On Di, 2014-07-01 at 10:31 +0200, Paolo Bonzini wrote: > Il 01/07/2014 10:26, Alexander Graf ha scritto: > >> > >>> Right. A quick fix would be to add a flag to force always using a shadow > >>> surface and set it in cirrus ... I'm not sure anybody will notice the > >>> performance difference. > >> I suspect we can rip out hw cursor emulation and nobody will notice the > >> difference either ... > > > > Very likely ;). Though I think we're better off keeping it around to > > make sure we're still compatible with ancient guests (Windows 3.1 might > > use it too). Making it slow however shouldn't make any difference at all. > > If you tell me what to look at, I legally own a Windows 98 CD (also NT4 > but I have to dig it out) and can test it later this week.
/me has win98 too, that doesn't boot after install though. So if you can try nt4 that would be great. Test is simple: switch into 16bpp mode (that one uses shared surface in git master), check whenever the mouse pointer is present. Bonus points for compiling with DEBUG_CIRRUS and checking the logs to see whenever nt4 actually enables the hw cursor (sr registers, index 0x12, bit 0). cheers, Gerd
