darwin: do not create double-buffered offscreen pixel formats

http://xquartz.macosforge.org/trac/ticket/536 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com> (cherry picked from commit 5fdf1f784b)
darwin: Eliminate a possible race condition while destroying a surface
2012-09-24 16:07:14 -07:00 · 2012-05-16 18:59:30 -07:00 · 2012-05-16 18:59:28 -07:00 · 2012-04-24 00:27:44 -07:00 · 2012-04-23 16:18:36 -07:00 · 2012-04-22 20:41:36 -07:00
37 changed files with 440 additions and 128 deletions
--- a/2
+++ b/2
@@ -183,7 +183,7 @@ ultrix-gcc:

 # Rules for making release tarballs

-VERSION=7.11.1
+VERSION=7.11.2
 DIRECTORY = Mesa-$(VERSION)
 LIB_NAME = MesaLib-$(VERSION)
 GLUT_NAME = MesaGLUT-$(VERSION)
--- a/configure.ac
+++ b/configure.ac
@@ -75,13 +75,13 @@ AC_COMPILE_IFELSE(
       not clang
 #endif
 ]])],
-[CLANG=yes], [CLANG=no])
+[acv_mesa_CLANG=yes], [acv_mesa_CLANG=no])

-AC_MSG_RESULT([$CLANG])
+AC_MSG_RESULT([$acv_mesa_CLANG])

 dnl If we're using GCC, make sure that it is at least version 3.3.0.  Older
 dnl versions are explictly not supported.
-if test "x$GCC" = xyes -a "x$CLANG" = xno; then
+if test "x$GCC" = xyes -a "x$acv_mesa_CLANG" = xno; then
    AC_MSG_CHECKING([whether gcc version is sufficient])
    major=0
    minor=0
--- a/docs/news.html
+++ b/docs/news.html
@@ -11,6 +11,23 @@
 <H1>News</H1>


+<h2>November 27, 2011</h2>
+
+<p>
+<a href="relnotes-7.11.2.html">Mesa 7.11.2</a> is released.
+This release was made primarily to fix build problems with 7.11.1 on
+Mandriva and to fix problems related to glCopyTexImage to
+luminance-alpha textures.  That later was believed to have been fixed
+in 7.11.1 but was not.
+</p>
+
+<h2>November 17, 2011</h2>
+
+<p>
+<a href="relnotes-7.11.1.html">Mesa 7.11.1</a> is released.  This is a bug fix
+release.
+</p>
+
 <h2>July 31, 2011</h2>

 <p>
--- a/docs/relnotes-7.11.1.html
+++ b/docs/relnotes-7.11.1.html
@@ -28,10 +28,14 @@ for DRI hardware acceleration.

 <h2>MD5 checksums</h2>
 <pre>
-TBD
+ac0181a4076770fb657c1169af43aa09  MesaLib-7.11.1.tar.gz
+a77307102cee844ff6544ffa8fafeac1  MesaLib-7.11.1.tar.bz2
+dfcb11516c1730f3981b55a65a835623  MesaLib-7.11.1.zip
+2cb2b9ecb4fb7d1a6be69346ee886952  MesaGLUT-7.11.1.tar.gz
+3f54e314290d4dacbab089839197080b  MesaGLUT-7.11.1.tar.bz2
+5d66c7ee8c5cc2f27e1ffb037ad4172c  MesaGLUT-7.11.1.zip
 </pre>

-
 <h2>New features</h2>
 <p>None.</p>

--- a/docs/relnotes-7.11.2.html
+++ b/docs/relnotes-7.11.2.html
@@ -0,0 +1,85 @@
+<HTML>
+
+<head>
+<TITLE>Mesa Release Notes</TITLE>
+<link rel="stylesheet" type="text/css" href="mesa.css">
+<meta http-equiv="content-type" content="text/html; charset=utf-8" />
+</head>
+
+<BODY>
+
+<body bgcolor="#eeeeee">
+
+<H1>Mesa 7.11.2 Release Notes / November 27, 2011</H1>
+
+<p>
+Mesa 7.11.2 is a bug fix release which fixes bugs found since the 7.11 release.
+</p>
+<p>
+Mesa 7.11 implements the OpenGL 2.1 API, but the version reported by
+glGetString(GL_VERSION) depends on the particular driver being used.
+Some drivers don't support all the features required in OpenGL 2.1.
+</p>
+<p>
+See the <a href="install.html">Compiling/Installing page</a> for prerequisites
+for DRI hardware acceleration.
+</p>
+
+
+<h2>MD5 checksums</h2>
+<pre>
+b9e84efee3931c0acbccd1bb5a860554  MesaLib-7.11.2.tar.gz
+0837c52698fe3252369c3fdb5195afcc  MesaLib-7.11.2.tar.bz2
+141273c274d12e0d2bafb497fe937da3  MesaLib-7.11.2.zip
+39ae9926794794503815ffdc069521eb  MesaGLUT-7.11.2.tar.gz
+35ca3a0b54cb6f9d2e0e4eae8f6bb95e  MesaGLUT-7.11.2.tar.bz2
+f8705fcff2510b6c39cd27b575c05dba  MesaGLUT-7.11.2.zip
+</pre>
+
+<h2>New features</h2>
+<p>None.</p>
+
+<h2>Bug fixes</h2>
+
+<p>This list is likely incomplete.</p>
+
+<ul>
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=43143">Bug 43143</a> - Mesa 7.11.1 fails to build at main/dlist.c:4532 with error message: "format not a string literal and no format arguments"</li>
+
+<li>Incorrect handling of CopyTexImage from RGBA window to LA texture.</li>
+
+<!-- <li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=">Bug </a> - </li> -->
+</ul>
+
+
+<h2>Changes</h2>
+<p>The full set of changes can be viewed by using the following GIT command:</p>
+
+<pre>
+  git log mesa-7.11.1..mesa-7.11.2
+</pre>
+
+<p>Brian Paul (4):
+<ul>
+  <li>mesa: stop using ctx-&gt;Driver.CopyTexImage1D/2D() hooks</li>
+  <li>mesa: fix format selection for meta CopyTexSubImage()</li>
+  <li>docs: update news.html and relnotes.html for 7.11.1 release</li>
+  <li>mesa: use format string in _mesa_error() call to silence warning</li>
+</ul></p>
+
+<p>Chad Versace (3):
+<ul>
+  <li>intel: Simplify stencil detiling arithmetic</li>
+  <li>intel: Fix region dimensions for stencil buffers received from DDX</li>
+  <li>intel: Fix separate stencil in builtin DRI2 backend</li>
+</ul></p>
+
+<p>Ian Romanick (3):
+<ul>
+  <li>docs: Add 7.11.1 release md5sums</li>
+  <li>mesa: set version string to 7.11.2-devel</li>
+  <li>mesa: Bump version to 7.11.2 (final)</li>
+</ul></p>
+
+</body>
+</html>
--- a/docs/relnotes.html
+++ b/docs/relnotes.html
@@ -13,6 +13,8 @@ The release notes summarize what's new or changed in each Mesa release.
 </p>

 <UL>
+<LI><A HREF="relnotes-7.11.2.html">7.11.2 release notes</A>
+<LI><A HREF="relnotes-7.11.1.html">7.11.1 release notes</A>
 <LI><A HREF="relnotes-7.11.html">7.11 release notes</A>
 <LI><A HREF="relnotes-7.10.3.html">7.10.3 release notes</A>
 <LI><A HREF="relnotes-7.10.2.html">7.10.2 release notes</A>
--- a/src/gallium/auxiliary/draw/draw_llvm.c
+++ b/src/gallium/auxiliary/draw/draw_llvm.c
@@ -96,7 +96,7 @@ draw_llvm_generate_elts(struct draw_llvm *llvm, struct draw_llvm_variant *var);
 * Create LLVM type for struct draw_jit_texture
 */
 static LLVMTypeRef
-create_jit_texture_type(struct gallivm_state *gallivm)
+create_jit_texture_type(struct gallivm_state *gallivm, const char *struct_name)
 {
   LLVMTargetDataRef target = gallivm->target;
   LLVMTypeRef texture_type;
@@ -120,13 +120,21 @@ create_jit_texture_type(struct gallivm_state *gallivm)
   elem_types[DRAW_JIT_TEXTURE_BORDER_COLOR] = 
      LLVMArrayType(LLVMFloatTypeInContext(gallivm->context), 4);

+#if HAVE_LLVM >= 0x0300
+   texture_type = LLVMStructCreateNamed(gallivm->context, struct_name);
+   LLVMStructSetBody(texture_type, elem_types,
+                     Elements(elem_types), 0);
+#else
   texture_type = LLVMStructTypeInContext(gallivm->context, elem_types,
                                          Elements(elem_types), 0);

+   LLVMAddTypeName(gallivm->module, struct_name, texture_type);
+
   /* Make sure the target's struct layout cache doesn't return
    * stale/invalid data.
    */
   LLVMInvalidateStructLayout(gallivm->target, texture_type);
+#endif

   LP_CHECK_MEMBER_OFFSET(struct draw_jit_texture, width,
                          target, texture_type,
@@ -176,7 +184,7 @@ create_jit_texture_type(struct gallivm_state *gallivm)
 */
 static LLVMTypeRef
 create_jit_context_type(struct gallivm_state *gallivm,
-                        LLVMTypeRef texture_type)
+                        LLVMTypeRef texture_type, const char *struct_name)
 {
   LLVMTargetDataRef target = gallivm->target;
   LLVMTypeRef float_type = LLVMFloatTypeInContext(gallivm->context);
@@ -189,11 +197,17 @@ create_jit_context_type(struct gallivm_state *gallivm,
   elem_types[3] = LLVMPointerType(float_type, 0); /* viewport */
   elem_types[4] = LLVMArrayType(texture_type,
                                 PIPE_MAX_VERTEX_SAMPLERS); /* textures */
-
+#if HAVE_LLVM >= 0x0300
+   context_type = LLVMStructCreateNamed(gallivm->context, struct_name);
+   LLVMStructSetBody(context_type, elem_types,
+                     Elements(elem_types), 0);
+#else
   context_type = LLVMStructTypeInContext(gallivm->context, elem_types,
                                          Elements(elem_types), 0);
+   LLVMAddTypeName(gallivm->module, struct_name, context_type);

   LLVMInvalidateStructLayout(gallivm->target, context_type);
+#endif

   LP_CHECK_MEMBER_OFFSET(struct draw_jit_context, vs_constants,
                          target, context_type, 0);
@@ -215,7 +229,7 @@ create_jit_context_type(struct gallivm_state *gallivm,
 * Create LLVM type for struct pipe_vertex_buffer
 */
 static LLVMTypeRef
-create_jit_vertex_buffer_type(struct gallivm_state *gallivm)
+create_jit_vertex_buffer_type(struct gallivm_state *gallivm, const char *struct_name)
 {
   LLVMTargetDataRef target = gallivm->target;
   LLVMTypeRef elem_types[3];
@@ -225,10 +239,17 @@ create_jit_vertex_buffer_type(struct gallivm_state *gallivm)
   elem_types[1] = LLVMInt32TypeInContext(gallivm->context);
   elem_types[2] = LLVMPointerType(LLVMInt8TypeInContext(gallivm->context), 0); /* vs_constants */

+#if HAVE_LLVM >= 0x0300
+   vb_type = LLVMStructCreateNamed(gallivm->context, struct_name);
+   LLVMStructSetBody(vb_type, elem_types,
+                     Elements(elem_types), 0);
+#else
   vb_type = LLVMStructTypeInContext(gallivm->context, elem_types,
                                     Elements(elem_types), 0);
+   LLVMAddTypeName(gallivm->module, struct_name, vb_type);

   LLVMInvalidateStructLayout(gallivm->target, vb_type);
+#endif

   LP_CHECK_MEMBER_OFFSET(struct pipe_vertex_buffer, stride,
                          target, vb_type, 0);
@@ -258,10 +279,17 @@ create_jit_vertex_header(struct gallivm_state *gallivm, int data_elems)
   elem_types[1]  = LLVMArrayType(LLVMFloatTypeInContext(gallivm->context), 4);
   elem_types[2]  = LLVMArrayType(elem_types[1], data_elems);

+#if HAVE_LLVM >= 0x0300
+   vertex_header = LLVMStructCreateNamed(gallivm->context, struct_name);
+   LLVMStructSetBody(vertex_header, elem_types,
+                     Elements(elem_types), 0);
+#else
   vertex_header = LLVMStructTypeInContext(gallivm->context, elem_types,
                                           Elements(elem_types), 0);
+   LLVMAddTypeName(gallivm->module, struct_name, vertex_header);

   LLVMInvalidateStructLayout(gallivm->target, vertex_header);
+#endif

   /* these are bit-fields and we can't take address of them
      LP_CHECK_MEMBER_OFFSET(struct vertex_header, clipmask,
@@ -284,8 +312,6 @@ create_jit_vertex_header(struct gallivm_state *gallivm, int data_elems)
                          target, vertex_header,
                          DRAW_JIT_VERTEX_DATA);

-   LLVMAddTypeName(gallivm->module, struct_name, vertex_header);
-
   return vertex_header;
 }

@@ -299,19 +325,15 @@ create_jit_types(struct draw_llvm *llvm)
   struct gallivm_state *gallivm = llvm->gallivm;
   LLVMTypeRef texture_type, context_type, buffer_type, vb_type;

-   texture_type = create_jit_texture_type(gallivm);
-   LLVMAddTypeName(gallivm->module, "texture", texture_type);
+   texture_type = create_jit_texture_type(gallivm, "texture");

-   context_type = create_jit_context_type(gallivm, texture_type);
-   LLVMAddTypeName(gallivm->module, "draw_jit_context", context_type);
+   context_type = create_jit_context_type(gallivm, texture_type, "draw_jit_context");
   llvm->context_ptr_type = LLVMPointerType(context_type, 0);

   buffer_type = LLVMPointerType(LLVMIntTypeInContext(gallivm->context, 8), 0);
-   LLVMAddTypeName(gallivm->module, "buffer", buffer_type);
   llvm->buffer_ptr_type = LLVMPointerType(buffer_type, 0);

-   vb_type = create_jit_vertex_buffer_type(gallivm);
-   LLVMAddTypeName(gallivm->module, "pipe_vertex_buffer", vb_type);
+   vb_type = create_jit_vertex_buffer_type(gallivm, "pipe_vertex_buffer");
   llvm->vb_ptr_type = LLVMPointerType(vb_type, 0);
 }

--- a/src/gallium/auxiliary/gallivm/lp_bld_debug.cpp
+++ b/src/gallium/auxiliary/gallivm/lp_bld_debug.cpp
@@ -27,17 +27,23 @@

 #include <llvm-c/Core.h>
 #include <llvm/Target/TargetMachine.h>
-#include <llvm/Target/TargetRegistry.h>
-#include <llvm/Target/TargetSelect.h>
 #include <llvm/Target/TargetInstrInfo.h>
 #include <llvm/Support/raw_ostream.h>
 #include <llvm/Support/MemoryObject.h>

+#if HAVE_LLVM >= 0x0300
+#include <llvm/Support/TargetRegistry.h>
+#include <llvm/Support/TargetSelect.h>
+#else /* HAVE_LLVM < 0x0300 */
+#include <llvm/Target/TargetRegistry.h>
+#include <llvm/Target/TargetSelect.h>
+#endif /* HAVE_LLVM < 0x0300 */
+
 #if HAVE_LLVM >= 0x0209
 #include <llvm/Support/Host.h>
-#else
+#else /* HAVE_LLVM < 0x0209 */
 #include <llvm/System/Host.h>
-#endif
+#endif /* HAVE_LLVM < 0x0209 */

 #if HAVE_LLVM >= 0x0207
 #include <llvm/MC/MCDisassembler.h>
@@ -193,14 +199,23 @@ lp_disassemble(const void* func)

   InitializeAllDisassemblers();

+#if HAVE_LLVM >= 0x0300
+   OwningPtr<const MCAsmInfo> AsmInfo(T->createMCAsmInfo(Triple));
+#else
   OwningPtr<const MCAsmInfo> AsmInfo(T->createAsmInfo(Triple));
+#endif

   if (!AsmInfo) {
      debug_printf("error: no assembly info for target %s\n", Triple.c_str());
      return;
   }

+#if HAVE_LLVM >= 0x0300
+   const MCSubtargetInfo *STI = T->createMCSubtargetInfo(Triple, sys::getHostCPUName(), "");
+   OwningPtr<const MCDisassembler> DisAsm(T->createMCDisassembler(*STI));
+#else 
   OwningPtr<const MCDisassembler> DisAsm(T->createMCDisassembler());
+#endif 
   if (!DisAsm) {
      debug_printf("error: no disassembler for target %s\n", Triple.c_str());
      return;
@@ -213,7 +228,11 @@ lp_disassemble(const void* func)
 #else
   int AsmPrinterVariant = AsmInfo->getAssemblerDialect();
 #endif
-#if HAVE_LLVM >= 0x0208
+
+#if HAVE_LLVM >= 0x0300
+   OwningPtr<MCInstPrinter> Printer(
+         T->createMCInstPrinter(AsmPrinterVariant, *AsmInfo, *STI));
+#elif HAVE_LLVM >= 0x0208
   OwningPtr<MCInstPrinter> Printer(
         T->createMCInstPrinter(AsmPrinterVariant, *AsmInfo));
 #else
@@ -253,7 +272,11 @@ lp_disassemble(const void* func)

      if (!DisAsm->getInstruction(Inst, Size, memoryObject,
                                 pc,
-                                 nulls())) {
+#if HAVE_LLVM >= 0x0300
+				  nulls(), nulls())) {
+#else
+				  nulls())) {
+#endif
         debug_printf("invalid\n");
         pc += 1;
      }
@@ -276,7 +299,9 @@ lp_disassemble(const void* func)
       * Print the instruction.
       */

-#if HAVE_LLVM >= 0x208
+#if HAVE_LLVM >= 0x0300
+      Printer->printInst(&Inst, Out, "");
+#elif HAVE_LLVM >= 0x208
      Printer->printInst(&Inst, Out);
 #else
      Printer->printInst(&Inst);
@@ -289,7 +314,11 @@ lp_disassemble(const void* func)

      pc += Size;

+#if HAVE_LLVM >= 0x0300
+      const MCInstrDesc &TID = TII->get(Inst.getOpcode());
+#else
      const TargetInstrDesc &TID = TII->get(Inst.getOpcode());
+#endif

      /*
       * Keep track of forward jumps to a nearby address.
--- a/src/gallium/auxiliary/gallivm/lp_bld_type.c
+++ b/src/gallium/auxiliary/gallivm/lp_bld_type.c
@@ -325,16 +325,18 @@ lp_typekind_name(LLVMTypeKind t)
      return "LLVMArrayTypeKind";
   case LLVMPointerTypeKind:
      return "LLVMPointerTypeKind";
+#if HAVE_LLVM < 0x0300
   case LLVMOpaqueTypeKind:
      return "LLVMOpaqueTypeKind";
+#endif
   case LLVMVectorTypeKind:
      return "LLVMVectorTypeKind";
   case LLVMMetadataTypeKind:
      return "LLVMMetadataTypeKind";
-   /* Only in LLVM 2.7 and later???
+#if HAVE_LLVM == 0x0207
   case LLVMUnionTypeKind:
      return "LLVMUnionTypeKind";
-   */
+#endif
   default:
      return "unknown LLVMTypeKind";
   }
--- a/src/gallium/drivers/llvmpipe/lp_jit.c
+++ b/src/gallium/drivers/llvmpipe/lp_jit.c
@@ -68,10 +68,17 @@ lp_jit_create_types(struct llvmpipe_context *lp)
      elem_types[LP_JIT_TEXTURE_BORDER_COLOR] = 
         LLVMArrayType(LLVMFloatTypeInContext(lc), 4);

+#if HAVE_LLVM >= 0x0300
+   texture_type = LLVMStructCreateNamed(gallivm->context, "texture");
+   LLVMStructSetBody(texture_type, elem_types,
+                     Elements(elem_types), 0);
+#else
      texture_type = LLVMStructTypeInContext(lc, elem_types,
                                             Elements(elem_types), 0);
+      LLVMAddTypeName(gallivm->module, "texture", texture_type);

      LLVMInvalidateStructLayout(gallivm->target, texture_type);
+#endif

      LP_CHECK_MEMBER_OFFSET(struct lp_jit_texture, width,
                             gallivm->target, texture_type,
@@ -112,8 +119,6 @@ lp_jit_create_types(struct llvmpipe_context *lp)

      LP_CHECK_STRUCT_SIZE(struct lp_jit_texture,
                           gallivm->target, texture_type);
-
-      LLVMAddTypeName(gallivm->module, "texture", texture_type);
   }

   /* struct lp_jit_context */
@@ -129,11 +134,19 @@ lp_jit_create_types(struct llvmpipe_context *lp)
      elem_types[LP_JIT_CTX_TEXTURES] = LLVMArrayType(texture_type,
                                                      PIPE_MAX_SAMPLERS);

+#if HAVE_LLVM >= 0x0300
+   context_type = LLVMStructCreateNamed(gallivm->context, "context");
+   LLVMStructSetBody(context_type, elem_types,
+                     Elements(elem_types), 0);
+#else
      context_type = LLVMStructTypeInContext(lc, elem_types,
                                             Elements(elem_types), 0);

      LLVMInvalidateStructLayout(gallivm->target, context_type);

+      LLVMAddTypeName(gallivm->module, "context", context_type);
+#endif
+
      LP_CHECK_MEMBER_OFFSET(struct lp_jit_context, constants,
                             gallivm->target, context_type,
                             LP_JIT_CTX_CONSTANTS);
@@ -155,8 +168,6 @@ lp_jit_create_types(struct llvmpipe_context *lp)
      LP_CHECK_STRUCT_SIZE(struct lp_jit_context,
                           gallivm->target, context_type);

-      LLVMAddTypeName(gallivm->module, "context", context_type);
-
      lp->jit_context_ptr_type = LLVMPointerType(context_type, 0);
   }

--- a/src/gallium/drivers/nv50/nv50_context.c
+++ b/src/gallium/drivers/nv50/nv50_context.c
@@ -60,13 +60,13 @@ nv50_texture_barrier(struct pipe_context *pipe)
 void
 nv50_default_flush_notify(struct nouveau_channel *chan)
 {
-   struct nv50_context *nv50 = chan->user_private;
+   struct nv50_screen *screen = chan->user_private;

-   if (!nv50)
+   if (!screen)
      return;

-   nouveau_fence_update(&nv50->screen->base, TRUE);
-   nouveau_fence_next(&nv50->screen->base);
+   nouveau_fence_update(&screen->base, TRUE);
+   nouveau_fence_next(&screen->base);
 }

 static void
@@ -100,10 +100,8 @@ nv50_destroy(struct pipe_context *pipe)

   draw_destroy(nv50->draw);

-   if (nv50->screen->cur_ctx == nv50) {
-      nv50->screen->base.channel->user_private = NULL;
+   if (nv50->screen->cur_ctx == nv50)
      nv50->screen->cur_ctx = NULL;
-   }

   FREE(nv50);
 }
@@ -140,7 +138,6 @@ nv50_create(struct pipe_screen *pscreen, void *priv)

   if (!screen->cur_ctx)
      screen->cur_ctx = nv50;
-   screen->base.channel->user_private = nv50;
   screen->base.channel->flush_notify = nv50_default_flush_notify;

   nv50_init_query_functions(nv50);
--- a/src/gallium/drivers/nv50/nv50_screen.c
+++ b/src/gallium/drivers/nv50/nv50_screen.c
@@ -215,6 +215,8 @@ nv50_screen_destroy(struct pipe_screen *pscreen)
      nouveau_fence_wait(screen->base.fence.current);
      nouveau_fence_ref (NULL, &screen->base.fence.current);
   }
+   if (screen->base.channel)
+      screen->base.channel->user_private = NULL;

   nouveau_bo_ref(NULL, &screen->code);
   nouveau_bo_ref(NULL, &screen->tls_bo);
@@ -304,6 +306,7 @@ nv50_screen_create(struct pipe_winsys *ws, struct nouveau_device *dev)
      FAIL_SCREEN_INIT("nouveau_screen_init failed: %d\n", ret);

   chan = screen->base.channel;
+   chan->user_private = screen;

   pscreen->winsys = ws;
   pscreen->destroy = nv50_screen_destroy;
--- a/src/gallium/drivers/nv50/nv50_state_validate.c
+++ b/src/gallium/drivers/nv50/nv50_state_validate.c
@@ -282,8 +282,7 @@ nv50_switch_pipe_context(struct nv50_context *ctx_to)
   if (!ctx_to->zsa)
      ctx_to->dirty &= ~NV50_NEW_ZSA;

-   ctx_to->screen->base.channel->user_private = ctx_to->screen->cur_ctx =
-      ctx_to;
+   ctx_to->screen->cur_ctx = ctx_to;
 }

 static struct state_validate {
--- a/src/gallium/drivers/nv50/nv50_vbo.c
+++ b/src/gallium/drivers/nv50/nv50_vbo.c
@@ -389,11 +389,11 @@ nv50_prim_gl(unsigned prim)
 static void
 nv50_draw_vbo_flush_notify(struct nouveau_channel *chan)
 {
-   struct nv50_context *nv50 = chan->user_private;
+   struct nv50_screen *screen = chan->user_private;

-   nouveau_fence_update(&nv50->screen->base, TRUE);
+   nouveau_fence_update(&screen->base, TRUE);

-   nv50_bufctx_emit_relocs(nv50);
+   nv50_bufctx_emit_relocs(screen->cur_ctx);
 }

 static void
@@ -650,7 +650,6 @@ nv50_draw_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
   nv50_state_validate(nv50);

   chan->flush_notify = nv50_draw_vbo_flush_notify;
-   chan->user_private = nv50;

   if (nv50->vbo_fifo) {
      nv50_push_vbo(nv50, info);
--- a/src/gallium/drivers/nvc0/nvc0_context.c
+++ b/src/gallium/drivers/nvc0/nvc0_context.c
@@ -89,10 +89,8 @@ nvc0_destroy(struct pipe_context *pipe)

   draw_destroy(nvc0->draw);

-   if (nvc0->screen->cur_ctx == nvc0) {
-      nvc0->screen->base.channel->user_private = NULL;
+   if (nvc0->screen->cur_ctx == nvc0)
      nvc0->screen->cur_ctx = NULL;
-   }

   FREE(nvc0);
 }
@@ -100,13 +98,13 @@ nvc0_destroy(struct pipe_context *pipe)
 void
 nvc0_default_flush_notify(struct nouveau_channel *chan)
 {
-   struct nvc0_context *nvc0 = chan->user_private;
+   struct nvc0_screen *screen = chan->user_private;

-   if (!nvc0)
+   if (!screen)
      return;

-   nouveau_fence_update(&nvc0->screen->base, TRUE);
-   nouveau_fence_next(&nvc0->screen->base);
+   nouveau_fence_update(&screen->base, TRUE);
+   nouveau_fence_next(&screen->base);
 }

 struct pipe_context *
@@ -141,7 +139,6 @@ nvc0_create(struct pipe_screen *pscreen, void *priv)

   if (!screen->cur_ctx)
      screen->cur_ctx = nvc0;
-   screen->base.channel->user_private = nvc0;
   screen->base.channel->flush_notify = nvc0_default_flush_notify;

   nvc0_init_query_functions(nvc0);
--- a/src/gallium/drivers/nvc0/nvc0_screen.c
+++ b/src/gallium/drivers/nvc0/nvc0_screen.c
@@ -198,8 +198,12 @@ nvc0_screen_destroy(struct pipe_screen *pscreen)
 {
   struct nvc0_screen *screen = nvc0_screen(pscreen);

-   nouveau_fence_wait(screen->base.fence.current);
-   nouveau_fence_ref(NULL, &screen->base.fence.current);
+   if (screen->base.fence.current) {
+      nouveau_fence_wait(screen->base.fence.current);
+      nouveau_fence_ref(NULL, &screen->base.fence.current);
+   }
+   if (screen->base.channel)
+      screen->base.channel->user_private = NULL;

   nouveau_bo_ref(NULL, &screen->text);
   nouveau_bo_ref(NULL, &screen->tls);
@@ -362,6 +366,7 @@ nvc0_screen_create(struct pipe_winsys *ws, struct nouveau_device *dev)
      return NULL;
   }
   chan = screen->base.channel;
+   chan->user_private = screen;

   pscreen->winsys = ws;
   pscreen->destroy = nvc0_screen_destroy;
--- a/src/gallium/drivers/nvc0/nvc0_state_validate.c
+++ b/src/gallium/drivers/nvc0/nvc0_state_validate.c
@@ -428,8 +428,7 @@ nvc0_switch_pipe_context(struct nvc0_context *ctx_to)
   if (!ctx_to->zsa)
      ctx_to->dirty &= ~NVC0_NEW_ZSA;

-   ctx_to->screen->base.channel->user_private = ctx_to->screen->cur_ctx =
-      ctx_to;
+   ctx_to->screen->cur_ctx = ctx_to;
 }

 static struct state_validate {
--- a/src/gallium/drivers/nvc0/nvc0_vbo.c
+++ b/src/gallium/drivers/nvc0/nvc0_vbo.c
@@ -367,11 +367,11 @@ nvc0_prim_gl(unsigned prim)
 static void
 nvc0_draw_vbo_flush_notify(struct nouveau_channel *chan)
 {
-   struct nvc0_context *nvc0 = chan->user_private;
+   struct nvc0_screen *screen = chan->user_private;

-   nouveau_fence_update(&nvc0->screen->base, TRUE);
+   nouveau_fence_update(&screen->base, TRUE);

-   nvc0_bufctx_emit_relocs(nvc0);
+   nvc0_bufctx_emit_relocs(screen->cur_ctx);
 }

 static void
@@ -587,7 +587,6 @@ nvc0_draw_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
   nvc0_state_validate(nvc0);

   chan->flush_notify = nvc0_draw_vbo_flush_notify;
-   chan->user_private = nvc0;

   if (nvc0->vbo_fifo) {
      nvc0_push_vbo(nvc0, info);
--- a/src/glsl/ast.h
+++ b/src/glsl/ast.h
@@ -372,7 +372,7 @@ struct ast_type_qualifier {
    * \note
    * This field is only valid if \c explicit_location is set.
    */
-   unsigned location;
+   int location;

   /**
    * Return true if and only if an interpolation qualifier is present.
--- a/src/glsl/ast_to_hir.cpp
+++ b/src/glsl/ast_to_hir.cpp
@@ -1140,7 +1140,7 @@ ast_expression::hir(exec_list *instructions,
 	 error_emitted = true;
      }

-      type = op[0]->type;
+      type = error_emitted ? glsl_type::error_type : op[0]->type;
      result = new(ctx) ir_expression(ir_unop_bit_not, type, op[0], NULL);
      break;

--- a/src/glsl/lower_vec_index_to_swizzle.cpp
+++ b/src/glsl/lower_vec_index_to_swizzle.cpp
@@ -33,6 +33,7 @@
 #include "ir_visitor.h"
 #include "ir_optimization.h"
 #include "glsl_types.h"
+#include "main/macros.h"

 /**
 * Visitor class for replacing expressions with ir_constant values.
@@ -76,8 +77,25 @@ ir_vec_index_to_swizzle_visitor::convert_vec_index_to_swizzle(ir_rvalue *ir)

   void *ctx = ralloc_parent(ir);
   this->progress = true;
-   return new(ctx) ir_swizzle(deref->array,
-			      ir_constant->value.i[0], 0, 0, 0, 1);
+
+   /* Page 40 of the GLSL 1.20 spec says:
+    *
+    *     "When indexing with non-constant expressions, behavior is undefined
+    *     if the index is negative, or greater than or equal to the size of
+    *     the vector."
+    *
+    * The quoted spec text mentions non-constant expressions, but this code
+    * operates on constants.  These constants are the result of non-constant
+    * expressions that have been optimized to constants.  The common case here
+    * is a loop counter from an unrolled loop that is used to index a vector.
+    *
+    * The ir_swizzle constructor gets angry if the index is negative or too
+    * large.  For simplicity sake, just clamp the index to [0, size-1].
+    */
+   const int i = MIN2(MAX2(ir_constant->value.i[0], 0),
+		      (deref->array->type->vector_elements - 1));
+
+   return new(ctx) ir_swizzle(deref->array, i, 0, 0, 0, 1);
 }

 ir_visitor_status
--- a/src/glw/GLwDrawA.c
+++ b/src/glw/GLwDrawA.c
@@ -270,7 +270,7 @@ static XtResource backgroundResources[] = {
 static XtResource otherColorResources[] = {
  {XmNforeground,XmCForeground,XmRPixel, 
   sizeof(Pixel),XtOffset(GLwDrawingAreaWidget,primitive.foreground),
-   XmRString,(XtPointer)"lighgrey"},
+   XmRString,(XtPointer)"lightgrey"},
   /*XmRCallProc, (XtPointer) _XmForegroundColorDefault},*/

  {XmNhighlightColor,XmCHighlightColor,XmRPixel,sizeof(Pixel),
--- a/src/glx/apple/apple_glx_drawable.c
+++ b/src/glx/apple/apple_glx_drawable.c
@@ -135,6 +135,7 @@ release_drawable(struct apple_glx_drawable *d)
 static bool
 destroy_drawable(struct apple_glx_drawable *d)
 {
+   int err;

   d->lock(d);

@@ -172,6 +173,15 @@ destroy_drawable(struct apple_glx_drawable *d)

   apple_glx_diagnostic("%s: freeing %p\n", __func__, (void *) d);

+   /* Stupid recursive locks */
+   while (pthread_mutex_unlock(&d->mutex) == 0);
+
+   err = pthread_mutex_destroy(&d->mutex);
+   if (err) {
+      fprintf(stderr, "pthread_mutex_destroy error: %s\n", strerror(err));
+      abort();
+   }
+   
   free(d);

   /* So that the locks are balanced and the caller correctly unlocks. */
--- a/src/glx/apple/apple_glx_surface.c
+++ b/src/glx/apple/apple_glx_surface.c
@@ -206,6 +206,7 @@ apple_glx_surface_destroy(unsigned int uid)
   if (d) {
      d->types.surface.pending_destroy = true;
      d->release(d);
+
      /* 
       * We release 2 references to the surface.  One was acquired by
       * the find, and the other was leftover from a context, or 
@@ -216,8 +217,9 @@ apple_glx_surface_destroy(unsigned int uid)
       * to actually destroy it when the pending_destroy is processed
       * by a glViewport callback (see apple_glx_context_update()).
       */
-      d->destroy(d);
-
-      d->unlock(d);
+      if (!d->destroy(d)) {
+          /* apple_glx_drawable_find_by_uid returns a locked drawable */
+          d->unlock(d);
+      }
   }
 }
--- a/src/glx/apple/apple_visual.c
+++ b/src/glx/apple/apple_visual.c
@@ -116,7 +116,7 @@ apple_visual_create_pfobj(CGLPixelFormatObj * pfobj, const struct glx_config * m
      *uses_stereo = false;
   }

-   if (mode->doubleBufferMode) {
+   if (!offscreen && mode->doubleBufferMode) {
      attr[numattr++] = kCGLPFADoubleBuffer;
      *double_buffered = true;
   }
--- a/src/mapi/glapi/glapi_gentable.c
+++ b/src/mapi/glapi/glapi_gentable.c
@@ -77,7 +77,7 @@ __glapi_gentable_set_remaining_noop(struct _glapi_table *disp) {

 struct _glapi_table *
 _glapi_create_table_from_handle(void *handle, const char *symbol_prefix) {
-    struct _glapi_table *disp = calloc(1, sizeof(struct _glapi_table));
+    struct _glapi_table *disp = calloc(_glapi_get_dispatch_table_size(), sizeof(void *));
    char symboln[512];

    if(!disp)
--- a/src/mesa/drivers/common/meta.c
+++ b/src/mesa/drivers/common/meta.c
@@ -2842,6 +2842,16 @@ copy_tex_sub_image(struct gl_context *ctx,

   /* Choose format/type for temporary image buffer */
   format = _mesa_get_format_base_format(texImage->TexFormat);
+   if (format == GL_LUMINANCE ||
+       format == GL_LUMINANCE_ALPHA ||
+       format == GL_INTENSITY) {
+      /* We don't want to use GL_LUMINANCE, GL_INTENSITY, etc. for the
+       * temp image buffer because glReadPixels will do L=R+G+B which is
+       * not what we want (should be L=R).
+       */
+      format = GL_RGBA;
+   }
+
   type = get_temp_image_type(ctx, format);
   bpp = _mesa_bytes_per_pixel(format, type);
   if (bpp <= 0) {
--- a/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
+++ b/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
@@ -283,10 +283,10 @@ brw_create_constant_surface(struct brw_context *brw,

   surf[1] = bo->offset; /* reloc */

-   surf[2] = (((w & 0x7f) - 1) << BRW_SURFACE_WIDTH_SHIFT |
-	      (((w >> 7) & 0x1fff) - 1) << BRW_SURFACE_HEIGHT_SHIFT);
+   surf[2] = ((w & 0x7f) << BRW_SURFACE_WIDTH_SHIFT |
+	      ((w >> 7) & 0x1fff) << BRW_SURFACE_HEIGHT_SHIFT);

-   surf[3] = ((((w >> 20) & 0x7f) - 1) << BRW_SURFACE_DEPTH_SHIFT |
+   surf[3] = (((w >> 20) & 0x7f) << BRW_SURFACE_DEPTH_SHIFT |
 	      (width * 16 - 1) << BRW_SURFACE_PITCH_SHIFT);

   surf[4] = 0;
--- a/src/mesa/drivers/dri/i965/gen7_viewport_state.c
+++ b/src/mesa/drivers/dri/i965/gen7_viewport_state.c
@@ -36,7 +36,7 @@ prepare_sf_clip_viewport(struct brw_context *brw)
   const GLfloat *v = ctx->Viewport._WindowMap.m;
   struct gen7_sf_clip_viewport *vp;

-   vp = brw_state_batch(brw, sizeof(vp), 64, &brw->sf.vp_offset);
+   vp = brw_state_batch(brw, sizeof(*vp), 64, &brw->sf.vp_offset);
   /* Also assign to clip.vp_offset in case something uses it. */
   brw->clip.vp_offset = brw->sf.vp_offset;

--- a/src/mesa/drivers/dri/intel/intel_context.c
+++ b/src/mesa/drivers/dri/intel/intel_context.c
@@ -1334,27 +1334,28 @@ intel_process_dri2_buffer_with_separate_stencil(struct intel_context *intel,
 	      buffer->cpp, buffer->pitch);
   }

-   /*
-    * The stencil buffer has quirky pitch requirements.  From Section
-    * 2.11.5.6.2.1 3DSTATE_STENCIL_BUFFER, field "Surface Pitch":
-    *    The pitch must be set to 2x the value computed based on width, as
-    *    the stencil buffer is stored with two rows interleaved.
-    * If we neglect to double the pitch, then drm_intel_gem_bo_map_gtt()
-    * maps the memory incorrectly.
-    *
-    * To satisfy the pitch requirement, the X driver hackishly allocated
-    * the gem buffer with bpp doubled and height halved. So buffer->cpp is
-    * correct, but drawable->height is not.
-    */
-   int buffer_height = drawable->h;
+   int buffer_width;
+   int buffer_height;
   if (buffer->attachment == __DRI_BUFFER_STENCIL) {
-      buffer_height /= 2;
+      /* The stencil buffer has quirky pitch requirements.  From Section
+       * 2.11.5.6.2.1 3DSTATE_STENCIL_BUFFER, field "Surface Pitch":
+       *    The pitch must be set to 2x the value computed based on width, as
+       *    the stencil buffer is stored with two rows interleaved.
+       *
+       * To satisfy the pitch requirement, the X driver allocated the region
+       * with the following dimensions.
+       */
+       buffer_width = ALIGN(drawable->w, 64);
+       buffer_height = ALIGN(ALIGN(drawable->h, 2) / 2, 64);
+   } else {
+       buffer_width = drawable->w;
+       buffer_height = drawable->h;
   }

   struct intel_region *region =
      intel_region_alloc_for_handle(intel->intelScreen,
 				    buffer->cpp,
-				    drawable->w,
+				    buffer_width,
 				    buffer_height,
 				    buffer->pitch / buffer->cpp,
 				    buffer->name,
--- a/src/mesa/drivers/dri/intel/intel_screen.c
+++ b/src/mesa/drivers/dri/intel/intel_screen.c
@@ -108,13 +108,15 @@ intelDRI2Flush(__DRIdrawable *drawable)
   GET_CURRENT_CONTEXT(ctx);
   struct intel_context *intel = intel_context(ctx);

-   if (intel->gen < 4)
-      INTEL_FIREVERTICES(intel);
+   if (intel != NULL) {
+      if (intel->gen < 4)
+	 INTEL_FIREVERTICES(intel);

-   intel->need_throttle = GL_TRUE;
+      intel->need_throttle = GL_TRUE;

-   if (intel->batch.used)
-      intel_batchbuffer_flush(intel);
+      if (intel->batch.used)
+	 intel_batchbuffer_flush(intel);
+   }
 }

 static const struct __DRI2flushExtensionRec intelFlushExtension = {
@@ -785,6 +787,54 @@ struct intel_buffer {
   struct intel_region *region;
 };

+/**
+ * \brief Get tiling format for a DRI buffer.
+ *
+ * \param attachment is the buffer's attachmet point, such as
+ *        __DRI_BUFFER_DEPTH.
+ * \param out_tiling is the returned tiling format for buffer.
+ * \return false if attachment is unrecognized or is incompatible with screen.
+ */
+static bool
+intel_get_dri_buffer_tiling(struct intel_screen *screen,
+                            uint32_t attachment,
+                            uint32_t *out_tiling)
+{
+   if (screen->gen < 4) {
+      *out_tiling = I915_TILING_X;
+      return true;
+   }
+
+   switch (attachment) {
+   case __DRI_BUFFER_DEPTH:
+   case __DRI_BUFFER_DEPTH_STENCIL:
+   case __DRI_BUFFER_HIZ:
+      *out_tiling = I915_TILING_Y;
+      return true;
+   case __DRI_BUFFER_ACCUM:
+   case __DRI_BUFFER_FRONT_LEFT:
+   case __DRI_BUFFER_FRONT_RIGHT:
+   case __DRI_BUFFER_BACK_LEFT:
+   case __DRI_BUFFER_BACK_RIGHT:
+   case __DRI_BUFFER_FAKE_FRONT_LEFT:
+   case __DRI_BUFFER_FAKE_FRONT_RIGHT:
+      *out_tiling = I915_TILING_X;
+      return true;
+   case __DRI_BUFFER_STENCIL:
+      /* The stencil buffer is W tiled. However, we request from the kernel
+       * a non-tiled buffer because the GTT is incapable of W fencing.
+       */
+      *out_tiling = I915_TILING_NONE;
+      return true;
+   default:
+      if(unlikely(INTEL_DEBUG & DEBUG_DRI)) {
+	 fprintf(stderr, "error: %s: unrecognized DRI buffer attachment 0x%x\n",
+	         __FUNCTION__, attachment);
+      }
+       return false;
+   }
+}
+
 static __DRIbuffer *
 intelAllocateBuffer(__DRIscreen *screen,
 		    unsigned attachment, unsigned format,
@@ -792,22 +842,45 @@ intelAllocateBuffer(__DRIscreen *screen,
 {
   struct intel_buffer *intelBuffer;
   struct intel_screen *intelScreen = screen->private;
+
   uint32_t tiling;
+   uint32_t region_width;
+   uint32_t region_height;
+   uint32_t region_cpp;
+
+   bool ok = true;
+
+   ok = intel_get_dri_buffer_tiling(intelScreen, attachment, &tiling);
+   if (!ok)
+      return NULL;

   intelBuffer = CALLOC(sizeof *intelBuffer);
   if (intelBuffer == NULL)
      return NULL;

-   if ((attachment == __DRI_BUFFER_DEPTH ||
-	attachment == __DRI_BUFFER_STENCIL ||
-	attachment == __DRI_BUFFER_DEPTH_STENCIL) &&
-       intelScreen->gen >= 4)
-      tiling = I915_TILING_Y;
-   else
-      tiling = I915_TILING_X;
+   if (attachment == __DRI_BUFFER_STENCIL) {
+      /* The stencil buffer has quirky pitch requirements.  From Vol 2a,
+       * 11.5.6.2.1 3DSTATE_STENCIL_BUFFER, field "Surface Pitch":
+       *    The pitch must be set to 2x the value computed based on width, as
+       *    the stencil buffer is stored with two rows interleaved.
+       * To accomplish this, we resort to the nasty hack of doubling the
+       * region's cpp and halving its height.
+       */
+      region_width = ALIGN(width, 64);
+      region_height = ALIGN(ALIGN(height, 2) / 2, 64);
+      region_cpp = format / 4;
+   } else {
+      region_width = width;
+      region_height = height;
+      region_cpp = format / 8;
+   }

-   intelBuffer->region = intel_region_alloc(intelScreen, tiling,
-					    format / 8, width, height, GL_TRUE);
+   intelBuffer->region = intel_region_alloc(intelScreen,
+                                            tiling,
+                                            region_cpp,
+                                            region_width,
+                                            region_height,
+                                            true);
   
   if (intelBuffer->region == NULL) {
 	   FREE(intelBuffer);
--- a/src/mesa/drivers/dri/intel/intel_span.c
+++ b/src/mesa/drivers/dri/intel/intel_span.c
@@ -141,10 +141,9 @@ intel_set_span_functions(struct intel_context *intel,
   struct intel_renderbuffer *irb = intel_renderbuffer(rb);		\
   uint8_t *buf = irb->region->buffer->virtual;				\
   unsigned stride = irb->region->pitch;				\
-   unsigned height = 2 * irb->region->height;				\
   bool flip = rb->Name == 0;						\
   int y_scale = flip ? -1 : 1;						\
-   int y_bias = flip ? (height - 1) : 0;				\
+   int y_bias = flip ? (rb->Height - 1) : 0;				\

 #undef Y_FLIP
 #define Y_FLIP(y) (y_scale * (y) + y_bias)
--- a/src/mesa/main/dlist.c
+++ b/src/mesa/main/dlist.c
@@ -4529,7 +4529,7 @@ static GLvoid *copy_data(const GLvoid *data, GLsizei size, const char *func)

   image = malloc(size);
   if (!image) {
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, func);
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "%s", func);
      return NULL;
   }
   memcpy(image, data, size);
--- a/src/mesa/main/texgetimage.c
+++ b/src/mesa/main/texgetimage.c
@@ -713,6 +713,9 @@ _mesa_GetnTexImageARB( GLenum target, GLint level, GLenum format,
   texObj = _mesa_get_current_tex_object(ctx, target);
   texImage = _mesa_select_tex_image(ctx, texObj, target, level);

+   if (_mesa_is_zero_size_texture(texImage))
+      return;
+
   if (MESA_VERBOSE & (VERBOSE_API | VERBOSE_TEXTURE)) {
      _mesa_debug(ctx, "glGetTexImage(tex %u) format = %s, w=%d, h=%d,"
                  " dstFmt=0x%x, dstType=0x%x\n",
@@ -846,6 +849,9 @@ _mesa_GetnCompressedTexImageARB(GLenum target, GLint level, GLsizei bufSize,
   texObj = _mesa_get_current_tex_object(ctx, target);
   texImage = _mesa_select_tex_image(ctx, texObj, target, level);

+   if (_mesa_is_zero_size_texture(texImage))
+      return;
+
   if (MESA_VERBOSE & (VERBOSE_API | VERBOSE_TEXTURE)) {
      _mesa_debug(ctx,
                  "glGetCompressedTexImage(tex %u) format = %s, w=%d, h=%d\n",
--- a/src/mesa/main/teximage.c
+++ b/src/mesa/main/teximage.c
@@ -2816,29 +2816,43 @@ copyteximage(struct gl_context *ctx, GLuint dims,
 	 _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyTexImage%uD", dims);
      }
      else {
-         gl_format texFormat;
-
-         if (texImage->Data) {
-            ctx->Driver.FreeTexImageData( ctx, texImage );
-         }
-
-         ASSERT(texImage->Data == NULL);
-
-         texFormat = _mesa_choose_texture_format(ctx, texObj, target, level,
-                                                 internalFormat, GL_NONE,
-                                                 GL_NONE);
+         /* choose actual hw format */
+         gl_format texFormat = _mesa_choose_texture_format(ctx, texObj,
+                                                           target, level,
+                                                           internalFormat,
+                                                           GL_NONE, GL_NONE);

         if (legal_texture_size(ctx, texFormat, width, height, 1)) {
+            GLint srcX = x, srcY = y, dstX = 0, dstY = 0;
+
+            /* Free old texture image */
+            ctx->Driver.FreeTexImageData(ctx, texImage);
+
            _mesa_init_teximage_fields(ctx, target, texImage, width, height, 1,
                                       border, internalFormat, texFormat);

-            ASSERT(ctx->Driver.CopyTexImage2D);
-            if (dims == 1)
-               ctx->Driver.CopyTexImage1D(ctx, target, level, internalFormat,
-                                          x, y, width, border);
-            else
-               ctx->Driver.CopyTexImage2D(ctx, target, level, internalFormat,
-                                          x, y, width, height, border);
+            /* Allocate texture memory (no pixel data yet) */
+            if (dims == 1) {
+               ctx->Driver.TexImage1D(ctx, target, level, internalFormat,
+                                      width, border, GL_NONE, GL_NONE, NULL,
+                                      &ctx->Unpack, texObj, texImage);
+            }
+            else {
+               ctx->Driver.TexImage2D(ctx, target, level, internalFormat,
+                                      width, height, border, GL_NONE, GL_NONE,
+                                      NULL, &ctx->Unpack, texObj, texImage);
+            }
+
+            if (_mesa_clip_copytexsubimage(ctx, &dstX, &dstY, &srcX, &srcY,
+                                           &width, &height)) {
+               if (dims == 1)
+                  ctx->Driver.CopyTexSubImage1D(ctx, target, level, dstX,
+                                                srcX, srcY, width);
+                                                
+               else
+                  ctx->Driver.CopyTexSubImage2D(ctx, target, level, dstX, dstY,
+                                                srcX, srcY, width, height);
+            }

            check_gen_mipmap(ctx, target, texObj, level);

@@ -2849,6 +2863,7 @@ copyteximage(struct gl_context *ctx, GLuint dims,
            ctx->NewState |= _NEW_TEXTURE;
         }
         else {
+            /* probably too large of image */
            _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyTexImage%uD", dims);
         }
      }
--- a/src/mesa/main/teximage.h
+++ b/src/mesa/main/teximage.h
@@ -42,6 +42,14 @@ _mesa_alloc_texmemory(GLsizei bytes);
 extern void
 _mesa_free_texmemory(void *m);

+/** Is any of the dimensions of given texture equal to zero? */
+static inline GLboolean
+_mesa_is_zero_size_texture(const struct gl_texture_image *texImage)
+{
+   return (texImage->Width == 0 ||
+           texImage->Height == 0 ||
+           texImage->Depth == 0);
+}

 /** \name Internal functions */
 /*@{*/
--- a/src/mesa/main/version.h
+++ b/src/mesa/main/version.h
@@ -34,8 +34,8 @@ struct gl_context;
 /* Mesa version */
 #define MESA_MAJOR 7
 #define MESA_MINOR 11
-#define MESA_PATCH 1
-#define MESA_VERSION_STRING "7.11.1"
+#define MESA_PATCH 2
+#define MESA_VERSION_STRING "7.11.2"

 /* To make version comparison easy */
 #define MESA_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + (c))
Author	SHA1	Message	Date
Jonas Maebe	1bdf31e8da	darwin: do not create double-buffered offscreen pixel formats http://xquartz.macosforge.org/trac/ticket/536 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com> (cherry picked from commit `5fdf1f784b`)	2012-09-24 16:07:14 -07:00
Jeremy Huddleston	ecf0bef4cd	darwin: Eliminate a possible race condition while destroying a surface Introduced by: `c60ffd2840` Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> (cherry picked from commit `d65bd195ec`)	2012-05-16 18:59:30 -07:00
Jeremy Huddleston	60fa18b90f	darwin: Unlock our mutex before destroying it http://xquartz.macosforge.org/trac/ticket/575 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> (cherry picked from commit `a73a800b32`)	2012-05-16 18:59:28 -07:00
Jeremy Huddleston	62058968e7	darwin: Eliminate a pthread mutex leak Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> Tested-by: Charles Davis <cdavis@mines.edu> (cherry picked from commit `1a33c1b2b8`)	2012-04-24 00:27:44 -07:00
Jonas Maebe	1fa6c87c88	apple: Fix a use after free Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> (cherry picked from commit `c60ffd2840`)	2012-04-23 16:18:36 -07:00
Jonas Maebe	00d310a130	glapi: Correct size of allocated _glapi_table struct The __glapi_gentable_set_remaining_noop() routine treats the _glapi_struct as an array of _glapi_get_dispatch_table_size() pointers, so we have to allocate _glapi_get_dispatch_table_size()sizeof(void) bytes rather than sizeof(struct _glapi_struct) bytes. Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> (cherry picked from commit `8d09f4d0cc`)	2012-04-22 20:41:36 -07:00
Jeremy Huddleston	906f670f1a	configure.ac: Don't use $CLANG since it will collide with the static analyzer. We just prefix the $CLANG environment variable in configure.ac with acv_mesa_ Found by: tinderbox Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> (cherry picked from commit `b728eefb06`)	2012-01-27 09:05:10 -08:00
Anuj Phogat	9ae2499be2	mesa: Add condition in glGetTexImage for zero size textures TestMipMaps() function in src/OGLconform/textureNPOT.c calls glTexImage2D() with width = 0. Texture with zero size skips miptree allocation due to a condition in function _mesa_store_teximage3d(). While calling glGetTexImage() it results in assertion failure in intel_map_texture_image() due to null mt pointer. This patch fixes the issue by detecting the zero size texture early in glGetTexImage and glGetCompressedTexImage functions. In such a case function simply returns doing nothing. Verified that below mentioned bug is fixed by this patch. https://bugs.freedesktop.org/show_bug.cgi?id=42334 NOTE: This is a candidate for stable branches Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Reviewed-by: Brian Paul <brianp@vmware.com> (cherry picked from commit `f1a9a9bcd1`) Conflicts: src/mesa/main/teximage.h	2012-01-19 12:56:15 -08:00
Anuj Phogat	9b3ac17991	intel: Fix segfault in glXSwapBuffers with no bound context Calling glXSwapBuffers with no bound context causes segmentation fault in function intelDRI2Flush. All the gl calls should be ignored after setting the current context to null. So the contents of framebuffer stay unchanged. But the driver should not seg fault. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=44614 Reported-by: Yi Sun <yi.sun@intel.com> Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Tested-by: Yi Sun <yi.sun@intel.com> (cherry picked from commit `dd7220652e`) Conflicts: src/mesa/drivers/dri/intel/intel_screen.c	2012-01-19 12:56:15 -08:00
Vinson Lee	d7d47d6c0b	i965: Fix Coverity wrong sizeof argument defect. NOTE: This is a candidate for stable release branches. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=42542 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (cherry picked from commit `3e18ad7fd7`)	2012-01-19 12:56:15 -08:00
Paul Berry	29f07f6461	i965 gen4-6: Fix off-by-one errors brw_create_constant_surface() Commit `9bdc44a528` (i965: Replace struct with bit shifting for WM pull constant surfaces) accidentally introduced off-by-one errors into the calculation of the surface width, height, and depth. This patch restores the correct computation. The reason this wasn't noticed by Piglit tests is that the size of our constant surfaces is always less than 2^20, therefore the off-by-one error was causing the "depth" field of the surface to be set to all 1's. The hardware interpreted this as an extremely large surface, so overflow checking was effectively disabled. No Piglit regressions on Sandy Bridge. NOTE: This is a candidate for the 7.11 and 8.0 branches. Reviewed-by: Eric Anholt <eric@anholt.net> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (cherry picked from commit `f6f43bd5a2`)	2012-01-19 12:56:15 -08:00
Brian Paul	1f18137d28	glw: s/lighgrey/lightgrey/ http://bugs.freedesktop.org/show_bug.cgi?id=44844	2012-01-16 15:07:23 -07:00
Ian Romanick	7722a54e60	glsl: Don't use base type for bit-not when there's an error Other parts of the compiler assume that expressions will have well-formed types or the error type. Just using the type of the thing being operated on can cause expressions like ~3.14 or ~false to not have a well-formed type. This could then result in an assertion failure in the context epxression handler. If there is an error processing the expression, set the type of the IR expression to error. Fixes piglit's bit-not-0[789].frag tests. NOTE: This is a candidate for the 7.11 branch. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=42755 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Cc: Vinson Lee <vlee@vmware.com> (cherry picked from commit `3946448951`)	2012-01-04 12:44:49 -08:00
Ian Romanick	bc507471db	glsl: Clamp vector indices when lowering to swizzles This prevents other code from seeing a swizzle of the 16th component of a vector, for example. NOTE: This is a candidate for the 7.11 branch. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=42517 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Paul Berry <stereotype441@gmail.com> Tested-by: Christian Holler <choller@mozilla.com> (cherry picked from commit `6f5c737970`)	2012-01-03 19:23:15 -08:00
Ian Romanick	42d1a7b3de	glsl: Silence "ast_to_hir.cpp:1984:25: warning: comparison of unsigned expression >= 0 is always true" ast_type_qualifier::location should have been a signed integer from the beginning, and the giant comment in apply_type_qualifier_to_variable explains why. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=40207 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (cherry picked from commit `f0dd21ac88`)	2011-12-22 17:01:53 -08:00
Marcin Slusarz	237d6fa97a	nv50/nvc0: fix crash when channel allocation fails	2011-12-11 17:00:42 +01:00
Christoph Bumiller	0e9b79c22a	nv50,nvc0: use screen instead of context for flush notifier Context may become NULL and we still have to be able to flush pending fences. (Cherry picked from commit `3069a7eaa5`) Fixes crash on screen shutdown: https://bugs.freedesktop.org/show_bug.cgi?id=43428	2011-12-11 16:58:09 +01:00
Brian Paul	7a18f005ed	gallivm: fix build with llvm 3.0svn https://bugs.freedesktop.org/show_bug.cgi?id=41065 (cherry picked from commit `51002968c9`)	2011-12-05 08:58:18 -07:00
Tobias Droste	95a3724c69	gallivm: fix build with LLVM 3.0svn LLVM 3.0svn added SubtargetInfo as additional parameter to createMCDisassembler() and createMCInstPrinter(). See revision 139237 of LLVM. Signed-off-by: Tobias Droste <tdroste@gmx.de> Signed-off-by: Brian Paul <brianp@vmware.com> (cherry picked from commit `1795372fee`)	2011-12-05 08:58:12 -07:00
Tobias Droste	98e8fb84de	gallivm: fix build with LLVM 3.0svn LLVM 3.0svn moved TargetRegistry.h and TargetSelect.h. See revision 138450 of LLVM. Signed-off-by: Tobias Droste <tdroste@gmx.de> (cherry picked from commit `4a468de2d7`)	2011-12-05 08:58:06 -07:00
Tobias Droste	5b5c436c65	llvmpipe: fix build with LLVM 3.0svn LLVM 3.0svn introduced a new type system. It defines a new way to create named structs and removes the (now not needed) LLVMInvalidateStructLayout function. See revision 134829 of LLVM. Signed-off-by: Tobias Droste <tdroste@gmx.de> Signed-off-by: Brian Paul <brianp@vmware.com> (cherry picked from commit `3143e95353`)	2011-12-05 08:58:00 -07:00
Vinson Lee	970df845ef	gallivm: Rename createAsmInfo to createMCAsmInfo with llvm-3.0. llvm-3.0svn r135219 renamed createAsmInfo to createMCAsmInfo in include/llvm/Target/TargetRegistry.h. (cherry picked from commit `9228bfb375`)	2011-12-05 08:57:49 -07:00
Vinson Lee	32f4a83fe8	gallivm: Re-enable LLVMUnionTypeKind case for llvm-2.7 only. LLVMUnionTypeKind is not in llvm-2.6, llvm-2.8, llvm-2.9, or llvm-3.0svn. (cherry picked from commit `1844ae7e7e`)	2011-12-05 08:57:42 -07:00
Vinson Lee	2c1998b3b9	gallivm: Remove LLVMOpaqueKindType case with llvm-3.0. llvm-3.0svn r134829 removed LLVMOpaqueKindType from enum LLVMTypeKind in include/llvm-c/Core.h. (cherry picked from commit `e4189f2e2e`)	2011-12-05 08:57:11 -07:00
Vinson Lee	a3337a66ef	gallivm: Rename TargetInstrDesc to MCInstrDesc when using llvm-3.0. llvm-3.0svn revision 134021 renamed TargetInstrDesc to MCInstrDesc. (cherry picked from commit `b61e56756c`)	2011-12-05 08:57:00 -07:00
Brian Paul	25ba654229	docs: update news, relnotes list for 7.11.2 release	2011-11-28 08:24:19 -07:00
Ian Romanick	733478975d	docs: Add 7.11.2 release md5sums	2011-11-27 14:06:59 -08:00
Ian Romanick	ea72e3cda8	docs: Add 7.11.2 release notes	2011-11-27 14:04:48 -08:00
Ian Romanick	835205e295	mesa: Bump version to 7.11.2 (final)	2011-11-27 14:04:05 -08:00
Chad Versace	2a45566587	intel: Fix separate stencil in builtin DRI2 backend intelAllocateBuffer() was oblivious to separate stencil buffers. This patch fixes it to allocate a non-tiled stencil buffer with special pitch, just as the DDX does. Without this, any app that attempted to create an EGL surface with stencil bits would crash. Of course, this affected only environments that used the builtin DRI2 backend, such as Android and Wayland. Fixes GLBenchmark2.1 on Android on gen7. Note: This is a candidate for the 7.11 branch. Tested-by: Louie Tsaie <louie.tsai@intel.com> Signed-off-by: Chad Versace <chad.versace@linux.intel.com> (cherry picked from commit `79653c12d6`) Conflicts: src/mesa/drivers/dri/intel/intel_screen.c	2011-11-21 16:11:50 -08:00
Chad Versace	68060d6bfd	intel: Fix region dimensions for stencil buffers received from DDX I changed the dimensions of the stencil buffer's region, as allocated by the DDX, at xf86-video-intel commit commit 3e55f3e88b40471706d5cd45c4df4010f8675c75 dri: Do not tile stencil buffer But I forgot to make the analogous update to the Intel DRI2 glue in Mesa. This patch makes that update. Surprisingly, the mismatch did not cause any bugs. But the mismatch, if left unfixed, would create bugs in the next commit. Note: This is a candidate for the 7.11 branch. Signed-off-by: Chad Versace <chad.versace@linux.intel.com> (cherry picked from commit `50b3356078`)	2011-11-21 16:10:49 -08:00
Chad Versace	fd35fab340	intel: Simplify stencil detiling arithmetic When calculating the y offset needed for detiling window system stencil buffers, replace the term region->height * 2 + region->height % 2 - 1 with rb->Height - 1 . The two terms are incidentally equivalent due to some out-of-date, incorrect code in the Intel DRI2 glue for DDX. (See intel_process_dri2_buffer_with_separate_stencil(), line ``buffer_height /= 2;``). Note: This is a candidate for the 7.11 branch (only the intel_span.c hunk). Signed-off-by: Chad Versace <chad.versace@linux.intel.com> (cherry picked from commit `dc4c3a31c6`) Conflicts: src/mesa/drivers/dri/intel/intel_fbo.c src/mesa/drivers/dri/intel/intel_span.c	2011-11-21 16:10:22 -08:00
Brian Paul	658d994473	mesa: use format string in _mesa_error() call to silence warning (cherry picked from commit `5485192fc8`)	2011-11-21 19:49:48 +01:00
Brian Paul	8a8ccc00b5	docs: update news.html and relnotes.html for 7.11.1 release	2011-11-21 09:16:38 -07:00
Brian Paul	21364bd09d	mesa: fix format selection for meta CopyTexSubImage() When we do a glReadPixels into the temporary buffer, we don't want to use GL_LUMINANCE, GL_LUMINANCE_ALPHA or GL_INTENSITY since they will compute L=R+G+B which is not what we want. This bug has existed all along but was only exposed by the elimination of the driver hook for glCopyTexImage() in `5874890c26`. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=39604 Tested-by: Ian Romanick <ian.d.romanick@intel.com> (cherry picked from commit `e4fdc95277`) (cherry-pick note: This patch and the one before it combine to fix CopyTexImage from RGBA window to LA texture. This was believed to already be fixed by commit `296e6b9`, but these changes were also necessary.)	2011-11-18 12:18:19 -08:00
Brian Paul	6b2c771ddd	mesa: stop using ctx->Driver.CopyTexImage1D/2D() hooks (cherry picked from commit `5874890c26`)	2011-11-18 12:12:22 -08:00
Ian Romanick	22ff4ec969	mesa: set version string to 7.11.2-devel	2011-11-18 11:06:33 -08:00
Ian Romanick	228da884c9	docs: Add 7.11.1 release md5sums	2011-11-17 12:12:34 -08:00