Correct big-memory bug.
 [rsc] --rw-rw-r-- M 852702 glenda sys 12395 Jan 19 11:45 sys/src/9/pc/vganvidia.c
	/n/sourcesdump/2006/0119/plan9/sys/src/9/pc/vganvidia.c:189,195 - 
	/n/sourcesdump/2006/0120/plan9/sys/src/9/pc/vganvidia.c:189,195
	  		 * have allocated less storage than aux/vga
	  		 * expected.
	  		 */
	- 		tmp = scr->storage - 96*1024;
	+ 		tmp = scr->apsize - 96*1024;
	  		p = (void*)((uchar*)scr->vaddr + tmp);
	  		vgaxo(Crtx, 0x30, 0x80|(tmp>>17));
	  		vgaxo(Crtx, 0x31, (tmp>>11)<<2);
	/n/sourcesdump/2006/0119/plan9/sys/src/9/pc/vganvidia.c:363,369 - 
	/n/sourcesdump/2006/0120/plan9/sys/src/9/pc/vganvidia.c:363,385
	  
	  	pitch = scr->gscreen->width*BY2WD;
	  
	- 	nv.dmabase = (void*)((uchar*)scr->vaddr + scr->storage - 128*1024);
	+ 	/*
	+ 	 * DMA is at the end of the virtual window,
	+ 	 * but we might have cut it short when mapping it.
	+ 	 */
	+ 	if(nv.dmabase == nil){
	+ 		if(scr->storage <= scr->apsize)
	+ 			nv.dmabase = (ulong*)((uchar*)scr->vaddr + scr->storage - 128*1024);
	+ 		else{
	+ 			nv.dmabase = (void*)vmap(scr->paddr + scr->storage - 128*1024, 128*1024);
	+ 			if(nv.dmabase == 0){
	+ 				hwaccel = 0;
	+ 				hwblank = 0;
	+ 				print("vmap nvidia dma failed\n");
	+ 				return;
	+ 			}
	+ 		}
	+ 	}
	  
	  	for(i=0; i<SKIPS; i++)
	  		nv.dmabase[i] = 0x00000000;