diff --git a/source/cl_main.c b/source/cl_main.c
index c2226c6..5a9a025 100644
--- a/source/cl_main.c
+++ b/source/cl_main.c
@@ -337,12 +337,12 @@ void SetPal (int i)
 
 void CL_CopyPlayerInfo (entity_t *ent, entity_t *player)
 {
-	memcpy (&ent->baseline, &player->baseline, sizeof(entity_state_t));
+	memcpy_vfpu(&ent->baseline, &player->baseline, sizeof(entity_state_t));
 
 	ent->msgtime = player->msgtime;
-	memcpy (ent->msg_origins, player->msg_origins, sizeof(ent->msg_origins));
+	memcpy_vfpu(ent->msg_origins, player->msg_origins, sizeof(ent->msg_origins));
 	VectorCopy (player->origin, ent->origin);
-	memcpy (ent->msg_angles, player->msg_angles, sizeof(ent->msg_angles));
+	memcpy_vfpu(ent->msg_angles, player->msg_angles, sizeof(ent->msg_angles));
 	VectorCopy (player->angles, ent->angles);
 
 	ent->model = (ent == &q3player_body.ent) ? cl.model_precache[cl_modelindex[mi_q3torso]] : cl.model_precache[cl_modelindex[mi_q3head]];
diff --git a/source/cl_parse.c b/source/cl_parse.c
index 0ea02cd..7811f91 100644
--- a/source/cl_parse.c
+++ b/source/cl_parse.c
@@ -205,7 +205,7 @@ void CL_KeepaliveMessage (void)
 
 // read messages from server, should just be nops
 	old = net_message;
-	memcpy (olddata, net_message.data, net_message.cursize);
+	memcpy_vfpu(olddata, net_message.data, net_message.cursize);
 
 	do
 	{
@@ -227,7 +227,7 @@ void CL_KeepaliveMessage (void)
 	} while (ret);
 
 	net_message = old;
-	memcpy (net_message.data, olddata, net_message.cursize);
+	memcpy_vfpu(net_message.data, olddata, net_message.cursize);
 
 // check time
 	time = Sys_FloatTime ();
diff --git a/source/cl_slist.c b/source/cl_slist.c
index bc5dc97..ccbfa3b 100644
--- a/source/cl_slist.c
+++ b/source/cl_slist.c
@@ -124,9 +124,9 @@ void SList_Switch (int a, int b)
 	if (b >= MAX_SERVER_LIST || b < 0)
 		Sys_Error ("SList_Switch: Bad index %d", b);
 
-	memcpy (&temp, &slist[a], sizeof(temp));
-	memcpy (&slist[a], &slist[b], sizeof(temp));
-	memcpy (&slist[b], &temp, sizeof(temp));
+	memcpy_vfpu(&temp, &slist[a], sizeof(temp));
+	memcpy_vfpu(&slist[a], &slist[b], sizeof(temp));
+	memcpy_vfpu(&slist[b], &temp, sizeof(temp));
 }
 
 int SList_Length (void)
diff --git a/source/cmd.c b/source/cmd.c
index 7f6d0b6..e1355ab 100644
--- a/source/cmd.c
+++ b/source/cmd.c
@@ -164,7 +164,7 @@ void Cbuf_Execute (void)
 		}
 			
 				
-		memcpy (line, text, i);
+		memcpy_vfpu(line, text, i);
 		line[i] = 0;
 		
 // delete the text from the command buffer and move remaining commands down
diff --git a/source/host_cmd.c b/source/host_cmd.c
index 8390f9d..93d905c 100644
--- a/source/host_cmd.c
+++ b/source/host_cmd.c
@@ -406,9 +406,9 @@ void Host_SavegameComment (char *text)
 
 	for (i=0 ; i<SAVEGAME_COMMENT_LENGTH ; i++)
 		text[i] = ' ';
-	memcpy (text, cl.levelname, strlen(cl.levelname));
+	memcpy_vfpu(text, cl.levelname, strlen(cl.levelname));
 	sprintf (kills,"kills:%3i/%3i", cl.stats[STAT_INSTA], cl.stats[STAT_ROUNDCHANGE]);
-	memcpy (text+22, kills, strlen(kills));
+	memcpy_vfpu(text+22, kills, strlen(kills));
 // convert space to _ to make stdio happy
 	for (i=0 ; i<SAVEGAME_COMMENT_LENGTH ; i++)
 		if (text[i] == ' ')
diff --git a/source/mathlib.c b/source/mathlib.c
index d864e14..bbb0cb4 100644
--- a/source/mathlib.c
+++ b/source/mathlib.c
@@ -183,7 +183,7 @@ void RotatePointAroundVector( vec3_t dst, const vec3_t dir, const vec3_t point,
 	m[1][2] = vf[1];
 	m[2][2] = vf[2];
 
-	memcpy( im, m, sizeof( im ) );
+	memcpy_vfpu( im, m, sizeof( im ) );
 
 	im[0][1] = m[1][0];
 	im[0][2] = m[2][0];
diff --git a/source/menu.c b/source/menu.c
index 9086f9a..a787862 100644
--- a/source/menu.c
+++ b/source/menu.c
@@ -286,16 +286,16 @@ void M_BuildTranslationTable(int top, int bottom)
 		identityTable[j] = j;
 	dest = translationTable;
 	source = identityTable;
-	memcpy (dest, source, 256);
+	memcpy_vfpu(dest, source, 256);
 
 	if (top < 128)	// the artists made some backwards ranges.  sigh.
-		memcpy (dest + TOP_RANGE, source + top, 16);
+		memcpy_vfpu(dest + TOP_RANGE, source + top, 16);
 	else
 		for (j=0 ; j<16 ; j++)
 			dest[TOP_RANGE+j] = source[top+15-j];
 
 	if (bottom < 128)
-		memcpy (dest + BOTTOM_RANGE, source + bottom, 16);
+		memcpy_vfpu(dest + BOTTOM_RANGE, source + bottom, 16);
 	else
 		for (j=0 ; j<16 ; j++)
 			dest[BOTTOM_RANGE+j] = source[bottom+15-j];
diff --git a/source/pr_cmds.c b/source/pr_cmds.c
index 80557a3..b63b8b1 100644
--- a/source/pr_cmds.c
+++ b/source/pr_cmds.c
@@ -940,7 +940,7 @@ int PF_newcheckclient (int check)
 	VectorAdd (ent->v.origin, ent->v.view_ofs, org);
 	leaf = Mod_PointInLeaf (org, sv.worldmodel);
 	pvs = Mod_LeafPVS (leaf, sv.worldmodel);
-	memcpy (checkpvs, pvs, (sv.worldmodel->numleafs+7)>>3 );
+	memcpy_vfpu(checkpvs, pvs, (sv.worldmodel->numleafs+7)>>3 );
 
 	return i;
 }
diff --git a/source/psp/input.cpp b/source/psp/input.cpp
index 58a4446..780cfa0 100644
--- a/source/psp/input.cpp
+++ b/source/psp/input.cpp
@@ -94,9 +94,9 @@ void IN_Init (void)
 	buttonToGameKeyMap[buttonMaskToShift(PSP_CTRL_RIGHT)]	= K_RIGHTARROW;
 	buttonToGameKeyMap[buttonMaskToShift(PSP_CTRL_DOWN)]	= K_DOWNARROW;
 	buttonToGameKeyMap[buttonMaskToShift(PSP_CTRL_LEFT)]	= K_LEFTARROW;
-	memcpy(buttonToConsoleKeyMap, buttonToGameKeyMap, sizeof(ButtonToKeyMap));
-	memcpy(buttonToMessageKeyMap, buttonToGameKeyMap, sizeof(ButtonToKeyMap));
-	memcpy(buttonToMenuKeyMap, buttonToGameKeyMap, sizeof(ButtonToKeyMap));
+	memcpy_vfpu(buttonToConsoleKeyMap, buttonToGameKeyMap, sizeof(ButtonToKeyMap));
+	memcpy_vfpu(buttonToMessageKeyMap, buttonToGameKeyMap, sizeof(ButtonToKeyMap));
+	memcpy_vfpu(buttonToMenuKeyMap, buttonToGameKeyMap, sizeof(ButtonToKeyMap));
 
 	// Game keys:
 	buttonToGameKeyMap[buttonMaskToShift(PSP_CTRL_LTRIGGER)]	= K_AUX1;
@@ -115,7 +115,7 @@ void IN_Init (void)
 	buttonToConsoleKeyMap[buttonMaskToShift(PSP_CTRL_SQUARE)]	= K_INS;
 
 	// Message keys:
-	memcpy(buttonToMessageKeyMap, buttonToConsoleKeyMap, sizeof(ButtonToKeyMap));
+	memcpy_vfpu(buttonToMessageKeyMap, buttonToConsoleKeyMap, sizeof(ButtonToKeyMap));
 
 	// Menu keys:
 	buttonToMenuKeyMap[buttonMaskToShift(PSP_CTRL_SQUARE)]	= K_INS;
diff --git a/source/psp/network_psp.cpp b/source/psp/network_psp.cpp
index 3f27025..e528080 100644
--- a/source/psp/network_psp.cpp
+++ b/source/psp/network_psp.cpp
@@ -77,7 +77,7 @@ static pdpStatStruct gPdpStat;
 pdpStatStruct *findPdpStat(int socket, pdpStatStruct *pdpStat)
 {
 	if(socket == pdpStat->pdpId) {
-		memcpy(&gPdpStat, pdpStat, sizeof(pdpStatStruct));
+		memcpy_vfpu(&gPdpStat, pdpStat, sizeof(pdpStatStruct));
 		return &gPdpStat;
 	}
 		if(pdpStat->next) return findPdpStat(socket, pdpStat->next);
@@ -863,7 +863,7 @@ namespace quake
 				pdpStatStruct *tempPdp = findPdpStat(socket, pdpStat);
 				if(tempPdp < 0) return -1;
 
-				memcpy(((struct sockaddr_adhoc *)addr)->mac, tempPdp->mac, 6);
+				memcpy_vfpu(((struct sockaddr_adhoc *)addr)->mac, tempPdp->mac, 6);
 				((struct sockaddr_adhoc *)addr)->port = tempPdp->port;
 				addr->sa_family = ADHOC_NET;
 				return 0;
diff --git a/source/psp/video_hardware_draw.cpp b/source/psp/video_hardware_draw.cpp
index 5725b1a..5cc4545 100644
--- a/source/psp/video_hardware_draw.cpp
+++ b/source/psp/video_hardware_draw.cpp
@@ -488,7 +488,7 @@ qpic_t	*Draw_CacheImg (char *path)
 	// the translatable player picture just for the menu
 	// configuration dialog
 	if (!strcmp (path, "gfx/menuplyr.lmp"))
-		memcpy (menuplyr_pixels, dat->data, dat->width*dat->height);
+		memcpy(menuplyr_pixels, dat->data, dat->width*dat->height);
 
 	pic->pic.width = dat->width;
 	pic->pic.height = dat->height;
@@ -3398,16 +3398,6 @@ void GL_Upload4(int texture_index, const byte *data, int width, int height)
 	memcpy(texture.palette, data + buffer_size, 16 * 4);
 	int i;
 
-	// Copy to VRAM?
-	/*if (texture.vram)
-	{
-		// Copy.
-		memcpy(texture.vram, texture.ram, buffer_size);
-
-		// Flush the data cache.
-		sceKernelDcacheWritebackRange(texture.vram, buffer_size);
-	}*/
-
 	// Flush the data cache.
 	sceKernelDcacheWritebackRange(texture.ram, buffer_size);
 }
diff --git a/source/psp/video_hardware_hlmdl.cpp b/source/psp/video_hardware_hlmdl.cpp
index e321c73..ee9dabf 100644
--- a/source/psp/video_hardware_hlmdl.cpp
+++ b/source/psp/video_hardware_hlmdl.cpp
@@ -130,7 +130,7 @@ qboolean Mod_LoadHLModel (model_t *mod, void *buffer)
 	model = static_cast<hlmodelcache_t*>(Hunk_Alloc(sizeof(hlmodelcache_t)));
 
 	header = static_cast<hlmdl_header_t*>(Hunk_Alloc(com_filesize));
-	memcpy(header, buffer, com_filesize);
+	memcpy_vfpu(header, buffer, com_filesize);
 
 	if (header->version != 10)
 	{
@@ -175,7 +175,7 @@ qboolean Mod_LoadHLModel (model_t *mod, void *buffer)
 	if (!mod->cache.data)
 		return qfalse;
 
-	memcpy (mod->cache.data, model, total);
+	memcpy_vfpu(mod->cache.data, model, total);
 
 	Hunk_FreeToLowMark (start);
 	return qtrue;
@@ -482,7 +482,7 @@ void HL_SetupBones(hlmodel_t *model)
         }
         else
         {
-            memcpy(transform_matrix[i], matrix, 12 * sizeof(float));
+            memcpy_vfpu(transform_matrix[i], matrix, 12 * sizeof(float));
         }
     }
 }
diff --git a/source/psp/video_hardware_images.cpp b/source/psp/video_hardware_images.cpp
index 74fb8e3..6e51743 100644
--- a/source/psp/video_hardware_images.cpp
+++ b/source/psp/video_hardware_images.cpp
@@ -260,7 +260,7 @@ byte* LoadPCX (FILE *f, int matchwidth, int matchheight)
 	image_width = pcx->xmax+1;
 	image_height = pcx->ymax+1;
 
-	memcpy(image_palette, palette, sizeof(palette));
+	memcpy_vfpu(image_palette, palette, sizeof(palette));
 	image_palette_type = PAL_RGB;
 
 	fclose (f);
@@ -304,7 +304,7 @@ byte *LoadWAL (char *name)
 	size = width * height;
 
 	data = static_cast<byte*>(malloc(size));
-	memcpy(data, (byte *)mt + ofs, size);
+	memcpy_vfpu(data, (byte *)mt + ofs, size);
 
 	image_palette_type = PAL_Q2;
 
diff --git a/source/psp/video_hardware_mhex2.cpp b/source/psp/video_hardware_mhex2.cpp
index a3b3b14..343a4d7 100644
--- a/source/psp/video_hardware_mhex2.cpp
+++ b/source/psp/video_hardware_mhex2.cpp
@@ -343,7 +343,7 @@ void GL_MakeAliasModelDisplayListsH2 (model_t *m, aliashdr_t *hdr)
 
 	int* cmds = static_cast<int*>(Hunk_Alloc (numcommands * 4));
 	paliashdr->commands = (byte *)cmds - (byte *)paliashdr;
-	memcpy (cmds, commands, numcommands * 4);
+	memcpy_vfpu(cmds, commands, numcommands * 4);
 
 	trivertx_t* verts = static_cast<trivertx_t*>(Hunk_Alloc (paliashdr->numposes * paliashdr->poseverts
 		* sizeof(trivertx_t) ));
@@ -420,7 +420,7 @@ void GL_MakeAliasModelDisplayLists (model_t *m, aliashdr_t *hdr)
 
 	cmds = static_cast<int*>(Hunk_Alloc (numcommands * 4));
 	paliashdr->commands = (byte *)cmds - (byte *)paliashdr;
-	memcpy (cmds, commands, numcommands * 4);
+	memcpy_vfpu(cmds, commands, numcommands * 4);
 
 	verts = static_cast<trivertx_t*>(Hunk_Alloc (paliashdr->numposes * paliashdr->poseverts
 		* sizeof(trivertx_t) ));
diff --git a/source/psp/video_hardware_model.cpp b/source/psp/video_hardware_model.cpp
index e56760b..842a8ad 100644
--- a/source/psp/video_hardware_model.cpp
+++ b/source/psp/video_hardware_model.cpp
@@ -537,13 +537,13 @@ void Mod_LoadTextures (lump_t *l)
 		loadmodel->textures[i] = tx;
 
 
-		memcpy (tx->name, mt->name, sizeof(tx->name));
+		memcpy_vfpu(tx->name, mt->name, sizeof(tx->name));
 		tx->width = mt->width;
 		tx->height = mt->height;
 		for (j=0 ; j<MIPLEVELS ; j++)
 			tx->offsets[j] = mt->offsets[j] + sizeof(texture_t) - sizeof(miptex_t);
 		// the pixels immediately follow the structures
-		memcpy ( tx_pixels, mt+1, pixels);
+		memcpy_vfpu( tx_pixels, mt+1, pixels);
 
 		int level = 0;
 		if (r_mipmaps.value > 0)
@@ -776,7 +776,7 @@ void Mod_LoadLighting (lump_t *l)
 			return;
 		}
 		loadmodel->lightdata = static_cast<byte*>(Hunk_AllocName ( l->filelen, loadname));
-		memcpy (loadmodel->lightdata, mod_base + l->fileofs, l->filelen);
+		memcpy_vfpu(loadmodel->lightdata, mod_base + l->fileofs, l->filelen);
 		return;
 	}
 
@@ -816,7 +816,7 @@ void Mod_LoadLighting (lump_t *l)
         loadmodel->lightdata = static_cast<byte*>(Hunk_AllocName ( l->filelen*3, litfilename));
         in = loadmodel->lightdata + l->filelen*2; // place the file at the end, so it will not be overwritten until the very last write
         out = loadmodel->lightdata;
-        memcpy (in, mod_base + l->fileofs, l->filelen);
+        memcpy_vfpu(in, mod_base + l->fileofs, l->filelen);
         for (i = 0;i < l->filelen;i++)
         {
             d = *in++;
@@ -853,7 +853,7 @@ void Mod_HL_LoadLighting (lump_t *l)
 	}
 
 	loadmodel->lightdata = static_cast<byte*>(Hunk_AllocName ( l->filelen, loadname));
-	memcpy (loadmodel->lightdata, mod_base + l->fileofs, l->filelen);
+	memcpy_vfpu(loadmodel->lightdata, mod_base + l->fileofs, l->filelen);
 }
 
 /*
@@ -871,7 +871,7 @@ void Mod_LoadVisibility (lump_t *l)
 	}
 
 	loadmodel->visdata = static_cast<byte*>(Hunk_AllocName ( l->filelen, loadname));
-	memcpy (loadmodel->visdata, mod_base + l->fileofs, l->filelen);
+	memcpy_vfpu(loadmodel->visdata, mod_base + l->fileofs, l->filelen);
 }
 
 /*
@@ -980,7 +980,7 @@ void Mod_LoadEntities (lump_t *l)
 	}
 
 	loadmodel->entities = static_cast<char*>(Hunk_AllocName ( l->filelen, entfilename));
-	memcpy (loadmodel->entities, mod_base + l->fileofs, l->filelen);
+	memcpy_vfpu(loadmodel->entities, mod_base + l->fileofs, l->filelen);
 
 	if (loadmodel->bspversion == HL_BSPVERSION || loadmodel->bspversion == NZP_BSPVERSION)
 		Mod_ParseWadsFromEntityLump(loadmodel->entities);
@@ -2324,7 +2324,7 @@ void Mod_LoadAliasModel (model_t *mod, void *buffer)
 	if (!mod->cache.data)
 		return;
 
-	memcpy (mod->cache.data, pheader, total);
+	memcpy_vfpu(mod->cache.data, pheader, total);
 
 	Hunk_FreeToLowMark (start);
 }
@@ -2496,7 +2496,7 @@ void Mod_LoadH2AliasModel (model_t *mod, void *buffer)
 	Cache_Alloc (&mod->cache, total, loadname);
 	if (!mod->cache.data)
 		return;
-	memcpy (mod->cache.data, pheader, total);
+	memcpy_vfpu(mod->cache.data, pheader, total);
 
 	Hunk_FreeToLowMark (start);
 }
@@ -2650,7 +2650,7 @@ void Mod_LoadQ2AliasModel (model_t *mod, void *buffer)
 	Cache_Alloc (&mod->cache, total, loadname);
 	if (!mod->cache.data)
 		return;
-	memcpy (mod->cache.data, pheader, total);
+	memcpy_vfpu(mod->cache.data, pheader, total);
 
 	Hunk_FreeToLowMark (start);
 }
@@ -3247,7 +3247,7 @@ void Mod_LoadQ3AliasModel (model_t *mod, void *buffer)
 	if (!mod->cache.data)
 		return;
 
-	memcpy (header, buffer, com_filesize);
+	memcpy_vfpu(header, buffer, com_filesize);
 	base = com_filesize;
 
 	mod->type = mod_md3;
diff --git a/source/psp/video_hardware_resample.cpp b/source/psp/video_hardware_resample.cpp
index bfe869b..4514880 100644
--- a/source/psp/video_hardware_resample.cpp
+++ b/source/psp/video_hardware_resample.cpp
@@ -74,7 +74,7 @@ static void Image_Resample32 (void *indata, int inwidth, int inheight,void *outd
 				{
 					inrow = (byte *) indata + inwidth4 * yi;
 					if (yi == oldy + 1)
-						memcpy(row1, row2, outwidth4);
+						memcpy_vfpu(row1, row2, outwidth4);
 					else
 					Image_Resample32LerpLine (inrow, row1, inwidth, outwidth);
 					Image_Resample32LerpLine (inrow + inwidth4, row2, inwidth, outwidth);
@@ -116,12 +116,12 @@ static void Image_Resample32 (void *indata, int inwidth, int inheight,void *outd
 				{
 					inrow = (byte *) indata + inwidth4 * yi;
 					if (yi == oldy+1)
-						memcpy(row1, row2, outwidth4);
+						memcpy_vfpu(row1, row2, outwidth4);
 					else
 						Image_Resample32LerpLine (inrow, row1, inwidth, outwidth);
 					oldy = yi;
 				}
-				memcpy(out, row1, outwidth4);
+				memcpy_vfpu(out, row1, outwidth4);
 			}
 		}
 		free(memalloc);
@@ -221,7 +221,7 @@ static void Image_Resample24 (void *indata, int inwidth, int inheight,
 				if (yi != oldy) {
 					inrow = (byte *) indata + inwidth3 * yi;
 					if (yi == oldy + 1)
-						memcpy(row1, row2, outwidth3);
+						memcpy_vfpu(row1, row2, outwidth3);
 					else
 						Image_Resample24LerpLine (inrow, row1, inwidth, outwidth);
 					Image_Resample24LerpLine (inrow + inwidth3, row2, inwidth, outwidth);
@@ -259,12 +259,12 @@ static void Image_Resample24 (void *indata, int inwidth, int inheight,
 				if (yi != oldy) {
 					inrow = (byte *) indata + inwidth3 * yi;
 					if (yi == oldy+1)
-						memcpy(row1, row2, outwidth3);
+						memcpy_vfpu(row1, row2, outwidth3);
 					else
 						Image_Resample24LerpLine (inrow, row1, inwidth, outwidth);
 					oldy = yi;
 				}
-				memcpy(out, row1, outwidth3);
+				memcpy_vfpu(out, row1, outwidth3);
 			}
 		}
 		free(memalloc);
diff --git a/source/snd_mem.c b/source/snd_mem.c
index 45b5a95..8af75ea 100644
--- a/source/snd_mem.c
+++ b/source/snd_mem.c
@@ -232,7 +232,7 @@ void DumpChunks(void)
 	data_p=iff_data;
 	do
 	{
-		memcpy (str, data_p, 4);
+		memcpy_vfpu(str, data_p, 4);
 		data_p += 4;
 		iff_chunk_len = GetLittleLong();
 		Con_Printf ("0x%x : %s (%d)\n", (int)(data_p - 4), str, iff_chunk_len);
diff --git a/source/sv_main.c b/source/sv_main.c
index c099b3c..2c4918c 100644
--- a/source/sv_main.c
+++ b/source/sv_main.c
@@ -263,7 +263,7 @@ void SV_ConnectClient (int clientnum)
 	netconnection = client->netconnection;
 
 	if (sv.loadgame)
-		memcpy (spawn_parms, client->spawn_parms, sizeof(spawn_parms));
+		memcpy_vfpu(spawn_parms, client->spawn_parms, sizeof(spawn_parms));
 	memset (client, 0, sizeof(*client));
 	client->netconnection = netconnection;
 
@@ -278,7 +278,7 @@ void SV_ConnectClient (int clientnum)
 	client->privileged = false;
 
 	if (sv.loadgame)
-		memcpy (client->spawn_parms, spawn_parms, sizeof(spawn_parms));
+		memcpy_vfpu(client->spawn_parms, spawn_parms, sizeof(spawn_parms));
 	else
 	{
 	// call the progs to get default spawn parms for the new client
diff --git a/source/sv_phys.c b/source/sv_phys.c
index 2dc009e..39c4aef 100644
--- a/source/sv_phys.c
+++ b/source/sv_phys.c
@@ -1713,7 +1713,7 @@ trace_t SV_Trace_Toss (edict_t *ent, edict_t *ignore)
 	save_frametime = host_frametime;
 	host_frametime = 0.05;
 
-	memcpy (&tempent, ent, sizeof(edict_t));
+	memcpy_vfpu(&tempent, ent, sizeof(edict_t));
 	tent = &tempent;
 
 	while (1)
diff --git a/source/zone.c b/source/zone.c
index b27eecb..c64e1ca 100644
--- a/source/zone.c
+++ b/source/zone.c
@@ -48,6 +48,340 @@ typedef struct
 void Cache_FreeLow (int new_low_hunk);
 void Cache_FreeHigh (int new_high_hunk);
 
+void* memcpy_vfpu( void* dst, void* src, unsigned int size )
+{
+	u8* src8 = (u8*)src;
+	u8* dst8 = (u8*)dst;
+
+	// < 8 isn't worth trying any optimisations...
+	if (size<8) goto bytecopy;
+
+	// < 64 means we don't gain anything from using vfpu...
+	if (size<64)
+	{
+		// Align dst on 4 bytes or just resume if already done
+		while (((((u32)dst8) & 0x3)!=0) && size) {
+			*dst8++ = *src8++;
+			size--;
+		}
+		if (size<4) goto bytecopy;
+
+		// We are dst aligned now and >= 4 bytes to copy
+		u32* src32 = (u32*)src8;
+		u32* dst32 = (u32*)dst8;
+		switch(((u32)src8)&0x3)
+		{
+			case 0:
+				while (size&0xC)
+				{
+					*dst32++ = *src32++;
+					size -= 4;
+				}
+				if (size==0) return (dst);		// fast out
+				while (size>=16)
+				{
+					*dst32++ = *src32++;
+					*dst32++ = *src32++;
+					*dst32++ = *src32++;
+					*dst32++ = *src32++;
+					size -= 16;
+				}
+				if (size==0) return (dst);		// fast out
+				src8 = (u8*)src32;
+				dst8 = (u8*)dst32;
+				break;
+			default:
+				{
+					register u32 a, b, c, d;
+					while (size>=4)
+					{
+						a = *src8++;
+						b = *src8++;
+						c = *src8++;
+						d = *src8++;
+						*dst32++ = (d << 24) | (c << 16) | (b << 8) | a;
+						size -= 4;
+					}
+					if (size==0) return (dst);		// fast out
+					dst8 = (u8*)dst32;
+				}
+				break;
+		}
+		goto bytecopy;
+	}
+
+	// Align dst on 16 bytes to gain from vfpu aligned stores
+	while ((((u32)dst8) & 0xF)!=0 && size) {
+		*dst8++ = *src8++;
+		size--;
+	}
+
+	// We use uncached dst to use VFPU writeback and free cpu cache for src only
+	u8* udst8 = (u8*)((u32)dst8 | 0x40000000);
+	// We need the 64 byte aligned address to make sure the dcache is invalidated correctly
+	u8* dst64a = ((u32)dst8&~0x3F);
+	// Invalidate the first line that matches up to the dst start
+	if (size>=64)
+	asm(".set	push\n"					// save assembler option
+		".set	noreorder\n"			// suppress reordering
+		"cache 0x1B, 0(%0)\n"
+		"addiu	%0, %0, 64\n"
+		"sync\n"
+		".set	pop\n"
+		:"+r"(dst64a));
+	switch(((u32)src8&0xF))
+	{
+		// src aligned on 16 bytes too? nice!
+		case 0:
+			while (size>=64)
+			{
+				asm(".set	push\n"					// save assembler option
+					".set	noreorder\n"			// suppress reordering
+					"cache	0x1B,  0(%2)\n"			// Dcache writeback invalidate
+					"lv.q	c000,  0(%1)\n"
+					"lv.q	c010, 16(%1)\n"
+					"lv.q	c020, 32(%1)\n"
+					"lv.q	c030, 48(%1)\n"
+					"sync\n"						// Wait for allegrex writeback
+					"sv.q	c000,  0(%0), wb\n"
+					"sv.q	c010, 16(%0), wb\n"
+					"sv.q	c020, 32(%0), wb\n"
+					"sv.q	c030, 48(%0), wb\n"
+					// Lots of variable updates... but get hidden in sv.q latency anyway
+					"addiu  %3, %3, -64\n"
+					"addiu	%2, %2, 64\n"
+					"addiu	%1, %1, 64\n"
+					"addiu	%0, %0, 64\n"
+					".set	pop\n"					// restore assembler option
+					:"+r"(udst8),"+r"(src8),"+r"(dst64a),"+r"(size)
+					:
+					:"memory"
+					);
+			}
+			if (size>16)
+			{
+				// Invalidate the last cache line where the max remaining 63 bytes are
+				asm(".set	push\n"					// save assembler option
+					".set	noreorder\n"			// suppress reordering
+					"cache	0x1B, 0(%0)\n"
+					"sync\n"
+					".set	pop\n"					// restore assembler option
+					::"r"(dst64a));
+				while (size>=16)
+				{
+					asm(".set	push\n"					// save assembler option
+						".set	noreorder\n"			// suppress reordering
+						"lv.q	c000, 0(%1)\n"
+						"sv.q	c000, 0(%0), wb\n"
+						// Lots of variable updates... but get hidden in sv.q latency anyway
+						"addiu	%2, %2, -16\n"
+						"addiu	%1, %1, 16\n"
+						"addiu	%0, %0, 16\n"
+						".set	pop\n"					// restore assembler option
+						:"+r"(udst8),"+r"(src8),"+r"(size)
+						:
+						:"memory"
+						);
+				}
+			}
+			asm(".set	push\n"					// save assembler option
+				".set	noreorder\n"			// suppress reordering
+				"vflush\n"						// Flush VFPU writeback cache
+				".set	pop\n"					// restore assembler option
+				);
+			dst8 = (u8*)((u32)udst8 & ~0x40000000);
+			break;
+		// src is only qword unaligned but word aligned? We can at least use ulv.q
+		case 4:
+		case 8:
+		case 12:
+			while (size>=64)
+			{
+				asm(".set	push\n"					// save assembler option
+					".set	noreorder\n"			// suppress reordering
+					"cache	0x1B,  0(%2)\n"			// Dcache writeback invalidate
+					"ulv.q	c000,  0(%1)\n"
+					"ulv.q	c010, 16(%1)\n"
+					"ulv.q	c020, 32(%1)\n"
+					"ulv.q	c030, 48(%1)\n"
+					"sync\n"						// Wait for allegrex writeback
+					"sv.q	c000,  0(%0), wb\n"
+					"sv.q	c010, 16(%0), wb\n"
+					"sv.q	c020, 32(%0), wb\n"
+					"sv.q	c030, 48(%0), wb\n"
+					// Lots of variable updates... but get hidden in sv.q latency anyway
+					"addiu  %3, %3, -64\n"
+					"addiu	%2, %2, 64\n"
+					"addiu	%1, %1, 64\n"
+					"addiu	%0, %0, 64\n"
+					".set	pop\n"					// restore assembler option
+					:"+r"(udst8),"+r"(src8),"+r"(dst64a),"+r"(size)
+					:
+					:"memory"
+					);
+			}
+			if (size>16)
+			// Invalidate the last cache line where the max remaining 63 bytes are
+			asm(".set	push\n"					// save assembler option
+				".set	noreorder\n"			// suppress reordering
+				"cache	0x1B, 0(%0)\n"
+				"sync\n"
+				".set	pop\n"					// restore assembler option
+				::"r"(dst64a));
+			while (size>=16)
+			{
+				asm(".set	push\n"					// save assembler option
+					".set	noreorder\n"			// suppress reordering
+					"ulv.q	c000, 0(%1)\n"
+					"sv.q	c000, 0(%0), wb\n"
+					// Lots of variable updates... but get hidden in sv.q latency anyway
+					"addiu	%2, %2, -16\n"
+					"addiu	%1, %1, 16\n"
+					"addiu	%0, %0, 16\n"
+					".set	pop\n"					// restore assembler option
+					:"+r"(udst8),"+r"(src8),"+r"(size)
+					:
+					:"memory"
+					);
+			}
+			asm(".set	push\n"					// save assembler option
+				".set	noreorder\n"			// suppress reordering
+				"vflush\n"						// Flush VFPU writeback cache
+				".set	pop\n"					// restore assembler option
+				);
+			dst8 = (u8*)((u32)udst8 & ~0x40000000);
+			break;
+		// src not aligned? too bad... have to use unaligned reads
+		default:
+			while (size>=64)
+			{
+				asm(".set	push\n"					// save assembler option
+					".set	noreorder\n"			// suppress reordering
+					"cache 0x1B,  0(%2)\n"
+
+					"lwr	 $8,  0(%1)\n"			//
+					"lwl	 $8,  3(%1)\n"			// $8  = *(s + 0)
+					"lwr	 $9,  4(%1)\n"			//
+					"lwl	 $9,  7(%1)\n"			// $9  = *(s + 4)
+					"lwr	$10,  8(%1)\n"			//
+					"lwl	$10, 11(%1)\n"			// $10 = *(s + 8)
+					"lwr	$11, 12(%1)\n"			//
+					"lwl	$11, 15(%1)\n"			// $11 = *(s + 12)
+					"mtv	 $8, s000\n"
+					"mtv	 $9, s001\n"
+					"mtv	$10, s002\n"
+					"mtv	$11, s003\n"
+
+					"lwr	 $8, 16(%1)\n"
+					"lwl	 $8, 19(%1)\n"
+					"lwr	 $9, 20(%1)\n"
+					"lwl	 $9, 23(%1)\n"
+					"lwr	$10, 24(%1)\n"
+					"lwl	$10, 27(%1)\n"
+					"lwr	$11, 28(%1)\n"
+					"lwl	$11, 31(%1)\n"
+					"mtv	 $8, s010\n"
+					"mtv	 $9, s011\n"
+					"mtv	$10, s012\n"
+					"mtv	$11, s013\n"
+
+					"lwr	 $8, 32(%1)\n"
+					"lwl	 $8, 35(%1)\n"
+					"lwr	 $9, 36(%1)\n"
+					"lwl	 $9, 39(%1)\n"
+					"lwr	$10, 40(%1)\n"
+					"lwl	$10, 43(%1)\n"
+					"lwr	$11, 44(%1)\n"
+					"lwl	$11, 47(%1)\n"
+					"mtv	 $8, s020\n"
+					"mtv	 $9, s021\n"
+					"mtv	$10, s022\n"
+					"mtv	$11, s023\n"
+
+					"lwr	 $8, 48(%1)\n"
+					"lwl	 $8, 51(%1)\n"
+					"lwr	 $9, 52(%1)\n"
+					"lwl	 $9, 55(%1)\n"
+					"lwr	$10, 56(%1)\n"
+					"lwl	$10, 59(%1)\n"
+					"lwr	$11, 60(%1)\n"
+					"lwl	$11, 63(%1)\n"
+					"mtv	 $8, s030\n"
+					"mtv	 $9, s031\n"
+					"mtv	$10, s032\n"
+					"mtv	$11, s033\n"
+
+					"sync\n"
+					"sv.q 	c000,  0(%0), wb\n"
+					"sv.q 	c010, 16(%0), wb\n"
+					"sv.q 	c020, 32(%0), wb\n"
+					"sv.q 	c030, 48(%0), wb\n"
+					// Lots of variable updates... but get hidden in sv.q latency anyway
+					"addiu	%3, %3, -64\n"
+					"addiu	%2, %2, 64\n"
+					"addiu	%1, %1, 64\n"
+					"addiu	%0, %0, 64\n"
+					".set	pop\n"					// restore assembler option
+					:"+r"(udst8),"+r"(src8),"+r"(dst64a),"+r"(size)
+					:
+					:"$8","$9","$10","$11","memory"
+					);
+			}
+			if (size>16)
+			// Invalidate the last cache line where the max remaining 63 bytes are
+			asm(".set	push\n"					// save assembler option
+				".set	noreorder\n"			// suppress reordering
+				"cache	0x1B, 0(%0)\n"
+				"sync\n"
+				".set	pop\n"					// restore assembler option
+				::"r"(dst64a));
+			while (size>=16)
+			{
+				asm(".set	push\n"					// save assembler option
+					".set	noreorder\n"			// suppress reordering
+					"lwr	 $8,  0(%1)\n"			//
+					"lwl	 $8,  3(%1)\n"			// $8  = *(s + 0)
+					"lwr	 $9,  4(%1)\n"			//
+					"lwl	 $9,  7(%1)\n"			// $9  = *(s + 4)
+					"lwr	$10,  8(%1)\n"			//
+					"lwl	$10, 11(%1)\n"			// $10 = *(s + 8)
+					"lwr	$11, 12(%1)\n"			//
+					"lwl	$11, 15(%1)\n"			// $11 = *(s + 12)
+					"mtv	 $8, s000\n"
+					"mtv	 $9, s001\n"
+					"mtv	$10, s002\n"
+					"mtv	$11, s003\n"
+
+					"sv.q	c000, 0(%0), wb\n"
+					// Lots of variable updates... but get hidden in sv.q latency anyway
+					"addiu	%2, %2, -16\n"
+					"addiu	%1, %1, 16\n"
+					"addiu	%0, %0, 16\n"
+					".set	pop\n"					// restore assembler option
+					:"+r"(udst8),"+r"(src8),"+r"(size)
+					:
+					:"$8","$9","$10","$11","memory"
+					);
+			}
+			asm(".set	push\n"					// save assembler option
+				".set	noreorder\n"			// suppress reordering
+				"vflush\n"						// Flush VFPU writeback cache
+				".set	pop\n"					// restore assembler option
+				);
+			dst8 = (u8*)((u32)udst8 & ~0x40000000);
+			break;
+	}
+
+bytecopy:
+	// Copy the remains byte per byte...
+	while (size--)
+	{
+		*dst8++ = *src8++;
+	}
+
+	return (dst);
+}
+
 
 /*
 ==============================================================================
@@ -374,7 +708,7 @@ void Hunk_Print (qboolean all)
 	//
 	// print the single block
 	//
-		memcpy (name, h->name, 8);
+		memcpy_vfpu(name, h->name, 8);
 		if (all)
 			Con_Printf ("%8p :%8i %8s\n",h, h->size, name);
 
diff --git a/source/zone.h b/source/zone.h
index 7655ad9..6445428 100644
--- a/source/zone.h
+++ b/source/zone.h
@@ -127,5 +127,5 @@ void *Cache_Alloc (cache_user_t *c, int size, char *name);
 
 void Cache_Report (void);
 
-
+void* memcpy_vfpu(void* dst, void* src, unsigned int size);