diff options
author | Oliver McFadden <z3ro.geek@gmail.com> | 2007-05-09 02:35:07 +0000 |
---|---|---|
committer | Oliver McFadden <z3ro.geek@gmail.com> | 2007-05-09 02:35:07 +0000 |
commit | 59494bafb482b30f3eef468d6f11396418a09a15 (patch) | |
tree | f922e619c75bbec34a04efcc693f27590ee26867 | |
parent | 33939fd29b01e1de7262eb042e4091f1caa8c367 (diff) |
r300: Use an array for the vertex program sources and corrected an error from
4960af08ad50a3a6ea039145de2698234c2cc892.
-rw-r--r-- | src/mesa/drivers/dri/r300/r300_state.c | 6 | ||||
-rw-r--r-- | src/mesa/drivers/dri/r300/r300_vertprog.c | 156 | ||||
-rw-r--r-- | src/mesa/drivers/dri/r300/r300_vertprog.h | 6 |
3 files changed, 84 insertions, 84 deletions
diff --git a/src/mesa/drivers/dri/r300/r300_state.c b/src/mesa/drivers/dri/r300/r300_state.c index 3e06c44381..96678b82a2 100644 --- a/src/mesa/drivers/dri/r300/r300_state.c +++ b/src/mesa/drivers/dri/r300/r300_state.c @@ -1573,9 +1573,9 @@ static void r300GenerateSimpleVertexShader(r300ContextPtr r300) #define WRITE_OP(oper,source1,source2,source3) {\ r300->state.vertex_shader.program.body.i[r300->state.vertex_shader.program_end].op=(oper); \ - r300->state.vertex_shader.program.body.i[r300->state.vertex_shader.program_end].src1=(source1); \ - r300->state.vertex_shader.program.body.i[r300->state.vertex_shader.program_end].src2=(source2); \ - r300->state.vertex_shader.program.body.i[r300->state.vertex_shader.program_end].src3=(source3); \ + r300->state.vertex_shader.program.body.i[r300->state.vertex_shader.program_end].src[0]=(source1); \ + r300->state.vertex_shader.program.body.i[r300->state.vertex_shader.program_end].src[1]=(source2); \ + r300->state.vertex_shader.program.body.i[r300->state.vertex_shader.program_end].src[2]=(source3); \ r300->state.vertex_shader.program_end++; \ } diff --git a/src/mesa/drivers/dri/r300/r300_vertprog.c b/src/mesa/drivers/dri/r300/r300_vertprog.c index e8917980ba..486a3f2f04 100644 --- a/src/mesa/drivers/dri/r300/r300_vertprog.c +++ b/src/mesa/drivers/dri/r300/r300_vertprog.c @@ -468,13 +468,13 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_ADD, u_temp_i, VSF_FLAG_ALL, VSF_OUT_CLASS_TMP); - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[2]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[2]), SWIZZLE_X, SWIZZLE_Y, SWIZZLE_Z, SWIZZLE_W, t_src_class(src[2].File), VSF_FLAG_NONE) | (src[2].RelAddr << 4); - o_inst->src2=ZERO_SRC_2; - o_inst->src3=ZERO_SRC_2; + o_inst->src[1]=ZERO_SRC_2; + o_inst->src[2]=ZERO_SRC_2; o_inst++; src[2].File=PROGRAM_TEMPORARY; @@ -490,13 +490,13 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_ADD, u_temp_i, VSF_FLAG_ALL, VSF_OUT_CLASS_TMP); - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), SWIZZLE_X, SWIZZLE_Y, SWIZZLE_Z, SWIZZLE_W, t_src_class(src[0].File), VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src2=ZERO_SRC_0; - o_inst->src3=ZERO_SRC_0; + o_inst->src[1]=ZERO_SRC_0; + o_inst->src[2]=ZERO_SRC_0; o_inst++; src[0].File=PROGRAM_TEMPORARY; @@ -511,9 +511,9 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct case OPCODE_POW: o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_POW, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src_scalar(vp, &src[0]); - o_inst->src2=ZERO_SRC_0; - o_inst->src3=t_src_scalar(vp, &src[1]); + o_inst->src[0]=t_src_scalar(vp, &src[0]); + o_inst->src[1]=ZERO_SRC_0; + o_inst->src[2]=t_src_scalar(vp, &src[1]); goto next; case OPCODE_MOV://ADD RESULT 1.X Y Z W PARAM 0{} {X Y Z W} PARAM 0{} {ZERO ZERO ZERO ZERO} @@ -521,17 +521,17 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct #if 1 o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_ADD, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=ZERO_SRC_0; - o_inst->src3=ZERO_SRC_0; + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=ZERO_SRC_0; + o_inst->src[2]=ZERO_SRC_0; #else hw_op=(src[0].File == PROGRAM_TEMPORARY) ? R300_VPI_OUT_OP_MAD_2 : R300_VPI_OUT_OP_MAD; o_inst->op=MAKE_VSF_OP(hw_op, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=ONE_SRC_0; - o_inst->src3=ZERO_SRC_0; + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=ONE_SRC_0; + o_inst->src[2]=ZERO_SRC_0; #endif goto next; @@ -543,15 +543,15 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(hw_op, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=ONE_SRC_0; - o_inst->src2=t_src(vp, &src[0]); - o_inst->src3=t_src(vp, &src[1]); + o_inst->src[0]=ONE_SRC_0; + o_inst->src[1]=t_src(vp, &src[0]); + o_inst->src[2]=t_src(vp, &src[1]); #else o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_ADD, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=t_src(vp, &src[1]); - o_inst->src3=ZERO_SRC_1; + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=t_src(vp, &src[1]); + o_inst->src[2]=ZERO_SRC_1; #endif goto next; @@ -563,9 +563,9 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(hw_op, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=t_src(vp, &src[1]); - o_inst->src3=t_src(vp, &src[2]); + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=t_src(vp, &src[1]); + o_inst->src[2]=t_src(vp, &src[2]); goto next; case OPCODE_MUL: /* HW mul can take third arg but appears to have some other limitations. */ @@ -574,17 +574,17 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(hw_op, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=t_src(vp, &src[1]); + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=t_src(vp, &src[1]); - o_inst->src3=ZERO_SRC_1; + o_inst->src[2]=ZERO_SRC_1; goto next; case OPCODE_DP3://DOT RESULT 1.X Y Z W PARAM 0{} {X Y Z ZERO} PARAM 0{} {X Y Z ZERO} o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_DOT, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), t_swizzle(GET_SWZ(src[0].Swizzle, 1)), t_swizzle(GET_SWZ(src[0].Swizzle, 2)), @@ -592,7 +592,7 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct t_src_class(src[0].File), src[0].NegateBase ? VSF_FLAG_XYZ : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src2=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), + o_inst->src[1]=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), t_swizzle(GET_SWZ(src[1].Swizzle, 0)), t_swizzle(GET_SWZ(src[1].Swizzle, 1)), t_swizzle(GET_SWZ(src[1].Swizzle, 2)), @@ -600,7 +600,7 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct t_src_class(src[1].File), src[1].NegateBase ? VSF_FLAG_XYZ : VSF_FLAG_NONE) | (src[1].RelAddr << 4); - o_inst->src3=ZERO_SRC_1; + o_inst->src[2]=ZERO_SRC_1; goto next; case OPCODE_SUB://ADD RESULT 1.X Y Z W TMP 0{} {X Y Z W} PARAM 1{X Y Z W } {X Y Z W} neg Xneg Yneg Zneg W @@ -610,9 +610,9 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(hw_op, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=ONE_SRC_0; - o_inst->src3=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=ONE_SRC_0; + o_inst->src[2]=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), t_swizzle(GET_SWZ(src[1].Swizzle, 0)), t_swizzle(GET_SWZ(src[1].Swizzle, 1)), t_swizzle(GET_SWZ(src[1].Swizzle, 2)), @@ -623,15 +623,15 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_ADD, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), t_swizzle(GET_SWZ(src[1].Swizzle, 0)), t_swizzle(GET_SWZ(src[1].Swizzle, 1)), t_swizzle(GET_SWZ(src[1].Swizzle, 2)), t_swizzle(GET_SWZ(src[1].Swizzle, 3)), t_src_class(src[1].File), (!src[1].NegateBase) ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[1].RelAddr << 4); - o_inst->src3=0; + o_inst->src[2]=0; #endif goto next; @@ -639,15 +639,15 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_MAX, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), t_swizzle(GET_SWZ(src[0].Swizzle, 1)), t_swizzle(GET_SWZ(src[0].Swizzle, 2)), t_swizzle(GET_SWZ(src[0].Swizzle, 3)), t_src_class(src[0].File), (!src[0].NegateBase) ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src3=0; + o_inst->src[2]=0; goto next; case OPCODE_FLR: @@ -657,16 +657,16 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_FRC, u_temp_i, t_dst_mask(vpi->DstReg.WriteMask), VSF_OUT_CLASS_TMP); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=ZERO_SRC_0; - o_inst->src3=ZERO_SRC_0; + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=ZERO_SRC_0; + o_inst->src[2]=ZERO_SRC_0; o_inst++; o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_ADD, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=MAKE_VSF_SOURCE(u_temp_i, + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=MAKE_VSF_SOURCE(u_temp_i, VSF_IN_COMPONENT_X, VSF_IN_COMPONENT_Y, VSF_IN_COMPONENT_Z, @@ -675,7 +675,7 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct /* Not 100% sure about this */ (!src[0].NegateBase) ? VSF_FLAG_ALL : VSF_FLAG_NONE/*VSF_FLAG_ALL*/); - o_inst->src3=ZERO_SRC_0; + o_inst->src[2]=ZERO_SRC_0; u_temp_i--; goto next; @@ -683,36 +683,36 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_LG2, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), t_src_class(src[0].File), src[0].NegateBase ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src2=ZERO_SRC_0; - o_inst->src3=ZERO_SRC_0; + o_inst->src[1]=ZERO_SRC_0; + o_inst->src[2]=ZERO_SRC_0; goto next; case OPCODE_LIT://LIT TMP 1.Y Z TMP 1{} {X W Z Y} TMP 1{} {Y W Z X} TMP 1{} {Y X Z W} o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_LIT, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); /* NOTE: Users swizzling might not work. */ - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), // x t_swizzle(GET_SWZ(src[0].Swizzle, 3)), // w VSF_IN_COMPONENT_ZERO, // z t_swizzle(GET_SWZ(src[0].Swizzle, 1)), // y t_src_class(src[0].File), src[0].NegateBase ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src2=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[1]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 1)), // y t_swizzle(GET_SWZ(src[0].Swizzle, 3)), // w VSF_IN_COMPONENT_ZERO, // z t_swizzle(GET_SWZ(src[0].Swizzle, 0)), // x t_src_class(src[0].File), src[0].NegateBase ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src3=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[2]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 1)), // y t_swizzle(GET_SWZ(src[0].Swizzle, 0)), // x VSF_IN_COMPONENT_ZERO, // z @@ -725,15 +725,15 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_DOT, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 0)), t_swizzle(GET_SWZ(src[0].Swizzle, 1)), t_swizzle(GET_SWZ(src[0].Swizzle, 2)), VSF_IN_COMPONENT_ONE, t_src_class(src[0].File), src[0].NegateBase ? VSF_FLAG_XYZ : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src2=t_src(vp, &src[1]); - o_inst->src3=ZERO_SRC_1; + o_inst->src[1]=t_src(vp, &src[1]); + o_inst->src[2]=ZERO_SRC_1; goto next; case OPCODE_XPD: @@ -745,7 +745,7 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_MAD, u_temp_i, t_dst_mask(vpi->DstReg.WriteMask), VSF_OUT_CLASS_TMP); - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 1)), // y t_swizzle(GET_SWZ(src[0].Swizzle, 2)), // z t_swizzle(GET_SWZ(src[0].Swizzle, 0)), // x @@ -753,7 +753,7 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct t_src_class(src[0].File), src[0].NegateBase ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src2=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), + o_inst->src[1]=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), t_swizzle(GET_SWZ(src[1].Swizzle, 2)), // z t_swizzle(GET_SWZ(src[1].Swizzle, 0)), // x t_swizzle(GET_SWZ(src[1].Swizzle, 1)), // y @@ -761,14 +761,14 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct t_src_class(src[1].File), src[1].NegateBase ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[1].RelAddr << 4); - o_inst->src3=ZERO_SRC_1; + o_inst->src[2]=ZERO_SRC_1; o_inst++; u_temp_i--; o_inst->op=MAKE_VSF_OP(R300_VPI_OUT_OP_MAD, t_dst_index(vp, &vpi->DstReg), t_dst_mask(vpi->DstReg.WriteMask), t_dst_class(vpi->DstReg.File)); - o_inst->src1=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), + o_inst->src[0]=MAKE_VSF_SOURCE(t_src_index(vp, &src[1]), t_swizzle(GET_SWZ(src[1].Swizzle, 1)), // y t_swizzle(GET_SWZ(src[1].Swizzle, 2)), // z t_swizzle(GET_SWZ(src[1].Swizzle, 0)), // x @@ -776,7 +776,7 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct t_src_class(src[1].File), (!src[1].NegateBase) ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[1].RelAddr << 4); - o_inst->src2=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), + o_inst->src[1]=MAKE_VSF_SOURCE(t_src_index(vp, &src[0]), t_swizzle(GET_SWZ(src[0].Swizzle, 2)), // z t_swizzle(GET_SWZ(src[0].Swizzle, 0)), // x t_swizzle(GET_SWZ(src[0].Swizzle, 1)), // y @@ -784,7 +784,7 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct t_src_class(src[0].File), src[0].NegateBase ? VSF_FLAG_ALL : VSF_FLAG_NONE) | (src[0].RelAddr << 4); - o_inst->src3=MAKE_VSF_SOURCE(u_temp_i+1, + o_inst->src[2]=MAKE_VSF_SOURCE(u_temp_i+1, VSF_IN_COMPONENT_X, VSF_IN_COMPONENT_Y, VSF_IN_COMPONENT_Z, @@ -810,21 +810,21 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct if(are_srcs_scalar){ switch(operands){ case 1: - o_inst->src1=t_src_scalar(vp, &src[0]); - o_inst->src2=ZERO_SRC_0; - o_inst->src3=ZERO_SRC_0; + o_inst->src[0]=t_src_scalar(vp, &src[0]); + o_inst->src[1]=ZERO_SRC_0; + o_inst->src[2]=ZERO_SRC_0; break; case 2: - o_inst->src1=t_src_scalar(vp, &src[0]); - o_inst->src2=t_src_scalar(vp, &src[1]); - o_inst->src3=ZERO_SRC_1; + o_inst->src[0]=t_src_scalar(vp, &src[0]); + o_inst->src[1]=t_src_scalar(vp, &src[1]); + o_inst->src[2]=ZERO_SRC_1; break; case 3: - o_inst->src1=t_src_scalar(vp, &src[0]); - o_inst->src2=t_src_scalar(vp, &src[1]); - o_inst->src3=t_src_scalar(vp, &src[2]); + o_inst->src[0]=t_src_scalar(vp, &src[0]); + o_inst->src[1]=t_src_scalar(vp, &src[1]); + o_inst->src[2]=t_src_scalar(vp, &src[2]); break; default: @@ -835,21 +835,21 @@ static void r300_translate_vertex_shader(struct r300_vertex_program *vp, struct }else{ switch(operands){ case 1: - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=ZERO_SRC_0; - o_inst->src3=ZERO_SRC_0; + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=ZERO_SRC_0; + o_inst->src[2]=ZERO_SRC_0; break; case 2: - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=t_src(vp, &src[1]); - o_inst->src3=ZERO_SRC_1; + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=t_src(vp, &src[1]); + o_inst->src[2]=ZERO_SRC_1; break; case 3: - o_inst->src1=t_src(vp, &src[0]); - o_inst->src2=t_src(vp, &src[1]); - o_inst->src3=t_src(vp, &src[2]); + o_inst->src[0]=t_src(vp, &src[0]); + o_inst->src[1]=t_src(vp, &src[1]); + o_inst->src[2]=t_src(vp, &src[2]); break; default: diff --git a/src/mesa/drivers/dri/r300/r300_vertprog.h b/src/mesa/drivers/dri/r300/r300_vertprog.h index 433e98960b..005fca976b 100644 --- a/src/mesa/drivers/dri/r300/r300_vertprog.h +++ b/src/mesa/drivers/dri/r300/r300_vertprog.h @@ -1,11 +1,11 @@ #ifndef __R300_VERTPROG_H_ #define __R300_VERTPROG_H_ +#include "r300_reg.h" + typedef struct { GLuint op; - GLuint src1; - GLuint src2; - GLuint src3; + GLuint src[3]; } VERTEX_SHADER_INSTRUCTION; #define VSF_FLAG_X 1 |