https://bugs.winehq.org/show_bug.cgi?id=43062
Bug ID: 43062 Summary: Installer from GOG.com displays black rectangles Product: Wine Version: 2.8 Hardware: x86 URL: https://www.gog.com/game/teenagent OS: Linux Status: NEW Keywords: Installer, regression Severity: minor Priority: P2 Component: winex11.drv Assignee: wine-bugs@winehq.org Reporter: gyebro69@gmail.com CC: julliard@winehq.org Regression SHA1: bb6a65e0ca884809fc6077604bdc1d6f1e348197 Distribution: ---
Created attachment 58232 --> https://bugs.winehq.org/attachment.cgi?id=58232 terminal output
Certain parts of the installer window for the games available on GOG.com are black. Plain terminal output doesn't reveal anything suspicious.
Reverting the following commit fixes the problem:
commit bb6a65e0ca884809fc6077604bdc1d6f1e348197 Author: Alexandre Julliard julliard@winehq.org Date: Thu May 18 13:35:02 2017 +0200
winex11: Fall back to the null driver for common device capabilities.
There are a number of games available for free on GOG.com (after registration): https://www.gog.com/games?price=free&sort=bestselling&page=1
I picked out the game 'Teenagent' to reproduce the issue because it has a small download size (19 MB): https://www.gog.com/game/teenagent
wine-2.8-133-g3f01ab4c39 xorg-server 1.19.3 OpenGL vendor string: NVIDIA Corporation OpenGL renderer string: GeForce GT 730/PCIe/SSE2 OpenGL core profile version string: 4.5.0 NVIDIA 381.22
https://bugs.winehq.org/show_bug.cgi?id=43062
--- Comment #1 from Béla Gyebrószki gyebro69@gmail.com --- Created attachment 58233 --> https://bugs.winehq.org/attachment.cgi?id=58233 screenshot
https://bugs.winehq.org/show_bug.cgi?id=43062
--- Comment #2 from Bruno Jesus 00cpxxx@gmail.com --- I can confirm this.
https://bugs.winehq.org/show_bug.cgi?id=43062
--- Comment #3 from Alexandre Julliard julliard@winehq.org --- That's a gdiplus bug, it uses SHADEBLENDCAPS to check if alpha blending is supported, but the meaning is different on display devices. Quoting MSDN: "For a display device, all blending operations are available; besides SB_NONE, the only return values are SB_CONST_ALPHA and SB_PIXEL_ALPHA, which indicate whether these operations are accelerated." It works OK with native gdiplus.
https://bugs.winehq.org/show_bug.cgi?id=43062
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |madewokherd@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=43062
--- Comment #4 from Vincent Povirk madewokherd@gmail.com --- What's the proper way to determine whether GdiAlphaBlend works? I believe the original motivation for this was printer devices that don't support it.
https://bugs.winehq.org/show_bug.cgi?id=43062
--- Comment #5 from Vincent Povirk madewokherd@gmail.com --- MSDN suggests we should check SHADEBLENDCAPS only for printer devices.
https://bugs.winehq.org/show_bug.cgi?id=43062
--- Comment #6 from Alexandre Julliard julliard@winehq.org --- Yes, something like GetDeviceCaps(graphics->hdc, TECHNOLOGY ) != DT_RASDISPLAY should work. We should probably also check the relevant SHADEBLENDCAPS flags instead of simply checking for non-zero.
https://bugs.winehq.org/show_bug.cgi?id=43062
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Fixed by SHA1| |b8a8e1bde9382897927945ec0b5 | |8b8a2fcfcfe59 Resolution|--- |FIXED
--- Comment #7 from Alexandre Julliard julliard@winehq.org --- Fixed by b8a8e1bde9382897927945ec0b58b8a2fcfcfe59. Thanks Vincent!
https://bugs.winehq.org/show_bug.cgi?id=43062
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #8 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 2.9.
-
wine-bugs＠winehq.org