http://bugs.winehq.org/show_bug.cgi?id=20565
Summary: Emu42: regression in StretchBlt causes black display background Product: Wine Version: 1.1.31 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: winex11.drv AssignedTo: wine-bugs@winehq.org ReportedBy: thomas@intevation.de
Emu42 1.12 contains a workaround for the BitBlt problem discussed in http://bugs.winehq.org/show_bug.cgi?id=4140 This bugzilla entry was split out of above entry, because it is a different problem.
As noted in http://hp.giesselink.com/Emu42/CHANGES.TXT Emu42 now uses StretchBlt instead, which worked fine until wine 1.1.30, but no longer in 1.1.31 (and still does not work in .32 and yesterday's git).
I did a git bisect on the regression and tracked it down to:
| 77b9b8a307711bd1f6da30f7d2d5e04faa3f5092 is first bad commit | commit 77b9b8a307711bd1f6da30f7d2d5e04faa3f5092 | Author: Alexandre Julliard julliard@winehq.org | Date: Wed Sep 30 20:39:51 2009 +0200 | | winex11: Move the DIB locking and the client-side optimizations into | BITBLT_InternalStretchBlt. | | This way they also apply to the non-stretching StretchBlt case.
After this commit the display looks exactly like in previous versions of Emu42 which use BitBlt: extremely dark red pixels on a black background (= unreadable)