December 2019 - Wine-Bugs - list.winehq.org

[Bug 48185] Applications crashing with "Not enough memory to run application"

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48185 Hari Saksena <hari.mail(a)protonmail.ch> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |hari.mail(a)protonmail.ch --- Comment #2 from Hari Saksena <hari.mail(a)protonmail.ch> --- Created attachment 65847 --> https://bugs.winehq.org/attachment.cgi?id=65847 Patch for regression b73cfa72edaed76744cfa9a19670bff3684c66ce Also encountering same problem. Reverting whole commit is a bad idea imo as that's breaking a lot of other applications. Did +virtual log and was able to pinpoint error to ullAvailPhys and ullAvailPageFile in GlobalMemoryStatusEx. Reverting just these two to old values seems to correct issue without breaking other applications. Attaching a patch for same. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 48054] gdi32:bitmap fails on Windows XP and 2003

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48054 --- Comment #3 from Gijs Vermeulen <gijsvrm(a)gmail.com> --- Looks like this was fixed by <https://source.winehq.org/git/wine.git/commit/22a4b3fc7e6f48c99934a49643a87…> -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 48216] Add a WineTest size, time and test failures comparison page to the TestBot

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48216 --- Comment #1 from François Gouget <fgouget(a)codeweavers.com> --- Note that tasks can have more than one report. So the (JobId, StepNo, TaskNo) triplet is not enough to fully identify the report to use to build the table. Instead it should be extended with the report name: (JobId, StepNo, TaskNo, ReportBaseName). But currently we never have more than one WineTest run per task because this ensures each WineTest run starts from a clean environment. So the (JobId, StepNo, TaskNo) triplet would be enough. Adding the ReportName would not be hard but may make things harder in case the report name changes for whatever reason. So it may be better to only add it when actually needed. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 48216] New: Add a WineTest size, time and test failures comparison page to the TestBot

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48216 Bug ID: 48216 Summary: Add a WineTest size, time and test failures comparison page to the TestBot Product: Wine-Testbot Version: unspecified Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: unknown Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- This would expend on the data collected for bug 48215. This would be a new page that lets the user pick two (or three or more) WineTest results and put them in the same size, run time and test unit failure tables as the JobDetails page would if they were all from the same job. For each column the user would pick a CommitId and then get a list of the available task reports (for instance "wxppro (32 bit)", "debian10 (32 bit Chinese:China)"). This would allow comparing a VM's 32 and 64 bit results (these run in separate jobs). Or compare a VM's latest result against its past result(s). Or compare the results of running the tests on Windows 8 with those of Windows 10 on the same hardware. Or comparing a WineTest run that timed out with one that did not. In effect this would be similar to bug 15477 but implemented in the TestBot rather than the on test.winehq.org, the advantage being that the TestBot has the infrastructure needed to dynamically build pages. One approach to implement this would be to pour all the report summary information into a big table. Something like (JobId, StepNo, TaskNo, CommitId, TestUnit, Failures, Size, RunTime). However given the performance issues with our current Perl ORM the performance may well be unacceptably slow (see bug 45023). Another approach would be to use the task-level summary files (described in bug 48215) and use a much smaller table to map commit ids to the tasks with relevant results. So this table would just contain (JobId, StepNo, TaskNo, CommitId) and would be used to know which tasks have results for a given commit id, and once the tasks have been selected, where to find the summary file containing the data to put in the tables. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 48215] New: Add size, time and test unit failures tabs to the JobDetails page

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48215 Bug ID: 48215 Summary: Add size, time and test unit failures tabs to the JobDetails page Product: Wine-Testbot Version: unspecified Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: unknown Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- Currently the JobDetails page only provides access to each task's reports and logs. This can make it hard to figure out which test units have failures (bug 47904) and harder still to compare the test unit failures across the tasks. This is ok when testing regular patches since each task typically only runs a single test unit (except for Wine tasks). But for WineTest tasks it would be useful to be able to compare which test units failed for each task. The current workaround is to use test.winehq.org for that, but there too comparing results as its limitations (see bug 15477), and sometimes the reports a rejected which is when it would be nice to be able to dig into them. Similarly it would be quite nice to be able to have a table showing the run time an report size across test unit and tasks. One way to present this would be to have a list of tabs at the top of the JobDetails page: Reports | Unit Failures | Size | Run time The Reports tab would show the usual task reports while the other tabs would show tables with one column per task, one line per test unit and the specified information in the cells. Ideally clicking on a column header would sort the table on that column. This is reasonably easy to do at the JobDetails level: when parsing the reports we can extract all the required information. All that would be required is to save it to a summary file (one line per test unit with the count of failures, size, run time). That would make it easy and quick for JobDetails to retrieve the information and then it's a matter of getting the HTML side in place. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 48175] AION (64 bit) - crashes in crysystem.dll.CryFree() due to high memory pointers allocated

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48175 Fred <dracoanarion(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |dracoanarion(a)gmail.com --- Comment #1 from Fred <dracoanarion(a)gmail.com> --- For information, I tried the wine-staging for Aion 64bit in the last day, and didn't encounter any issue after few hours playing on 2 different machines. This new version of the patch looks pretty good. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 47344] Amazon WorkSpaces Clipboard not accessible

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=47344 Damjan Jovanovic <damjan.jov(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |damjan.jov(a)gmail.com --- Comment #3 from Damjan Jovanovic <damjan.jov(a)gmail.com> --- Nothing is visible there. Maybe it's using the OLE clipboard? Try running it like this: WINEDEBUG='+ole,+clipboard,+timestamp,+tid' wine workspaces.exe &> log.txt and attach log.txt -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 11121] Night Watch crashes on startup (StarForce v3, loader: section SizeOfRawData vs. FileAlignment violation, zero padding of section remainder wipes out code executed at runtime)

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=11121 Anastasius Focht <focht(a)gmx.net> changed: What |Removed |Added ---------------------------------------------------------------------------- URL|http://www.gamershell.com/d |https://www.fileplanet.com/ |ownload_12193.shtml |archive/p-56695/Night-Watch | |-Updated-Demo/download --- Comment #20 from Anastasius Focht <focht(a)gmx.net> --- Hello David, --- quote --- Wow, that's a lot of info! Are you able to test if this still occurs with a newer version of Wine? --- quote --- yes, the issue is still present. I had to search a bit to find sites that still provide the demo download and the no-cd patch. https://web.archive.org/web/20191201155126/https://www.gamesindustry.biz/ar… -> https://www.fileplanet.com/archive/p-56695/Night-Watch-Updated-Demo/download For the no-cd patch use following search query: https://www.google.com/search?q=lonebullet+nightwatch I've snapshotted the patch using Internet Archive as well in case it goes away. --- Wine source: https://source.winehq.org/git/wine.git/blob/HEAD:/dlls/ntdll/virtual.c#l1414 --- snip --- 1414 static NTSTATUS map_image( HANDLE hmapping, ACCESS_MASK access, int fd, int top_down, unsigned short zero_bits_64, 1415 pe_image_info_t *image_info, int shared_fd, BOOL removable, PVOID *addr_ptr ) 1416 { ... 1510 /* map all the sections */ 1511 1512 for (i = pos = 0; i < nt->FileHeader.NumberOfSections; i++, sec++) 1513 { 1514 static const SIZE_T sector_align = 0x1ff; 1515 SIZE_T map_size, file_start, file_size, end; 1516 1517 if (!sec->Misc.VirtualSize) 1518 map_size = ROUND_SIZE( 0, sec->SizeOfRawData ); 1519 else 1520 map_size = ROUND_SIZE( 0, sec->Misc.VirtualSize ); 1521 1522 /* file positions are rounded to sector boundaries regardless of OptionalHeader.FileAlignment */ 1523 file_start = sec->PointerToRawData & ~sector_align; 1524 file_size = (sec->SizeOfRawData + (sec->PointerToRawData & sector_align) + sector_align) & ~sector_align; 1525 if (file_size > map_size) file_size = map_size; ... 1566 TRACE_(module)( "mapping section %.8s at %p off %x size %x virt %x flags %x\n", 1567 sec->Name, ptr + sec->VirtualAddress, 1568 sec->PointerToRawData, sec->SizeOfRawData, 1569 sec->Misc.VirtualSize, sec->Characteristics ); 1570 1571 if (!sec->PointerToRawData || !file_size) continue; 1572 1573 /* Note: if the section is not aligned properly map_file_into_view will magically 1574 * fall back to read(), so we don't need to check anything here. 1575 */ 1576 end = file_start + file_size; 1577 if (sec->PointerToRawData >= st.st_size || 1578 end > ((st.st_size + sector_align) & ~sector_align) || 1579 end < file_start || 1580 map_file_into_view( view, fd, sec->VirtualAddress, file_size, file_start, 1581 VPROT_COMMITTED | VPROT_READ | VPROT_WRITECOPY, 1582 removable ) != STATUS_SUCCESS) 1583 { 1584 ERR_(module)( "Could not map section %.8s, file probably truncated\n", sec->Name ); 1585 goto error; 1586 } 1587 1588 if (file_size & page_mask) 1589 { 1590 end = ROUND_SIZE( 0, file_size ); 1591 if (end > map_size) end = map_size; 1592 TRACE_(module)("clearing %p - %p\n", 1593 ptr + sec->VirtualAddress + file_size, 1594 ptr + sec->VirtualAddress + end ); 1595 memset( ptr + sec->VirtualAddress + file_size, 0, end - file_size ); 1596 } 1597 } ... --- snip --- Running the patched executable on top of demo install: --- snip --- $ pwd /home/focht/.wine/drive_c/Program Files (x86)/NightWatch/Run $ WINEDEBUG=+seh,+relay wine ./NWatch.exe >>log.txt 2>&1 ... 0040:Call KERNEL32.OutputDebugStringA(00ab0bd8 "WORLD VALID: no ADRESS 0\n") ret=00496b93 0040:warn:seh:OutputDebugStringA "WORLD VALID: no ADRESS 0\n" 0040:Ret KERNEL32.OutputDebugStringA() retval=00000000 ret=00496b93 0040:trace:seh:raise_exception code=c0000005 flags=0 addr=0x90a810 ip=0090a810 tid=0040 0040:trace:seh:raise_exception info[0]=00000001 0040:trace:seh:raise_exception info[1]=00000000 0040:trace:seh:raise_exception eax=00000000 ebx=593a3a88 ecx=593a3a88 edx=0000005f esi=593a3a88 edi=00000000 0040:trace:seh:raise_exception ebp=00000000 esp=0032f970 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00010246 0040:trace:seh:call_stack_handlers calling handler at 0x964e18 code=c0000005 flags=0 0040:trace:seh:call_stack_handlers handler at 0x964e18 returned 1 ... Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x0090a810). --- snip --- $ sha1sum night* 826a6318a016d86fab6172eab23b4a15b2b92f54 night_watch_demo_english.exe c485bd99eb076733a3c96c02716f2c4c760aa851 night-watch-v101-english-no-cdfixed-exe.rar $ du -sh night* 458M night_watch_demo_english.exe 2.1M night-watch-v101-english-no-cdfixed-exe.rar $ wine --version wine-4.21 Regards -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 48202] RedDragon app crashes on startup

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48202 --- Comment #5 from joaopa <jeremielapuree(a)yahoo.fr> --- Try vanilla wine-4.21. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019

[Bug 48214] New: RPGmaker input does not work after moving window

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=48214 Bug ID: 48214 Summary: RPGmaker input does not work after moving window Product: Wine Version: 4.0.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-dinput Assignee: wine-bugs(a)winehq.org Reporter: ilkka.prusi(a)gmail.com Distribution: --- In some games made with RPGMaker VX ACE the input does not work after you have moved window in desktop (input focus gone and back again). Symptom is that game character does not move, menu is not opened on pressing esc and so forth. This is repeatable on Debian bullseye with wine 4.0.2, Wayland + Gnome on x86-64 (Linux 5.4.1). To repeat simply grab the window with mouse, move it, click inside window and try keyboard input again. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 1, 2019