Increment version to 10.6.0-rc2

Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

VERSION

@@ -1 +1 @@
-10.6.0-devel
+10.6.0-rc2

configure.ac

@@ -649,6 +649,7 @@ if test "x$enable_asm" = xyes; then
 fi
 
 AC_CHECK_HEADER([xlocale.h], [DEFINES="$DEFINES -DHAVE_XLOCALE_H"])
+AC_CHECK_HEADER([sys/sysctl.h], [DEFINES="$DEFINES -DHAVE_SYS_SYSCTL_H"])
 AC_CHECK_FUNC([strtof], [DEFINES="$DEFINES -DHAVE_STRTOF"])
 
 dnl Check to see if dlopen is in default libraries (like Solaris, which

src/gallium/auxiliary/cso_cache/cso_context.c

@@ -300,6 +300,8 @@ void cso_destroy_context( struct cso_context *ctx )
    unsigned i, shader;
 
    if (ctx->pipe) {
+      ctx->pipe->set_index_buffer(ctx->pipe, NULL);
+
       ctx->pipe->bind_blend_state( ctx->pipe, NULL );
       ctx->pipe->bind_rasterizer_state( ctx->pipe, NULL );
 
@@ -326,13 +328,18 @@ void cso_destroy_context( struct cso_context *ctx )
 
       ctx->pipe->bind_depth_stencil_alpha_state( ctx->pipe, NULL );
       ctx->pipe->bind_fs_state( ctx->pipe, NULL );
+      ctx->pipe->set_constant_buffer(ctx->pipe, PIPE_SHADER_FRAGMENT, 0, NULL);
       ctx->pipe->bind_vs_state( ctx->pipe, NULL );
+      ctx->pipe->set_constant_buffer(ctx->pipe, PIPE_SHADER_VERTEX, 0, NULL);
       if (ctx->has_geometry_shader) {
          ctx->pipe->bind_gs_state(ctx->pipe, NULL);
+         ctx->pipe->set_constant_buffer(ctx->pipe, PIPE_SHADER_GEOMETRY, 0, NULL);
       }
       if (ctx->has_tessellation) {
          ctx->pipe->bind_tcs_state(ctx->pipe, NULL);
+         ctx->pipe->set_constant_buffer(ctx->pipe, PIPE_SHADER_TESS_CTRL, 0, NULL);
          ctx->pipe->bind_tes_state(ctx->pipe, NULL);
+         ctx->pipe->set_constant_buffer(ctx->pipe, PIPE_SHADER_TESS_EVAL, 0, NULL);
       }
       ctx->pipe->bind_vertex_elements_state( ctx->pipe, NULL );
 

src/gallium/auxiliary/util/u_math.h

@@ -42,6 +42,7 @@
 #include "pipe/p_compiler.h"
 
 #include "c99_math.h"
+#include <assert.h>
 #include <float.h>
 #include <stdarg.h>
 

src/gallium/drivers/freedreno/a3xx/fd3_program.c

@@ -413,12 +413,15 @@ fd3_program_emit(struct fd_ringbuffer *ring, struct fd3_emit *emit,
 				}
 			}
 
-			/* TODO: Figure out if there's a way to make it spit out 0's and
-			 * 1's for the .z and .w components.
+			/* Replace the .xy coordinates with S/T from the point sprite. Set
+			 * interpolation bits for .zw such that they become .01
 			 */
-			if (emit->sprite_coord_enable & (1 << sem2idx(fp->inputs[j].semantic)))
+			if (emit->sprite_coord_enable & (1 << sem2idx(fp->inputs[j].semantic))) {
 				vpsrepl[inloc / 16] |= (emit->sprite_coord_mode ? 0x0d : 0x09)
 					<< ((inloc % 16) * 2);
+				vinterp[(inloc + 2) / 16] |= 2 << (((inloc + 2) % 16) * 2);
+				vinterp[(inloc + 3) / 16] |= 3 << (((inloc + 3) % 16) * 2);
+			}
 		}
 
 		OUT_PKT0(ring, REG_A3XX_VPC_ATTR, 2);

src/gallium/drivers/freedreno/ir3/ir3_compiler.c

@@ -1615,7 +1615,7 @@ trans_samp(const struct instr_translater *t,
 			instr->cat1.src_type = type_mov;
 			instr->cat1.dst_type = type_mov;
 			add_dst_reg(ctx, instr, &tmp_dst, i);
-			add_src_reg(ctx, instr, &zero, 0);
+			add_src_reg(ctx, instr, &zero, zero.SwizzleX);
 			i++;
 		}
 		if (tgt->array) {
@@ -1669,15 +1669,18 @@ trans_samp(const struct instr_translater *t,
 	 */
 	if (inst->Instruction.Opcode == TGSI_OPCODE_TXD) {
 		while (collect->regs_count < 5)
-			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA), &zero, 0);
+			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA),
+					&zero, zero.SwizzleX);
 		for (i = 0; i < tgt->dims; i++)
 			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA), dpdx, i);
 		if (tgt->dims < 2)
-			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA), &zero, 0);
+			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA),
+					&zero, zero.SwizzleX);
 		for (i = 0; i < tgt->dims; i++)
 			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA), dpdy, i);
 		if (tgt->dims < 2)
-			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA), &zero, 0);
+			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA),
+					&zero, zero.SwizzleX);
 		tinf.src_wrmask |= ((1 << (2 * MAX2(tgt->dims, 2))) - 1) << 4;
 	}
 
@@ -1700,7 +1703,8 @@ trans_samp(const struct instr_translater *t,
 			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA),
 					offset, i);
 		if (tgt->dims < 2)
-			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA), &zero, 0);
+			ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA),
+					&zero, zero.SwizzleX);
 	}
 	if (inst->Instruction.Opcode == TGSI_OPCODE_TXB2)
 		ssa_src(ctx, ir3_reg_create(collect, 0, IR3_REG_SSA),

src/gallium/drivers/llvmpipe/lp_query.c

@@ -315,7 +315,7 @@ llvmpipe_check_render_cond(struct llvmpipe_context *lp)
 
    b = pipe->get_query_result(pipe, lp->render_cond_query, wait, (void*)&result);
    if (b)
-      return (!result == lp->render_cond_cond);
+      return ((!result) == lp->render_cond_cond);
    else
       return TRUE;
 }

src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp

@@ -1356,18 +1356,20 @@ Converter::srcToSym(tgsi::Instruction::SrcRegister src, int c)
 {
    const int swz = src.getSwizzle(c);
 
+   /* TODO: Use Array ID when it's available for the index */
    return makeSym(src.getFile(),
                   src.is2D() ? src.getIndex(1) : 0,
-                  src.isIndirect(0) ? -1 : src.getIndex(0), swz,
+                  src.getIndex(0), swz,
                   src.getIndex(0) * 16 + swz * 4);
 }
 
 Symbol *
 Converter::dstToSym(tgsi::Instruction::DstRegister dst, int c)
 {
+   /* TODO: Use Array ID when it's available for the index */
    return makeSym(dst.getFile(),
                   dst.is2D() ? dst.getIndex(1) : 0,
-                  dst.isIndirect(0) ? -1 : dst.getIndex(0), c,
+                  dst.getIndex(0), c,
                   dst.getIndex(0) * 16 + c * 4);
 }
 

src/gallium/drivers/nouveau/codegen/nv50_ir_lowering_gm107.cpp

@@ -240,6 +240,7 @@ GM107LoweringPass::visit(Instruction *i)
             Value *ptr = bld.mkOp2v(OP_SHL, TYPE_U32, bld.getSSA(),
                                     i->getIndirect(0, 0), bld.mkImm(4));
             i->setIndirect(0, 0, ptr);
+            i->op = OP_VFETCH;
          } else {
             i->op = OP_VFETCH;
             assert(prog->getType() != Program::TYPE_FRAGMENT); // INTERP

src/gallium/drivers/nouveau/codegen/nv50_ir_lowering_nvc0.cpp

@@ -100,8 +100,7 @@ void
 NVC0LegalizeSSA::handleFTZ(Instruction *i)
 {
    // Only want to flush float inputs
-   if (i->sType != TYPE_F32)
-      return;
+   assert(i->sType == TYPE_F32);
 
    // If we're already flushing denorms (and NaN's) to zero, no need for this.
    if (i->dnz)
@@ -129,7 +128,7 @@ NVC0LegalizeSSA::visit(BasicBlock *bb)
    Instruction *next;
    for (Instruction *i = bb->getEntry(); i; i = next) {
       next = i->next;
-      if (i->dType == TYPE_F32) {
+      if (i->sType == TYPE_F32) {
          if (prog->getType() != Program::TYPE_COMPUTE)
             handleFTZ(i);
          continue;
@@ -1751,6 +1750,7 @@ NVC0LoweringPass::visit(Instruction *i)
             Value *ptr = bld.mkOp2v(OP_SHL, TYPE_U32, bld.getSSA(),
                                     i->getIndirect(0, 0), bld.mkImm(4));
             i->setIndirect(0, 0, ptr);
+            i->op = OP_VFETCH;
          } else {
             i->op = OP_VFETCH;
             assert(prog->getType() != Program::TYPE_FRAGMENT); // INTERP

src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp

@@ -236,6 +236,9 @@ LoadPropagation::visit(BasicBlock *bb)
       if (i->op == OP_CALL) // calls have args as sources, they must be in regs
          continue;
 
+      if (i->op == OP_PFETCH) // pfetch expects arg1 to be a reg
+         continue;
+
       if (i->srcExists(1))
          checkSwapSrc01(i);
 
@@ -574,6 +577,11 @@ ConstantFolding::expr(Instruction *i,
    case OP_POPCNT:
       res.data.u32 = util_bitcount(a->data.u32 & b->data.u32);
       break;
+   case OP_PFETCH:
+      // The two arguments to pfetch are logically added together. Normally
+      // the second argument will not be constant, but that can happen.
+      res.data.u32 = a->data.u32 + b->data.u32;
+      break;
    default:
       return;
    }
@@ -588,7 +596,9 @@ ConstantFolding::expr(Instruction *i,
 
    i->getSrc(0)->reg.data = res.data;
 
-   if (i->op == OP_MAD || i->op == OP_FMA) {
+   switch (i->op) {
+   case OP_MAD:
+   case OP_FMA: {
       i->op = OP_ADD;
 
       i->setSrc(1, i->getSrc(0));
@@ -603,8 +613,14 @@ ConstantFolding::expr(Instruction *i,
          bld.setPosition(i, false);
          i->setSrc(1, bld.loadImm(NULL, res.data.u32));
       }
-   } else {
+      break;
+   }
+   case OP_PFETCH:
+      // Leave PFETCH alone... we just folded its 2 args into 1.
+      break;
+   default:
       i->op = i->saturate ? OP_SAT : OP_MOV; /* SAT handled by unary() */
+      break;
    }
    i->subOp = 0;
 }

src/gallium/drivers/nouveau/nv30/nv30_draw.c

@@ -71,12 +71,12 @@ nv30_render_allocate_vertices(struct vbuf_render *render,
    struct nv30_render *r = nv30_render(render);
    struct nv30_context *nv30 = r->nv30;
 
-   r->length = vertex_size * nr_vertices;
+   r->length = (uint32_t)vertex_size * (uint32_t)nr_vertices;
 
    if (r->offset + r->length >= render->max_vertex_buffer_bytes) {
       pipe_resource_reference(&r->buffer, NULL);
       r->buffer = pipe_buffer_create(&nv30->screen->base.base,
-                                     PIPE_BIND_VERTEX_BUFFER, 0,
+                                     PIPE_BIND_VERTEX_BUFFER, PIPE_USAGE_STREAM,
                                      render->max_vertex_buffer_bytes);
       if (!r->buffer)
          return FALSE;
@@ -91,10 +91,14 @@ static void *
 nv30_render_map_vertices(struct vbuf_render *render)
 {
    struct nv30_render *r = nv30_render(render);
-   char *map = pipe_buffer_map(&r->nv30->base.pipe, r->buffer,
-                               PIPE_TRANSFER_WRITE |
-                               PIPE_TRANSFER_UNSYNCHRONIZED, &r->transfer);
-   return map + r->offset;
+   char *map = pipe_buffer_map_range(
+         &r->nv30->base.pipe, r->buffer,
+         r->offset, r->length,
+         PIPE_TRANSFER_WRITE |
+         PIPE_TRANSFER_DISCARD_RANGE,
+         &r->transfer);
+   assert(map);
+   return map;
 }
 
 static void
@@ -103,6 +107,7 @@ nv30_render_unmap_vertices(struct vbuf_render *render,
 {
    struct nv30_render *r = nv30_render(render);
    pipe_buffer_unmap(&r->nv30->base.pipe, r->transfer);
+   r->transfer = NULL;
 }
 
 static void
@@ -126,7 +131,7 @@ nv30_render_draw_elements(struct vbuf_render *render,
    for (i = 0; i < r->vertex_info.num_attribs; i++) {
       PUSH_RESRC(push, NV30_3D(VTXBUF(i)), BUFCTX_VTXTMP,
                        nv04_resource(r->buffer), r->offset + r->vtxptr[i],
-                       NOUVEAU_BO_LOW | NOUVEAU_BO_RD, 0, 0);
+                       NOUVEAU_BO_LOW | NOUVEAU_BO_RD, 0, NV30_3D_VTXBUF_DMA1);
    }
 
    if (!nv30_state_validate(nv30, FALSE))
@@ -171,7 +176,7 @@ nv30_render_draw_arrays(struct vbuf_render *render, unsigned start, uint nr)
    for (i = 0; i < r->vertex_info.num_attribs; i++) {
       PUSH_RESRC(push, NV30_3D(VTXBUF(i)), BUFCTX_VTXTMP,
                        nv04_resource(r->buffer), r->offset + r->vtxptr[i],
-                       NOUVEAU_BO_LOW | NOUVEAU_BO_RD, 0, 0);
+                       NOUVEAU_BO_LOW | NOUVEAU_BO_RD, 0, NV30_3D_VTXBUF_DMA1);
    }
 
    if (!nv30_state_validate(nv30, FALSE))
@@ -213,22 +218,24 @@ static const struct {
    [TGSI_SEMANTIC_BCOLOR  ] = { EMIT_4F, INTERP_LINEAR     , 1, 3, 0x00000004 },
    [TGSI_SEMANTIC_FOG     ] = { EMIT_4F, INTERP_PERSPECTIVE, 5, 5, 0x00000010 },
    [TGSI_SEMANTIC_PSIZE   ] = { EMIT_1F_PSIZE, INTERP_POS  , 6, 6, 0x00000020 },
-   [TGSI_SEMANTIC_GENERIC ] = { EMIT_4F, INTERP_PERSPECTIVE, 8, 7, 0x00004000 }
+   [TGSI_SEMANTIC_TEXCOORD] = { EMIT_4F, INTERP_PERSPECTIVE, 8, 7, 0x00004000 },
 };
 
 static boolean
 vroute_add(struct nv30_render *r, uint attrib, uint sem, uint *idx)
 {
-   struct pipe_screen *pscreen = &r->nv30->screen->base.base;
+   struct nv30_screen *screen = r->nv30->screen;
    struct nv30_fragprog *fp = r->nv30->fragprog.program;
    struct vertex_info *vinfo = &r->vertex_info;
    enum pipe_format format;
    uint emit = EMIT_OMIT;
    uint result = *idx;
 
-   if (sem == TGSI_SEMANTIC_GENERIC && result >= 8) {
-      for (result = 0; result < 8; result++) {
-         if (fp->texcoord[result] == *idx) {
+   if (sem == TGSI_SEMANTIC_GENERIC) {
+      uint num_texcoords = (screen->eng3d->oclass < NV40_3D_CLASS) ? 8 : 10;
+      for (result = 0; result < num_texcoords; result++) {
+         if (fp->texcoord[result] == *idx + 8) {
+            sem = TGSI_SEMANTIC_TEXCOORD;
             emit = vroute[sem].emit;
             break;
          }
@@ -243,11 +250,11 @@ vroute_add(struct nv30_render *r, uint attrib, uint sem, uint *idx)
    draw_emit_vertex_attr(vinfo, emit, vroute[sem].interp, attrib);
    format = draw_translate_vinfo_format(emit);
 
-   r->vtxfmt[attrib] = nv30_vtxfmt(pscreen, format)->hw;
-   r->vtxptr[attrib] = vinfo->size | NV30_3D_VTXBUF_DMA1;
+   r->vtxfmt[attrib] = nv30_vtxfmt(&screen->base.base, format)->hw;
+   r->vtxptr[attrib] = vinfo->size;
    vinfo->size += draw_translate_vinfo_size(emit);
 
-   if (nv30_screen(pscreen)->eng3d->oclass < NV40_3D_CLASS) {
+   if (screen->eng3d->oclass < NV40_3D_CLASS) {
       r->vtxprog[attrib][0] = 0x001f38d8;
       r->vtxprog[attrib][1] = 0x0080001b | (attrib << 9);
       r->vtxprog[attrib][2] = 0x0836106c;
@@ -259,7 +266,12 @@ vroute_add(struct nv30_render *r, uint attrib, uint sem, uint *idx)
       r->vtxprog[attrib][3] = 0x6041ff80 | (result + vroute[sem].vp40) << 2;
    }
 
-   *idx = vroute[sem].ow40 << result;
+   if (result < 8)
+      *idx = vroute[sem].ow40 << result;
+   else {
+      assert(sem == TGSI_SEMANTIC_TEXCOORD);
+      *idx = 0x00001000 << (result - 8);
+   }
    return TRUE;
 }
 
@@ -313,7 +325,7 @@ nv30_render_validate(struct nv30_context *nv30)
 
    while (pntc && attrib < 16) {
       uint index = ffs(pntc) - 1; pntc &= ~(1 << index);
-      if (vroute_add(r, attrib, TGSI_SEMANTIC_GENERIC, &index)) {
+      if (vroute_add(r, attrib, TGSI_SEMANTIC_TEXCOORD, &index)) {
          vp_attribs |= (1 << attrib++);
          vp_results |= index;
       }
@@ -398,17 +410,17 @@ nv30_render_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
       if (nv30->vertprog.constbuf) {
          void *map = nv04_resource(nv30->vertprog.constbuf)->data;
          draw_set_mapped_constant_buffer(draw, PIPE_SHADER_VERTEX, 0,
-                                         map, nv30->vertprog.constbuf_nr);
+                                         map, nv30->vertprog.constbuf_nr * 16);
+      } else {
+         draw_set_mapped_constant_buffer(draw, PIPE_SHADER_VERTEX, 0, NULL, 0);
       }
    }
 
    for (i = 0; i < nv30->num_vtxbufs; i++) {
       const void *map = nv30->vtxbuf[i].user_buffer;
       if (!map) {
-         if (!nv30->vtxbuf[i].buffer) {
-            continue;
-         }
-         map = pipe_buffer_map(pipe, nv30->vtxbuf[i].buffer,
+         if (nv30->vtxbuf[i].buffer)
+            map = pipe_buffer_map(pipe, nv30->vtxbuf[i].buffer,
                                   PIPE_TRANSFER_UNSYNCHRONIZED |
                                   PIPE_TRANSFER_READ, &transfer[i]);
       }
@@ -418,9 +430,9 @@ nv30_render_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
    if (info->indexed) {
       const void *map = nv30->idxbuf.user_buffer;
       if (!map)
-         pipe_buffer_map(pipe, nv30->idxbuf.buffer,
-                                  PIPE_TRANSFER_UNSYNCHRONIZED |
-                                  PIPE_TRANSFER_READ, &transferi);
+         map = pipe_buffer_map(pipe, nv30->idxbuf.buffer,
+                               PIPE_TRANSFER_UNSYNCHRONIZED |
+                               PIPE_TRANSFER_READ, &transferi);
       draw_set_indexes(draw,
                        (ubyte *) map + nv30->idxbuf.offset,
                        nv30->idxbuf.index_size, ~0);
@@ -444,6 +456,12 @@ nv30_render_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
 static void
 nv30_render_destroy(struct vbuf_render *render)
 {
+   struct nv30_render *r = nv30_render(render);
+
+   if (r->transfer)
+      pipe_buffer_unmap(&r->nv30->base.pipe, r->transfer);
+   pipe_resource_reference(&r->buffer, NULL);
+   nouveau_heap_free(&r->vertprog);
    FREE(render);
 }
 

src/gallium/drivers/nouveau/nv30/nv30_fragprog.c

@@ -23,6 +23,7 @@
  *
  */
 
+#include "draw/draw_context.h"
 #include "tgsi/tgsi_parse.h"
 
 #include "nv_object.xml.h"
@@ -147,8 +148,12 @@ nv30_fp_state_delete(struct pipe_context *pipe, void *hwcso)
 
    pipe_resource_reference(&fp->buffer, NULL);
 
+   if (fp->draw)
+      draw_delete_fragment_shader(nv30_context(pipe)->draw, fp->draw);
+
    FREE((void *)fp->pipe.tokens);
    FREE(fp->insn);
+   FREE(fp->consts);
    FREE(fp);
 }
 

src/gallium/drivers/nouveau/nv30/nv30_state_validate.c

@@ -272,15 +272,13 @@ nv30_validate_clip(struct nv30_context *nv30)
    uint32_t clpd_enable = 0;
 
    for (i = 0; i < 6; i++) {
-      if (nv30->rast->pipe.clip_plane_enable & (1 << i)) {
-         if (nv30->dirty & NV30_NEW_CLIP) {
-            BEGIN_NV04(push, NV30_3D(VP_UPLOAD_CONST_ID), 5);
-            PUSH_DATA (push, i);
-            PUSH_DATAp(push, nv30->clip.ucp[i], 4);
-         }
-
-         clpd_enable |= 1 << (1 + 4*i);
+      if (nv30->dirty & NV30_NEW_CLIP) {
+         BEGIN_NV04(push, NV30_3D(VP_UPLOAD_CONST_ID), 5);
+         PUSH_DATA (push, i);
+         PUSH_DATAp(push, nv30->clip.ucp[i], 4);
       }
+      if (nv30->rast->pipe.clip_plane_enable & (1 << i))
+         clpd_enable |= 2 << (4*i);
    }
 
    BEGIN_NV04(push, NV30_3D(VP_CLIP_PLANES_ENABLE), 1);
@@ -389,7 +387,7 @@ static struct state_validate hwtnl_validate_list[] = {
     { nv30_validate_stipple,       NV30_NEW_STIPPLE },
     { nv30_validate_scissor,       NV30_NEW_SCISSOR | NV30_NEW_RASTERIZER },
     { nv30_validate_viewport,      NV30_NEW_VIEWPORT },
-    { nv30_validate_clip,          NV30_NEW_CLIP },
+    { nv30_validate_clip,          NV30_NEW_CLIP | NV30_NEW_RASTERIZER },
     { nv30_fragprog_validate,      NV30_NEW_FRAGPROG | NV30_NEW_FRAGCONST },
     { nv30_vertprog_validate,      NV30_NEW_VERTPROG | NV30_NEW_VERTCONST |
                                    NV30_NEW_FRAGPROG | NV30_NEW_RASTERIZER },

src/gallium/drivers/nouveau/nv30/nv30_vertprog.c

@@ -23,6 +23,7 @@
  *
  */
 
+#include "draw/draw_context.h"
 #include "util/u_dynarray.h"
 #include "tgsi/tgsi_parse.h"
 
@@ -237,6 +238,10 @@ nv30_vp_state_delete(struct pipe_context *pipe, void *hwcso)
 
    if (vp->translated)
       nv30_vertprog_destroy(vp);
+
+   if (vp->draw)
+      draw_delete_vertex_shader(nv30_context(pipe)->draw, vp->draw);
+
    FREE((void *)vp->pipe.tokens);
    FREE(vp);
 }

src/gallium/drivers/nouveau/nv30/nvfx_vertprog.c

@@ -872,9 +872,8 @@ nvfx_vertprog_parse_decl_output(struct nvfx_vpc *vpc,
       }
       break;
    case TGSI_SEMANTIC_EDGEFLAG:
-      /* not really an error just a fallback */
-      NOUVEAU_ERR("cannot handle edgeflag output\n");
-      return FALSE;
+      vpc->r_result[idx] = nvfx_reg(NVFXSR_NONE, 0);
+      return TRUE;
    default:
       NOUVEAU_ERR("bad output semantic\n");
       return FALSE;

src/gallium/drivers/nouveau/nvc0/nvc0_program.c

@@ -392,7 +392,7 @@ nvc0_gp_gen_header(struct nvc0_program *gp, struct nv50_ir_prog_info *info)
       break;
    }
 
-   gp->hdr[4] = info->prop.gp.maxVertices & 0x1ff;
+   gp->hdr[4] = MIN2(info->prop.gp.maxVertices, 1024);
 
    return nvc0_vtgp_gen_header(gp, info);
 }

src/gallium/drivers/radeonsi/si_state_shaders.c

@@ -835,8 +835,15 @@ static void si_update_spi_tmpring_size(struct si_context *sctx)
 			si_pm4_bind_state(sctx, ps, sctx->ps_shader->current->pm4);
 		if (si_update_scratch_buffer(sctx, sctx->gs_shader))
 			si_pm4_bind_state(sctx, gs, sctx->gs_shader->current->pm4);
-		if (si_update_scratch_buffer(sctx, sctx->vs_shader))
-			si_pm4_bind_state(sctx, vs, sctx->vs_shader->current->pm4);
+
+		/* VS can be bound as ES or VS. */
+		if (sctx->gs_shader) {
+			if (si_update_scratch_buffer(sctx, sctx->vs_shader))
+				si_pm4_bind_state(sctx, es, sctx->vs_shader->current->pm4);
+		} else {
+			if (si_update_scratch_buffer(sctx, sctx->vs_shader))
+				si_pm4_bind_state(sctx, vs, sctx->vs_shader->current->pm4);
+		}
 	}
 
 	/* The LLVM shader backend should be reporting aligned scratch_sizes. */

src/gallium/state_trackers/clover/core/error.hpp

@@ -26,6 +26,7 @@
 #include "CL/cl.h"
 
 #include <stdexcept>
+#include <string>
 
 namespace clover {
    class command_queue;

src/gallium/targets/osmesa/Makefile.am

@@ -42,7 +42,6 @@ nodist_EXTRA_lib@OSMESA_LIB@_la_SOURCES = dummy.cpp
 lib@OSMESA_LIB@_la_SOURCES = target.c
 
 lib@OSMESA_LIB@_la_LDFLAGS = \
-	-module \
 	-no-undefined \
 	-version-number @OSMESA_VERSION@ \
 	$(GC_SECTIONS) \

src/glsl/linker.cpp

@@ -2829,8 +2829,11 @@ link_shaders(struct gl_context *ctx, struct gl_shader_program *prog)
             link_intrastage_shaders(mem_ctx, ctx, prog, shader_list[stage],
                                     num_shaders[stage]);
 
-         if (!prog->LinkStatus)
+         if (!prog->LinkStatus) {
+            if (sh)
+               ctx->Driver.DeleteShader(ctx, sh);
             goto done;
+         }
 
          switch (stage) {
          case MESA_SHADER_VERTEX:
@@ -2843,8 +2846,11 @@ link_shaders(struct gl_context *ctx, struct gl_shader_program *prog)
             validate_fragment_shader_executable(prog, sh);
             break;
          }
-         if (!prog->LinkStatus)
+         if (!prog->LinkStatus) {
+            if (sh)
+               ctx->Driver.DeleteShader(ctx, sh);
             goto done;
+         }
 
          _mesa_reference_shader(ctx, &prog->_LinkedShaders[stage], sh);
       }

src/mesa/drivers/dri/i965/brw_fs.cpp

@@ -1050,7 +1050,7 @@ fs_visitor::implied_mrf_writes(fs_inst *inst)
    case FS_OPCODE_VARYING_PULL_CONSTANT_LOAD:
       return inst->mlen;
    case SHADER_OPCODE_GEN4_SCRATCH_WRITE:
-      return 2;
+      return inst->mlen;
    case SHADER_OPCODE_UNTYPED_ATOMIC:
    case SHADER_OPCODE_UNTYPED_SURFACE_READ:
    case SHADER_OPCODE_UNTYPED_SURFACE_WRITE:

src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp

@@ -2460,11 +2460,27 @@ vec4_visitor::emit_mcs_fetch(ir_texture *ir, src_reg coordinate, src_reg sampler
       new(mem_ctx) vec4_instruction(SHADER_OPCODE_TXF_MCS,
                                     dst_reg(this, glsl_type::uvec4_type));
    inst->base_mrf = 2;
-   inst->mlen = 1;
    inst->src[1] = sampler;
 
+   int param_base;
+
+   if (devinfo->gen >= 9) {
+      /* Gen9+ needs a message header in order to use SIMD4x2 mode */
+      vec4_instruction *header_inst = new(mem_ctx)
+         vec4_instruction(VS_OPCODE_SET_SIMD4X2_HEADER_GEN9,
+                          dst_reg(MRF, inst->base_mrf));
+
+      emit(header_inst);
+
+      inst->mlen = 2;
+      inst->header_size = 1;
+      param_base = inst->base_mrf + 1;
+   } else {
+      inst->mlen = 1;
+      param_base = inst->base_mrf;
+   }
+
    /* parameters are: u, v, r, lod; lod will always be zero due to api restrictions */
-   int param_base = inst->base_mrf;
    int coord_mask = (1 << ir->coordinate->type->vector_elements) - 1;
    int zero_mask = 0xf & ~coord_mask;
 

src/mesa/drivers/dri/swrast/swrast.c

@@ -62,7 +62,9 @@
 #include "swrast/s_context.h"
 
 #include <sys/types.h>
-#include <sys/sysctl.h>
+#ifdef HAVE_SYS_SYSCTL_H
+# include <sys/sysctl.h>
+#endif
 
 const __DRIextension **__driDriverGetExtensions_swrast(void);
 

src/mesa/drivers/osmesa/Makefile.am

@@ -39,7 +39,6 @@ nodist_EXTRA_lib@OSMESA_LIB@_la_SOURCES = dummy.cpp
 lib@OSMESA_LIB@_la_SOURCES = osmesa.c
 
 lib@OSMESA_LIB@_la_LDFLAGS = \
-	-module \
 	-no-undefined \
 	-version-number @OSMESA_VERSION@ \
 	$(GC_SECTIONS) \

src/mesa/main/api_exec.h

@@ -38,6 +38,9 @@ _mesa_initialize_exec_table(struct gl_context *ctx);
 extern void
 _mesa_initialize_dispatch_tables(struct gl_context *ctx);
 
+extern struct _glapi_table *
+_mesa_new_nop_table(unsigned numEntries);
+
 #ifdef __cplusplus
 } // extern "C"
 #endif

src/mesa/main/context.c

@@ -883,6 +883,19 @@ update_default_objects(struct gl_context *ctx)
 }
 
 
+/* XXX this is temporary and should be removed at some point in the
+ * future when there's a reasonable expectation that the libGL library
+ * contains the _glapi_new_nop_table() and _glapi_set_nop_handler()
+ * functions which were added in Mesa 10.6.
+ */
+#if !defined(_WIN32)
+/* Avoid libGL / driver ABI break */
+#define USE_GLAPI_NOP_FEATURES 0
+#else
+#define USE_GLAPI_NOP_FEATURES 1
+#endif
+
+
 /**
  * This function is called by the glapi no-op functions.  For each OpenGL
  * function/entrypoint there's a simple no-op function.  These "no-op"
@@ -898,6 +911,7 @@ update_default_objects(struct gl_context *ctx)
  *
  * \param name  the name of the OpenGL function
  */
+#if USE_GLAPI_NOP_FEATURES
 static void
 nop_handler(const char *name)
 {
@@ -914,6 +928,7 @@ nop_handler(const char *name)
    }
 #endif
 }
+#endif
 
 
 /**
@@ -923,11 +938,51 @@ nop_handler(const char *name)
 static void GLAPIENTRY
 nop_glFlush(void)
 {
-   /* don't record an error like we do in _mesa_generic_nop() */
+   /* don't record an error like we do in nop_handler() */
 }
 #endif
 
 
+#if !USE_GLAPI_NOP_FEATURES
+static int
+generic_nop(void)
+{
+   GET_CURRENT_CONTEXT(ctx);
+   _mesa_error(ctx, GL_INVALID_OPERATION,
+               "unsupported function called "
+               "(unsupported extension or deprecated function?)");
+   return 0;
+}
+#endif
+
+
+/**
+ * Create a new API dispatch table in which all entries point to the
+ * generic_nop() function.  This will not work on Windows because of
+ * the __stdcall convention which requires the callee to clean up the
+ * call stack.  That's impossible with one generic no-op function.
+ */
+struct _glapi_table *
+_mesa_new_nop_table(unsigned numEntries)
+{
+   struct _glapi_table *table;
+
+#if !USE_GLAPI_NOP_FEATURES
+   table = malloc(numEntries * sizeof(_glapi_proc));
+   if (table) {
+      _glapi_proc *entry = (_glapi_proc *) table;
+      unsigned i;
+      for (i = 0; i < numEntries; i++) {
+         entry[i] = (_glapi_proc) generic_nop;
+      }
+   }
+#else
+   table = _glapi_new_nop_table(numEntries);
+#endif
+   return table;
+}
+
+
 /**
  * Allocate and initialize a new dispatch table.  The table will be
  * populated with pointers to "no-op" functions.  In turn, the no-op
@@ -941,8 +996,9 @@ alloc_dispatch_table(void)
     * Mesa we do this to accommodate different versions of libGL and various
     * DRI drivers.
     */
-   GLint numEntries = MAX2(_glapi_get_dispatch_table_size(), _gloffset_COUNT);
-   struct _glapi_table *table = _glapi_new_nop_table(numEntries);
+   int numEntries = MAX2(_glapi_get_dispatch_table_size(), _gloffset_COUNT);
+
+   struct _glapi_table *table = _mesa_new_nop_table(numEntries);
 
 #if defined(_WIN32)
    if (table) {
@@ -966,7 +1022,9 @@ alloc_dispatch_table(void)
    }
 #endif
 
+#if USE_GLAPI_NOP_FEATURES
    _glapi_set_nop_handler(nop_handler);
+#endif
 
    return table;
 }

src/mesa/main/tests/dispatch_sanity.cpp

@@ -96,7 +96,7 @@ DispatchSanity_test::SetUp()
    _mesa_init_driver_functions(&driver_functions);
 
    const unsigned size = _glapi_get_dispatch_table_size();
-   nop_table = (_glapi_proc *) _glapi_new_nop_table(size);
+   nop_table = (_glapi_proc *) _mesa_new_nop_table(size);
 }
 
 void

src/mesa/state_tracker/st_context.c

@@ -376,12 +376,6 @@ void st_destroy_context( struct st_context *st )
    }
    pipe_surface_reference(&st->state.framebuffer.zsbuf, NULL);
 
-   pipe->set_index_buffer(pipe, NULL);
-
-   for (i = 0; i < PIPE_SHADER_TYPES; i++) {
-      pipe->set_constant_buffer(pipe, i, 0, NULL);
-   }
-
    _mesa_delete_program_cache(st->ctx, st->pixel_xfer.cache);
 
    _vbo_DestroyContext(st->ctx);

src/mesa/state_tracker/st_glsl_to_tgsi.cpp

@@ -5549,6 +5549,7 @@ get_mesa_program(struct gl_context *ctx,
     */
    _mesa_associate_uniform_storage(ctx, shader_program, prog->Parameters);
    if (!shader_program->LinkStatus) {
+      free_glsl_to_tgsi_visitor(v);
       return NULL;
    }