Can you check and make sure the MMIO and framebuffer are mapped? I belive its pMga->IOBase and pMga->FbBase. Maybe for some reason they didn't get mapped, which would explain why once you try to copy (swcursor) into the FB or 8x8 Mono blit the MMIO will crap out with null pointers.
ani On Tue, 30 Oct 2001, Romain Dolbeau wrote: > Hello, > > I'm still trying to get my Matrox Millenium to correctly > dual-head with my 3Dlabs Permedia3 on my PowerMac. > With the latest CVS, the server segfault in the MGA > driver. > > Note that single-head works fine with the patch to > matrofb Ani Joshi send me (as long as useFBDev is off). > I can even launch 2 copies of XFree86, one on the > Mil1 (vt7), one on the pm3 (vt8), and switch between > the two. I just can't get proper dual-head. > > I rebuilded a static/debug XFree86 and here's what > I get _without_ NoAccel and _without_ PciRetry: > > ##### > Program received signal SIGSEGV, Segmentation fault. > 0x10081b98 in Mga32SetupForMono8x8PatternFill (pScrn=0x10b49ad0, patx=2011020270, >paty=2011020270, fg=0, > bg=16777215, rop=3, planemask=4294967295) at mga_storm32.c:1574 > 1574 WAITFIFO(6); > (gdb) bt > #0 0x10081b98 in Mga32SetupForMono8x8PatternFill (pScrn=0x10b49ad0, >patx=2011020270, paty=2011020270, fg=0, > bg=16777215, rop=3, planemask=4294967295) at mga_storm32.c:1574 > #1 0x102152b8 in XAAFillMono8x8PatternRectsScreenOrigin (pScrn=0x10b49ad0, fg=0, >bg=16777215, rop=3, > planemask=4294967295, nBox=1, pBox=0x7ffff968, pattern0=2011020270, >pattern1=2011020270, xorigin=0, yorigin=0) > at xaaFillRect.c:438 > #2 0x10228ed8 in XAAPaintWindow (pWin=0x10b9b6c8, prgn=0x7ffff968, what=0) at >xaaPaintWin.c:139 > #3 0x105dc2cc in miSpritePaintWindowBackground (pWin=0x10b9b6c8, >pRegion=0x7ffff968, what=0) at misprite.c:840 > #4 0x105b2994 in miWindowExposures (pWin=0x10b9b6c8, prgn=0x7ffff968, >other_exposed=0x0) at miexpose.c:532 > #5 0x1037aca0 in xf86XVWindowExposures (pWin=0x10b9b6c8, reg1=0x7ffff968, reg2=0x0) >at xf86xv.c:1027 > #6 0x10434c0c in MapWindow (pWin=0x10b9b6c8, client=0x10b3bf40) at window.c:2865 > #7 0x1042f110 in InitRootWindow (pWin=0x10b9b6c8) at window.c:518 > #8 0x1041bd4c in main (argc=3, argv=0x7ffffa94, envp=0x7ffffaa4) at main.c:435 > ##### > > The same _with_ NoAccel, _without_ PciRetry ; I get a SegFault in > the same place and the same calling sequence with only PciRetry. > > ##### > Program received signal SIGSEGV, Segmentation fault. > 0x10066240 in xf86ReadMmio8 (base=0x30015000, offset=15370) > at ../../../../../../programs/Xserver/hw/xfree86/common/compiler.h:880 > 880 __asm__ __volatile__( > (gdb) bt > #0 0x10066240 in xf86ReadMmio8 (base=0x30015000, offset=15370) > at ../../../../../../programs/Xserver/hw/xfree86/common/compiler.h:880 > #1 0x10064d80 in MGA3026HideCursor (pScrn=0x10b49af8) at mga_dac3026.c:918 > #2 0x101f6720 in xf86SetCursor (pScreen=0x10b4e028, pCurs=0x10c675a8, x=505, y=377) >at xf86HWCurs.c:133 > #3 0x101f615c in xf86CursorSetCursor (pScreen=0x10b4e028, pCurs=0x10c675a8, x=512, >y=384) at xf86Cursor.c:335 > #4 0x105cbd84 in miPointerUpdate () at mipointer.c:352 > #5 0x105cb6bc in miPointerDisplayCursor (pScreen=0x10b4e028, pCursor=0x10c675a8) at >mipointer.c:175 > #6 0x1040ea58 in DefineInitialRootWindow (win=0x10b87a28) at events.c:2124 > #7 0x1041bd6c in main (argc=3, argv=0x7ffffa94, envp=0x7ffffaa4) at main.c:436 > ##### > > I've also attached the XF86COnfig-4 and XFree86.0.log for > the first trace. > > Thanks for any help/advices/comments, > > -- > DOLBEAU Romain | The Gods made Heavy Metal > ENS Cachan / Ker Lann | and it's never gonna die > Thesard IRISA / CAPS | -- Manowar, > [EMAIL PROTECTED] | 'The Gods made Heavy Metal' _______________________________________________ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
