March 2023 - Wine-Devel - list.winehq.org

Re: Should we stop linking to FAudio for our XAudio implementation?

by Esme Povirk

> Unfortunately FAudio cannot, counterintuitively, link to LGPL code, > because it needs to be able to run on embedded platforms, which (for > reasons I do not understand) cannot ship any LGPL code. It's quite simple. If a platform (such as Nintendo Switch) does not allow users to sideload their own code on it, it's not possible to comply with the terms of the LGPL while shipping LGPL code on that platform, because the user cannot install their own modified version of the LGPL library.

March 29, 2023

Re: Should we stop linking to FAudio for our XAudio implementation?

by Rémi Bernon

On 3/29/23 17:09, Dean Greer wrote: > Not only did FAudio link to gstreamer but prior to this it linked to FFmpeg. > > The gstreamer backend was removed in 22.02, FAudio, there assumption seems > to be that wine wouldn’t require the gstreamer backend and will be using > MediaFoundation… > Indeed, BTW it could also use that backend to get through to some GStreamer audio resampler plugin on Wine, although that adds a bit of indirection (maybe not so much). -- Rémi Bernon <rbernon(a)codeweavers.com>

March 29, 2023

Re: Should we stop linking to FAudio for our XAudio implementation?

by Dean Greer

Not only did FAudio link to gstreamer but prior to this it linked to FFmpeg. The gstreamer backend was removed in 22.02, FAudio, there assumption seems to be that wine wouldn’t require the gstreamer backend and will be using MediaFoundation… On Wed, Mar 29, 2023 at 2:52 AM Rémi Bernon <rbernon(a)codeweavers.com> wrote: > On 3/29/23 02:31, Zebediah Figura wrote: > > Hi all, > > > > This may seem something of a controversial question, but I am concerned > > that FAudio is not a good fit for Wine, mainly because of its licensing > > considerations. > > > > Recently I found a bug in FAudio wherein it does not properly handle > > unaligned buffers. While in the process of drafting a fix, I was stymied > > multiple times by its rather broken resampler, and when I sought to try > > to fix that, I found that it was a rather difficult endeavour, which led > > me to propose replacing it with an external library such as GStreamer. > > Unfortunately FAudio cannot, counterintuitively, link to LGPL code, > > because it needs to be able to run on embedded platforms, which (for > > reasons I do not understand) cannot ship any LGPL code. > > > > While I can to some degree respect the decision of FAudio to target > > platforms that forbid copyleft licensing, I do not think this aligns > > well with the interests of the Wine project—a project which specifically > > chose copyleft licensing. More personally, I do not want to have to fix > > the FAudio resampler or implement another resampler with a non-copyleft > > license (and I am not even particularly sure that I can, given my > > familiarity with resamplers in LGPL projects such as GStreamer and Wine). > > > > Besides this, it is probably worth noting that we do not seem to gain > > much from continuing to rely on FAudio. There would be the usual benefit > > of relying on upstream development, but development on FAudio seems very > > scarce, especially when restricted to parts relevant to Wine, and much > > of that is from Wine contributors anyway. (And, of course, due to our > > bundling practices, there is hardly the usual benefit of size or > > modularity in linking to external libraries.) > > > > Of course we would be well served by simply importing the current > > version of FAudio; I do not mean to propose that we reimplement XAudio > > entirely, but rather begin to modify it in ways that would not be > > acceptable upstream. > > > > This would mean replacing the resampler, as well as perhaps also the > > mixer, and probably with the ones from GStreamer, which is already a > > dependency of ours, and which has a very featureful resampler and mixer, > > and a much higher rate of development overall. > > > > Do others in the Wine community have thoughts or opinions on this matter? > > > > ἔρρωσθε, > > Zeb > > > > Fwiw FAudio has linked with GStreamer in the past, conditionally at least. > > -- > Rémi Bernon <rbernon(a)codeweavers.com> > > > >

March 29, 2023

Re: Should we stop linking to FAudio for our XAudio implementation?

by Rémi Bernon

On 3/29/23 02:31, Zebediah Figura wrote: > Hi all, > > This may seem something of a controversial question, but I am concerned > that FAudio is not a good fit for Wine, mainly because of its licensing > considerations. > > Recently I found a bug in FAudio wherein it does not properly handle > unaligned buffers. While in the process of drafting a fix, I was stymied > multiple times by its rather broken resampler, and when I sought to try > to fix that, I found that it was a rather difficult endeavour, which led > me to propose replacing it with an external library such as GStreamer. > Unfortunately FAudio cannot, counterintuitively, link to LGPL code, > because it needs to be able to run on embedded platforms, which (for > reasons I do not understand) cannot ship any LGPL code. > > While I can to some degree respect the decision of FAudio to target > platforms that forbid copyleft licensing, I do not think this aligns > well with the interests of the Wine project—a project which specifically > chose copyleft licensing. More personally, I do not want to have to fix > the FAudio resampler or implement another resampler with a non-copyleft > license (and I am not even particularly sure that I can, given my > familiarity with resamplers in LGPL projects such as GStreamer and Wine). > > Besides this, it is probably worth noting that we do not seem to gain > much from continuing to rely on FAudio. There would be the usual benefit > of relying on upstream development, but development on FAudio seems very > scarce, especially when restricted to parts relevant to Wine, and much > of that is from Wine contributors anyway. (And, of course, due to our > bundling practices, there is hardly the usual benefit of size or > modularity in linking to external libraries.) > > Of course we would be well served by simply importing the current > version of FAudio; I do not mean to propose that we reimplement XAudio > entirely, but rather begin to modify it in ways that would not be > acceptable upstream. > > This would mean replacing the resampler, as well as perhaps also the > mixer, and probably with the ones from GStreamer, which is already a > dependency of ours, and which has a very featureful resampler and mixer, > and a much higher rate of development overall. > > Do others in the Wine community have thoughts or opinions on this matter? > > ἔρρωσθε, > Zeb > Fwiw FAudio has linked with GStreamer in the past, conditionally at least. -- Rémi Bernon <rbernon(a)codeweavers.com>

March 29, 2023

Should we stop linking to FAudio for our XAudio implementation?

by Zebediah Figura

Hi all, This may seem something of a controversial question, but I am concerned that FAudio is not a good fit for Wine, mainly because of its licensing considerations. Recently I found a bug in FAudio wherein it does not properly handle unaligned buffers. While in the process of drafting a fix, I was stymied multiple times by its rather broken resampler, and when I sought to try to fix that, I found that it was a rather difficult endeavour, which led me to propose replacing it with an external library such as GStreamer. Unfortunately FAudio cannot, counterintuitively, link to LGPL code, because it needs to be able to run on embedded platforms, which (for reasons I do not understand) cannot ship any LGPL code. While I can to some degree respect the decision of FAudio to target platforms that forbid copyleft licensing, I do not think this aligns well with the interests of the Wine project—a project which specifically chose copyleft licensing. More personally, I do not want to have to fix the FAudio resampler or implement another resampler with a non-copyleft license (and I am not even particularly sure that I can, given my familiarity with resamplers in LGPL projects such as GStreamer and Wine). Besides this, it is probably worth noting that we do not seem to gain much from continuing to rely on FAudio. There would be the usual benefit of relying on upstream development, but development on FAudio seems very scarce, especially when restricted to parts relevant to Wine, and much of that is from Wine contributors anyway. (And, of course, due to our bundling practices, there is hardly the usual benefit of size or modularity in linking to external libraries.) Of course we would be well served by simply importing the current version of FAudio; I do not mean to propose that we reimplement XAudio entirely, but rather begin to modify it in ways that would not be acceptable upstream. This would mean replacing the resampler, as well as perhaps also the mixer, and probably with the ones from GStreamer, which is already a dependency of ours, and which has a very featureful resampler and mixer, and a much higher rate of development overall. Do others in the Wine community have thoughts or opinions on this matter? ἔρρωσθε, Zeb

March 29, 2023

Re: TestBot News: Missing dlls and VM updates

by Bernhard Übelacker

Am 19.03.23 um 13:29 schrieb Bernhard Übelacker: > Am 12.01.23 um 14:34 schrieb Francois Gouget: >> On Thu, 12 Jan 2023, Francois Gouget wrote: >> [...] >>>> And does it happen on a specific wine test? >>> >>> It's enough to run the amstream:amstream test. >> >> More precisely it crashes in the gst_init_check() call! >> That is, the test works till line 464: >> >> 464 check_interface(filter, &IID_IMediaSeeking, FALSE); >> >> And the crash happens in dlls/winegstreamer/wg_parser.c in >> init_gstreamer_once(): >> >> 1673 if (!gst_init_check(&argc, &argv, &err)) >> >> I guess the good news is that I can crash the X server with: >> >> gst-play-1.0 no-such-file.avi >> >> And it turns out to be caused by the Xv support so a workaround is to >> uninstall the gstreamer1.0-vaapi package. Given that it's pretty basic >> functionality that's broken (though maybe it's a bad interaction with >> QXL) there's hope that it will be fixed in time. > > Hello Francois, > sorry for the delay. I have tried to reproduce and found > the X-server hitting an assert. > > Unfortunately I found in the upstream Xserver tracker > no exact matching report and no fix. > > Therefore I have opened following issue: > https://gitlab.freedesktop.org/xorg/xserver/-/issues/1534 > > There is also a patch that would avoid hitting the assert, > instead return DRI2Authenticate with FALSE. > Maybe having that applied to a locally built xserver-xorg-core package > could be another workaround instead of uninstalling gstreamer1.0-vaapi. > > Unfortunately the QXL driver package was removed from Debian testing > most part of the last year, maybe therefore got less tested: > https://tracker.debian.org/pkg/xserver-xorg-video-qxl > > Kind regards, > Bernhard Hello Francois, I wondered how this worked in Bullseye/stable and reported another issue to libva. https://github.com/intel/libva/issues/700 Let's see how the reactions are. Kind regards, Bernhard

March 19, 2023

Re: TestBot News: Missing dlls and VM updates

by Bernhard Übelacker

Am 12.01.23 um 14:34 schrieb Francois Gouget: > On Thu, 12 Jan 2023, Francois Gouget wrote: > [...] >>> And does it happen on a specific wine test? >> >> It's enough to run the amstream:amstream test. > > More precisely it crashes in the gst_init_check() call! > That is, the test works till line 464: > > 464 check_interface(filter, &IID_IMediaSeeking, FALSE); > > And the crash happens in dlls/winegstreamer/wg_parser.c in > init_gstreamer_once(): > > 1673 if (!gst_init_check(&argc, &argv, &err)) > > I guess the good news is that I can crash the X server with: > > gst-play-1.0 no-such-file.avi > > And it turns out to be caused by the Xv support so a workaround is to > uninstall the gstreamer1.0-vaapi package. Given that it's pretty basic > functionality that's broken (though maybe it's a bad interaction with > QXL) there's hope that it will be fixed in time. Hello Francois, sorry for the delay. I have tried to reproduce and found the X-server hitting an assert. Unfortunately I found in the upstream Xserver tracker no exact matching report and no fix. Therefore I have opened following issue: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1534 There is also a patch that would avoid hitting the assert, instead return DRI2Authenticate with FALSE. Maybe having that applied to a locally built xserver-xorg-core package could be another workaround instead of uninstalling gstreamer1.0-vaapi. Unfortunately the QXL driver package was removed from Debian testing most part of the last year, maybe therefore got less tested: https://tracker.debian.org/pkg/xserver-xorg-video-qxl Kind regards, Bernhard

March 19, 2023

Wine staging 8.4 release

by Alistair Leslie-Hughes

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.4 (532 patches are applied to wine vanilla) The patchinstall.sh script is no longer available and users should use the staging/patchinstall.py instead. Upstreamed (Either directly from staging or fixed with a similar patch). * None Added: * None Updated: * None Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

March 18, 2023

wine-7.0.2-rc1 Available

by Michael Stefaniuc

Hello, I have pushed wine-7.0.2-rc1 (f25a27c2e57c9f8c061746a813ace405fbd39073) to https://github.com/mstefani/wine-stable/tree/oldstable https://people.redhat.com/mstefani/wine/stable/ And created the draft merge request https://gitlab.winehq.org/wine/wine/-/merge_requests/2402 Nothing fancy so here we go: What's new in this release: - Various bug fixes ---------------------------------------------------------------- Bugs fixed in 7.0.2 (total 43): - #36711 Sid Meier's Civilization IV crashes while loading saved game (GOG.com complete, Colonization, Warlords, Beyond the Sword)('winetricks -q msxml3' is a workaround) - #39346 Lego Rock Raiders crashes on startup (d3drm_device_init doesn't handle device version 3 case) - #40444 Find doesn't jump to a found item in WinSCP when editing a file - #42751 Multiple applications crash on unimplemented function msvcr{100,110}.dll.?_Schedule(a)_StructuredTaskCollection@details(a)Concurrency@@QAEXPAV_UnrealizedChore(a)23@@Z (Gas Guzzlers Combat Carnage, Clip Studio Paint, Cube World, Intel ACAT) - #42857 Can't select a fixture in FreeStyler's fixture editor - #44814 cygwin's bsdtar.exe needs a better ntdll.NtQueryEaFile stub - #46822 Edit control in ADL search dialog gets initially not drawn in DC++ 0.868, regression - #47375 Multiple applications crash on unimplemented function msvcr120.dll.?_Schedule(a)_StructuredTaskCollection@details(a)Concurrency@@QEAAXPEAV_UnrealizedChore(a)23@@Z (Photoshop C 2018, CLIP Studio Paint) - #48487 MSYS based Git 2.25 installer reports 'could not decode hex <hexstring>' (needs support for CryptStringToBinaryW CRYPT_STRING_HEX) - #48600 Bloomberg Terminal needs unimplemented function dbghelp.dll.SymSrvGetFileIndexInfo - #49285 PNotesPortable crashes inside WritePrivateProfileStructW - #50948 taskmgr.exe: wrong memory usage unit (GB => MB) - #51788 windowscodecs:wmpformat test_decode() fails randomly - #51842 include/winnt.h:403:50: error: size of unnamed array is negative - #52457 CNG Encryption Failure (BCryptEncrypt) - #52534 ListView: multi select never sends LVN_ODSTATECHANGED - #52545 ACE rights parser should allow octal and decimal formats - #52686 DateAdd not implemented - #52831 Kernel32::GetSystemPowerStatus returns invalid data if /sys/class/power_supply/BAT0 is missing - #52841 Leverless arcade controller SOCD cleaning does not work - #52972 wineusb.sys: add_usb_device erroneously releases critical section on failure path - #52975 VTFEdit fails to start: err:module:LdrInitializeThunk "MSVCP80.dll" failed to initialize, aborting - #53003 XM6 TypeG: some MIDI note-on messages aren't sent to MIDI devices (MIDI running status isn't supported in the MIDI mapper) - #53005 quartz:systemclock Fails after timeGetTime() wraps - #53027 Port Royale 4 crashes with wined3d_texture* - #53032 winedevice.exe segfaults on exit when built with GCC - #53051 Kvaser software crashes on unimplemented function ntoskrnl.exe.ExNotifyCallback - #53123 ListView: Ctrl+Shift+LMouse multiselect on LVS_OWNERDATA listviews erroneously sends LVN_ODSTATECHANGED - #53169 Rich edit control becomes unstable or trips assertion after ITextRange::SetFont is called - #53170 Rich edit control becomes unstable or trips assertion after changing TextFont properties - #53234 nsi:nsi - test_ndis_ifinfo() fails randomly on Windows 10 1507 - #53310 Roblox Fails With HttpSendRequest - #53335 Crash when converting an rich text editor already containing text into a password input - #53455 shell32:recyclebin - test_query_recyclebin() gets unexpected recyclebin size in Wine - #53569 postgresql installer 9.3 needs correct string length returned from fso.GetTempName - #53587 comctl32 GetWindowSubclass crash in Notepad++ (v8.4.3 and 8.4.4 only) - #53614 KeePassXC crashes on unimplemented function msvcp140.dll.?ReportUnhandledError(a)_ExceptionHolder@details(a)Concurrency@@AEAAXXZ - #53645 AIMP 3 causes memory leaks - #53706 NtDeviceIoControlFile() is quite slow when called with and invalid handle - #53714 SQLGetInfo(W) doesn't fill StringLength when InfoValue is NULL - #53800 CRYPT_GetBuiltinDecoder Unsupported decoder for lpszStructType 1.3.6.1.4.1.311.2.1.4 - #53821 winbase.h: ReOpenFile declaration missing (and not available in any other header) - #53832 wbemprox: where-clause seems to be case-sensitive ---------------------------------------------------------------- Changes since 7.0.1: Akihiro Sagawa (2): midimap: Handle MIDI running status. mciqtz32: Fix thread handle leak if a second play. Alex Henrie (3): ntoskrnl: Add ExNotifyCallback stub. ntdll: Don't hard-code the battery and AC adapter names on Linux. user32: Plug icon handle leak in MDI_AugmentFrameMenu (Coverity). Alexandre Julliard (4): shell32: Don't return an uninitialized pointer on error in enum_trash_items(). urlmon/tests: Fix race condition in abort test. mountmgr: Don't crash if connecting to dbus fails. gdiplus: Get the compositing mode directly from the graphics object. Alistair Leslie-Hughes (8): d3drm: Support IDirect3D3 when creating device. ntdll: Check return parameter before use. ntdll: RtlFirstFreeAce only return FALSE on error. ntdll/tests: Add RtlFirstFreeAce tests. dsdmo: Add Echo FX Support. dsdmo: Add Compressor FX Support. odbc: SQLGetInfoW allow NULL parameter InfoValue. crypt32: Only report Unimplemented decoder when not found in external dll. Angelo Haller (9): comctl32/tests: Add hold_key and release_key functions. comctl32/tests: Add multi select tests for ownerdata listviews. comctl32/tests: Add change sequence for ownerdata listviews. comctl32/listview: Send one deselect all items notification for LVS_OWNERDATA listviews. comctl32/listview: Move sending LVN_ODSTATECHANGED notifications to a function. comctl32/listview: Send LVN_ODSTATECHANGED only for LVS_OWNERDATA listviews. comctl32/listview: Send LVN_ODSTATECHANGED notification for LVS_OWNERDATA listview on selection changes. comctl32/listview: Don't send LVN_ODSTATECHANGED for empty ranges. comctl32/tests: Add more ownerdata listview tests. Claire Girka (1): winepulse: Fix memory leak in pulse_create_stream in some error cases. Daniel Lehman (1): kernelbase: Pass va_list copy to internal RtlFormatMessage. David Curtiss (1): ws2_32: Allow getsockname after AcceptEx. David Kahurani (2): win32u: Use allocator mechanism to allocate DC attributes. ieframe: Fix memory leak in open_dde_url. Dmitry Timoshkov (2): comctl32/tests: Add iItem and uChanged to LVN_ITEMCHANGING/LVN_ITEMCHANGED context. sechost: Support decimal or octal access rights in ConvertStringSecurityDescriptorToSecurityDescriptor. Eric Pouech (2): wldap32: Fix compilation in 32-bit. winedbg: Fix retrieving integral values. Esme Povirk (1): wmphoto: Zero user buffer before calling jxrlib Copy. Fabian Maurer (1): kernel32: Prevent null pointer dereference in WritePrivateProfileStructW. Gijs Vermeulen (2): dbghelp: Add SymSrvGetFileIndexInfo stub. msvcp140: Add _ExceptionHolder::ReportUnhandledError stub. Hans Leidekker (4): wininet: Remove assertion that is no longer valid in Windows 10. wininet: Always reset the second input buffer in netcon_secure_connect_setup(). bcrypt: Force symmetric key reset if necessary. wbemprox: WQL string comparisons are case insensitive. Hugh McMaster (1): conhost: Set correct face name when the specified font is not available. Huw D. M. Davies (1): nsi: Use the correct members for the error and discard counts. Jeff Smith (2): d3drm/tests: Clean up some magic values. d3drm: Fix components of palette built when loading texture from file. Jinoh Kang (3): wineusb.sys: Remove erroneous call to LeaveCriticalSection() in add_usb_device. riched20: Commit and wrap paragraphs after changing text properties. riched20: Ensure MEPF_COMPLEX is unset when in password input mode. Kevin Puetz (2): winecrt0: Remove free_delay_imports. vbscript: Don't abort when trying to TRACE() functions with many arguments. Limstash Wong (1): taskmgr: Fix the unit in memory usage graph on performance tab. Louis Lenders (1): msvcr80: Update manifest resource. Martin Storsjö (2): include: Add declarations of _alloca and alloca for MSVC mode. ntdll: Add SEH unwind info in ARM assembly functions/trampolines. Mihail Ivanchev (1): include: Add prototype for ReOpenFile(). Nikolay Sivov (2): vbscript: Implement DateSerial(). vbscript: Implement DateAdd(). Paul Gofman (3): msxml3: Use node_get_text() in domelem_get_nodeTypedValue() if no type is present. crypt32: Support CRYPT_STRING_HEX in CryptStringToBinary(). winhttp/tests: Avoid race condition on closing connection in server_thread(). Piotr Caban (4): msvcr120: Add missing _StructuredTaskCollection destructor spec file entry. msvcrt: Fix _wputenv_s return value on error. msvcrt: Fix _putenv_s return value on error. msvcrt: Fix _wputenv_s invalid argument handling. Robert Wilhelm (4): scrrun: Added DateCreated property for IFile. scrrun: Add test for correct string length returned from GetTempName. scrrun: Correct string length returned from GetTempName. scrrun: Avoid memory leak in error case. Rémi Bernon (2): winebus.sys: Implement SOCD neutral cleaning for hatswitches. server: Avoid using pointer value after realloc. Santino Mazza (3): riched20/tests: When selection is out of view, the scrollbar should move. riched20: Ensure that cursor is visible when selecting text. jscript: Return ParseProcedureText when compile fails. Sebastian Lackner (1): ntdll: Improve stub of NtQueryEaFile. Tim Clem (1): kernelbase: Don't assume the length of the process image name. Torge Matthies (8): msvcr110: Add _StructuredTaskCollection::_StructuredTaskCollection_ctor stub. msvcr120: Add _StructuredTaskCollection::_StructuredTaskCollection_dtor stub. msvcr100: Add _StructuredTaskCollection::_Schedule stub. msvcr110: Add _StructuredTaskCollection::_Schedule_loc stub. msvcr100: Implement invalid_multiple_scheduling exception. msvcr100: Factor out the mapping of a context to a scheduler. msvcr100: Implement _StructuredTaskCollection::_Schedule and _Schedule_loc. msvcp90: Fix segfault in bad_alloc exception default constructor. Zebediah Figura (7): quartz/systemclock: Use timeGetTime() to retrieve the current time. wined3d: Correctly handle 3D textures in wined3d_view_{load, invalidate}_location(). ntdll: Add a fast path for NtDeviceIoControlFile(INVALID_HANDLE_VALUE). xslt: Use win32config.h from the correct directory. ntdll: Calculate the necessary length after calling snprintf() in SystemWineVersionInformation. msvcrt: Add SEH annotations for sse2_memcpy(). libwine: Avoid comparing the result of pointer arithmetic to zero. Zhiyi Zhang (14): shell32: Correct SHOpenFolderAndSelectItems() prototype. include: Add SHOpenFolderAndSelectItems() declaration and flags. shell32/tests: Add SHOpenFolderAndSelectItems() tests. shell32: Implement SHOpenFolderAndSelectItems(). comctl32/tests: Add GetWindowSubclass() tests. comctl32: Check reference data pointer before using it. comctl32/tests: Test that changing focus for edit controls doesn't repaint immediately. comctl32/edit: Don't repaint immediately when getting focus. winex11.drv: Handle X errors from glXCopyContext(). winex11.drv: Don't compare error event serial if it's zero. winegstreamer: Always check the return value from mf_media_type_from_wg_format(). gdi32/tests: Add more tests for GetTextMetrics(). win32u: Choose a smaller ppem to avoid exceeding the requested font height. win32u: Fix a memory leak. bye michael

March 16, 2023

Re: fileapi.h / Kernel32: GetTempPath2A/W() implementation

by Stefan Dösinger

Am Donnerstag, 16. März 2023, 18:52:33 EAT schrieb Gabriel Ivăncescu: > > From GetTempPath2W() docs: "When calling this function from a process > > > > running as SYSTEM it will return the path C:\Windows\SystemTemp" Do we even care for the purpose of GetTempPath2? Wine doesn't really do multiple users (it all lives in the context of your own Linux user), and it handles different privilege levels only as much as necessary. In that sense a real SYSTEM user doesn't exist, and depending on your point of view everything runs as SYSTEM or nothing runs as SYSTEM.

March 16, 2023