docs: Update 13.0.0 release notes

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

VERSION

@@ -1 +1 @@
-13.0.0-rc2
+13.0.0

bin/.cherry-ignore

@@ -0,0 +1,2 @@
+# Commit was picked with -x
+907ace57986733add2aebfa9dd7c83c67efed70e mapi: automake: set VISIBILITY_CFLAGS for shared glapi

docs/relnotes/13.0.0.html

@@ -14,7 +14,7 @@
 <iframe src="../contents.html"></iframe>
 <div class="content">
 
-<h1>Mesa 13.0.0 Release Notes / TBD</h1>
+<h1>Mesa 13.0.0 Release Notes / November 1, 2016</h1>
 
 <p>
 Mesa 13.0.0 is a new development release.
@@ -74,11 +74,236 @@ Note: some of the new features are only available with certain drivers.
 
 <h2>Bug fixes</h2>
 
-TBD.
+<ul>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=61907">Bug 61907</a> - Indirect rendering of multi-texture vertex arrays broken</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=69622">Bug 69622</a> - eglTerminate then eglMakeCurrent crahes</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=71759">Bug 71759</a> - Intel driver fails with &quot;intel_do_flush_locked failed: No such file or directory&quot; if buffer imported with EGL_NATIVE_PIXMAP_KHR</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=83036">Bug 83036</a> - [ILK]Piglit spec_ARB_copy_image_arb_copy_image-formats fails</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89599">Bug 89599</a> - symbol 'x86_64_entry_start' is already defined when building with LLVM/clang</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90513">Bug 90513</a> - Odd gray and red flicker in The Talos Principle on GK104</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91342">Bug 91342</a> - Very dark textures on some objects in indoors environments in Postal 2</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92306">Bug 92306</a> - GL Excess demo renders incorrectly on nv43</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94148">Bug 94148</a> - Framebuffer considered invalid when a draw call is done before glCheckFramebufferStatus</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94354">Bug 94354</a> - R9285 Unigine Valley perf regression since radeonsi: use re-Z</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94561">Bug 94561</a> - [llvmpipe] PIPE_CAP_VIDEO_MEMORY reports negative value on 32 bits (with 16GB ram)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94627">Bug 94627</a> - Game Risen on wine black grass</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=94681">Bug 94681</a> - dEQP-GLES31.functional.ssbo.layout.random.all_shared_buffer.23 takes 25 minutes to compile</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=95000">Bug 95000</a> - deqp: assert in dEQP-GLES3.functional.vertex_arrays.single_attribute.strides.fixed.user_ptr_stride17_components2_quads1</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=95130">Bug 95130</a> - Derivatives of gl_Color wrong when helper pixels used</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=95246">Bug 95246</a> - Segfault in glBindFramebuffer()</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=95419">Bug 95419</a> - [HSW][regression][bisect] RPG Maker game gives &quot;invalid floating point operation&quot; at startup</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=95462">Bug 95462</a> - [BXT,BSW] arb_gpu_shader_fp64 causes gpu hang</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=95529">Bug 95529</a> - [regression, bisected] Image corruption in Chrome</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96235">Bug 96235</a> - st_nir.h:34: error: redefinition of typedef ‘nir_shader’</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96274">Bug 96274</a> - [NVC0] Failure when compiling compute shader: Assertion `bb-&gt;getFirst()-&gt;serial &lt;= bb-&gt;getExit()-&gt;serial' failed</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96285">Bug 96285</a> - Mesa build broken</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96299">Bug 96299</a> - [vulkan] 64 regressions due to mesa d5f2f32</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96343">Bug 96343</a> - oom since st/mesa: implement PBO downloads for ReadPixels</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96346">Bug 96346</a> - [SNB,CTS] es2-cts.gtf.gl.atan regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96349">Bug 96349</a> - [CTS,SKL,BSW,BDW,KBL,BXT] es31-cts.arrays_of_arrays.interactionuniformbuffers3</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96351">Bug 96351</a> - [CTS,SKL,KBL,BXT] es2-cts.gtf.gl2extensiontests.egl_image.egl_image</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96358">Bug 96358</a> - SSO: wrong interface validation between GS and VS (regresion due to latest gles 3.1)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96425">Bug 96425</a> - [bisected] occasional dark render in The Talos Principle</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96484">Bug 96484</a> - [vulkan] deqp-vk.glsl.builtin.precision.sin / cos regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96504">Bug 96504</a> - [vulkancts] compute tests crash</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96516">Bug 96516</a> - [bisected: 482526] &quot;clover: Update OpenCL version string to match OpenGL&quot;: clover's build fails because of missing git_sha1.h</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96528">Bug 96528</a> - Location qualifier segfaults during shader compilation</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96541">Bug 96541</a> - Tonga Unreal elemental bad rendering since radeonsi: Decompress DCC textures in a render feedback loop</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96565">Bug 96565</a> - Clive Barker's Jericho displays strange,vivid colors when motion blur enabled</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96607">Bug 96607</a> - [bisected] texture misrender / flicker in The Talos Principle on SKL</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96617">Bug 96617</a> - gl_SecondaryFragDataEXT doesn't work for extended blend func</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96629">Bug 96629</a> - dEQP-GLES2.functional.texture.completeness.cube.not_positive_level_0: Assertion `width &gt;= 1' failed.</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96639">Bug 96639</a> - st/mesa: transfer_map with too-high level with dEQP-GLES2.functional.texture.completeness.cube.extra_level</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96674">Bug 96674</a> - [SNB, ILK] spec.ext_image_dma_buf_import.ext_image_dma_buf_import-sample_nv1</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96729">Bug 96729</a> - Wrong shader compilation error message</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96762">Bug 96762</a> - [radeonsi,apitrace] Firewatch: nothing rendered in scrollable (text) areas</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96765">Bug 96765</a> - BindFragDataLocationIndexed on array fragment shader output.</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96770">Bug 96770</a> - include/GL/mesa_glinterop.h:62: error: redefinition of typedef ‘GLXContext’</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96782">Bug 96782</a> - [regression bisected] R600 fp64 and glsl-4.00 piglit failures</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96791">Bug 96791</a> - Cannot use image from swapchains for sampling</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96825">Bug 96825</a> - anv_device.c:31:27: fatal error: anv_timestamp.h: No such file or directory</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96835">Bug 96835</a> - &quot;gallium: Force blend color to 16-byte alignment&quot; crash with &quot;-march=native -O3&quot; causes some 32bit games to crash</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96850">Bug 96850</a> - Crucible tests fail for 32bit mesa</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96878">Bug 96878</a> - [Bisected: cc2d0e6][HSW] &quot;GPU HANG&quot; msg after autologin to gnome-session</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96908">Bug 96908</a> - [radeonsi] MSAA causes graphical artifacts</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96911">Bug 96911</a> - webgl2 conformance2/textures/misc/tex-mipmap-levels.html crashes 12.1 Intel driver</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96949">Bug 96949</a> - [regression] Piglit numSamples assertion failures with 9a23a177b90</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96950">Bug 96950</a> - Another regression from bc4e0c486: vbo: Use a bitmask to track the active arrays in vbo_exec*.</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=96971">Bug 96971</a> - invariant qualifier is not valid for shader inputs</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97019">Bug 97019</a> - [clover] build failure in llvm/codegen/native.cpp:129:52</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97032">Bug 97032</a> - [BDW,SKL] piglit.spec.arb_gpu_shader5.arb_gpu_shader5-interpolateatcentroid-flat</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97033">Bug 97033</a> - [BDW,SKL] piglit.spec.arb_gpu_shader_fp64.varying-packing.simple regressions</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97039">Bug 97039</a> - The Talos Principle and Serious Sam 3 GPU faults</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97083">Bug 97083</a> - [IVB,BYT] GPU hang on deqp-gles31.functional.separate.shader.random</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97140">Bug 97140</a> - dd_draw.c:949:11: error: implicit declaration of function 'fmemopen' is invalid in C99 [-Werror,-Wimplicit-function-declaration]</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97207">Bug 97207</a> - [IVY BRIDGE] Fragment shader discard writing to depth</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97214">Bug 97214</a> - X not running with error &quot;Failed to make EGL context current&quot;</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97225">Bug 97225</a> - [i965 on HD4600 Haswell] xcom switch to ingame cinematics cause segmentation fault</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97231">Bug 97231</a> - GL_DEPTH_CLAMP doesn't clamp to the far plane</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97233">Bug 97233</a> - vkQuake VkSpecializationMapEntry related bug</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97260">Bug 97260</a> - R9 290 low performance in Linux 4.7</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97267">Bug 97267</a> - [BDW] GL45-CTS.texture_cube_map_array.sampling asserts inside brw_fs.cpp</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97278">Bug 97278</a> - [vulkancts,HSW] all vulkancts tests assert on HSW</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97285">Bug 97285</a> - Darkness in Dota 2 after Patch &quot;Make Gallium's BlitFramebuffer follow the GL 4.4 sRGB rules&quot;</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97286">Bug 97286</a> - `make check` fails uniform-initializer-test</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97305">Bug 97305</a> - Gallium: TBOs and images set the offset in elements, not bytes</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97307">Bug 97307</a> - glsl/glcpp/tests/glcpp-test regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97309">Bug 97309</a> - piglit.spec.glsl-1_30.compiler.switch-statement.switch-case-duplicated.vert regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97322">Bug 97322</a> - GenerateMipmap creates wrong mipmap for sRGB texture</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97331">Bug 97331</a> - glDrawElementsBaseVertex doesn't work in display list on i915</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97351">Bug 97351</a> - DrawElementsBaseVertex with VBO ignores base vertex on Intel GMA 9xx in some cases</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97413">Bug 97413</a> - BioShock Infinite crashes on startup with Mesa Git version, R7 370</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97426">Bug 97426</a> - glScissor gives vertically inverted result</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97448">Bug 97448</a> - [HSW] deqp-vk.api_.copy_and_blit.image_to_image_stencil regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97476">Bug 97476</a> - Shader binaries should not be stored in the PipelineCache</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97477">Bug 97477</a> - i915g: gl_FragCoord is always (0.0, max_y)</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97513">Bug 97513</a> - clover reports wrong device pointer size</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97549">Bug 97549</a> - [SNB, BXT] up to 40% perf drop from &quot;loader/dri3: Overhaul dri3_update_num_back&quot; commit</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97587">Bug 97587</a> - make check nir/tests/control_flow_tests regression</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97761">Bug 97761</a> - es2-cts.gtf.gl2extensiontests.egl_image_external.testsimpleunassociated crashes</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97773">Bug 97773</a> - New Mesa master now results in warnings in glrender (and subsurfaces and simple-egl), black screen</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97779">Bug 97779</a> - [regression, bisected][BDW, GPU hang] stuck on render ring, always reproducible</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97790">Bug 97790</a> - Vulkan cts regressions due to 24be63066</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97804">Bug 97804</a> - Later precision statement isn't overriding earlier one</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97808">Bug 97808</a> - &quot;tgsi/scan: don't set interp flags for inputs only used by INTERP instructions&quot; causes glitches in wine with gallium nine</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97887">Bug 97887</a> - llvm segfault in janusvr -render vive</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97894">Bug 97894</a> - Crash in u_transfer_unmap_vtbl when unmapping a buffer mapped in different context</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97952">Bug 97952</a> - /usr/include/string.h:518:12: error: exception specification in declaration does not match previous declaration</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97969">Bug 97969</a> - [radeonsi, bisected: fb827c0] Video decoding shows green artifacts</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=97976">Bug 97976</a> - VCE regression BO to small for addr since winsys/amdgpu: enable buffer allocation from slabs</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98005">Bug 98005</a> - VCE dual instance encoding inconsistent since st/va: enable dual instances encode by sync surface</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98025">Bug 98025</a> - [radeonsi] incorrect primitive restart index used</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98128">Bug 98128</a> - nir/tests/control_flow_tests.cpp:79:73: error: ‘nir_loop_first_cf_node’ was not declared in this scope</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98131">Bug 98131</a> - Compiler should reject lowp/mediump qualifiers on atomic_uints</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98133">Bug 98133</a> - GetSynciv should raise an error if bufSize &lt; 0</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98134">Bug 98134</a> - dEQP-GLES31.functional.debug.negative_coverage.get_error.buffer.draw_buffers wants a different GL error code</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98135">Bug 98135</a> - dEQP-GLES31.functional.debug.negative_coverage.get_error.shader.transform_feedback_varyings wants a different GL error code</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98167">Bug 98167</a> - [vulkan, radv] missing libgcrypt and openssl devel results in linker error in libvulkan_common</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98172">Bug 98172</a> - Concurrent call to glClientWaitSync results in segfault in one of the waiters.</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98244">Bug 98244</a> - dEQP: textureOffset(sampler2DArrayShadow, ...) should not exist.</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98264">Bug 98264</a> - Build broken for i965 due to multiple deifnitions of intelFenceExtension</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98307">Bug 98307</a> - &quot;st/glsl_to_tgsi: explicitly track all input and output declaration&quot; broke flightgear colors on rs780</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98326">Bug 98326</a> - [dEQP, EGL] pbuffer depth/stencil tests fail</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98415">Bug 98415</a> - Vulkan Driver JSON file contains incorrect field</li>
+
+<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=98431">Bug 98431</a> - UnrealEngine v4 demos startup fails to blorp blit assert</li>
+
+</ul>
+
 
 <h2>Changes</h2>
 
-TBD.
+Mesa no longer depends on libudev.
 
 </div>
 </body>

src/amd/common/ac_nir_to_llvm.c

@@ -3520,7 +3520,8 @@ static void visit_tex(struct nir_to_llvm_context *ctx, nir_tex_instr *instr)
 		LLVMValueRef z = LLVMBuildExtractElement(ctx->builder, result, two, "");
 		z = LLVMBuildSDiv(ctx->builder, z, six, "");
 		result = LLVMBuildInsertElement(ctx->builder, result, z, two, "");
-	}
+	} else if (instr->dest.ssa.num_components != 4)
+		result = trim_vector(ctx, result, instr->dest.ssa.num_components);
 
 write_result:
 	if (result) {

src/amd/vulkan/dev_icd.json.in

@@ -2,6 +2,6 @@
     "file_format_version": "1.0.0",
     "ICD": {
         "library_path": "@build_libdir@/libvulkan_radeon.so",
-        "abi_versions": "1.0.3"
+        "api_version": "1.0.3"
     }
 }

src/amd/vulkan/radeon_icd.json

@@ -2,6 +2,6 @@
     "file_format_version": "1.0.0",
     "ICD": {
         "library_path": "libvulkan_radeon.so",
-        "abi_versions": "1.0.3"
+        "api_version": "1.0.3"
     }
 }

src/amd/vulkan/radv_device.c

@@ -44,12 +44,6 @@
 #include "util/debug.h"
 struct radv_dispatch_table dtable;
 
-struct radv_fence {
-	struct radeon_winsys_fence *fence;
-	bool submitted;
-	bool signalled;
-};
-
 static VkResult
 radv_physical_device_init(struct radv_physical_device *device,
 			  struct radv_instance *instance,

src/amd/vulkan/radv_private.h

@@ -1206,6 +1206,13 @@ void radv_initialise_cmask(struct radv_cmd_buffer *cmd_buffer,
 			   struct radv_image *image, uint32_t value);
 void radv_initialize_dcc(struct radv_cmd_buffer *cmd_buffer,
 			 struct radv_image *image, uint32_t value);
+
+struct radv_fence {
+	struct radeon_winsys_fence *fence;
+	bool submitted;
+	bool signalled;
+};
+
 #define RADV_DEFINE_HANDLE_CASTS(__radv_type, __VkType)		\
 								\
 	static inline struct __radv_type *			\

src/amd/vulkan/radv_wsi.c

@@ -318,13 +318,21 @@ VkResult radv_AcquireNextImageKHR(
 	VkSwapchainKHR                               _swapchain,
 	uint64_t                                     timeout,
 	VkSemaphore                                  semaphore,
-	VkFence                                      fence,
+	VkFence                                      _fence,
 	uint32_t*                                    pImageIndex)
 {
 	RADV_FROM_HANDLE(wsi_swapchain, swapchain, _swapchain);
+	RADV_FROM_HANDLE(radv_fence, fence, _fence);
 
-	return swapchain->acquire_next_image(swapchain, timeout, semaphore,
-					     pImageIndex);
+	VkResult result = swapchain->acquire_next_image(swapchain, timeout, semaphore,
+	                                                pImageIndex);
+
+	if (fence && result == VK_SUCCESS) {
+		fence->submitted = true;
+		fence->signalled = true;
+	}
+
+	return result;
 }
 
 VkResult radv_QueuePresentKHR(

src/compiler/glsl/ast_to_hir.cpp

@@ -4330,6 +4330,8 @@ handle_tess_ctrl_shader_output_decl(struct _mesa_glsl_parse_state *state,
    if (var->data.patch)
       return;
 
+   var->data.tess_varying_implicit_sized_array = var->type->is_unsized_array();
+
    validate_layout_qualifier_vertex_count(state, loc, var, num_vertices,
                                           &state->tcs_output_size,
                                           "tessellation control shader output");
@@ -4366,6 +4368,7 @@ handle_tess_shader_input_decl(struct _mesa_glsl_parse_state *state,
    if (var->type->is_unsized_array()) {
       var->type = glsl_type::get_array_instance(var->type->fields.array,
             state->Const.MaxPatchVertices);
+      var->data.tess_varying_implicit_sized_array = true;
    } else if (var->type->length != state->Const.MaxPatchVertices) {
       _mesa_glsl_error(&loc, state,
                        "per-vertex tessellation shader input arrays must be "

src/compiler/glsl/glsl_symbol_table.cpp

@@ -126,7 +126,7 @@ void glsl_symbol_table::pop_scope()
 
 bool glsl_symbol_table::name_declared_this_scope(const char *name)
 {
-   return _mesa_symbol_table_symbol_scope(table, -1, name) == 0;
+   return _mesa_symbol_table_symbol_scope(table, name) == 0;
 }
 
 bool glsl_symbol_table::add_variable(ir_variable *v)
@@ -152,7 +152,7 @@ bool glsl_symbol_table::add_variable(ir_variable *v)
 	 symbol_table_entry *entry = new(mem_ctx) symbol_table_entry(v);
 	 if (existing != NULL)
 	    entry->f = existing->f;
-	 int added = _mesa_symbol_table_add_symbol(table, -1, v->name, entry);
+	 int added = _mesa_symbol_table_add_symbol(table, v->name, entry);
 	 assert(added == 0);
 	 (void)added;
 	 return true;
@@ -162,13 +162,13 @@ bool glsl_symbol_table::add_variable(ir_variable *v)
 
    /* 1.20+ rules: */
    symbol_table_entry *entry = new(mem_ctx) symbol_table_entry(v);
-   return _mesa_symbol_table_add_symbol(table, -1, v->name, entry) == 0;
+   return _mesa_symbol_table_add_symbol(table, v->name, entry) == 0;
 }
 
 bool glsl_symbol_table::add_type(const char *name, const glsl_type *t)
 {
    symbol_table_entry *entry = new(mem_ctx) symbol_table_entry(t);
-   return _mesa_symbol_table_add_symbol(table, -1, name, entry) == 0;
+   return _mesa_symbol_table_add_symbol(table, name, entry) == 0;
 }
 
 bool glsl_symbol_table::add_interface(const char *name, const glsl_type *i,
@@ -180,7 +180,7 @@ bool glsl_symbol_table::add_interface(const char *name, const glsl_type *i,
       symbol_table_entry *entry =
          new(mem_ctx) symbol_table_entry(i, mode);
       bool add_interface_symbol_result =
-         _mesa_symbol_table_add_symbol(table, -1, name, entry) == 0;
+         _mesa_symbol_table_add_symbol(table, name, entry) == 0;
       assert(add_interface_symbol_result);
       return add_interface_symbol_result;
    } else {
@@ -199,7 +199,7 @@ bool glsl_symbol_table::add_function(ir_function *f)
       }
    }
    symbol_table_entry *entry = new(mem_ctx) symbol_table_entry(f);
-   return _mesa_symbol_table_add_symbol(table, -1, f->name, entry) == 0;
+   return _mesa_symbol_table_add_symbol(table, f->name, entry) == 0;
 }
 
 bool glsl_symbol_table::add_default_precision_qualifier(const char *type_name,
@@ -213,13 +213,16 @@ bool glsl_symbol_table::add_default_precision_qualifier(const char *type_name,
    symbol_table_entry *entry =
       new(mem_ctx) symbol_table_entry(default_specifier);
 
-   return _mesa_symbol_table_add_symbol(table, -1, name, entry) == 0;
+   if (!get_entry(name))
+      return _mesa_symbol_table_add_symbol(table, name, entry) == 0;
+
+   return _mesa_symbol_table_replace_symbol(table, name, entry) == 0;
 }
 
 void glsl_symbol_table::add_global_function(ir_function *f)
 {
    symbol_table_entry *entry = new(mem_ctx) symbol_table_entry(f);
-   int added = _mesa_symbol_table_add_global_symbol(table, -1, f->name, entry);
+   int added = _mesa_symbol_table_add_global_symbol(table, f->name, entry);
    assert(added == 0);
    (void)added;
 }
@@ -261,7 +264,7 @@ int glsl_symbol_table::get_default_precision_qualifier(const char *type_name)
 symbol_table_entry *glsl_symbol_table::get_entry(const char *name)
 {
    return (symbol_table_entry *)
-      _mesa_symbol_table_find_symbol(table, -1, name);
+      _mesa_symbol_table_find_symbol(table, name);
 }
 
 void

src/compiler/glsl/ir.h

@@ -832,6 +832,12 @@ public:
 
       unsigned implicit_sized_array:1;
 
+      /**
+       * Is this a non-patch TCS output / TES input array that was implicitly
+       * sized to gl_MaxPatchVertices?
+       */
+      unsigned tess_varying_implicit_sized_array:1;
+
       /**
        * Whether this is a fragment shader output implicitly initialized with
        * the previous contents of the specified render target at the

src/compiler/glsl/ir_print_visitor.cpp

@@ -130,14 +130,14 @@ ir_print_visitor::unique_name(ir_variable *var)
 
    /* If there's no conflict, just use the original name */
    const char* name = NULL;
-   if (_mesa_symbol_table_find_symbol(this->symbols, -1, var->name) == NULL) {
+   if (_mesa_symbol_table_find_symbol(this->symbols, var->name) == NULL) {
       name = var->name;
    } else {
       static unsigned i = 1;
       name = ralloc_asprintf(this->mem_ctx, "%s@%u", var->name, ++i);
    }
    _mesa_hash_table_insert(this->printable_names, var, (void *) name);
-   _mesa_symbol_table_add_symbol(this->symbols, -1, name, var);
+   _mesa_symbol_table_add_symbol(this->symbols, name, var);
    return name;
 }
 

src/compiler/glsl/linker.cpp

@@ -3574,6 +3574,7 @@ static gl_shader_variable *
 create_shader_variable(struct gl_shader_program *shProg,
                        const ir_variable *in,
                        const char *name, const glsl_type *type,
+                       const glsl_type *interface_type,
                        bool use_implicit_location, int location,
                        const glsl_type *outermost_struct_type)
 {
@@ -3631,7 +3632,7 @@ create_shader_variable(struct gl_shader_program *shProg,
 
    out->type = type;
    out->outermost_struct_type = outermost_struct_type;
-   out->interface_type = in->get_interface_type();
+   out->interface_type = interface_type;
    out->component = in->data.location_frac;
    out->index = in->data.index;
    out->patch = in->data.patch;
@@ -3643,8 +3644,21 @@ create_shader_variable(struct gl_shader_program *shProg,
    return out;
 }
 
+static const glsl_type *
+resize_to_max_patch_vertices(const struct gl_context *ctx,
+                             const glsl_type *type)
+{
+   if (!type)
+      return NULL;
+
+   return glsl_type::get_array_instance(type->fields.array,
+                                        ctx->Const.MaxPatchVertices);
+}
+
 static bool
-add_shader_variable(struct gl_shader_program *shProg, struct set *resource_set,
+add_shader_variable(const struct gl_context *ctx,
+                    struct gl_shader_program *shProg,
+                    struct set *resource_set,
                     unsigned stage_mask,
                     GLenum programInterface, ir_variable *var,
                     const char *name, const glsl_type *type,
@@ -3673,7 +3687,7 @@ add_shader_variable(struct gl_shader_program *shProg, struct set *resource_set,
       for (unsigned i = 0; i < type->length; i++) {
          const struct glsl_struct_field *field = &type->fields.structure[i];
          char *field_name = ralloc_asprintf(shProg, "%s.%s", name, field->name);
-         if (!add_shader_variable(shProg, resource_set,
+         if (!add_shader_variable(ctx, shProg, resource_set,
                                   stage_mask, programInterface,
                                   var, field_name, field->type,
                                   use_implicit_location, field_location,
@@ -3687,6 +3701,29 @@ add_shader_variable(struct gl_shader_program *shProg, struct set *resource_set,
    }
 
    default: {
+      const glsl_type *interface_type = var->get_interface_type();
+
+      /* Unsized (non-patch) TCS output/TES input arrays are implicitly
+       * sized to gl_MaxPatchVertices.  Internally, we shrink them to a
+       * smaller size.
+       *
+       * This can cause trouble with SSO programs.  Since the TCS declares
+       * the number of output vertices, we can always shrink TCS output
+       * arrays.  However, the TES might not be linked with a TCS, in
+       * which case it won't know the size of the patch.  In other words,
+       * the TCS and TES may disagree on the (smaller) array sizes.  This
+       * can result in the resource names differing across stages, causing
+       * SSO validation failures and other cascading issues.
+       *
+       * Expanding the array size to the full gl_MaxPatchVertices fixes
+       * these issues.  It's also what program interface queries expect,
+       * as that is the official size of the array.
+       */
+      if (var->data.tess_varying_implicit_sized_array) {
+         type = resize_to_max_patch_vertices(ctx, type);
+         interface_type = resize_to_max_patch_vertices(ctx, interface_type);
+      }
+
       /* Issue #16 of the ARB_program_interface_query spec says:
        *
        * "* If a variable is a member of an interface block without an
@@ -3699,8 +3736,7 @@ add_shader_variable(struct gl_shader_program *shProg, struct set *resource_set,
        */
       const char *prefixed_name = (var->data.from_named_ifc_block &&
                                    !is_gl_identifier(var->name))
-         ? ralloc_asprintf(shProg, "%s.%s", var->get_interface_type()->name,
-                           name)
+         ? ralloc_asprintf(shProg, "%s.%s", interface_type->name, name)
          : name;
 
       /* The ARB_program_interface_query spec says:
@@ -3711,6 +3747,7 @@ add_shader_variable(struct gl_shader_program *shProg, struct set *resource_set,
        */
       gl_shader_variable *sha_v =
          create_shader_variable(shProg, var, prefixed_name, type,
+                                interface_type,
                                 use_implicit_location, location,
                                 outermost_struct_type);
       if (!sha_v)
@@ -3723,7 +3760,8 @@ add_shader_variable(struct gl_shader_program *shProg, struct set *resource_set,
 }
 
 static bool
-add_interface_variables(struct gl_shader_program *shProg,
+add_interface_variables(const struct gl_context *ctx,
+                        struct gl_shader_program *shProg,
                         struct set *resource_set,
                         unsigned stage, GLenum programInterface)
 {
@@ -3774,7 +3812,7 @@ add_interface_variables(struct gl_shader_program *shProg,
          (stage == MESA_SHADER_VERTEX && var->data.mode == ir_var_shader_in) ||
          (stage == MESA_SHADER_FRAGMENT && var->data.mode == ir_var_shader_out);
 
-      if (!add_shader_variable(shProg, resource_set,
+      if (!add_shader_variable(ctx, shProg, resource_set,
                                1 << stage, programInterface,
                                var, var->name, var->type, vs_input_or_fs_output,
                                var->data.location - loc_bias))
@@ -3784,7 +3822,9 @@ add_interface_variables(struct gl_shader_program *shProg,
 }
 
 static bool
-add_packed_varyings(struct gl_shader_program *shProg, struct set *resource_set,
+add_packed_varyings(const struct gl_context *ctx,
+                    struct gl_shader_program *shProg,
+                    struct set *resource_set,
                     int stage, GLenum type)
 {
    struct gl_linked_shader *sh = shProg->_LinkedShaders[stage];
@@ -3810,7 +3850,7 @@ add_packed_varyings(struct gl_shader_program *shProg, struct set *resource_set,
          if (type == iface) {
             const int stage_mask =
                build_stageref(shProg, var->name, var->data.mode);
-            if (!add_shader_variable(shProg, resource_set,
+            if (!add_shader_variable(ctx, shProg, resource_set,
                                      stage_mask,
                                      iface, var, var->name, var->type, false,
                                      var->data.location - VARYING_SLOT_VAR0))
@@ -3822,7 +3862,9 @@ add_packed_varyings(struct gl_shader_program *shProg, struct set *resource_set,
 }
 
 static bool
-add_fragdata_arrays(struct gl_shader_program *shProg, struct set *resource_set)
+add_fragdata_arrays(const struct gl_context *ctx,
+                    struct gl_shader_program *shProg,
+                    struct set *resource_set)
 {
    struct gl_linked_shader *sh = shProg->_LinkedShaders[MESA_SHADER_FRAGMENT];
 
@@ -3834,7 +3876,7 @@ add_fragdata_arrays(struct gl_shader_program *shProg, struct set *resource_set)
       if (var) {
          assert(var->data.mode == ir_var_shader_out);
 
-         if (!add_shader_variable(shProg, resource_set,
+         if (!add_shader_variable(ctx, shProg, resource_set,
                                   1 << MESA_SHADER_FRAGMENT,
                                   GL_PROGRAM_OUTPUT, var, var->name, var->type,
                                   true, var->data.location - FRAG_RESULT_DATA0))
@@ -4093,24 +4135,24 @@ build_program_resource_list(struct gl_context *ctx,
 
    /* Program interface needs to expose varyings in case of SSO. */
    if (shProg->SeparateShader) {
-      if (!add_packed_varyings(shProg, resource_set,
+      if (!add_packed_varyings(ctx, shProg, resource_set,
                                input_stage, GL_PROGRAM_INPUT))
          return;
 
-      if (!add_packed_varyings(shProg, resource_set,
+      if (!add_packed_varyings(ctx, shProg, resource_set,
                                output_stage, GL_PROGRAM_OUTPUT))
          return;
    }
 
-   if (!add_fragdata_arrays(shProg, resource_set))
+   if (!add_fragdata_arrays(ctx, shProg, resource_set))
       return;
 
    /* Add inputs and outputs to the resource list. */
-   if (!add_interface_variables(shProg, resource_set,
+   if (!add_interface_variables(ctx, shProg, resource_set,
                                 input_stage, GL_PROGRAM_INPUT))
       return;
 
-   if (!add_interface_variables(shProg, resource_set,
+   if (!add_interface_variables(ctx, shProg, resource_set,
                                 output_stage, GL_PROGRAM_OUTPUT))
       return;
 

src/compiler/glsl/lower_blend_equation_advanced.cpp

@@ -308,12 +308,18 @@ calc_blend_result(ir_factory f,
    f.emit(assign(dst_alpha, swizzle_w(fb)));
    f.emit(if_tree(equal(dst_alpha, imm1(0)),
                      assign(dst_rgb, imm3(0)),
-                     assign(dst_rgb, div(swizzle_xyz(fb), dst_alpha))));
+                     assign(dst_rgb, csel(equal(swizzle_xyz(fb),
+                                                swizzle(fb, SWIZZLE_WWWW, 3)),
+                                          imm3(1),
+                                          div(swizzle_xyz(fb), dst_alpha)))));
 
    f.emit(assign(src_alpha, swizzle_w(src)));
    f.emit(if_tree(equal(src_alpha, imm1(0)),
                      assign(src_rgb, imm3(0)),
-                     assign(src_rgb, div(swizzle_xyz(src), src_alpha))));
+                     assign(src_rgb, csel(equal(swizzle_xyz(src),
+                                                swizzle(src, SWIZZLE_WWWW, 3)),
+                                          imm3(1),
+                                          div(swizzle_xyz(src), src_alpha)))));
 
    ir_variable *factor = f.make_temp(glsl_type::vec3_type, "__blend_factor");
 

src/compiler/glsl/lower_named_interface_blocks.cpp

@@ -193,6 +193,8 @@ flatten_named_interface_blocks_declarations::run(exec_list *instructions)
             new_var->data.patch = iface_t->fields.structure[i].patch;
             new_var->data.stream = var->data.stream;
             new_var->data.how_declared = var->data.how_declared;
+            new_var->data.tess_varying_implicit_sized_array =
+               var->data.tess_varying_implicit_sized_array;
             new_var->data.from_named_ifc_block = 1;
 
             new_var->init_interface_type(var->type);

src/egl/drivers/dri2/egl_dri2.c

@@ -241,6 +241,15 @@ dri2_add_config(_EGLDisplay *disp, const __DRIconfig *dri_config, int id,
             return NULL;
          break;
 
+      case __DRI_ATTRIB_MAX_PBUFFER_WIDTH:
+         _eglSetConfigKey(&base, EGL_MAX_PBUFFER_WIDTH,
+                          _EGL_MAX_PBUFFER_WIDTH);
+         break;
+      case __DRI_ATTRIB_MAX_PBUFFER_HEIGHT:
+         _eglSetConfigKey(&base, EGL_MAX_PBUFFER_HEIGHT,
+                          _EGL_MAX_PBUFFER_HEIGHT);
+         break;
+
       default:
          key = dri2_to_egl_attribute_map[attrib];
          if (key != 0)

src/egl/drivers/dri2/egl_dri2_fallbacks.h

@@ -66,7 +66,8 @@ dri2_fallback_swap_buffers_with_damage(_EGLDriver *drv, _EGLDisplay *dpy,
                                       _EGLSurface *surf,
                                       const EGLint *rects, EGLint n_rects)
 {
-   return EGL_FALSE;
+   struct dri2_egl_display *dri2_dpy = dri2_egl_display(dpy);
+   return dri2_dpy->vtbl->swap_buffers(drv, dpy, surf);
 }
 
 static inline EGLBoolean

src/egl/drivers/dri2/platform_android.c

@@ -766,8 +766,6 @@ droid_add_configs_for_visuals(_EGLDriver *drv, _EGLDisplay *dpy)
      EGL_NATIVE_VISUAL_TYPE, 0,
      EGL_FRAMEBUFFER_TARGET_ANDROID, EGL_TRUE,
      EGL_RECORDABLE_ANDROID, EGL_TRUE,
-     EGL_MAX_PBUFFER_WIDTH, _EGL_MAX_PBUFFER_WIDTH,
-     EGL_MAX_PBUFFER_HEIGHT, _EGL_MAX_PBUFFER_HEIGHT,
      EGL_NONE
    };
    unsigned int format_count[ARRAY_SIZE(visuals)] = { 0 };

src/egl/main/eglapi.c

@@ -734,7 +734,9 @@ eglCreateContext(EGLDisplay dpy, EGLConfig config, EGLContext share_list,
 
    _EGL_CHECK_DISPLAY(disp, EGL_NO_CONTEXT, drv);
 
-   if (!config && !disp->Extensions.KHR_no_config_context)
+   if (config != EGL_NO_CONFIG_KHR)
+      _EGL_CHECK_CONFIG(disp, conf, EGL_NO_CONTEXT, drv);
+   else if (!disp->Extensions.KHR_no_config_context)
       RETURN_EGL_ERROR(disp, EGL_BAD_CONFIG, EGL_NO_CONTEXT);
 
    if (!share && share_list != EGL_NO_CONTEXT)

src/egl/main/eglsurface.c

@@ -262,9 +262,13 @@ _eglInitSurface(_EGLSurface *surf, _EGLDisplay *dpy, EGLint type,
 {
    const char *func;
    EGLint renderBuffer = EGL_BACK_BUFFER;
-   EGLint swapBehavior = EGL_BUFFER_PRESERVED;
+   EGLint swapBehavior = EGL_BUFFER_DESTROYED;
    EGLint err;
 
+   /* Swap behavior can be preserved only if config supports this. */
+   if (conf->SurfaceType & EGL_SWAP_BEHAVIOR_PRESERVED_BIT)
+      swapBehavior = EGL_BUFFER_PRESERVED;
+
    switch (type) {
    case EGL_WINDOW_BIT:
       func = "eglCreateWindowSurface";

src/gallium/drivers/nouveau/codegen/nv50_ir_emit_gk110.cpp

@@ -727,7 +727,7 @@ void
 CodeEmitterGK110::emitIMAD(const Instruction *i)
 {
    uint8_t addOp =
-      (i->src(2).mod.neg() << 1) | (i->src(0).mod.neg() ^ i->src(1).mod.neg());
+      i->src(2).mod.neg() | ((i->src(0).mod.neg() ^ i->src(1).mod.neg()) << 1);
 
    emitForm_21(i, 0x100, 0xa00);
 
@@ -760,7 +760,7 @@ CodeEmitterGK110::emitISAD(const Instruction *i)
 void
 CodeEmitterGK110::emitSHLADD(const Instruction *i)
 {
-   uint8_t addOp = (i->src(2).mod.neg() << 1) | i->src(0).mod.neg();
+   uint8_t addOp = (i->src(0).mod.neg() << 1) | i->src(2).mod.neg();
    const ImmediateValue *imm = i->src(1).get()->asImm();
    assert(imm);
 

src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp

@@ -737,7 +737,7 @@ void
 CodeEmitterNVC0::emitIMAD(const Instruction *i)
 {
    uint8_t addOp =
-      (i->src(2).mod.neg() << 1) | (i->src(0).mod.neg() ^ i->src(1).mod.neg());
+      i->src(2).mod.neg() | ((i->src(0).mod.neg() ^ i->src(1).mod.neg()) << 1);
 
    assert(i->encSize == 8);
    emitForm_A(i, HEX64(20000000, 00000003));
@@ -762,7 +762,7 @@ CodeEmitterNVC0::emitIMAD(const Instruction *i)
 void
 CodeEmitterNVC0::emitSHLADD(const Instruction *i)
 {
-   uint8_t addOp = (i->src(2).mod.neg() << 1) | i->src(0).mod.neg();
+   uint8_t addOp = (i->src(0).mod.neg() << 1) | i->src(2).mod.neg();
    const ImmediateValue *imm = i->src(1).get()->asImm();
    assert(imm);
 

src/gallium/drivers/nouveau/nvc0/nvc0_tex.c

@@ -607,7 +607,7 @@ void nvc0_validate_textures(struct nvc0_context *nvc0)
 
    /* Invalidate all CP textures because they are aliased. */
    for (int i = 0; i < nvc0->num_textures[5]; i++)
-      nouveau_bufctx_reset(nvc0->bufctx_3d, NVC0_BIND_CP_TEX(i));
+      nouveau_bufctx_reset(nvc0->bufctx_cp, NVC0_BIND_CP_TEX(i));
    nvc0->textures_dirty[5] = ~0;
    nvc0->dirty_cp |= NVC0_NEW_CP_TEXTURES;
 }

src/gallium/drivers/radeon/r600_texture.c

@@ -826,8 +826,9 @@ static void r600_texture_alloc_cmask_separate(struct r600_common_screen *rscreen
 	}
 
 	rtex->cmask_buffer = (struct r600_resource *)
-		pipe_buffer_create(&rscreen->b, PIPE_BIND_CUSTOM,
-				   PIPE_USAGE_DEFAULT, rtex->cmask.size);
+		r600_aligned_buffer_create(&rscreen->b, 0, PIPE_USAGE_DEFAULT,
+					   rtex->cmask.size,
+					   rtex->cmask.alignment);
 	if (rtex->cmask_buffer == NULL) {
 		rtex->cmask.size = 0;
 		return;
@@ -2442,29 +2443,29 @@ static void si_set_optimal_micro_tile_mode(struct r600_common_screen *rscreen,
 		switch (rtex->last_msaa_resolve_target_micro_mode) {
 		case 0: /* displayable */
 			switch (rtex->surface.bpe) {
-			case 8:
+			case 1:
                             rtex->surface.tiling_index[0] = 10;
                             break;
-			case 16:
+			case 2:
                             rtex->surface.tiling_index[0] = 11;
                             break;
-			default: /* 32, 64 */
+			default: /* 4, 8 */
                             rtex->surface.tiling_index[0] = 12;
                             break;
 			}
 			break;
 		case 1: /* thin */
 			switch (rtex->surface.bpe) {
-			case 8:
+			case 1:
                                 rtex->surface.tiling_index[0] = 14;
                                 break;
-			case 16:
+			case 2:
                                 rtex->surface.tiling_index[0] = 15;
                                 break;
-			case 32:
+			case 4:
                                 rtex->surface.tiling_index[0] = 16;
                                 break;
-			default: /* 64, 128 */
+			default: /* 8, 16 */
                                 rtex->surface.tiling_index[0] = 17;
                                 break;
 			}

src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c

@@ -491,23 +491,32 @@ static void emit_lsb(const struct lp_build_tgsi_action *action,
 		     struct lp_build_emit_data *emit_data)
 {
 	struct gallivm_state *gallivm = bld_base->base.gallivm;
+	LLVMBuilderRef builder = gallivm->builder;
 	LLVMValueRef args[2] = {
 		emit_data->args[0],
 
 		/* The value of 1 means that ffs(x=0) = undef, so LLVM won't
 		 * add special code to check for x=0. The reason is that
 		 * the LLVM behavior for x=0 is different from what we
-		 * need here.
-		 *
-		 * The hardware already implements the correct behavior.
+		 * need here. However, LLVM also assumes that ffs(x) is
+		 * in [0, 31], but GLSL expects that ffs(0) = -1, so
+		 * a conditional assignment to handle 0 is still required.
 		 */
 		LLVMConstInt(LLVMInt1TypeInContext(gallivm->context), 1, 0)
 	};
 
-	emit_data->output[emit_data->chan] =
+	LLVMValueRef lsb =
 		lp_build_intrinsic(gallivm->builder, "llvm.cttz.i32",
 				emit_data->dst_type, args, ARRAY_SIZE(args),
 				LLVMReadNoneAttribute);
+
+	/* TODO: We need an intrinsic to skip this conditional. */
+	/* Check for zero: */
+	emit_data->output[emit_data->chan] =
+		LLVMBuildSelect(builder,
+				LLVMBuildICmp(builder, LLVMIntEQ, args[0],
+					      bld_base->uint_bld.zero, ""),
+				lp_build_const_int32(gallivm, -1), lsb, "");
 }
 
 /* Find the last bit set. */

src/gallium/drivers/radeonsi/si_state.c

@@ -3925,6 +3925,14 @@ static void si_init_config(struct si_context *sctx)
 	si_pm4_set_reg(pm4, R_028408_VGT_INDX_OFFSET, 0);
 
 	if (sctx->b.chip_class >= CIK) {
+		/* If this is 0, Bonaire can hang even if GS isn't being used.
+		 * Other chips are unaffected. These are suboptimal values,
+		 * but we don't use on-chip GS.
+		 */
+		si_pm4_set_reg(pm4, R_028A44_VGT_GS_ONCHIP_CNTL,
+			       S_028A44_ES_VERTS_PER_SUBGRP(64) |
+			       S_028A44_GS_PRIMS_PER_SUBGRP(4));
+
 		si_pm4_set_reg(pm4, R_00B51C_SPI_SHADER_PGM_RSRC3_LS, S_00B51C_CU_EN(0xffff));
 		si_pm4_set_reg(pm4, R_00B41C_SPI_SHADER_PGM_RSRC3_HS, 0);
 		si_pm4_set_reg(pm4, R_00B31C_SPI_SHADER_PGM_RSRC3_ES, S_00B31C_CU_EN(0xffff));

src/gallium/state_trackers/nine/nine_state.c

@@ -126,7 +126,6 @@ prepare_vs_constants_userbuf_swvp(struct NineDevice9 *device)
         cb.user_buffer = state->vs_const_i;
 
         state->pipe.cb2_swvp = cb;
-        state->changed.vs_const_i = 0;
     }
 
     if (state->changed.vs_const_b || state->changed.group & NINE_STATE_SWVP) {
@@ -138,7 +137,6 @@ prepare_vs_constants_userbuf_swvp(struct NineDevice9 *device)
         cb.user_buffer = state->vs_const_b;
 
         state->pipe.cb3_swvp = cb;
-        state->changed.vs_const_b = 0;
     }
 
     if (!device->driver_caps.user_cbufs) {
@@ -236,14 +234,30 @@ prepare_vs_constants_userbuf(struct NineDevice9 *device)
     if (state->changed.vs_const_i || state->changed.group & NINE_STATE_SWVP) {
         int *idst = (int *)&state->vs_const_f[4 * device->max_vs_const_f];
         memcpy(idst, state->vs_const_i, NINE_MAX_CONST_I * sizeof(int[4]));
-        state->changed.vs_const_i = 0;
     }
 
     if (state->changed.vs_const_b || state->changed.group & NINE_STATE_SWVP) {
         int *idst = (int *)&state->vs_const_f[4 * device->max_vs_const_f];
         uint32_t *bdst = (uint32_t *)&idst[4 * NINE_MAX_CONST_I];
         memcpy(bdst, state->vs_const_b, NINE_MAX_CONST_B * sizeof(BOOL));
-        state->changed.vs_const_b = 0;
+    }
+
+    if (device->state.changed.vs_const_i) {
+        struct nine_range *r = device->state.changed.vs_const_i;
+        struct nine_range *p = r;
+        while (p->next)
+            p = p->next;
+        nine_range_pool_put_chain(&device->range_pool, r, p);
+        device->state.changed.vs_const_i = NULL;
+    }
+
+    if (device->state.changed.vs_const_b) {
+        struct nine_range *r = device->state.changed.vs_const_b;
+        struct nine_range *p = r;
+        while (p->next)
+            p = p->next;
+        nine_range_pool_put_chain(&device->range_pool, r, p);
+        device->state.changed.vs_const_b = NULL;
     }
 
     if (!cb.buffer_size)
@@ -290,23 +304,6 @@ prepare_vs_constants_userbuf(struct NineDevice9 *device)
         device->state.changed.vs_const_f = NULL;
     }
 
-    if (device->state.changed.vs_const_i) {
-        struct nine_range *r = device->state.changed.vs_const_i;
-        struct nine_range *p = r;
-        while (p->next)
-            p = p->next;
-        nine_range_pool_put_chain(&device->range_pool, r, p);
-        device->state.changed.vs_const_i = NULL;
-    }
-
-    if (device->state.changed.vs_const_b) {
-        struct nine_range *r = device->state.changed.vs_const_b;
-        struct nine_range *p = r;
-        while (p->next)
-            p = p->next;
-        nine_range_pool_put_chain(&device->range_pool, r, p);
-        device->state.changed.vs_const_b = NULL;
-    }
     state->changed.group &= ~NINE_STATE_VS_CONST;
     state->commit |= NINE_STATE_COMMIT_CONST_VS;
 }

src/gallium/state_trackers/nine/surface9.c

@@ -447,6 +447,7 @@ NineSurface9_LockRect( struct NineSurface9 *This,
     } else {
         u_box_origin_2d(This->desc.Width, This->desc.Height, &box);
     }
+    box.z = This->layer;
 
     user_warn(This->desc.Format == D3DFMT_NULL);
 

src/gallium/state_trackers/nine/volume9.c

@@ -373,7 +373,7 @@ NineVolume9_UnlockBox( struct NineVolume9 *This )
                                         This->layer_stride_conversion,
                                         0, 0, 0,
                                         This->desc.Width, This->desc.Height,
-                                        This->desc.Height);
+                                        This->desc.Depth);
 
         if (!This->data)
             pipe_transfer_unmap(This->pipe, transfer);

src/gallium/state_trackers/omx/vid_dec.c

@@ -425,24 +425,19 @@ void vid_dec_NeedTarget(vid_dec_PrivateType *priv)
    struct pipe_video_buffer templat = {};
    struct vl_screen *omx_screen;
    struct pipe_screen *pscreen;
-   omx_base_video_PortType *port;
 
    omx_screen = priv->screen;
-   port = (omx_base_video_PortType *)priv->ports[OMX_BASE_FILTER_INPUTPORT_INDEX];
-
    assert(omx_screen);
-   assert(port);
 
    pscreen = omx_screen->pscreen;
-
    assert(pscreen);
 
    if (!priv->target) {
       memset(&templat, 0, sizeof(templat));
 
       templat.chroma_format = PIPE_VIDEO_CHROMA_FORMAT_420;
-      templat.width = port->sPortParam.format.video.nFrameWidth;
-      templat.height = port->sPortParam.format.video.nFrameHeight;
+      templat.width = priv->codec->width;
+      templat.height = priv->codec->height;
       templat.buffer_format = pscreen->get_video_param(
             pscreen,
             PIPE_VIDEO_PROFILE_UNKNOWN,
@@ -626,7 +621,7 @@ static void vid_dec_FrameDecoded(OMX_COMPONENTTYPE *comp, OMX_BUFFERHEADERTYPE*
    }
 
    if (input->pInputPortPrivate) {
-      if (output->pInputPortPrivate) {
+      if (output->pInputPortPrivate && !priv->disable_tunnel) {
          struct pipe_video_buffer *tmp, *vbuf, *new_vbuf;
 
          tmp = output->pOutputPortPrivate;

src/gallium/state_trackers/omx/vid_dec.h

@@ -129,6 +129,7 @@ DERIVEDCLASS(vid_dec_PrivateType, omx_base_filter_PrivateType)
    bool frame_started; \
    unsigned bytes_left; \
    const void *slice; \
+   bool disable_tunnel; \
    struct vl_compositor compositor; \
    struct vl_compositor_state cstate;
 ENDCLASS(vid_dec_PrivateType)

src/gallium/state_trackers/omx/vid_dec_h264.c

@@ -105,13 +105,6 @@ static void vid_dec_h264_BeginFrame(vid_dec_PrivateType *priv)
    if (priv->frame_started)
       return;
 
-   vid_dec_NeedTarget(priv);
-   if (priv->first_buf_in_frame)
-      priv->timestamp = priv->timestamps[0];
-   priv->first_buf_in_frame = false;
-
-   priv->picture.h264.num_ref_frames = priv->picture.h264.pps->sps->max_num_ref_frames;
-
    if (!priv->codec) {
       struct pipe_video_codec templat = {};
       omx_base_video_PortType *port;
@@ -128,6 +121,15 @@ static void vid_dec_h264_BeginFrame(vid_dec_PrivateType *priv)
 
       priv->codec = priv->pipe->create_video_codec(priv->pipe, &templat);
    }
+
+   vid_dec_NeedTarget(priv);
+
+   if (priv->first_buf_in_frame)
+      priv->timestamp = priv->timestamps[0];
+   priv->first_buf_in_frame = false;
+
+   priv->picture.h264.num_ref_frames = priv->picture.h264.pps->sps->max_num_ref_frames;
+
    priv->picture.h264.slice_count = 0;
    priv->codec->begin_frame(priv->codec, priv->target, &priv->picture.base);
    priv->frame_started = true;

src/gallium/state_trackers/omx/vid_dec_h265.c

@@ -612,13 +612,10 @@ static void vid_dec_h265_BeginFrame(vid_dec_PrivateType *priv)
    if (priv->frame_started)
       return;
 
-   vid_dec_NeedTarget(priv);
-   if (priv->first_buf_in_frame)
-      priv->timestamp = priv->timestamps[0];
-   priv->first_buf_in_frame = false;
-
    if (!priv->codec) {
       struct pipe_video_codec templat = {};
+      omx_base_video_PortType *port = (omx_base_video_PortType *)
+         priv->ports[OMX_BASE_FILTER_INPUTPORT_INDEX];
 
       templat.profile = priv->profile;
       templat.entrypoint = PIPE_VIDEO_ENTRYPOINT_BITSTREAM;
@@ -628,7 +625,21 @@ static void vid_dec_h265_BeginFrame(vid_dec_PrivateType *priv)
       templat.height = priv->codec_data.h265.pic_height_in_luma_samples;
       templat.level =  priv->codec_data.h265.level_idc;
       priv->codec = priv->pipe->create_video_codec(priv->pipe, &templat);
+
+      /* disable transcode tunnel if video size is different from coded size */
+      if (priv->codec_data.h265.pic_width_in_luma_samples !=
+          port->sPortParam.format.video.nFrameWidth ||
+          priv->codec_data.h265.pic_height_in_luma_samples !=
+          port->sPortParam.format.video.nFrameHeight)
+         priv->disable_tunnel = true;
    }
+
+   vid_dec_NeedTarget(priv);
+
+   if (priv->first_buf_in_frame)
+      priv->timestamp = priv->timestamps[0];
+   priv->first_buf_in_frame = false;
+
    priv->codec->begin_frame(priv->codec, priv->target, &priv->picture.base);
    priv->frame_started = true;
 }

src/gallium/winsys/amdgpu/drm/amdgpu_surface.c

@@ -303,6 +303,20 @@ static void set_micro_tile_mode(struct radeon_surf *surf,
       surf->micro_tile_mode = G_009910_MICRO_TILE_MODE(tile_mode);
 }
 
+static unsigned cik_get_macro_tile_index(struct radeon_surf *surf)
+{
+	unsigned index, tileb;
+
+	tileb = 8 * 8 * surf->bpe;
+	tileb = MIN2(surf->tile_split, tileb);
+
+	for (index = 0; tileb > 64; index++)
+		tileb >>= 1;
+
+	assert(index < 16);
+	return index;
+}
+
 static int amdgpu_surface_init(struct radeon_winsys *rws,
                                struct radeon_surf *surf)
 {
@@ -475,6 +489,9 @@ static int amdgpu_surface_init(struct radeon_winsys *rws,
             AddrSurfInfoIn.tileIndex = 10; /* 2D displayable */
          else
             AddrSurfInfoIn.tileIndex = 14; /* 2D non-displayable */
+
+         /* Addrlib doesn't set this if tileIndex is forced like above. */
+         AddrSurfInfoOut.macroModeIndex = cik_get_macro_tile_index(surf);
       }
    }
 

src/intel/blorp/blorp.c

@@ -165,21 +165,15 @@ nir_uniform_type_size(const struct glsl_type *type)
 }
 
 const unsigned *
-brw_blorp_compile_nir_shader(struct blorp_context *blorp, struct nir_shader *nir,
-                             const struct brw_wm_prog_key *wm_key,
-                             bool use_repclear,
-                             struct brw_blorp_prog_data *prog_data,
-                             unsigned *program_size)
+blorp_compile_fs(struct blorp_context *blorp, void *mem_ctx,
+                 struct nir_shader *nir,
+                 const struct brw_wm_prog_key *wm_key,
+                 bool use_repclear,
+                 struct brw_blorp_prog_data *prog_data,
+                 unsigned *program_size)
 {
    const struct brw_compiler *compiler = blorp->compiler;
 
-   void *mem_ctx = ralloc_context(NULL);
-
-   /* Calling brw_preprocess_nir and friends is destructive and, if cloning is
-    * enabled, may end up completely replacing the nir_shader.  Therefore, we
-    * own it and might as well put it in our context for easy cleanup.
-    */
-   ralloc_steal(mem_ctx, nir);
    nir->options =
       compiler->glsl_compiler_options[MESA_SHADER_FRAGMENT].NirOptions;
 

src/intel/blorp/blorp_blit.c

@@ -997,7 +997,7 @@ blorp_nir_manual_blend_bilinear(nir_builder *b, nir_ssa_def *pos,
  * of samples).
  */
 static nir_shader *
-brw_blorp_build_nir_shader(struct blorp_context *blorp,
+brw_blorp_build_nir_shader(struct blorp_context *blorp, void *mem_ctx,
                            const struct brw_blorp_blit_prog_key *key)
 {
    const struct gen_device_info *devinfo = blorp->isl_dev->info;
@@ -1044,7 +1044,7 @@ brw_blorp_build_nir_shader(struct blorp_context *blorp,
           (key->dst_samples <= 1));
 
    nir_builder b;
-   nir_builder_init_simple_shader(&b, NULL, MESA_SHADER_FRAGMENT, NULL);
+   nir_builder_init_simple_shader(&b, mem_ctx, MESA_SHADER_FRAGMENT, NULL);
 
    struct brw_blorp_blit_vars v;
    brw_blorp_blit_vars_init(&b, &v, key);
@@ -1233,6 +1233,8 @@ brw_blorp_get_blit_kernel(struct blorp_context *blorp,
                             &params->wm_prog_kernel, &params->wm_prog_data))
       return;
 
+   void *mem_ctx = ralloc_context(NULL);
+
    const unsigned *program;
    unsigned program_size;
    struct brw_blorp_prog_data prog_data;
@@ -1240,7 +1242,7 @@ brw_blorp_get_blit_kernel(struct blorp_context *blorp,
    /* Try and compile with NIR first.  If that fails, fall back to the old
     * method of building shaders manually.
     */
-   nir_shader *nir = brw_blorp_build_nir_shader(blorp, prog_key);
+   nir_shader *nir = brw_blorp_build_nir_shader(blorp, mem_ctx, prog_key);
    struct brw_wm_prog_key wm_key;
    brw_blorp_init_wm_prog_key(&wm_key);
    wm_key.tex.compressed_multisample_layout_mask =
@@ -1248,13 +1250,15 @@ brw_blorp_get_blit_kernel(struct blorp_context *blorp,
    wm_key.tex.msaa_16 = prog_key->tex_samples == 16;
    wm_key.multisample_fbo = prog_key->rt_samples > 1;
 
-   program = brw_blorp_compile_nir_shader(blorp, nir, &wm_key, false,
-                                          &prog_data, &program_size);
+   program = blorp_compile_fs(blorp, mem_ctx, nir, &wm_key, false,
+                              &prog_data, &program_size);
 
    blorp->upload_shader(blorp, prog_key, sizeof(*prog_key),
                         program, program_size,
                         &prog_data, sizeof(prog_data),
                         &params->wm_prog_kernel, &params->wm_prog_data);
+
+   ralloc_free(mem_ctx);
 }
 
 static void
@@ -1761,10 +1765,14 @@ surf_convert_to_uncompressed(const struct isl_device *isl_dev,
    surf_convert_to_single_slice(isl_dev, info);
 
    if (width || height) {
+#ifndef NDEBUG
+      uint32_t right_edge_px = info->tile_x_sa + *x + *width;
+      uint32_t bottom_edge_px = info->tile_y_sa + *y + *height;
       assert(*width % fmtl->bw == 0 ||
-             *x + *width == info->surf.logical_level0_px.width);
+             right_edge_px == info->surf.logical_level0_px.width);
       assert(*height % fmtl->bh == 0 ||
-             *y + *height == info->surf.logical_level0_px.height);
+             bottom_edge_px == info->surf.logical_level0_px.height);
+#endif
       *width = DIV_ROUND_UP(*width, fmtl->bw);
       *height = DIV_ROUND_UP(*height, fmtl->bh);
    }

src/intel/blorp/blorp_clear.c

@@ -56,6 +56,7 @@ blorp_params_get_clear_kernel(struct blorp_context *blorp,
 
    nir_builder b;
    nir_builder_init_simple_shader(&b, NULL, MESA_SHADER_FRAGMENT, NULL);
+   nir_builder_init_simple_shader(&b, mem_ctx, MESA_SHADER_FRAGMENT, NULL);
    b.shader->info.name = ralloc_strdup(b.shader, "BLORP-clear");
 
    nir_variable *v_color = nir_variable_create(b.shader, nir_var_shader_in,
@@ -76,8 +77,8 @@ blorp_params_get_clear_kernel(struct blorp_context *blorp,
    struct brw_blorp_prog_data prog_data;
    unsigned program_size;
    const unsigned *program =
-      brw_blorp_compile_nir_shader(blorp, b.shader, &wm_key, use_replicated_data,
-                                   &prog_data, &program_size);
+      blorp_compile_fs(blorp, mem_ctx, b.shader, &wm_key, use_replicated_data,
+                       &prog_data, &program_size);
 
    blorp->upload_shader(blorp, &blorp_key, sizeof(blorp_key),
                         program, program_size,

src/intel/blorp/blorp_priv.h

@@ -310,11 +310,12 @@ struct brw_blorp_blit_prog_key
 void brw_blorp_init_wm_prog_key(struct brw_wm_prog_key *wm_key);
 
 const unsigned *
-brw_blorp_compile_nir_shader(struct blorp_context *blorp, struct nir_shader *nir,
-                             const struct brw_wm_prog_key *wm_key,
-                             bool use_repclear,
-                             struct brw_blorp_prog_data *prog_data,
-                             unsigned *program_size);
+blorp_compile_fs(struct blorp_context *blorp, void *mem_ctx,
+                 struct nir_shader *nir,
+                 const struct brw_wm_prog_key *wm_key,
+                 bool use_repclear,
+                 struct brw_blorp_prog_data *prog_data,
+                 unsigned *program_size);
 
 /** \} */
 

src/intel/vulkan/dev_icd.json.in

@@ -2,6 +2,6 @@
     "file_format_version": "1.0.0",
     "ICD": {
         "library_path": "@build_libdir@/libvulkan_intel.so",
-        "abi_versions": "1.0.3"
+        "api_version": "1.0.3"
     }
 }

src/intel/vulkan/intel_icd.json.in

@@ -2,6 +2,6 @@
     "file_format_version": "1.0.0",
     "ICD": {
         "library_path": "@install_libdir@/libvulkan_intel.so",
-        "abi_versions": "1.0.3"
+        "api_version": "1.0.3"
     }
 }

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

@@ -492,19 +492,6 @@ type_size_scalar(const struct glsl_type *type)
    return 0;
 }
 
-/**
- * Returns the number of scalar components needed to store type, assuming
- * that vectors are padded out to vec4.
- *
- * This has the packing rules of type_size_vec4(), but counts components
- * similar to type_size_scalar().
- */
-extern "C" int
-type_size_vec4_times_4(const struct glsl_type *type)
-{
-   return 4 * type_size_vec4(type);
-}
-
 /* Attribute arrays are loaded as one vec4 per element (or matrix column),
  * except for double-precision types, which are loaded as one dvec4.
  */
@@ -1687,6 +1674,12 @@ fs_visitor::assign_gs_urb_setup()
 void
 fs_visitor::split_virtual_grfs()
 {
+   /* Compact the register file so we eliminate dead vgrfs.  This
+    * only defines split points for live registers, so if we have
+    * too large dead registers they will hit assertions later.
+    */
+   compact_virtual_grfs();
+
    int num_vars = this->alloc.count;
 
    /* Count the total number of registers */

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

@@ -324,8 +324,6 @@ public:
 
    fs_reg *nir_locals;
    fs_reg *nir_ssa_values;
-   fs_reg nir_inputs;
-   fs_reg nir_outputs;
    fs_reg *nir_system_values;
 
    bool failed;

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

@@ -400,34 +400,33 @@ fs_generator::generate_mov_indirect(fs_inst *inst,
       indirect_byte_offset =
          retype(spread(indirect_byte_offset, 2), BRW_REGISTER_TYPE_UW);
 
-      struct brw_reg ind_src;
-      if (devinfo->gen < 8) {
-         /* From the Haswell PRM section "Register Region Restrictions":
-          *
-          *    "The lower bits of the AddressImmediate must not overflow to
-          *    change the register address.  The lower 5 bits of Address
-          *    Immediate when added to lower 5 bits of address register gives
-          *    the sub-register offset. The upper bits of Address Immediate
-          *    when added to upper bits of address register gives the register
-          *    address. Any overflow from sub-register offset is dropped."
-          *
-          * This restriction is only listed in the Haswell PRM but emperical
-          * testing indicates that it applies on all older generations and is
-          * lifted on Broadwell.
-          *
-          * Since the indirect may cause us to cross a register boundary, this
-          * makes the base offset almost useless.  We could try and do
-          * something clever where we use a actual base offset if
-          * base_offset % 32 == 0 but that would mean we were generating
-          * different code depending on the base offset.  Instead, for the
-          * sake of consistency, we'll just do the add ourselves.
-          */
-         brw_ADD(p, addr, indirect_byte_offset, brw_imm_uw(imm_byte_offset));
-         ind_src = brw_VxH_indirect(0, 0);
-      } else {
-         brw_MOV(p, addr, indirect_byte_offset);
-         ind_src = brw_VxH_indirect(0, imm_byte_offset);
-      }
+      /* There are a number of reasons why we don't use the base offset here.
+       * One reason is that the field is only 9 bits which means we can only
+       * use it to access the first 16 GRFs.  Also, from the Haswell PRM
+       * section "Register Region Restrictions":
+       *
+       *    "The lower bits of the AddressImmediate must not overflow to
+       *    change the register address.  The lower 5 bits of Address
+       *    Immediate when added to lower 5 bits of address register gives
+       *    the sub-register offset. The upper bits of Address Immediate
+       *    when added to upper bits of address register gives the register
+       *    address. Any overflow from sub-register offset is dropped."
+       *
+       * Since the indirect may cause us to cross a register boundary, this
+       * makes the base offset almost useless.  We could try and do something
+       * clever where we use a actual base offset if base_offset % 32 == 0 but
+       * that would mean we were generating different code depending on the
+       * base offset.  Instead, for the sake of consistency, we'll just do the
+       * add ourselves.  This restriction is only listed in the Haswell PRM
+       * but empirical testing indicates that it applies on all older
+       * generations and is lifted on Broadwell.
+       *
+       * In the end, while base_offset is nice to look at in the generated
+       * code, using it saves us 0 instructions and would require quite a bit
+       * of case-by-case work.  It's just not worth it.
+       */
+      brw_ADD(p, addr, indirect_byte_offset, brw_imm_uw(imm_byte_offset));
+      struct brw_reg ind_src = brw_VxH_indirect(0, 0);
 
       brw_inst *mov = brw_MOV(p, dst, retype(ind_src, dst.type));
 

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

@@ -48,56 +48,18 @@ fs_visitor::emit_nir_code()
    }
 }
 
-void
-fs_visitor::nir_setup_single_output_varying(fs_reg *reg,
-                                            const glsl_type *type,
-                                            unsigned *location)
-{
-   if (type->is_array() || type->is_matrix()) {
-      const struct glsl_type *elem_type = glsl_get_array_element(type);
-      const unsigned length = glsl_get_length(type);
-
-      for (unsigned i = 0; i < length; i++) {
-         nir_setup_single_output_varying(reg, elem_type, location);
-      }
-   } else if (type->is_record()) {
-      for (unsigned i = 0; i < type->length; i++) {
-         const struct glsl_type *field_type = type->fields.structure[i].type;
-         nir_setup_single_output_varying(reg, field_type, location);
-      }
-   } else {
-      assert(type->is_scalar() || type->is_vector());
-      unsigned num_iter = 1;
-      if (type->is_dual_slot())
-         num_iter = 2;
-      for (unsigned count = 0; count < num_iter; count++) {
-         this->outputs[*location] = *reg;
-         *reg = offset(*reg, bld, 4);
-         (*location)++;
-      }
-   }
-}
-
 void
 fs_visitor::nir_setup_outputs()
 {
    if (stage == MESA_SHADER_TESS_CTRL || stage == MESA_SHADER_FRAGMENT)
       return;
 
-   nir_outputs = bld.vgrf(BRW_REGISTER_TYPE_F, nir->num_outputs);
-
    nir_foreach_variable(var, &nir->outputs) {
-      switch (stage) {
-      case MESA_SHADER_VERTEX:
-      case MESA_SHADER_TESS_EVAL:
-      case MESA_SHADER_GEOMETRY: {
-         fs_reg reg = offset(nir_outputs, bld, var->data.driver_location);
-         unsigned location = var->data.location;
-         nir_setup_single_output_varying(&reg, var->type, &location);
-         break;
-      }
-      default:
-         unreachable("unhandled shader stage");
+      const unsigned vec4s = type_size_vec4(var->type);
+      fs_reg reg = bld.vgrf(BRW_REGISTER_TYPE_F, 4 * vec4s);
+      for (unsigned i = 0; i < vec4s; i++) {
+         if (outputs[var->data.driver_location + i].file == BAD_FILE)
+            outputs[var->data.driver_location + i] = offset(reg, bld, 4 * i);
       }
    }
 }
@@ -4251,12 +4213,11 @@ fs_visitor::nir_emit_intrinsic(const fs_builder &bld, nir_intrinsic_instr *instr
 
    case nir_intrinsic_store_output: {
       fs_reg src = get_nir_src(instr->src[0]);
-      fs_reg new_dest = offset(retype(nir_outputs, src.type), bld,
-                               instr->const_index[0]);
 
       nir_const_value *const_offset = nir_src_as_const_value(instr->src[1]);
       assert(const_offset && "Indirect output stores not allowed");
-      new_dest = offset(new_dest, bld, const_offset->u32[0]);
+      fs_reg new_dest = retype(offset(outputs[instr->const_index[0]], bld,
+                                      4 * const_offset->u32[0]), src.type);
 
       unsigned num_components = instr->num_components;
       unsigned first_component = nir_intrinsic_component(instr);

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

@@ -327,16 +327,11 @@ void
 brw_nir_lower_vue_outputs(nir_shader *nir,
                           bool is_scalar)
 {
-   if (is_scalar) {
-      nir_assign_var_locations(&nir->outputs, &nir->num_outputs,
-                               VARYING_SLOT_VAR0,
-                               type_size_vec4_times_4);
-      nir_lower_io(nir, nir_var_shader_out, type_size_vec4_times_4, 0);
-   } else {
-      nir_foreach_variable(var, &nir->outputs)
-         var->data.driver_location = var->data.location;
-      nir_lower_io(nir, nir_var_shader_out, type_size_vec4, 0);
+   nir_foreach_variable(var, &nir->outputs) {
+      var->data.driver_location = var->data.location;
    }
+
+   nir_lower_io(nir, nir_var_shader_out, type_size_vec4, 0);
 }
 
 void

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

@@ -34,7 +34,6 @@ extern "C" {
 int type_size_scalar(const struct glsl_type *type);
 int type_size_vec4(const struct glsl_type *type);
 int type_size_dvec4(const struct glsl_type *type);
-int type_size_vec4_times_4(const struct glsl_type *type);
 int type_size_vs_input(const struct glsl_type *type);
 
 static inline int

src/mesa/main/buffers.c

@@ -389,17 +389,48 @@ draw_buffers(struct gl_context *ctx, struct gl_framebuffer *fb,
 
    /* complicated error checking... */
    for (output = 0; output < n; output++) {
-      /* Section 4.2 (Whole Framebuffer Operations) of the OpenGL 3.0
+      destMask[output] = draw_buffer_enum_to_bitmask(ctx, buffers[output]);
+
+      /* From the OpenGL 3.0 specification, page 258:
+       * "Each buffer listed in bufs must be one of the values from tables
+       *  4.5 or 4.6.  Otherwise, an INVALID_ENUM error is generated.
+       */
+      if (destMask[output] == BAD_MASK) {
+         _mesa_error(ctx, GL_INVALID_ENUM, "%s(invalid buffer %s)",
+                     caller, _mesa_enum_to_string(buffers[output]));
+         return;
+      }
+
+      /* From the OpenGL 4.0 specification, page 256:
+       * "For both the default framebuffer and framebuffer objects, the
+       *  constants FRONT, BACK, LEFT, RIGHT, and FRONT_AND_BACK are not
+       *  valid in the bufs array passed to DrawBuffers, and will result in
+       *  the error INVALID_ENUM. This restriction is because these
+       *  constants may themselves refer to multiple buffers, as shown in
+       *  table 4.4."
+       *  Previous versions of the OpenGL specification say INVALID_OPERATION,
+       *  but the Khronos conformance tests expect INVALID_ENUM.
+       */
+      if (_mesa_bitcount(destMask[output]) > 1) {
+         _mesa_error(ctx, GL_INVALID_ENUM, "%s(invalid buffer %s)",
+                     caller, _mesa_enum_to_string(buffers[output]));
+         return;
+      }
+
+      /* Section 4.2 (Whole Framebuffer Operations) of the OpenGL ES 3.0
        * specification says:
        *
-       *     "Each buffer listed in bufs must be BACK, NONE, or one of the values
-       *      from table 4.3 (NONE, COLOR_ATTACHMENTi)"
+       *     "If the GL is bound to a draw framebuffer object, the ith buffer
+       *     listed in bufs must be COLOR_ATTACHMENTi or NONE . Specifying a
+       *     buffer out of order, BACK , or COLOR_ATTACHMENTm where m is greater
+       *     than or equal to the value of MAX_- COLOR_ATTACHMENTS , will
+       *     generate the error INVALID_OPERATION .
        */
-      if (_mesa_is_gles3(ctx) && buffers[output] != GL_NONE &&
-          buffers[output] != GL_BACK &&
+      if (_mesa_is_gles3(ctx) && _mesa_is_user_fbo(fb) &&
+          buffers[output] != GL_NONE &&
           (buffers[output] < GL_COLOR_ATTACHMENT0 ||
            buffers[output] >= GL_COLOR_ATTACHMENT0 + ctx->Const.MaxColorAttachments)) {
-         _mesa_error(ctx, GL_INVALID_ENUM, "glDrawBuffers(buffer)");
+         _mesa_error(ctx, GL_INVALID_OPERATION, "glDrawBuffers(buffer)");
          return;
       }
 
@@ -423,34 +454,6 @@ draw_buffers(struct gl_context *ctx, struct gl_framebuffer *fb,
             return;
          }
 
-         destMask[output] = draw_buffer_enum_to_bitmask(ctx, buffers[output]);
-
-         /* From the OpenGL 3.0 specification, page 258:
-          * "Each buffer listed in bufs must be one of the values from tables
-          *  4.5 or 4.6.  Otherwise, an INVALID_ENUM error is generated.
-          */
-         if (destMask[output] == BAD_MASK) {
-            _mesa_error(ctx, GL_INVALID_ENUM, "%s(invalid buffer %s)",
-                        caller, _mesa_enum_to_string(buffers[output]));
-            return;
-         }
-
-         /* From the OpenGL 4.0 specification, page 256:
-          * "For both the default framebuffer and framebuffer objects, the
-          *  constants FRONT, BACK, LEFT, RIGHT, and FRONT_AND_BACK are not
-          *  valid in the bufs array passed to DrawBuffers, and will result in
-          *  the error INVALID_ENUM. This restriction is because these
-          *  constants may themselves refer to multiple buffers, as shown in
-          *  table 4.4."
-          *  Previous versions of the OpenGL specification say INVALID_OPERATION,
-          *  but the Khronos conformance tests expect INVALID_ENUM.
-          */
-         if (_mesa_bitcount(destMask[output]) > 1) {
-            _mesa_error(ctx, GL_INVALID_ENUM, "%s(invalid buffer %s)",
-                        caller, _mesa_enum_to_string(buffers[output]));
-            return;
-         }
-
          /* From the OpenGL 3.0 specification, page 259:
           * "If the GL is bound to the default framebuffer and DrawBuffers is
           *  supplied with a constant (other than NONE) that does not indicate

src/mesa/program/program_lexer.l

@@ -123,7 +123,7 @@ handle_ident(struct asm_parser_state *state, const char *text, YYSTYPE *lval)
 {
    lval->string = strdup(text);
 
-   return (_mesa_symbol_table_find_symbol(state->st, 0, text) == NULL)
+   return (_mesa_symbol_table_find_symbol(state->st, text) == NULL)
       ? IDENTIFIER : USED_IDENTIFIER;
 }
 

src/mesa/program/program_parse.y

@@ -724,7 +724,7 @@ extSwizSel: INTEGER
 srcReg: USED_IDENTIFIER /* temporaryReg | progParamSingle */
 	{
 	   struct asm_symbol *const s = (struct asm_symbol *)
-	      _mesa_symbol_table_find_symbol(state->st, 0, $1);
+              _mesa_symbol_table_find_symbol(state->st, $1);
 
 	   free($1);
 
@@ -812,7 +812,7 @@ dstReg: resultBinding
 	| USED_IDENTIFIER /* temporaryReg | vertexResultReg */
 	{
 	   struct asm_symbol *const s = (struct asm_symbol *)
-	      _mesa_symbol_table_find_symbol(state->st, 0, $1);
+              _mesa_symbol_table_find_symbol(state->st, $1);
 
 	   free($1);
 
@@ -841,7 +841,7 @@ dstReg: resultBinding
 progParamArray: USED_IDENTIFIER
 	{
 	   struct asm_symbol *const s = (struct asm_symbol *)
-	      _mesa_symbol_table_find_symbol(state->st, 0, $1);
+              _mesa_symbol_table_find_symbol(state->st, $1);
 
 	   free($1);
 
@@ -914,7 +914,7 @@ addrRegNegOffset: INTEGER
 addrReg: USED_IDENTIFIER
 	{
 	   struct asm_symbol *const s = (struct asm_symbol *)
-	      _mesa_symbol_table_find_symbol(state->st, 0, $1);
+              _mesa_symbol_table_find_symbol(state->st, $1);
 
 	   free($1);
 
@@ -2028,9 +2028,9 @@ legacyTexUnitNum: INTEGER
 ALIAS_statement: ALIAS IDENTIFIER '=' USED_IDENTIFIER
 	{
 	   struct asm_symbol *exist = (struct asm_symbol *)
-	      _mesa_symbol_table_find_symbol(state->st, 0, $2);
+              _mesa_symbol_table_find_symbol(state->st, $2);
 	   struct asm_symbol *target = (struct asm_symbol *)
-	      _mesa_symbol_table_find_symbol(state->st, 0, $4);
+              _mesa_symbol_table_find_symbol(state->st, $4);
 
 	   free($4);
 
@@ -2046,7 +2046,7 @@ ALIAS_statement: ALIAS IDENTIFIER '=' USED_IDENTIFIER
 		      "undefined variable binding in ALIAS statement");
 	      YYERROR;
 	   } else {
-	      _mesa_symbol_table_add_symbol(state->st, 0, $2, target);
+              _mesa_symbol_table_add_symbol(state->st, $2, target);
 	   }
 	}
 	;
@@ -2235,7 +2235,7 @@ declare_variable(struct asm_parser_state *state, char *name, enum asm_type t,
 {
    struct asm_symbol *s = NULL;
    struct asm_symbol *exist = (struct asm_symbol *)
-      _mesa_symbol_table_find_symbol(state->st, 0, name);
+      _mesa_symbol_table_find_symbol(state->st, name);
 
 
    if (exist != NULL) {
@@ -2273,7 +2273,7 @@ declare_variable(struct asm_parser_state *state, char *name, enum asm_type t,
 	 break;
       }
 
-      _mesa_symbol_table_add_symbol(state->st, 0, s->name, s);
+      _mesa_symbol_table_add_symbol(state->st, s->name, s);
       s->next = state->sym;
       state->sym = s;
    }

src/mesa/program/symbol_table.c

@@ -26,6 +26,9 @@
 #include "../../util/hash_table.h"
 
 struct symbol {
+   /** Symbol name. */
+   char *name;
+
     /**
      * Link to the next symbol in the table with the same name
      *
@@ -34,7 +37,6 @@ struct symbol {
      */
     struct symbol *next_with_same_name;
 
-
     /**
      * Link to the next symbol in the table with the same scope
      *
@@ -43,21 +45,6 @@ struct symbol {
      */
     struct symbol *next_with_same_scope;
 
-
-    /**
-     * Header information for the list of symbols with the same name.
-     */
-    struct symbol_header *hdr;
-
-
-    /**
-     * Name space of the symbol
-     *
-     * Name space are arbitrary user assigned integers.  No two symbols can
-     * exist in the same name space at the same scope level.
-     */
-    int name_space;
-
     /** Scope depth where this symbol was defined. */
     unsigned depth;
 
@@ -68,20 +55,6 @@ struct symbol {
 };
 
 
-/**
- */
-struct symbol_header {
-    /** Linkage in list of all headers in a given symbol table. */
-    struct symbol_header *next;
-
-    /** Symbol name. */
-    char *name;
-
-    /** Linked list of symbols with the same name. */
-    struct symbol *symbols;
-};
-
-
 /**
  * Element of the scope stack.
  */
@@ -104,41 +77,10 @@ struct _mesa_symbol_table {
     /** Top of scope stack. */
     struct scope_level *current_scope;
 
-    /** List of all symbol headers in the table. */
-    struct symbol_header *hdr;
-
     /** Current scope depth. */
     unsigned depth;
 };
 
-
-static void
-check_symbol_table(struct _mesa_symbol_table *table)
-{
-#if !defined(NDEBUG)
-    struct scope_level *scope;
-
-    for (scope = table->current_scope; scope != NULL; scope = scope->next) {
-        struct symbol *sym;
-
-        for (sym = scope->symbols
-             ; sym != NULL
-             ; sym = sym->next_with_same_name) {
-            const struct symbol_header *const hdr = sym->hdr;
-            struct symbol *sym2;
-
-            for (sym2 = hdr->symbols
-                 ; sym2 != NULL
-                 ; sym2 = sym2->next_with_same_name) {
-                assert(sym2->hdr == hdr);
-            }
-        }
-    }
-#else
-    (void) table;
-#endif /* !defined(NDEBUG) */
-}
-
 void
 _mesa_symbol_table_pop_scope(struct _mesa_symbol_table *table)
 {
@@ -152,18 +94,22 @@ _mesa_symbol_table_pop_scope(struct _mesa_symbol_table *table)
 
     while (sym != NULL) {
         struct symbol *const next = sym->next_with_same_scope;
-        struct symbol_header *const hdr = sym->hdr;
-
-        assert(hdr->symbols == sym);
-
-        hdr->symbols = sym->next_with_same_name;
+        struct hash_entry *hte = _mesa_hash_table_search(table->ht,
+                                                         sym->name);
+        if (sym->next_with_same_name) {
+           /* If there is a symbol with this name in an outer scope update
+            * the hash table to point to it.
+            */
+           hte->key = sym->next_with_same_name->name;
+           hte->data = sym->next_with_same_name;
+        } else {
+           _mesa_hash_table_remove(table->ht, hte);
+           free(sym->name);
+        }
 
         free(sym);
-
         sym = next;
     }
-
-    check_symbol_table(table);
 }
 
 
@@ -171,7 +117,6 @@ void
 _mesa_symbol_table_push_scope(struct _mesa_symbol_table *table)
 {
     struct scope_level *const scope = calloc(1, sizeof(*scope));
-    
     if (scope == NULL) {
        _mesa_error_no_memory(__func__);
        return;
@@ -183,11 +128,11 @@ _mesa_symbol_table_push_scope(struct _mesa_symbol_table *table)
 }
 
 
-static struct symbol_header *
+static struct symbol *
 find_symbol(struct _mesa_symbol_table *table, const char *name)
 {
    struct hash_entry *entry = _mesa_hash_table_search(table->ht, name);
-   return entry ? (struct symbol_header *) entry->data : NULL;
+   return entry ? (struct symbol *) entry->data : NULL;
 }
 
 
@@ -201,200 +146,140 @@ find_symbol(struct _mesa_symbol_table *table, const char *name)
  */
 int
 _mesa_symbol_table_symbol_scope(struct _mesa_symbol_table *table,
-				int name_space, const char *name)
+                                const char *name)
 {
-    struct symbol_header *const hdr = find_symbol(table, name);
-    struct symbol *sym;
+   struct symbol *const sym = find_symbol(table, name);
 
-    if (hdr != NULL) {
-       for (sym = hdr->symbols; sym != NULL; sym = sym->next_with_same_name) {
-	  assert(sym->hdr == hdr);
+   if (sym) {
+      assert(sym->depth <= table->depth);
+      return sym->depth - table->depth;
+   }
 
-	  if ((name_space == -1) || (sym->name_space == name_space)) {
-	     assert(sym->depth <= table->depth);
-	     return sym->depth - table->depth;
-	  }
-       }
-    }
-
-    return -1;
+   return -1;
 }
 
 
 void *
 _mesa_symbol_table_find_symbol(struct _mesa_symbol_table *table,
-                               int name_space, const char *name)
+                               const char *name)
 {
-    struct symbol_header *const hdr = find_symbol(table, name);
+   struct symbol *const sym = find_symbol(table, name);
+   if (sym)
+      return sym->data;
 
-    if (hdr != NULL) {
-        struct symbol *sym;
-
-
-        for (sym = hdr->symbols; sym != NULL; sym = sym->next_with_same_name) {
-            assert(sym->hdr == hdr);
-
-            if ((name_space == -1) || (sym->name_space == name_space)) {
-                return sym->data;
-            }
-        }
-    }
-
-    return NULL;
+   return NULL;
 }
 
 
 int
 _mesa_symbol_table_add_symbol(struct _mesa_symbol_table *table,
-                              int name_space, const char *name,
-                              void *declaration)
+                              const char *name, void *declaration)
 {
-    struct symbol_header *hdr;
-    struct symbol *sym;
+   struct symbol *new_sym;
+   struct symbol *sym = find_symbol(table, name);
 
-    check_symbol_table(table);
+   if (sym && sym->depth == table->depth)
+      return -1;
 
-    hdr = find_symbol(table, name);
+   new_sym = calloc(1, sizeof(*sym));
+   if (new_sym == NULL) {
+      _mesa_error_no_memory(__func__);
+      return -1;
+   }
 
-    check_symbol_table(table);
+   if (sym) {
+      /* Store link to symbol in outer scope with the same name */
+      new_sym->next_with_same_name = sym;
+      new_sym->name = sym->name;
+   } else {
+      new_sym->name = strdup(name);
+      if (new_sym->name == NULL) {
+         free(new_sym);
+         _mesa_error_no_memory(__func__);
+         return -1;
+      }
+   }
 
-    if (hdr == NULL) {
-       hdr = calloc(1, sizeof(*hdr));
-       if (hdr == NULL) {
-          _mesa_error_no_memory(__func__);
-          return -1;
-       }
+   new_sym->next_with_same_scope = table->current_scope->symbols;
+   new_sym->data = declaration;
+   new_sym->depth = table->depth;
 
-       hdr->name = strdup(name);
-       if (hdr->name == NULL) {
-          free(hdr);
-          _mesa_error_no_memory(__func__);
-          return -1;
-       }
+   table->current_scope->symbols = new_sym;
 
-       _mesa_hash_table_insert(table->ht, hdr->name, hdr);
-       hdr->next = table->hdr;
-       table->hdr = hdr;
-    }
+   _mesa_hash_table_insert(table->ht, new_sym->name, new_sym);
 
-    check_symbol_table(table);
+   return 0;
+}
 
-    /* If the symbol already exists in this namespace at this scope, it cannot
-     * be added to the table.
-     */
-    for (sym = hdr->symbols
-	 ; (sym != NULL) && (sym->name_space != name_space)
-	 ; sym = sym->next_with_same_name) {
-       /* empty */
-    }
+int
+_mesa_symbol_table_replace_symbol(struct _mesa_symbol_table *table,
+                                  const char *name,
+                                  void *declaration)
+{
+    struct symbol *sym = find_symbol(table, name);
 
-    if (sym && (sym->depth == table->depth))
+    /* If the symbol doesn't exist, it cannot be replaced. */
+    if (sym == NULL)
        return -1;
 
-    sym = calloc(1, sizeof(*sym));
-    if (sym == NULL) {
-       _mesa_error_no_memory(__func__);
-       return -1;
-    }
-
-    sym->next_with_same_name = hdr->symbols;
-    sym->next_with_same_scope = table->current_scope->symbols;
-    sym->hdr = hdr;
-    sym->name_space = name_space;
     sym->data = declaration;
-    sym->depth = table->depth;
-
-    assert(sym->hdr == hdr);
-
-    hdr->symbols = sym;
-    table->current_scope->symbols = sym;
-
-    check_symbol_table(table);
     return 0;
 }
 
-
 int
 _mesa_symbol_table_add_global_symbol(struct _mesa_symbol_table *table,
-				     int name_space, const char *name,
-				     void *declaration)
+                                     const char *name, void *declaration)
 {
-    struct symbol_header *hdr;
-    struct symbol *sym;
-    struct symbol *curr;
-    struct scope_level *top_scope;
+   struct scope_level *top_scope;
+   struct symbol *inner_sym = NULL;
+   struct symbol *sym = find_symbol(table, name);
 
-    check_symbol_table(table);
+   while (sym) {
+      if (sym->depth == 0)
+         return -1;
 
-    hdr = find_symbol(table, name);
+      inner_sym = sym;
 
-    check_symbol_table(table);
+      /* Get symbol from the outer scope with the same name */
+      sym = sym->next_with_same_name;
+   }
 
-    if (hdr == NULL) {
-        hdr = calloc(1, sizeof(*hdr));
-        if (hdr == NULL) {
-           _mesa_error_no_memory(__func__);
-           return -1;
-        }
+   /* Find the top-level scope */
+   for (top_scope = table->current_scope; top_scope->next != NULL;
+        top_scope = top_scope->next) {
+      /* empty */
+   }
 
-        hdr->name = strdup(name);
+   sym = calloc(1, sizeof(*sym));
+   if (sym == NULL) {
+      _mesa_error_no_memory(__func__);
+      return -1;
+   }
 
-        _mesa_hash_table_insert(table->ht, hdr->name, hdr);
-        hdr->next = table->hdr;
-        table->hdr = hdr;
-    }
+   if (inner_sym) {
+      /* In case we add the global out of order store a link to the global
+       * symbol in global.
+       */
+      inner_sym->next_with_same_name = sym;
 
-    check_symbol_table(table);
+      sym->name = inner_sym->name;
+   } else {
+      sym->name = strdup(name);
+      if (sym->name == NULL) {
+         free(sym);
+         _mesa_error_no_memory(__func__);
+         return -1;
+      }
+   }
 
-    /* If the symbol already exists in this namespace at this scope, it cannot
-     * be added to the table.
-     */
-    for (sym = hdr->symbols
-	 ; (sym != NULL) && (sym->name_space != name_space)
-	 ; sym = sym->next_with_same_name) {
-       /* empty */
-    }
+   sym->next_with_same_scope = top_scope->symbols;
+   sym->data = declaration;
 
-    if (sym && sym->depth == 0)
-       return -1;
+   top_scope->symbols = sym;
 
-    /* Find the top-level scope */
-    for (top_scope = table->current_scope
-	 ; top_scope->next != NULL
-	 ; top_scope = top_scope->next) {
-       /* empty */
-    }
+   _mesa_hash_table_insert(table->ht, sym->name, sym);
 
-    sym = calloc(1, sizeof(*sym));
-    if (sym == NULL) {
-       _mesa_error_no_memory(__func__);
-       return -1;
-    }
-
-    sym->next_with_same_scope = top_scope->symbols;
-    sym->hdr = hdr;
-    sym->name_space = name_space;
-    sym->data = declaration;
-
-    assert(sym->hdr == hdr);
-
-    /* Since next_with_same_name is ordered by scope, we need to append the
-     * new symbol to the _end_ of the list.
-     */
-    if (hdr->symbols == NULL) {
-       hdr->symbols = sym;
-    } else {
-       for (curr = hdr->symbols
-	    ; curr->next_with_same_name != NULL
-	    ; curr = curr->next_with_same_name) {
-	  /* empty */
-       }
-       curr->next_with_same_name = sym;
-    }
-    top_scope->symbols = sym;
-
-    check_symbol_table(table);
-    return 0;
+   return 0;
 }
 
 
@@ -418,19 +303,10 @@ _mesa_symbol_table_ctor(void)
 void
 _mesa_symbol_table_dtor(struct _mesa_symbol_table *table)
 {
-   struct symbol_header *hdr;
-   struct symbol_header *next;
-
    while (table->current_scope != NULL) {
       _mesa_symbol_table_pop_scope(table);
    }
 
-   for (hdr = table->hdr; hdr != NULL; hdr = next) {
-       next = hdr->next;
-       free(hdr->name);
-       free(hdr);
-   }
-
    _mesa_hash_table_destroy(table->ht, NULL);
    free(table);
 }

src/mesa/program/symbol_table.h

@@ -30,17 +30,22 @@ extern void _mesa_symbol_table_push_scope(struct _mesa_symbol_table *table);
 extern void _mesa_symbol_table_pop_scope(struct _mesa_symbol_table *table);
 
 extern int _mesa_symbol_table_add_symbol(struct _mesa_symbol_table *symtab,
-    int name_space, const char *name, void *declaration);
+                                         const char *name, void *declaration);
 
-extern int _mesa_symbol_table_add_global_symbol(
-    struct _mesa_symbol_table *symtab, int name_space, const char *name,
-    void *declaration);
+extern int _mesa_symbol_table_replace_symbol(struct _mesa_symbol_table *table,
+                                             const char *name,
+                                             void *declaration);
+
+extern int
+_mesa_symbol_table_add_global_symbol(struct _mesa_symbol_table *symtab,
+                                     const char *name,
+                                     void *declaration);
 
 extern int _mesa_symbol_table_symbol_scope(struct _mesa_symbol_table *table,
-    int name_space, const char *name);
+                                           const char *name);
 
-extern void *_mesa_symbol_table_find_symbol(
-    struct _mesa_symbol_table *symtab, int name_space, const char *name);
+extern void *_mesa_symbol_table_find_symbol(struct _mesa_symbol_table *symtab,
+                                            const char *name);
 
 extern struct _mesa_symbol_table *_mesa_symbol_table_ctor(void);
 

src/mesa/state_tracker/st_cb_syncobj.c

@@ -41,16 +41,21 @@ struct st_sync_object {
    struct gl_sync_object b;
 
    struct pipe_fence_handle *fence;
+   mtx_t mutex; /**< protects "fence" */
 };
 
 
 static struct gl_sync_object * st_new_sync_object(struct gl_context *ctx,
                                                   GLenum type)
 {
-   if (type == GL_SYNC_FENCE)
-      return (struct gl_sync_object*)CALLOC_STRUCT(st_sync_object);
-   else
+   if (type == GL_SYNC_FENCE) {
+      struct st_sync_object *so = CALLOC_STRUCT(st_sync_object);
+
+      mtx_init(&so->mutex, mtx_plain);
+      return &so->b;
+   } else {
       return NULL;
+   }
 }
 
 static void st_delete_sync_object(struct gl_context *ctx,
@@ -60,6 +65,7 @@ static void st_delete_sync_object(struct gl_context *ctx,
    struct st_sync_object *so = (struct st_sync_object*)obj;
 
    screen->fence_reference(screen, &so->fence, NULL);
+   mtx_destroy(&so->mutex);
    free(so->b.Label);
    free(so);
 }
@@ -76,24 +82,6 @@ static void st_fence_sync(struct gl_context *ctx, struct gl_sync_object *obj,
    pipe->flush(pipe, &so->fence, PIPE_FLUSH_DEFERRED);
 }
 
-static void st_check_sync(struct gl_context *ctx, struct gl_sync_object *obj)
-{
-   struct pipe_context *pipe = st_context(ctx)->pipe;
-   struct pipe_screen *screen = pipe->screen;
-   struct st_sync_object *so = (struct st_sync_object*)obj;
-
-   /* If the fence doesn't exist, assume it's signalled. */
-   if (!so->fence) {
-      so->b.StatusFlag = GL_TRUE;
-      return;
-   }
-
-   if (screen->fence_finish(screen, pipe, so->fence, 0)) {
-      screen->fence_reference(screen, &so->fence, NULL);
-      so->b.StatusFlag = GL_TRUE;
-   }
-}
-
 static void st_client_wait_sync(struct gl_context *ctx,
                                 struct gl_sync_object *obj,
                                 GLbitfield flags, GLuint64 timeout)
@@ -101,13 +89,22 @@ static void st_client_wait_sync(struct gl_context *ctx,
    struct pipe_context *pipe = st_context(ctx)->pipe;
    struct pipe_screen *screen = pipe->screen;
    struct st_sync_object *so = (struct st_sync_object*)obj;
+   struct pipe_fence_handle *fence = NULL;
 
    /* If the fence doesn't exist, assume it's signalled. */
+   mtx_lock(&so->mutex);
    if (!so->fence) {
+      mtx_unlock(&so->mutex);
       so->b.StatusFlag = GL_TRUE;
       return;
    }
 
+   /* We need a local copy of the fence pointer, so that we can call
+    * fence_finish unlocked.
+    */
+   screen->fence_reference(screen, &fence, so->fence);
+   mtx_unlock(&so->mutex);
+
    /* Section 4.1.2 of OpenGL 4.5 (Compatibility Profile) says:
     *    [...] if ClientWaitSync is called and all of the following are true:
     *    - the SYNC_FLUSH_COMMANDS_BIT bit is set in flags,
@@ -120,11 +117,18 @@ static void st_client_wait_sync(struct gl_context *ctx,
     * Assume GL_SYNC_FLUSH_COMMANDS_BIT is always set, because applications
     * forget to set it.
     */
-   if (so->fence &&
-       screen->fence_finish(screen, pipe, so->fence, timeout)) {
+   if (screen->fence_finish(screen, pipe, fence, timeout)) {
+      mtx_lock(&so->mutex);
       screen->fence_reference(screen, &so->fence, NULL);
+      mtx_unlock(&so->mutex);
       so->b.StatusFlag = GL_TRUE;
    }
+   screen->fence_reference(screen, &fence, NULL);
+}
+
+static void st_check_sync(struct gl_context *ctx, struct gl_sync_object *obj)
+{
+   st_client_wait_sync(ctx, obj, 0, 0);
 }
 
 static void st_server_wait_sync(struct gl_context *ctx,

src/mesa/state_tracker/st_draw.c

@@ -126,6 +126,30 @@ setup_index_buffer(struct st_context *st,
 }
 
 
+/**
+ * Set the restart index.
+ */
+static void
+setup_primitive_restart(struct gl_context *ctx,
+                        const struct _mesa_index_buffer *ib,
+                        struct pipe_draw_info *info)
+{
+   if (ctx->Array._PrimitiveRestart) {
+      info->restart_index = _mesa_primitive_restart_index(ctx, ib->type);
+
+      /* Enable primitive restart only when the restart index can have an
+       * effect. This is required for correctness in radeonsi VI support.
+       * Other hardware may also benefit from taking a faster, non-restart path
+       * when possible.
+       */
+      if ((ib->type == GL_UNSIGNED_INT) ||
+          (ib->type == GL_UNSIGNED_SHORT && info->restart_index <= 0xffff) ||
+          (ib->type == GL_UNSIGNED_BYTE && info->restart_index <= 0xff))
+         info->primitive_restart = true;
+   }
+}
+
+
 /**
  * Translate OpenGL primtive type (GL_POINTS, GL_TRIANGLE_STRIP, etc) to
  * the corresponding Gallium type.
@@ -205,19 +229,7 @@ st_draw_vbo(struct gl_context *ctx,
       /* The VBO module handles restart for the non-indexed GLDrawArrays
        * so we only set these fields for indexed drawing:
        */
-      if (ctx->Array._PrimitiveRestart) {
-         info.restart_index = _mesa_primitive_restart_index(ctx, ib->type);
-
-         /* Enable primitive restart only when the restart index can have an
-          * effect. This is required for correctness in radeonsi VI support,
-          * though other hardware may also benefit from taking a faster,
-          * non-restart path when possible.
-          */
-         if ((ibuffer.index_size >= 4) ||
-             (ibuffer.index_size >= 2 && info.restart_index <= 0xffff) ||
-             (info.restart_index <= 0xff))
-            info.primitive_restart = true;
-      }
+      setup_primitive_restart(ctx, ib, &info);
    }
    else {
       /* Transform feedback drawing is always non-indexed. */
@@ -310,6 +322,9 @@ st_indirect_draw_vbo(struct gl_context *ctx,
       }
 
       info.indexed = TRUE;
+
+      /* Primitive restart is not handled by the VBO module in this case. */
+      setup_primitive_restart(ctx, ib, &info);
    }
 
    info.mode = translate_prim(ctx, mode);
@@ -317,10 +332,6 @@ st_indirect_draw_vbo(struct gl_context *ctx,
    info.indirect = st_buffer_object(indirect_data)->buffer;
    info.indirect_offset = indirect_offset;
 
-   /* Primitive restart is not handled by the VBO module in this case. */
-   info.primitive_restart = ctx->Array._PrimitiveRestart;
-   info.restart_index = ctx->Array.RestartIndex;
-
    if (ST_DEBUG & DEBUG_DRAW) {
       debug_printf("st/draw indirect: mode %s drawcount %d indexed %d\n",
                    u_prim_name(info.mode),

src/util/bitscan.h

@@ -31,6 +31,7 @@
 
 #include <assert.h>
 #include <stdint.h>
+#include <string.h>
 
 #if defined(_MSC_VER)
 #include <intrin.h>

src/vulkan/wsi/wsi_common_wayland.c

@@ -702,6 +702,9 @@ wsi_wl_surface_create_swapchain(VkIcdSurfaceBase *icd_surface,
    if (chain == NULL)
       return VK_ERROR_OUT_OF_HOST_MEMORY;
 
+   bool alpha = pCreateInfo->compositeAlpha ==
+                      VK_COMPOSITE_ALPHA_PRE_MULTIPLIED_BIT_KHR;
+
    chain->base.device = device;
    chain->base.destroy = wsi_wl_swapchain_destroy;
    chain->base.get_images = wsi_wl_swapchain_get_images;
@@ -711,7 +714,7 @@ wsi_wl_surface_create_swapchain(VkIcdSurfaceBase *icd_surface,
    chain->surface = surface->surface;
    chain->extent = pCreateInfo->imageExtent;
    chain->vk_format = pCreateInfo->imageFormat;
-   chain->drm_format = wl_drm_format_for_vk_format(chain->vk_format, false);
+   chain->drm_format = wl_drm_format_for_vk_format(chain->vk_format, alpha);
 
    chain->present_mode = pCreateInfo->presentMode;
    chain->fifo_ready = true;

src/vulkan/wsi/wsi_common_x11.c

@@ -471,6 +471,7 @@ struct x11_swapchain {
    xcb_connection_t *                           conn;
    xcb_window_t                                 window;
    xcb_gc_t                                     gc;
+   uint32_t                                     depth;
    VkExtent2D                                   extent;
    uint32_t                                     image_count;
 
@@ -625,7 +626,6 @@ x11_image_init(VkDevice device_h, struct x11_swapchain *chain,
    uint32_t row_pitch;
    uint32_t offset;
    uint32_t bpp = 32;
-   uint32_t depth = 24;
    int fd;
    uint32_t size;
 
@@ -651,7 +651,7 @@ x11_image_init(VkDevice device_h, struct x11_swapchain *chain,
                                           pCreateInfo->imageExtent.width,
                                           pCreateInfo->imageExtent.height,
                                           row_pitch,
-                                          depth, bpp, fd);
+                                          chain->depth, bpp, fd);
    xcb_discard_reply(chain->conn, cookie.sequence);
 
    int fence_fd = xshmfence_alloc_shm();
@@ -752,18 +752,29 @@ x11_surface_create_swapchain(VkIcdSurfaceBase *icd_surface,
    if (chain == NULL)
       return VK_ERROR_OUT_OF_HOST_MEMORY;
 
+   xcb_connection_t *conn = x11_surface_get_connection(icd_surface);
+   xcb_window_t window = x11_surface_get_window(icd_surface);
+   xcb_get_geometry_reply_t *geometry =
+      xcb_get_geometry_reply(conn, xcb_get_geometry(conn, window), NULL);
+
+   if (geometry == NULL)
+      return VK_ERROR_SURFACE_LOST_KHR;
+
    chain->base.device = device;
    chain->base.destroy = x11_swapchain_destroy;
    chain->base.get_images = x11_get_images;
    chain->base.acquire_next_image = x11_acquire_next_image;
    chain->base.queue_present = x11_queue_present;
    chain->base.image_fns = image_fns;
-   chain->conn = x11_surface_get_connection(icd_surface);
-   chain->window = x11_surface_get_window(icd_surface);
+   chain->conn = conn;
+   chain->window = window;
+   chain->depth = geometry->depth;
    chain->extent = pCreateInfo->imageExtent;
    chain->image_count = num_images;
    chain->send_sbc = 0;
 
+   free(geometry);
+
    chain->event_id = xcb_generate_id(chain->conn);
    xcb_present_select_input(chain->conn, chain->event_id, chain->window,
                             XCB_PRESENT_EVENT_MASK_CONFIGURE_NOTIFY |