nv30/nv40: upload shaders directly from rankine/curie init

Also removes A8 surface shader hack, and implements proper shaders for
the specific purpose.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

8 files changed, 583 insertions, 570 deletions

diff --git a/src/Makefile.am b/src/Makefile.am
index 879f79c..0bdd780 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -48,8 +48,6 @@ nouveau_drv_la_SOURCES = \
 			 nv04_xv_blit.c \
 			 nv10_exa.c \
 			 nv10_xv_ovl.c \
-			 nv30_shaders.c \
-			 nv30_shaders.h \
 			 nv30_exa.c \
 			 nv30_xv_tex.c \
 			 nv40_exa.c \
diff --git a/src/nv04_accel.h b/src/nv04_accel.h
index e1b4d8f..7100e85 100644
--- a/src/nv04_accel.h
+++ b/src/nv04_accel.h
@@ -1,11 +1,22 @@
 #ifndef __NV04_ACCEL_H__
 #define __NV04_ACCEL_H__
 
+#include "hwdefs/nv_object.xml.h"
+#include "hwdefs/nv01_2d.xml.h"
+
 #define XV_TABLE_SIZE 512
 
 /* scratch buffer offsets */
-#define FRAGPROG 0x00000000
-#define XV_TABLE 0x00001000
+#define PFP_PASS          0x00000000
+#define PFP_S             0x00000100
+#define PFP_C             0x00000200
+#define PFP_CCA           0x00000300
+#define PFP_CCASA         0x00000400
+#define PFP_S_A8          0x00000500
+#define PFP_C_A8          0x00000600
+#define PFP_NV12_BILINEAR 0x00000700
+#define PFP_NV12_BICUBIC  0x00000800
+#define XV_TABLE          0x00001000
 
 /* subchannel assignments */
 #define SUBC_M2MF(mthd)  0, (mthd)
@@ -37,4 +48,35 @@
 #define NV30_3D(mthd)    SUBC_3D(NV30_3D_##mthd)
 #define NV40_3D(mthd)    SUBC_3D(NV40_3D_##mthd)
 
+static __inline__ void
+PUSH_DATAu(struct nouveau_pushbuf *push, struct nouveau_bo *bo,
+	   unsigned delta, unsigned dwords)
+{
+	const uint32_t domain = NOUVEAU_BO_VRAM | NOUVEAU_BO_WR;
+	struct nouveau_pushbuf_refn refs[] = { { bo, domain } };
+	unsigned pitch = ((dwords * 4) + 63) & ~63;
+
+	if (nouveau_pushbuf_space(push, 32 + dwords, 2, 0) ||
+	    nouveau_pushbuf_refn (push, refs, 1))
+		return;
+
+	BEGIN_NV04(push, NV01_SUBC(MISC, OBJECT), 1);
+	PUSH_DATA (push, NvClipRectangle);
+	BEGIN_NV04(push, NV01_CLIP(POINT), 2);
+	PUSH_DATA (push, (0 << 16) | 0);
+	PUSH_DATA (push, (1 << 16) | dwords);
+	BEGIN_NV04(push, NV04_SF2D(FORMAT), 4);
+	PUSH_DATA (push, NV04_SURFACE_2D_FORMAT_A8R8G8B8);
+	PUSH_DATA (push, (pitch << 16) | pitch);
+	PUSH_RELOC(push, bo, delta, NOUVEAU_BO_LOW, 0, 0);
+	PUSH_RELOC(push, bo, delta, NOUVEAU_BO_LOW, 0, 0);
+	BEGIN_NV04(push, NV01_IFC(OPERATION), 5);
+	PUSH_DATA (push, NV01_IFC_OPERATION_SRCCOPY);
+	PUSH_DATA (push, NV01_IFC_COLOR_FORMAT_A8R8G8B8);
+	PUSH_DATA (push, (0 << 16) | 0);
+	PUSH_DATA (push, (1 << 16) | dwords);
+	PUSH_DATA (push, (1 << 16) | dwords);
+	BEGIN_NV04(push, NV01_IFC(COLOR(0)), dwords);
+}
+
 #endif
diff --git a/src/nv30_exa.c b/src/nv30_exa.c
index 7ac318c..0df2d0f 100644
--- a/src/nv30_exa.c
+++ b/src/nv30_exa.c
@@ -24,7 +24,6 @@
  */
 
 #include "nv_include.h"
-#include "nv30_shaders.h"
 
 #include "hwdefs/nv_object.xml.h"
 #include "hwdefs/nv30-40_3d.xml.h"
@@ -85,48 +84,6 @@ NV30_GetPictSurfaceFormat(int format)
 	return NULL;
 }
 
-enum {
-	NV30EXA_FPID_PASS_COL0 = 0,
-	NV30EXA_FPID_PASS_TEX0 = 1,
-	NV30EXA_FPID_COMPOSITE_MASK = 2,
-	NV30EXA_FPID_COMPOSITE_MASK_SA_CA = 3,
-	NV30EXA_FPID_COMPOSITE_MASK_CA = 4,
-	NV30EXA_FPID_MAX = 5
-} NV30EXA_FPID;
-
-static nv_shader_t *nv40_fp_map[NV30EXA_FPID_MAX] = {
-	&nv30_fp_pass_col0,
-	&nv30_fp_pass_tex0,
-	&nv30_fp_composite_mask,
-	&nv30_fp_composite_mask_sa_ca,
-	&nv30_fp_composite_mask_ca
-};
-
-static nv_shader_t *nv40_fp_map_a8[NV30EXA_FPID_MAX];
-
-static void
-NV30EXAHackupA8Shaders(ScrnInfoPtr pScrn)
-{
-	int s;
-
-	for (s = 0; s < NV30EXA_FPID_MAX; s++) {
-		nv_shader_t *def, *a8;
-
-		def = nv40_fp_map[s];
-		a8 = calloc(1, sizeof(nv_shader_t));
-		a8->card_priv.NV30FP.num_regs = def->card_priv.NV30FP.num_regs;
-		a8->size = def->size + 4;
-		memcpy(a8->data, def->data, def->size * sizeof(uint32_t));
-		nv40_fp_map_a8[s] = a8;
-
-		a8->data[a8->size - 8 + 0] &= ~0x00000081;
-		a8->data[a8->size - 4 + 0]  = 0x01401e81;
-		a8->data[a8->size - 4 + 1]  = 0x1c9dfe00;
-		a8->data[a8->size - 4 + 2]  = 0x0001c800;
-		a8->data[a8->size - 4 + 3]  = 0x0001c800;
-	}
-}
-
 /* should be in nouveau_reg.h at some point.. */
 #define NV30_3D_TEX_SWIZZLE_UNIT_S0_X_ZERO	 0
 #define NV30_3D_TEX_SWIZZLE_UNIT_S0_X_ONE	 1
@@ -462,7 +419,7 @@ NV30EXAPrepareComposite(int op, PicturePtr psPict,
 	NVPtr pNv = NVPTR(pScrn);
 	nv_pict_op_t *blend = NV30_GetPictOpRec(op);
 	struct nouveau_pushbuf *push = pNv->pushbuf;
-	int fpid = NV30EXA_FPID_PASS_COL0;
+	uint32_t fragprog;
 	NV30EXA_STATE;
 
 	if (!PUSH_SPACE(push, 128))
@@ -481,26 +438,37 @@ NV30EXAPrepareComposite(int op, PicturePtr psPict,
 		if (!NV30EXATexture(pScrn, pmPix, pmPict, 1))
 			return FALSE;
 
+		if (pdPict->format == PICT_a8) {
+			fragprog = PFP_C_A8;
+		} else
 		if (pmPict->componentAlpha && PICT_FORMAT_RGB(pmPict->format)) {
 			if (blend->src_alpha)
-				fpid = NV30EXA_FPID_COMPOSITE_MASK_SA_CA;
+				fragprog = PFP_CCASA;
 			else
-				fpid = NV30EXA_FPID_COMPOSITE_MASK_CA;
+				fragprog = PFP_CCA;
 		} else {
-			fpid = NV30EXA_FPID_COMPOSITE_MASK;
+			fragprog = PFP_C;
 		}
 
 		state->have_mask = TRUE;
 	} else {
-		fpid = NV30EXA_FPID_PASS_TEX0;
-
+		if (pdPict->format == PICT_a8)
+			fragprog = PFP_S_A8;
+		else
+			fragprog = PFP_S;
 		state->have_mask = FALSE;
 	}
 
-	if (!NV30_LoadFragProg(pScrn, (pdPict->format == PICT_a8) ?
-			       nv40_fp_map_a8[fpid] : nv40_fp_map[fpid]))
-		return FALSE;
-
+	BEGIN_NV04(push, NV30_3D(FP_ACTIVE_PROGRAM), 1);
+	PUSH_MTHD (push, NV30_3D(FP_ACTIVE_PROGRAM), pNv->scratch, fragprog,
+			 NOUVEAU_BO_VRAM | NOUVEAU_BO_RD | NOUVEAU_BO_LOW |
+			 NOUVEAU_BO_OR,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA0,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA1);
+	BEGIN_NV04(push, NV30_3D(FP_REG_CONTROL), 1);
+	PUSH_DATA (push, 0x0001000f);
+	BEGIN_NV04(push, NV30_3D(FP_CONTROL), 1);
+	PUSH_DATA (push, 0x00000000);
 	BEGIN_NV04(push, NV30_3D(TEX_UNITS_ENABLE), 1);
 	PUSH_DATA (push, pmPict ? 3 : 1);
 
@@ -629,10 +597,8 @@ NVAccelInitNV30TCL(ScrnInfoPtr pScrn)
 	struct nouveau_pushbuf *push = pNv->pushbuf;
 	struct nv04_fifo *fifo = pNv->channel->data;
 	uint32_t class = 0, chipset;
-	int next_hw_offset = FRAGPROG, i;
+	int i;
 
-	if (!nv40_fp_map_a8[0])
-		NV30EXAHackupA8Shaders(pScrn);
 	NVXVComputeBicubicFilter(pNv->scratch, XV_TABLE, XV_TABLE_SIZE);
 
 #define NV30TCL_CHIPSET_3X_MASK 0x00000003
@@ -702,11 +668,11 @@ NVAccelInitNV30TCL(ScrnInfoPtr pScrn)
 	PUSH_DATA (push, 0);
 	PUSH_DATA (push, 0x3f800000);
 	BEGIN_NV04(push, SUBC_3D(0x1f80), 16);
-	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); 
-	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); 
+	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0);
+	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0);
 	PUSH_DATA (push, 0x0000ffff);
-	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); 
-	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); 
+	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0);
+	PUSH_DATA (push, 0); PUSH_DATA (push, 0); PUSH_DATA (push, 0);
 
 	BEGIN_NV04(push, SUBC_3D(0x120), 3);
 	PUSH_DATA (push, 0);
@@ -721,6 +687,8 @@ NVAccelInitNV30TCL(ScrnInfoPtr pScrn)
 	BEGIN_NV04(push, SUBC_3D(0x1d88), 1);
 	PUSH_DATA (push, 0x00001200);
 
+	BEGIN_NV04(push, NV30_3D(MULTISAMPLE_CONTROL), 1);
+	PUSH_DATA (push, 0xffff0000);
 	BEGIN_NV04(push, NV30_3D(RC_ENABLE), 1);
 	PUSH_DATA (push, 0);
 
@@ -857,12 +825,219 @@ NVAccelInitNV30TCL(ScrnInfoPtr pScrn)
 	PUSH_DATA (push, 4096<<16);
 	PUSH_DATA (push, 4096<<16);
 
-	for (i = 0; i < NV30EXA_FPID_MAX; i++) {
-		NV30_UploadFragProg(pNv, nv40_fp_map[i], &next_hw_offset);
-		NV30_UploadFragProg(pNv, nv40_fp_map_a8[i], &next_hw_offset);
-	}
-	NV30_UploadFragProg(pNv, &nv30_fp_yv12_bicubic, &next_hw_offset);
-	NV30_UploadFragProg(pNv, &nv30_fp_yv12_bilinear, &next_hw_offset);
+	PUSH_DATAu(push, pNv->scratch, PFP_PASS, 1 * 4);
+	PUSH_DATA (push, 0x01403e81); /* mov r0, a[col0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_S, 2 * 4);
+	PUSH_DATA (push, 0x17009e00); /* tex r0, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x01401e81); /* mov r0, r0 */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_S_A8, 2 * 4);
+	PUSH_DATA (push, 0x17009000); /* tex r0.w, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x01401e81); /* mov r0, r0.w */
+	PUSH_DATA (push, 0x1c9dfe00);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_C, 3 * 4);
+	PUSH_DATA (push, 0x1702b102); /* texc0 r1.w, a[tex1], t[1] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x17009e00); /* tex r0 (ne0.w), a[tex0], t[0] */
+	PUSH_DATA (push, 0x1ff5c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r0, r1.w */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x0001fe04);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_C_A8, 3 * 4);
+	PUSH_DATA (push, 0x1702b102); /* texc0 r1.w, a[tex1], t[1] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x17009000); /* tex r0.w (ne0.w), a[tex0], t[0] */
+	PUSH_DATA (push, 0x1ff5c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r0.w, r1.w */
+	PUSH_DATA (push, 0x1c9dfe00);
+	PUSH_DATA (push, 0x0001fe04);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_CCA, 3 * 4);
+	PUSH_DATA (push, 0x17009f00); /* texc0 r0, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x1702be02); /* tex r1 (ne0), a[tex1], t[1] */
+	PUSH_DATA (push, 0x1c95c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r0, r1 */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x0001c804);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_CCASA, 3 * 4);
+	PUSH_DATA (push, 0x17009102); /* texc0 r1.w, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x1702be00); /* tex r0 (ne0.w), a[tex1], t[1] */
+	PUSH_DATA (push, 0x1ff5c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r1.w, r0 */
+	PUSH_DATA (push, 0x1c9dfe04);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_NV12_BILINEAR, 8 * 4);
+	PUSH_DATA (push, 0x17028200); /* texr r0.x, a[tex0], t[1] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x04000e02); /* madr r1.xyz, r0.x, imm.x, imm.yzww */
+	PUSH_DATA (push, 0x1c9c0000);
+	PUSH_DATA (push, 0x00000002);
+	PUSH_DATA (push, 0x0001f202);
+	PUSH_DATA (push, 0x3f9507c8); /* { 1.16, -0.87, 0.53, -1.08 } */
+	PUSH_DATA (push, 0xbf5ee393);
+	PUSH_DATA (push, 0x3f078fef);
+	PUSH_DATA (push, 0xbf8a6762);
+	PUSH_DATA (push, 0x1704ac80); /* texr r0.yz, a[tex1], t[2] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x04000e02); /* madr r1.xyz, r0.y, imm, r1 */
+	PUSH_DATA (push, 0x1c9cab00);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c804);
+	PUSH_DATA (push, 0x00000000); /* { 0.00, -0.39, 2.02, 0.00 } */
+	PUSH_DATA (push, 0xbec890d6);
+	PUSH_DATA (push, 0x40011687);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x04000e81); /* madr r0.xyz, r0.z, imm, r1 */
+	PUSH_DATA (push, 0x1c9d5500);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c804);
+	PUSH_DATA (push, 0x3fcc432d); /* { 1.60, -0.81, 0.00, 0.00 } */
+	PUSH_DATA (push, 0xbf501a37);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_NV12_BICUBIC, 24 * 4);
+	PUSH_DATA (push, 0x01008604); /* movr r2.xy, a[tex0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x03000600); /* addr r0.xy, r2, imm.x */
+	PUSH_DATA (push, 0x1c9dc808);
+	PUSH_DATA (push, 0x00000002);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3f000000); /* { 0.50, 0.00, 0.00, 0.00 } */
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x17000e06); /* texr r3.xyz, r0, t[0] */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x17000e00); /* texr r0.xyz, r0.y, t[0] */
+	PUSH_DATA (push, 0x1c9caa00);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x02000a02); /* mulr r1.xz, r3.xxyy, imm.xxyy */
+	PUSH_DATA (push, 0x1c9ca00c);
+	PUSH_DATA (push, 0x0000a002);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0xbf800000); /* { -1.00, 1.00, 0.00, 0.00 } */
+	PUSH_DATA (push, 0x3f800000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x02001402); /* mulr r1.yw, r0.xxyy, imm.xxyy */
+	PUSH_DATA (push, 0x1c9ca000);
+	PUSH_DATA (push, 0x0000a002);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0xbf800000); /* { -1.00, 1.00, 0.00, 0.00 } */
+	PUSH_DATA (push, 0x3f800000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x03001e04); /* addr r2, r2.xyxy, r1 */
+	PUSH_DATA (push, 0x1c9c8808);
+	PUSH_DATA (push, 0x0001c804);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x17020200); /* texr r0.x, r2, t[1] */
+	PUSH_DATA (push, 0x1c9dc808);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x17020402); /* texr r1.y, r2.xwxw, t[1] */
+	PUSH_DATA (push, 0x1c9d9808);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x17020202); /* texr r1.x, r2.zyxy, t[1] */
+	PUSH_DATA (push, 0x1c9c8c08);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x1f400280); /* lrph r0.x, r0.z, r0, r1.y */
+	PUSH_DATA (push, 0x1c9d5400);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0000aa04);
+	PUSH_DATA (push, 0x17020400); /* texr r0.y, r2.zwzz, t[1] */
+	PUSH_DATA (push, 0x1c9d5c08);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x1f400480); /* lrph r0.y, r0.z, r1.x, r0 */
+	PUSH_DATA (push, 0x1c9d5400);
+	PUSH_DATA (push, 0x00000004);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x1f400280); /* lrph r0.x, r3.z, r0, r0.y */
+	PUSH_DATA (push, 0x1c9d540c);
+	PUSH_DATA (push, 0x0001c900);
+	PUSH_DATA (push, 0x0000ab00);
+	PUSH_DATA (push, 0x04400e80); /* madh r0.xyz, r0.x, imm.x, imm.yzww */
+	PUSH_DATA (push, 0x1c9c0100);
+	PUSH_DATA (push, 0x00000002);
+	PUSH_DATA (push, 0x0001f202);
+	PUSH_DATA (push, 0x3f9507c8); /* { 1.16, -0.87, 0.53, -1.08 } */
+	PUSH_DATA (push, 0xbf5ee393);
+	PUSH_DATA (push, 0x3f078fef);
+	PUSH_DATA (push, 0xbf8a6762);
+	PUSH_DATA (push, 0x1704ac02); /* texr r1.yz, a[tex1], t[2] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x04400e80); /* madh r0.xyz, r1.y, imm, r0 */
+	PUSH_DATA (push, 0x1c9caa04);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c900);
+	PUSH_DATA (push, 0x00000000); /* { 0.00, -0.39, 2.02, 0.00 } */
+	PUSH_DATA (push, 0xbec890d6);
+	PUSH_DATA (push, 0x40011687);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x04400e81); /* madh r0.xyz, r1.z, imm, r0 */
+	PUSH_DATA (push, 0x1c9d5404);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c900);
+	PUSH_DATA (push, 0x3fcc432d); /* { 1.60, -0.81, 0.00, 0.00 } */
+	PUSH_DATA (push, 0xbf501a37);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
 
 	return TRUE;
 }
diff --git a/src/nv30_shaders.c b/src/nv30_shaders.c
deleted file mode 100644
index 6a72d8d..0000000
--- a/src/nv30_shaders.c
+++ /dev/null
@@ -1,344 +0,0 @@
-/*
- * Copyright 2007 Nouveau Project
- *
- * 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 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
- * THE AUTHORS 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.
- */
-
-
-#include "nv30_shaders.h"
-
-#include "hwdefs/nv30-40_3d.xml.h"
-#include "nv04_accel.h"
-
-void NV30_UploadFragProg(NVPtr pNv, nv_shader_t *shader, int *hw_offset)
-{
-	uint32_t data, i;
-	uint32_t *map;
-
-	shader->hw_id = *hw_offset;
-
-	map = pNv->scratch->map + *hw_offset;
-	for (i = 0; i < shader->size; i++) {
-		data = shader->data[i];
-#if (X_BYTE_ORDER != X_LITTLE_ENDIAN)
-		data = ((data >> 16) | ((data & 0xffff) << 16));
-#endif
-		map[i] = data;
-	}
-
-	*hw_offset += (shader->size * sizeof(uint32_t));
-	*hw_offset = (*hw_offset + 63) & ~63;
-}
-
-void NV40_UploadVtxProg(NVPtr pNv, nv_shader_t *shader, int *hw_id)
-{
-	struct nouveau_pushbuf *push = pNv->pushbuf;
-	int i;
-
-	shader->hw_id = *hw_id;
-
-	BEGIN_NV04(push, NV30_3D(VP_UPLOAD_FROM_ID), 1);
-	PUSH_DATA (push, (shader->hw_id));
-	for (i=0; i<shader->size; i+=4) {
-		BEGIN_NV04(push, NV30_3D(VP_UPLOAD_INST(0)), 4);
-		PUSH_DATA (push, shader->data[i + 0]);
-		PUSH_DATA (push, shader->data[i + 1]);
-		PUSH_DATA (push, shader->data[i + 2]);
-		PUSH_DATA (push, shader->data[i + 3]);
-		(*hw_id)++;
-	}
-}
-
-Bool
-NV30_LoadFragProg(ScrnInfoPtr pScrn, nv_shader_t *shader)
-{
-	NVPtr pNv = NVPTR(pScrn);
-	struct nouveau_pushbuf *push = pNv->pushbuf;
-
-	BEGIN_NV04(push, NV30_3D(FP_ACTIVE_PROGRAM), 1);
-	PUSH_MTHD (push, NV30_3D(FP_ACTIVE_PROGRAM), pNv->scratch,
-			 shader->hw_id, NOUVEAU_BO_VRAM | NOUVEAU_BO_RD |
-			 NOUVEAU_BO_LOW | NOUVEAU_BO_OR,
-			 NV30_3D_FP_ACTIVE_PROGRAM_DMA0,
-			 NV30_3D_FP_ACTIVE_PROGRAM_DMA1);
-	BEGIN_NV04(push, NV30_3D(FP_REG_CONTROL), 1);
-	PUSH_DATA (push, (1 << 16) | 0xf);
-	BEGIN_NV04(push, NV30_3D(MULTISAMPLE_CONTROL), 1);
-	PUSH_DATA (push, 0xffff0000);
-	BEGIN_NV04(push, NV30_3D(FP_CONTROL),1);
-	PUSH_DATA (push, (shader->card_priv.NV30FP.num_regs - 1) / 2);
-
-	return TRUE;
-}
-
-void
-NV40_LoadVtxProg(ScrnInfoPtr pScrn, nv_shader_t *shader)
-{
-	NVPtr pNv = NVPTR(pScrn);
-	struct nouveau_pushbuf *push = pNv->pushbuf;
-
-	BEGIN_NV04(push, NV30_3D(VP_START_FROM_ID), 1);
-	PUSH_DATA (push, (shader->hw_id));
-	BEGIN_NV04(push, NV40_3D(VP_ATTRIB_EN), 2);
-	PUSH_DATA (push, shader->card_priv.NV30VP.vp_in_reg);
-	PUSH_DATA (push, shader->card_priv.NV30VP.vp_out_reg);
-}
-
-Bool
-NV40_LoadFragProg(ScrnInfoPtr pScrn, nv_shader_t *shader)
-{
-	NVPtr pNv = NVPTR(pScrn);
-	struct nouveau_pushbuf *push = pNv->pushbuf;
-
-	BEGIN_NV04(push, NV30_3D(FP_ACTIVE_PROGRAM), 1);
-	PUSH_MTHD (push, NV30_3D(FP_ACTIVE_PROGRAM), pNv->scratch,
-			 shader->hw_id, NOUVEAU_BO_VRAM | NOUVEAU_BO_GART |
-			 NOUVEAU_BO_RD | NOUVEAU_BO_LOW | NOUVEAU_BO_OR,
-			 NV30_3D_FP_ACTIVE_PROGRAM_DMA0,
-			 NV30_3D_FP_ACTIVE_PROGRAM_DMA1);
-	BEGIN_NV04(push, NV30_3D(FP_CONTROL), 1);
-	PUSH_DATA (push, shader->card_priv.NV30FP.num_regs <<
-			 NV40_3D_FP_CONTROL_TEMP_COUNT__SHIFT);
-
-	return TRUE;
-}
-
-/*******************************************************************************
- * NV40/G70 vertex shaders
- */
-
-nv_shader_t nv40_vp_exa_render = {
-	.card_priv.NV30VP.vp_in_reg  = 0x00000309,
-	.card_priv.NV30VP.vp_out_reg = 0x0000c001,
-	.size = (3*4),
-	.data = {
-		/* MOV result.position, vertex.position */
-		0x40041c6c, 0x0040000d, 0x8106c083, 0x6041ff80,
-		/* MOV result.texcoord[0], vertex.texcoord[0] */
-		0x401f9c6c, 0x0040080d, 0x8106c083, 0x6041ff9c,
-		/* MOV result.texcoord[1], vertex.texcoord[1] */
-		0x401f9c6c, 0x0040090d, 0x8106c083, 0x6041ffa1,
-	}
-};
-
-/*******************************************************************************
- * NV30/NV40/G70 fragment shaders
- */
-
-nv_shader_t nv30_fp_pass_col0 = {
-	.card_priv.NV30FP.num_regs = 2,
-	.size = (1*4),
-	.data = {
-		/* MOV R0, fragment.color */
-		0x01403e81, 0x1c9dc801, 0x0001c800, 0x3fe1c800, 
-	}
-};
-
-nv_shader_t nv30_fp_pass_tex0 = {
-	.card_priv.NV30FP.num_regs = 2,
-	.size = (2*4),
-	.data = {
-		/* TEX R0, fragment.texcoord[0], texture[0], 2D */
-		0x17009e00, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* MOV R0, R0 */
-		0x01401e81, 0x1c9dc800, 0x0001c800, 0x0001c800,
-	}
-};
-
-nv_shader_t nv30_fp_composite_mask = {
-	.card_priv.NV30FP.num_regs = 2,
-	.size = (3*4),
-	.data = {
-		/* TEXC0 R1.w         , fragment.texcoord[1], texture[1], 2D */
-		0x1702b102, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* TEX   R0 (NE0.wwww), fragment.texcoord[0], texture[0], 2D */
-		0x17009e00, 0x1ff5c801, 0x0001c800, 0x3fe1c800,
-		/* MUL   R0           , R0, R1.w */
-		0x02001e81, 0x1c9dc800, 0x0001fe04, 0x0001c800,
-	}
-};
-
-nv_shader_t nv30_fp_composite_mask_sa_ca = {
-	.card_priv.NV30FP.num_regs = 2,
-	.size = (3*4),
-	.data = {
-		/* TEXC0 R1.w         , fragment.texcoord[0], texture[0], 2D */
-		0x17009102, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* TEX   R0 (NE0.wwww), fragment.texcoord[1], texture[1], 2D */
-		0x1702be00, 0x1ff5c801, 0x0001c800, 0x3fe1c800,
-		/* MUL   R0           , R1,wwww, R0 */
-		0x02001e81, 0x1c9dfe04, 0x0001c800, 0x0001c800,
-	}
-};
-
-nv_shader_t nv30_fp_composite_mask_ca = {
-	.card_priv.NV30FP.num_regs = 2,
-	.size = (3*4),
-	.data = {
-		/* TEXC0 R0           , fragment.texcoord[0], texture[0], 2D */
-		0x17009f00, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* TEX   R1 (NE0.xyzw), fragment.texcoord[1], texture[1], 2D */
-		0x1702be02, 0x1c95c801, 0x0001c800, 0x3fe1c800,
-		/* MUL   R0           , R0, R1 */
-		0x02001e81, 0x1c9dc800, 0x0001c804, 0x0001c800,
-	}
-};
-
-nv_shader_t nv40_vp_video = {
-	.card_priv.NV30VP.vp_in_reg  = 0x00000309,
-	.card_priv.NV30VP.vp_out_reg = 0x0000c001,
-	.size = (3*4),
-	.data = {
-		/* MOV result.position, vertex.position */
-		0x40041c6c, 0x0040000d, 0x8106c083, 0x6041ff80,
-		/* MOV result.texcoord[0], vertex.texcoord[0] */
-		0x401f9c6c, 0x0040080d, 0x8106c083, 0x6041ff9c,
-		/* MOV result.texcoord[1], vertex.texcoord[1] */
-		0x401f9c6c, 0x0040090d, 0x8106c083, 0x6041ffa1,
-	}
-};
-
-nv_shader_t nv40_fp_yv12_bicubic = {
-	.card_priv.NV30FP.num_regs = 4,
-	.size = (29*4),
-	.data = {
-		/* INST 0: MOVR R0.xy (TR0.xyzw), attrib.texcoord[0] */
-		0x01008600, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* INST 1: ADDR R0.z (TR0.xyzw), R0.yyyy, { 0.50, 0.00, 0.00, 0.00 }.xxxx */
-		0x03000800, 0x1c9caa00, 0x00000002, 0x0001c800,
-		0x3f000000, 0x00000000, 0x00000000, 0x00000000,
-		/* INST 2: ADDR R1.x (TR0.xyzw), R0, { 0.50, 0.00, 0.00, 0.00 }.xxxx */
-		0x03000202, 0x1c9dc800, 0x00000002, 0x0001c800,
-		0x3f000000, 0x00000000, 0x00000000, 0x00000000,
-		/* INST 3: TEXRC0 R1.xyz (TR0.xyzw), R0.zzzz, texture[0] */
-		0x17000f82, 0x1c9d5400, 0x0001c800, 0x0001c800,
-		/* INST 4: MULR R2.yw (TR0.xyzw), R1.xxyy, { -1.00, 1.00, 0.00, 0.00 }.xxyy */
-		0x02001404, 0x1c9ca104, 0x0000a002, 0x0001c800,
-		0xbf800000, 0x3f800000, 0x00000000, 0x00000000,
-		/* INST 5: TEXR R3.xyz (TR0.xyzw), R1, texture[0] */
-		0x17000e86, 0x1c9dc804, 0x0001c800, 0x0001c800,
-		/* INST 6: MULR R2.xz (TR0.xyzw), R3.xxyy, { -1.00, 1.00, 0.00, 0.00 }.xxyy */
-		0x02000a04, 0x1c9ca10c, 0x0000a002, 0x0001c800,
-		0xbf800000, 0x3f800000, 0x00000000, 0x00000000,
-		/* INST 7: ADDR R2 (TR0.xyzw), R0.xyxy, R2 */
-		0x03001e04, 0x1c9c8800, 0x0001c808, 0x0001c800,
-		/* INST 8: TEXR R1.y (TR0.xyzw), R2.zwzz, -texture[1] */
-		0x17020402, 0x1c9d5c08, 0x0001c800, 0x0001c800,
-		/* INST 9: MADH R1.x (TR0.xyzw), -R1.zzzz, R1.yyyy, R1.yyyy */
-		0x04400282, 0x1c9f5504, 0x0000aa04, 0x0000aa04,
-		/* INST 10: TEXR R0.y (TR0.xyzw), R2.xwxw, -texture[1] */
-		0x17020400, 0x1c9d9808, 0x0001c800, 0x0001c800,
-		/* INST 11: MADH R0.w (TR0.xyzw), -R1.zzzz, R0.yyyy, R0.yyyy */
-		0x04401080, 0x1c9f5504, 0x0000aa00, 0x0000aa00,
-		/* INST 12: TEXR R0.x (TR0.xyzw), R2.zyxy, texture[1] */
-		0x17020200, 0x1c9c8c08, 0x0001c800, 0x0001c800,
-		/* INST 13: MADH R1.x (TR0.xyzw), R1.zzzz, R0, R1 */
-		0x04400282, 0x1c9d5504, 0x0001c800, 0x0001c904,
-		/* INST 14: TEXR R0.x (NE0.zzzz), R2, texture[1] */
-		0x17020200, 0x1555c808, 0x0001c800, 0x0001c800,
-		/* INST 15: MADH R0.x (TR0.xyzw), R1.zzzz, R0, R0.wwww */
-		0x04400280, 0x1c9d5504, 0x0001c800, 0x0001ff00,
-		/* INST 16: MADH R0.w (TR0.xyzw), -R3.zzzz, R1.xxxx, R1.xxxx */
-		0x04401080, 0x1c9f550c, 0x00000104, 0x00000104,
-		/* INST 17: TEXR R0.yz (TR0.xyzw), attrib.texcoord[1], abs(texture[2]) */
-		0x1704ac80, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* INST 18: MADH R0.x (TR0.xyzw), R3.zzzz, R0, R0.wwww */
-		0x04400280, 0x1c9d550c, 0x0001c900, 0x0001ff00,
-		/* INST 19: MADH R1.xyz (TR0.xyzw), R0.xxxx, { 1.16, -0.87, 0.53, -1.08 }.xxxx, { 1.16, -0.87, 0.53, -1.08 }.yzww */
-		0x04400e82, 0x1c9c0100, 0x00000002, 0x0001f202,
-		0x3f9507c8, 0xbf5ee393, 0x3f078fef, 0xbf8a6762,
-		/* INST 20: MADH R1.xyz (TR0.xyzw), R0.yyyy, { 0.00, -0.39, 2.02, 0.00 }, R1 */
-		0x04400e82, 0x1c9cab00, 0x0001c802, 0x0001c904,
-		0x00000000, 0xbec890d6, 0x40011687, 0x00000000,
-		/* INST 21: MADH R0.xyz (TR0.xyzw), R0.zzzz, { 1.60, -0.81, 0.00, 0.00 }, R1 + END */
-		0x04400e81, 0x1c9d5500, 0x0001c802, 0x0001c904,
-		0x3fcc432d, 0xbf501a37, 0x00000000, 0x00000000,
-	}
-};
-
-nv_shader_t nv30_fp_yv12_bicubic = {
-	.card_priv.NV30FP.num_regs = 4,
-	.size = (24*4),
-	.data = {
-		/* INST 0: MOVR R2.xy (TR0.xyzw), attrib.texcoord[0] */
-		0x01008604, 0x1c9dc801, 0x0001c800, 0x0001c800,
-		/* INST 1: ADDR R0.xy (TR0.xyzw), R2, { 0.50, 0.00, 0.00, 0.00 }.xxxx */
-		0x03000600, 0x1c9dc808, 0x00000002, 0x0001c800,
-		0x3f000000, 0x00000000, 0x00000000, 0x00000000,
-		/* INST 2: TEXR R3.xyz (TR0.xyzw), R0, texture[0] */
-		0x17000e06, 0x1c9dc800, 0x0001c800, 0x0001c800,
-		/* INST 3: TEXR R0.xyz (TR0.xyzw), R0.yyyy, texture[0] */
-		0x17000e00, 0x1c9caa00, 0x0001c800, 0x0001c800,
-		/* INST 4: MULR R1.xz (TR0.xyzw), R3.xxyy, { -1.00, 1.00, 0.00, 0.00 }.xxyy */
-		0x02000a02, 0x1c9ca00c, 0x0000a002, 0x0001c800,
-		0xbf800000, 0x3f800000, 0x00000000, 0x00000000,
-		/* INST 5: MULR R1.yw (TR0.xyzw), R0.xxyy, { -1.00, 1.00, 0.00, 0.00 }.xxyy */
-		0x02001402, 0x1c9ca000, 0x0000a002, 0x0001c800,
-		0xbf800000, 0x3f800000, 0x00000000, 0x00000000,
-		/* INST 6: ADDR R2 (TR0.xyzw), R2.xyxy, R1 */
-		0x03001e04, 0x1c9c8808, 0x0001c804, 0x0001c800,
-		/* INST 7: TEXR R0.x (TR0.xyzw), R2, texture[1] */
-		0x17020200, 0x1c9dc808, 0x0001c800, 0x0001c800,
-		/* INST 8: TEXR R1.y (TR0.xyzw), R2.xwxw, texture[1] */
-		0x17020402, 0x1c9d9808, 0x0001c800, 0x0001c800,
-		/* INST 9: TEXR R1.x (TR0.xyzw), R2.zyxy, texture[1] */
-		0x17020202, 0x1c9c8c08, 0x0001c800, 0x0001c800,
-		/* INST 10: LRPH R0.x (TR0.xyzw), R0.zzzz, R0, R1.yyyy */
-		0x1f400280, 0x1c9d5400, 0x0001c800, 0x0000aa04,
-		/* INST 11: TEXR R0.y (TR0.xyzw), R2.zwzz, texture[1] */
-		0x17020400, 0x1c9d5c08, 0x0001c800, 0x0001c800,
-		/* INST 12: LRPH R0.y (TR0.xyzw), R0.zzzz, R1.xxxx, R0 */
-		0x1f400480, 0x1c9d5400, 0x00000004, 0x0001c800,
-		/* INST 13: LRPH R0.x (TR0.xyzw), R3.zzzz, R0, R0.yyyy */
-		0x1f400280, 0x1c9d540c, 0x0001c900, 0x0000ab00,
-		/* INST 14: MADH R0.xyz (TR0.xyzw), R0.xxxx, { 1.16, -0.87, 0.53, -1.08 }.xxxx, { 1.16, -0.87, 0.53, -1.08 }.yzww */
-		0x04400e80, 0x1c9c0100, 0x00000002, 0x0001f202,
-		0x3f9507c8, 0xbf5ee393, 0x3f078fef, 0xbf8a6762,
-		/* INST 15: TEXR R1.yz (TR0.xyzw), attrib.texcoord[1], abs(texture[2]) */
-		0x1704ac02, 0x1c9dc801, 0x0001c800, 0x0001c800,
-		/* INST 16: MADH R0.xyz (TR0.xyzw), R1.yyyy, { 0.00, -0.39, 2.02, 0.00 }, R0 */
-		0x04400e80, 0x1c9caa04, 0x0001c802, 0x0001c900,
-		0x00000000, 0xbec890d6, 0x40011687, 0x00000000,
-		/* INST 17: MADH R0.xyz (TR0.xyzw), R1.zzzz, { 1.60, -0.81, 0.00, 0.00 }, R0 + END */
-		0x04400e81, 0x1c9d5404, 0x0001c802, 0x0001c900,
-		0x3fcc432d, 0xbf501a37, 0x00000000, 0x00000000,
-	}
-};
-
-nv_shader_t nv30_fp_yv12_bilinear = {
-	.card_priv.NV30FP.num_regs = 2,
-	.size = (8*4),
-	.data = {
-		/* INST 0: TEXR R0.x (TR0.xyzw), attrib.texcoord[0], abs(texture[1]) */
-		0x17028200, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* INST 1: MADR R1.xyz (TR0.xyzw), R0.xxxx, { 1.16, -0.87, 0.53, -1.08 }.xxxx, { 1.16, -0.87, 0.53, -1.08 }.yzww */
-		0x04000e02, 0x1c9c0000, 0x00000002, 0x0001f202,
-		0x3f9507c8, 0xbf5ee393, 0x3f078fef, 0xbf8a6762,
-		/* INST 2: TEXR R0.yz (TR0.xyzw), attrib.texcoord[1], abs(texture[2]) */
-		0x1704ac80, 0x1c9dc801, 0x0001c800, 0x3fe1c800,
-		/* INST 3: MADR R1.xyz (TR0.xyzw), R0.yyyy, { 0.00, -0.39, 2.02, 0.00 }, R1 */
-		0x04000e02, 0x1c9cab00, 0x0001c802, 0x0001c804,
-		0x00000000, 0xbec890d6, 0x40011687, 0x00000000,
-		/* INST 4: MADR R0.xyz (TR0.xyzw), R0.zzzz, { 1.60, -0.81, 0.00, 0.00 }, R1 + END */
-		0x04000e81, 0x1c9d5500, 0x0001c802, 0x0001c804,
-		0x3fcc432d, 0xbf501a37, 0x00000000, 0x00000000,
-	}
-};
-
-
diff --git a/src/nv30_shaders.h b/src/nv30_shaders.h
deleted file mode 100644
index 3e666fa..0000000
--- a/src/nv30_shaders.h
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * Copyright 2007 Nouveau Project
- *
- * 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 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
- * THE AUTHORS 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 __NV30_SHADERS_H__
-#define __NV30_SHADERS_H__
-
-#define NV_SHADER_MAX_PROGRAM_LENGTH 256
-
-#include "nv_include.h"
-
-typedef struct nv_shader {
-	uint32_t hw_id;
-	uint32_t size;
-	union {
-		struct {
-			uint32_t vp_in_reg;
-			uint32_t vp_out_reg;
-		} NV30VP;
-		struct  {
-			uint32_t num_regs;
-		} NV30FP;
-	} card_priv;
-	uint32_t data[NV_SHADER_MAX_PROGRAM_LENGTH];
-} nv_shader_t;
-
-void NV30_UploadFragProg(NVPtr pNv, nv_shader_t *shader, int *hw_offset);
-void NV40_UploadVtxProg(NVPtr pNv, nv_shader_t *shader, int *hw_id);
-void NV40_LoadVtxProg(ScrnInfoPtr pScrn, nv_shader_t *shader);
-Bool NV40_LoadFragProg(ScrnInfoPtr pScrn, nv_shader_t *shader);
-Bool NV30_LoadFragProg(ScrnInfoPtr pScrn, nv_shader_t *shader);
-
-
-/*******************************************************************************
- * NV40/G70 vertex shaders
- */
-
-nv_shader_t nv40_vp_exa_render;
-nv_shader_t nv40_vp_video;
-
-/*******************************************************************************
- * NV30/NV40/G70 fragment shaders
- */
-
-nv_shader_t nv30_fp_pass_col0;
-nv_shader_t nv30_fp_pass_tex0;
-nv_shader_t nv30_fp_composite_mask;
-nv_shader_t nv30_fp_composite_mask_sa_ca;
-nv_shader_t nv30_fp_composite_mask_ca;
-nv_shader_t nv30_fp_yv12_bicubic;
-nv_shader_t nv30_fp_yv12_bilinear;
-nv_shader_t nv40_fp_yv12_bicubic;
-
-#endif
diff --git a/src/nv30_xv_tex.c b/src/nv30_xv_tex.c
index 88fed3e..7ab0520 100644
--- a/src/nv30_xv_tex.c
+++ b/src/nv30_xv_tex.c
@@ -36,8 +36,6 @@
 #include "nv_include.h"
 #include "nv_dma.h"
 
-#include "nv30_shaders.h"
-
 #include "hwdefs/nv30-40_3d.xml.h"
 #include "nv04_accel.h"
 
@@ -259,10 +257,17 @@ NV30PutTextureImage(ScrnInfoPtr pScrn, struct nouveau_bo *src, int src_offset,
 	if (drw_w / 2 < src_w || drw_h / 2 < src_h)
 		bicubic = FALSE;
 
-	if (!NV30_LoadFragProg(pScrn, bicubic ?
-			       &nv30_fp_yv12_bicubic :
-			       &nv30_fp_yv12_bilinear))
-		return BadImplementation;
+	BEGIN_NV04(push, NV30_3D(FP_ACTIVE_PROGRAM), 1);
+	PUSH_MTHD (push, NV30_3D(FP_ACTIVE_PROGRAM), pNv->scratch,
+			 bicubic ? PFP_NV12_BICUBIC : PFP_NV12_BILINEAR,
+			 NOUVEAU_BO_VRAM | NOUVEAU_BO_RD | NOUVEAU_BO_LOW |
+			 NOUVEAU_BO_OR,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA0,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA1);
+	BEGIN_NV04(push, NV30_3D(FP_REG_CONTROL), 1);
+	PUSH_DATA (push, 0x0001000f);
+	BEGIN_NV04(push, NV30_3D(FP_CONTROL), 1);
+	PUSH_DATA (push, 0x00000001);
 
 	nouveau_pushbuf_bufctx(push, pNv->bufctx);
 	if (nouveau_pushbuf_validate(push)) {
diff --git a/src/nv40_exa.c b/src/nv40_exa.c
index 42fc428..c4249a8 100644
--- a/src/nv40_exa.c
+++ b/src/nv40_exa.c
@@ -21,7 +21,6 @@
  */
 
 #include "nv_include.h"
-#include "nv30_shaders.h"
 
 #include "hwdefs/nv_object.xml.h"
 #include "hwdefs/nv30-40_3d.xml.h"
@@ -80,48 +79,6 @@ NV40_GetPictSurfaceFormat(int format)
 	return NULL;
 }
 
-enum {
-	NV40EXA_FPID_PASS_COL0 = 0,
-	NV40EXA_FPID_PASS_TEX0 = 1,
-	NV40EXA_FPID_COMPOSITE_MASK = 2,
-	NV40EXA_FPID_COMPOSITE_MASK_SA_CA = 3,
-	NV40EXA_FPID_COMPOSITE_MASK_CA = 4,
-	NV40EXA_FPID_MAX = 5
-} NV40EXA_FPID;
-
-static nv_shader_t *nv40_fp_map[NV40EXA_FPID_MAX] = {
-	&nv30_fp_pass_col0,
-	&nv30_fp_pass_tex0,
-	&nv30_fp_composite_mask,
-	&nv30_fp_composite_mask_sa_ca,
-	&nv30_fp_composite_mask_ca
-};
-
-static nv_shader_t *nv40_fp_map_a8[NV40EXA_FPID_MAX];
-
-static void
-NV40EXAHackupA8Shaders(ScrnInfoPtr pScrn)
-{
-	int s;
-
-	for (s = 0; s < NV40EXA_FPID_MAX; s++) {
-		nv_shader_t *def, *a8;
-
-		def = nv40_fp_map[s];
-		a8 = calloc(1, sizeof(nv_shader_t));
-		a8->card_priv.NV30FP.num_regs = def->card_priv.NV30FP.num_regs;
-		a8->size = def->size + 4;
-		memcpy(a8->data, def->data, def->size * sizeof(uint32_t));
-		nv40_fp_map_a8[s] = a8;
-
-		a8->data[a8->size - 8 + 0] &= ~0x00000081;
-		a8->data[a8->size - 4 + 0]  = 0x01401e81;
-		a8->data[a8->size - 4 + 1]  = 0x1c9dfe00;
-		a8->data[a8->size - 4 + 2]  = 0x0001c800;
-		a8->data[a8->size - 4 + 3]  = 0x0001c800;
-	}
-}
-
 #define _(r,tf,ts0x,ts0y,ts0z,ts0w,ts1x,ts1y,ts1z,ts1w)                        \
   {                                                                            \
   PICT_##r, NV40_3D_TEX_FORMAT_FORMAT_##tf,                                    \
@@ -406,7 +363,7 @@ NV40EXAPrepareComposite(int op, PicturePtr psPict,
 	NVPtr pNv = NVPTR(pScrn);
 	nv_pict_op_t *blend = NV40_GetPictOpRec(op);
 	struct nouveau_pushbuf *push = pNv->pushbuf;
-	int fpid = NV40EXA_FPID_PASS_COL0;
+	uint32_t fragprog;
 	NV40EXA_STATE;
 
 	if (!PUSH_SPACE(push, 128))
@@ -421,31 +378,39 @@ NV40EXAPrepareComposite(int op, PicturePtr psPict,
 	    !NV40EXATexture(pScrn, psPix, psPict, 0))
 		return FALSE;
 
-	NV40_LoadVtxProg(pScrn, &nv40_vp_exa_render);
 	if (pmPict) {
 		if (!NV40EXATexture(pScrn, pmPix, pmPict, 1))
 			return FALSE;
 
+		if (pdPict->format == PICT_a8) {
+			fragprog = PFP_C_A8;
+		} else
 		if (pmPict->componentAlpha && PICT_FORMAT_RGB(pmPict->format)) {
 			if (blend->src_alpha)
-				fpid = NV40EXA_FPID_COMPOSITE_MASK_SA_CA;
+				fragprog = PFP_CCASA;
 			else
-				fpid = NV40EXA_FPID_COMPOSITE_MASK_CA;
+				fragprog = PFP_CCA;
 		} else {
-			fpid = NV40EXA_FPID_COMPOSITE_MASK;
+			fragprog = PFP_C;
 		}
 
 		state->have_mask = TRUE;
 	} else {
-		fpid = NV40EXA_FPID_PASS_TEX0;
-
+		if (pdPict->format == PICT_a8)
+			fragprog = PFP_S_A8;
+		else
+			fragprog = PFP_S;
 		state->have_mask = FALSE;
 	}
 
-
-	if (!NV40_LoadFragProg(pScrn, (pdPict->format == PICT_a8) ?
-			       nv40_fp_map_a8[fpid] : nv40_fp_map[fpid]))
-		return FALSE;
+	BEGIN_NV04(push, NV30_3D(FP_ACTIVE_PROGRAM), 1);
+	PUSH_MTHD (push, NV30_3D(FP_ACTIVE_PROGRAM), pNv->scratch, fragprog,
+			 NOUVEAU_BO_VRAM | NOUVEAU_BO_RD | NOUVEAU_BO_LOW |
+			 NOUVEAU_BO_OR,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA0,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA1);
+	BEGIN_NV04(push, NV30_3D(FP_CONTROL), 1);
+	PUSH_DATA (push, 0x02000000);
 
 	/* Appears to be some kind of cache flush, needed here at least
 	 * sometimes.. funky text rendering otherwise :)
@@ -580,10 +545,8 @@ NVAccelInitNV40TCL(ScrnInfoPtr pScrn)
 	struct nouveau_pushbuf *push = pNv->pushbuf;
 	struct nv04_fifo *fifo = pNv->channel->data;
 	uint32_t class = 0, chipset;
-	int next_hw_id = 0, next_hw_offset = FRAGPROG, i;
+	int i;
 
-	if (!nv40_fp_map_a8[0])
-		NV40EXAHackupA8Shaders(pScrn);
 	NVXVComputeBicubicFilter(pNv->scratch, XV_TABLE, XV_TABLE_SIZE);
 
 	chipset = pNv->dev->chipset;
@@ -667,7 +630,7 @@ NVAccelInitNV40TCL(ScrnInfoPtr pScrn)
 	PUSH_DATA (push, 0);
 	BEGIN_NV04(push, NV30_3D(DEPTH_WRITE_ENABLE), 2);
 	PUSH_DATA (push, 0);
-	PUSH_DATA (push, 0); 
+	PUSH_DATA (push, 0);
 	BEGIN_NV04(push, NV30_3D(COLOR_MASK), 1);
 	PUSH_DATA (push, 0x01010101); /* TR,TR,TR,TR */
 	BEGIN_NV04(push, NV30_3D(CULL_FACE_ENABLE), 1);
@@ -714,15 +677,262 @@ NVAccelInitNV40TCL(ScrnInfoPtr pScrn)
 	PUSH_DATA (push, (4095 << 16));
 	PUSH_DATA (push, (4095 << 16));
 
-	NV40_UploadVtxProg(pNv, &nv40_vp_exa_render, &next_hw_id);
-	for (i = 0; i < NV40EXA_FPID_MAX; i++) {
-		NV30_UploadFragProg(pNv, nv40_fp_map[i], &next_hw_offset);
-		NV30_UploadFragProg(pNv, nv40_fp_map_a8[i], &next_hw_offset);
-	}
-
-	NV40_UploadVtxProg(pNv, &nv40_vp_video, &next_hw_id);
-	NV30_UploadFragProg(pNv, &nv40_fp_yv12_bicubic, &next_hw_offset);
-	NV30_UploadFragProg(pNv, &nv30_fp_yv12_bilinear, &next_hw_offset);
-
+	BEGIN_NV04(push, NV30_3D(VP_UPLOAD_FROM_ID), 1);
+	PUSH_DATA (push, 0);
+	BEGIN_NV04(push, NV30_3D(VP_UPLOAD_INST(0)), 4);
+	PUSH_DATA (push, 0x40041c6c); /* mov o[pos], a[0] */
+	PUSH_DATA (push, 0x0040000d);
+	PUSH_DATA (push, 0x8106c083);
+	PUSH_DATA (push, 0x6041ff80);
+	BEGIN_NV04(push, NV30_3D(VP_UPLOAD_INST(0)), 4);
+	PUSH_DATA (push, 0x401f9c6c); /* mov o[tex0], a[8] */
+	PUSH_DATA (push, 0x0040080d);
+	PUSH_DATA (push, 0x8106c083);
+	PUSH_DATA (push, 0x6041ff9c);
+	BEGIN_NV04(push, NV30_3D(VP_UPLOAD_INST(0)), 4);
+	PUSH_DATA (push, 0x401f9c6c); /* mov o[tex1], a[9] */
+	PUSH_DATA (push, 0x0040090d);
+	PUSH_DATA (push, 0x8106c083);
+	PUSH_DATA (push, 0x6041ffa1);
+	BEGIN_NV04(push, NV30_3D(VP_START_FROM_ID), 1);
+	PUSH_DATA (push, 0);
+	BEGIN_NV04(push, NV40_3D(VP_ATTRIB_EN), 2);
+	PUSH_DATA (push, 0x00000309);
+	PUSH_DATA (push, 0x0000c001);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_PASS, 1 * 4);
+	PUSH_DATA (push, 0x01403e81); /* mov r0, a[col0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_S, 2 * 4);
+	PUSH_DATA (push, 0x17009e00); /* tex r0, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x01401e81); /* mov r0, r0 */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_S_A8, 2 * 4);
+	PUSH_DATA (push, 0x17009000); /* tex r0.w, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x01401e81); /* mov r0, r0.w */
+	PUSH_DATA (push, 0x1c9dfe00);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_C, 3 * 4);
+	PUSH_DATA (push, 0x1702b102); /* texc0 r1.w, a[tex1], t[1] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x17009e00); /* tex r0 (ne0.w), a[tex0], t[0] */
+	PUSH_DATA (push, 0x1ff5c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r0, r1.w */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x0001fe04);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_C_A8, 3 * 4);
+	PUSH_DATA (push, 0x1702b102); /* texc0 r1.w, a[tex1], t[1] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x17009000); /* tex r0.w (ne0.w), a[tex0], t[0] */
+	PUSH_DATA (push, 0x1ff5c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r0.w, r1.w */
+	PUSH_DATA (push, 0x1c9dfe00);
+	PUSH_DATA (push, 0x0001fe04);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_CCA, 3 * 4);
+	PUSH_DATA (push, 0x17009f00); /* texc0 r0, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x1702be02); /* tex r1 (ne0), a[tex1], t[1] */
+	PUSH_DATA (push, 0x1c95c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r0, r1 */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x0001c804);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_CCASA, 3 * 4);
+	PUSH_DATA (push, 0x17009102); /* texc0 r1.w, a[tex0], t[0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x1702be00); /* tex r0 (ne0.w), a[tex1], t[1] */
+	PUSH_DATA (push, 0x1ff5c801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x02001e81); /* mul r0, r1.w, r0 */
+	PUSH_DATA (push, 0x1c9dfe04);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+
+	PUSH_DATAu(push, pNv->scratch, PFP_NV12_BILINEAR, 8 * 4);
+	PUSH_DATA (push, 0x17028200); /* texr r0.x, a[tex0], t[1] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x04000e02); /* madr r1.xyz, r0.x, imm.x, imm.yzww */
+	PUSH_DATA (push, 0x1c9c0000);
+	PUSH_DATA (push, 0x00000002);
+	PUSH_DATA (push, 0x0001f202);
+	PUSH_DATA (push, 0x3f9507c8); /* { 1.16, -0.87, 0.53, -1.08 } */
+	PUSH_DATA (push, 0xbf5ee393);
+	PUSH_DATA (push, 0x3f078fef);
+	PUSH_DATA (push, 0xbf8a6762);
+	PUSH_DATA (push, 0x1704ac80); /* texr r0.yz, a[tex1], t[2] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x04000e02); /* madr r1.xyz, r0.y, imm, r1 */
+	PUSH_DATA (push, 0x1c9cab00);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c804);
+	PUSH_DATA (push, 0x00000000); /* { 0.00, -0.39, 2.02, 0.00 } */
+	PUSH_DATA (push, 0xbec890d6);
+	PUSH_DATA (push, 0x40011687);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x04000e81); /* madr r0.xyz, r0.z, imm, r1 */
+	PUSH_DATA (push, 0x1c9d5500);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c804);
+	PUSH_DATA (push, 0x3fcc432d); /* { 1.60, -0.81, 0.00, 0.00 } */
+	PUSH_DATA (push, 0xbf501a37);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+
+
+	PUSH_DATAu(push, pNv->scratch, PFP_NV12_BICUBIC, 29 * 4);
+	PUSH_DATA (push, 0x01008600); /* movr r0.xy, a[tex0] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x03000800); /* addr r0.z, r0.y, imm.x */
+	PUSH_DATA (push, 0x1c9caa00);
+	PUSH_DATA (push, 0x00000002);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3f000000); /* { 0.50, 0.00, 0.00, 0.00 } */
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x03000202); /* addr r1.x, r0, imm.x */
+	PUSH_DATA (push, 0x1c9dc800);
+	PUSH_DATA (push, 0x00000002);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3f000000); /* { 0.50, 0.00, 0.00, 0.00 } */
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x17000f82); /* texrc0 r1.xyz, r0.z, t[0] */
+	PUSH_DATA (push, 0x1c9d5400);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x02001404); /* mulr r2.yw, r1.xxyy, imm.xxyy */
+	PUSH_DATA (push, 0x1c9ca104);
+	PUSH_DATA (push, 0x0000a002);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0xbf800000); /* { -1.00, 1.00, 0.00, 0.00 } */
+	PUSH_DATA (push, 0x3f800000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x17000e86); /* texr r3.xyz, r1, t[0] */
+	PUSH_DATA (push, 0x1c9dc804);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x02000a04); /* mulr r2.xz, r3.xxyy, imm.xxyy */
+	PUSH_DATA (push, 0x1c9ca10c);
+	PUSH_DATA (push, 0x0000a002);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0xbf800000); /* { -1.00, 1.00, 0.00, 0.00 } */
+	PUSH_DATA (push, 0x3f800000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x03001e04); /* addr r2, r0.xyxy, r2 */
+	PUSH_DATA (push, 0x1c9c8800);
+	PUSH_DATA (push, 0x0001c808);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x17020402); /* texr r1.y, r2.zwzz, -t[1] */
+	PUSH_DATA (push, 0x1c9d5c08);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x04400282); /* madh r1.x, -r1.z, r1.y, r1.y */
+	PUSH_DATA (push, 0x1c9f5504);
+	PUSH_DATA (push, 0x0000aa04);
+	PUSH_DATA (push, 0x0000aa04);
+	PUSH_DATA (push, 0x17020400); /* texr r0.y, r2.xwxw, -t[1] */
+	PUSH_DATA (push, 0x1c9d9808);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x04401080); /* madh r0.w, -r1.z, r0.y, r0.y */
+	PUSH_DATA (push, 0x1c9f5504);
+	PUSH_DATA (push, 0x0000aa00);
+	PUSH_DATA (push, 0x0000aa00);
+	PUSH_DATA (push, 0x17020200); /* texr r0.x, r2.zyxy, t[1] */
+	PUSH_DATA (push, 0x1c9c8c08);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x04400282); /* madh r1.x, r1.z, r0, r1 */
+	PUSH_DATA (push, 0x1c9d5504);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c904);
+	PUSH_DATA (push, 0x17020200); /* texr r0.x (NE0.z), r2, t[1] */
+	PUSH_DATA (push, 0x1555c808);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x04400280); /* madh r0.x, r1.z, r0, r0.w */
+	PUSH_DATA (push, 0x1c9d5504);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x0001ff00);
+	PUSH_DATA (push, 0x04401080); /* madh r0.w, -r3.z, r1.x, r1.x */
+	PUSH_DATA (push, 0x1c9f550c);
+	PUSH_DATA (push, 0x00000104);
+	PUSH_DATA (push, 0x00000104);
+	PUSH_DATA (push, 0x1704ac80); /* texr r0.yz, a[tex1], t[2] */
+	PUSH_DATA (push, 0x1c9dc801);
+	PUSH_DATA (push, 0x0001c800);
+	PUSH_DATA (push, 0x3fe1c800);
+	PUSH_DATA (push, 0x04400280); /* madh r0.x, r3.z, r0, r0.w */
+	PUSH_DATA (push, 0x1c9d550c);
+	PUSH_DATA (push, 0x0001c900);
+	PUSH_DATA (push, 0x0001ff00);
+	PUSH_DATA (push, 0x04400e82); /* madh r1.xyz, r0.x, imm.x, imm.yzww */
+	PUSH_DATA (push, 0x1c9c0100);
+	PUSH_DATA (push, 0x00000002);
+	PUSH_DATA (push, 0x0001f202);
+	PUSH_DATA (push, 0x3f9507c8); /* { 1.16, -0.87, 0.53, -1.08 } */
+	PUSH_DATA (push, 0xbf5ee393);
+	PUSH_DATA (push, 0x3f078fef);
+	PUSH_DATA (push, 0xbf8a6762);
+	PUSH_DATA (push, 0x04400e82); /* madh r1.xyz, r0.y, imm, r1 */
+	PUSH_DATA (push, 0x1c9cab00);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c904);
+	PUSH_DATA (push, 0x00000000); /* { 0.00, -0.39, 2.02, 0.00 } */
+	PUSH_DATA (push, 0xbec890d6);
+	PUSH_DATA (push, 0x40011687);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x04400e81); /* madh r0.xyz, r0.z, imm, r1 */
+	PUSH_DATA (push, 0x1c9d5500);
+	PUSH_DATA (push, 0x0001c802);
+	PUSH_DATA (push, 0x0001c904);
+	PUSH_DATA (push, 0x3fcc432d); /* { 1.60, -0.81, 0.00, 0.00 } */
+	PUSH_DATA (push, 0xbf501a37);
+	PUSH_DATA (push, 0x00000000);
+	PUSH_DATA (push, 0x00000000);
 	return TRUE;
 }
diff --git a/src/nv40_xv_tex.c b/src/nv40_xv_tex.c
index 78dd0d6..2184678 100644
--- a/src/nv40_xv_tex.c
+++ b/src/nv40_xv_tex.c
@@ -36,8 +36,6 @@
 #include "nv_include.h"
 #include "nv_dma.h"
 
-#include "nv30_shaders.h"
-
 #include "hwdefs/nv30-40_3d.xml.h"
 #include "nv04_accel.h"
 
@@ -236,17 +234,18 @@ NV40PutTextureImage(ScrnInfoPtr pScrn,
 		return BadImplementation;
 	}
 
-	NV40_LoadVtxProg(pScrn, &nv40_vp_video);
-
 	if (drw_w / 2 < src_w || drw_h / 2 < src_h)
 		bicubic = FALSE;
 
-	if (!NV40_LoadFragProg(pScrn, bicubic ?
-			       &nv40_fp_yv12_bicubic :
-			       &nv30_fp_yv12_bilinear)) {
-		PUSH_RESET(push);
-		return BadImplementation;
-	}
+	BEGIN_NV04(push, NV30_3D(FP_ACTIVE_PROGRAM), 1);
+	PUSH_MTHD (push, NV30_3D(FP_ACTIVE_PROGRAM), pNv->scratch,
+			 bicubic ? PFP_NV12_BICUBIC : PFP_NV12_BILINEAR,
+			 NOUVEAU_BO_VRAM | NOUVEAU_BO_RD | NOUVEAU_BO_LOW |
+			 NOUVEAU_BO_OR,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA0,
+			 NV30_3D_FP_ACTIVE_PROGRAM_DMA1);
+	BEGIN_NV04(push, NV30_3D(FP_CONTROL), 1);
+	PUSH_DATA (push, 0x04000000);
 
 	/* Appears to be some kind of cache flush, needed here at least
 	 * sometimes.. funky text rendering otherwise :)