | |staller/8.7.0.0039/Logos-x8 | |6.msi CC| |focht@gmx.net Summary|Logos Bible Software |Logos 8 Bible Software |crashes when entering text |crashes when entering text |in passage guide search |in passage guide search |field |field
--- Comment #16 from Anastasius Focht focht@gmx.net --- Hello folks,
I can't reproduce the problem, the search works fine here.
Stable download link via Internet Archive:
https://web.archive.org/web/20210209172851/https://downloads.logoscdn.com/LB...
Prerequisites:
* 'winetricks -q dotnet472' * 'winetricks -q arial' (work around crash in 'LogosCEF.exe' browser process)
$ sha1sum Logos-x86.msi 659900692c79c0df046f413eb33e1d729487fe19 Logos-x86.msi
$ du -sh Logos-x86.msi 187M Logos-x86.msi
$ wine --version wine-6.2
Regards
https://bugs.winehq.org/show_bug.cgi?id=49919
Frank franksauer@cox.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |franksauer@cox.net
--- Comment #17 from Frank franksauer@cox.net --- I don't personally recall having this issue, but I just tested it again on Wine 7.0 RC5 Staging. I have successfully entered multiple Verses into the Passage Guide and did not experience a single crash.
Can others verify that this is no longer an issue?
https://bugs.winehq.org/show_bug.cgi?id=49919
m0rvj johnpgoodman@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED
--- Comment #18 from m0rvj johnpgoodman@gmail.com --- Yes, I can't say when it was fixed but I haven't seen this bug in ages. I say mark as fixed.
https://bugs.winehq.org/show_bug.cgi?id=49919
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #19 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 7.0-rc6.