summaryrefslogtreecommitdiff
path: root/shared-core/via_drm.h
blob: e6a8ec647f8fc4b9040774bf9402477cfc5d1306 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
/*
 * Copyright 1998-2003 VIA Technologies, Inc. All Rights Reserved.
 * Copyright 2001-2003 S3 Graphics, Inc. All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sub license,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice (including the
 * next paragraph) shall be included in all copies or substantial portions
 * of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
 * VIA, S3 GRAPHICS, AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
 * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
 * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
 * DEALINGS IN THE SOFTWARE.
 */
#ifndef _VIA_DRM_H_
#define _VIA_DRM_H_

/* WARNING: These defines must be the same as what the Xserver uses.
 * if you change them, you must change the defines in the Xserver.
 */

#ifndef _VIA_DEFINES_
#define _VIA_DEFINES_


#if !defined(__KERNEL__) && !defined(_KERNEL)
#include "via_drmclient.h"
#endif

/*
 * With the arrival of libdrm there is a need to version this file.
 * As usual, bump MINOR for new features, MAJOR for changes that create
 * backwards incompatibilities, (which should be avoided whenever possible).
 */

#define VIA_DRM_DRIVER_DATE		"20070202"

#define VIA_DRM_DRIVER_MAJOR		2
#define VIA_DRM_DRIVER_MINOR		11
#define VIA_DRM_DRIVER_PATCHLEVEL	1
#define VIA_DRM_DRIVER_VERSION	  (((VIA_DRM_DRIVER_MAJOR) << 16) | (VIA_DRM_DRIVER_MINOR))

#define VIA_NR_SAREA_CLIPRECTS		8
#define VIA_NR_XVMC_PORTS	       10
#define VIA_NR_XVMC_LOCKS	       5
#define VIA_MAX_CACHELINE_SIZE	  64
#define XVMCLOCKPTR(saPriv,lockNo)					\
	((volatile struct drm_hw_lock *)(((((unsigned long) (saPriv)->XvMCLockArea) + \
				      (VIA_MAX_CACHELINE_SIZE - 1)) &	\
				     ~(VIA_MAX_CACHELINE_SIZE - 1)) +	\
				    VIA_MAX_CACHELINE_SIZE*(lockNo)))
#define VIA_NR_TEX_REGIONS 64

#endif

#define DRM_VIA_FENCE_TYPE_ACCEL 0x00000002

/* VIA specific ioctls */
#define DRM_VIA_ALLOCMEM	0x00
#define DRM_VIA_FREEMEM		0x01
#define DRM_VIA_AGP_INIT	0x02
#define DRM_VIA_FB_INIT		0x03
#define DRM_VIA_MAP_INIT	0x04
#define DRM_VIA_DEC_FUTEX       0x05
#define NOT_USED
#define DRM_VIA_DMA_INIT	0x07
#define DRM_VIA_CMDBUFFER	0x08
#define DRM_VIA_FLUSH		0x09
#define DRM_VIA_PCICMD		0x0a
#define DRM_VIA_CMDBUF_SIZE	0x0b
#define NOT_USED
#define DRM_VIA_WAIT_IRQ	0x0d
#define DRM_VIA_DMA_BLIT	0x0e
#define DRM_VIA_BLIT_SYNC       0x0f

#define DRM_IOCTL_VIA_ALLOCMEM	  DRM_IOWR(DRM_COMMAND_BASE + DRM_VIA_ALLOCMEM, drm_via_mem_t)
#define DRM_IOCTL_VIA_FREEMEM	  DRM_IOW( DRM_COMMAND_BASE + DRM_VIA_FREEMEM, drm_via_mem_t)
#define DRM_IOCTL_VIA_AGP_INIT	  DRM_IOWR(DRM_COMMAND_BASE + DRM_VIA_AGP_INIT, drm_via_agp_t)
#define DRM_IOCTL_VIA_FB_INIT	  DRM_IOWR(DRM_COMMAND_BASE + DRM_VIA_FB_INIT, drm_via_fb_t)
#define DRM_IOCTL_VIA_MAP_INIT	  DRM_IOWR(DRM_COMMAND_BASE + DRM_VIA_MAP_INIT, drm_via_init_t)
#define DRM_IOCTL_VIA_DEC_FUTEX   DRM_IOW( DRM_COMMAND_BASE + DRM_VIA_DEC_FUTEX, drm_via_futex_t)
#define DRM_IOCTL_VIA_DMA_INIT	  DRM_IOWR(DRM_COMMAND_BASE + DRM_VIA_DMA_INIT, drm_via_dma_init_t)
#define DRM_IOCTL_VIA_CMDBUFFER	  DRM_IOW( DRM_COMMAND_BASE + DRM_VIA_CMDBUFFER, drm_via_cmdbuffer_t)
#define DRM_IOCTL_VIA_FLUSH	  DRM_IO(  DRM_COMMAND_BASE + DRM_VIA_FLUSH)
#define DRM_IOCTL_VIA_PCICMD	  DRM_IOW( DRM_COMMAND_BASE + DRM_VIA_PCICMD, drm_via_cmdbuffer_t)
#define DRM_IOCTL_VIA_CMDBUF_SIZE DRM_IOWR( DRM_COMMAND_BASE + DRM_VIA_CMDBUF_SIZE, \
					    drm_via_cmdbuf_size_t)
#define DRM_IOCTL_VIA_WAIT_IRQ    DRM_IOWR( DRM_COMMAND_BASE + DRM_VIA_WAIT_IRQ, drm_via_irqwait_t)
#define DRM_IOCTL_VIA_DMA_BLIT    DRM_IOW(DRM_COMMAND_BASE + DRM_VIA_DMA_BLIT, drm_via_dmablit_t)
#define DRM_IOCTL_VIA_BLIT_SYNC   DRM_IOW(DRM_COMMAND_BASE + DRM_VIA_BLIT_SYNC, drm_via_blitsync_t)

/* Indices into buf.Setup where various bits of state are mirrored per
 * context and per buffer.  These can be fired at the card as a unit,
 * or in a piecewise fashion as required.
 */

#define VIA_TEX_SETUP_SIZE 8

/* Flags for clear ioctl
 */
#define VIA_FRONT   0x1
#define VIA_BACK    0x2
#define VIA_DEPTH   0x4
#define VIA_STENCIL 0x8

#define VIA_MEM_VIDEO   0	/* matches drm constant */
#define VIA_MEM_AGP     1	/* matches drm constant */
#define VIA_MEM_SYSTEM  2
#define VIA_MEM_MIXED   3
#define VIA_MEM_UNKNOWN 4

typedef struct {
	uint32_t offset;
	uint32_t size;
} drm_via_agp_t;

typedef struct {
	uint32_t offset;
	uint32_t size;
} drm_via_fb_t;

typedef struct {
	uint32_t context;
	uint32_t type;
	uint32_t size;
	unsigned long index;
	unsigned long offset;
} drm_via_mem_t;

typedef struct _drm_via_init {
	enum {
		VIA_INIT_MAP = 0x01,
		VIA_CLEANUP_MAP = 0x02
	} func;

	unsigned long sarea_priv_offset;
	unsigned long fb_offset;
	unsigned long mmio_offset;
	unsigned long agpAddr;
} drm_via_init_t;

typedef struct _drm_via_futex {
	enum {
		VIA_FUTEX_WAIT = 0x00,
		VIA_FUTEX_WAKE = 0X01
	} func;
	uint32_t ms;
	uint32_t lock;
	uint32_t val;
} drm_via_futex_t;

typedef struct _drm_via_dma_init {
	enum {
		VIA_INIT_DMA = 0x01,
		VIA_CLEANUP_DMA = 0x02,
		VIA_DMA_INITIALIZED = 0x03
	} func;

	unsigned long offset;
	unsigned long size;
	unsigned long reg_pause_addr;
} drm_via_dma_init_t;

typedef struct _drm_via_cmdbuffer {
	char __user *buf;
	unsigned long size;
} drm_via_cmdbuffer_t;

/* Warning: If you change the SAREA structure you must change the Xserver
 * structure as well */

typedef struct _drm_via_tex_region {
	unsigned char next, prev;	/* indices to form a circular LRU  */
	unsigned char inUse;	/* owned by a client, or free? */
	int age;		/* tracked by clients to update local LRU's */
} drm_via_tex_region_t;

typedef struct _drm_via_sarea {
	unsigned int dirty;
	unsigned int nbox;
	struct drm_clip_rect boxes[VIA_NR_SAREA_CLIPRECTS];
	drm_via_tex_region_t texList[VIA_NR_TEX_REGIONS + 1];
	int texAge;		/* last time texture was uploaded */
	int ctxOwner;		/* last context to upload state */
	int vertexPrim;

	/*
	 * Below is for XvMC.
	 * We want the lock integers alone on, and aligned to, a cache line.
	 * Therefore this somewhat strange construct.
	 */

	char XvMCLockArea[VIA_MAX_CACHELINE_SIZE * (VIA_NR_XVMC_LOCKS + 1)];

	unsigned int XvMCDisplaying[VIA_NR_XVMC_PORTS];
	unsigned int XvMCSubPicOn[VIA_NR_XVMC_PORTS];
	unsigned int XvMCCtxNoGrabbed;	/* Last context to hold decoder */

	/* Used by the 3d driver only at this point, for pageflipping:
	 */
	unsigned int pfCurrentOffset;
} drm_via_sarea_t;

typedef struct _drm_via_cmdbuf_size {
	enum {
		VIA_CMDBUF_SPACE = 0x01,
		VIA_CMDBUF_LAG = 0x02
	} func;
	int wait;
	uint32_t size;
} drm_via_cmdbuf_size_t;

typedef enum {
	VIA_IRQ_ABSOLUTE = 0x0,
	VIA_IRQ_RELATIVE = 0x1,
	VIA_IRQ_SIGNAL = 0x10000000,
	VIA_IRQ_FORCE_SEQUENCE = 0x20000000
} via_irq_seq_type_t;

#define VIA_IRQ_FLAGS_MASK 0xF0000000

enum drm_via_irqs {
	drm_via_irq_hqv0 = 0,
	drm_via_irq_hqv1,
	drm_via_irq_dma0_dd,
	drm_via_irq_dma0_td,
	drm_via_irq_dma1_dd,
	drm_via_irq_dma1_td,
	drm_via_irq_num
};

struct drm_via_wait_irq_request {
	unsigned irq;
	via_irq_seq_type_t type;
	uint32_t sequence;
	uint32_t signal;
};

typedef union drm_via_irqwait {
	struct drm_via_wait_irq_request request;
	struct drm_wait_vblank_reply reply;
} drm_via_irqwait_t;

typedef struct drm_via_blitsync {
	uint32_t sync_handle;
	unsigned engine;
} drm_via_blitsync_t;

/*
 * Below,"flags" is currently unused but will be used for possible future
 * extensions like kernel space bounce buffers for bad alignments and
 * blit engine busy-wait polling for better latency in the absence of
 * interrupts.
 */

typedef struct drm_via_dmablit {
	uint32_t num_lines;
	uint32_t line_length;

	uint32_t fb_addr;
	uint32_t fb_stride;

	unsigned char *mem_addr;
	uint32_t mem_stride;

	uint32_t flags;
	int to_fb;

	drm_via_blitsync_t sync;
} drm_via_dmablit_t;


#endif				/* _VIA_DRM_H_ */
867' href='#n867'>867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713
/* r128_state.c -- State support for r128 -*- linux-c -*-
 * Created: Thu Jan 27 02:53:43 2000 by gareth@valinux.com
 */
/*
 * Copyright 2000 VA Linux Systems, Inc., Sunnyvale, California.
 * All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice (including the next
 * paragraph) shall be included in all copies or substantial portions of the
 * Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
 * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
 * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
 * DEALINGS IN THE SOFTWARE.
 *
 * Authors:
 *    Gareth Hughes <gareth@valinux.com>
 */

#include "drmP.h"
#include "drm.h"
#include "r128_drm.h"
#include "r128_drv.h"

/* ================================================================
 * CCE hardware state programming functions
 */

static void r128_emit_clip_rects(drm_r128_private_t * dev_priv,
				 drm_clip_rect_t * boxes, int count)
{
	u32 aux_sc_cntl = 0x00000000;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING((count < 3 ? count : 3) * 5 + 2);

	if (count >= 1) {
		OUT_RING(CCE_PACKET0(R128_AUX1_SC_LEFT, 3));
		OUT_RING(boxes[0].x1);
		OUT_RING(boxes[0].x2 - 1);
		OUT_RING(boxes[0].y1);
		OUT_RING(boxes[0].y2 - 1);

		aux_sc_cntl |= (R128_AUX1_SC_EN | R128_AUX1_SC_MODE_OR);
	}
	if (count >= 2) {
		OUT_RING(CCE_PACKET0(R128_AUX2_SC_LEFT, 3));
		OUT_RING(boxes[1].x1);
		OUT_RING(boxes[1].x2 - 1);
		OUT_RING(boxes[1].y1);
		OUT_RING(boxes[1].y2 - 1);

		aux_sc_cntl |= (R128_AUX2_SC_EN | R128_AUX2_SC_MODE_OR);
	}
	if (count >= 3) {
		OUT_RING(CCE_PACKET0(R128_AUX3_SC_LEFT, 3));
		OUT_RING(boxes[2].x1);
		OUT_RING(boxes[2].x2 - 1);
		OUT_RING(boxes[2].y1);
		OUT_RING(boxes[2].y2 - 1);

		aux_sc_cntl |= (R128_AUX3_SC_EN | R128_AUX3_SC_MODE_OR);
	}

	OUT_RING(CCE_PACKET0(R128_AUX_SC_CNTL, 0));
	OUT_RING(aux_sc_cntl);

	ADVANCE_RING();
}

static __inline__ void r128_emit_core(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	drm_r128_context_regs_t *ctx = &sarea_priv->context_state;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING(2);

	OUT_RING(CCE_PACKET0(R128_SCALE_3D_CNTL, 0));
	OUT_RING(ctx->scale_3d_cntl);

	ADVANCE_RING();
}

static __inline__ void r128_emit_context(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	drm_r128_context_regs_t *ctx = &sarea_priv->context_state;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING(13);

	OUT_RING(CCE_PACKET0(R128_DST_PITCH_OFFSET_C, 11));
	OUT_RING(ctx->dst_pitch_offset_c);
	OUT_RING(ctx->dp_gui_master_cntl_c);
	OUT_RING(ctx->sc_top_left_c);
	OUT_RING(ctx->sc_bottom_right_c);
	OUT_RING(ctx->z_offset_c);
	OUT_RING(ctx->z_pitch_c);
	OUT_RING(ctx->z_sten_cntl_c);
	OUT_RING(ctx->tex_cntl_c);
	OUT_RING(ctx->misc_3d_state_cntl_reg);
	OUT_RING(ctx->texture_clr_cmp_clr_c);
	OUT_RING(ctx->texture_clr_cmp_msk_c);
	OUT_RING(ctx->fog_color_c);

	ADVANCE_RING();
}

static __inline__ void r128_emit_setup(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	drm_r128_context_regs_t *ctx = &sarea_priv->context_state;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING(3);

	OUT_RING(CCE_PACKET1(R128_SETUP_CNTL, R128_PM4_VC_FPU_SETUP));
	OUT_RING(ctx->setup_cntl);
	OUT_RING(ctx->pm4_vc_fpu_setup);

	ADVANCE_RING();
}

static __inline__ void r128_emit_masks(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	drm_r128_context_regs_t *ctx = &sarea_priv->context_state;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING(5);

	OUT_RING(CCE_PACKET0(R128_DP_WRITE_MASK, 0));
	OUT_RING(ctx->dp_write_mask);

	OUT_RING(CCE_PACKET0(R128_STEN_REF_MASK_C, 1));
	OUT_RING(ctx->sten_ref_mask_c);
	OUT_RING(ctx->plane_3d_mask_c);

	ADVANCE_RING();
}

static __inline__ void r128_emit_window(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	drm_r128_context_regs_t *ctx = &sarea_priv->context_state;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING(2);

	OUT_RING(CCE_PACKET0(R128_WINDOW_XY_OFFSET, 0));
	OUT_RING(ctx->window_xy_offset);

	ADVANCE_RING();
}

static __inline__ void r128_emit_tex0(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	drm_r128_context_regs_t *ctx = &sarea_priv->context_state;
	drm_r128_texture_regs_t *tex = &sarea_priv->tex_state[0];
	int i;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING(7 + R128_MAX_TEXTURE_LEVELS);

	OUT_RING(CCE_PACKET0(R128_PRIM_TEX_CNTL_C,
			     2 + R128_MAX_TEXTURE_LEVELS));
	OUT_RING(tex->tex_cntl);
	OUT_RING(tex->tex_combine_cntl);
	OUT_RING(ctx->tex_size_pitch_c);
	for (i = 0; i < R128_MAX_TEXTURE_LEVELS; i++) {
		OUT_RING(tex->tex_offset[i]);
	}

	OUT_RING(CCE_PACKET0(R128_CONSTANT_COLOR_C, 1));
	OUT_RING(ctx->constant_color_c);
	OUT_RING(tex->tex_border_color);

	ADVANCE_RING();
}

static __inline__ void r128_emit_tex1(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	drm_r128_texture_regs_t *tex = &sarea_priv->tex_state[1];
	int i;
	RING_LOCALS;
	DRM_DEBUG("    %s\n", __FUNCTION__);

	BEGIN_RING(5 + R128_MAX_TEXTURE_LEVELS);

	OUT_RING(CCE_PACKET0(R128_SEC_TEX_CNTL_C, 1 + R128_MAX_TEXTURE_LEVELS));
	OUT_RING(tex->tex_cntl);
	OUT_RING(tex->tex_combine_cntl);
	for (i = 0; i < R128_MAX_TEXTURE_LEVELS; i++) {
		OUT_RING(tex->tex_offset[i]);
	}

	OUT_RING(CCE_PACKET0(R128_SEC_TEXTURE_BORDER_COLOR_C, 0));
	OUT_RING(tex->tex_border_color);

	ADVANCE_RING();
}

static __inline__ void r128_emit_state(drm_r128_private_t * dev_priv)
{
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	unsigned int dirty = sarea_priv->dirty;

	DRM_DEBUG("%s: dirty=0x%08x\n", __FUNCTION__, dirty);

	if (dirty & R128_UPLOAD_CORE) {
		r128_emit_core(dev_priv);
		sarea_priv->dirty &= ~R128_UPLOAD_CORE;
	}

	if (dirty & R128_UPLOAD_CONTEXT) {
		r128_emit_context(dev_priv);
		sarea_priv->dirty &= ~R128_UPLOAD_CONTEXT;
	}

	if (dirty & R128_UPLOAD_SETUP) {
		r128_emit_setup(dev_priv);
		sarea_priv->dirty &= ~R128_UPLOAD_SETUP;
	}

	if (dirty & R128_UPLOAD_MASKS) {
		r128_emit_masks(dev_priv);
		sarea_priv->dirty &= ~R128_UPLOAD_MASKS;
	}

	if (dirty & R128_UPLOAD_WINDOW) {
		r128_emit_window(dev_priv);
		sarea_priv->dirty &= ~R128_UPLOAD_WINDOW;
	}

	if (dirty & R128_UPLOAD_TEX0) {
		r128_emit_tex0(dev_priv);
		sarea_priv->dirty &= ~R128_UPLOAD_TEX0;
	}

	if (dirty & R128_UPLOAD_TEX1) {
		r128_emit_tex1(dev_priv);
		sarea_priv->dirty &= ~R128_UPLOAD_TEX1;
	}

	/* Turn off the texture cache flushing */
	sarea_priv->context_state.tex_cntl_c &= ~R128_TEX_CACHE_FLUSH;

	sarea_priv->dirty &= ~R128_REQUIRE_QUIESCENCE;
}

#if R128_PERFORMANCE_BOXES
/* ================================================================
 * Performance monitoring functions
 */

static void r128_clear_box(drm_r128_private_t * dev_priv,
			   int x, int y, int w, int h, int r, int g, int b)
{
	u32 pitch, offset;
	u32 fb_bpp, color;
	RING_LOCALS;

	switch (dev_priv->fb_bpp) {
	case 16:
		fb_bpp = R128_GMC_DST_16BPP;
		color = (((r & 0xf8) << 8) |
			 ((g & 0xfc) << 3) | ((b & 0xf8) >> 3));
		break;
	case 24:
		fb_bpp = R128_GMC_DST_24BPP;
		color = ((r << 16) | (g << 8) | b);
		break;
	case 32:
		fb_bpp = R128_GMC_DST_32BPP;
		color = (((0xff) << 24) | (r << 16) | (g << 8) | b);
		break;
	default:
		return;
	}

	offset = dev_priv->back_offset;
	pitch = dev_priv->back_pitch >> 3;

	BEGIN_RING(6);

	OUT_RING(CCE_PACKET3(R128_CNTL_PAINT_MULTI, 4));
	OUT_RING(R128_GMC_DST_PITCH_OFFSET_CNTL |
		 R128_GMC_BRUSH_SOLID_COLOR |
		 fb_bpp |
		 R128_GMC_SRC_DATATYPE_COLOR |
		 R128_ROP3_P |
		 R128_GMC_CLR_CMP_CNTL_DIS | R128_GMC_AUX_CLIP_DIS);

	OUT_RING((pitch << 21) | (offset >> 5));
	OUT_RING(color);

	OUT_RING((x << 16) | y);
	OUT_RING((w << 16) | h);

	ADVANCE_RING();
}

static void r128_cce_performance_boxes(drm_r128_private_t * dev_priv)
{
	if (atomic_read(&dev_priv->idle_count) == 0) {
		r128_clear_box(dev_priv, 64, 4, 8, 8, 0, 255, 0);
	} else {
		atomic_set(&dev_priv->idle_count, 0);
	}
}

#endif

/* ================================================================
 * CCE command dispatch functions
 */

static void r128_print_dirty(const char *msg, unsigned int flags)
{
	DRM_INFO("%s: (0x%x) %s%s%s%s%s%s%s%s%s\n",
		 msg,
		 flags,
		 (flags & R128_UPLOAD_CORE) ? "core, " : "",
		 (flags & R128_UPLOAD_CONTEXT) ? "context, " : "",
		 (flags & R128_UPLOAD_SETUP) ? "setup, " : "",
		 (flags & R128_UPLOAD_TEX0) ? "tex0, " : "",
		 (flags & R128_UPLOAD_TEX1) ? "tex1, " : "",
		 (flags & R128_UPLOAD_MASKS) ? "masks, " : "",
		 (flags & R128_UPLOAD_WINDOW) ? "window, " : "",
		 (flags & R128_UPLOAD_CLIPRECTS) ? "cliprects, " : "",
		 (flags & R128_REQUIRE_QUIESCENCE) ? "quiescence, " : "");
}

static void r128_cce_dispatch_clear(drm_device_t * dev,
				    drm_r128_clear_t * clear)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	int nbox = sarea_priv->nbox;
	drm_clip_rect_t *pbox = sarea_priv->boxes;
	unsigned int flags = clear->flags;
	int i;
	RING_LOCALS;
	DRM_DEBUG("%s\n", __FUNCTION__);

	if (dev_priv->page_flipping && dev_priv->current_page == 1) {
		unsigned int tmp = flags;

		flags &= ~(R128_FRONT | R128_BACK);
		if (tmp & R128_FRONT)
			flags |= R128_BACK;
		if (tmp & R128_BACK)
			flags |= R128_FRONT;
	}

	for (i = 0; i < nbox; i++) {
		int x = pbox[i].x1;
		int y = pbox[i].y1;
		int w = pbox[i].x2 - x;
		int h = pbox[i].y2 - y;

		DRM_DEBUG("dispatch clear %d,%d-%d,%d flags 0x%x\n",
			  pbox[i].x1, pbox[i].y1, pbox[i].x2,
			  pbox[i].y2, flags);

		if (flags & (R128_FRONT | R128_BACK)) {
			BEGIN_RING(2);

			OUT_RING(CCE_PACKET0(R128_DP_WRITE_MASK, 0));
			OUT_RING(clear->color_mask);

			ADVANCE_RING();
		}

		if (flags & R128_FRONT) {
			BEGIN_RING(6);

			OUT_RING(CCE_PACKET3(R128_CNTL_PAINT_MULTI, 4));
			OUT_RING(R128_GMC_DST_PITCH_OFFSET_CNTL |
				 R128_GMC_BRUSH_SOLID_COLOR |
				 (dev_priv->color_fmt << 8) |
				 R128_GMC_SRC_DATATYPE_COLOR |
				 R128_ROP3_P |
				 R128_GMC_CLR_CMP_CNTL_DIS |
				 R128_GMC_AUX_CLIP_DIS);

			OUT_RING(dev_priv->front_pitch_offset_c);
			OUT_RING(clear->clear_color);

			OUT_RING((x << 16) | y);
			OUT_RING((w << 16) | h);

			ADVANCE_RING();
		}

		if (flags & R128_BACK) {
			BEGIN_RING(6);

			OUT_RING(CCE_PACKET3(R128_CNTL_PAINT_MULTI, 4));
			OUT_RING(R128_GMC_DST_PITCH_OFFSET_CNTL |
				 R128_GMC_BRUSH_SOLID_COLOR |
				 (dev_priv->color_fmt << 8) |
				 R128_GMC_SRC_DATATYPE_COLOR |
				 R128_ROP3_P |
				 R128_GMC_CLR_CMP_CNTL_DIS |
				 R128_GMC_AUX_CLIP_DIS);

			OUT_RING(dev_priv->back_pitch_offset_c);
			OUT_RING(clear->clear_color);

			OUT_RING((x << 16) | y);
			OUT_RING((w << 16) | h);

			ADVANCE_RING();
		}

		if (flags & R128_DEPTH) {
			BEGIN_RING(6);

			OUT_RING(CCE_PACKET3(R128_CNTL_PAINT_MULTI, 4));
			OUT_RING(R128_GMC_DST_PITCH_OFFSET_CNTL |
				 R128_GMC_BRUSH_SOLID_COLOR |
				 (dev_priv->depth_fmt << 8) |
				 R128_GMC_SRC_DATATYPE_COLOR |
				 R128_ROP3_P |
				 R128_GMC_CLR_CMP_CNTL_DIS |
				 R128_GMC_AUX_CLIP_DIS | R128_GMC_WR_MSK_DIS);

			OUT_RING(dev_priv->depth_pitch_offset_c);
			OUT_RING(clear->clear_depth);

			OUT_RING((x << 16) | y);
			OUT_RING((w << 16) | h);

			ADVANCE_RING();
		}
	}
}

static void r128_cce_dispatch_swap(drm_device_t * dev)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	int nbox = sarea_priv->nbox;
	drm_clip_rect_t *pbox = sarea_priv->boxes;
	int i;
	RING_LOCALS;
	DRM_DEBUG("%s\n", __FUNCTION__);

#if R128_PERFORMANCE_BOXES
	/* Do some trivial performance monitoring...
	 */
	r128_cce_performance_boxes(dev_priv);
#endif

	for (i = 0; i < nbox; i++) {
		int x = pbox[i].x1;
		int y = pbox[i].y1;
		int w = pbox[i].x2 - x;
		int h = pbox[i].y2 - y;

		BEGIN_RING(7);

		OUT_RING(CCE_PACKET3(R128_CNTL_BITBLT_MULTI, 5));
		OUT_RING(R128_GMC_SRC_PITCH_OFFSET_CNTL |
			 R128_GMC_DST_PITCH_OFFSET_CNTL |
			 R128_GMC_BRUSH_NONE |
			 (dev_priv->color_fmt << 8) |
			 R128_GMC_SRC_DATATYPE_COLOR |
			 R128_ROP3_S |
			 R128_DP_SRC_SOURCE_MEMORY |
			 R128_GMC_CLR_CMP_CNTL_DIS |
			 R128_GMC_AUX_CLIP_DIS | R128_GMC_WR_MSK_DIS);

		/* Make this work even if front & back are flipped:
		 */
		if (dev_priv->current_page == 0) {
			OUT_RING(dev_priv->back_pitch_offset_c);
			OUT_RING(dev_priv->front_pitch_offset_c);
		} else {
			OUT_RING(dev_priv->front_pitch_offset_c);
			OUT_RING(dev_priv->back_pitch_offset_c);
		}

		OUT_RING((x << 16) | y);
		OUT_RING((x << 16) | y);
		OUT_RING((w << 16) | h);

		ADVANCE_RING();
	}

	/* Increment the frame counter.  The client-side 3D driver must
	 * throttle the framerate by waiting for this value before
	 * performing the swapbuffer ioctl.
	 */
	dev_priv->sarea_priv->last_frame++;

	BEGIN_RING(2);

	OUT_RING(CCE_PACKET0(R128_LAST_FRAME_REG, 0));
	OUT_RING(dev_priv->sarea_priv->last_frame);

	ADVANCE_RING();
}

static void r128_cce_dispatch_flip(drm_device_t * dev)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	RING_LOCALS;
	DRM_DEBUG("%s: page=%d pfCurrentPage=%d\n",
		  __FUNCTION__,
		  dev_priv->current_page, dev_priv->sarea_priv->pfCurrentPage);

#if R128_PERFORMANCE_BOXES
	/* Do some trivial performance monitoring...
	 */
	r128_cce_performance_boxes(dev_priv);
#endif

	BEGIN_RING(4);

	R128_WAIT_UNTIL_PAGE_FLIPPED();
	OUT_RING(CCE_PACKET0(R128_CRTC_OFFSET, 0));

	if (dev_priv->current_page == 0) {
		OUT_RING(dev_priv->back_offset);
	} else {
		OUT_RING(dev_priv->front_offset);
	}

	ADVANCE_RING();

	/* Increment the frame counter.  The client-side 3D driver must
	 * throttle the framerate by waiting for this value before
	 * performing the swapbuffer ioctl.
	 */
	dev_priv->sarea_priv->last_frame++;
	dev_priv->sarea_priv->pfCurrentPage = dev_priv->current_page =
	    1 - dev_priv->current_page;

	BEGIN_RING(2);

	OUT_RING(CCE_PACKET0(R128_LAST_FRAME_REG, 0));
	OUT_RING(dev_priv->sarea_priv->last_frame);

	ADVANCE_RING();
}

static void r128_cce_dispatch_vertex(drm_device_t * dev, drm_buf_t * buf)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	drm_r128_buf_priv_t *buf_priv = buf->dev_private;
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	int format = sarea_priv->vc_format;
	int offset = buf->bus_address;
	int size = buf->used;
	int prim = buf_priv->prim;
	int i = 0;
	RING_LOCALS;
	DRM_DEBUG("buf=%d nbox=%d\n", buf->idx, sarea_priv->nbox);

	if (0)
		r128_print_dirty("dispatch_vertex", sarea_priv->dirty);

	if (buf->used) {
		buf_priv->dispatched = 1;

		if (sarea_priv->dirty & ~R128_UPLOAD_CLIPRECTS) {
			r128_emit_state(dev_priv);
		}

		do {
			/* Emit the next set of up to three cliprects */
			if (i < sarea_priv->nbox) {
				r128_emit_clip_rects(dev_priv,
						     &sarea_priv->boxes[i],
						     sarea_priv->nbox - i);
			}

			/* Emit the vertex buffer rendering commands */
			BEGIN_RING(5);

			OUT_RING(CCE_PACKET3(R128_3D_RNDR_GEN_INDX_PRIM, 3));
			OUT_RING(offset);
			OUT_RING(size);
			OUT_RING(format);
			OUT_RING(prim | R128_CCE_VC_CNTL_PRIM_WALK_LIST |
				 (size << R128_CCE_VC_CNTL_NUM_SHIFT));

			ADVANCE_RING();

			i += 3;
		} while (i < sarea_priv->nbox);
	}

	if (buf_priv->discard) {
		buf_priv->age = dev_priv->sarea_priv->last_dispatch;

		/* Emit the vertex buffer age */
		BEGIN_RING(2);

		OUT_RING(CCE_PACKET0(R128_LAST_DISPATCH_REG, 0));
		OUT_RING(buf_priv->age);

		ADVANCE_RING();

		buf->pending = 1;
		buf->used = 0;
		/* FIXME: Check dispatched field */
		buf_priv->dispatched = 0;
	}

	dev_priv->sarea_priv->last_dispatch++;

	sarea_priv->dirty &= ~R128_UPLOAD_CLIPRECTS;
	sarea_priv->nbox = 0;
}

static void r128_cce_dispatch_indirect(drm_device_t * dev,
				       drm_buf_t * buf, int start, int end)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	drm_r128_buf_priv_t *buf_priv = buf->dev_private;
	RING_LOCALS;
	DRM_DEBUG("indirect: buf=%d s=0x%x e=0x%x\n", buf->idx, start, end);

	if (start != end) {
		int offset = buf->bus_address + start;
		int dwords = (end - start + 3) / sizeof(u32);

		/* Indirect buffer data must be an even number of
		 * dwords, so if we've been given an odd number we must
		 * pad the data with a Type-2 CCE packet.
		 */
		if (dwords & 1) {
			u32 *data = (u32 *)
			    ((char *)dev->agp_buffer_map->handle
			     + buf->offset + start);
			data[dwords++] = cpu_to_le32(R128_CCE_PACKET2);
		}

		buf_priv->dispatched = 1;

		/* Fire off the indirect buffer */
		BEGIN_RING(3);

		OUT_RING(CCE_PACKET0(R128_PM4_IW_INDOFF, 1));
		OUT_RING(offset);
		OUT_RING(dwords);

		ADVANCE_RING();
	}

	if (buf_priv->discard) {
		buf_priv->age = dev_priv->sarea_priv->last_dispatch;

		/* Emit the indirect buffer age */
		BEGIN_RING(2);

		OUT_RING(CCE_PACKET0(R128_LAST_DISPATCH_REG, 0));
		OUT_RING(buf_priv->age);

		ADVANCE_RING();

		buf->pending = 1;
		buf->used = 0;
		/* FIXME: Check dispatched field */
		buf_priv->dispatched = 0;
	}

	dev_priv->sarea_priv->last_dispatch++;
}

static void r128_cce_dispatch_indices(drm_device_t * dev,
				      drm_buf_t * buf,
				      int start, int end, int count)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	drm_r128_buf_priv_t *buf_priv = buf->dev_private;
	drm_r128_sarea_t *sarea_priv = dev_priv->sarea_priv;
	int format = sarea_priv->vc_format;
	int offset = dev->agp_buffer_map->offset - dev_priv->cce_buffers_offset;
	int prim = buf_priv->prim;
	u32 *data;
	int dwords;
	int i = 0;
	RING_LOCALS;
	DRM_DEBUG("indices: s=%d e=%d c=%d\n", start, end, count);

	if (0)
		r128_print_dirty("dispatch_indices", sarea_priv->dirty);

	if (start != end) {
		buf_priv->dispatched = 1;

		if (sarea_priv->dirty & ~R128_UPLOAD_CLIPRECTS) {
			r128_emit_state(dev_priv);
		}

		dwords = (end - start + 3) / sizeof(u32);

		data = (u32 *) ((char *)dev->agp_buffer_map->handle
				+ buf->offset + start);

		data[0] = cpu_to_le32(CCE_PACKET3(R128_3D_RNDR_GEN_INDX_PRIM,
						  dwords - 2));

		data[1] = cpu_to_le32(offset);
		data[2] = cpu_to_le32(R128_MAX_VB_VERTS);
		data[3] = cpu_to_le32(format);
		data[4] = cpu_to_le32((prim | R128_CCE_VC_CNTL_PRIM_WALK_IND |
				       (count << 16)));

		if (count & 0x1) {
#ifdef __LITTLE_ENDIAN
			data[dwords - 1] &= 0x0000ffff;
#else
			data[dwords - 1] &= 0xffff0000;
#endif
		}

		do {
			/* Emit the next set of up to three cliprects */
			if (i < sarea_priv->nbox) {
				r128_emit_clip_rects(dev_priv,
						     &sarea_priv->boxes[i],
						     sarea_priv->nbox - i);
			}

			r128_cce_dispatch_indirect(dev, buf, start, end);

			i += 3;
		} while (i < sarea_priv->nbox);
	}

	if (buf_priv->discard) {
		buf_priv->age = dev_priv->sarea_priv->last_dispatch;

		/* Emit the vertex buffer age */
		BEGIN_RING(2);

		OUT_RING(CCE_PACKET0(R128_LAST_DISPATCH_REG, 0));
		OUT_RING(buf_priv->age);

		ADVANCE_RING();

		buf->pending = 1;
		/* FIXME: Check dispatched field */
		buf_priv->dispatched = 0;
	}

	dev_priv->sarea_priv->last_dispatch++;

	sarea_priv->dirty &= ~R128_UPLOAD_CLIPRECTS;
	sarea_priv->nbox = 0;
}

static int r128_cce_dispatch_blit(DRMFILE filp,
				  drm_device_t * dev, drm_r128_blit_t * blit)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	drm_device_dma_t *dma = dev->dma;
	drm_buf_t *buf;
	drm_r128_buf_priv_t *buf_priv;
	u32 *data;
	int dword_shift, dwords;
	RING_LOCALS;
	DRM_DEBUG("\n");

	/* The compiler won't optimize away a division by a variable,
	 * even if the only legal values are powers of two.  Thus, we'll
	 * use a shift instead.
	 */
	switch (blit->format) {
	case R128_DATATYPE_ARGB8888:
		dword_shift = 0;
		break;
	case R128_DATATYPE_ARGB1555:
	case R128_DATATYPE_RGB565:
	case R128_DATATYPE_ARGB4444:
	case R128_DATATYPE_YVYU422:
	case R128_DATATYPE_VYUY422:
		dword_shift = 1;
		break;
	case R128_DATATYPE_CI8:
	case R128_DATATYPE_RGB8:
		dword_shift = 2;
		break;
	default:
		DRM_ERROR("invalid blit format %d\n", blit->format);
		return DRM_ERR(EINVAL);
	}

	/* Flush the pixel cache, and mark the contents as Read Invalid.
	 * This ensures no pixel data gets mixed up with the texture
	 * data from the host data blit, otherwise part of the texture
	 * image may be corrupted.
	 */
	BEGIN_RING(2);

	OUT_RING(CCE_PACKET0(R128_PC_GUI_CTLSTAT, 0));
	OUT_RING(R128_PC_RI_GUI | R128_PC_FLUSH_GUI);

	ADVANCE_RING();

	/* Dispatch the indirect buffer.
	 */
	buf = dma->buflist[blit->idx];
	buf_priv = buf->dev_private;

	if (buf->filp != filp) {
		DRM_ERROR("process %d using buffer owned by %p\n",
			  DRM_CURRENTPID, buf->filp);
		return DRM_ERR(EINVAL);
	}
	if (buf->pending) {
		DRM_ERROR("sending pending buffer %d\n", blit->idx);
		return DRM_ERR(EINVAL);
	}

	buf_priv->discard = 1;

	dwords = (blit->width * blit->height) >> dword_shift;

	data = (u32 *) ((char *)dev->agp_buffer_map->handle + buf->offset);

	data[0] = cpu_to_le32(CCE_PACKET3(R128_CNTL_HOSTDATA_BLT, dwords + 6));
	data[1] = cpu_to_le32((R128_GMC_DST_PITCH_OFFSET_CNTL |
			       R128_GMC_BRUSH_NONE |
			       (blit->format << 8) |
			       R128_GMC_SRC_DATATYPE_COLOR |
			       R128_ROP3_S |
			       R128_DP_SRC_SOURCE_HOST_DATA |
			       R128_GMC_CLR_CMP_CNTL_DIS |
			       R128_GMC_AUX_CLIP_DIS | R128_GMC_WR_MSK_DIS));

	data[2] = cpu_to_le32((blit->pitch << 21) | (blit->offset >> 5));
	data[3] = cpu_to_le32(0xffffffff);
	data[4] = cpu_to_le32(0xffffffff);
	data[5] = cpu_to_le32((blit->y << 16) | blit->x);
	data[6] = cpu_to_le32((blit->height << 16) | blit->width);
	data[7] = cpu_to_le32(dwords);

	buf->used = (dwords + 8) * sizeof(u32);

	r128_cce_dispatch_indirect(dev, buf, 0, buf->used);

	/* Flush the pixel cache after the blit completes.  This ensures
	 * the texture data is written out to memory before rendering
	 * continues.
	 */
	BEGIN_RING(2);

	OUT_RING(CCE_PACKET0(R128_PC_GUI_CTLSTAT, 0));
	OUT_RING(R128_PC_FLUSH_GUI);

	ADVANCE_RING();

	return 0;
}

/* ================================================================
 * Tiled depth buffer management
 *
 * FIXME: These should all set the destination write mask for when we
 * have hardware stencil support.
 */

static int r128_cce_dispatch_write_span(drm_device_t * dev,
					drm_r128_depth_t * depth)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	int count, x, y;
	u32 *buffer;
	u8 *mask;
	int i, buffer_size, mask_size;
	RING_LOCALS;
	DRM_DEBUG("\n");

	count = depth->n;
	if (count > 4096 || count <= 0)
		return DRM_ERR(EMSGSIZE);

	if (DRM_COPY_FROM_USER(&x, depth->x, sizeof(x))) {
		return DRM_ERR(EFAULT);
	}
	if (DRM_COPY_FROM_USER(&y, depth->y, sizeof(y))) {
		return DRM_ERR(EFAULT);
	}

	buffer_size = depth->n * sizeof(u32);
	buffer = drm_alloc(buffer_size, DRM_MEM_BUFS);
	if (buffer == NULL)
		return DRM_ERR(ENOMEM);
	if (DRM_COPY_FROM_USER(buffer, depth->buffer, buffer_size)) {
		drm_free(buffer, buffer_size, DRM_MEM_BUFS);
		return DRM_ERR(EFAULT);
	}

	mask_size = depth->n * sizeof(u8);
	if (depth->mask) {
		mask = drm_alloc(mask_size, DRM_MEM_BUFS);
		if (mask == NULL) {
			drm_free(buffer, buffer_size, DRM_MEM_BUFS);
			return DRM_ERR(ENOMEM);
		}
		if (DRM_COPY_FROM_USER(mask, depth->mask, mask_size)) {
			drm_free(buffer, buffer_size, DRM_MEM_BUFS);
			drm_free(mask, mask_size, DRM_MEM_BUFS);
			return DRM_ERR(EFAULT);
		}

		for (i = 0; i < count; i++, x++) {
			if (mask[i]) {
				BEGIN_RING(6);

				OUT_RING(CCE_PACKET3(R128_CNTL_PAINT_MULTI, 4));
				OUT_RING(R128_GMC_DST_PITCH_OFFSET_CNTL |
					 R128_GMC_BRUSH_SOLID_COLOR |
					 (dev_priv->depth_fmt << 8) |
					 R128_GMC_SRC_DATATYPE_COLOR |
					 R128_ROP3_P |
					 R128_GMC_CLR_CMP_CNTL_DIS |
					 R128_GMC_WR_MSK_DIS);

				OUT_RING(dev_priv->depth_pitch_offset_c);
				OUT_RING(buffer[i]);

				OUT_RING((x << 16) | y);
				OUT_RING((1 << 16) | 1);

				ADVANCE_RING();
			}
		}

		drm_free(mask, mask_size, DRM_MEM_BUFS);
	} else {
		for (i = 0; i < count; i++, x++) {
			BEGIN_RING(6);

			OUT_RING(CCE_PACKET3(R128_CNTL_PAINT_MULTI, 4));
			OUT_RING(R128_GMC_DST_PITCH_OFFSET_CNTL |
				 R128_GMC_BRUSH_SOLID_COLOR |
				 (dev_priv->depth_fmt << 8) |
				 R128_GMC_SRC_DATATYPE_COLOR |
				 R128_ROP3_P |
				 R128_GMC_CLR_CMP_CNTL_DIS |
				 R128_GMC_WR_MSK_DIS);

			OUT_RING(dev_priv->depth_pitch_offset_c);
			OUT_RING(buffer[i]);

			OUT_RING((x << 16) | y);
			OUT_RING((1 << 16) | 1);

			ADVANCE_RING();
		}
	}

	drm_free(buffer, buffer_size, DRM_MEM_BUFS);

	return 0;
}

static int r128_cce_dispatch_write_pixels(drm_device_t * dev,
					  drm_r128_depth_t * depth)
{
	drm_r128_private_t *dev_priv = dev->dev_private;
	int count, *x, *y;
	u32 *buffer;
	u8 *mask;
	int i, xbuf_size, ybuf_size, buffer_size, mask_size;
	RING_LOCALS;
	DRM_DEBUG("\n");

	count = depth->n;