SVGA: double buffering

1 files changed, 13 insertions, 1 deletions

diff --git a/vxd_vdd.c b/vxd_vdd.c
index 422b1c4..af7b200 100644
--- a/vxd_vdd.c
+++ b/vxd_vdd.c
@@ -92,6 +92,18 @@ DWORD map_pm16(DWORD vm, DWORD linear, DWORD size)
 	return selector << 16;

 }

 

+void update_pm16(DWORD vm, DWORD oldmap, DWORD linear, DWORD size)

+{

+	WORD selector = oldmap >> 16;

+	DWORD hi  = 0;

+	DWORD low = 0;

+	

+	dbg_printf("update_pm16: %ld\n", linear);

+	

+	_BuildDescriptorDWORDs(linear, RoundToPages(size), 0x92, 0x80, 0, &hi, &low);

+	_SetDescriptor(selector, vm, hi, low, 0);

+}

+

 /*

 You can implement all the VESA support entirely in your mini-VDD. Doing so will

 cause VESA applications to run more efficiently since all of the VESA support is

@@ -128,7 +140,7 @@ See also VDD_REGISTER_DISPLAY_DRIVER_INFO.
 **/

 VDDPROC(REGISTER_DISPLAY_DRIVER, register_display_driver)

 {

-	dbg_printf(dbg_register, state->Client_EBX, state->Client_ECX, ThisVM);

+	dbg_printf("register: ebx = %ld, ecx = %ld, VM = %lX\n", state->Client_EBX, state->Client_ECX, ThisVM);

 	

 	if(hda->vram_pm16 == 0)

 	{