February 2024 - Wine-GitLab

[PATCH 0/1] MR5023: quartz: Release critical section before IVideoWindow_put_Visible
by CKolle (＠CKolle) Feb. 6, 2024

Feb. 6, 2024

The IVideoWindow_put_Visible runs the NtUserShowWindow which may trigger the VideoWindow_NotifyOwnerMessage with a WM_PAINT message and afterwards try to send a WM_ERASEBKND. This will cause the VideoWindow_NotifyOwnerMessage to be stuck waiting on the lock. This causes a deadlock because the IVideoWindow_put_Visible is never able to return, as it is waiting for the WM_ERASEBKND message to go through in the send_erase function, which can only go through once the WM_PAINT has returned. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5023

3 4

[PATCH 0/1] MR5038: dmusic: Fix loading wave data from soundfont.
by Yuxuan Shui (＠yshui) Feb. 6, 2024

Feb. 6, 2024

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/5038

4 3

[PATCH v3 0/3] MR5035: Change SetLastError to RestoreLastError if appropriate
by Alfred Agrell (＠Alcaro) Feb. 6, 2024

Feb. 6, 2024

Theoretically, this allows debuggers to tell them apart. https://devblogs.microsoft.com/oldnewthing/20110429-00/?p=10803 But in practice, this is more about cleanliness than anything else; I'm not aware of anyone running SetLastError-aware debuggers in Wine. Feel free to reject if this is a waste of time. The changeable SetLastError calls were found by this script: ```python #!/usr/bin/env python3 import os import re for (dirpath, dirnames, filenames) in os.walk("."): for fn in filenames: if not fn.endswith(".c") and not fn.endswith(".h"): continue text = open(dirpath+"/"+fn,"rt").read() text = text.replace("RtlSetLastWin32Error","SetLastError") text = text.replace("RtlGetLastWin32Error","GetLastError") if "SetLastError" not in text: continue varname = None for line in text.split("\n"): if "GetLastError" in line: if m := re.search(r"([A-Za-z0-9_]+) *= *GetLastError", line): prevline = line varname = m[1] if line == "}": varname = None if varname is not None and "SetLastError" in line: if m := re.search(r"SetLastError\( *([A-Za-z0-9_]+) *\)", line): if m[1] == varname: print(dirpath+"/"+fn, prevline, line) ``` The script also flags https://gitlab.winehq.org/wine/wine/-/blob/master/dlls/mpr/wnet.c#L2811, which is half set and half restore; I chose to left it unchanged. Idea from https://gitlab.winehq.org/wine/wine/-/merge_requests/5020/diffs#6c845445f68… -- v3: win32u: Use RtlRestoreLastWin32Error instead of RtlSetLastWin32Error if appropriate everywhere: Change SetLastError to RestoreLastError if appropriate https://gitlab.winehq.org/wine/wine/-/merge_requests/5035

3 5

[PATCH v4 0/10] MR4954: kernelbase: Reimplement UrlCanonicalize().
by Zebediah Figura (＠zfigura) Feb. 6, 2024

Feb. 6, 2024

Normally I am a strong proponent of adjusting implementations one small step at a time, but this is a bit of an extreme case. The current state of UrlCanonicalize() is a prototypical example of "spaghetti code": the implementation has been repeatedly adjusted to fix a new discovered edge case, without properly testing the scope of the new logic, with the effect that the current logic is a scattered mess of conditions that interact in unpredictable ways. To be fair, the actual function is much more complicated than one would anticipate, and the number of new weird edge cases I found while trying to flesh out existing logic was rather exhausting. I initially tried to "fix" the existing implementation one step at a time. After racking up dozens of commits without an end in sight, though, and having to waste time carefully unravelling the broken code in the right order so as to avoid temporary failures, I decided instead to just fix everything at once, effectively rewriting the function from scratch, and this proved to be much more productive. Hopefully the huge swath of new tests is enough to prove that this new implementation really is correct, and has no more spaghetti than what Microsoft made necessary. -- v4: shlwapi/tests: Add many more tests for UrlCanonicalize(). kernelbase: Use scheme_is_opaque() in UrlIs(). kernelbase: Reimplement UrlCanonicalize(). kernelbase: Do not canonicalize the relative part in UrlCombine(). kernelbase: Do not use isalnum() with Unicode characters. https://gitlab.winehq.org/wine/wine/-/merge_requests/4954

3 13

[PATCH v5 0/4] MR4761: ntdll: Implement NtContinueEx()
by Dāvis Mosāns Feb. 6, 2024

Feb. 6, 2024

This MR implements `NtContinueEx(PCONTEXT, PCONTINUE_OPTIONS)` which was added in Windows 10 20H1. Also added basic test reusing existing `test_continue()` (included from !4720) League of Legends game hooks `NtContinue()` and `NtContinueEx()` by putting `jmp` instruction in front of it. Note that LoL doesn't actually use `NtContinueEx()` itself and game will work fine without it but if game doesn't find `NtContinueEx()` it will permanently ban your account due to detecting "scripting program" -- v5: ntdll/tests: Implement APC test for NtContinue() ntdll/tests: Add basic test for NtContinueEx() ntdll: Implement NtContinueEx() ntdll/tests: Unify APC test functions https://gitlab.winehq.org/wine/wine/-/merge_requests/4761

1 5

Re: [PATCH v11 0/7] MR4082: gdiplus: Implement font linking in GdipDrawString.
by Esme Povirk (＠madewokherd) Feb. 6, 2024

Feb. 6, 2024

On Tue Feb 6 18:52:35 2024 +0000, Esme Povirk wrote: > No, that doesn't seem right either, I need to think about this. Thought about it, I do think it should be ||. We should break if all of the text has been measured, and we should break if the current section doesn't fully fit in to_measure_length. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4082#note_60325

1 0

Re: [PATCH v11 0/7] MR4082: gdiplus: Implement font linking in GdipDrawString.
by Esme Povirk (＠madewokherd) Feb. 6, 2024

Feb. 6, 2024

On Tue Feb 6 18:52:00 2024 +0000, Esme Povirk wrote: > I think this should be || No, that doesn't seem right either, I need to think about this. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4082#note_60324

1 0

Re: [PATCH v11 0/7] MR4082: gdiplus: Implement font linking in GdipDrawString.
by Esme Povirk (＠madewokherd) Feb. 6, 2024

Feb. 6, 2024

Esme Povirk (@madewokherd) commented about dlls/gdiplus/graphics.c: > + to_measure_length = min(length - (i - index), section->end - i); > + > + get_font_hfont(graphics, section->font, NULL, &hfont, NULL, NULL); > + oldhfont = SelectObject(graphics->hdc, hfont); > + GetTextExtentExPointW(graphics->hdc, &string[i], to_measure_length, max_ext, &fitaux, NULL, &sizeaux); > + SelectObject(graphics->hdc, oldhfont); > + DeleteObject(hfont); > + > + max_ext -= sizeaux.cx; > + if (fit) > + *fit += fitaux; > + size->cx += sizeaux.cx; > + size->cy = max(size->cy, sizeaux.cy); > + > + i += to_measure_length; > + if ((i - index) >= length && fitaux < to_measure_length) break; I think this should be || -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4082#note_60323

1 0

Re: [PATCH v11 0/7] MR4082: gdiplus: Implement font linking in GdipDrawString.
by Esme Povirk (＠madewokherd) Feb. 6, 2024

Feb. 6, 2024

On Tue Feb 6 15:01:50 2024 +0000, Santino Mazza wrote: > I tested GdipMeasureString with different strings that require font > linking and it behaves correctly. It's strange because I also tested > with the exact same string the game uses and with the generic > typographic format, but I still can't get the bug to reproduce. > Apparently when the game tries to draw that specific string, > GdipMeasureString returns a bounding box with width 0, which causes > draw_string_callback to not increase the X position, so it overwrites > the other word. Again this doesn't happen if I use font_link_get_text_extent_point. The difference might have to do with scaling? It looks like GdipMeasureString scales the size, and I don't think font_link_get_text_extent_point would. Come to think of it, it's probably necessary to account for text rotation when in draw_string_callback (an arbitrary affine transform can be applied to the text, although I think at the moment our handling of anything beyond rotation is a bit broken). I think it might be a good idea to add output parameters for dx and dy to draw_driver_string. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4082#note_60317

1 0

[PATCH v2 0/1] MR5024: msvcrt: Adjust _gmtime64_s() accepted time limits.
by Paul Gofman (＠gofman) Feb. 6, 2024

Feb. 6, 2024

Fixes a crash in Microsoft Flight Simulator (which passes some currently invalid values to ucrtbase._gmtime64() and doesn't expect NULL result). It looks like any msvcrt version on modern Windows accepts wider range of values. Meanwhile, the exact range is different for ucrtbase vs older CRT versions (the exact values for ucrtbase also vary before Win10 1909). Also, older msvcrt.dll matches what we have in Wine now. So the patch keeps old msvcrt behavior, adds universally correct limits for msvcrt 70-120 and the latest (Win10 1909+) for ucrtbase. -- v2: msvcrt: Adjust _gmtime64_s() accepted time limits. https://gitlab.winehq.org/wine/wine/-/merge_requests/5024

3 6