i965: Fix ENDLOOP to only patch up this loop's BREAK and CONT.

[mesa.git] / src / gallium / drivers / nv30 / nv30_fragprog.c

diff --git a/src/gallium/drivers/nv30/nv30_fragprog.c b/src/gallium/drivers/nv30/nv30_fragprog.c
index 14dc884b3acfbe53746b63146c518385d0dd5f95..2c432c6dfa7caff343a5b7effa8eb0bea3a4ca5d 100644 (file)
--- a/src/gallium/drivers/nv30/nv30_fragprog.c
+++ b/src/gallium/drivers/nv30/nv30_fragprog.c
@@ -1,7 +1,7 @@
 #include "pipe/p_context.h"
 #include "pipe/p_defines.h"
 #include "pipe/p_state.h"
-#include "pipe/p_inlines.h"
+#include "util/u_inlines.h"
 
 #include "pipe/p_shader_tokens.h"
 #include "tgsi/tgsi_dump.h"
@@ -237,20 +237,20 @@ tgsi_src(struct nv30_fpc *fpc, const struct tgsi_full_src_register *fsrc)
 {
        struct nv30_sreg src;
 
-       switch (fsrc->SrcRegister.File) {
+       switch (fsrc->Register.File) {
        case TGSI_FILE_INPUT:
                src = nv30_sr(NV30SR_INPUT,
-                             fpc->attrib_map[fsrc->SrcRegister.Index]);
+                             fpc->attrib_map[fsrc->Register.Index]);
                break;
        case TGSI_FILE_CONSTANT:
-               src = constant(fpc, fsrc->SrcRegister.Index, NULL);
+               src = constant(fpc, fsrc->Register.Index, NULL);
                break;
        case TGSI_FILE_IMMEDIATE:
-               assert(fsrc->SrcRegister.Index < fpc->nr_imm);
-               src = fpc->imm[fsrc->SrcRegister.Index];
+               assert(fsrc->Register.Index < fpc->nr_imm);
+               src = fpc->imm[fsrc->Register.Index];
                break;
        case TGSI_FILE_TEMPORARY:
-               src = nv30_sr(NV30SR_TEMP, fsrc->SrcRegister.Index + 1);
+               src = nv30_sr(NV30SR_TEMP, fsrc->Register.Index + 1);
                if (fpc->high_temp < src.index)
                        fpc->high_temp = src.index;
                break;
@@ -258,7 +258,7 @@ tgsi_src(struct nv30_fpc *fpc, const struct tgsi_full_src_register *fsrc)
         * Luckily fragprog results are just temp regs..
         */
        case TGSI_FILE_OUTPUT:
-               if (fsrc->SrcRegister.Index == fpc->colour_id)
+               if (fsrc->Register.Index == fpc->colour_id)
                        return nv30_sr(NV30SR_OUTPUT, 0);
                else
                        return nv30_sr(NV30SR_OUTPUT, 1);
@@ -268,12 +268,12 @@ tgsi_src(struct nv30_fpc *fpc, const struct tgsi_full_src_register *fsrc)
                break;
        }
 
-       src.abs = fsrc->SrcRegister.Absolute;
-       src.negate = fsrc->SrcRegister.Negate;
-       src.swz[0] = fsrc->SrcRegister.SwizzleX;
-       src.swz[1] = fsrc->SrcRegister.SwizzleY;
-       src.swz[2] = fsrc->SrcRegister.SwizzleZ;
-       src.swz[3] = fsrc->SrcRegister.SwizzleW;
+       src.abs = fsrc->Register.Absolute;
+       src.negate = fsrc->Register.Negate;
+       src.swz[0] = fsrc->Register.SwizzleX;
+       src.swz[1] = fsrc->Register.SwizzleY;
+       src.swz[2] = fsrc->Register.SwizzleZ;
+       src.swz[3] = fsrc->Register.SwizzleW;
        return src;
 }
 
@@ -364,7 +364,7 @@ nv30_fragprog_parse_instruction(struct nv30_fpc *fpc,
                const struct tgsi_full_src_register *fsrc;
 
                fsrc = &finst->Src[i];
-               if (fsrc->SrcRegister.File == TGSI_FILE_TEMPORARY) {
+               if (fsrc->Register.File == TGSI_FILE_TEMPORARY) {
                        src[i] = tgsi_src(fpc, fsrc);
                }
        }
@@ -374,7 +374,7 @@ nv30_fragprog_parse_instruction(struct nv30_fpc *fpc,
 
                fsrc = &finst->Src[i];
 
-               switch (fsrc->SrcRegister.File) {
+               switch (fsrc->Register.File) {
                case TGSI_FILE_INPUT:
                case TGSI_FILE_CONSTANT:
                case TGSI_FILE_TEMPORARY:
@@ -385,14 +385,14 @@ nv30_fragprog_parse_instruction(struct nv30_fpc *fpc,
                        break;
                }
 
-               switch (fsrc->SrcRegister.File) {
+               switch (fsrc->Register.File) {
                case TGSI_FILE_INPUT:
-                       if (ai == -1 || ai == fsrc->SrcRegister.Index) {
-                               ai = fsrc->SrcRegister.Index;
+                       if (ai == -1 || ai == fsrc->Register.Index) {
+                               ai = fsrc->Register.Index;
                                src[i] = tgsi_src(fpc, fsrc);
                        } else {
                                NOUVEAU_MSG("extra src attr %d\n",
-                                        fsrc->SrcRegister.Index);
+                                        fsrc->Register.Index);
                                src[i] = temp(fpc);
                                arith(fpc, 0, MOV, src[i], MASK_ALL,
                                      tgsi_src(fpc, fsrc), none, none);
@@ -400,8 +400,8 @@ nv30_fragprog_parse_instruction(struct nv30_fpc *fpc,
                        break;
                case TGSI_FILE_CONSTANT:
                case TGSI_FILE_IMMEDIATE:
-                       if (ci == -1 || ci == fsrc->SrcRegister.Index) {
-                               ci = fsrc->SrcRegister.Index;
+                       if (ci == -1 || ci == fsrc->Register.Index) {
+                               ci = fsrc->Register.Index;
                                src[i] = tgsi_src(fpc, fsrc);
                        } else {
                                src[i] = temp(fpc);
@@ -413,7 +413,7 @@ nv30_fragprog_parse_instruction(struct nv30_fpc *fpc,
                        /* handled above */
                        break;
                case TGSI_FILE_SAMPLER:
-                       unit = fsrc->SrcRegister.Index;
+                       unit = fsrc->Register.Index;
                        break;
                case TGSI_FILE_OUTPUT:
                        break;
@@ -435,10 +435,11 @@ nv30_fragprog_parse_instruction(struct nv30_fpc *fpc,
                arith(fpc, sat, ADD, dst, mask, src[0], src[1], none);
                break;
        case TGSI_OPCODE_CMP:
-               tmp = temp(fpc);
-               arith(fpc, sat, MOV, dst, mask, src[2], none, none);
+               tmp = nv30_sr(NV30SR_NONE, 0);
                tmp.cc_update = 1;
                arith(fpc, 0, MOV, tmp, 0xf, src[0], none, none);
+               dst.cc_test = NV30_VP_INST_COND_GE;
+               arith(fpc, sat, MOV, dst, mask, src[2], none, none);
                dst.cc_test = NV30_VP_INST_COND_LT;
                arith(fpc, sat, MOV, dst, mask, src[1], none, none);
                break;
@@ -517,13 +518,28 @@ nv30_fragprog_parse_instruction(struct nv30_fpc *fpc,
                arith(fpc, sat, RSQ, dst, mask, abs(swz(src[0], X, X, X, X)), none, none);
                break;
        case TGSI_OPCODE_SCS:
-               if (mask & MASK_X) {
-                       arith(fpc, sat, COS, dst, MASK_X,
-                             swz(src[0], X, X, X, X), none, none);
+               /* avoid overwriting the source */
+               if(src[0].swz[SWZ_X] != SWZ_X)
+               {
+                       if (mask & MASK_X) {
+                               arith(fpc, sat, COS, dst, MASK_X,
+                                     swz(src[0], X, X, X, X), none, none);
+                       }
+                       if (mask & MASK_Y) {
+                               arith(fpc, sat, SIN, dst, MASK_Y,
+                                     swz(src[0], X, X, X, X), none, none);
+                       }
                }
-               if (mask & MASK_Y) {
-                       arith(fpc, sat, SIN, dst, MASK_Y,
-                             swz(src[0], X, X, X, X), none, none);
+               else
+               {
+                       if (mask & MASK_Y) {
+                               arith(fpc, sat, SIN, dst, MASK_Y,
+                                     swz(src[0], X, X, X, X), none, none);
+                       }
+                       if (mask & MASK_X) {
+                               arith(fpc, sat, COS, dst, MASK_X,
+                                     swz(src[0], X, X, X, X), none, none);
+                       }
                }
                break;
        case TGSI_OPCODE_SIN:
@@ -821,7 +837,7 @@ nv30_fragprog_validate(struct nv30_context *nv30)
        fp->buffer = pscreen->buffer_create(pscreen, 0x100, 0, fp->insn_len * 4);
        nv30_fragprog_upload(nv30, fp);
 
-       so = so_new(8, 1);
+       so = so_new(4, 4, 1);
        so_method(so, nv30->screen->rankine, NV34TCL_FP_ACTIVE_PROGRAM, 1);
        so_reloc (so, nouveau_bo(fp->buffer), 0, NOUVEAU_BO_VRAM |
                      NOUVEAU_BO_GART | NOUVEAU_BO_RD | NOUVEAU_BO_LOW |
@@ -870,6 +886,12 @@ void
 nv30_fragprog_destroy(struct nv30_context *nv30,
                      struct nv30_fragment_program *fp)
 {
+       if (fp->buffer)
+               pipe_buffer_reference(&fp->buffer, NULL);
+
+       if (fp->so)
+               so_ref(NULL, &fp->so);
+
        if (fp->insn_len)
                FREE(fp->insn);
 }