docs: Update 10.5.0 release notes

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

Makefile.am

@@ -21,7 +21,7 @@
 
 SUBDIRS = src
 
-DISTCHECK_CONFIGURE_FLAGS = \
+AM_DISTCHECK_CONFIGURE_FLAGS = \
 	--enable-dri3 \
 	--enable-gallium-tests \
 	--enable-gbm \

VERSION

@@ -1 +1 @@
-10.5.0-devel
+10.5.0

bin/.cherry-ignore

@@ -0,0 +1,12 @@
+# Cherry-picked without -x
+# nir: resolve nir.h dependency list (fix make distcheck)
+556fc4b84df99a1cd4b18c11fb16f7854a948b2a 
+
+# nir: add missing header to the sources list
+72e602905dd9d86450a936d5a22bf21758844b38
+
+# configure: rework wayland_scanner handling(fix make distcheck)
+153539bd9d4445b504110958306f00632222f840
+
+# auxiliary/vl: bring back the VL code for the dri targets
+c39dbfdd0f764b1aaa7319b4694e7335692993dd

configure.ac

@@ -1572,14 +1572,23 @@ if test -n "$with_gallium_drivers" -a "x$with_gallium_drivers" != xswrast; then
     fi
 fi
 
+if test "x$enable_dri" = xyes -o \
+        "x$enable_xvmc" = xyes -o \
+        "x$enable_vdpau" = xyes -o \
+        "x$enable_omx" = xyes -o \
+        "x$enable_va" = xyes; then
+    need_gallium_vl=yes
+fi
+AM_CONDITIONAL(NEED_GALLIUM_VL, test "x$need_gallium_vl" = xyes)
+
 if test "x$enable_xvmc" = xyes -o \
         "x$enable_vdpau" = xyes -o \
         "x$enable_omx" = xyes -o \
         "x$enable_va" = xyes; then
     PKG_CHECK_MODULES([VL], [x11-xcb xcb xcb-dri2 >= $XCBDRI2_REQUIRED])
-    need_gallium_vl=yes
+    need_gallium_vl_winsys=yes
 fi
-AM_CONDITIONAL(NEED_GALLIUM_VL, test "x$need_gallium_vl" = xyes)
+AM_CONDITIONAL(NEED_GALLIUM_VL_WINSYS, test "x$need_gallium_vl_winsys" = xyes)
 
 if test "x$enable_xvmc" = xyes; then
     PKG_CHECK_MODULES([XVMC], [xvmc >= $XVMC_REQUIRED])
@@ -1730,6 +1739,13 @@ if test "x$with_egl_platforms" != "x" -a "x$enable_egl" != xyes; then
     AC_MSG_ERROR([cannot build egl state tracker without EGL library])
 fi
 
+PKG_CHECK_MODULES([WAYLAND_SCANNER], [wayland_scanner],
+        WAYLAND_SCANNER=`$PKG_CONFIG --variable=wayland_scanner wayland_scanner`,
+        WAYLAND_SCANNER='')
+if test "x$WAYLAND_SCANNER" = x; then
+    AC_PATH_PROG([WAYLAND_SCANNER], [wayland-scanner])
+fi
+
 # Do per-EGL platform setups and checks
 egl_platforms=`IFS=', '; echo $with_egl_platforms`
 for plat in $egl_platforms; do
@@ -1737,9 +1753,9 @@ for plat in $egl_platforms; do
 	wayland)
 		PKG_CHECK_MODULES([WAYLAND], [wayland-client >= $WAYLAND_REQUIRED wayland-server >= $WAYLAND_REQUIRED])
 
-                WAYLAND_PREFIX=`$PKG_CONFIG --variable=prefix wayland-client`
-                AC_PATH_PROG([WAYLAND_SCANNER], [wayland-scanner],,
-                             [${WAYLAND_PREFIX}/bin$PATH_SEPARATOR$PATH])
+		if test "x$WAYLAND_SCANNER" = x; then
+			AC_MSG_ERROR([wayland-scanner is needed to compile the wayland egl platform])
+		fi
 		;;
 
 	x11)
@@ -1770,7 +1786,7 @@ done
 # libEGL wants to default to the first platform specified in
 # ./configure.  parse that here.
 if test "x$egl_platforms" != "x"; then
-    FIRST_PLATFORM_CAPS=`echo $egl_platforms | sed 's| .*||' | tr 'a-z' 'A-Z'`
+    FIRST_PLATFORM_CAPS=`echo $egl_platforms | sed 's| .*||' | tr '[[a-z]]' '[[A-Z]]'`
     EGL_NATIVE_PLATFORM="_EGL_PLATFORM_$FIRST_PLATFORM_CAPS"
 else
     EGL_NATIVE_PLATFORM="_EGL_INVALID_PLATFORM"

docs/relnotes/10.5.0.html

@@ -14,7 +14,7 @@
 <iframe src="../contents.html"></iframe>
 <div class="content">
 
-<h1>Mesa 10.5.0 Release Notes / TBD</h1>
+<h1>Mesa 10.5.0 Release Notes / March 06, 2015</h1>
 
 <p>
 Mesa 10.5.0 is a new development release.
@@ -55,7 +55,150 @@ Note: some of the new features are only available with certain drivers.
 
 <h2>Bug fixes</h2>
 
-TBD.
+<p>This list is likely incomplete.</p>
+
+<ul>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=10370">Bug 10370</a> - Incorrect pixels read back if draw bitmap texture through Display list</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=45348">Bug 45348</a> - [swrast] piglit fbo-drawbuffers-arbfp regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=60879">Bug 60879</a> - [radeonsi] X11 can't start with acceleration enabled</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=67672">Bug 67672</a> - [llvmpipe] lp_test_arit fails on old CPUs</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=77544">Bug 77544</a> - i965: Try to use LINE instructions to perform MAD with immediate arguments</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=78770">Bug 78770</a> - [SNB bisected]Webglc conformance/textures/texture-size-limit.html fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=80568">Bug 80568</a> - [gen4] GPU Crash During Google Chrome Operation</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=82477">Bug 82477</a> - [softpipe] piglit fp-long-alu regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=82585">Bug 82585</a> - geometry shader with optional out variable segfaults</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=82991">Bug 82991</a> - Inverted bumpmap in webgl applications</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=83463">Bug 83463</a> - [swrast] piglit glsl-vs-clamp-1 regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=83500">Bug 83500</a> - si_dma_copy_tile causes GPU hangs</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=83510">Bug 83510</a> - Graphical glitches in Unreal Engine 4</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=83908">Bug 83908</a> - [i965] Incorrect icon colors in Steam Big Picture</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84212">Bug 84212</a> - [BSW]ES3-CTS.shaders.loops.do_while_dynamic_iterations.vector_counter_vertex fails and causes GPU hang</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84651">Bug 84651</a> - Distorted graphics or black window when running Battle.net app on Intel hardware via wine</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84777">Bug 84777</a> - [BSW]Piglit spec_glsl-1.50_execution_geometry-basic fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85367">Bug 85367</a> - [gen4] GPU hang in glmark-es2</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85467">Bug 85467</a> - [llvmpipe] piglit gl-1.0-dlist-beginend failure with llvm-3.6.0svn</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85529">Bug 85529</a> - Surfaces not drawn in Unvanquished</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85647">Bug 85647</a> - Random radeonsi crashes with mesa 10.3.x</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85696">Bug 85696</a> - r600g+nine: Bioshock shader failure after 7b1c0cbc90d456384b0950ad21faa3c61a6b43ff</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86089">Bug 86089</a> - [r600g][mesa 10.4.0-dev] shader failure - r600_sb::bc_finalizer::cf_peephole() when starting Second Life</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86618">Bug 86618</a> - [NV96] neg modifiers not working in MIN and MAX operations</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86760">Bug 86760</a> - mesa doesn't build: recipe for target 'r600_llvm.lo' failed</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86764">Bug 86764</a> - [SNB+ Bisected]Piglit glean/pointSprite fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86788">Bug 86788</a> - (bisected) 32bit UrbanTerror 4.1 timedemo sse4.1 segfault...</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86811">Bug 86811</a> - [BDW/BSW Bisected]Piglit spec_arb_shading_language_packing_execution_built-in-functions_vs-unpackSnorm4x8 fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86837">Bug 86837</a> - kodi segfault since auxiliary/vl: rework the build of the VL code</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86939">Bug 86939</a> - test_vf_float_conversions.cpp:63:12: error: expected primary-expression before ‘union’</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86944">Bug 86944</a> - glsl_parser_extras.cpp&quot;, line 1455: Error: Badly formed expression. (Oracle Studio)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86958">Bug 86958</a> - lp_bld_misc.cpp:503:40: error: no matching function for call to ‘llvm::EngineBuilder::setMCJITMemoryManager(ShaderMemoryManager*&amp;)’</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86969">Bug 86969</a> - _drm_intel_gem_bo_references() function takes half the CPU with Witcher2 game</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87076">Bug 87076</a> - Dead Island needs allow_glsl_extension_directive_midshader</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87516">Bug 87516</a> - glProgramBinary violates spec</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87619">Bug 87619</a> - Changes to state such as render targets change fragment shader without marking it dirty.</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87658">Bug 87658</a> - [llvmpipe] SEGV in sse2_has_daz on ancient Pentium4-M</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87694">Bug 87694</a> - [SNB] Crash in brw_begin_transform_feedback</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87886">Bug 87886</a> - constant fps drops with Intel and Radeon</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87887">Bug 87887</a> - [i965 Bisected]ES2-CTS.gtf.GL.cos.cos_float_vert_xvary fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=87913">Bug 87913</a> - CPU cacheline size of 0 can be returned by CPUID leaf 0x80000006 in some virtual machines</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88079">Bug 88079</a> - dEQP-GLES3.functional.fbo.completeness.renderable.renderbuffer.color0 tests fail due to enabling of GL_RGB and GL_RGBA</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88170">Bug 88170</a> - 32 bits opengl apps crash with latest llvm 3.6 git / mesa git / radeonsi</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88219">Bug 88219</a> - include/c11/threads_posix.h:197: undefined reference to `pthread_mutex_lock'</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88227">Bug 88227</a> - Radeonsi: High GTT usage in Prison Architect large map</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88248">Bug 88248</a> - Calling glClear while there is an occlusion query in progress messes up the results</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88335">Bug 88335</a> - format_pack.c:9567:22: error: expected ')'</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88385">Bug 88385</a> - [SNB+ Bisected]Ogles3conform ES3-CTS.gtf.GL3Tests.packed_pixels.packed_pixels core dumped</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88467">Bug 88467</a> - nir.c:140: error: ‘nir_src’ has no member named ‘ssa’</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88478">Bug 88478</a> - #error &quot;&lt;malloc.h&gt; has been replaced by &lt;stdlib.h&gt;&quot;</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88519">Bug 88519</a> - sha1.c:210:22: error: 'grcy_md_hd_t' undeclared (first use in this function)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88523">Bug 88523</a> - sha1.c:37: error: 'SHA1_CTX' undeclared (first use in this function)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88561">Bug 88561</a> - [radeonsi][regression,bisected] Depth test/buffer issues in Portal</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88658">Bug 88658</a> - (bisected) Slow video playback on Kabini</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88662">Bug 88662</a> - unaligned access to gl_dlist_node</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88783">Bug 88783</a> - FTBFS: Clover: src/gallium/state_trackers/clover/llvm/invocation.cpp:335:49: error: no matching function for call to 'llvm::TargetLibraryInfo::TargetLibraryInfo(llvm::Triple)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88792">Bug 88792</a> - [BDW/BSW Bisected]Piglit spec_ARB_pixel_buffer_object_pbo-read-argb8888 fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88806">Bug 88806</a> - nir/nir_constant_expressions.c:2754:15: error: controlling expression type 'unsigned int' not compatible with any generic association type</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88841">Bug 88841</a> - [SNB/IVB/HSW/BDW Bisected]Piglit spec_EGL_NOK_texture_from_pixmap_basic fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88852">Bug 88852</a> - macros.h(181) : error C2143: syntax error : missing '{' before 'enum [tag]'</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88905">Bug 88905</a> - [SNB+ Bisected]Ogles3conform ES3-CTS.gtf.GL3Tests.packed_pixels.packed_pixels fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88930">Bug 88930</a> - [osmesa] osbuffer-&gt;textures should be indexed by attachment type</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=88962">Bug 88962</a> - [osmesa] Crash on postprocessing if z buffer is NULL</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89032">Bug 89032</a> - [BDW/BSW/SKL Bisected]Piglit spec_OpenGL_1.1_infinite-spot-light fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89037">Bug 89037</a> - [SKL]Piglit spec_EXT_texture_array_copyteximage_1D_ARRAY_samples=2 sporadically causes GPU hang</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89068">Bug 89068</a> - glTexImage2D regression by texstore_rgba switch to _mesa_format_convert</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89069">Bug 89069</a> - Lack of grass in The Talos Principle on radeonsi (native\wine\nine)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89180">Bug 89180</a> - [IVB regression] Rendering issues in Mass Effect through VMware Workstation</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86330">Bug 86330</a> - lp_bld_debug.cpp:112: multiple definition of `raw_debug_ostream::write_impl(char const*, unsigned long)'</li>
+
+</ul>
+
 
 <h2>Changes</h2>
 

install-lib-links.mk

@@ -3,9 +3,9 @@
 
 if BUILD_SHARED
 if HAVE_COMPAT_SYMLINKS
-all-local : .libs/install-mesa-links
+all-local : .install-mesa-links
 
-.libs/install-mesa-links : $(lib_LTLIBRARIES)
+.install-mesa-links : $(lib_LTLIBRARIES)
 	$(AM_V_GEN)$(MKDIR_P) $(top_builddir)/$(LIB_DIR);	\
 	for f in $(join $(addsuffix .libs/,$(dir $(lib_LTLIBRARIES))),$(notdir $(lib_LTLIBRARIES:%.la=%.$(LIB_EXT)*))); do \
 		if test -h .libs/$$f; then			\
@@ -19,6 +19,7 @@ clean-local:
 	for f in $(notdir $(lib_LTLIBRARIES:%.la=.libs/%.$(LIB_EXT)*)); do \
 		$(RM) $(top_builddir)/$(LIB_DIR)/$$f;	\
 	done;
+	$(RM) .install-mesa-links
 
 endif
 endif

src/egl/drivers/dri2/egl_dri2.c

@@ -534,7 +534,7 @@ dri2_setup_screen(_EGLDisplay *disp)
          disp->Extensions.KHR_gl_texture_2D_image = EGL_TRUE;
          disp->Extensions.KHR_gl_texture_cubemap_image = EGL_TRUE;
       }
-#ifdef HAVE_DRM_PLATFORM
+#ifdef HAVE_LIBDRM
       if (dri2_dpy->image->base.version >= 8 &&
           dri2_dpy->image->createImageFromDmaBufs) {
          disp->Extensions.EXT_image_dma_buf_import = EGL_TRUE;
@@ -1335,7 +1335,7 @@ dri2_create_image_khr_renderbuffer(_EGLDisplay *disp, _EGLContext *ctx,
    return dri2_create_image_from_dri(disp, dri_image);
 }
 
-#ifdef HAVE_DRM_PLATFORM
+#ifdef HAVE_LIBDRM
 static _EGLImage *
 dri2_create_image_mesa_drm_buffer(_EGLDisplay *disp, _EGLContext *ctx,
 				  EGLClientBuffer buffer, const EGLint *attr_list)
@@ -1571,7 +1571,7 @@ dri2_create_wayland_buffer_from_image(_EGLDriver *drv, _EGLDisplay *dpy,
    return dri2_dpy->vtbl->create_wayland_buffer_from_image(drv, dpy, img);
 }
 
-#ifdef HAVE_DRM_PLATFORM
+#ifdef HAVE_LIBDRM
 static EGLBoolean
 dri2_check_dma_buf_attribs(const _EGLImageAttribs *attrs)
 {
@@ -1821,7 +1821,7 @@ dri2_create_image_khr(_EGLDriver *drv, _EGLDisplay *disp,
       return dri2_create_image_khr_texture(disp, ctx, target, buffer, attr_list);
    case EGL_GL_RENDERBUFFER_KHR:
       return dri2_create_image_khr_renderbuffer(disp, ctx, buffer, attr_list);
-#ifdef HAVE_DRM_PLATFORM
+#ifdef HAVE_LIBDRM
    case EGL_DRM_BUFFER_MESA:
       return dri2_create_image_mesa_drm_buffer(disp, ctx, buffer, attr_list);
 #endif
@@ -1829,7 +1829,7 @@ dri2_create_image_khr(_EGLDriver *drv, _EGLDisplay *disp,
    case EGL_WAYLAND_BUFFER_WL:
       return dri2_create_image_wayland_wl_buffer(disp, ctx, buffer, attr_list);
 #endif
-#ifdef HAVE_DRM_PLATFORM
+#ifdef HAVE_LIBDRM
    case EGL_LINUX_DMA_BUF_EXT:
       return dri2_create_image_dma_buf(disp, ctx, buffer, attr_list);
 #endif
@@ -1853,7 +1853,7 @@ dri2_destroy_image_khr(_EGLDriver *drv, _EGLDisplay *disp, _EGLImage *image)
    return EGL_TRUE;
 }
 
-#ifdef HAVE_DRM_PLATFORM
+#ifdef HAVE_LIBDRM
 static _EGLImage *
 dri2_create_drm_image_mesa(_EGLDriver *drv, _EGLDisplay *disp,
 			   const EGLint *attr_list)
@@ -2215,7 +2215,7 @@ _eglBuiltInDriverDRI2(const char *args)
    dri2_drv->base.API.CreateImageKHR = dri2_create_image;
    dri2_drv->base.API.DestroyImageKHR = dri2_destroy_image_khr;
    dri2_drv->base.API.CreateWaylandBufferFromImageWL = dri2_create_wayland_buffer_from_image;
-#ifdef HAVE_DRM_PLATFORM
+#ifdef HAVE_LIBDRM
    dri2_drv->base.API.CreateDRMImageMESA = dri2_create_drm_image_mesa;
    dri2_drv->base.API.ExportDRMImageMESA = dri2_export_drm_image_mesa;
 #endif

src/egl/drivers/dri2/platform_wayland.c

@@ -292,6 +292,26 @@ get_back_bo(struct dri2_egl_surface *dri2_surf)
    struct dri2_egl_display *dri2_dpy =
       dri2_egl_display(dri2_surf->base.Resource.Display);
    int i;
+   unsigned int dri_image_format;
+
+   /* currently supports three WL DRM formats,
+    * WL_DRM_FORMAT_ARGB8888, WL_DRM_FORMAT_XRGB8888,
+    * and WL_DRM_FORMAT_RGB565
+    */
+   switch (dri2_surf->format) {
+   case WL_DRM_FORMAT_ARGB8888:
+      dri_image_format = __DRI_IMAGE_FORMAT_ARGB8888;
+      break;
+   case WL_DRM_FORMAT_XRGB8888:
+      dri_image_format = __DRI_IMAGE_FORMAT_XRGB8888;
+      break;
+   case WL_DRM_FORMAT_RGB565:
+      dri_image_format = __DRI_IMAGE_FORMAT_RGB565;
+      break;
+   default:
+      /* format is not supported */
+      return -1;
+   }
 
    /* We always want to throttle to some event (either a frame callback or
     * a sync request) after the commit so that we can be sure the
@@ -322,7 +342,7 @@ get_back_bo(struct dri2_egl_surface *dri2_surf)
          dri2_dpy->image->createImage(dri2_dpy->dri_screen,
                                       dri2_surf->base.Width,
                                       dri2_surf->base.Height,
-                                      __DRI_IMAGE_FORMAT_ARGB8888,
+                                      dri_image_format,
                                       __DRI_IMAGE_USE_SHARE,
                                       NULL);
       dri2_surf->back->age = 0;
@@ -462,11 +482,26 @@ dri2_wl_get_buffers(__DRIdrawable * driDrawable,
                     unsigned int *attachments, int count,
                     int *out_count, void *loaderPrivate)
 {
+   struct dri2_egl_surface *dri2_surf = loaderPrivate;
    unsigned int *attachments_with_format;
    __DRIbuffer *buffer;
-   const unsigned int format = 32;
+   unsigned int bpp;
+
    int i;
 
+   switch (dri2_surf->format) {
+   case WL_DRM_FORMAT_ARGB8888:
+   case WL_DRM_FORMAT_XRGB8888:
+      bpp = 32;
+      break;
+   case WL_DRM_FORMAT_RGB565:
+      bpp = 16;
+      break;
+   default:
+      /* format is not supported */
+      return NULL;
+   }
+
    attachments_with_format = calloc(count, 2 * sizeof(unsigned int));
    if (!attachments_with_format) {
       *out_count = 0;
@@ -475,7 +510,7 @@ dri2_wl_get_buffers(__DRIdrawable * driDrawable,
 
    for (i = 0; i < count; ++i) {
       attachments_with_format[2*i] = attachments[i];
-      attachments_with_format[2*i + 1] = format;
+      attachments_with_format[2*i + 1] = bpp;
    }
 
    buffer =

src/gallium/auxiliary/Makefile.am

@@ -53,24 +53,40 @@ libgalliumvl_stub_la_SOURCES = \
 
 if NEED_GALLIUM_VL
 
-noinst_LTLIBRARIES += libgalliumvl.la
-
-libgalliumvl_la_CFLAGS = \
+COMMON_VL_CFLAGS = \
 	$(AM_CFLAGS) \
 	$(VL_CFLAGS) \
+	$(DRI2PROTO_CFLAGS) \
 	$(LIBDRM_CFLAGS) \
 	$(GALLIUM_PIPE_LOADER_DEFINES) \
 	-DPIPE_SEARCH_DIR=\"$(libdir)/gallium-pipe\"
 
 if HAVE_GALLIUM_STATIC_TARGETS
-libgalliumvl_la_CFLAGS += \
+COMMON_VL_CFLAGS += \
 	-DGALLIUM_STATIC_TARGETS=1
 
 endif # HAVE_GALLIUM_STATIC_TARGETS
 
+noinst_LTLIBRARIES += libgalliumvl.la
+
+libgalliumvl_la_CFLAGS = \
+	$(COMMON_VL_CFLAGS)
+
 libgalliumvl_la_SOURCES = \
 	$(VL_SOURCES)
 
+if NEED_GALLIUM_VL_WINSYS
+
+noinst_LTLIBRARIES += libgalliumvlwinsys.la
+
+libgalliumvlwinsys_la_CFLAGS = \
+	$(COMMON_VL_CFLAGS)
+
+libgalliumvlwinsys_la_SOURCES = \
+	$(VL_WINSYS_SOURCES)
+
+endif
+
 endif
 
 EXTRA_DIST = \

src/gallium/auxiliary/Makefile.sources

@@ -334,10 +334,13 @@ VL_SOURCES := \
 	vl/vl_video_buffer.h \
 	vl/vl_vlc.h \
 	vl/vl_winsys.h \
-	vl/vl_winsys_dri.c \
 	vl/vl_zscan.c \
 	vl/vl_zscan.h
 
+# XXX: Nuke this as our dri targets no longer depend on VL.
+VL_WINSYS_SOURCES := \
+	vl/vl_winsys_dri.c
+
 VL_STUB_SOURCES := \
 	vl/vl_stubs.c
 

src/gallium/auxiliary/os/os_misc.c

@@ -118,7 +118,7 @@ os_get_total_physical_memory(uint64_t *size)
    *size = phys_pages * page_size;
    return (phys_pages > 0 && page_size > 0);
 #elif defined(PIPE_OS_APPLE) || defined(PIPE_OS_BSD)
-   size_t len = sizeof(size);
+   size_t len = sizeof(*size);
    int mib[2];
 
    mib[0] = CTL_HW;
@@ -134,7 +134,7 @@ os_get_total_physical_memory(uint64_t *size)
 #error Unsupported *BSD
 #endif
 
-   return (sysctl(mib, 2, &size, &len, NULL, 0) == 0);
+   return (sysctl(mib, 2, size, &len, NULL, 0) == 0);
 #elif defined(PIPE_OS_HAIKU)
    system_info info;
    status_t ret;

src/gallium/drivers/nouveau/nvc0/nvc0_shader_state.c

@@ -252,7 +252,12 @@ nvc0_tfb_validate(struct nvc0_context *nvc0)
 
    for (b = 0; b < nvc0->num_tfbbufs; ++b) {
       struct nvc0_so_target *targ = nvc0_so_target(nvc0->tfbbuf[b]);
-      struct nv04_resource *buf = nv04_resource(targ->pipe.buffer);
+      struct nv04_resource *buf;
+
+      if (!targ) {
+         IMMED_NVC0(push, NVC0_3D(TFB_BUFFER_ENABLE(b)), 0);
+         continue;
+      }
 
       if (tfb)
          targ->stride = tfb->stride[b];
@@ -260,6 +265,8 @@ nvc0_tfb_validate(struct nvc0_context *nvc0)
       if (!(nvc0->tfbbuf_dirty & (1 << b)))
          continue;
 
+      buf = nv04_resource(targ->pipe.buffer);
+
       if (!targ->clean)
          nvc0_query_fifo_wait(push, targ->pq);
       BEGIN_NVC0(push, NVC0_3D(TFB_BUFFER_ENABLE(b)), 5);

src/gallium/drivers/nouveau/nvc0/nvc0_state.c

@@ -1089,9 +1089,11 @@ nvc0_set_transform_feedback_targets(struct pipe_context *pipe,
       pipe_so_target_reference(&nvc0->tfbbuf[i], targets[i]);
    }
    for (; i < nvc0->num_tfbbufs; ++i) {
-      nvc0->tfbbuf_dirty |= 1 << i;
-      nvc0_so_target_save_offset(pipe, nvc0->tfbbuf[i], i, &serialize);
-      pipe_so_target_reference(&nvc0->tfbbuf[i], NULL);
+      if (nvc0->tfbbuf[i]) {
+         nvc0->tfbbuf_dirty |= 1 << i;
+         nvc0_so_target_save_offset(pipe, nvc0->tfbbuf[i], i, &serialize);
+         pipe_so_target_reference(&nvc0->tfbbuf[i], NULL);
+      }
    }
    nvc0->num_tfbbufs = num_targets;
 

src/gallium/drivers/nouveau/nvc0/nvc0_surface.c

@@ -1401,11 +1401,14 @@ nvc0_blit(struct pipe_context *pipe, const struct pipe_blit_info *info)
       } else
       if (!nv50_2d_src_format_faithful(info->src.format)) {
          if (!util_format_is_luminance(info->src.format)) {
+            if (!nv50_2d_dst_format_ops_supported(info->dst.format))
+               eng3d = TRUE;
+            else
             if (util_format_is_intensity(info->src.format))
                eng3d = info->src.format != PIPE_FORMAT_I8_UNORM;
             else
-            if (!nv50_2d_dst_format_ops_supported(info->dst.format))
-               eng3d = TRUE;
+            if (util_format_is_alpha(info->src.format))
+               eng3d = info->src.format != PIPE_FORMAT_A8_UNORM;
             else
                eng3d = !nv50_2d_format_supported(info->src.format);
          }

src/gallium/drivers/r600/sb/sb_sched.cpp

@@ -266,7 +266,7 @@ bool rp_gpr_tracker::try_reserve(alu_node* n) {
 
 	for (i = 0; i < nsrc; ++i) {
 		value *v = n->src[i];
-		if (v->is_readonly()) {
+		if (v->is_readonly() || v->is_undef()) {
 			const_count++;
 			if (trans && const_count == 3)
 				break;
@@ -295,7 +295,7 @@ bool rp_gpr_tracker::try_reserve(alu_node* n) {
 	if (need_unreserve && i--) {
 		do {
 			value *v = n->src[i];
-			if (!v->is_readonly()) {
+			if (!v->is_readonly() && !v->is_undef()) {
 			if (i == 1 && opt)
 				continue;
 			unreserve(bs_cycle(trans, bs, i), n->bc.src[i].sel,

src/gallium/drivers/radeon/radeon_llvm_emit.c

@@ -80,10 +80,6 @@ void radeon_llvm_shader_type(LLVMValueRef F, unsigned type)
 	sprintf(Str, "%1d", llvm_type);
 
 	LLVMAddTargetDependentFunctionAttr(F, "ShaderType", Str);
-
-	if (type != TGSI_PROCESSOR_COMPUTE) {
-		LLVMAddTargetDependentFunctionAttr(F, "unsafe-fp-math", "true");
-	}
 }
 
 static void init_r600_target()

src/gallium/drivers/radeonsi/si_state.c

@@ -735,12 +735,16 @@ static void si_delete_rs_state(struct pipe_context *ctx, void *state)
  */
 static void si_update_dsa_stencil_ref(struct si_context *sctx)
 {
-	struct si_pm4_state *pm4 = CALLOC_STRUCT(si_pm4_state);
+	struct si_pm4_state *pm4;
 	struct pipe_stencil_ref *ref = &sctx->stencil_ref;
-        struct si_state_dsa *dsa = sctx->queued.named.dsa;
+	struct si_state_dsa *dsa = sctx->queued.named.dsa;
 
-        if (pm4 == NULL)
-                return;
+	if (!dsa)
+		return;
+
+	pm4 = CALLOC_STRUCT(si_pm4_state);
+	if (pm4 == NULL)
+		return;
 
 	si_pm4_set_reg(pm4, R_028430_DB_STENCILREFMASK,
 		       S_028430_STENCILTESTVAL(ref->ref_value[0]) |

src/gallium/drivers/radeonsi/si_state_draw.c

@@ -367,24 +367,21 @@ void si_emit_cache_flush(struct r600_common_context *sctx, struct r600_atom *ato
 {
 	struct radeon_winsys_cs *cs = sctx->rings.gfx.cs;
 	uint32_t cp_coher_cntl = 0;
-	uint32_t sqc_caches = 0;
 	uint32_t compute =
 		PKT3_SHADER_TYPE_S(!!(sctx->flags & SI_CONTEXT_FLAG_COMPUTE));
 
 	/* SI has a bug that it always flushes ICACHE and KCACHE if either
-	 * bit is set. An alternative way is to write SQC_CACHES. */
-	if (sctx->chip_class == SI &&
-	    sctx->flags & BOTH_ICACHE_KCACHE &&
-	    (sctx->flags & BOTH_ICACHE_KCACHE) != BOTH_ICACHE_KCACHE) {
-		sqc_caches =
-			S_008C08_INST_INVALIDATE(!!(sctx->flags & SI_CONTEXT_INV_ICACHE)) |
-			S_008C08_DATA_INVALIDATE(!!(sctx->flags & SI_CONTEXT_INV_KCACHE));
-	} else {
-		if (sctx->flags & SI_CONTEXT_INV_ICACHE)
-			cp_coher_cntl |= S_0085F0_SH_ICACHE_ACTION_ENA(1);
-		if (sctx->flags & SI_CONTEXT_INV_KCACHE)
-			cp_coher_cntl |= S_0085F0_SH_KCACHE_ACTION_ENA(1);
-	}
+	 * bit is set. An alternative way is to write SQC_CACHES, but that
+	 * doesn't seem to work reliably. Since the bug doesn't affect
+	 * correctness (it only does more work than necessary) and
+	 * the performance impact is likely negligible, there is no plan
+	 * to fix it.
+	 */
+
+	if (sctx->flags & SI_CONTEXT_INV_ICACHE)
+		cp_coher_cntl |= S_0085F0_SH_ICACHE_ACTION_ENA(1);
+	if (sctx->flags & SI_CONTEXT_INV_KCACHE)
+		cp_coher_cntl |= S_0085F0_SH_KCACHE_ACTION_ENA(1);
 
 	if (sctx->flags & SI_CONTEXT_INV_TC_L1)
 		cp_coher_cntl |= S_0085F0_TCL1_ACTION_ENA(1);
@@ -451,10 +448,6 @@ void si_emit_cache_flush(struct r600_common_context *sctx, struct r600_atom *ato
 	 * It looks like SURFACE_SYNC flushes caches immediately and doesn't
 	 * wait for any engines. This should be last.
 	 */
-	if (sqc_caches) {
-		r600_write_config_reg(cs, R_008C08_SQC_CACHES, sqc_caches);
-		cs->buf[cs->cdw-3] |= compute; /* set the compute bit in the header */
-	}
 	if (cp_coher_cntl) {
 		if (sctx->chip_class >= CIK) {
 			radeon_emit(cs, PKT3(PKT3_ACQUIRE_MEM, 5, 0) | compute);

src/gallium/drivers/radeonsi/si_state_shaders.c

@@ -661,9 +661,11 @@ bcolor:
 			}
 		}
 
-		if (j == vsinfo->num_outputs) {
-			/* No corresponding output found, load defaults into input */
-			tmp |= S_028644_OFFSET(0x20);
+		if (j == vsinfo->num_outputs && !G_028644_PT_SPRITE_TEX(tmp)) {
+			/* No corresponding output found, load defaults into input.
+			 * Don't set any other bits.
+			 * (FLAT_SHADE=1 completely changes behavior) */
+			tmp = S_028644_OFFSET(0x20);
 		}
 
 		si_pm4_set_reg(pm4,

src/gallium/state_trackers/omx/vid_dec_h264.c

@@ -706,6 +706,11 @@ static void slice_header(vid_dec_PrivateType *priv, struct vl_rbsp *rbsp,
       if (pic_order_cnt_lsb != priv->codec_data.h264.pic_order_cnt_lsb)
          vid_dec_h264_EndFrame(priv);
 
+      if (IdrPicFlag) {
+         priv->codec_data.h264.pic_order_cnt_msb = 0;
+         priv->codec_data.h264.pic_order_cnt_lsb = 0;
+      }
+
       if ((pic_order_cnt_lsb < priv->codec_data.h264.pic_order_cnt_lsb) &&
           (priv->codec_data.h264.pic_order_cnt_lsb - pic_order_cnt_lsb) >= (max_pic_order_cnt_lsb / 2))
          pic_order_cnt_msb = priv->codec_data.h264.pic_order_cnt_msb + max_pic_order_cnt_lsb;

src/gallium/targets/dri/Makefile.am

@@ -44,7 +44,7 @@ gallium_dri_la_LIBADD = \
 	$(top_builddir)/src/mesa/drivers/dri/common/libdricommon.la \
 	$(top_builddir)/src/mesa/drivers/dri/common/libmegadriver_stub.la \
 	$(top_builddir)/src/gallium/state_trackers/dri/libdri.la \
-	$(top_builddir)/src/gallium/auxiliary/libgalliumvl_stub.la \
+	$(top_builddir)/src/gallium/auxiliary/libgalliumvl.la \
 	$(top_builddir)/src/gallium/auxiliary/libgallium.la \
 	$(top_builddir)/src/gallium/drivers/galahad/libgalahad.la \
 	$(top_builddir)/src/gallium/drivers/noop/libnoop.la \

src/gallium/targets/omx/Makefile.am

@@ -24,6 +24,7 @@ endif # HAVE_LD_VERSION_SCRIPT
 
 libomx_mesa_la_LIBADD = \
 	$(top_builddir)/src/gallium/state_trackers/omx/libomxtracker.la \
+	$(top_builddir)/src/gallium/auxiliary/libgalliumvlwinsys.la \
 	$(top_builddir)/src/gallium/auxiliary/libgalliumvl.la \
 	$(top_builddir)/src/gallium/auxiliary/libgallium.la \
 	$(top_builddir)/src/util/libmesautil.la \

src/gallium/targets/pipe-loader/Makefile.am

@@ -46,6 +46,11 @@ PIPE_LIBS += \
 	$(top_builddir)/src/gallium/auxiliary/libgalliumvl_stub.la
 endif
 
+if NEED_GALLIUM_VL_WINSYS
+PIPE_LIBS+= \
+	$(top_builddir)/src/gallium/auxiliary/libgalliumvlwinsys.la
+endif
+
 PIPE_LIBS += \
 	$(top_builddir)/src/gallium/auxiliary/libgallium.la \
 	$(top_builddir)/src/util/libmesautil.la \

src/gallium/targets/va/Makefile.am

@@ -25,6 +25,7 @@ endif # HAVE_LD_VERSION_SCRIPT
 # NOTE: gallium_drv_video does not use(link against) libva
 gallium_drv_video_la_LIBADD = \
 	$(top_builddir)/src/gallium/state_trackers/va/libvatracker.la \
+	$(top_builddir)/src/gallium/auxiliary/libgalliumvlwinsys.la \
 	$(top_builddir)/src/gallium/auxiliary/libgalliumvl.la \
 	$(top_builddir)/src/gallium/auxiliary/libgallium.la \
 	$(top_builddir)/src/util/libmesautil.la \

src/gallium/targets/vdpau/Makefile.am

@@ -30,6 +30,7 @@ endif # HAVE_LD_DYNAMIC_LIST
 # NOTE: libvdpau_gallium does not use(link against) libvdpau
 libvdpau_gallium_la_LIBADD = \
 	$(top_builddir)/src/gallium/state_trackers/vdpau/libvdpautracker.la \
+	$(top_builddir)/src/gallium/auxiliary/libgalliumvlwinsys.la \
 	$(top_builddir)/src/gallium/auxiliary/libgalliumvl.la \
 	$(top_builddir)/src/gallium/auxiliary/libgallium.la \
 	$(top_builddir)/src/util/libmesautil.la \

src/gallium/targets/xvmc/Makefile.am

@@ -24,6 +24,7 @@ endif # HAVE_LD_VERSION_SCRIPT
 
 libXvMCgallium_la_LIBADD = \
 	$(top_builddir)/src/gallium/state_trackers/xvmc/libxvmctracker.la \
+	$(top_builddir)/src/gallium/auxiliary/libgalliumvlwinsys.la \
 	$(top_builddir)/src/gallium/auxiliary/libgalliumvl.la \
 	$(top_builddir)/src/gallium/auxiliary/libgallium.la \
 	$(top_builddir)/src/util/libmesautil.la \

src/glsl/Makefile.am

@@ -240,7 +240,7 @@ nir/nir_opcodes.h: nir/nir_opcodes.py nir/nir_opcodes_h.py
 	$(MKDIR_P) nir;							\
 	$(PYTHON2) $(PYTHON_FLAGS) $(srcdir)/nir/nir_opcodes_h.py > $@
 
-nir/nir.h: $(top_builddir)/src/glsl/nir/nir_opcodes.h
+nir/nir.h: nir/nir_opcodes.h
 
 nir/nir_opcodes.c: nir/nir_opcodes.py nir/nir_opcodes_c.py
 	$(MKDIR_P) nir;							\

src/glsl/Makefile.sources

@@ -21,6 +21,7 @@ NIR_FILES = \
 	nir/glsl_to_nir.h \
 	nir/nir.c \
 	nir/nir.h \
+	nir/nir_constant_expressions.h \
 	nir/nir_dominance.c \
 	nir/nir_from_ssa.c \
 	nir/nir_intrinsics.c \

src/glsl/builtin_type_macros.h

@@ -110,39 +110,39 @@ DECL_TYPE(sampler2DRectShadow,    GL_SAMPLER_2D_RECT_SHADOW,        GLSL_TYPE_SA
 
 DECL_TYPE(samplerExternalOES,     GL_SAMPLER_EXTERNAL_OES,          GLSL_TYPE_SAMPLER, GLSL_SAMPLER_DIM_EXTERNAL, 0, 0, GLSL_TYPE_FLOAT)
 
-DECL_TYPE(image1D,         GL_IMAGE_1D,                                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 0, GLSL_TYPE_FLOAT);
-DECL_TYPE(image2D,         GL_IMAGE_2D,                                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 0, GLSL_TYPE_FLOAT);
-DECL_TYPE(image3D,         GL_IMAGE_3D,                                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_3D,     0, 0, GLSL_TYPE_FLOAT);
-DECL_TYPE(image2DRect,     GL_IMAGE_2D_RECT,                           GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_RECT,   0, 0, GLSL_TYPE_FLOAT);
-DECL_TYPE(imageCube,       GL_IMAGE_CUBE,                              GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 0, GLSL_TYPE_FLOAT);
-DECL_TYPE(imageBuffer,     GL_IMAGE_BUFFER,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_BUF,    0, 0, GLSL_TYPE_FLOAT);
-DECL_TYPE(image1DArray,    GL_IMAGE_1D_ARRAY,                          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 1, GLSL_TYPE_FLOAT);
-DECL_TYPE(image2DArray,    GL_IMAGE_2D_ARRAY,                          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 1, GLSL_TYPE_FLOAT);
-DECL_TYPE(imageCubeArray,  GL_IMAGE_CUBE_MAP_ARRAY,                    GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 1, GLSL_TYPE_FLOAT);
-DECL_TYPE(image2DMS,       GL_IMAGE_2D_MULTISAMPLE,                    GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 0, GLSL_TYPE_FLOAT);
-DECL_TYPE(image2DMSArray,  GL_IMAGE_2D_MULTISAMPLE_ARRAY,              GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 1, GLSL_TYPE_FLOAT);
-DECL_TYPE(iimage1D,        GL_INT_IMAGE_1D,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 0, GLSL_TYPE_INT);
-DECL_TYPE(iimage2D,        GL_INT_IMAGE_2D,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 0, GLSL_TYPE_INT);
-DECL_TYPE(iimage3D,        GL_INT_IMAGE_3D,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_3D,     0, 0, GLSL_TYPE_INT);
-DECL_TYPE(iimage2DRect,    GL_INT_IMAGE_2D_RECT,                       GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_RECT,   0, 0, GLSL_TYPE_INT);
-DECL_TYPE(iimageCube,      GL_INT_IMAGE_CUBE,                          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 0, GLSL_TYPE_INT);
-DECL_TYPE(iimageBuffer,    GL_INT_IMAGE_BUFFER,                        GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_BUF,    0, 0, GLSL_TYPE_INT);
-DECL_TYPE(iimage1DArray,   GL_INT_IMAGE_1D_ARRAY,                      GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 1, GLSL_TYPE_INT);
-DECL_TYPE(iimage2DArray,   GL_INT_IMAGE_2D_ARRAY,                      GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 1, GLSL_TYPE_INT);
-DECL_TYPE(iimageCubeArray, GL_INT_IMAGE_CUBE_MAP_ARRAY,                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 1, GLSL_TYPE_INT);
-DECL_TYPE(iimage2DMS,      GL_INT_IMAGE_2D_MULTISAMPLE,                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 0, GLSL_TYPE_INT);
-DECL_TYPE(iimage2DMSArray, GL_INT_IMAGE_2D_MULTISAMPLE_ARRAY,          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 1, GLSL_TYPE_INT);
-DECL_TYPE(uimage1D,        GL_UNSIGNED_INT_IMAGE_1D,                   GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 0, GLSL_TYPE_UINT);
-DECL_TYPE(uimage2D,        GL_UNSIGNED_INT_IMAGE_2D,                   GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 0, GLSL_TYPE_UINT);
-DECL_TYPE(uimage3D,        GL_UNSIGNED_INT_IMAGE_3D,                   GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_3D,     0, 0, GLSL_TYPE_UINT);
-DECL_TYPE(uimage2DRect,    GL_UNSIGNED_INT_IMAGE_2D_RECT,              GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_RECT,   0, 0, GLSL_TYPE_UINT);
-DECL_TYPE(uimageCube,      GL_UNSIGNED_INT_IMAGE_CUBE,                 GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 0, GLSL_TYPE_UINT);
-DECL_TYPE(uimageBuffer,    GL_UNSIGNED_INT_IMAGE_BUFFER,               GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_BUF,    0, 0, GLSL_TYPE_UINT);
-DECL_TYPE(uimage1DArray,   GL_UNSIGNED_INT_IMAGE_1D_ARRAY,             GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 1, GLSL_TYPE_UINT);
-DECL_TYPE(uimage2DArray,   GL_UNSIGNED_INT_IMAGE_2D_ARRAY,             GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 1, GLSL_TYPE_UINT);
-DECL_TYPE(uimageCubeArray, GL_UNSIGNED_INT_IMAGE_CUBE_MAP_ARRAY,       GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 1, GLSL_TYPE_UINT);
-DECL_TYPE(uimage2DMS,      GL_UNSIGNED_INT_IMAGE_2D_MULTISAMPLE,       GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 0, GLSL_TYPE_UINT);
-DECL_TYPE(uimage2DMSArray, GL_UNSIGNED_INT_IMAGE_2D_MULTISAMPLE_ARRAY, GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 1, GLSL_TYPE_UINT);
+DECL_TYPE(image1D,         GL_IMAGE_1D,                                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 0, GLSL_TYPE_FLOAT)
+DECL_TYPE(image2D,         GL_IMAGE_2D,                                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 0, GLSL_TYPE_FLOAT)
+DECL_TYPE(image3D,         GL_IMAGE_3D,                                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_3D,     0, 0, GLSL_TYPE_FLOAT)
+DECL_TYPE(image2DRect,     GL_IMAGE_2D_RECT,                           GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_RECT,   0, 0, GLSL_TYPE_FLOAT)
+DECL_TYPE(imageCube,       GL_IMAGE_CUBE,                              GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 0, GLSL_TYPE_FLOAT)
+DECL_TYPE(imageBuffer,     GL_IMAGE_BUFFER,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_BUF,    0, 0, GLSL_TYPE_FLOAT)
+DECL_TYPE(image1DArray,    GL_IMAGE_1D_ARRAY,                          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 1, GLSL_TYPE_FLOAT)
+DECL_TYPE(image2DArray,    GL_IMAGE_2D_ARRAY,                          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 1, GLSL_TYPE_FLOAT)
+DECL_TYPE(imageCubeArray,  GL_IMAGE_CUBE_MAP_ARRAY,                    GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 1, GLSL_TYPE_FLOAT)
+DECL_TYPE(image2DMS,       GL_IMAGE_2D_MULTISAMPLE,                    GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 0, GLSL_TYPE_FLOAT)
+DECL_TYPE(image2DMSArray,  GL_IMAGE_2D_MULTISAMPLE_ARRAY,              GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 1, GLSL_TYPE_FLOAT)
+DECL_TYPE(iimage1D,        GL_INT_IMAGE_1D,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 0, GLSL_TYPE_INT)
+DECL_TYPE(iimage2D,        GL_INT_IMAGE_2D,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 0, GLSL_TYPE_INT)
+DECL_TYPE(iimage3D,        GL_INT_IMAGE_3D,                            GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_3D,     0, 0, GLSL_TYPE_INT)
+DECL_TYPE(iimage2DRect,    GL_INT_IMAGE_2D_RECT,                       GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_RECT,   0, 0, GLSL_TYPE_INT)
+DECL_TYPE(iimageCube,      GL_INT_IMAGE_CUBE,                          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 0, GLSL_TYPE_INT)
+DECL_TYPE(iimageBuffer,    GL_INT_IMAGE_BUFFER,                        GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_BUF,    0, 0, GLSL_TYPE_INT)
+DECL_TYPE(iimage1DArray,   GL_INT_IMAGE_1D_ARRAY,                      GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 1, GLSL_TYPE_INT)
+DECL_TYPE(iimage2DArray,   GL_INT_IMAGE_2D_ARRAY,                      GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 1, GLSL_TYPE_INT)
+DECL_TYPE(iimageCubeArray, GL_INT_IMAGE_CUBE_MAP_ARRAY,                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 1, GLSL_TYPE_INT)
+DECL_TYPE(iimage2DMS,      GL_INT_IMAGE_2D_MULTISAMPLE,                GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 0, GLSL_TYPE_INT)
+DECL_TYPE(iimage2DMSArray, GL_INT_IMAGE_2D_MULTISAMPLE_ARRAY,          GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 1, GLSL_TYPE_INT)
+DECL_TYPE(uimage1D,        GL_UNSIGNED_INT_IMAGE_1D,                   GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 0, GLSL_TYPE_UINT)
+DECL_TYPE(uimage2D,        GL_UNSIGNED_INT_IMAGE_2D,                   GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 0, GLSL_TYPE_UINT)
+DECL_TYPE(uimage3D,        GL_UNSIGNED_INT_IMAGE_3D,                   GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_3D,     0, 0, GLSL_TYPE_UINT)
+DECL_TYPE(uimage2DRect,    GL_UNSIGNED_INT_IMAGE_2D_RECT,              GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_RECT,   0, 0, GLSL_TYPE_UINT)
+DECL_TYPE(uimageCube,      GL_UNSIGNED_INT_IMAGE_CUBE,                 GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 0, GLSL_TYPE_UINT)
+DECL_TYPE(uimageBuffer,    GL_UNSIGNED_INT_IMAGE_BUFFER,               GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_BUF,    0, 0, GLSL_TYPE_UINT)
+DECL_TYPE(uimage1DArray,   GL_UNSIGNED_INT_IMAGE_1D_ARRAY,             GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_1D,     0, 1, GLSL_TYPE_UINT)
+DECL_TYPE(uimage2DArray,   GL_UNSIGNED_INT_IMAGE_2D_ARRAY,             GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_2D,     0, 1, GLSL_TYPE_UINT)
+DECL_TYPE(uimageCubeArray, GL_UNSIGNED_INT_IMAGE_CUBE_MAP_ARRAY,       GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_CUBE,   0, 1, GLSL_TYPE_UINT)
+DECL_TYPE(uimage2DMS,      GL_UNSIGNED_INT_IMAGE_2D_MULTISAMPLE,       GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 0, GLSL_TYPE_UINT)
+DECL_TYPE(uimage2DMSArray, GL_UNSIGNED_INT_IMAGE_2D_MULTISAMPLE_ARRAY, GLSL_TYPE_IMAGE, GLSL_SAMPLER_DIM_MS,     0, 1, GLSL_TYPE_UINT)
 
 DECL_TYPE(atomic_uint, GL_UNSIGNED_INT_ATOMIC_COUNTER, GLSL_TYPE_ATOMIC_UINT, 1, 1)
 

src/glsl/ir_constant_expression.cpp

@@ -44,7 +44,7 @@ static int isnormal(double x)
 {
    return _fpclass(x) == _FPCLASS_NN || _fpclass(x) == _FPCLASS_PN;
 }
-#elif defined(__SUNPRO_CC)
+#elif defined(__SUNPRO_CC) && !defined(isnormal)
 #include <ieeefp.h>
 static int isnormal(double x)
 {

src/glsl/opt_algebraic.cpp

@@ -696,48 +696,65 @@ ir_algebraic_visitor::handle_expression(ir_expression *ir)
        * a saturate operation
        */
       for (int op = 0; op < 2; op++) {
-         ir_expression *minmax = op_expr[op];
+         ir_expression *inner_expr = op_expr[op];
          ir_constant *outer_const = op_const[1 - op];
          ir_expression_operation op_cond = (ir->operation == ir_binop_max) ?
             ir_binop_min : ir_binop_max;
 
-         if (!minmax || !outer_const || (minmax->operation != op_cond))
+         if (!inner_expr || !outer_const || (inner_expr->operation != op_cond))
             continue;
 
+         /* One of these has to be a constant */
+         if (!inner_expr->operands[0]->as_constant() &&
+             !inner_expr->operands[1]->as_constant())
+            break;
+
          /* Found a min(max) combination. Now try to see if its operands
           * meet our conditions that we can do just a single saturate operation
           */
          for (int minmax_op = 0; minmax_op < 2; minmax_op++) {
-            ir_rvalue *inner_val_a = minmax->operands[minmax_op];
-            ir_rvalue *inner_val_b = minmax->operands[1 - minmax_op];
+            ir_rvalue *x = inner_expr->operands[minmax_op];
+            ir_rvalue *y = inner_expr->operands[1 - minmax_op];
 
-            if (!inner_val_a || !inner_val_b)
+            ir_constant *inner_const = y->as_constant();
+            if (!inner_const)
                continue;
 
-            /* Found a {min|max} ({max|min} (x, 0.0), 1.0) operation and its variations */
-            if ((outer_const->is_one() && inner_val_a->is_zero()) ||
-                (inner_val_a->is_one() && outer_const->is_zero()))
-               return saturate(inner_val_b);
+            /* min(max(x, 0.0), 1.0) is sat(x) */
+            if (ir->operation == ir_binop_min &&
+                inner_const->is_zero() &&
+                outer_const->is_one())
+               return saturate(x);
 
-            /* Found a {min|max} ({max|min} (x, 0.0), b) where b < 1.0
-             * and its variations
-             */
-            if (is_less_than_one(outer_const) && inner_val_b->is_zero())
-               return expr(ir_binop_min, saturate(inner_val_a), outer_const);
+            /* max(min(x, 1.0), 0.0) is sat(x) */
+            if (ir->operation == ir_binop_max &&
+                inner_const->is_one() &&
+                outer_const->is_zero())
+               return saturate(x);
 
-            if (!inner_val_b->as_constant())
-               continue;
+            /* min(max(x, 0.0), b) where b < 1.0 is sat(min(x, b)) */
+            if (ir->operation == ir_binop_min &&
+                inner_const->is_zero() &&
+                is_less_than_one(outer_const))
+               return saturate(expr(ir_binop_min, x, outer_const));
 
-            if (is_less_than_one(inner_val_b->as_constant()) && outer_const->is_zero())
-               return expr(ir_binop_min, saturate(inner_val_a), inner_val_b);
+            /* max(min(x, b), 0.0) where b < 1.0 is sat(min(x, b)) */
+            if (ir->operation == ir_binop_max &&
+                is_less_than_one(inner_const) &&
+                outer_const->is_zero())
+               return saturate(expr(ir_binop_min, x, inner_const));
 
-            /* Found a {min|max} ({max|min} (x, b), 1.0), where b > 0.0
-             * and its variations
-             */
-            if (outer_const->is_one() && is_greater_than_zero(inner_val_b->as_constant()))
-               return expr(ir_binop_max, saturate(inner_val_a), inner_val_b);
-            if (inner_val_b->as_constant()->is_one() && is_greater_than_zero(outer_const))
-               return expr(ir_binop_max, saturate(inner_val_a), outer_const);
+            /* max(min(x, 1.0), b) where b > 0.0 is sat(max(x, b)) */
+            if (ir->operation == ir_binop_max &&
+                inner_const->is_one() &&
+                is_greater_than_zero(outer_const))
+               return saturate(expr(ir_binop_max, x, outer_const));
+
+            /* min(max(x, b), 1.0) where b > 0.0 is sat(max(x, b)) */
+            if (ir->operation == ir_binop_min &&
+                is_greater_than_zero(inner_const) &&
+                outer_const->is_one())
+               return saturate(expr(ir_binop_max, x, inner_const));
          }
       }
 

src/glsl/opt_copy_propagation.cpp

@@ -128,6 +128,9 @@ ir_copy_propagation_visitor::visit_enter(ir_function_signature *ir)
 
    visit_list_elements(this, &ir->body);
 
+   ralloc_free(this->acp);
+   ralloc_free(this->kills);
+
    this->kills = orig_kills;
    this->acp = orig_acp;
    this->killed_all = orig_killed_all;
@@ -215,7 +218,7 @@ ir_copy_propagation_visitor::handle_if_block(exec_list *instructions)
 
    /* Populate the initial acp with a copy of the original */
    foreach_in_list(acp_entry, a, orig_acp) {
-      this->acp->push_tail(new(this->mem_ctx) acp_entry(a->lhs, a->rhs));
+      this->acp->push_tail(new(this->acp) acp_entry(a->lhs, a->rhs));
    }
 
    visit_list_elements(this, instructions);
@@ -226,12 +229,15 @@ ir_copy_propagation_visitor::handle_if_block(exec_list *instructions)
 
    exec_list *new_kills = this->kills;
    this->kills = orig_kills;
+   ralloc_free(this->acp);
    this->acp = orig_acp;
    this->killed_all = this->killed_all || orig_killed_all;
 
    foreach_in_list(kill_entry, k, new_kills) {
       kill(k->var);
    }
+
+   ralloc_free(new_kills);
 }
 
 ir_visitor_status
@@ -269,6 +275,7 @@ ir_copy_propagation_visitor::visit_enter(ir_loop *ir)
 
    exec_list *new_kills = this->kills;
    this->kills = orig_kills;
+   ralloc_free(this->acp);
    this->acp = orig_acp;
    this->killed_all = this->killed_all || orig_killed_all;
 
@@ -276,6 +283,8 @@ ir_copy_propagation_visitor::visit_enter(ir_loop *ir)
       kill(k->var);
    }
 
+   ralloc_free(new_kills);
+
    /* already descended into the children. */
    return visit_continue_with_parent;
 }
@@ -294,7 +303,7 @@ ir_copy_propagation_visitor::kill(ir_variable *var)
 
    /* Add the LHS variable to the list of killed variables in this block.
     */
-   this->kills->push_tail(new(this->mem_ctx) kill_entry(var));
+   this->kills->push_tail(new(this->kills) kill_entry(var));
 }
 
 /**
@@ -322,7 +331,7 @@ ir_copy_propagation_visitor::add_copy(ir_assignment *ir)
 	 ir->condition = new(ralloc_parent(ir)) ir_constant(false);
 	 this->progress = true;
       } else {
-	 entry = new(this->mem_ctx) acp_entry(lhs_var, rhs_var);
+	 entry = new(this->acp) acp_entry(lhs_var, rhs_var);
 	 this->acp->push_tail(entry);
       }
    }

src/glsl/opt_copy_propagation_elements.cpp

@@ -156,6 +156,9 @@ ir_copy_propagation_elements_visitor::visit_enter(ir_function_signature *ir)
 
    visit_list_elements(this, &ir->body);
 
+   ralloc_free(this->acp);
+   ralloc_free(this->kills);
+
    this->kills = orig_kills;
    this->acp = orig_acp;
    this->killed_all = orig_killed_all;
@@ -173,9 +176,9 @@ ir_copy_propagation_elements_visitor::visit_leave(ir_assignment *ir)
       kill_entry *k;
 
       if (lhs)
-	 k = new(mem_ctx) kill_entry(var, ir->write_mask);
+	 k = new(this->kills) kill_entry(var, ir->write_mask);
       else
-	 k = new(mem_ctx) kill_entry(var, ~0);
+	 k = new(this->kills) kill_entry(var, ~0);
 
       kill(k);
    }
@@ -334,7 +337,7 @@ ir_copy_propagation_elements_visitor::handle_if_block(exec_list *instructions)
 
    /* Populate the initial acp with a copy of the original */
    foreach_in_list(acp_entry, a, orig_acp) {
-      this->acp->push_tail(new(this->mem_ctx) acp_entry(a));
+      this->acp->push_tail(new(this->acp) acp_entry(a));
    }
 
    visit_list_elements(this, instructions);
@@ -345,6 +348,7 @@ ir_copy_propagation_elements_visitor::handle_if_block(exec_list *instructions)
 
    exec_list *new_kills = this->kills;
    this->kills = orig_kills;
+   ralloc_free(this->acp);
    this->acp = orig_acp;
    this->killed_all = this->killed_all || orig_killed_all;
 
@@ -354,6 +358,8 @@ ir_copy_propagation_elements_visitor::handle_if_block(exec_list *instructions)
    foreach_in_list_safe(kill_entry, k, new_kills) {
       kill(k);
    }
+
+   ralloc_free(new_kills);
 }
 
 ir_visitor_status
@@ -391,6 +397,7 @@ ir_copy_propagation_elements_visitor::visit_enter(ir_loop *ir)
 
    exec_list *new_kills = this->kills;
    this->kills = orig_kills;
+   ralloc_free(this->acp);
    this->acp = orig_acp;
    this->killed_all = this->killed_all || orig_killed_all;
 
@@ -398,6 +405,8 @@ ir_copy_propagation_elements_visitor::visit_enter(ir_loop *ir)
       kill(k);
    }
 
+   ralloc_free(new_kills);
+
    /* already descended into the children. */
    return visit_continue_with_parent;
 }
@@ -423,6 +432,7 @@ ir_copy_propagation_elements_visitor::kill(kill_entry *k)
    if (k->next)
       k->remove();
 
+   ralloc_steal(this->kills, k);
    this->kills->push_tail(k);
 }
 

src/glx/apple/Makefile.am

@@ -3,6 +3,7 @@ EXTRA_DIST = RELEASE_NOTES
 noinst_LTLIBRARIES = libappleglx.la
 
 AM_CFLAGS = \
+	-I$(top_srcdir)/src \
 	-I$(top_srcdir)/include \
 	-I$(top_srcdir)/src/glx \
 	-I$(top_srcdir)/src/mesa \

src/glx/dri_common_query_renderer.c

@@ -65,10 +65,23 @@ dri2_convert_glx_query_renderer_attribs(int attribute)
    return -1;
 }
 
+/* Convert internal dri context profile bits into GLX context profile bits */
+static inline void
+dri_convert_context_profile_bits(int attribute, unsigned int *value)
+{
+   if (attribute == GLX_RENDERER_PREFERRED_PROFILE_MESA) {
+      if (value[0] == (1U << __DRI_API_OPENGL_CORE))
+         value[0] = GLX_CONTEXT_CORE_PROFILE_BIT_ARB;
+      else if (value[0] == (1U << __DRI_API_OPENGL))
+         value[0] = GLX_CONTEXT_COMPATIBILITY_PROFILE_BIT_ARB;
+   }
+}
+
 _X_HIDDEN int
 dri2_query_renderer_integer(struct glx_screen *base, int attribute,
                             unsigned int *value)
 {
+   int ret;
    struct dri2_screen *const psc = (struct dri2_screen *) base;
 
    /* Even though there are invalid values (and
@@ -81,8 +94,11 @@ dri2_query_renderer_integer(struct glx_screen *base, int attribute,
    if (psc->rendererQuery == NULL)
       return -1;
 
-   return psc->rendererQuery->queryInteger(psc->driScreen, dri_attribute,
-                                           value);
+   ret = psc->rendererQuery->queryInteger(psc->driScreen, dri_attribute,
+                                          value);
+   dri_convert_context_profile_bits(attribute, value);
+
+   return ret;
 }
 
 _X_HIDDEN int
@@ -108,6 +124,7 @@ _X_HIDDEN int
 dri3_query_renderer_integer(struct glx_screen *base, int attribute,
                             unsigned int *value)
 {
+   int ret;
    struct dri3_screen *const psc = (struct dri3_screen *) base;
 
    /* Even though there are invalid values (and
@@ -120,8 +137,11 @@ dri3_query_renderer_integer(struct glx_screen *base, int attribute,
    if (psc->rendererQuery == NULL)
       return -1;
 
-   return psc->rendererQuery->queryInteger(psc->driScreen, dri_attribute,
-                                           value);
+   ret = psc->rendererQuery->queryInteger(psc->driScreen, dri_attribute,
+                                          value);
+   dri_convert_context_profile_bits(attribute, value);
+
+   return ret;
 }
 
 _X_HIDDEN int
@@ -147,6 +167,7 @@ _X_HIDDEN int
 drisw_query_renderer_integer(struct glx_screen *base, int attribute,
                              unsigned int *value)
 {
+   int ret;
    struct drisw_screen *const psc = (struct drisw_screen *) base;
 
    /* Even though there are invalid values (and
@@ -159,8 +180,11 @@ drisw_query_renderer_integer(struct glx_screen *base, int attribute,
    if (psc->rendererQuery == NULL)
       return -1;
 
-   return psc->rendererQuery->queryInteger(psc->driScreen, dri_attribute,
-                                           value);
+   ret = psc->rendererQuery->queryInteger(psc->driScreen, dri_attribute,
+                                          value);
+   dri_convert_context_profile_bits(attribute, value);
+
+   return ret;
 }
 
 _X_HIDDEN int

src/glx/xfont.c

@@ -221,7 +221,10 @@ DRI_glXUseXFont(struct glx_context *CC, Font font, int first, int count, int lis
    XGCValues values;
    unsigned long valuemask;
    XFontStruct *fs;
+
+#if !defined(GLX_USE_APPLEGL)
    __GLXDRIdrawable *glxdraw;
+#endif
 
    GLint swapbytes, lsbfirst, rowlength;
    GLint skiprows, skippixels, alignment;
@@ -234,9 +237,11 @@ DRI_glXUseXFont(struct glx_context *CC, Font font, int first, int count, int lis
    dpy = CC->currentDpy;
    win = CC->currentDrawable;
 
+#if !defined(GLX_USE_APPLEGL)
    glxdraw = GetGLXDRIDrawable(CC->currentDpy, CC->currentDrawable);
    if (glxdraw)
       win = glxdraw->xDrawable;
+#endif
 
    fs = XQueryFont(dpy, font);
    if (!fs) {

src/mapi/Makefile.am

@@ -66,7 +66,7 @@ if HAVE_SHARED_GLAPI
 BUILT_SOURCES += shared-glapi/glapi_mapi_tmp.h
 
 lib_LTLIBRARIES += shared-glapi/libglapi.la
-shared_glapi_libglapi_la_SOURCES = $(MAPI_GLAPI_FILES)
+shared_glapi_libglapi_la_SOURCES = $(MAPI_GLAPI_FILES) shared-glapi/glapi_mapi_tmp.h
 shared_glapi_libglapi_la_CPPFLAGS = \
 	$(AM_CPPFLAGS) \
 	-DMAPI_MODE_GLAPI \

src/mesa/Android.gen.mk

@@ -122,5 +122,5 @@ format_info_deps := \
 	$(LOCAL_PATH)/main/format_parser.py \
 	$(FORMAT_INFO)
 
-$(intermediates)/main/format_info.c: $(format_info_deps)
+$(intermediates)/main/format_info.h: $(format_info_deps)
 	@$(MESA_PYTHON2) $(FORMAT_INFO) $< > $@

src/mesa/Makefile.am

@@ -80,7 +80,7 @@ EXTRA_DIST = \
 
 BUILT_SOURCES = \
 	main/get_hash.h \
-        main/format_info.c \
+	main/format_info.h \
 	main/git_sha1.h \
 	main/format_pack.c \
 	main/format_unpack.c \
@@ -94,13 +94,13 @@ CLEANFILES = \
 GET_HASH_GEN = main/get_hash_generator.py
 
 main/get_hash.h: ../mapi/glapi/gen/gl_and_es_API.xml main/get_hash_params.py 	\
-		 $(GET_HASH_GEN) Makefile
+		 $(GET_HASH_GEN)
 	$(AM_V_GEN)set -e;						\
 	$(PYTHON2) $(PYTHON_FLAGS) $(srcdir)/$(GET_HASH_GEN)		\
 		-f $< > $@.tmp;						\
 	mv $@.tmp $@;
 
-main/format_info.c: main/formats.csv                                    \
+main/format_info.h: main/formats.csv                                    \
                     main/format_parser.py main/format_info.py
 	$(AM_V_GEN)set -e;						\
 	$(PYTHON2) $(PYTHON_FLAGS) $(srcdir)/main/format_info.py        \
@@ -123,7 +123,7 @@ main/format_unpack.c: main/format_unpack.py main/formats.csv	\
 			$(srcdir)/main/formats.csv			\
 		| $(INDENT) $(INDENT_FLAGS) > $@;
 
-main/formats.c: main/format_info.c
+main/formats.c: main/format_info.h
 
 noinst_LTLIBRARIES = $(ARCH_LIBS)
 if NEED_LIBMESA

src/mesa/Makefile.sources

@@ -87,6 +87,7 @@ MAIN_FILES = \
 	main/ffvertex_prog.h \
 	main/fog.c \
 	main/fog.h \
+	main/format_info.h \
 	main/format_pack.h \
 	main/format_pack.c \
 	main/format_unpack.h \
@@ -101,6 +102,7 @@ MAIN_FILES = \
 	main/framebuffer.h \
 	main/get.c \
 	main/get.h \
+	main/get_hash.h \
 	main/genmipmap.c \
 	main/genmipmap.h \
 	main/getstring.c \

src/mesa/SConscript

@@ -60,7 +60,7 @@ get_hash_header = env.CodeGenerate(
 )
 
 format_info = env.CodeGenerate(
-      target = 'main/format_info.c',
+      target = 'main/format_info.h',
       script = 'main/format_info.py',
       source = 'main/formats.csv',
       command = python_cmd + ' $SCRIPT ' + ' $SOURCE > $TARGET'

src/mesa/drivers/common/meta.c

@@ -827,15 +827,18 @@ _mesa_meta_end(struct gl_context *ctx)
    const GLbitfield state = save->SavedState;
    int i;
 
-   /* After starting a new occlusion query, initialize the results to the
-    * values saved previously. The driver will then continue to increment
-    * these values.
-    */
+   /* Grab the result of the old occlusion query before starting it again. The
+    * old result is added to the result of the new query so the driver will
+    * continue adding where it left off. */
    if (state & MESA_META_OCCLUSION_QUERY) {
       if (save->CurrentOcclusionObject) {
-         _mesa_BeginQuery(save->CurrentOcclusionObject->Target,
-                          save->CurrentOcclusionObject->Id);
-         ctx->Query.CurrentOcclusionObject->Result = save->CurrentOcclusionObject->Result;
+         struct gl_query_object *q = save->CurrentOcclusionObject;
+         GLuint64EXT result;
+         if (!q->Ready)
+            ctx->Driver.WaitQuery(ctx, q);
+         result = q->Result;
+         _mesa_BeginQuery(q->Target, q->Id);
+         ctx->Query.CurrentOcclusionObject->Result += result;
       }
    }
 

src/mesa/drivers/common/meta_tex_subimage.c

@@ -44,7 +44,7 @@
 
 static struct gl_texture_image *
 create_texture_for_pbo(struct gl_context *ctx, bool create_pbo,
-                       GLenum pbo_target, int width, int height, int depth,
+                       GLenum pbo_target, int width, int height,
                        GLenum format, GLenum type, const void *pixels,
                        const struct gl_pixelstore_attrib *packing,
                        GLuint *tmp_pbo, GLuint *tmp_tex)
@@ -57,8 +57,7 @@ create_texture_for_pbo(struct gl_context *ctx, bool create_pbo,
    struct gl_texture_image *tex_image;
    bool read_only;
 
-   if ((packing->ImageHeight != 0 && packing->ImageHeight != height) ||
-       packing->SwapBytes ||
+   if (packing->SwapBytes ||
        packing->LsbFirst ||
        packing->Invert)
       return NULL;
@@ -99,14 +98,16 @@ create_texture_for_pbo(struct gl_context *ctx, bool create_pbo,
 
    _mesa_GenTextures(1, tmp_tex);
    tex_obj = _mesa_lookup_texture(ctx, *tmp_tex);
-   tex_obj->Target = depth > 1 ? GL_TEXTURE_2D_ARRAY : GL_TEXTURE_2D;
-   tex_obj->Immutable = GL_TRUE;
    _mesa_initialize_texture_object(ctx, tex_obj, *tmp_tex, GL_TEXTURE_2D);
+   /* This must be set after _mesa_initialize_texture_object, not before. */
+   tex_obj->Immutable = GL_TRUE;
+   /* This is required for interactions with ARB_texture_view. */
+   tex_obj->NumLayers = 1;
 
    internal_format = _mesa_get_format_base_format(pbo_format);
 
    tex_image = _mesa_get_tex_image(ctx, tex_obj, tex_obj->Target, 0);
-   _mesa_init_teximage_fields(ctx, tex_image, width, height, depth,
+   _mesa_init_teximage_fields(ctx, tex_image, width, height, 1,
                               0, internal_format, pbo_format);
 
    read_only = pbo_target == GL_PIXEL_UNPACK_BUFFER;
@@ -136,7 +137,7 @@ _mesa_meta_pbo_TexSubImage(struct gl_context *ctx, GLuint dims,
    struct gl_texture_image *pbo_tex_image;
    GLenum status;
    bool success = false;
-   int z;
+   int z, iters;
 
    /* XXX: This should probably be passed in from somewhere */
    const char *where = "_mesa_meta_pbo_TexSubImage";
@@ -166,9 +167,14 @@ _mesa_meta_pbo_TexSubImage(struct gl_context *ctx, GLuint dims,
       return true;
    }
 
+   /* Only accept tightly packed pixels from the user. */
+   if (packing->ImageHeight != 0 && packing->ImageHeight != height)
+      return false;
+
+   /* For arrays, use a tall (height * depth) 2D texture. */
    pbo_tex_image = create_texture_for_pbo(ctx, create_pbo,
                                           GL_PIXEL_UNPACK_BUFFER,
-                                          width, height, depth,
+                                          width, height * depth,
                                           format, type, pixels, packing,
                                           &pbo, &pbo_tex);
    if (!pbo_tex_image)
@@ -184,12 +190,6 @@ _mesa_meta_pbo_TexSubImage(struct gl_context *ctx, GLuint dims,
    _mesa_BindFramebuffer(GL_READ_FRAMEBUFFER, fbos[0]);
    _mesa_BindFramebuffer(GL_DRAW_FRAMEBUFFER, fbos[1]);
 
-   if (tex_image->TexObject->Target == GL_TEXTURE_1D_ARRAY) {
-      assert(depth == 1);
-      depth = height;
-      height = 1;
-   }
-
    _mesa_meta_bind_fbo_image(GL_READ_FRAMEBUFFER, GL_COLOR_ATTACHMENT0,
                              pbo_tex_image, 0);
    /* If this passes on the first layer it should pass on the others */
@@ -213,7 +213,10 @@ _mesa_meta_pbo_TexSubImage(struct gl_context *ctx, GLuint dims,
                                   GL_COLOR_BUFFER_BIT, GL_NEAREST))
       goto fail;
 
-   for (z = 1; z < depth; z++) {
+   iters = tex_image->TexObject->Target == GL_TEXTURE_1D_ARRAY ?
+           height : depth;
+
+   for (z = 1; z < iters; z++) {
       _mesa_meta_bind_fbo_image(GL_READ_FRAMEBUFFER, GL_COLOR_ATTACHMENT0,
                                 pbo_tex_image, z);
       _mesa_meta_bind_fbo_image(GL_DRAW_FRAMEBUFFER, GL_COLOR_ATTACHMENT0,
@@ -221,11 +224,18 @@ _mesa_meta_pbo_TexSubImage(struct gl_context *ctx, GLuint dims,
 
       _mesa_update_state(ctx);
 
-      _mesa_meta_BlitFramebuffer(ctx, ctx->ReadBuffer, ctx->DrawBuffer,
-                                 0, 0, width, height,
-                                 xoffset, yoffset,
-                                 xoffset + width, yoffset + height,
-                                 GL_COLOR_BUFFER_BIT, GL_NEAREST);
+      if (tex_image->TexObject->Target == GL_TEXTURE_1D_ARRAY)
+         _mesa_meta_BlitFramebuffer(ctx, ctx->ReadBuffer, ctx->DrawBuffer,
+                                    0, z, width, z + 1,
+                                    xoffset, yoffset,
+                                    xoffset + width, yoffset + 1,
+                                    GL_COLOR_BUFFER_BIT, GL_NEAREST);
+      else
+         _mesa_meta_BlitFramebuffer(ctx, ctx->ReadBuffer, ctx->DrawBuffer,
+                                    0, z * height, width, (z + 1) * height,
+                                    xoffset, yoffset,
+                                    xoffset + width, yoffset + height,
+                                    GL_COLOR_BUFFER_BIT, GL_NEAREST);
    }
 
    success = true;
@@ -252,7 +262,7 @@ _mesa_meta_pbo_GetTexSubImage(struct gl_context *ctx, GLuint dims,
    struct gl_texture_image *pbo_tex_image;
    GLenum status;
    bool success = false;
-   int z;
+   int z, iters;
 
    /* XXX: This should probably be passed in from somewhere */
    const char *where = "_mesa_meta_pbo_GetTexSubImage";
@@ -282,8 +292,13 @@ _mesa_meta_pbo_GetTexSubImage(struct gl_context *ctx, GLuint dims,
       return true;
    }
 
+   /* Only accept tightly packed pixels from the user. */
+   if (packing->ImageHeight != 0 && packing->ImageHeight != height)
+      return false;
+
+   /* For arrays, use a tall (height * depth) 2D texture. */
    pbo_tex_image = create_texture_for_pbo(ctx, false, GL_PIXEL_PACK_BUFFER,
-                                          width, height, depth,
+                                          width, height * depth,
                                           format, type, pixels, packing,
                                           &pbo, &pbo_tex);
    if (!pbo_tex_image)
@@ -294,12 +309,6 @@ _mesa_meta_pbo_GetTexSubImage(struct gl_context *ctx, GLuint dims,
 
    _mesa_GenFramebuffers(2, fbos);
 
-   if (tex_image && tex_image->TexObject->Target == GL_TEXTURE_1D_ARRAY) {
-      assert(depth == 1);
-      depth = height;
-      height = 1;
-   }
-
    /* If we were given a texture, bind it to the read framebuffer.  If not,
     * we're doing a ReadPixels and we should just use whatever framebuffer
     * the client has bound.
@@ -333,7 +342,12 @@ _mesa_meta_pbo_GetTexSubImage(struct gl_context *ctx, GLuint dims,
                                   GL_COLOR_BUFFER_BIT, GL_NEAREST))
       goto fail;
 
-   for (z = 1; z < depth; z++) {
+   if (tex_image && tex_image->TexObject->Target == GL_TEXTURE_1D_ARRAY)
+      iters = height;
+   else
+      iters = depth;
+
+   for (z = 1; z < iters; z++) {
       _mesa_meta_bind_fbo_image(GL_READ_FRAMEBUFFER, GL_COLOR_ATTACHMENT0,
                                 tex_image, zoffset + z);
       _mesa_meta_bind_fbo_image(GL_DRAW_FRAMEBUFFER, GL_COLOR_ATTACHMENT0,
@@ -341,11 +355,18 @@ _mesa_meta_pbo_GetTexSubImage(struct gl_context *ctx, GLuint dims,
 
       _mesa_update_state(ctx);
 
-      _mesa_meta_BlitFramebuffer(ctx, ctx->ReadBuffer, ctx->DrawBuffer,
-                                 xoffset, yoffset,
-                                 xoffset + width, yoffset + height,
-                                 0, 0, width, height,
-                                 GL_COLOR_BUFFER_BIT, GL_NEAREST);
+      if (tex_image->TexObject->Target == GL_TEXTURE_1D_ARRAY)
+         _mesa_meta_BlitFramebuffer(ctx, ctx->ReadBuffer, ctx->DrawBuffer,
+                                    xoffset, yoffset,
+                                    xoffset + width, yoffset + 1,
+                                    0, z, width, z + 1,
+                                    GL_COLOR_BUFFER_BIT, GL_NEAREST);
+      else
+         _mesa_meta_BlitFramebuffer(ctx, ctx->ReadBuffer, ctx->DrawBuffer,
+                                    xoffset, yoffset,
+                                    xoffset + width, yoffset + height,
+                                    0, z * height, width, (z + 1) * height,
+                                    GL_COLOR_BUFFER_BIT, GL_NEAREST);
    }
 
    success = true;

src/mesa/drivers/dri/common/xmlpool/Makefile.am

@@ -52,7 +52,7 @@ POT=xmlpool.pot
 
 .PHONY: all clean pot po mo
 
-EXTRA_DIST = gen_xmlpool.py t_options.h $(POS) SConscript
+EXTRA_DIST = gen_xmlpool.py options.h t_options.h $(POS) SConscript
 BUILT_SOURCES = options.h
 CLEANFILES = $(MOS) options.h
 

src/mesa/drivers/dri/i965/Makefile.am

@@ -54,6 +54,7 @@ TEST_LIBS = \
 
 TESTS = \
 	test_fs_cmod_propagation \
+	test_fs_saturate_propagation \
         test_eu_compact \
 	test_vf_float_conversions \
         test_vec4_copy_propagation \
@@ -64,26 +65,32 @@ check_PROGRAMS = $(TESTS)
 test_fs_cmod_propagation_SOURCES = \
 	test_fs_cmod_propagation.cpp
 test_fs_cmod_propagation_LDADD = \
-	$(TEST_LIBS) \
-	$(top_builddir)/src/gtest/libgtest.la
+	$(top_builddir)/src/gtest/libgtest.la \
+	$(TEST_LIBS)
+
+test_fs_saturate_propagation_SOURCES = \
+	test_fs_saturate_propagation.cpp
+test_fs_saturate_propagation_LDADD = \
+	$(top_builddir)/src/gtest/libgtest.la \
+	$(TEST_LIBS)
 
 test_vf_float_conversions_SOURCES = \
 	test_vf_float_conversions.cpp
 test_vf_float_conversions_LDADD = \
-	$(TEST_LIBS) \
-	$(top_builddir)/src/gtest/libgtest.la
+	$(top_builddir)/src/gtest/libgtest.la \
+	$(TEST_LIBS)
 
 test_vec4_register_coalesce_SOURCES = \
 	test_vec4_register_coalesce.cpp
 test_vec4_register_coalesce_LDADD = \
-        $(TEST_LIBS) \
-        $(top_builddir)/src/gtest/libgtest.la
+        $(top_builddir)/src/gtest/libgtest.la \
+        $(TEST_LIBS)
 
 test_vec4_copy_propagation_SOURCES = \
 	test_vec4_copy_propagation.cpp
 test_vec4_copy_propagation_LDADD = \
-        $(TEST_LIBS) \
-        $(top_builddir)/src/gtest/libgtest.la
+        $(top_builddir)/src/gtest/libgtest.la \
+        $(TEST_LIBS)
 
 test_eu_compact_SOURCES = \
 	test_eu_compact.c

src/mesa/drivers/dri/i965/brw_defines.h

@@ -551,6 +551,7 @@
 #define BRW_SURFACE_PITCH_MASK		INTEL_MASK(19, 3)
 #define BRW_SURFACE_TILED		(1 << 1)
 #define BRW_SURFACE_TILED_Y		(1 << 0)
+#define HSW_SURFACE_IS_INTEGER_FORMAT   (1 << 18)
 
 /* Surface state DW4 */
 #define BRW_SURFACE_MIN_LOD_SHIFT	28
@@ -910,6 +911,7 @@ enum opcode {
 
    SHADER_OPCODE_URB_WRITE_SIMD8,
 
+   VEC4_OPCODE_MOV_BYTES,
    VEC4_OPCODE_PACK_BYTES,
    VEC4_OPCODE_UNPACK_UNIFORM,
 

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

@@ -64,10 +64,10 @@ opt_saturate_propagation_local(fs_visitor *v, bblock_t *block)
 
       bool interfered = false;
       foreach_inst_in_block_reverse_starting_from(fs_inst, scan_inst, inst, block) {
-         if (scan_inst->dst.file == GRF &&
-             scan_inst->dst.reg == inst->src[0].reg &&
-             scan_inst->dst.reg_offset == inst->src[0].reg_offset &&
-             !scan_inst->is_partial_write()) {
+         if (scan_inst->overwrites_reg(inst->src[0])) {
+            if (scan_inst->is_partial_write())
+               break;
+
             if (scan_inst->saturate) {
                inst->saturate = false;
                progress = true;
@@ -81,12 +81,16 @@ opt_saturate_propagation_local(fs_visitor *v, bblock_t *block)
             break;
          }
          for (int i = 0; i < scan_inst->sources; i++) {
-            if ((scan_inst->opcode != BRW_OPCODE_MOV || !scan_inst->saturate) &&
-                scan_inst->src[i].file == GRF &&
+            if (scan_inst->src[i].file == GRF &&
                 scan_inst->src[i].reg == inst->src[0].reg &&
                 scan_inst->src[i].reg_offset == inst->src[0].reg_offset) {
-               interfered = true;
-               break;
+               if (scan_inst->opcode != BRW_OPCODE_MOV ||
+                   !scan_inst->saturate ||
+                   scan_inst->src[0].abs ||
+                   scan_inst->src[0].negate) {
+                  interfered = true;
+                  break;
+               }
             }
          }
 

src/mesa/drivers/dri/i965/brw_sampler_state.c

@@ -208,7 +208,7 @@ upload_default_color(struct brw_context *brw,
    struct gl_texture_unit *texUnit = &ctx->Texture.Unit[unit];
    struct gl_texture_object *texObj = texUnit->_Current;
    struct gl_texture_image *firstImage = texObj->Image[0][texObj->BaseLevel];
-   float color[4];
+   union gl_color_union color;
 
    switch (firstImage->_BaseFormat) {
    case GL_DEPTH_COMPONENT:
@@ -216,40 +216,40 @@ upload_default_color(struct brw_context *brw,
        * R channel, while the hardware uses A.  Spam R into all the
        * channels for safety.
        */
-      color[0] = sampler->BorderColor.f[0];
-      color[1] = sampler->BorderColor.f[0];
-      color[2] = sampler->BorderColor.f[0];
-      color[3] = sampler->BorderColor.f[0];
+      color.ui[0] = sampler->BorderColor.ui[0];
+      color.ui[1] = sampler->BorderColor.ui[0];
+      color.ui[2] = sampler->BorderColor.ui[0];
+      color.ui[3] = sampler->BorderColor.ui[0];
       break;
    case GL_ALPHA:
-      color[0] = 0.0;
-      color[1] = 0.0;
-      color[2] = 0.0;
-      color[3] = sampler->BorderColor.f[3];
+      color.ui[0] = 0u;
+      color.ui[1] = 0u;
+      color.ui[2] = 0u;
+      color.ui[3] = sampler->BorderColor.ui[3];
       break;
    case GL_INTENSITY:
-      color[0] = sampler->BorderColor.f[0];
-      color[1] = sampler->BorderColor.f[0];
-      color[2] = sampler->BorderColor.f[0];
-      color[3] = sampler->BorderColor.f[0];
+      color.ui[0] = sampler->BorderColor.ui[0];
+      color.ui[1] = sampler->BorderColor.ui[0];
+      color.ui[2] = sampler->BorderColor.ui[0];
+      color.ui[3] = sampler->BorderColor.ui[0];
       break;
    case GL_LUMINANCE:
-      color[0] = sampler->BorderColor.f[0];
-      color[1] = sampler->BorderColor.f[0];
-      color[2] = sampler->BorderColor.f[0];
-      color[3] = 1.0;
+      color.ui[0] = sampler->BorderColor.ui[0];
+      color.ui[1] = sampler->BorderColor.ui[0];
+      color.ui[2] = sampler->BorderColor.ui[0];
+      color.ui[3] = float_as_int(1.0);
       break;
    case GL_LUMINANCE_ALPHA:
-      color[0] = sampler->BorderColor.f[0];
-      color[1] = sampler->BorderColor.f[0];
-      color[2] = sampler->BorderColor.f[0];
-      color[3] = sampler->BorderColor.f[3];
+      color.ui[0] = sampler->BorderColor.ui[0];
+      color.ui[1] = sampler->BorderColor.ui[0];
+      color.ui[2] = sampler->BorderColor.ui[0];
+      color.ui[3] = sampler->BorderColor.ui[3];
       break;
    default:
-      color[0] = sampler->BorderColor.f[0];
-      color[1] = sampler->BorderColor.f[1];
-      color[2] = sampler->BorderColor.f[2];
-      color[3] = sampler->BorderColor.f[3];
+      color.ui[0] = sampler->BorderColor.ui[0];
+      color.ui[1] = sampler->BorderColor.ui[1];
+      color.ui[2] = sampler->BorderColor.ui[2];
+      color.ui[3] = sampler->BorderColor.ui[3];
       break;
    }
 
@@ -258,18 +258,79 @@ upload_default_color(struct brw_context *brw,
     * the border color alpha to 1.0 in that case.
     */
    if (firstImage->_BaseFormat == GL_RGB)
-      color[3] = 1.0;
+      color.ui[3] = float_as_int(1.0);
 
    if (brw->gen >= 8) {
       /* On Broadwell, the border color is represented as four 32-bit floats,
        * integers, or unsigned values, interpreted according to the surface
-       * format.  This matches the sampler->BorderColor union exactly.  Since
-       * we use floats both here and in the above reswizzling code, we preserve
-       * the original bit pattern.  So we actually handle all three formats.
+       * format.  This matches the sampler->BorderColor union exactly; just
+       * memcpy the values.
        */
-      float *sdc = brw_state_batch(brw, AUB_TRACE_SAMPLER_DEFAULT_COLOR,
-                                   4 * 4, 64, sdc_offset);
-      COPY_4FV(sdc, color);
+      uint32_t *sdc = brw_state_batch(brw, AUB_TRACE_SAMPLER_DEFAULT_COLOR,
+                                      4 * 4, 64, sdc_offset);
+      memcpy(sdc, color.ui, 4 * 4);
+   } else if (brw->is_haswell && texObj->_IsIntegerFormat) {
+      /* Haswell's integer border color support is completely insane:
+       * SAMPLER_BORDER_COLOR_STATE is 20 DWords.  The first four are
+       * for float colors.  The next 12 DWords are MBZ and only exist to
+       * pad it out to a 64 byte cacheline boundary.  DWords 16-19 then
+       * contain integer colors; these are only used if SURFACE_STATE
+       * has the "Integer Surface Format" bit set.  Even then, the
+       * arrangement of the RGBA data devolves into madness.
+       */
+      uint32_t *sdc = brw_state_batch(brw, AUB_TRACE_SAMPLER_DEFAULT_COLOR,
+                                      20 * 4, 512, sdc_offset);
+      memset(sdc, 0, 20 * 4);
+      sdc = &sdc[16];
+
+      mesa_format format = firstImage->TexFormat;
+      int bits_per_channel = _mesa_get_format_bits(format, GL_RED_BITS);
+
+      /* From the Haswell PRM, "Command Reference: Structures", Page 36:
+       * "If any color channel is missing from the surface format,
+       *  corresponding border color should be programmed as zero and if
+       *  alpha channel is missing, corresponding Alpha border color should
+       *  be programmed as 1."
+       */
+      unsigned c[4] = { 0, 0, 0, 1 };
+      for (int i = 0; i < 4; i++) {
+         if (_mesa_format_has_color_component(format, i))
+            c[i] = color.ui[i];
+      }
+
+      switch (bits_per_channel) {
+      case 8:
+         /* Copy RGBA in order. */
+         for (int i = 0; i < 4; i++)
+            ((uint8_t *) sdc)[i] = c[i];
+         break;
+      case 10:
+         /* R10G10B10A2_UINT is treated like a 16-bit format. */
+      case 16:
+         ((uint16_t *) sdc)[0] = c[0]; /* R -> DWord 0, bits 15:0  */
+         ((uint16_t *) sdc)[1] = c[1]; /* G -> DWord 0, bits 31:16 */
+         /* DWord 1 is Reserved/MBZ! */
+         ((uint16_t *) sdc)[4] = c[2]; /* B -> DWord 2, bits 15:0  */
+         ((uint16_t *) sdc)[5] = c[3]; /* A -> DWord 3, bits 31:16 */
+         break;
+      case 32:
+         if (firstImage->_BaseFormat == GL_RG) {
+            /* Careful inspection of the tables reveals that for RG32 formats,
+             * the green channel needs to go where blue normally belongs.
+             */
+            sdc[0] = c[0];
+            sdc[2] = c[1];
+            sdc[3] = 1;
+         } else {
+            /* Copy RGBA in order. */
+            for (int i = 0; i < 4; i++)
+               sdc[i] = c[i];
+         }
+         break;
+      default:
+         assert(!"Invalid number of bits per channel in integer format.");
+         break;
+      }
    } else if (brw->gen == 5 || brw->gen == 6) {
       struct gen5_sampler_default_color *sdc;
 
@@ -278,39 +339,39 @@ upload_default_color(struct brw_context *brw,
 
       memset(sdc, 0, sizeof(*sdc));
 
-      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[0], color[0]);
-      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[1], color[1]);
-      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[2], color[2]);
-      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[3], color[3]);
+      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[0], color.f[0]);
+      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[1], color.f[1]);
+      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[2], color.f[2]);
+      UNCLAMPED_FLOAT_TO_UBYTE(sdc->ub[3], color.f[3]);
 
-      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[0], color[0]);
-      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[1], color[1]);
-      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[2], color[2]);
-      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[3], color[3]);
+      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[0], color.f[0]);
+      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[1], color.f[1]);
+      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[2], color.f[2]);
+      UNCLAMPED_FLOAT_TO_USHORT(sdc->us[3], color.f[3]);
 
-      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[0], color[0]);
-      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[1], color[1]);
-      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[2], color[2]);
-      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[3], color[3]);
+      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[0], color.f[0]);
+      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[1], color.f[1]);
+      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[2], color.f[2]);
+      UNCLAMPED_FLOAT_TO_SHORT(sdc->s[3], color.f[3]);
 
-      sdc->hf[0] = _mesa_float_to_half(color[0]);
-      sdc->hf[1] = _mesa_float_to_half(color[1]);
-      sdc->hf[2] = _mesa_float_to_half(color[2]);
-      sdc->hf[3] = _mesa_float_to_half(color[3]);
+      sdc->hf[0] = _mesa_float_to_half(color.f[0]);
+      sdc->hf[1] = _mesa_float_to_half(color.f[1]);
+      sdc->hf[2] = _mesa_float_to_half(color.f[2]);
+      sdc->hf[3] = _mesa_float_to_half(color.f[3]);
 
       sdc->b[0] = sdc->s[0] >> 8;
       sdc->b[1] = sdc->s[1] >> 8;
       sdc->b[2] = sdc->s[2] >> 8;
       sdc->b[3] = sdc->s[3] >> 8;
 
-      sdc->f[0] = color[0];
-      sdc->f[1] = color[1];
-      sdc->f[2] = color[2];
-      sdc->f[3] = color[3];
+      sdc->f[0] = color.f[0];
+      sdc->f[1] = color.f[1];
+      sdc->f[2] = color.f[2];
+      sdc->f[3] = color.f[3];
    } else {
       float *sdc = brw_state_batch(brw, AUB_TRACE_SAMPLER_DEFAULT_COLOR,
 			           4 * 4, 32, sdc_offset);
-      memcpy(sdc, color, 4 * 4);
+      memcpy(sdc, color.f, 4 * 4);
    }
 }
 

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

@@ -470,6 +470,8 @@ brw_instruction_name(enum opcode op)
    case SHADER_OPCODE_URB_WRITE_SIMD8:
       return "gen8_urb_write_simd8";
 
+   case VEC4_OPCODE_MOV_BYTES:
+      return "mov_bytes";
    case VEC4_OPCODE_PACK_BYTES:
       return "pack_bytes";
    case VEC4_OPCODE_UNPACK_UNIFORM:

src/mesa/drivers/dri/i965/brw_vec4_generator.cpp

@@ -1512,6 +1512,22 @@ vec4_generator::generate_code(const cfg_t *cfg)
          generate_unpack_flags(inst, dst);
          break;
 
+      case VEC4_OPCODE_MOV_BYTES: {
+         /* Moves the low byte from each channel, using an Align1 access mode
+          * and a <4,1,0> source region.
+          */
+         assert(src[0].type == BRW_REGISTER_TYPE_UB ||
+                src[0].type == BRW_REGISTER_TYPE_B);
+
+         brw_set_default_access_mode(p, BRW_ALIGN_1);
+         src[0].vstride = BRW_VERTICAL_STRIDE_4;
+         src[0].width = BRW_WIDTH_1;
+         src[0].hstride = BRW_HORIZONTAL_STRIDE_0;
+         brw_MOV(p, dst, src[0]);
+         brw_set_default_access_mode(p, BRW_ALIGN_16);
+         break;
+      }
+
       case VEC4_OPCODE_PACK_BYTES: {
          /* Is effectively:
           *

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

@@ -485,7 +485,7 @@ vec4_visitor::emit_unpack_unorm_4x8(const dst_reg &dst, src_reg src0)
 
    shifted.type = BRW_REGISTER_TYPE_UB;
    dst_reg f(this, glsl_type::vec4_type);
-   emit(MOV(f, src_reg(shifted)));
+   emit(VEC4_OPCODE_MOV_BYTES, f, src_reg(shifted));
 
    emit(MUL(dst, src_reg(f), src_reg(1.0f / 255.0f)));
 }
@@ -507,7 +507,7 @@ vec4_visitor::emit_unpack_snorm_4x8(const dst_reg &dst, src_reg src0)
 
    shifted.type = BRW_REGISTER_TYPE_B;
    dst_reg f(this, glsl_type::vec4_type);
-   emit(MOV(f, src_reg(shifted)));
+   emit(VEC4_OPCODE_MOV_BYTES, f, src_reg(shifted));
 
    dst_reg scaled(this, glsl_type::vec4_type);
    emit(MUL(scaled, src_reg(f), src_reg(1.0f / 127.0f)));

src/mesa/drivers/dri/i965/brw_vec4_vp.cpp

@@ -43,8 +43,7 @@ vec4_visitor::emit_vp_sop(enum brw_conditional_mod conditional_mod,
 {
    vec4_instruction *inst;
 
-   inst = emit(BRW_OPCODE_CMP, dst_null_d(), src0, src1);
-   inst->conditional_mod = conditional_mod;
+   inst = emit(CMP(dst_null_f(), src0, src1, conditional_mod));
 
    inst = emit(BRW_OPCODE_SEL, dst, one, src_reg(0.0f));
    inst->predicate = BRW_PREDICATE_NORMAL;

src/mesa/drivers/dri/i965/brw_wm_surface_state.c

@@ -199,6 +199,14 @@ brw_get_texture_swizzle(const struct gl_context *ctx,
       swizzles[1] = SWIZZLE_ZERO;
       swizzles[2] = SWIZZLE_ZERO;
       break;
+   case GL_LUMINANCE:
+      if (t->_IsIntegerFormat) {
+         swizzles[0] = SWIZZLE_X;
+         swizzles[1] = SWIZZLE_X;
+         swizzles[2] = SWIZZLE_X;
+         swizzles[3] = SWIZZLE_ONE;
+      }
+      break;
    case GL_RED:
    case GL_RG:
    case GL_RGB:

src/mesa/drivers/dri/i965/gen7_wm_surface_state.c

@@ -321,6 +321,9 @@ gen7_update_texture_surface(struct gl_context *ctx,
    surf[3] = SET_FIELD(effective_depth - 1, BRW_SURFACE_DEPTH) |
              (mt->pitch - 1);
 
+   if (brw->is_haswell && tObj->_IsIntegerFormat)
+      surf[3] |= HSW_SURFACE_IS_INTEGER_FORMAT;
+
    surf[4] = gen7_surface_msaa_bits(mt->num_samples, mt->msaa_layout) |
              SET_FIELD(tObj->MinLayer, GEN7_SURFACE_MIN_ARRAY_ELEMENT) |
              SET_FIELD((effective_depth - 1),

src/mesa/drivers/dri/i965/gen8_depth_state.c

@@ -190,6 +190,18 @@ gen8_emit_depth_stencil_hiz(struct brw_context *brw,
    case GL_TEXTURE_3D:
       assert(mt);
       depth = MAX2(mt->logical_depth0, 1);
+      surftype = translate_tex_target(gl_target);
+      break;
+   case GL_TEXTURE_1D_ARRAY:
+   case GL_TEXTURE_1D:
+      if (brw->gen >= 9) {
+         /* WaDisable1DDepthStencil. Skylake+ doesn't support 1D depth
+          * textures but it does allow pretending it's a 2D texture
+          * instead.
+          */
+         surftype = BRW_SURFACE_2D;
+         break;
+      }
       /* fallthrough */
    default:
       surftype = translate_tex_target(gl_target);

src/mesa/drivers/dri/i965/intel_fbo.c

@@ -885,6 +885,13 @@ intel_blit_framebuffer(struct gl_context *ctx,
    if (mask == 0x0)
       return;
 
+   mask = _mesa_meta_BlitFramebuffer(ctx, readFb, drawFb,
+                                     srcX0, srcY0, srcX1, srcY1,
+                                     dstX0, dstY0, dstX1, dstY1,
+                                     mask, filter);
+   if (mask == 0x0)
+      return;
+
    if (brw->gen >= 8 && (mask & GL_STENCIL_BUFFER_BIT)) {
       brw_meta_fbo_stencil_blit(brw_context(ctx), readFb, drawFb,
                                 srcX0, srcY0, srcX1, srcY1,
@@ -902,13 +909,6 @@ intel_blit_framebuffer(struct gl_context *ctx,
    if (mask == 0x0)
       return;
 
-   mask = _mesa_meta_BlitFramebuffer(ctx, readFb, drawFb,
-                                     srcX0, srcY0, srcX1, srcY1,
-                                     dstX0, dstY0, dstX1, dstY1,
-                                     mask, filter);
-   if (mask == 0x0)
-      return;
-
    _swrast_BlitFramebuffer(ctx, readFb, drawFb,
                            srcX0, srcY0, srcX1, srcY1,
                            dstX0, dstY0, dstX1, dstY1,

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

@@ -0,0 +1,439 @@
+/*
+ * Copyright © 2015 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <gtest/gtest.h>
+#include "brw_fs.h"
+#include "brw_cfg.h"
+#include "program/program.h"
+
+class saturate_propagation_test : public ::testing::Test {
+   virtual void SetUp();
+
+public:
+   struct brw_context *brw;
+   struct gl_context *ctx;
+   struct brw_wm_prog_data *prog_data;
+   struct gl_shader_program *shader_prog;
+   struct brw_fragment_program *fp;
+   fs_visitor *v;
+};
+
+class saturate_propagation_fs_visitor : public fs_visitor
+{
+public:
+   saturate_propagation_fs_visitor(struct brw_context *brw,
+                                   struct brw_wm_prog_data *prog_data,
+                                   struct gl_shader_program *shader_prog)
+      : fs_visitor(brw, NULL, NULL, prog_data, shader_prog, NULL, 8) {}
+};
+
+
+void saturate_propagation_test::SetUp()
+{
+   brw = (struct brw_context *)calloc(1, sizeof(*brw));
+   ctx = &brw->ctx;
+
+   fp = ralloc(NULL, struct brw_fragment_program);
+   prog_data = ralloc(NULL, struct brw_wm_prog_data);
+   shader_prog = ralloc(NULL, struct gl_shader_program);
+
+   v = new saturate_propagation_fs_visitor(brw, prog_data, shader_prog);
+
+   _mesa_init_fragment_program(ctx, &fp->program, GL_FRAGMENT_SHADER, 0);
+
+   brw->gen = 4;
+}
+
+static fs_inst *
+instruction(bblock_t *block, int num)
+{
+   fs_inst *inst = (fs_inst *)block->start();
+   for (int i = 0; i < num; i++) {
+      inst = (fs_inst *)inst->next;
+   }
+   return inst;
+}
+
+static bool
+saturate_propagation(fs_visitor *v)
+{
+   const bool print = false;
+
+   if (print) {
+      fprintf(stderr, "= Before =\n");
+      v->cfg->dump(v);
+   }
+
+   bool ret = v->opt_saturate_propagation();
+
+   if (print) {
+      fprintf(stderr, "\n= After =\n");
+      v->cfg->dump(v);
+   }
+
+   return ret;
+}
+
+TEST_F(saturate_propagation_test, basic)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_ADD, dst0, src0, src1);
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+
+   /* = Before =
+    *
+    * 0: add(8)        dst0  src0  src1
+    * 1: mov.sat(8)    dst1  dst0
+    *
+    * = After =
+    * 0: add.sat(8)    dst0  src0  src1
+    * 1: mov(8)        dst1  dst0
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+
+   EXPECT_TRUE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_TRUE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_FALSE(instruction(block0, 1)->saturate);
+}
+
+TEST_F(saturate_propagation_test, other_non_saturated_use)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg dst2 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_ADD, dst0, src0, src1);
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+   v->emit(BRW_OPCODE_ADD, dst2, dst0, src0);
+
+   /* = Before =
+    *
+    * 0: add(8)        dst0  src0  src1
+    * 1: mov.sat(8)    dst1  dst0
+    * 2: add(8)        dst2  dst0  src0
+    *
+    * = After =
+    * (no changes)
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+
+   EXPECT_FALSE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_FALSE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_TRUE(instruction(block0, 1)->saturate);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 2)->opcode);
+}
+
+TEST_F(saturate_propagation_test, predicated_instruction)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_ADD, dst0, src0, src1)
+      ->predicate = BRW_PREDICATE_NORMAL;
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+
+   /* = Before =
+    *
+    * 0: (+f0) add(8)  dst0  src0  src1
+    * 1: mov.sat(8)    dst1  dst0
+    *
+    * = After =
+    * (no changes)
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+
+   EXPECT_FALSE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_FALSE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_TRUE(instruction(block0, 1)->saturate);
+}
+
+TEST_F(saturate_propagation_test, neg_mov_sat)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_ADD, dst0, src0, src1);
+   dst0.negate = true;
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+
+   /* = Before =
+    *
+    * 0: add(8)        dst0  src0  src1
+    * 1: mov.sat(8)    dst1  -dst0
+    *
+    * = After =
+    * (no changes)
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+
+   EXPECT_FALSE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_FALSE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_TRUE(instruction(block0, 1)->saturate);
+}
+
+TEST_F(saturate_propagation_test, abs_mov_sat)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_ADD, dst0, src0, src1);
+   dst0.abs = true;
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+
+   /* = Before =
+    *
+    * 0: add(8)        dst0  src0  src1
+    * 1: mov.sat(8)    dst1  (abs)dst0
+    *
+    * = After =
+    * (no changes)
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+
+   EXPECT_FALSE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(1, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_FALSE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_TRUE(instruction(block0, 1)->saturate);
+}
+
+TEST_F(saturate_propagation_test, producer_saturates)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg dst2 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_ADD, dst0, src0, src1)
+      ->saturate = true;
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+   v->emit(BRW_OPCODE_MOV, dst2, dst0);
+
+   /* = Before =
+    *
+    * 0: add.sat(8)    dst0  src0  src1
+    * 1: mov.sat(8)    dst1  dst0
+    * 2: mov(8)        dst2  dst0
+    *
+    * = After =
+    * 0: add.sat(8)    dst0  src0  src1
+    * 1: mov(8)        dst1  dst0
+    * 2: mov(8)        dst2  dst0
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+
+   EXPECT_TRUE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_TRUE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_FALSE(instruction(block0, 1)->saturate);
+}
+
+TEST_F(saturate_propagation_test, intervening_saturating_copy)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg dst2 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_ADD, dst0, src0, src1);
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+   v->emit(BRW_OPCODE_MOV, dst2, dst0)
+      ->saturate = true;
+
+   /* = Before =
+    *
+    * 0: add(8)    dst0  src0  src1
+    * 1: mov.sat(8)    dst1  dst0
+    * 2: mov.sat(8)    dst2  dst0
+    *
+    * = After =
+    * 0: add.sat(8)    dst0  src0  src1
+    * 1: mov(8)        dst1  dst0
+    * 2: mov(8)        dst2  dst0
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+
+   EXPECT_TRUE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_TRUE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_FALSE(instruction(block0, 1)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 2)->opcode);
+   EXPECT_FALSE(instruction(block0, 2)->saturate);
+}
+
+TEST_F(saturate_propagation_test, intervening_dest_write)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::vec4_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   fs_reg src2 = v->vgrf(glsl_type::vec2_type);
+   v->emit(BRW_OPCODE_ADD, offset(dst0, 2), src0, src1);
+   v->emit(SHADER_OPCODE_TEX, dst0, src2)
+      ->regs_written = 4;
+   v->emit(BRW_OPCODE_MOV, dst1, offset(dst0, 2))
+      ->saturate = true;
+
+   /* = Before =
+    *
+    * 0: add(8)        dst0+2  src0    src1
+    * 1: tex(8) rlen 4 dst0+0  src2
+    * 2: mov.sat(8)    dst1    dst0+2
+    *
+    * = After =
+    * (no changes)
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+
+   EXPECT_FALSE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode);
+   EXPECT_FALSE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(SHADER_OPCODE_TEX, instruction(block0, 1)->opcode);
+   EXPECT_FALSE(instruction(block0, 0)->saturate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 2)->opcode);
+   EXPECT_TRUE(instruction(block0, 2)->saturate);
+}
+
+TEST_F(saturate_propagation_test, mul_neg_mov_sat_mov_sat)
+{
+   fs_reg dst0 = v->vgrf(glsl_type::float_type);
+   fs_reg dst1 = v->vgrf(glsl_type::float_type);
+   fs_reg dst2 = v->vgrf(glsl_type::float_type);
+   fs_reg src0 = v->vgrf(glsl_type::float_type);
+   fs_reg src1 = v->vgrf(glsl_type::float_type);
+   v->emit(BRW_OPCODE_MUL, dst0, src0, src1);
+   dst0.negate = true;
+   v->emit(BRW_OPCODE_MOV, dst1, dst0)
+      ->saturate = true;
+   dst0.negate = false;
+   v->emit(BRW_OPCODE_MOV, dst2, dst0)
+      ->saturate = true;
+
+   /* = Before =
+    *
+    * 0: mul(8)        dst0  src0  src1
+    * 1: mov.sat(8)    dst1  -dst0
+    * 2: mov.sat(8)    dst2  dst0
+    *
+    * = After =
+    * (no changes)
+    */
+
+   v->calculate_cfg();
+   bblock_t *block0 = v->cfg->blocks[0];
+
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+
+   EXPECT_FALSE(saturate_propagation(v));
+   EXPECT_EQ(0, block0->start_ip);
+   EXPECT_EQ(2, block0->end_ip);
+   EXPECT_EQ(BRW_OPCODE_MUL, instruction(block0, 0)->opcode);
+   EXPECT_FALSE(instruction(block0, 0)->saturate);
+   EXPECT_FALSE(instruction(block0, 0)->src[1].negate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode);
+   EXPECT_TRUE(instruction(block0, 1)->saturate);
+   EXPECT_TRUE(instruction(block0, 1)->src[0].negate);
+   EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 2)->opcode);
+   EXPECT_TRUE(instruction(block0, 2)->saturate);
+}

src/mesa/main/.gitignore

@@ -1,13 +1,11 @@
 api_exec.c
 dispatch.h
 enums.c
-get_es1.c
-get_es2.c
 git_sha1.h
 git_sha1.h.tmp
 remap_helper.h
 get_hash.h
 get_hash.h.tmp
-format_info.c
+format_info.h
 format_pack.c
 format_unpack.c

src/mesa/main/bufferobj.c

@@ -1226,7 +1226,7 @@ _mesa_DeleteBuffers(GLsizei n, const GLuint *ids)
             }
          }
 
-         if (ctx->UniformBuffer == bufObj) {
+         if (ctx->AtomicBuffer == bufObj) {
             _mesa_BindBuffer( GL_ATOMIC_COUNTER_BUFFER, 0 );
          }
 

src/mesa/main/fbobject.c

@@ -780,6 +780,18 @@ test_attachment_completeness(const struct gl_context *ctx, GLenum format,
             att->Complete = GL_FALSE;
             return;
          }
+
+         /* OES_texture_float allows creation and use of floating point
+          * textures with GL_FLOAT, GL_HALF_FLOAT but it does not allow
+          * these textures to be used as a render target, this is done via
+          * GL_EXT_color_buffer(_half)_float with set of new sized types.
+          */
+         if (_mesa_is_gles(ctx) && (texImage->TexObject->_IsFloat ||
+                                    texImage->TexObject->_IsHalfFloat)) {
+            att_incomplete("bad internal format");
+            att->Complete = GL_FALSE;
+            return;
+         }
       }
       else if (format == GL_DEPTH) {
          if (baseFormat == GL_DEPTH_COMPONENT) {

src/mesa/main/format_utils.c

@@ -1423,8 +1423,8 @@ convert_int(void *void_dst, int num_dst_channels,
  * \param[in]  count             the number of pixels to convert
  */
 void
-_mesa_swizzle_and_convert(void *void_dst, GLenum dst_type, int num_dst_channels,
-                          const void *void_src, GLenum src_type, int num_src_channels,
+_mesa_swizzle_and_convert(void *void_dst, enum mesa_array_format_datatype dst_type, int num_dst_channels,
+                          const void *void_src, enum mesa_array_format_datatype src_type, int num_src_channels,
                           const uint8_t swizzle[4], bool normalized, int count)
 {
    if (swizzle_convert_try_memcpy(void_dst, dst_type, num_dst_channels,

src/mesa/main/formats.c

@@ -75,7 +75,7 @@ struct gl_format_info
    mesa_array_format ArrayFormat;
 };
 
-#include "format_info.c"
+#include "format_info.h"
 
 static const struct gl_format_info *
 _mesa_get_format_info(mesa_format format)

src/mesa/main/readpix.c

@@ -605,12 +605,15 @@ read_rgba_pixels( struct gl_context *ctx,
 done_swap:
    /* Handle byte swapping if required */
    if (packing->SwapBytes) {
-      int components = _mesa_components_in_format(format);
       GLint swapSize = _mesa_sizeof_packed_type(type);
-      if (swapSize == 2)
-         _mesa_swap2((GLushort *) dst, width * height * components);
-      else if (swapSize == 4)
-         _mesa_swap4((GLuint *) dst, width * height * components);
+      if (swapSize == 2 || swapSize == 4) {
+         int swapsPerPixel = _mesa_bytes_per_pixel(format, type) / swapSize;
+         assert(_mesa_bytes_per_pixel(format, type) % swapSize == 0);
+         if (swapSize == 2)
+            _mesa_swap2((GLushort *) dst, width * height * swapsPerPixel);
+         else if (swapSize == 4)
+            _mesa_swap4((GLuint *) dst, width * height * swapsPerPixel);
+      }
    }
 
 done_unmap:

src/mesa/main/texgetimage.c

@@ -511,12 +511,15 @@ get_tex_rgba_uncompressed(struct gl_context *ctx, GLuint dimensions,
    do_swap:
       /* Handle byte swapping if required */
       if (ctx->Pack.SwapBytes) {
-         int components = _mesa_components_in_format(format);
          GLint swapSize = _mesa_sizeof_packed_type(type);
-         if (swapSize == 2)
-            _mesa_swap2((GLushort *) dest, width * height * components);
-         else if (swapSize == 4)
-            _mesa_swap4((GLuint *) dest, width * height * components);
+         if (swapSize == 2 || swapSize == 4) {
+            int swapsPerPixel = _mesa_bytes_per_pixel(format, type) / swapSize;
+            assert(_mesa_bytes_per_pixel(format, type) % swapSize == 0);
+            if (swapSize == 2)
+               _mesa_swap2((GLushort *) dest, width * height * swapsPerPixel);
+            else if (swapSize == 4)
+               _mesa_swap4((GLuint *) dest, width * height * swapsPerPixel);
+         }
       }
 
       /* Unmap the src texture buffer */
@@ -740,7 +743,7 @@ _mesa_GetCompressedTexImage_sw(struct gl_context *ctx,
       GLubyte *src;
 
       /* map src texture buffer */
-      ctx->Driver.MapTextureImage(ctx, texImage, 0,
+      ctx->Driver.MapTextureImage(ctx, texImage, slice,
                                   0, 0, texImage->Width, texImage->Height,
                                   GL_MAP_READ_BIT, &src, &srcRowStride);
 
@@ -752,7 +755,7 @@ _mesa_GetCompressedTexImage_sw(struct gl_context *ctx,
             src += srcRowStride;
          }
 
-         ctx->Driver.UnmapTextureImage(ctx, texImage, 0);
+         ctx->Driver.UnmapTextureImage(ctx, texImage, slice);
 
          /* Advance to next slice */
          dest += store.TotalBytesPerRow * (store.TotalRowsPerSlice - store.CopyRowsPerSlice);

src/mesa/main/teximage.c

@@ -2491,8 +2491,8 @@ texsubimage_error_check(struct gl_context *ctx, GLuint dimensions,
    }
 
    if (error_check_subtexture_dimensions(ctx, dimensions,
-                                         texImage, xoffset, yoffset, 0,
-                                         width, height, 1,
+                                         texImage, xoffset, yoffset, zoffset,
+                                         width, height, depth,
                                          dsa ? "glTextureSubImage" :
                                          "glTexSubImage")) {
       return GL_TRUE;

src/mesa/main/texstore.c

@@ -73,6 +73,7 @@
 #include "texstore.h"
 #include "enums.h"
 #include "glformats.h"
+#include "pixeltransfer.h"
 #include "../../gallium/auxiliary/util/u_format_rgb9e5.h"
 #include "../../gallium/auxiliary/util/u_format_r11g11b10f.h"
 
@@ -675,12 +676,13 @@ texstore_compressed(TEXSTORE_PARAMS)
 static GLboolean
 texstore_rgba(TEXSTORE_PARAMS)
 {
-   void *tempImage = NULL;
+   void *tempImage = NULL, *tempRGBA = NULL;
    int srcRowStride, img;
-   GLubyte *src;
+   GLubyte *src, *dst;
    uint32_t srcMesaFormat;
    uint8_t rebaseSwizzle[4];
    bool needRebase;
+   bool transferOpsDone = false;
 
    /* We have to handle MESA_FORMAT_YCBCR manually because it is a special case
     * and _mesa_format_convert does not support it. In this case the we only
@@ -709,6 +711,11 @@ texstore_rgba(TEXSTORE_PARAMS)
       if (!tempImage)
          return GL_FALSE;
 
+      /* _mesa_unpack_color_index_to_rgba_ubyte has handled transferops
+       * if needed.
+       */
+      transferOpsDone = true;
+
       /* Now we only have to adjust our src info for a conversion from
        * the RGBA ubyte and then we continue as usual.
        */
@@ -721,15 +728,19 @@ texstore_rgba(TEXSTORE_PARAMS)
        */
       GLint swapSize = _mesa_sizeof_packed_type(srcType);
       if (swapSize == 2 || swapSize == 4) {
-         int components = _mesa_components_in_format(srcFormat);
-         int elementCount = srcWidth * srcHeight * components;
-         tempImage = malloc(elementCount * swapSize);
+         int bytesPerPixel = _mesa_bytes_per_pixel(srcFormat, srcType);
+         assert(bytesPerPixel % swapSize == 0);
+         int swapsPerPixel = bytesPerPixel / swapSize;
+         int elementCount = srcWidth * srcHeight * srcDepth;
+         tempImage = malloc(elementCount * bytesPerPixel);
          if (!tempImage)
             return GL_FALSE;
          if (swapSize == 2)
-            _mesa_swap2_copy(tempImage, (GLushort *) srcAddr, elementCount);
+            _mesa_swap2_copy(tempImage, (GLushort *) srcAddr,
+                             elementCount * swapsPerPixel);
          else
-            _mesa_swap4_copy(tempImage, (GLuint *) srcAddr, elementCount);
+            _mesa_swap4_copy(tempImage, (GLuint *) srcAddr,
+                             elementCount * swapsPerPixel);
          srcAddr = tempImage;
       }
    }
@@ -737,13 +748,52 @@ texstore_rgba(TEXSTORE_PARAMS)
    srcRowStride =
       _mesa_image_row_stride(srcPacking, srcWidth, srcFormat, srcType);
 
+   srcMesaFormat = _mesa_format_from_format_and_type(srcFormat, srcType);
+   dstFormat = _mesa_get_srgb_format_linear(dstFormat);
+
+   /* If we have transferOps then we need to convert to RGBA float first,
+      then apply transferOps, then do the conversion to dst
+    */
+   if (!transferOpsDone &&
+       _mesa_texstore_needs_transfer_ops(ctx, baseInternalFormat, dstFormat)) {
+      /* Allocate RGBA float image */
+      int elementCount = srcWidth * srcHeight * srcDepth;
+      tempRGBA = malloc(4 * elementCount * sizeof(float));
+      if (!tempRGBA) {
+         free(tempImage);
+         free(tempRGBA);
+         return GL_FALSE;
+      }
+
+      /* Convert from src to RGBA float */
+      src = (GLubyte *) srcAddr;
+      dst = (GLubyte *) tempRGBA;
+      for (img = 0; img < srcDepth; img++) {
+         _mesa_format_convert(dst, RGBA32_FLOAT, 4 * srcWidth * sizeof(float),
+                              src, srcMesaFormat, srcRowStride,
+                              srcWidth, srcHeight, NULL);
+         src += srcHeight * srcRowStride;
+         dst += srcHeight * 4 * srcWidth * sizeof(float);
+      }
+
+      /* Apply transferOps */
+      _mesa_apply_rgba_transfer_ops(ctx, ctx->_ImageTransferState, elementCount,
+                                    (float(*)[4]) tempRGBA);
+
+      /* Now we have to adjust our src info for a conversion from
+       * the RGBA float image and then we continue as usual.
+       */
+      srcAddr = tempRGBA;
+      srcFormat = GL_RGBA;
+      srcType = GL_FLOAT;
+      srcRowStride = srcWidth * 4 * sizeof(float);
+      srcMesaFormat = RGBA32_FLOAT;
+   }
+
    src = (GLubyte *)
       _mesa_image_address(dims, srcPacking, srcAddr, srcWidth, srcHeight,
                           srcFormat, srcType, 0, 0, 0);
 
-   srcMesaFormat = _mesa_format_from_format_and_type(srcFormat, srcType);
-   dstFormat = _mesa_get_srgb_format_linear(dstFormat);
-
    if (_mesa_get_format_base_format(dstFormat) != baseInternalFormat) {
       needRebase =
          _mesa_compute_rgba2base2rgba_component_mapping(baseInternalFormat,
@@ -761,6 +811,7 @@ texstore_rgba(TEXSTORE_PARAMS)
    }
 
    free(tempImage);
+   free(tempRGBA);
 
    return GL_TRUE;
 }

src/mesa/program/arbprogparse.c

@@ -85,9 +85,6 @@ _mesa_parse_arb_fragment_program(struct gl_context* ctx, GLenum target,
       return;
    }
 
-   if ((ctx->_Shader->Flags & GLSL_NO_OPT) == 0)
-      _mesa_optimize_program(ctx, &prog);
-
    free(program->Base.String);
 
    /* Copy the relevant contents of the arb_program struct into the

src/mesa/state_tracker/st_cb_drawpixels.c

@@ -1100,7 +1100,7 @@ st_DrawPixels(struct gl_context *ctx, GLint x, GLint y,
    const GLfloat *color;
    struct pipe_context *pipe = st->pipe;
    GLboolean write_stencil = GL_FALSE, write_depth = GL_FALSE;
-   struct pipe_sampler_view *sv[2];
+   struct pipe_sampler_view *sv[2] = { NULL };
    int num_sampler_view = 1;
    struct st_fp_variant *fpv;
    struct gl_pixelstore_attrib clippedUnpack;
@@ -1154,8 +1154,9 @@ st_DrawPixels(struct gl_context *ctx, GLint x, GLint y,
 
       color = NULL;
       if (st->pixel_xfer.pixelmap_enabled) {
-	  sv[1] = st->pixel_xfer.pixelmap_sampler_view;
-	  num_sampler_view++;
+         pipe_sampler_view_reference(&sv[1],
+                                     st->pixel_xfer.pixelmap_sampler_view);
+         num_sampler_view++;
       }
    }
 
@@ -1176,7 +1177,8 @@ st_DrawPixels(struct gl_context *ctx, GLint x, GLint y,
             if (write_stencil) {
                enum pipe_format stencil_format =
                      util_format_stencil_only(pt->format);
-
+               /* we should not be doing pixel map/transfer (see above) */
+               assert(num_sampler_view == 1);
                sv[1] = st_create_texture_sampler_view_format(st->pipe, pt,
                                                              stencil_format);
                num_sampler_view++;
@@ -1467,7 +1469,7 @@ st_CopyPixels(struct gl_context *ctx, GLint srcx, GLint srcy,
    struct st_renderbuffer *rbRead;
    void *driver_vp, *driver_fp;
    struct pipe_resource *pt;
-   struct pipe_sampler_view *sv[2];
+   struct pipe_sampler_view *sv[2] = { NULL };
    int num_sampler_view = 1;
    GLfloat *color;
    enum pipe_format srcFormat;
@@ -1516,7 +1518,8 @@ st_CopyPixels(struct gl_context *ctx, GLint srcx, GLint srcy,
       driver_vp = make_passthrough_vertex_shader(st, GL_FALSE);
 
       if (st->pixel_xfer.pixelmap_enabled) {
-         sv[1] = st->pixel_xfer.pixelmap_sampler_view;
+         pipe_sampler_view_reference(&sv[1],
+                                     st->pixel_xfer.pixelmap_sampler_view);
          num_sampler_view++;
       }
    }

src/mesa/state_tracker/st_cb_xformfb.c

@@ -122,7 +122,7 @@ st_begin_transform_feedback(struct gl_context *ctx, GLenum mode,
    for (i = 0; i < max_num_targets; i++) {
       struct st_buffer_object *bo = st_buffer_object(sobj->base.Buffers[i]);
 
-      if (bo) {
+      if (bo && bo->buffer) {
          /* Check whether we need to recreate the target. */
          if (!sobj->targets[i] ||
              sobj->targets[i] == sobj->draw_count ||

src/mesa/swrast/s_span.c

@@ -933,19 +933,19 @@ clamp_colors(SWspan *span)
  * \param output  which fragment program color output is being processed
  */
 static inline void
-convert_color_type(SWspan *span, GLenum newType, GLuint output)
+convert_color_type(SWspan *span, GLenum srcType, GLenum newType, GLuint output)
 {
    GLvoid *src, *dst;
 
-   if (output > 0 || span->array->ChanType == GL_FLOAT) {
+   if (output > 0 || srcType == GL_FLOAT) {
       src = span->array->attribs[VARYING_SLOT_COL0 + output];
       span->array->ChanType = GL_FLOAT;
    }
-   else if (span->array->ChanType == GL_UNSIGNED_BYTE) {
+   else if (srcType == GL_UNSIGNED_BYTE) {
       src = span->array->rgba8;
    }
    else {
-      ASSERT(span->array->ChanType == GL_UNSIGNED_SHORT);
+      ASSERT(srcType == GL_UNSIGNED_SHORT);
       src = span->array->rgba16;
    }
 
@@ -979,7 +979,7 @@ shade_texture_span(struct gl_context *ctx, SWspan *span)
        ctx->ATIFragmentShader._Enabled) {
       /* programmable shading */
       if (span->primitive == GL_BITMAP && span->array->ChanType != GL_FLOAT) {
-         convert_color_type(span, GL_FLOAT, 0);
+         convert_color_type(span, span->array->ChanType, GL_FLOAT, 0);
       }
       else {
          span->array->rgba = (void *) span->array->attribs[VARYING_SLOT_COL0];
@@ -1314,6 +1314,8 @@ _swrast_write_rgba_span( struct gl_context *ctx, SWspan *span)
       const GLboolean multiFragOutputs = 
          _swrast_use_fragment_program(ctx)
          && fp->Base.OutputsWritten >= (1 << FRAG_RESULT_DATA0);
+      /* Save srcColorType because convert_color_type() can change it */
+      const GLenum srcColorType = span->array->ChanType;
       GLuint buf;
 
       for (buf = 0; buf < numBuffers; buf++) {
@@ -1325,17 +1327,18 @@ _swrast_write_rgba_span( struct gl_context *ctx, SWspan *span)
             /* re-use one of the attribute array buffers for rgbaSave */
             GLchan (*rgbaSave)[4] = (GLchan (*)[4]) span->array->attribs[0];
             struct swrast_renderbuffer *srb = swrast_renderbuffer(rb);
-            GLenum colorType = srb->ColorType;
+            const GLenum dstColorType = srb->ColorType;
 
-            assert(colorType == GL_UNSIGNED_BYTE ||
-                   colorType == GL_FLOAT);
+            assert(dstColorType == GL_UNSIGNED_BYTE ||
+                   dstColorType == GL_FLOAT);
 
             /* set span->array->rgba to colors for renderbuffer's datatype */
-            if (span->array->ChanType != colorType) {
-               convert_color_type(span, colorType, 0);
+            if (srcColorType != dstColorType) {
+               convert_color_type(span, srcColorType, dstColorType,
+                                  multiFragOutputs ? buf : 0);
             }
             else {
-               if (span->array->ChanType == GL_UNSIGNED_BYTE) {
+               if (srcColorType == GL_UNSIGNED_BYTE) {
                   span->array->rgba = span->array->rgba8;
                }
                else {

src/mesa/vbo/vbo_attrib_tmp.h

@@ -210,6 +210,7 @@ static inline float conv_i2_to_norm_float(const struct gl_context *ctx, int i2)
       }								\
    } else if ((type) == GL_UNSIGNED_INT_10F_11F_11F_REV) {	\
       float res[4];						\
+      res[3] = 1;                                               \
       r11g11b10f_to_float3((arg), res);				\
       ATTR##val##FV((attr), res);				\
    } else							\

src/util/u_atomic.h

@@ -177,7 +177,7 @@ char _InterlockedCompareExchange8(char volatile *Destination8, char Exchange8, c
    sizeof(*v) == sizeof(uint64_t) ? atomic_inc_64((uint64_t *)(v)) : \
                                     (assert(!"should not get here"), 0))
 
-#define p_atomic_inc_return(v) ((typeof(*v)) \
+#define p_atomic_inc_return(v) ((__typeof(*v)) \
    sizeof(*v) == sizeof(uint8_t)  ? atomic_inc_8_nv ((uint8_t  *)(v)) : \
    sizeof(*v) == sizeof(uint16_t) ? atomic_inc_16_nv((uint16_t *)(v)) : \
    sizeof(*v) == sizeof(uint32_t) ? atomic_inc_32_nv((uint32_t *)(v)) : \
@@ -191,14 +191,14 @@ char _InterlockedCompareExchange8(char volatile *Destination8, char Exchange8, c
    sizeof(*v) == sizeof(uint64_t) ? atomic_dec_64((uint64_t *)(v)) : \
                                     (assert(!"should not get here"), 0))
 
-#define p_atomic_dec_return(v) ((typeof(*v)) \
+#define p_atomic_dec_return(v) ((__typeof(*v)) \
    sizeof(*v) == sizeof(uint8_t)  ? atomic_dec_8_nv ((uint8_t  *)(v)) : \
    sizeof(*v) == sizeof(uint16_t) ? atomic_dec_16_nv((uint16_t *)(v)) : \
    sizeof(*v) == sizeof(uint32_t) ? atomic_dec_32_nv((uint32_t *)(v)) : \
    sizeof(*v) == sizeof(uint64_t) ? atomic_dec_64_nv((uint64_t *)(v)) : \
                                     (assert(!"should not get here"), 0))
 
-#define p_atomic_cmpxchg(v, old, _new) ((typeof(*v)) \
+#define p_atomic_cmpxchg(v, old, _new) ((__typeof(*v)) \
    sizeof(*v) == sizeof(uint8_t)  ? atomic_cas_8 ((uint8_t  *)(v), (uint8_t )(old), (uint8_t )(_new)) : \
    sizeof(*v) == sizeof(uint16_t) ? atomic_cas_16((uint16_t *)(v), (uint16_t)(old), (uint16_t)(_new)) : \
    sizeof(*v) == sizeof(uint32_t) ? atomic_cas_32((uint32_t *)(v), (uint32_t)(old), (uint32_t)(_new)) : \