mesa/texformat: use format conversion function in _mesa_choose_tex_format

This function's cases for non-generic compressed formats duplicate
the GL to MESA translation in _mesa_glenum_to_compressed_format().
This patch replaces the switch cases with a call to the translation
function. This change teaches this function about ASTC, thus enabling
ASTC for glTex*Storage*() calls.

Reviewed-by: Chad Versace <chad.versace@intel.com>
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>

Android.common.mk

@@ -42,7 +42,6 @@ LOCAL_CFLAGS += \
 	-DANDROID_VERSION=0x0$(MESA_ANDROID_MAJOR_VERSION)0$(MESA_ANDROID_MINOR_VERSION)
 
 LOCAL_CFLAGS += \
-	-D__STDC_LIMIT_MACROS \
 	-DHAVE___BUILTIN_EXPECT \
 	-DHAVE___BUILTIN_FFS \
 	-DHAVE___BUILTIN_FFSLL \
@@ -71,7 +70,7 @@ endif
 
 ifeq ($(MESA_ENABLE_LLVM),true)
 LOCAL_CFLAGS += \
-	-DHAVE_LLVM=0x0305 -DMESA_LLVM_VERSION_PATCH=2 \
+	-DHAVE_LLVM=0x0305 -DLLVM_VERSION_PATCH=2 \
 	-D__STDC_CONSTANT_MACROS \
 	-D__STDC_FORMAT_MACROS \
 	-D__STDC_LIMIT_MACROS

VERSION

@@ -1 +1 @@
-11.0.5
+11.1.0-devel

bin/.cherry-ignore

@@ -1,2 +0,0 @@
-# The commit base differs greatly between 11.0 and master
-2832ca95ecce064c7d841a3a374c2179f56161be glsl: fix stream qualifier for blocks with an instance name

configure.ac

@@ -9,6 +9,7 @@ dnl Copyright © 2009-2014 Jon TURNEY
 dnl Copyright © 2011-2012 Benjamin Franzke
 dnl Copyright © 2008-2014 David Airlie
 dnl Copyright © 2009-2013 Brian Paul
+dnl Copyright © 2003-2007 Keith Packard, Daniel Stone
 dnl
 dnl Permission is hereby granted, free of charge, to any person obtaining a
 dnl copy of this software and associated documentation files (the "Software"),
@@ -106,8 +107,6 @@ AC_SYS_LARGEFILE
 LT_PREREQ([2.2])
 LT_INIT([disable-static])
 
-AC_CHECK_PROG(RM, rm, [rm -f])
-
 AX_PROG_BISON([],
               AS_IF([test ! -f "$srcdir/src/glsl/glcpp/glcpp-parse.c"],
                     [AC_MSG_ERROR([bison not found - unable to compile glcpp-parse.y])]))
@@ -989,6 +988,144 @@ fi
 
 AC_SUBST([MESA_LLVM])
 
+# SHA1 hashing
+AC_ARG_WITH([sha1],
+        [AS_HELP_STRING([--with-sha1=libc|libmd|libnettle|libgcrypt|libcrypto|libsha1|CommonCrypto|CryptoAPI],
+        [choose SHA1 implementation])])
+case "x$with_sha1" in
+x | xlibc | xlibmd | xlibnettle | xlibgcrypt | xlibcrypto | xlibsha1 | xCommonCrypto | xCryptoAPI)
+  ;;
+*)
+        AC_MSG_ERROR([Illegal value for --with-sha1: $with_sha1])
+esac
+
+AC_CHECK_FUNC([SHA1Init], [HAVE_SHA1_IN_LIBC=yes])
+if test "x$with_sha1" = x && test "x$HAVE_SHA1_IN_LIBC" = xyes; then
+	with_sha1=libc
+fi
+if test "x$with_sha1" = xlibc && test "x$HAVE_SHA1_IN_LIBC" != xyes; then
+	AC_MSG_ERROR([sha1 in libc requested but not found])
+fi
+if test "x$with_sha1" = xlibc; then
+	AC_DEFINE([HAVE_SHA1_IN_LIBC], [1],
+		[Use libc SHA1 functions])
+	SHA1_LIBS=""
+fi
+AC_CHECK_FUNC([CC_SHA1_Init], [HAVE_SHA1_IN_COMMONCRYPTO=yes])
+if test "x$with_sha1" = x && test "x$HAVE_SHA1_IN_COMMONCRYPTO" = xyes; then
+	with_sha1=CommonCrypto
+fi
+if test "x$with_sha1" = xCommonCrypto && test "x$HAVE_SHA1_IN_COMMONCRYPTO" != xyes; then
+	AC_MSG_ERROR([CommonCrypto requested but not found])
+fi
+if test "x$with_sha1" = xCommonCrypto; then
+	AC_DEFINE([HAVE_SHA1_IN_COMMONCRYPTO], [1],
+		[Use CommonCrypto SHA1 functions])
+	SHA1_LIBS=""
+fi
+dnl stdcall functions cannot be tested with AC_CHECK_LIB
+AC_CHECK_HEADER([wincrypt.h], [HAVE_SHA1_IN_CRYPTOAPI=yes], [], [#include <windows.h>])
+if test "x$with_sha1" = x && test "x$HAVE_SHA1_IN_CRYPTOAPI" = xyes; then
+	with_sha1=CryptoAPI
+fi
+if test "x$with_sha1" = xCryptoAPI && test "x$HAVE_SHA1_IN_CRYPTOAPI" != xyes; then
+	AC_MSG_ERROR([CryptoAPI requested but not found])
+fi
+if test "x$with_sha1" = xCryptoAPI; then
+	AC_DEFINE([HAVE_SHA1_IN_CRYPTOAPI], [1],
+		[Use CryptoAPI SHA1 functions])
+	SHA1_LIBS=""
+fi
+AC_CHECK_LIB([md], [SHA1Init], [HAVE_LIBMD=yes])
+if test "x$with_sha1" = x && test "x$HAVE_LIBMD" = xyes; then
+	with_sha1=libmd
+fi
+if test "x$with_sha1" = xlibmd && test "x$HAVE_LIBMD" != xyes; then
+	AC_MSG_ERROR([libmd requested but not found])
+fi
+if test "x$with_sha1" = xlibmd; then
+	AC_DEFINE([HAVE_SHA1_IN_LIBMD], [1],
+	          [Use libmd SHA1 functions])
+	SHA1_LIBS=-lmd
+fi
+PKG_CHECK_MODULES([LIBSHA1], [libsha1], [HAVE_LIBSHA1=yes], [HAVE_LIBSHA1=no])
+if test "x$with_sha1" = x && test "x$HAVE_LIBSHA1" = xyes; then
+   with_sha1=libsha1
+fi
+if test "x$with_sha1" = xlibsha1 && test "x$HAVE_LIBSHA1" != xyes; then
+	AC_MSG_ERROR([libsha1 requested but not found])
+fi
+if test "x$with_sha1" = xlibsha1; then
+	AC_DEFINE([HAVE_SHA1_IN_LIBSHA1], [1],
+	          [Use libsha1 for SHA1])
+	SHA1_LIBS=-lsha1
+fi
+AC_CHECK_LIB([nettle], [nettle_sha1_init], [HAVE_LIBNETTLE=yes])
+if test "x$with_sha1" = x && test "x$HAVE_LIBNETTLE" = xyes; then
+	with_sha1=libnettle
+fi
+if test "x$with_sha1" = xlibnettle && test "x$HAVE_LIBNETTLE" != xyes; then
+	AC_MSG_ERROR([libnettle requested but not found])
+fi
+if test "x$with_sha1" = xlibnettle; then
+	AC_DEFINE([HAVE_SHA1_IN_LIBNETTLE], [1],
+	          [Use libnettle SHA1 functions])
+	SHA1_LIBS=-lnettle
+fi
+AC_CHECK_LIB([gcrypt], [gcry_md_open], [HAVE_LIBGCRYPT=yes])
+if test "x$with_sha1" = x && test "x$HAVE_LIBGCRYPT" = xyes; then
+	with_sha1=libgcrypt
+fi
+if test "x$with_sha1" = xlibgcrypt && test "x$HAVE_LIBGCRYPT" != xyes; then
+	AC_MSG_ERROR([libgcrypt requested but not found])
+fi
+if test "x$with_sha1" = xlibgcrypt; then
+	AC_DEFINE([HAVE_SHA1_IN_LIBGCRYPT], [1],
+	          [Use libgcrypt SHA1 functions])
+	SHA1_LIBS=-lgcrypt
+fi
+# We don't need all of the OpenSSL libraries, just libcrypto
+AC_CHECK_LIB([crypto], [SHA1_Init], [HAVE_LIBCRYPTO=yes])
+PKG_CHECK_MODULES([OPENSSL], [openssl], [HAVE_OPENSSL_PKC=yes],
+                  [HAVE_OPENSSL_PKC=no])
+if test "x$HAVE_LIBCRYPTO" = xyes || test "x$HAVE_OPENSSL_PKC" = xyes; then
+	if test "x$with_sha1" = x; then
+		with_sha1=libcrypto
+	fi
+else
+	if test "x$with_sha1" = xlibcrypto; then
+		AC_MSG_ERROR([OpenSSL libcrypto requested but not found])
+	fi
+fi
+if test "x$with_sha1" = xlibcrypto; then
+	if test "x$HAVE_LIBCRYPTO" = xyes; then
+		SHA1_LIBS=-lcrypto
+	else
+		SHA1_LIBS="$OPENSSL_LIBS"
+		SHA1_CFLAGS="$OPENSSL_CFLAGS"
+	fi
+fi
+AC_MSG_CHECKING([for SHA1 implementation])
+AC_MSG_RESULT([$with_sha1])
+AC_SUBST(SHA1_LIBS)
+AC_SUBST(SHA1_CFLAGS)
+
+# Allow user to configure out the shader-cache feature
+AC_ARG_ENABLE([shader-cache],
+    AS_HELP_STRING([--disable-shader-cache], [Disable binary shader cache]),
+    [enable_shader_cache="$enableval"],
+    [if test "x$with_sha1" != "x"; then
+        enable_shader_cache=yes
+     else
+        enable_shader_cache=no
+     fi])
+if test "x$with_sha1" = "x"; then
+    if test "x$enable_shader_cache" = "xyes"; then
+        AC_MSG_ERROR([Cannot enable shader cache (no SHA-1 implementation found)])
+    fi
+fi
+AM_CONDITIONAL([ENABLE_SHADER_CACHE], [test x$enable_shader_cache = xyes])
+
 case "$host_os" in
 linux*)
     need_pci_id=yes ;;
@@ -1152,16 +1289,6 @@ AC_SUBST(GLX_TLS, ${GLX_USE_TLS})
 AS_IF([test "x$GLX_USE_TLS" = xyes -a "x$ax_pthread_ok" = xyes],
       [DEFINES="${DEFINES} -DGLX_USE_TLS"])
 
-dnl Read-only text section on x86 hardened platforms
-AC_ARG_ENABLE([glx-read-only-text],
-    [AS_HELP_STRING([--enable-glx-read-only-text],
-        [Disable writable .text section on x86 (decreases performance) @<:@default=disabled@:>@])],
-    [enable_glx_read_only_text="$enableval"],
-    [enable_glx_read_only_text=no])
-if test "x$enable_glx_read_only_text" = xyes; then
-    DEFINES="$DEFINES -DGLX_X86_READONLY_TEXT"
-fi
-
 dnl
 dnl More DRI setup
 dnl
@@ -2190,6 +2317,7 @@ AC_CONFIG_FILES([Makefile
 		src/gallium/auxiliary/Makefile
 		src/gallium/auxiliary/pipe-loader/Makefile
 		src/gallium/drivers/freedreno/Makefile
+		src/gallium/drivers/ddebug/Makefile
 		src/gallium/drivers/i915/Makefile
 		src/gallium/drivers/ilo/Makefile
 		src/gallium/drivers/llvmpipe/Makefile
@@ -2357,6 +2485,12 @@ else
     echo "        Gallium:         no"
 fi
 
+dnl Shader cache
+echo ""
+echo "        Shader cache:    $enable_shader_cache"
+if test "x$enable_shader_cache" = "xyes"; then
+    echo "        With SHA1 from:  $with_sha1"
+fi
 
 dnl Libraries
 echo ""

docs/GL3.txt

@@ -196,7 +196,7 @@ GL 4.5, GLSL 4.50:
   GL_ARB_get_texture_sub_image                         DONE (all drivers)
   GL_ARB_shader_texture_image_samples                  not started
   GL_ARB_texture_barrier                               DONE (nv50, nvc0, r600, radeonsi)
-  GL_KHR_context_flush_control                         DONE (all - but needs GLX/EXT extension to be useful)
+  GL_KHR_context_flush_control                         DONE (all - but needs GLX/EGL extension to be useful)
   GL_KHR_robust_buffer_access_behavior                 not started
   GL_KHR_robustness                                    90% done (the ARB variant)
   GL_EXT_shader_integer_mix                            DONE (all drivers that support GLSL)

docs/relnotes/11.0.0.html

@@ -14,7 +14,7 @@
 <iframe src="../contents.html"></iframe>
 <div class="content">
 
-<h1>Mesa 11.0.0 Release Notes / September 12, 2015</h1>
+<h1>Mesa 11.0.0 Release Notes / TBD</h1>
 
 <p>
 Mesa 11.0.0 is a new development release.
@@ -33,8 +33,7 @@ because compatibility contexts are not supported.
 
 <h2>SHA256 checksums</h2>
 <pre>
-7d7e4ddffa3b162506efa01e2cc41e329caa4995336b92e5cc21f2e1fb36c1b3  mesa-11.0.0.tar.gz
-e095a3eb2eca9dfde7efca8946527c8ae20a0cc938a8c78debc7f158ad44af32  mesa-11.0.0.tar.xz
+TBD.
 </pre>
 
 
@@ -84,175 +83,13 @@ Note: some of the new features are only available with certain drivers.
 <li>EGL 1.5 on r600, radeonsi, nv50, nvc0</li>
 </ul>
 
-
 <h2>Bug fixes</h2>
 
-<p>This list is likely incomplete.</p>
-
-<ul>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=51658">Bug 51658</a> - r200 (&amp; possibly radeon) DRI fixes for gnome shell on Mesa 8.0.3</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=65525">Bug 65525</a> - [llvmpipe] lp_scene.h:210:lp_scene_alloc: Assertion `size &lt;= (64 * 1024)' failed.</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=66346">Bug 66346</a> - shader_query.cpp:49: error: invalid conversion from 'void*' to 'GLuint'</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=73512">Bug 73512</a> - [clover] mesa.icd. should contain full path</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=73528">Bug 73528</a> - Deferred lighting in Second Life causes system hiccups and screen flickering</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=74329">Bug 74329</a> - Please expose OES_texture_float and OES_texture_half_float on the ES3 context</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=80500">Bug 80500</a> - Flickering shadows in unreleased title trace</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=82186">Bug 82186</a> - [r600g] BARTS GPU lockup with minecraft shaders</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84225">Bug 84225</a> - Allow constant-index-expression sampler array indexing with GLSL-ES &lt; 300</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=84677">Bug 84677</a> - Triangle disappears with glPolygonMode GL_LINE</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=85252">Bug 85252</a> - Segfault in compiler while processing ternary operator with void arguments</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=89131">Bug 89131</a> - [Bisected] Graphical corruption in Weston,  shows old framebuffer pieces</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90000">Bug 90000</a> - [i965 Bisected NIR] Piglit/gglean_fragprog1-z-write_test fail</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90073">Bug 90073</a> - Leaks in xcb_dri3_open_reply_fds() and get_render_node_from_id_path_tag</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90249">Bug 90249</a> - Fails to build egl_dri2 on osx</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90310">Bug 90310</a> - Fails to build gallium_dri.so at linking stage with clang because of multiple redefinitions</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90347">Bug 90347</a> - [NVE0+] Failure to insert texbar under some circumstances (causing bad colors in Terasology)</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90466">Bug 90466</a> - arm: linker error ndefined reference to `nir_metadata_preserve'</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90520">Bug 90520</a> - Register spilling clobbers registers used elsewhere in the shader</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90537">Bug 90537</a> - radeonsi bo/va conflict on RADEON_GEM_VA (rscreen-&gt;ws-&gt;buffer_from_handle returns NULL)</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90547">Bug 90547</a> - [BDW/BSW/SKL Bisected]Piglit/glean&#64;vertprog1-rsq_test_2_(reciprocal_square_root_of_negative_value) fais</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90580">Bug 90580</a> - [HSW bisected] integer multiplication bug</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90600">Bug 90600</a> - IOError: [Errno 2] No such file or directory: 'gl_API.xml'</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90621">Bug 90621</a> - Mesa fail to build from git</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90629">Bug 90629</a> - [i965] SIMD16 dual_source_blend assertion `src[i].file != GRF || src[i].width == dst.width' failed</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90691">Bug 90691</a> - [BSW]Piglit/spec/nv_conditional_render/dlist fails intermittently</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90728">Bug 90728</a> - dvd playback with vlc and vdpau causes segmentation fault</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90734">Bug 90734</a> - glBufferSubData is corrupting data when buffer is &gt; 32k</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90748">Bug 90748</a> - [BDW Bisected]dEQP-GLES3.functional.fbo.completeness.renderable.texture.depth.rg_half_float_oes fails</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90749">Bug 90749</a> - [BDW Bisected]dEQP-GLES3.functional.rasterization.fbo.rbo_multisample_max.primitives.lines_wide fails</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90751">Bug 90751</a> - [BDW Bisected]dEQP-GLES3.functional.fbo.completeness.renderable.texture.stencil.stencil_index8 fails</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90797">Bug 90797</a> - [ALL bisected] Mesa change cause performance case manhattan fail.</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90817">Bug 90817</a> - swrast fails to load with certain remote X servers</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90830">Bug 90830</a> - [bsw bisected regression] GPU hang for spec.arb_gpu_shader5.execution.sampler_array_indexing.vs-nonzero-base</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90839">Bug 90839</a> - [10.5.5/10.6 regression, bisected] PBO glDrawPixels no longer using blit fastpath</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90873">Bug 90873</a> - Kernel hang, TearFree On, Mate desktop environment</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90887">Bug 90887</a> - PhiMovesPass in register allocator broken</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90895">Bug 90895</a> - [IVB/HSW/BDW/BSW Bisected] GLB2.7 Egypt, GfxBench3.0 T-Rex &amp; ALU and many SynMark cases performance reduced by 10-23%</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90902">Bug 90902</a> - [bsw][regression] dEQP: &quot;Found invalid pixel values&quot;</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90903">Bug 90903</a> - egl_dri2.c:dri2_load fails to load libglapi on osx</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90904">Bug 90904</a> - OSX: EXC_BAD_ACCESS when using translate_sse + gallium + softpipe/llvmpipe</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90905">Bug 90905</a> - mesa: Finish subdir-objects transition</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=90925">Bug 90925</a> - &quot;high fidelity&quot;: Segfault in _mesa_program_resource_find_name</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91022">Bug 91022</a> - [g45 g965 bisected] assertions generated from textureGrad cube samplers fix</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91047">Bug 91047</a> - [SNB Bisected] Messed up Fog in Super Smash Bros. Melee in Dolphin</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91056">Bug 91056</a> - The Bard's Tale (2005, native)  has rendering issues</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91077">Bug 91077</a> - dri2_glx.c:1186: undefined reference to `loader_open_device'</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91099">Bug 91099</a> - [llvmpipe] piglit glsl-max-varyings &gt;max_varying_components regression</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91101">Bug 91101</a> - [softpipe] piglit glsl-1.50&#64;execution&#64;geometry&#64;max-input-components regression</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91117">Bug 91117</a> - Nimbus (running in wine) has rendering issues, objects are semi-transparent</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91124">Bug 91124</a> - Civilization V (in Wine) has rendering issues: text missing, menu bar corrupted</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91173">Bug 91173</a> - Oddworld: Stranger's Wrath HD: disfigured models in wrong colors</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91193">Bug 91193</a> - [290x] Dota2 reborn ingame rendering breaks with git-af4b9c7</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91222">Bug 91222</a> - lp_test_format regression on CentOS 7</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91226">Bug 91226</a> - Crash in glLinkProgram (NEW)</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91231">Bug 91231</a> - [NV92] Psychonauts (native) segfaults on start when DRI3 enabled</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91254">Bug 91254</a> - (regresion) video using VA-API on Intel slow and freeze system with mesa 10.6 or 10.6.1</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91290">Bug 91290</a> - SIGSEGV glcpp/glcpp-parse.y:1077</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91292">Bug 91292</a> - [BDW+] glVertexAttribDivisor not working in combination with glPolygonMode</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91337">Bug 91337</a> - OSMesaGetProcAdress(&quot;OSMesaPixelStore&quot;) returns nil</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91418">Bug 91418</a> - Visual Studio 2015 vsnprintf build error</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91425">Bug 91425</a> - [regression, bisected] Piglit spec/ext_packed_float/ getteximage-invalid-format-for-packed-type fails</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91441">Bug 91441</a> - make check DispatchSanity_test.GL30 regression</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91444">Bug 91444</a> - regression bisected radeonsi: don't change pipe_resource in resource_copy_region</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91461">Bug 91461</a> - gl_TessLevel* writes have no effect for all but the last TCS invocation</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91513">Bug 91513</a> - [IVB/HSW/BDW/SKL Bisected] Lightsmark performance reduced by 7%-10%</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91526">Bug 91526</a> - World of Warcraft (on Wine) has UI corruption with nouveau</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91544">Bug 91544</a> - [i965, regression, bisected] regression of several tests in 93977d3a151675946c03e</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91551">Bug 91551</a> - DXTn compressed normal maps produce severe artifacts on all NV5x and NVDx chipsets</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91570">Bug 91570</a> - Upgrading mesa to 10.6 causes segfault in OpenGL applications with GeForce4 MX 440 / AGP 8X</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91591">Bug 91591</a> - rounding.h:102:2: error: #error &quot;Unsupported or undefined LONG_BIT&quot;</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91610">Bug 91610</a> - [BSW] GPU hang for spec.shaders.point-vertex-id gl_instanceid divisor</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91673">Bug 91673</a> - Segfault when calling glTexSubImage2D on storage texture to bound FBO</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91726">Bug 91726</a> - R600 asserts in tgsi_cmp/make_src_for_op3</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91847">Bug 91847</a> - glGenerateTextureMipmap not working (no errors) unless glActiveTexture(GL_TEXTURE1) is called before</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91857">Bug 91857</a> - Mesa 10.6.3 linker is slow</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91881">Bug 91881</a> - regression: GPU lockups since mesa-11.0.0_rc1 on RV620 (r600) driver</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91890">Bug 91890</a> - [nve7] witcher2: blurry image &amp; DATA_ERRORs (class 0xa097 mthd 0x2380/0x238c)</li>
-
-</ul>
-
+TBD.
 
 <h2>Changes</h2>
 
-<li>Removed the EGL loader from the Linux SCons build.</li>
+TBD.
 
 </div>
 </body>

docs/relnotes/11.0.1.html

@@ -1,134 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
-<html lang="en">
-<head>
-  <meta http-equiv="content-type" content="text/html; charset=utf-8">
-  <title>Mesa Release Notes</title>
-  <link rel="stylesheet" type="text/css" href="../mesa.css">
-</head>
-<body>
-
-<div class="header">
-  <h1>The Mesa 3D Graphics Library</h1>
-</div>
-
-<iframe src="../contents.html"></iframe>
-<div class="content">
-
-<h1>Mesa 11.0.1 Release Notes / September 26, 2015</h1>
-
-<p>
-Mesa 11.0.1 is a bug fix release which fixes bugs found since the 11.0.0 release.
-</p>
-<p>
-Mesa 11.0.1 implements the OpenGL 4.1 API, but the version reported by
-glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
-glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
-Some drivers don't support all the features required in OpenGL 4.1.  OpenGL
-4.1 is <strong>only</strong> available if requested at context creation
-because compatibility contexts are not supported.
-</p>
-
-
-<h2>SHA256 checksums</h2>
-<pre>
-6dab262877e12c0546a0e2970c6835a0f217e6d4026ccecb3cd5dd733d1ce867  mesa-11.0.1.tar.gz
-43d0dfcd1f1e36f07f8228cd76d90175d3fc74c1ed25d7071794a100a98ef2a6  mesa-11.0.1.tar.xz
-</pre>
-
-
-<h2>New features</h2>
-<p>None</p>
-
-<h2>Bug fixes</h2>
-
-<p>This list is likely incomplete.</p>
-
-<ul>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38109">Bug 38109</a> - i915 driver crashes if too few vertices are submitted (Mesa 7.10.2)</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91114">Bug 91114</a> - ES3-CTS.gtf.GL3Tests.shadow.shadow_execution_vert fails</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91716">Bug 91716</a> - [bisected] piglit.shaders.glsl-vs-int-attrib regresses on 32 bit BYT, HSW, IVB, SNB</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91719">Bug 91719</a> - [SNB,HSW,BYT] dEQP regressions associated with using NIR for vertex shaders</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92009">Bug 92009</a> - ES3-CTS.gtf.GL3Tests.packed_pixels.packed_pixels fails</li>
-
-</ul>
-
-
-<h2>Changes</h2>
-
-<p>Antia Puentes (2):</p>
-<ul>
-  <li>i965/vec4: Fix saturation errors when coalescing registers</li>
-  <li>i965/vec4_nir: Load constants as integers</li>
-</ul>
-
-<p>Anuj Phogat (1):</p>
-<ul>
-  <li>meta: Abort meta pbo path if TexSubImage need signed unsigned conversion</li>
-</ul>
-
-<p>Emil Velikov (2):</p>
-<ul>
-  <li>docs: add sha256 checksums for 11.0.0</li>
-  <li>Update version to 11.0.1</li>
-</ul>
-
-<p>Iago Toral Quiroga (1):</p>
-<ul>
-  <li>mesa: Fix GL_FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE for default framebuffer.</li>
-</ul>
-
-<p>Ian Romanick (5):</p>
-<ul>
-  <li>t_dd_dmatmp: Make "count" actually be the count</li>
-  <li>t_dd_dmatmp: Clean up improper code formatting from previous patch</li>
-  <li>t_dd_dmatmp: Use '&amp; 3' instead of '% 4' everywhere</li>
-  <li>t_dd_dmatmp: Pull out common 'count -= count &amp; 3' code</li>
-  <li>t_dd_dmatmp: Use addition instead of subtraction in loop bounds</li>
-</ul>
-
-<p>Ilia Mirkin (6):</p>
-<ul>
-  <li>st/mesa: avoid integer overflows with buffers &gt;= 512MB</li>
-  <li>nv50, nvc0: fix max texture buffer size to 128M elements</li>
-  <li>freedreno/a3xx: fix blending of L8 format</li>
-  <li>nv50,nvc0: detect underlying resource changes and update tic</li>
-  <li>nv50,nvc0: flush texture cache in presence of coherent bufs</li>
-  <li>radeonsi: load fmask ptr relative to the resources array</li>
-</ul>
-
-<p>Jason Ekstrand (2):</p>
-<ul>
-  <li>nir: Fix a bunch of ralloc parenting errors</li>
-  <li>i965/vec4: Don't reswizzle hardware registers</li>
-</ul>
-
-<p>Jeremy Huddleston (1):</p>
-<ul>
-  <li>configure.ac: Add support to enable read-only text segment on x86.</li>
-</ul>
-
-<p>Ray Strode (1):</p>
-<ul>
-  <li>gbm: convert gbm bo format to fourcc format on dma-buf import</li>
-</ul>
-
-<p>Tapani Pälli (2):</p>
-<ul>
-  <li>mesa: fix errors when reading depth with glReadPixels</li>
-  <li>i965: fix textureGrad for cubemaps</li>
-</ul>
-
-<p>Ulrich Weigand (1):</p>
-<ul>
-  <li>mesa: Fix texture compression on big-endian systems</li>
-</ul>
-
-
-</div>
-</body>
-</html>

docs/relnotes/11.0.2.html

@@ -1,85 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
-<html lang="en">
-<head>
-  <meta http-equiv="content-type" content="text/html; charset=utf-8">
-  <title>Mesa Release Notes</title>
-  <link rel="stylesheet" type="text/css" href="../mesa.css">
-</head>
-<body>
-
-<div class="header">
-  <h1>The Mesa 3D Graphics Library</h1>
-</div>
-
-<iframe src="../contents.html"></iframe>
-<div class="content">
-
-<h1>Mesa 11.0.2 Release Notes / September 28, 2015</h1>
-
-<p>
-Mesa 11.0.2 is a bug fix release which fixes bugs found since the 11.0.1 release.
-</p>
-<p>
-Mesa 11.0.2 implements the OpenGL 4.1 API, but the version reported by
-glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
-glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
-Some drivers don't support all the features required in OpenGL 4.1.  OpenGL
-4.1 is <strong>only</strong> available if requested at context creation
-because compatibility contexts are not supported.
-</p>
-
-
-<h2>SHA256 checksums</h2>
-<pre>
-45170773500d6ae2f9eb93fc85efee69f7c97084411ada4eddf92f78bca56d20  mesa-11.0.2.tar.gz
-fce11fb27eb87adf1e620a76455d635c6136dfa49ae58c53b34ef8d0c7b7eae4  mesa-11.0.2.tar.xz
-</pre>
-
-
-<h2>New features</h2>
-<p>None</p>
-
-<h2>Bug fixes</h2>
-
-<p>This list is likely incomplete.</p>
-
-<ul>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91582">Bug 91582</a> - [bisected] Regression in DEQP gles2.functional.negative_api.texture.texsubimage2d_neg_offset</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91970">Bug 91970</a> - [BSW regression] dEQP-GLES3.functional.shaders.precision.int.highp_mul_vertex</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92095">Bug 92095</a> - [Regression, bisected] arb_shader_atomic_counters.compiler.builtins.frag</li>
-
-</ul>
-
-
-<h2>Changes</h2>
-
-<p>Eduardo Lima Mitev (3):</p>
-<ul>
-  <li>mesa: Fix order of format+type and internal format checks for glTexImageXD ops</li>
-  <li>mesa: Move _mesa_base_tex_format() from teximage to glformats files</li>
-  <li>mesa: Use the effective internal format instead for validation</li>
-</ul>
-
-<p>Emil Velikov (2):</p>
-<ul>
-  <li>docs: add sha256 checksums for 11.0.1</li>
-  <li>Update version to 11.0.2</li>
-</ul>
-
-<p>Kristian Høgsberg Kristensen (1):</p>
-<ul>
-  <li>i965: Respect stride and subreg_offset for ATTR registers</li>
-</ul>
-
-<p>Matt Turner (1):</p>
-<ul>
-  <li>glsl: Expose gl_MaxTess{Control,Evaluation}AtomicCounters.</li>
-</ul>
-
-
-</div>
-</body>
-</html>

docs/relnotes/11.0.3.html

@@ -1,185 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
-<html lang="en">
-<head>
-  <meta http-equiv="content-type" content="text/html; charset=utf-8">
-  <title>Mesa Release Notes</title>
-  <link rel="stylesheet" type="text/css" href="../mesa.css">
-</head>
-<body>
-
-<div class="header">
-  <h1>The Mesa 3D Graphics Library</h1>
-</div>
-
-<iframe src="../contents.html"></iframe>
-<div class="content">
-
-<h1>Mesa 11.0.3 Release Notes / October 10, 2015</h1>
-
-<p>
-Mesa 11.0.3 is a bug fix release which fixes bugs found since the 11.0.2 release.
-</p>
-<p>
-Mesa 11.0.3 implements the OpenGL 4.1 API, but the version reported by
-glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
-glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
-Some drivers don't support all the features required in OpenGL 4.1.  OpenGL
-4.1 is <strong>only</strong> available if requested at context creation
-because compatibility contexts are not supported.
-</p>
-
-
-<h2>SHA256 checksums</h2>
-<pre>
-c2210e3daecc10ed9fdcea500327652ed6effc2f47c4b9cee63fb08f560d7117  mesa-11.0.3.tar.gz
-ab2992eece21adc23c398720ef8c6933cb69ea42e1b2611dc09d031e17e033d6  mesa-11.0.3.tar.xz
-</pre>
-
-
-<h2>New features</h2>
-<p>None</p>
-
-<h2>Bug fixes</h2>
-
-<p>This list is likely incomplete.</p>
-
-<ul>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=55552">Bug 55552</a> - Compile errors with --enable-mangling</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=71789">Bug 71789</a> - [r300g] Visuals not found in (default) depth = 24</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91044">Bug 91044</a> - piglit spec/egl_khr_create_context/valid debug flag gles* fail</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=91596">Bug 91596</a> - EGL_KHR_gl_colorspace (v2) causes problem with Android-x86 GUI</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91718">Bug 91718</a> - piglit.spec.arb_shader_image_load_store.invalid causes intermittent GPU HANG</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92072">Bug 92072</a> - Wine breakage since d082c5324 (st/mesa: don't call st_validate_state in BlitFramebuffer)</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92265">Bug 92265</a> - Black windows in weston after update mesa to 11.0.2-1</li>
-
-</ul>
-
-
-<h2>Changes</h2>
-
-<p>Brian Paul (1):</p>
-<ul>
-  <li>st/mesa: try PIPE_BIND_RENDER_TARGET when choosing float texture formats</li>
-</ul>
-
-<p>Daniel Scharrer (1):</p>
-<ul>
-  <li>mesa: Add abs input modifier to base for POW in ffvertex_prog</li>
-</ul>
-
-<p>Emil Velikov (3):</p>
-<ul>
-  <li>docs: add sha256 checksums for 11.0.2</li>
-  <li>Revert "nouveau: make sure there's always room to emit a fence"</li>
-  <li>Update version to 11.0.3</li>
-</ul>
-
-<p>Francisco Jerez (1):</p>
-<ul>
-  <li>i965/fs: Fix hang on IVB and VLV with image format mismatch.</li>
-</ul>
-
-<p>Ian Romanick (1):</p>
-<ul>
-  <li>meta: Handle array textures in scaled MSAA blits</li>
-</ul>
-
-<p>Ilia Mirkin (6):</p>
-<ul>
-  <li>nouveau: be more careful about freeing temporary transfer buffers</li>
-  <li>nouveau: delay deleting buffer with unflushed fence</li>
-  <li>nouveau: wait to unref the transfer's bo until it's no longer used</li>
-  <li>nv30: pretend to have packed texture/surface formats</li>
-  <li>nv30: always go through translate module on big-endian</li>
-  <li>nouveau: make sure there's always room to emit a fence</li>
-</ul>
-
-<p>Jason Ekstrand (1):</p>
-<ul>
-  <li>mesa: Correctly handle GL_BGRA_EXT in ES3 format_and_type checks</li>
-</ul>
-
-<p>Kyle Brenneman (3):</p>
-<ul>
-  <li>glx: Fix build errors with --enable-mangling (v2)</li>
-  <li>mapi: Make _glapi_get_stub work with "gl" or "mgl" prefix.</li>
-  <li>glx: Don't hard-code the name "libGL.so.1" in driOpenDriver (v3)</li>
-</ul>
-
-<p>Leo Liu (1):</p>
-<ul>
-  <li>radeon/vce: fix vui time_scale zero error</li>
-</ul>
-
-<p>Marek Olšák (21):</p>
-<ul>
-  <li>st/mesa: fix front buffer regression after dropping st_validate_state in Blit</li>
-  <li>radeonsi: handle index buffer alloc failures</li>
-  <li>radeonsi: handle constant buffer alloc failures</li>
-  <li>gallium/radeon: handle buffer_map staging buffer failures better</li>
-  <li>gallium/radeon: handle buffer alloc failures in r600_draw_rectangle</li>
-  <li>gallium/radeon: add a fail path for depth MSAA texture readback</li>
-  <li>radeonsi: report alloc failure from si_shader_binary_read</li>
-  <li>radeonsi: add malloc fail paths to si_create_shader_state</li>
-  <li>radeonsi: skip drawing if the tess factor ring allocation fails</li>
-  <li>radeonsi: skip drawing if GS ring allocations fail</li>
-  <li>radeonsi: handle shader precompile failures</li>
-  <li>radeonsi: handle fixed-func TCS shader create failure</li>
-  <li>radeonsi: skip drawing if VS, TCS, TES, GS fail to compile or upload</li>
-  <li>radeonsi: skip drawing if PS fails to compile or upload</li>
-  <li>radeonsi: skip drawing if updating the scratch buffer fails</li>
-  <li>radeonsi: don't forget to update scratch relocations for LS, HS, ES shaders</li>
-  <li>radeonsi: handle dummy constant buffer allocation failure</li>
-  <li>gallium/u_blitter: handle allocation failures</li>
-  <li>radeonsi: add scratch buffer to the buffer list when it's re-allocated</li>
-  <li>st/dri: don't use _ctx in client_wait_sync</li>
-  <li>egl/dri2: don't require a context for ClientWaitSync (v2)</li>
-</ul>
-
-<p>Matthew Waters (1):</p>
-<ul>
-  <li>egl: rework handling EGL_CONTEXT_FLAGS</li>
-</ul>
-
-<p>Michel Dänzer (1):</p>
-<ul>
-  <li>st/dri: Use packed RGB formats</li>
-</ul>
-
-<p>Roland Scheidegger (1):</p>
-<ul>
-  <li>mesa: fix mipmap generation for immutable, compressed textures</li>
-</ul>
-
-<p>Tom Stellard (3):</p>
-<ul>
-  <li>gallium/radeon: Use call_once() when initailizing LLVM targets</li>
-  <li>gallivm: Allow drivers and state trackers to initialize gallivm LLVM targets v2</li>
-  <li>radeon/llvm: Initialize gallivm targets when initializing the AMDGPU target v2</li>
-</ul>
-
-<p>Varad Gautam (1):</p>
-<ul>
-  <li>egl: restore surface type before linking config to its display</li>
-</ul>
-
-<p>Ville Syrjälä (3):</p>
-<ul>
-  <li>i830: Fix collision between I830_UPLOAD_RASTER_RULES and I830_UPLOAD_TEX(0)</li>
-  <li>i915: Fix texcoord vs. varying collision in fragment programs</li>
-  <li>i915: Remember to call intel_prepare_render() before blitting</li>
-</ul>
-
-
-</div>
-</body>
-</html>

docs/relnotes/11.0.4.html

@@ -1,168 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
-<html lang="en">
-<head>
-  <meta http-equiv="content-type" content="text/html; charset=utf-8">
-  <title>Mesa Release Notes</title>
-  <link rel="stylesheet" type="text/css" href="../mesa.css">
-</head>
-<body>
-
-<div class="header">
-  <h1>The Mesa 3D Graphics Library</h1>
-</div>
-
-<iframe src="../contents.html"></iframe>
-<div class="content">
-
-<h1>Mesa 11.0.4 Release Notes / October 24, 2015</h1>
-
-<p>
-Mesa 11.0.4 is a bug fix release which fixes bugs found since the 11.0.3 release.
-</p>
-<p>
-Mesa 11.0.4 implements the OpenGL 4.1 API, but the version reported by
-glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
-glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
-Some drivers don't support all the features required in OpenGL 4.1.  OpenGL
-4.1 is <strong>only</strong> available if requested at context creation
-because compatibility contexts are not supported.
-</p>
-
-
-<h2>SHA256 checksums</h2>
-<pre>
-ed412ca6a46d1bd055120e5c12806c15419ae8c4dd6d3f6ea20a83091d5c78bf  mesa-11.0.4.tar.gz
-40201bf7fc6fa12a6d9edfe870b41eb4dd6669154e3c42c48a96f70805f5483d  mesa-11.0.4.tar.xz
-</pre>
-
-
-<h2>New features</h2>
-<p>None</p>
-
-<h2>Bug fixes</h2>
-
-<p>This list is likely incomplete.</p>
-
-<ul>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86281">Bug 86281</a> - brw_meta_fast_clear (brw=brw&#64;entry=0x7fffd4097a08, fb=fb&#64;entry=0x7fffd40fa900, buffers=buffers&#64;entry=2, partial_clear=partial_clear&#64;entry=false)</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=86720">Bug 86720</a> - [radeon] Europa Universalis 4 freezing during game start (10.3.3+, still broken on 11.0.2)</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91788">Bug 91788</a> - [HSW Regression] Synmark2_v6 Multithread performance case FPS reduced by 36%</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92304">Bug 92304</a> - [cts] cts.shaders.negative conformance tests fail</li>
-
-</ul>
-
-
-<h2>Changes</h2>
-
-<p>Alejandro Piñeiro (2):</p>
-<ul>
-  <li>i965/vec4: check writemask when bailing out at register coalesce</li>
-  <li>i965/vec4: fill src_reg type using the constructor type parameter</li>
-</ul>
-
-<p>Brian Paul (2):</p>
-<ul>
-  <li>vbo: fix incorrect switch statement in init_mat_currval()</li>
-  <li>mesa: fix incorrect opcode in save_BlendFunci()</li>
-</ul>
-
-<p>Chih-Wei Huang (3):</p>
-<ul>
-  <li>mesa: android: Fix the incorrect path of sse_minmax.c</li>
-  <li>nv50/ir: use C++11 standard std::unordered_map if possible</li>
-  <li>nv30: include the header of ffs prototype</li>
-</ul>
-
-<p>Chris Wilson (1):</p>
-<ul>
-  <li>i965: Remove early release of DRI2 miptree</li>
-</ul>
-
-<p>Dave Airlie (1):</p>
-<ul>
-  <li>mesa/uniforms: fix get_uniform for doubles (v2)</li>
-</ul>
-
-<p>Emil Velikov (1):</p>
-<ul>
-  <li>docs: add sha256 checksums for 11.0.3</li>
-</ul>
-
-<p>Francisco Jerez (5):</p>
-<ul>
-  <li>i965: Don't tell the hardware about our UAV access.</li>
-  <li>mesa: Expose function to calculate whether a shader image unit is valid.</li>
-  <li>mesa: Skip redundant texture completeness checking during image validation.</li>
-  <li>i965: Use _mesa_is_image_unit_valid() instead of gl_image_unit::_Valid.</li>
-  <li>mesa: Get rid of texture-dependent image unit derived state.</li>
-</ul>
-
-<p>Ian Romanick (8):</p>
-<ul>
-  <li>glsl: Allow built-in functions as constant expressions in OpenGL ES 1.00</li>
-  <li>ff_fragment_shader: Use binding to set the sampler unit</li>
-  <li>glsl/linker: Use constant_initializer instead of constant_value to initialize uniforms</li>
-  <li>glsl: Use constant_initializer instead of constant_value to determine whether to keep an unused uniform</li>
-  <li>glsl: Only set ir_variable::constant_value for const-decorated variables</li>
-  <li>glsl: Restrict initializers for global variables to constant expression in ES</li>
-  <li>glsl: Add method to determine whether an expression contains the sequence operator</li>
-  <li>glsl: In later GLSL versions, sequence operator is cannot be a constant expression</li>
-</ul>
-
-<p>Ilia Mirkin (1):</p>
-<ul>
-  <li>nouveau: make sure there's always room to emit a fence</li>
-</ul>
-
-<p>Indrajit Das (1):</p>
-<ul>
-  <li>st/va: Used correct parameter to derive the value of the "h" variable in vlVaCreateImage</li>
-</ul>
-
-<p>Jonathan Gray (1):</p>
-<ul>
-  <li>configure.ac: ensure RM is set</li>
-</ul>
-
-<p>Krzysztof Sobiecki (1):</p>
-<ul>
-  <li>st/fbo: use pipe_surface_release instead of pipe_surface_reference</li>
-</ul>
-
-<p>Leo Liu (1):</p>
-<ul>
-  <li>st/omx/dec/h264: fix field picture type 0 poc disorder</li>
-</ul>
-
-<p>Marek Olšák (3):</p>
-<ul>
-  <li>st/mesa: fix clip state dependencies</li>
-  <li>radeonsi: fix a GS copy shader leak</li>
-  <li>gallium: add PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT</li>
-</ul>
-
-<p>Nicolai Hähnle (1):</p>
-<ul>
-  <li>u_vbuf: fix vb slot assignment for translated buffers</li>
-</ul>
-
-<p>Rob Clark (1):</p>
-<ul>
-  <li>freedreno/a3xx: cache-flush is needed after MEM_WRITE</li>
-</ul>
-
-<p>Tapani Pälli (3):</p>
-<ul>
-  <li>mesa: add GL_UNSIGNED_INT_24_8 to _mesa_pack_depth_span</li>
-  <li>mesa: Set api prefix to version string when overriding version</li>
-  <li>mesa: fix ARRAY_SIZE query for GetProgramResourceiv</li>
-</ul>
-
-
-</div>
-</body>
-</html>

docs/relnotes/11.0.5.html

@@ -1,173 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
-<html lang="en">
-<head>
-  <meta http-equiv="content-type" content="text/html; charset=utf-8">
-  <title>Mesa Release Notes</title>
-  <link rel="stylesheet" type="text/css" href="../mesa.css">
-</head>
-<body>
-
-<div class="header">
-  <h1>The Mesa 3D Graphics Library</h1>
-</div>
-
-<iframe src="../contents.html"></iframe>
-<div class="content">
-
-<h1>Mesa 11.0.5 Release Notes / November 11, 2015</h1>
-
-<p>
-Mesa 11.0.5 is a bug fix release which fixes bugs found since the 11.0.4 release.
-</p>
-<p>
-Mesa 11.0.5 implements the OpenGL 4.1 API, but the version reported by
-glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
-glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
-Some drivers don't support all the features required in OpenGL 4.1.  OpenGL
-4.1 is <strong>only</strong> available if requested at context creation
-because compatibility contexts are not supported.
-</p>
-
-
-<h2>SHA256 checksums</h2>
-<pre>
-TBD
-</pre>
-
-
-<h2>New features</h2>
-<p>None</p>
-
-<h2>Bug fixes</h2>
-
-<p>This list is likely incomplete.</p>
-
-<ul>
-
-<ul>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=91993">Bug 91993</a> - Graphical glitch in Astromenace (open-source game).</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92214">Bug 92214</a> - Flightgear crashes during splashboot with R600 driver, LLVM 3.7.0 and mesa 11.0.2</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92437">Bug 92437</a> - osmesa: Expose GL entry points for Windows build, via .def file</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92476">Bug 92476</a> - [cts] ES2-CTS.gtf.GL2ExtensionTests.egl_image.egl_image fails</li>
-
-<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=92623">Bug 92623</a> - Differences in prog_data ignored when caching fragment programs (causes hangs)</li>
-
-</ul>
-
-
-<h2>Changes</h2>
-
-<p>Alex Deucher (1):</p>
-<ul>
-  <li>radeon/uvd: don't expose HEVC on old UVD hw (v3)</li>
-</ul>
-
-<p>Ben Widawsky (1):</p>
-<ul>
-  <li>i965/skl: Add GT4 PCI IDs</li>
-</ul>
-
-<p>Emil Velikov (4):</p>
-<ul>
-  <li>docs: add sha256 checksums for 11.0.4</li>
-  <li>cherry-ignore: ignore a possible wrong nomination</li>
-  <li>Revert "mesa/glformats: Undo code changes from _mesa_base_tex_format() move"</li>
-  <li>Update version to 11.0.5</li>
-</ul>
-
-<p>Emmanuel Gil Peyrot (1):</p>
-<ul>
-  <li>gbm.h: Add a missing stddef.h include for size_t.</li>
-</ul>
-
-<p>Eric Anholt (1):</p>
-<ul>
-  <li>vc4: When the create ioctl fails, free our cache and try again.</li>
-</ul>
-
-<p>Ian Romanick (1):</p>
-<ul>
-  <li>i965: Fix is-renderable check in intel_image_target_renderbuffer_storage</li>
-</ul>
-
-<p>Ilia Mirkin (3):</p>
-<ul>
-  <li>nvc0: respect edgeflag attribute width</li>
-  <li>nouveau: set MaxDrawBuffers to the same value as MaxColorAttachments</li>
-  <li>nouveau: relax fence emit space assert</li>
-</ul>
-
-<p>Ivan Kalvachev (1):</p>
-<ul>
-  <li>r600g: Fix special negative immediate constants when using ABS modifier.</li>
-</ul>
-
-<p>Jason Ekstrand (2):</p>
-<ul>
-  <li>nir/lower_vec_to_movs: Pass the shader around directly</li>
-  <li>nir: Report progress from lower_vec_to_movs().</li>
-</ul>
-
-<p>Jose Fonseca (2):</p>
-<ul>
-  <li>gallivm: Translate all util_cpu_caps bits to LLVM attributes.</li>
-  <li>gallivm: Explicitly disable unsupported CPU features.</li>
-</ul>
-
-<p>Julien Isorce (4):</p>
-<ul>
-  <li>st/va: pass picture desc to begin and decode</li>
-  <li>nvc0: fix crash when nv50_miptree_from_handle fails</li>
-  <li>st/va: do not destroy old buffer when new one failed</li>
-  <li>st/va: add more errors checks in vlVaBufferSetNumElements and vlVaMapBuffer</li>
-</ul>
-
-<p>Kenneth Graunke (6):</p>
-<ul>
-  <li>i965: Fix missing BRW_NEW_*_PROG_DATA flagging caused by cache reuse.</li>
-  <li>nir: Report progress from nir_split_var_copies().</li>
-  <li>nir: Properly invalidate metadata in nir_split_var_copies().</li>
-  <li>nir: Properly invalidate metadata in nir_opt_copy_prop().</li>
-  <li>nir: Properly invalidate metadata in nir_lower_vec_to_movs().</li>
-  <li>nir: Properly invalidate metadata in nir_opt_remove_phis().</li>
-</ul>
-
-<p>Marek Olšák (1):</p>
-<ul>
-  <li>radeonsi: add register definitions for Stoney</li>
-</ul>
-
-<p>Nanley Chery (1):</p>
-<ul>
-  <li>mesa/glformats: Undo code changes from _mesa_base_tex_format() move</li>
-</ul>
-
-<p>Nicolai Hähnle (1):</p>
-<ul>
-  <li>st/mesa: fix mipmap generation for immutable textures with incomplete pyramids</li>
-</ul>
-
-<p>Nigel Stewart (1):</p>
-<ul>
-  <li>osmesa: Expose GL entry points for Windows build via DEF file.</li>
-</ul>
-
-<p>Roland Scheidegger (1):</p>
-<ul>
-  <li>gallivm: disable f16c when not using AVX</li>
-</ul>
-
-<p>Samuel Li (2):</p>
-<ul>
-  <li>radeonsi: add support for Stoney asics (v3)</li>
-  <li>radeonsi: add Stoney pci ids</li>
-</ul>
-
-
-</div>
-</body>
-</html>

docs/relnotes/11.1.0.html

@@ -0,0 +1,60 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
+<html lang="en">
+<head>
+  <meta http-equiv="content-type" content="text/html; charset=utf-8">
+  <title>Mesa Release Notes</title>
+  <link rel="stylesheet" type="text/css" href="../mesa.css">
+</head>
+<body>
+
+<div class="header">
+  <h1>The Mesa 3D Graphics Library</h1>
+</div>
+
+<iframe src="../contents.html"></iframe>
+<div class="content">
+
+<h1>Mesa 11.1.0 Release Notes / TBD</h1>
+
+<p>
+Mesa 11.1.0 is a new development release.
+People who are concerned with stability and reliability should stick
+with a previous release or wait for Mesa 11.1.1.
+</p>
+<p>
+Mesa 11.1.0 implements the OpenGL 4.1 API, but the version reported by
+glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
+glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
+Some drivers don't support all the features required in OpenGL 4.1.  OpenGL
+4.1 is <strong>only</strong> available if requested at context creation
+because compatibility contexts are not supported.
+</p>
+
+
+<h2>SHA256 checksums</h2>
+<pre>
+TBD.
+</pre>
+
+
+<h2>New features</h2>
+
+<p>
+Note: some of the new features are only available with certain drivers.
+</p>
+
+<ul>
+TBD.
+</ul>
+
+<h2>Bug fixes</h2>
+
+TBD.
+
+<h2>Changes</h2>
+
+TBD.
+
+</div>
+</body>
+</html>

include/pci_ids/i965_pci_ids.h

@@ -124,10 +124,6 @@ CHIPSET(0x1921, skl_gt2, "Intel(R) Skylake ULT GT2F")
 CHIPSET(0x1926, skl_gt3, "Intel(R) Skylake ULT GT3")
 CHIPSET(0x192A, skl_gt3, "Intel(R) Skylake SRV GT3")
 CHIPSET(0x192B, skl_gt3, "Intel(R) Skylake Halo GT3")
-CHIPSET(0x1932, skl_gt4, "Intel(R) Skylake GT4")
-CHIPSET(0x193A, skl_gt4, "Intel(R) Skylake GT4")
-CHIPSET(0x193B, skl_gt4, "Intel(R) Skylake GT4")
-CHIPSET(0x193D, skl_gt4, "Intel(R) Skylake GT4")
 CHIPSET(0x22B0, chv,     "Intel(R) HD Graphics (Cherryview)")
 CHIPSET(0x22B1, chv,     "Intel(R) HD Graphics (Cherryview)")
 CHIPSET(0x22B2, chv,     "Intel(R) HD Graphics (Cherryview)")

include/pci_ids/radeonsi_pci_ids.h

@@ -181,5 +181,3 @@ CHIPSET(0x9876, CARRIZO_, CARRIZO)
 CHIPSET(0x9877, CARRIZO_, CARRIZO)
 
 CHIPSET(0x7300, FIJI_, FIJI)
-
-CHIPSET(0x98E4, STONEY_, STONEY)

src/egl/drivers/dri2/egl_dri2.c

@@ -312,8 +312,6 @@ dri2_add_config(_EGLDisplay *disp, const __DRIconfig *dri_config, int id,
          else
             conf->dri_single_config = dri_config;
       }
-
-      conf->base.SurfaceType = 0;
       conf->base.ConfigID = config_id;
 
       _eglLinkConfig(&conf->base);
@@ -2386,18 +2384,13 @@ dri2_client_wait_sync(_EGLDriver *drv, _EGLDisplay *dpy, _EGLSync *sync,
    unsigned wait_flags = 0;
    EGLint ret = EGL_CONDITION_SATISFIED_KHR;
 
-   /* The EGL_KHR_fence_sync spec states:
-    *
-    *    "If no context is current for the bound API,
-    *     the EGL_SYNC_FLUSH_COMMANDS_BIT_KHR bit is ignored.
-    */
-   if (dri2_ctx && flags & EGL_SYNC_FLUSH_COMMANDS_BIT_KHR)
+   if (flags & EGL_SYNC_FLUSH_COMMANDS_BIT_KHR)
       wait_flags |= __DRI2_FENCE_FLAG_FLUSH_COMMANDS;
 
    /* the sync object should take a reference while waiting */
    dri2_egl_ref_sync(dri2_sync);
 
-   if (dri2_dpy->fence->client_wait_sync(dri2_ctx ? dri2_ctx->dri_context : NULL,
+   if (dri2_dpy->fence->client_wait_sync(dri2_ctx->dri_context,
                                          dri2_sync->fence, wait_flags,
                                          timeout))
       dri2_sync->base.SyncStatus = EGL_SIGNALED_KHR;

src/egl/main/eglcontext.c

@@ -152,51 +152,12 @@ _eglParseContextAttribList(_EGLContext *ctx, _EGLDisplay *dpy,
 
          /* The EGL_KHR_create_context spec says:
           *
-          *     "If the EGL_CONTEXT_OPENGL_DEBUG_BIT_KHR flag bit is set in
-          *     EGL_CONTEXT_FLAGS_KHR, then a <debug context> will be created.
-          *     [...]
-          *     In some cases a debug context may be identical to a non-debug
-          *     context. This bit is supported for OpenGL and OpenGL ES
-          *     contexts."
+          *     "Flags are only defined for OpenGL context creation, and
+          *     specifying a flags value other than zero for other types of
+          *     contexts, including OpenGL ES contexts, will generate an
+          *     error."
           */
-         if ((val & EGL_CONTEXT_OPENGL_DEBUG_BIT_KHR) &&
-             (api != EGL_OPENGL_API && api != EGL_OPENGL_ES_API)) {
-            err = EGL_BAD_ATTRIBUTE;
-            break;
-         }
-
-         /* The EGL_KHR_create_context spec says:
-          *
-          *     "If the EGL_CONTEXT_OPENGL_FORWARD_COMPATIBLE_BIT_KHR flag bit
-          *     is set in EGL_CONTEXT_FLAGS_KHR, then a <forward-compatible>
-          *     context will be created. Forward-compatible contexts are
-          *     defined only for OpenGL versions 3.0 and later. They must not
-          *     support functionality marked as <deprecated> by that version of
-          *     the API, while a non-forward-compatible context must support
-          *     all functionality in that version, deprecated or not. This bit
-          *     is supported for OpenGL contexts, and requesting a
-          *     forward-compatible context for OpenGL versions less than 3.0
-          *     will generate an error."
-          */
-         if ((val & EGL_CONTEXT_OPENGL_FORWARD_COMPATIBLE_BIT_KHR) &&
-             (api != EGL_OPENGL_API || ctx->ClientMajorVersion < 3)) {
-            err = EGL_BAD_ATTRIBUTE;
-            break;
-         }
-
-         /* The EGL_KHR_create_context_spec says:
-          *
-          *     "If the EGL_CONTEXT_OPENGL_ROBUST_ACCESS_BIT_KHR bit is set in
-          *     EGL_CONTEXT_FLAGS_KHR, then a context supporting <robust buffer
-          *     access> will be created. Robust buffer access is defined in the
-          *     GL_ARB_robustness extension specification, and the resulting
-          *     context must also support either the GL_ARB_robustness
-          *     extension, or a version of OpenGL incorporating equivalent
-          *     functionality. This bit is supported for OpenGL contexts.
-          */
-         if ((val & EGL_CONTEXT_OPENGL_ROBUST_ACCESS_BIT_KHR) &&
-             (api != EGL_OPENGL_API ||
-              !dpy->Extensions.EXT_create_context_robustness)) {
+         if (api != EGL_OPENGL_API && val != 0) {
             err = EGL_BAD_ATTRIBUTE;
             break;
          }

src/gallium/Makefile.am

@@ -11,6 +11,7 @@ SUBDIRS += auxiliary
 ##
 
 SUBDIRS += \
+	drivers/ddebug \
 	drivers/noop \
 	drivers/trace \
 	drivers/rbug

src/gallium/auxiliary/Makefile.am

@@ -38,23 +38,18 @@ libgallium_la_SOURCES += \
 
 endif
 
-MKDIR_GEN = $(AM_V_at)$(MKDIR_P) $(@D)
-PYTHON_GEN =  $(AM_V_GEN)$(PYTHON2) $(PYTHON_FLAGS)
+indices/u_indices_gen.c: $(srcdir)/indices/u_indices_gen.py
+	$(AM_V_at)$(MKDIR_P) indices
+	$(AM_V_GEN) $(PYTHON2) $< > $@
 
-indices/u_indices_gen.c: indices/u_indices_gen.py
-	$(MKDIR_GEN)
-	$(PYTHON_GEN) $(srcdir)/indices/u_indices_gen.py > $@
+indices/u_unfilled_gen.c: $(srcdir)/indices/u_unfilled_gen.py
+	$(AM_V_at)$(MKDIR_P) indices
+	$(AM_V_GEN) $(PYTHON2) $< > $@
 
-indices/u_unfilled_gen.c: indices/u_unfilled_gen.py
-	$(MKDIR_GEN)
-	$(PYTHON_GEN) $(srcdir)/indices/u_unfilled_gen.py > $@
+util/u_format_table.c: $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format_pack.py $(srcdir)/util/u_format_parse.py $(srcdir)/util/u_format.csv
+	$(AM_V_at)$(MKDIR_P) util
+	$(AM_V_GEN) $(PYTHON2) $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format.csv > $@
 
-util/u_format_table.c: util/u_format_table.py \
-                       util/u_format_pack.py \
-                       util/u_format_parse.py \
-                       util/u_format.csv
-	$(MKDIR_GEN)
-	$(PYTHON_GEN) $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format.csv > $@
 
 noinst_LTLIBRARIES += libgalliumvl_stub.la
 libgalliumvl_stub_la_SOURCES = \

src/gallium/auxiliary/gallivm/lp_bld_const.c

@@ -311,7 +311,7 @@ lp_build_const_elem(struct gallivm_state *gallivm,
    else {
       double dscale = lp_const_scale(type);
 
-      elem = LLVMConstInt(elem_type, (long long) round(val*dscale), 0);
+      elem = LLVMConstInt(elem_type, round(val*dscale), 0);
    }
 
    return elem;

src/gallium/auxiliary/gallivm/lp_bld_init.c

@@ -427,7 +427,6 @@ lp_build_init(void)
        */
       util_cpu_caps.has_avx = 0;
       util_cpu_caps.has_avx2 = 0;
-      util_cpu_caps.has_f16c = 0;
    }
 
 #ifdef PIPE_ARCH_PPC_64
@@ -459,9 +458,7 @@ lp_build_init(void)
    util_cpu_caps.has_sse3 = 0;
    util_cpu_caps.has_ssse3 = 0;
    util_cpu_caps.has_sse4_1 = 0;
-   util_cpu_caps.has_sse4_2 = 0;
    util_cpu_caps.has_avx = 0;
-   util_cpu_caps.has_avx2 = 0;
    util_cpu_caps.has_f16c = 0;
 #endif
 

src/gallium/auxiliary/gallivm/lp_bld_limits.h

@@ -137,8 +137,6 @@ gallivm_get_shader_param(enum pipe_shader_cap param)
    case PIPE_SHADER_CAP_TGSI_DFRACEXP_DLDEXP_SUPPORTED:
    case PIPE_SHADER_CAP_TGSI_FMA_SUPPORTED:
       return 0;
-   case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-      return 32;
    }
    /* if we get here, we missed a shader cap above (and should have seen
     * a compiler warning.)

src/gallium/auxiliary/gallivm/lp_bld_misc.cpp

@@ -81,8 +81,6 @@
 #  pragma pop_macro("DEBUG")
 #endif
 
-#include "c11/threads.h"
-#include "os/os_thread.h"
 #include "pipe/p_config.h"
 #include "util/u_debug.h"
 #include "util/u_cpu_detect.h"
@@ -105,33 +103,6 @@ static LLVMEnsureMultithreaded lLVMEnsureMultithreaded;
 
 }
 
-static once_flag init_native_targets_once_flag;
-
-static void init_native_targets()
-{
-   // If we have a native target, initialize it to ensure it is linked in and
-   // usable by the JIT.
-   llvm::InitializeNativeTarget();
-
-   llvm::InitializeNativeTargetAsmPrinter();
-
-   llvm::InitializeNativeTargetDisassembler();
-}
-
-/**
- * The llvm target registry is not thread-safe, so drivers and state-trackers
- * that want to initialize targets should use the gallivm_init_llvm_targets()
- * function to safely initialize targets.
- *
- * LLVM targets should be initialized before the driver or state-tracker tries
- * to access the registry.
- */
-extern "C" void
-gallivm_init_llvm_targets(void)
-{
-   call_once(&init_native_targets_once_flag, init_native_targets);
-}
-
 extern "C" void
 lp_set_target_options(void)
 {
@@ -144,7 +115,13 @@ lp_set_target_options(void)
    llvm::DisablePrettyStackTrace = true;
 #endif
 
-   gallivm_init_llvm_targets();
+   // If we have a native target, initialize it to ensure it is linked in and
+   // usable by the JIT.
+   llvm::InitializeNativeTarget();
+
+   llvm::InitializeNativeTargetAsmPrinter();
+
+   llvm::InitializeNativeTargetDisassembler();
 }
 
 
@@ -497,48 +474,20 @@ lp_build_create_jit_compiler_for_module(LLVMExecutionEngineRef *OutJIT,
 #endif
    }
 
-   llvm::SmallVector<std::string, 16> MAttrs;
-
-#if defined(PIPE_ARCH_X86) || defined(PIPE_ARCH_X86_64)
-   /*
-    * We need to unset attributes because sometimes LLVM mistakenly assumes
-    * certain features are present given the processor name.
-    *
-    * https://bugs.freedesktop.org/show_bug.cgi?id=92214
-    * http://llvm.org/PR25021
-    * http://llvm.org/PR19429
-    * http://llvm.org/PR16721
-    */
-   MAttrs.push_back(util_cpu_caps.has_sse    ? "+sse"    : "-sse"   );
-   MAttrs.push_back(util_cpu_caps.has_sse2   ? "+sse2"   : "-sse2"  );
-   MAttrs.push_back(util_cpu_caps.has_sse3   ? "+sse3"   : "-sse3"  );
-   MAttrs.push_back(util_cpu_caps.has_ssse3  ? "+ssse3"  : "-ssse3" );
-#if HAVE_LLVM >= 0x0304
-   MAttrs.push_back(util_cpu_caps.has_sse4_1 ? "+sse4.1" : "-sse4.1");
-#else
-   MAttrs.push_back(util_cpu_caps.has_sse4_1 ? "+sse41"  : "-sse41" );
-#endif
-#if HAVE_LLVM >= 0x0304
-   MAttrs.push_back(util_cpu_caps.has_sse4_2 ? "+sse4.2" : "-sse4.2");
-#else
-   MAttrs.push_back(util_cpu_caps.has_sse4_2 ? "+sse42"  : "-sse42" );
-#endif
-   /*
-    * AVX feature is not automatically detected from CPUID by the X86 target
-    * yet, because the old (yet default) JIT engine is not capable of
-    * emitting the opcodes. On newer llvm versions it is and at least some
-    * versions (tested with 3.3) will emit avx opcodes without this anyway.
-    */
-   MAttrs.push_back(util_cpu_caps.has_avx  ? "+avx"  : "-avx");
-   MAttrs.push_back(util_cpu_caps.has_f16c ? "+f16c" : "-f16c");
-   MAttrs.push_back(util_cpu_caps.has_avx2 ? "+avx2" : "-avx2");
-#endif
-
-#if defined(PIPE_ARCH_PPC)
-   MAttrs.push_back(util_cpu_caps.has_altivec ? "+altivec" : "-altivec");
-#endif
-
-   builder.setMAttrs(MAttrs);
+   llvm::SmallVector<std::string, 1> MAttrs;
+   if (util_cpu_caps.has_avx) {
+      /*
+       * AVX feature is not automatically detected from CPUID by the X86 target
+       * yet, because the old (yet default) JIT engine is not capable of
+       * emitting the opcodes. On newer llvm versions it is and at least some
+       * versions (tested with 3.3) will emit avx opcodes without this anyway.
+       */
+      MAttrs.push_back("+avx");
+      if (util_cpu_caps.has_f16c) {
+         MAttrs.push_back("+f16c");
+      }
+      builder.setMAttrs(MAttrs);
+   }
 
 #if HAVE_LLVM >= 0x0305
    StringRef MCPU = llvm::sys::getHostCPUName();

src/gallium/auxiliary/gallivm/lp_bld_misc.h

@@ -41,8 +41,6 @@ extern "C" {
 
 struct lp_generated_code;
 
-extern void
-gallivm_init_llvm_targets(void);
 
 extern void
 lp_set_target_options(void);

src/gallium/auxiliary/nir/tgsi_to_nir.c

@@ -24,6 +24,7 @@
 
 #include "util/ralloc.h"
 #include "glsl/nir/nir.h"
+#include "glsl/nir/nir_control_flow.h"
 #include "glsl/nir/nir_builder.h"
 #include "glsl/list.h"
 #include "glsl/shader_enums.h"
@@ -64,24 +65,24 @@ struct ttn_compile {
    nir_register *addr_reg;
 
    /**
-    * Stack of cf_node_lists where instructions should be pushed as we pop
+    * Stack of nir_cursors where instructions should be pushed as we pop
     * back out of the control flow stack.
     *
     * For each IF/ELSE/ENDIF block, if_stack[if_stack_pos] has where the else
     * instructions should be placed, and if_stack[if_stack_pos - 1] has where
     * the next instructions outside of the if/then/else block go.
     */
-   struct exec_list **if_stack;
+   nir_cursor *if_stack;
    unsigned if_stack_pos;
 
    /**
-    * Stack of cf_node_lists where instructions should be pushed as we pop
+    * Stack of nir_cursors where instructions should be pushed as we pop
     * back out of the control flow stack.
     *
     * loop_stack[loop_stack_pos - 1] contains the cf_node_list for the outside
     * of the loop.
     */
-   struct exec_list **loop_stack;
+   nir_cursor *loop_stack;
    unsigned loop_stack_pos;
 
    /* How many TGSI_FILE_IMMEDIATE vec4s have been parsed so far. */
@@ -307,7 +308,7 @@ ttn_emit_immediate(struct ttn_compile *c)
    for (i = 0; i < 4; i++)
       load_const->value.u[i] = tgsi_imm->u[i].Uint;
 
-   nir_instr_insert_after_cf_list(b->cf_node_list, &load_const->instr);
+   nir_builder_instr_insert(b, &load_const->instr);
 }
 
 static nir_src
@@ -363,7 +364,7 @@ ttn_src_for_file_and_index(struct ttn_compile *c, unsigned file, unsigned index,
          load->variables[0] = ttn_array_deref(c, load, var, offset, indirect);
 
          nir_ssa_dest_init(&load->instr, &load->dest, 4, NULL);
-         nir_instr_insert_after_cf_list(b->cf_node_list, &load->instr);
+         nir_builder_instr_insert(b, &load->instr);
 
          src = nir_src_for_ssa(&load->dest.ssa);
 
@@ -414,7 +415,7 @@ ttn_src_for_file_and_index(struct ttn_compile *c, unsigned file, unsigned index,
       load->num_components = ncomp;
 
       nir_ssa_dest_init(&load->instr, &load->dest, ncomp, NULL);
-      nir_instr_insert_after_cf_list(b->cf_node_list, &load->instr);
+      nir_builder_instr_insert(b, &load->instr);
 
       src = nir_src_for_ssa(&load->dest.ssa);
       break;
@@ -476,7 +477,7 @@ ttn_src_for_file_and_index(struct ttn_compile *c, unsigned file, unsigned index,
          srcn++;
       }
       nir_ssa_dest_init(&load->instr, &load->dest, 4, NULL);
-      nir_instr_insert_after_cf_list(b->cf_node_list, &load->instr);
+      nir_builder_instr_insert(b, &load->instr);
 
       src = nir_src_for_ssa(&load->dest.ssa);
       break;
@@ -552,7 +553,7 @@ ttn_get_dest(struct ttn_compile *c, struct tgsi_full_dst_register *tgsi_fdst)
 
          load->dest = nir_dest_for_reg(reg);
 
-         nir_instr_insert_after_cf_list(b->cf_node_list, &load->instr);
+         nir_builder_instr_insert(b, &load->instr);
       } else {
          assert(!tgsi_dst->Indirect);
          dest.dest.reg.reg = c->temp_regs[index].reg;
@@ -667,7 +668,7 @@ ttn_alu(nir_builder *b, nir_op op, nir_alu_dest dest, nir_ssa_def **src)
       instr->src[i].src = nir_src_for_ssa(src[i]);
 
    instr->dest = dest;
-   nir_instr_insert_after_cf_list(b->cf_node_list, &instr->instr);
+   nir_builder_instr_insert(b, &instr->instr);
 }
 
 static void
@@ -683,7 +684,7 @@ ttn_move_dest_masked(nir_builder *b, nir_alu_dest dest,
    mov->src[0].src = nir_src_for_ssa(def);
    for (unsigned i = def->num_components; i < 4; i++)
       mov->src[0].swizzle[i] = def->num_components - 1;
-   nir_instr_insert_after_cf_list(b->cf_node_list, &mov->instr);
+   nir_builder_instr_insert(b, &mov->instr);
 }
 
 static void
@@ -902,7 +903,7 @@ ttn_kill(nir_builder *b, nir_op op, nir_alu_dest dest, nir_ssa_def **src)
 {
    nir_intrinsic_instr *discard =
       nir_intrinsic_instr_create(b->shader, nir_intrinsic_discard);
-   nir_instr_insert_after_cf_list(b->cf_node_list, &discard->instr);
+   nir_builder_instr_insert(b, &discard->instr);
 }
 
 static void
@@ -912,7 +913,7 @@ ttn_kill_if(nir_builder *b, nir_op op, nir_alu_dest dest, nir_ssa_def **src)
    nir_intrinsic_instr *discard =
       nir_intrinsic_instr_create(b->shader, nir_intrinsic_discard_if);
    discard->src[0] = nir_src_for_ssa(cmp);
-   nir_instr_insert_after_cf_list(b->cf_node_list, &discard->instr);
+   nir_builder_instr_insert(b, &discard->instr);
 }
 
 static void
@@ -921,7 +922,7 @@ ttn_if(struct ttn_compile *c, nir_ssa_def *src, bool is_uint)
    nir_builder *b = &c->build;
 
    /* Save the outside-of-the-if-statement node list. */
-   c->if_stack[c->if_stack_pos] = b->cf_node_list;
+   c->if_stack[c->if_stack_pos] = b->cursor;
    c->if_stack_pos++;
 
    src = ttn_channel(b, src, X);
@@ -932,11 +933,11 @@ ttn_if(struct ttn_compile *c, nir_ssa_def *src, bool is_uint)
    } else {
       if_stmt->condition = nir_src_for_ssa(nir_fne(b, src, nir_imm_int(b, 0)));
    }
-   nir_cf_node_insert_end(b->cf_node_list, &if_stmt->cf_node);
+   nir_builder_cf_insert(b, &if_stmt->cf_node);
 
-   nir_builder_insert_after_cf_list(b, &if_stmt->then_list);
+   b->cursor = nir_after_cf_list(&if_stmt->then_list);
 
-   c->if_stack[c->if_stack_pos] = &if_stmt->else_list;
+   c->if_stack[c->if_stack_pos] = nir_after_cf_list(&if_stmt->else_list);
    c->if_stack_pos++;
 }
 
@@ -945,7 +946,7 @@ ttn_else(struct ttn_compile *c)
 {
    nir_builder *b = &c->build;
 
-   nir_builder_insert_after_cf_list(b, c->if_stack[c->if_stack_pos - 1]);
+   b->cursor = c->if_stack[c->if_stack_pos - 1];
 }
 
 static void
@@ -954,7 +955,7 @@ ttn_endif(struct ttn_compile *c)
    nir_builder *b = &c->build;
 
    c->if_stack_pos -= 2;
-   nir_builder_insert_after_cf_list(b, c->if_stack[c->if_stack_pos]);
+   b->cursor = c->if_stack[c->if_stack_pos];
 }
 
 static void
@@ -963,27 +964,27 @@ ttn_bgnloop(struct ttn_compile *c)
    nir_builder *b = &c->build;
 
    /* Save the outside-of-the-loop node list. */
-   c->loop_stack[c->loop_stack_pos] = b->cf_node_list;
+   c->loop_stack[c->loop_stack_pos] = b->cursor;
    c->loop_stack_pos++;
 
    nir_loop *loop = nir_loop_create(b->shader);
-   nir_cf_node_insert_end(b->cf_node_list, &loop->cf_node);
+   nir_builder_cf_insert(b, &loop->cf_node);
 
-   nir_builder_insert_after_cf_list(b, &loop->body);
+   b->cursor = nir_after_cf_list(&loop->body);
 }
 
 static void
 ttn_cont(nir_builder *b)
 {
    nir_jump_instr *instr = nir_jump_instr_create(b->shader, nir_jump_continue);
-   nir_instr_insert_after_cf_list(b->cf_node_list, &instr->instr);
+   nir_builder_instr_insert(b, &instr->instr);
 }
 
 static void
 ttn_brk(nir_builder *b)
 {
    nir_jump_instr *instr = nir_jump_instr_create(b->shader, nir_jump_break);
-   nir_instr_insert_after_cf_list(b->cf_node_list, &instr->instr);
+   nir_builder_instr_insert(b, &instr->instr);
 }
 
 static void
@@ -992,7 +993,7 @@ ttn_endloop(struct ttn_compile *c)
    nir_builder *b = &c->build;
 
    c->loop_stack_pos--;
-   nir_builder_insert_after_cf_list(b, c->loop_stack[c->loop_stack_pos]);
+   b->cursor = c->loop_stack[c->loop_stack_pos];
 }
 
 static void
@@ -1279,7 +1280,7 @@ ttn_tex(struct ttn_compile *c, nir_alu_dest dest, nir_ssa_def **src)
    assert(src_number == num_srcs);
 
    nir_ssa_dest_init(&instr->instr, &instr->dest, 4, NULL);
-   nir_instr_insert_after_cf_list(b->cf_node_list, &instr->instr);
+   nir_builder_instr_insert(b, &instr->instr);
 
    /* Resolve the writemask on the texture op. */
    ttn_move_dest(b, dest, &instr->dest.ssa);
@@ -1318,10 +1319,10 @@ ttn_txq(struct ttn_compile *c, nir_alu_dest dest, nir_ssa_def **src)
    txs->src[0].src_type = nir_tex_src_lod;
 
    nir_ssa_dest_init(&txs->instr, &txs->dest, 3, NULL);
-   nir_instr_insert_after_cf_list(b->cf_node_list, &txs->instr);
+   nir_builder_instr_insert(b, &txs->instr);
 
    nir_ssa_dest_init(&qlv->instr, &qlv->dest, 1, NULL);
-   nir_instr_insert_after_cf_list(b->cf_node_list, &qlv->instr);
+   nir_builder_instr_insert(b, &qlv->instr);
 
    ttn_move_dest_masked(b, dest, &txs->dest.ssa, TGSI_WRITEMASK_XYZ);
    ttn_move_dest_masked(b, dest, &qlv->dest.ssa, TGSI_WRITEMASK_W);
@@ -1730,7 +1731,7 @@ ttn_emit_instruction(struct ttn_compile *c)
       store->variables[0] = ttn_array_deref(c, store, var, offset, indirect);
       store->src[0] = nir_src_for_reg(dest.dest.reg.reg);
 
-      nir_instr_insert_after_cf_list(b->cf_node_list, &store->instr);
+      nir_builder_instr_insert(b, &store->instr);
    }
 }
 
@@ -1759,11 +1760,26 @@ ttn_add_output_stores(struct ttn_compile *c)
          store->const_index[0] = loc;
          store->src[0].reg.reg = c->output_regs[loc].reg;
          store->src[0].reg.base_offset = c->output_regs[loc].offset;
-         nir_instr_insert_after_cf_list(b->cf_node_list, &store->instr);
+         nir_builder_instr_insert(b, &store->instr);
       }
    }
 }
 
+static gl_shader_stage
+tgsi_processor_to_shader_stage(unsigned processor)
+{
+   switch (processor) {
+   case TGSI_PROCESSOR_FRAGMENT:  return MESA_SHADER_FRAGMENT;
+   case TGSI_PROCESSOR_VERTEX:    return MESA_SHADER_VERTEX;
+   case TGSI_PROCESSOR_GEOMETRY:  return MESA_SHADER_GEOMETRY;
+   case TGSI_PROCESSOR_TESS_CTRL: return MESA_SHADER_TESS_CTRL;
+   case TGSI_PROCESSOR_TESS_EVAL: return MESA_SHADER_TESS_EVAL;
+   case TGSI_PROCESSOR_COMPUTE:   return MESA_SHADER_COMPUTE;
+   default:
+      unreachable("invalid TGSI processor");
+   };
+}
+
 struct nir_shader *
 tgsi_to_nir(const void *tgsi_tokens,
             const nir_shader_compiler_options *options)
@@ -1775,17 +1791,19 @@ tgsi_to_nir(const void *tgsi_tokens,
    int ret;
 
    c = rzalloc(NULL, struct ttn_compile);
-   s = nir_shader_create(NULL, options);
+
+   tgsi_scan_shader(tgsi_tokens, &scan);
+   c->scan = &scan;
+
+   s = nir_shader_create(NULL, tgsi_processor_to_shader_stage(scan.processor),
+                         options);
 
    nir_function *func = nir_function_create(s, "main");
    nir_function_overload *overload = nir_function_overload_create(func);
    nir_function_impl *impl = nir_function_impl_create(overload);
 
    nir_builder_init(&c->build, impl);
-   nir_builder_insert_after_cf_list(&c->build, &impl->body);
-
-   tgsi_scan_shader(tgsi_tokens, &scan);
-   c->scan = &scan;
+   c->build.cursor = nir_after_cf_list(&impl->body);
 
    s->num_inputs = scan.file_max[TGSI_FILE_INPUT] + 1;
    s->num_uniforms = scan.const_file_max[0] + 1;
@@ -1801,10 +1819,10 @@ tgsi_to_nir(const void *tgsi_tokens,
    c->num_samp_types = scan.file_max[TGSI_FILE_SAMPLER_VIEW] + 1;
    c->samp_types = rzalloc_array(c, nir_alu_type, c->num_samp_types);
 
-   c->if_stack = rzalloc_array(c, struct exec_list *,
+   c->if_stack = rzalloc_array(c, nir_cursor,
                                (scan.opcode_count[TGSI_OPCODE_IF] +
                                 scan.opcode_count[TGSI_OPCODE_UIF]) * 2);
-   c->loop_stack = rzalloc_array(c, struct exec_list *,
+   c->loop_stack = rzalloc_array(c, nir_cursor,
                                  scan.opcode_count[TGSI_OPCODE_BGNLOOP]);
 
    ret = tgsi_parse_init(&parser, tgsi_tokens);

src/gallium/auxiliary/pipebuffer/pb_bufmgr.h

@@ -166,11 +166,6 @@ pb_cache_manager_create(struct pb_manager *provider,
                         unsigned bypass_usage,
                         uint64_t maximum_cache_size);
 
-/**
- * Remove a buffer from the cache, but keep it alive.
- */
-void
-pb_cache_manager_remove_buffer(struct pb_buffer *buf);
 
 struct pb_fence_ops;
 

src/gallium/auxiliary/pipebuffer/pb_bufmgr_cache.c

@@ -104,42 +104,18 @@ pb_cache_manager(struct pb_manager *mgr)
 }
 
 
-static void
-_pb_cache_manager_remove_buffer_locked(struct pb_cache_buffer *buf)
-{
-   struct pb_cache_manager *mgr = buf->mgr;
-
-   if (buf->head.next) {
-      LIST_DEL(&buf->head);
-      assert(mgr->numDelayed);
-      --mgr->numDelayed;
-      mgr->cache_size -= buf->base.size;
-   }
-   buf->mgr = NULL;
-}
-
-void
-pb_cache_manager_remove_buffer(struct pb_buffer *pb_buf)
-{
-   struct pb_cache_buffer *buf = (struct pb_cache_buffer*)pb_buf;
-   struct pb_cache_manager *mgr = buf->mgr;
-
-   if (!mgr)
-      return;
-
-   pipe_mutex_lock(mgr->mutex);
-   _pb_cache_manager_remove_buffer_locked(buf);
-   pipe_mutex_unlock(mgr->mutex);
-}
-
 /**
  * Actually destroy the buffer.
  */
 static inline void
 _pb_cache_buffer_destroy(struct pb_cache_buffer *buf)
 {
-   if (buf->mgr)
-      _pb_cache_manager_remove_buffer_locked(buf);
+   struct pb_cache_manager *mgr = buf->mgr;
+
+   LIST_DEL(&buf->head);
+   assert(mgr->numDelayed);
+   --mgr->numDelayed;
+   mgr->cache_size -= buf->base.size;
    assert(!pipe_is_referenced(&buf->base.reference));
    pb_reference(&buf->buffer, NULL);
    FREE(buf);
@@ -180,12 +156,6 @@ pb_cache_buffer_destroy(struct pb_buffer *_buf)
    struct pb_cache_buffer *buf = pb_cache_buffer(_buf);   
    struct pb_cache_manager *mgr = buf->mgr;
 
-   if (!mgr) {
-      pb_reference(&buf->buffer, NULL);
-      FREE(buf);
-      return;
-   }
-
    pipe_mutex_lock(mgr->mutex);
    assert(!pipe_is_referenced(&buf->base.reference));
    

src/gallium/auxiliary/target-helpers/inline_debug_helper.h

@@ -11,6 +11,10 @@
  * one or more debug driver: rbug, trace.
  */
 
+#ifdef GALLIUM_DDEBUG
+#include "ddebug/dd_public.h"
+#endif
+
 #ifdef GALLIUM_TRACE
 #include "trace/tr_public.h"
 #endif
@@ -30,6 +34,10 @@
 static inline struct pipe_screen *
 debug_screen_wrap(struct pipe_screen *screen)
 {
+#if defined(GALLIUM_DDEBUG)
+   screen = ddebug_screen_create(screen);
+#endif
+
 #if defined(GALLIUM_RBUG)
    screen = rbug_screen_create(screen);
 #endif

src/gallium/auxiliary/tgsi/tgsi_exec.h

@@ -463,8 +463,6 @@ tgsi_exec_get_shader_param(enum pipe_shader_cap param)
    case PIPE_SHADER_CAP_TGSI_DROUND_SUPPORTED:
    case PIPE_SHADER_CAP_TGSI_FMA_SUPPORTED:
       return 0;
-   case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-      return 32;
    }
    /* if we get here, we missed a shader cap above (and should have seen
     * a compiler warning.)

src/gallium/auxiliary/util/u_blitter.c

@@ -1190,8 +1190,6 @@ static void blitter_draw(struct blitter_context_priv *ctx,
 
    u_upload_data(ctx->upload, 0, sizeof(ctx->vertices), ctx->vertices,
                  &vb.buffer_offset, &vb.buffer);
-   if (!vb.buffer)
-      return;
    u_upload_unmap(ctx->upload);
 
    pipe->set_vertex_buffers(pipe, ctx->base.vb_slot, 1, &vb);
@@ -2091,9 +2089,6 @@ void util_blitter_clear_buffer(struct blitter_context *blitter,
 
    u_upload_data(ctx->upload, 0, num_channels*4, clear_value,
                  &vb.buffer_offset, &vb.buffer);
-   if (!vb.buffer)
-      goto out;
-
    vb.stride = 0;
 
    blitter_set_running_flag(ctx);
@@ -2117,7 +2112,6 @@ void util_blitter_clear_buffer(struct blitter_context *blitter,
 
    util_draw_arrays(pipe, PIPE_PRIM_POINTS, 0, size / 4);
 
-out:
    blitter_restore_vertex_states(ctx);
    blitter_restore_render_cond(ctx);
    blitter_unset_running_flag(ctx);

src/gallium/auxiliary/util/u_blitter.h

@@ -372,30 +372,28 @@ void util_blitter_custom_resolve_color(struct blitter_context *blitter,
  *
  * States not listed here are not affected by util_blitter. */
 
-static inline
-void util_blitter_save_blend(struct blitter_context *blitter,
-                             void *state)
+static inline void
+util_blitter_save_blend(struct blitter_context *blitter, void *state)
 {
    blitter->saved_blend_state = state;
 }
 
-static inline
-void util_blitter_save_depth_stencil_alpha(struct blitter_context *blitter,
-                                           void *state)
+static inline void
+util_blitter_save_depth_stencil_alpha(struct blitter_context *blitter,
+                                      void *state)
 {
    blitter->saved_dsa_state = state;
 }
 
-static inline
-void util_blitter_save_vertex_elements(struct blitter_context *blitter,
-                                       void *state)
+static inline void
+util_blitter_save_vertex_elements(struct blitter_context *blitter, void *state)
 {
    blitter->saved_velem_state = state;
 }
 
-static inline
-void util_blitter_save_stencil_ref(struct blitter_context *blitter,
-                                   const struct pipe_stencil_ref *state)
+static inline void
+util_blitter_save_stencil_ref(struct blitter_context *blitter,
+                              const struct pipe_stencil_ref *state)
 {
    blitter->saved_stencil_ref = *state;
 }
@@ -407,23 +405,20 @@ void util_blitter_save_rasterizer(struct blitter_context *blitter,
    blitter->saved_rs_state = state;
 }
 
-static inline
-void util_blitter_save_fragment_shader(struct blitter_context *blitter,
-                                       void *fs)
+static inline void
+util_blitter_save_fragment_shader(struct blitter_context *blitter, void *fs)
 {
    blitter->saved_fs = fs;
 }
 
-static inline
-void util_blitter_save_vertex_shader(struct blitter_context *blitter,
-                                     void *vs)
+static inline void
+util_blitter_save_vertex_shader(struct blitter_context *blitter, void *vs)
 {
    blitter->saved_vs = vs;
 }
 
-static inline
-void util_blitter_save_geometry_shader(struct blitter_context *blitter,
-                                       void *gs)
+static inline void
+util_blitter_save_geometry_shader(struct blitter_context *blitter, void *gs)
 {
    blitter->saved_gs = gs;
 }
@@ -442,24 +437,24 @@ util_blitter_save_tesseval_shader(struct blitter_context *blitter,
    blitter->saved_tes = sh;
 }
 
-static inline
-void util_blitter_save_framebuffer(struct blitter_context *blitter,
-                                   const struct pipe_framebuffer_state *state)
+static inline void
+util_blitter_save_framebuffer(struct blitter_context *blitter,
+                              const struct pipe_framebuffer_state *state)
 {
    blitter->saved_fb_state.nr_cbufs = 0; /* It's ~0 now, meaning it's unsaved. */
    util_copy_framebuffer_state(&blitter->saved_fb_state, state);
 }
 
-static inline
-void util_blitter_save_viewport(struct blitter_context *blitter,
-                                struct pipe_viewport_state *state)
+static inline void
+util_blitter_save_viewport(struct blitter_context *blitter,
+                           struct pipe_viewport_state *state)
 {
    blitter->saved_viewport = *state;
 }
 
-static inline
-void util_blitter_save_scissor(struct blitter_context *blitter,
-                               struct pipe_scissor_state *state)
+static inline void
+util_blitter_save_scissor(struct blitter_context *blitter,
+                          struct pipe_scissor_state *state)
 {
    blitter->saved_scissor = *state;
 }

src/gallium/auxiliary/util/u_format_rgb9e5.h

@@ -21,7 +21,8 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
-/* Copied from EXT_texture_shared_exponent and edited. */
+/* Copied from EXT_texture_shared_exponent and edited, getting rid of
+ * expensive float math bits too. */
 
 #ifndef RGB9E5_H
 #define RGB9E5_H
@@ -39,7 +40,6 @@
 #define RGB9E5_MANTISSA_VALUES       (1<<RGB9E5_MANTISSA_BITS)
 #define MAX_RGB9E5_MANTISSA          (RGB9E5_MANTISSA_VALUES-1)
 #define MAX_RGB9E5                   (((float)MAX_RGB9E5_MANTISSA)/RGB9E5_MANTISSA_VALUES * (1<<MAX_RGB9E5_EXP))
-#define EPSILON_RGB9E5               ((1.0/RGB9E5_MANTISSA_VALUES) / (1<<RGB9E5_EXP_BIAS))
 
 typedef union {
    unsigned int raw;
@@ -74,63 +74,59 @@ typedef union {
    } field;
 } rgb9e5;
 
-static inline float rgb9e5_ClampRange(float x)
-{
-   if (x > 0.0f) {
-      if (x >= MAX_RGB9E5) {
-         return MAX_RGB9E5;
-      } else {
-         return x;
-      }
-   } else {
-      /* NaN gets here too since comparisons with NaN always fail! */
-      return 0.0;
-   }
-}
 
-/* Ok, FloorLog2 is not correct for the denorm and zero values, but we
-   are going to do a max of this value with the minimum rgb9e5 exponent
-   that will hide these problem cases. */
-static inline int rgb9e5_FloorLog2(float x)
+static inline int rgb9e5_ClampRange(float x)
 {
    float754 f;
-
+   float754 max;
    f.value = x;
-   return (f.field.biasedexponent - 127);
+   max.value = MAX_RGB9E5;
+
+   if (f.raw > 0x7f800000)
+  /* catches neg, NaNs */
+      return 0;
+   else if (f.raw >= max.raw)
+      return max.raw;
+   else
+      return f.raw;
 }
 
 static inline unsigned float3_to_rgb9e5(const float rgb[3])
 {
    rgb9e5 retval;
-   float maxrgb;
-   int rm, gm, bm;
-   float rc, gc, bc;
-   int exp_shared, maxm;
-   double denom;
+   int rm, gm, bm, exp_shared;
+   float754 revdenom = {0};
+   float754 rc, bc, gc, maxrgb;
 
-   rc = rgb9e5_ClampRange(rgb[0]);
-   gc = rgb9e5_ClampRange(rgb[1]);
-   bc = rgb9e5_ClampRange(rgb[2]);
+   rc.raw = rgb9e5_ClampRange(rgb[0]);
+   gc.raw = rgb9e5_ClampRange(rgb[1]);
+   bc.raw = rgb9e5_ClampRange(rgb[2]);
+   maxrgb.raw = MAX3(rc.raw, gc.raw, bc.raw);
 
-   maxrgb = MAX3(rc, gc, bc);
-   exp_shared = MAX2(-RGB9E5_EXP_BIAS-1, rgb9e5_FloorLog2(maxrgb)) + 1 + RGB9E5_EXP_BIAS;
+   /*
+    * Compared to what the spec suggests, instead of conditionally adjusting
+    * the exponent after the fact do it here by doing the equivalent of +0.5 -
+    * the int add will spill over into the exponent in this case.
+    */
+   maxrgb.raw += maxrgb.raw & (1 << (23-9));
+   exp_shared = MAX2((maxrgb.raw >> 23), -RGB9E5_EXP_BIAS - 1 + 127) +
+                1 + RGB9E5_EXP_BIAS - 127;
+   revdenom.field.biasedexponent = 127 - (exp_shared - RGB9E5_EXP_BIAS -
+                                          RGB9E5_MANTISSA_BITS) + 1;
    assert(exp_shared <= RGB9E5_MAX_VALID_BIASED_EXP);
-   assert(exp_shared >= 0);
-   /* This exp2 function could be replaced by a table. */
-   denom = exp2(exp_shared - RGB9E5_EXP_BIAS - RGB9E5_MANTISSA_BITS);
 
-   maxm = (int) floor(maxrgb / denom + 0.5);
-   if (maxm == MAX_RGB9E5_MANTISSA+1) {
-      denom *= 2;
-      exp_shared += 1;
-      assert(exp_shared <= RGB9E5_MAX_VALID_BIASED_EXP);
-   } else {
-      assert(maxm <= MAX_RGB9E5_MANTISSA);
-   }
-
-   rm = (int) floor(rc / denom + 0.5);
-   gm = (int) floor(gc / denom + 0.5);
-   bm = (int) floor(bc / denom + 0.5);
+   /*
+    * The spec uses strict round-up behavior (d3d10 disagrees, but in any case
+    * must match what is done above for figuring out exponent).
+    * We avoid the doubles ((int) rc * revdenom + 0.5) by doing the rounding
+    * ourselves (revdenom was adjusted by +1, above).
+    */
+   rm = (int) (rc.value * revdenom.value);
+   gm = (int) (gc.value * revdenom.value);
+   bm = (int) (bc.value * revdenom.value);
+   rm = (rm & 1) + (rm >> 1);
+   gm = (gm & 1) + (gm >> 1);
+   bm = (bm & 1) + (bm >> 1);
 
    assert(rm <= MAX_RGB9E5_MANTISSA);
    assert(gm <= MAX_RGB9E5_MANTISSA);
@@ -151,15 +147,15 @@ static inline void rgb9e5_to_float3(unsigned rgb, float retval[3])
 {
    rgb9e5 v;
    int exponent;
-   float scale;
+   float754 scale = {0};
 
    v.raw = rgb;
    exponent = v.field.biasedexponent - RGB9E5_EXP_BIAS - RGB9E5_MANTISSA_BITS;
-   scale = exp2f(exponent);
+   scale.field.biasedexponent = exponent + 127;
 
-   retval[0] = v.field.r * scale;
-   retval[1] = v.field.g * scale;
-   retval[2] = v.field.b * scale;
+   retval[0] = v.field.r * scale.value;
+   retval[1] = v.field.g * scale.value;
+   retval[2] = v.field.b * scale.value;
 }
 
 #endif

src/gallium/auxiliary/util/u_tests.c

@@ -457,7 +457,7 @@ null_constant_buffer(struct pipe_context *ctx)
 void
 util_run_tests(struct pipe_screen *screen)
 {
-   struct pipe_context *ctx = screen->context_create(screen, NULL);
+   struct pipe_context *ctx = screen->context_create(screen, NULL, 0);
 
    tgsi_vs_window_space_position(ctx);
    null_sampler_view(ctx, TGSI_TEXTURE_2D);

src/gallium/auxiliary/util/u_vbuf.c

@@ -545,7 +545,6 @@ u_vbuf_translate_find_free_vb_slots(struct u_vbuf *mgr,
 
          index = ffs(unused_vb_mask) - 1;
          fallback_vbs[type] = index;
-         unused_vb_mask &= ~(1 << index);
          /*printf("found slot=%i for type=%i\n", index, type);*/
       }
    }

src/gallium/auxiliary/vl/vl_mpeg12_decoder.c

@@ -1120,7 +1120,7 @@ vl_create_mpeg12_decoder(struct pipe_context *context,
 
    dec->base = *templat;
    dec->base.context = context;
-   dec->context = context->screen->context_create(context->screen, NULL);
+   dec->context = context->screen->context_create(context->screen, NULL, 0);
 
    dec->base.destroy = vl_mpeg12_destroy;
    dec->base.begin_frame = vl_mpeg12_begin_frame;

src/gallium/docs/source/screen.rst

@@ -355,10 +355,6 @@ to be 0.
   are supported.
 * ``PIPE_SHADER_CAP_TGSI_ANY_INOUT_DECL_RANGE``: Whether the driver doesn't
   ignore tgsi_declaration_range::Last for shader inputs and outputs.
-* ``PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT``: This is the maximum number
-  of iterations that loops are allowed to have to be unrolled. It is only
-  a hint to state trackers. Whether any loops will be unrolled is not
-  guaranteed.
 
 
 .. _pipe_compute_cap:

src/gallium/drivers/ddebug/Makefile.am

@@ -0,0 +1,9 @@
+include Makefile.sources
+include $(top_srcdir)/src/gallium/Automake.inc
+
+AM_CFLAGS = \
+	$(GALLIUM_DRIVER_CFLAGS)
+
+noinst_LTLIBRARIES = libddebug.la
+
+libddebug_la_SOURCES = $(C_SOURCES)

src/gallium/drivers/ddebug/Makefile.sources

@@ -0,0 +1,6 @@
+C_SOURCES := \
+	dd_pipe.h \
+	dd_public.h \
+	dd_context.c \
+	dd_draw.c \
+	dd_screen.c

src/gallium/drivers/ddebug/dd_context.c

@@ -0,0 +1,771 @@
+/**************************************************************************
+ *
+ * Copyright 2015 Advanced Micro Devices, Inc.
+ * Copyright 2008 VMware, Inc.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * on the rights to use, copy, modify, merge, publish, distribute, sub
+ * license, and/or sell copies of the Software, and to permit persons to whom
+ * the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHOR(S) AND/OR THEIR SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ **************************************************************************/
+
+#include "dd_pipe.h"
+#include "tgsi/tgsi_parse.h"
+#include "util/u_memory.h"
+
+
+static void
+safe_memcpy(void *dst, const void *src, size_t size)
+{
+   if (src)
+      memcpy(dst, src, size);
+   else
+      memset(dst, 0, size);
+}
+
+
+/********************************************************************
+ * queries
+ */
+
+static struct dd_query *
+dd_query(struct pipe_query *query)
+{
+   return (struct dd_query *)query;
+}
+
+static struct pipe_query *
+dd_query_unwrap(struct pipe_query *query)
+{
+   if (query) {
+      return dd_query(query)->query;
+   } else {
+      return NULL;
+   }
+}
+
+static struct pipe_query *
+dd_context_create_query(struct pipe_context *_pipe, unsigned query_type,
+                        unsigned index)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+   struct pipe_query *query;
+
+   query = pipe->create_query(pipe, query_type, index);
+
+   /* Wrap query object. */
+   if (query) {
+      struct dd_query *dd_query = CALLOC_STRUCT(dd_query);
+      if (dd_query) {
+         dd_query->type = query_type;
+         dd_query->query = query;
+         query = (struct pipe_query *)dd_query;
+      } else {
+         pipe->destroy_query(pipe, query);
+         query = NULL;
+      }
+   }
+
+   return query;
+}
+
+static void
+dd_context_destroy_query(struct pipe_context *_pipe,
+                         struct pipe_query *query)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->destroy_query(pipe, dd_query_unwrap(query));
+   FREE(query);
+}
+
+static boolean
+dd_context_begin_query(struct pipe_context *_pipe, struct pipe_query *query)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   return pipe->begin_query(pipe, dd_query_unwrap(query));
+}
+
+static void
+dd_context_end_query(struct pipe_context *_pipe, struct pipe_query *query)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   pipe->end_query(pipe, dd_query_unwrap(query));
+}
+
+static boolean
+dd_context_get_query_result(struct pipe_context *_pipe,
+                            struct pipe_query *query, boolean wait,
+                            union pipe_query_result *result)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   return pipe->get_query_result(pipe, dd_query_unwrap(query), wait, result);
+}
+
+static void
+dd_context_render_condition(struct pipe_context *_pipe,
+                            struct pipe_query *query, boolean condition,
+                            uint mode)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   pipe->render_condition(pipe, dd_query_unwrap(query), condition, mode);
+   dctx->render_cond.query = dd_query(query);
+   dctx->render_cond.condition = condition;
+   dctx->render_cond.mode = mode;
+}
+
+
+/********************************************************************
+ * constant (immutable) non-shader states
+ */
+
+#define DD_CSO_CREATE(name, shortname) \
+   static void * \
+   dd_context_create_##name##_state(struct pipe_context *_pipe, \
+                                    const struct pipe_##name##_state *state) \
+   { \
+      struct pipe_context *pipe = dd_context(_pipe)->pipe; \
+      struct dd_state *hstate = CALLOC_STRUCT(dd_state); \
+ \
+      if (!hstate) \
+         return NULL; \
+      hstate->cso = pipe->create_##name##_state(pipe, state); \
+      hstate->state.shortname = *state; \
+      return hstate; \
+   }
+
+#define DD_CSO_BIND(name, shortname) \
+   static void \
+   dd_context_bind_##name##_state(struct pipe_context *_pipe, void *state) \
+   { \
+      struct dd_context *dctx = dd_context(_pipe); \
+      struct pipe_context *pipe = dctx->pipe; \
+      struct dd_state *hstate = state; \
+ \
+      dctx->shortname = hstate; \
+      pipe->bind_##name##_state(pipe, hstate ? hstate->cso : NULL); \
+   }
+
+#define DD_CSO_DELETE(name) \
+   static void \
+   dd_context_delete_##name##_state(struct pipe_context *_pipe, void *state) \
+   { \
+      struct dd_context *dctx = dd_context(_pipe); \
+      struct pipe_context *pipe = dctx->pipe; \
+      struct dd_state *hstate = state; \
+ \
+      pipe->delete_##name##_state(pipe, hstate->cso); \
+      FREE(hstate); \
+   }
+
+#define DD_CSO_WHOLE(name, shortname) \
+   DD_CSO_CREATE(name, shortname) \
+   DD_CSO_BIND(name, shortname) \
+   DD_CSO_DELETE(name)
+
+DD_CSO_WHOLE(blend, blend)
+DD_CSO_WHOLE(rasterizer, rs)
+DD_CSO_WHOLE(depth_stencil_alpha, dsa)
+
+DD_CSO_CREATE(sampler, sampler)
+DD_CSO_DELETE(sampler)
+
+static void
+dd_context_bind_sampler_states(struct pipe_context *_pipe, unsigned shader,
+                               unsigned start, unsigned count, void **states)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   memcpy(&dctx->sampler_states[shader][start], states,
+          sizeof(void*) * count);
+
+   if (states) {
+      void *samp[PIPE_MAX_SAMPLERS];
+      int i;
+
+      for (i = 0; i < count; i++) {
+         struct dd_state *s = states[i];
+         samp[i] = s ? s->cso : NULL;
+      }
+
+      pipe->bind_sampler_states(pipe, shader, start, count, samp);
+   }
+   else
+      pipe->bind_sampler_states(pipe, shader, start, count, NULL);
+}
+
+static void *
+dd_context_create_vertex_elements_state(struct pipe_context *_pipe,
+                                        unsigned num_elems,
+                                        const struct pipe_vertex_element *elems)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+   struct dd_state *hstate = CALLOC_STRUCT(dd_state);
+
+   if (!hstate)
+      return NULL;
+   hstate->cso = pipe->create_vertex_elements_state(pipe, num_elems, elems);
+   memcpy(hstate->state.velems.velems, elems, sizeof(elems[0]) * num_elems);
+   hstate->state.velems.count = num_elems;
+   return hstate;
+}
+
+DD_CSO_BIND(vertex_elements, velems)
+DD_CSO_DELETE(vertex_elements)
+
+
+/********************************************************************
+ * shaders
+ */
+
+#define DD_SHADER(NAME, name) \
+   static void * \
+   dd_context_create_##name##_state(struct pipe_context *_pipe, \
+                                    const struct pipe_shader_state *state) \
+   { \
+      struct pipe_context *pipe = dd_context(_pipe)->pipe; \
+      struct dd_state *hstate = CALLOC_STRUCT(dd_state); \
+ \
+      if (!hstate) \
+         return NULL; \
+      hstate->cso = pipe->create_##name##_state(pipe, state); \
+      hstate->state.shader = *state; \
+      hstate->state.shader.tokens = tgsi_dup_tokens(state->tokens); \
+      return hstate; \
+   } \
+    \
+   static void \
+   dd_context_bind_##name##_state(struct pipe_context *_pipe, void *state) \
+   { \
+      struct dd_context *dctx = dd_context(_pipe); \
+      struct pipe_context *pipe = dctx->pipe; \
+      struct dd_state *hstate = state; \
+   \
+      dctx->shaders[PIPE_SHADER_##NAME] = hstate; \
+      pipe->bind_##name##_state(pipe, hstate ? hstate->cso : NULL); \
+   } \
+    \
+   static void \
+   dd_context_delete_##name##_state(struct pipe_context *_pipe, void *state) \
+   { \
+      struct dd_context *dctx = dd_context(_pipe); \
+      struct pipe_context *pipe = dctx->pipe; \
+      struct dd_state *hstate = state; \
+   \
+      pipe->delete_##name##_state(pipe, hstate->cso); \
+      tgsi_free_tokens(hstate->state.shader.tokens); \
+      FREE(hstate); \
+   }
+
+DD_SHADER(FRAGMENT, fs)
+DD_SHADER(VERTEX, vs)
+DD_SHADER(GEOMETRY, gs)
+DD_SHADER(TESS_CTRL, tcs)
+DD_SHADER(TESS_EVAL, tes)
+
+
+/********************************************************************
+ * immediate states
+ */
+
+#define DD_IMM_STATE(name, type, deref, ref) \
+   static void \
+   dd_context_set_##name(struct pipe_context *_pipe, type deref) \
+   { \
+      struct dd_context *dctx = dd_context(_pipe); \
+      struct pipe_context *pipe = dctx->pipe; \
+ \
+      dctx->name = deref; \
+      pipe->set_##name(pipe, ref); \
+   }
+
+DD_IMM_STATE(blend_color, const struct pipe_blend_color, *state, state)
+DD_IMM_STATE(stencil_ref, const struct pipe_stencil_ref, *state, state)
+DD_IMM_STATE(clip_state, const struct pipe_clip_state, *state, state)
+DD_IMM_STATE(sample_mask, unsigned, sample_mask, sample_mask)
+DD_IMM_STATE(min_samples, unsigned, min_samples, min_samples)
+DD_IMM_STATE(framebuffer_state, const struct pipe_framebuffer_state, *state, state)
+DD_IMM_STATE(polygon_stipple, const struct pipe_poly_stipple, *state, state)
+
+static void
+dd_context_set_constant_buffer(struct pipe_context *_pipe,
+                               uint shader, uint index,
+                               struct pipe_constant_buffer *constant_buffer)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->constant_buffers[shader][index], constant_buffer,
+          sizeof(*constant_buffer));
+   pipe->set_constant_buffer(pipe, shader, index, constant_buffer);
+}
+
+static void
+dd_context_set_scissor_states(struct pipe_context *_pipe,
+                              unsigned start_slot, unsigned num_scissors,
+                              const struct pipe_scissor_state *states)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->scissors[start_slot], states,
+               sizeof(*states) * num_scissors);
+   pipe->set_scissor_states(pipe, start_slot, num_scissors, states);
+}
+
+static void
+dd_context_set_viewport_states(struct pipe_context *_pipe,
+                               unsigned start_slot, unsigned num_viewports,
+                               const struct pipe_viewport_state *states)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->viewports[start_slot], states,
+               sizeof(*states) * num_viewports);
+   pipe->set_viewport_states(pipe, start_slot, num_viewports, states);
+}
+
+static void dd_context_set_tess_state(struct pipe_context *_pipe,
+                                      const float default_outer_level[4],
+                                      const float default_inner_level[2])
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   memcpy(dctx->tess_default_levels, default_outer_level, sizeof(float) * 4);
+   memcpy(dctx->tess_default_levels+4, default_inner_level, sizeof(float) * 2);
+   pipe->set_tess_state(pipe, default_outer_level, default_inner_level);
+}
+
+
+/********************************************************************
+ * views
+ */
+
+static struct pipe_surface *
+dd_context_create_surface(struct pipe_context *_pipe,
+                          struct pipe_resource *resource,
+                          const struct pipe_surface *surf_tmpl)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+   struct pipe_surface *view =
+      pipe->create_surface(pipe, resource, surf_tmpl);
+
+   if (!view)
+      return NULL;
+   view->context = _pipe;
+   return view;
+}
+
+static void
+dd_context_surface_destroy(struct pipe_context *_pipe,
+                           struct pipe_surface *surf)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->surface_destroy(pipe, surf);
+}
+
+static struct pipe_sampler_view *
+dd_context_create_sampler_view(struct pipe_context *_pipe,
+                               struct pipe_resource *resource,
+                               const struct pipe_sampler_view *templ)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+   struct pipe_sampler_view *view =
+      pipe->create_sampler_view(pipe, resource, templ);
+
+   if (!view)
+      return NULL;
+   view->context = _pipe;
+   return view;
+}
+
+static void
+dd_context_sampler_view_destroy(struct pipe_context *_pipe,
+                                struct pipe_sampler_view *view)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->sampler_view_destroy(pipe, view);
+}
+
+static struct pipe_image_view *
+dd_context_create_image_view(struct pipe_context *_pipe,
+                             struct pipe_resource *resource,
+                             const struct pipe_image_view *templ)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+   struct pipe_image_view *view =
+      pipe->create_image_view(pipe, resource, templ);
+
+   if (!view)
+      return NULL;
+   view->context = _pipe;
+   return view;
+}
+
+static void
+dd_context_image_view_destroy(struct pipe_context *_pipe,
+                              struct pipe_image_view *view)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->image_view_destroy(pipe, view);
+}
+
+static struct pipe_stream_output_target *
+dd_context_create_stream_output_target(struct pipe_context *_pipe,
+                                       struct pipe_resource *res,
+                                       unsigned buffer_offset,
+                                       unsigned buffer_size)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+   struct pipe_stream_output_target *view =
+      pipe->create_stream_output_target(pipe, res, buffer_offset,
+                                        buffer_size);
+
+   if (!view)
+      return NULL;
+   view->context = _pipe;
+   return view;
+}
+
+static void
+dd_context_stream_output_target_destroy(struct pipe_context *_pipe,
+                                        struct pipe_stream_output_target *target)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->stream_output_target_destroy(pipe, target);
+}
+
+
+/********************************************************************
+ * set states
+ */
+
+static void
+dd_context_set_sampler_views(struct pipe_context *_pipe, unsigned shader,
+                             unsigned start, unsigned num,
+                             struct pipe_sampler_view **views)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->sampler_views[shader][start], views,
+               sizeof(views[0]) * num);
+   pipe->set_sampler_views(pipe, shader, start, num, views);
+}
+
+static void
+dd_context_set_shader_images(struct pipe_context *_pipe, unsigned shader,
+                             unsigned start, unsigned num,
+                             struct pipe_image_view **views)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->shader_images[shader][start], views,
+               sizeof(views[0]) * num);
+   pipe->set_shader_images(pipe, shader, start, num, views);
+}
+
+static void
+dd_context_set_shader_buffers(struct pipe_context *_pipe, unsigned shader,
+                              unsigned start, unsigned num_buffers,
+                              struct pipe_shader_buffer *buffers)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->shader_buffers[shader][start], buffers,
+               sizeof(buffers[0]) * num_buffers);
+   pipe->set_shader_buffers(pipe, shader, start, num_buffers, buffers);
+}
+
+static void
+dd_context_set_vertex_buffers(struct pipe_context *_pipe,
+                              unsigned start, unsigned num_buffers,
+                              const struct pipe_vertex_buffer *buffers)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->vertex_buffers[start], buffers,
+               sizeof(buffers[0]) * num_buffers);
+   pipe->set_vertex_buffers(pipe, start, num_buffers, buffers);
+}
+
+static void
+dd_context_set_index_buffer(struct pipe_context *_pipe,
+                            const struct pipe_index_buffer *ib)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   safe_memcpy(&dctx->index_buffer, ib, sizeof(*ib));
+   pipe->set_index_buffer(pipe, ib);
+}
+
+static void
+dd_context_set_stream_output_targets(struct pipe_context *_pipe,
+                                     unsigned num_targets,
+                                     struct pipe_stream_output_target **tgs,
+                                     const unsigned *offsets)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   dctx->num_so_targets = num_targets;
+   safe_memcpy(dctx->so_targets, tgs, sizeof(*tgs) * num_targets);
+   safe_memcpy(dctx->so_offsets, offsets, sizeof(*offsets) * num_targets);
+   pipe->set_stream_output_targets(pipe, num_targets, tgs, offsets);
+}
+
+static void
+dd_context_destroy(struct pipe_context *_pipe)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   pipe->destroy(pipe);
+   FREE(dctx);
+}
+
+
+/********************************************************************
+ * transfer
+ */
+
+static void *
+dd_context_transfer_map(struct pipe_context *_pipe,
+                        struct pipe_resource *resource, unsigned level,
+                        unsigned usage, const struct pipe_box *box,
+                        struct pipe_transfer **transfer)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   return pipe->transfer_map(pipe, resource, level, usage, box, transfer);
+}
+
+static void
+dd_context_transfer_flush_region(struct pipe_context *_pipe,
+                                 struct pipe_transfer *transfer,
+                                 const struct pipe_box *box)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->transfer_flush_region(pipe, transfer, box);
+}
+
+static void
+dd_context_transfer_unmap(struct pipe_context *_pipe,
+                          struct pipe_transfer *transfer)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->transfer_unmap(pipe, transfer);
+}
+
+static void
+dd_context_transfer_inline_write(struct pipe_context *_pipe,
+                                 struct pipe_resource *resource,
+                                 unsigned level, unsigned usage,
+                                 const struct pipe_box *box,
+                                 const void *data, unsigned stride,
+                                 unsigned layer_stride)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->transfer_inline_write(pipe, resource, level, usage, box, data,
+                               stride, layer_stride);
+}
+
+
+/********************************************************************
+ * miscellaneous
+ */
+
+static void
+dd_context_texture_barrier(struct pipe_context *_pipe)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->texture_barrier(pipe);
+}
+
+static void
+dd_context_memory_barrier(struct pipe_context *_pipe, unsigned flags)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->memory_barrier(pipe, flags);
+}
+
+static void
+dd_context_get_sample_position(struct pipe_context *_pipe,
+                               unsigned sample_count, unsigned sample_index,
+                               float *out_value)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   return pipe->get_sample_position(pipe, sample_count, sample_index,
+                                    out_value);
+}
+
+static void
+dd_context_invalidate_resource(struct pipe_context *_pipe,
+                               struct pipe_resource *resource)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   pipe->invalidate_resource(pipe, resource);
+}
+
+static enum pipe_reset_status
+dd_context_get_device_reset_status(struct pipe_context *_pipe)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   return pipe->get_device_reset_status(pipe);
+}
+
+static void
+dd_context_dump_debug_state(struct pipe_context *_pipe, FILE *stream,
+                            unsigned flags)
+{
+   struct pipe_context *pipe = dd_context(_pipe)->pipe;
+
+   return pipe->dump_debug_state(pipe, stream, flags);
+}
+
+struct pipe_context *
+dd_context_create(struct dd_screen *dscreen, struct pipe_context *pipe)
+{
+   struct dd_context *dctx;
+
+   if (!pipe)
+      return NULL;
+
+   dctx = CALLOC_STRUCT(dd_context);
+   if (!dctx) {
+      pipe->destroy(pipe);
+      return NULL;
+   }
+
+   dctx->pipe = pipe;
+   dctx->base.priv = pipe->priv; /* expose wrapped priv data */
+   dctx->base.screen = &dscreen->base;
+
+   dctx->base.destroy = dd_context_destroy;
+
+   CTX_INIT(render_condition);
+   CTX_INIT(create_query);
+   CTX_INIT(destroy_query);
+   CTX_INIT(begin_query);
+   CTX_INIT(end_query);
+   CTX_INIT(get_query_result);
+   CTX_INIT(create_blend_state);
+   CTX_INIT(bind_blend_state);
+   CTX_INIT(delete_blend_state);
+   CTX_INIT(create_sampler_state);
+   CTX_INIT(bind_sampler_states);
+   CTX_INIT(delete_sampler_state);
+   CTX_INIT(create_rasterizer_state);
+   CTX_INIT(bind_rasterizer_state);
+   CTX_INIT(delete_rasterizer_state);
+   CTX_INIT(create_depth_stencil_alpha_state);
+   CTX_INIT(bind_depth_stencil_alpha_state);
+   CTX_INIT(delete_depth_stencil_alpha_state);
+   CTX_INIT(create_fs_state);
+   CTX_INIT(bind_fs_state);
+   CTX_INIT(delete_fs_state);
+   CTX_INIT(create_vs_state);
+   CTX_INIT(bind_vs_state);
+   CTX_INIT(delete_vs_state);
+   CTX_INIT(create_gs_state);
+   CTX_INIT(bind_gs_state);
+   CTX_INIT(delete_gs_state);
+   CTX_INIT(create_tcs_state);
+   CTX_INIT(bind_tcs_state);
+   CTX_INIT(delete_tcs_state);
+   CTX_INIT(create_tes_state);
+   CTX_INIT(bind_tes_state);
+   CTX_INIT(delete_tes_state);
+   CTX_INIT(create_vertex_elements_state);
+   CTX_INIT(bind_vertex_elements_state);
+   CTX_INIT(delete_vertex_elements_state);
+   CTX_INIT(set_blend_color);
+   CTX_INIT(set_stencil_ref);
+   CTX_INIT(set_sample_mask);
+   CTX_INIT(set_min_samples);
+   CTX_INIT(set_clip_state);
+   CTX_INIT(set_constant_buffer);
+   CTX_INIT(set_framebuffer_state);
+   CTX_INIT(set_polygon_stipple);
+   CTX_INIT(set_scissor_states);
+   CTX_INIT(set_viewport_states);
+   CTX_INIT(set_sampler_views);
+   CTX_INIT(set_tess_state);
+   CTX_INIT(set_shader_buffers);
+   CTX_INIT(set_shader_images);
+   CTX_INIT(set_vertex_buffers);
+   CTX_INIT(set_index_buffer);
+   CTX_INIT(create_stream_output_target);
+   CTX_INIT(stream_output_target_destroy);
+   CTX_INIT(set_stream_output_targets);
+   CTX_INIT(create_sampler_view);
+   CTX_INIT(sampler_view_destroy);
+   CTX_INIT(create_surface);
+   CTX_INIT(surface_destroy);
+   CTX_INIT(create_image_view);
+   CTX_INIT(image_view_destroy);
+   CTX_INIT(transfer_map);
+   CTX_INIT(transfer_flush_region);
+   CTX_INIT(transfer_unmap);
+   CTX_INIT(transfer_inline_write);
+   CTX_INIT(texture_barrier);
+   CTX_INIT(memory_barrier);
+   /* create_video_codec */
+   /* create_video_buffer */
+   /* create_compute_state */
+   /* bind_compute_state */
+   /* delete_compute_state */
+   /* set_compute_resources */
+   /* set_global_binding */
+   CTX_INIT(get_sample_position);
+   CTX_INIT(invalidate_resource);
+   CTX_INIT(get_device_reset_status);
+   CTX_INIT(dump_debug_state);
+
+   dd_init_draw_functions(dctx);
+
+   dctx->sample_mask = ~0;
+   return &dctx->base;
+}

src/gallium/drivers/ddebug/dd_draw.c

@@ -0,0 +1,807 @@
+/**************************************************************************
+ *
+ * Copyright 2015 Advanced Micro Devices, Inc.
+ * Copyright 2008 VMware, Inc.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * on the rights to use, copy, modify, merge, publish, distribute, sub
+ * license, and/or sell copies of the Software, and to permit persons to whom
+ * the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHOR(S) AND/OR THEIR SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ **************************************************************************/
+
+#include "dd_pipe.h"
+
+#include "util/u_dump.h"
+#include "util/u_format.h"
+#include "tgsi/tgsi_scan.h"
+#include "os/os_process.h"
+#include <errno.h>
+#include <sys/stat.h>
+
+
+enum call_type
+{
+   CALL_DRAW_VBO,
+   CALL_RESOURCE_COPY_REGION,
+   CALL_BLIT,
+   CALL_FLUSH_RESOURCE,
+   CALL_CLEAR,
+   CALL_CLEAR_BUFFER,
+   CALL_CLEAR_RENDER_TARGET,
+   CALL_CLEAR_DEPTH_STENCIL,
+};
+
+struct call_resource_copy_region
+{
+   struct pipe_resource *dst;
+   unsigned dst_level;
+   unsigned dstx, dsty, dstz;
+   struct pipe_resource *src;
+   unsigned src_level;
+   const struct pipe_box *src_box;
+};
+
+struct call_clear
+{
+   unsigned buffers;
+   const union pipe_color_union *color;
+   double depth;
+   unsigned stencil;
+};
+
+struct call_clear_buffer
+{
+   struct pipe_resource *res;
+   unsigned offset;
+   unsigned size;
+   const void *clear_value;
+   int clear_value_size;
+};
+
+struct dd_call
+{
+   enum call_type type;
+
+   union {
+      struct pipe_draw_info draw_vbo;
+      struct call_resource_copy_region resource_copy_region;
+      struct pipe_blit_info blit;
+      struct pipe_resource *flush_resource;
+      struct call_clear clear;
+      struct call_clear_buffer clear_buffer;
+   } info;
+};
+
+
+static FILE *
+dd_get_file_stream(struct dd_context *dctx)
+{
+   struct pipe_screen *screen = dctx->pipe->screen;
+   static unsigned index;
+   char proc_name[128], dir[256], name[512];
+   FILE *f;
+
+   if (!os_get_process_name(proc_name, sizeof(proc_name))) {
+      fprintf(stderr, "dd: can't get the process name\n");
+      return NULL;
+   }
+
+   snprintf(dir, sizeof(dir), "%s/"DD_DIR, debug_get_option("HOME", "."));
+
+   if (mkdir(dir, 0774) && errno != EEXIST) {
+      fprintf(stderr, "dd: can't create a directory (%i)\n", errno);
+      return NULL;
+   }
+
+   snprintf(name, sizeof(name), "%s/%s_%u_%08u", dir, proc_name, getpid(), index++);
+   f = fopen(name, "w");
+   if (!f) {
+      fprintf(stderr, "dd: can't open file %s\n", name);
+      return NULL;
+   }
+
+   fprintf(f, "Driver vendor: %s\n", screen->get_vendor(screen));
+   fprintf(f, "Device vendor: %s\n", screen->get_device_vendor(screen));
+   fprintf(f, "Device name: %s\n\n", screen->get_name(screen));
+   return f;
+}
+
+static void
+dd_close_file_stream(FILE *f)
+{
+   fclose(f);
+}
+
+static unsigned
+dd_num_active_viewports(struct dd_context *dctx)
+{
+   struct tgsi_shader_info info;
+   const struct tgsi_token *tokens;
+
+   if (dctx->shaders[PIPE_SHADER_GEOMETRY])
+      tokens = dctx->shaders[PIPE_SHADER_GEOMETRY]->state.shader.tokens;
+   else if (dctx->shaders[PIPE_SHADER_TESS_EVAL])
+      tokens = dctx->shaders[PIPE_SHADER_TESS_EVAL]->state.shader.tokens;
+   else if (dctx->shaders[PIPE_SHADER_VERTEX])
+      tokens = dctx->shaders[PIPE_SHADER_VERTEX]->state.shader.tokens;
+   else
+      return 1;
+
+   tgsi_scan_shader(tokens, &info);
+   return info.writes_viewport_index ? PIPE_MAX_VIEWPORTS : 1;
+}
+
+#define COLOR_RESET	"\033[0m"
+#define COLOR_SHADER	"\033[1;32m"
+#define COLOR_STATE	"\033[1;33m"
+
+#define DUMP(name, var) do { \
+   fprintf(f, COLOR_STATE #name ": " COLOR_RESET); \
+   util_dump_##name(f, var); \
+   fprintf(f, "\n"); \
+} while(0)
+
+#define DUMP_I(name, var, i) do { \
+   fprintf(f, COLOR_STATE #name " %i: " COLOR_RESET, i); \
+   util_dump_##name(f, var); \
+   fprintf(f, "\n"); \
+} while(0)
+
+#define DUMP_M(name, var, member) do { \
+   fprintf(f, "  " #member ": "); \
+   util_dump_##name(f, (var)->member); \
+   fprintf(f, "\n"); \
+} while(0)
+
+#define DUMP_M_ADDR(name, var, member) do { \
+   fprintf(f, "  " #member ": "); \
+   util_dump_##name(f, &(var)->member); \
+   fprintf(f, "\n"); \
+} while(0)
+
+static void
+print_named_value(FILE *f, const char *name, int value)
+{
+   fprintf(f, COLOR_STATE "%s" COLOR_RESET " = %i\n", name, value);
+}
+
+static void
+print_named_xvalue(FILE *f, const char *name, int value)
+{
+   fprintf(f, COLOR_STATE "%s" COLOR_RESET " = 0x%08x\n", name, value);
+}
+
+static void
+util_dump_uint(FILE *f, unsigned i)
+{
+   fprintf(f, "%u", i);
+}
+
+static void
+util_dump_hex(FILE *f, unsigned i)
+{
+   fprintf(f, "0x%x", i);
+}
+
+static void
+util_dump_double(FILE *f, double d)
+{
+   fprintf(f, "%f", d);
+}
+
+static void
+util_dump_format(FILE *f, enum pipe_format format)
+{
+   fprintf(f, "%s", util_format_name(format));
+}
+
+static void
+util_dump_color_union(FILE *f, const union pipe_color_union *color)
+{
+   fprintf(f, "{f = {%f, %f, %f, %f}, ui = {%u, %u, %u, %u}",
+           color->f[0], color->f[1], color->f[2], color->f[3],
+           color->ui[0], color->ui[1], color->ui[2], color->ui[3]);
+}
+
+static void
+util_dump_query(FILE *f, struct dd_query *query)
+{
+   if (query->type >= PIPE_QUERY_DRIVER_SPECIFIC)
+      fprintf(f, "PIPE_QUERY_DRIVER_SPECIFIC + %i",
+              query->type - PIPE_QUERY_DRIVER_SPECIFIC);
+   else
+      fprintf(f, "%s", util_dump_query_type(query->type, false));
+}
+
+static void
+dd_dump_render_condition(struct dd_context *dctx, FILE *f)
+{
+   if (dctx->render_cond.query) {
+      fprintf(f, "render condition:\n");
+      DUMP_M(query, &dctx->render_cond, query);
+      DUMP_M(uint, &dctx->render_cond, condition);
+      DUMP_M(uint, &dctx->render_cond, mode);
+      fprintf(f, "\n");
+   }
+}
+
+static void
+dd_dump_draw_vbo(struct dd_context *dctx, struct pipe_draw_info *info, FILE *f)
+{
+   int sh, i;
+   const char *shader_str[PIPE_SHADER_TYPES];
+
+   shader_str[PIPE_SHADER_VERTEX] = "VERTEX";
+   shader_str[PIPE_SHADER_TESS_CTRL] = "TESS_CTRL";
+   shader_str[PIPE_SHADER_TESS_EVAL] = "TESS_EVAL";
+   shader_str[PIPE_SHADER_GEOMETRY] = "GEOMETRY";
+   shader_str[PIPE_SHADER_FRAGMENT] = "FRAGMENT";
+   shader_str[PIPE_SHADER_COMPUTE] = "COMPUTE";
+
+   DUMP(draw_info, info);
+   if (info->indexed) {
+      DUMP(index_buffer, &dctx->index_buffer);
+      if (dctx->index_buffer.buffer)
+         DUMP_M(resource, &dctx->index_buffer, buffer);
+   }
+   if (info->count_from_stream_output)
+      DUMP_M(stream_output_target, info,
+             count_from_stream_output);
+   if (info->indirect)
+      DUMP_M(resource, info, indirect);
+   fprintf(f, "\n");
+
+   /* TODO: dump active queries */
+
+   dd_dump_render_condition(dctx, f);
+
+   for (i = 0; i < PIPE_MAX_ATTRIBS; i++)
+      if (dctx->vertex_buffers[i].buffer ||
+          dctx->vertex_buffers[i].user_buffer) {
+         DUMP_I(vertex_buffer, &dctx->vertex_buffers[i], i);
+         if (dctx->vertex_buffers[i].buffer)
+            DUMP_M(resource, &dctx->vertex_buffers[i], buffer);
+      }
+
+   if (dctx->velems) {
+      print_named_value(f, "num vertex elements",
+                        dctx->velems->state.velems.count);
+      for (i = 0; i < dctx->velems->state.velems.count; i++) {
+         fprintf(f, "  ");
+         DUMP_I(vertex_element, &dctx->velems->state.velems.velems[i], i);
+      }
+   }
+
+   print_named_value(f, "num stream output targets", dctx->num_so_targets);
+   for (i = 0; i < dctx->num_so_targets; i++)
+      if (dctx->so_targets[i]) {
+         DUMP_I(stream_output_target, dctx->so_targets[i], i);
+         DUMP_M(resource, dctx->so_targets[i], buffer);
+         fprintf(f, "  offset = %i\n", dctx->so_offsets[i]);
+      }
+
+   fprintf(f, "\n");
+   for (sh = 0; sh < PIPE_SHADER_TYPES; sh++) {
+      if (sh == PIPE_SHADER_COMPUTE)
+         continue;
+
+      if (sh == PIPE_SHADER_TESS_CTRL &&
+          !dctx->shaders[PIPE_SHADER_TESS_CTRL] &&
+          dctx->shaders[PIPE_SHADER_TESS_EVAL])
+         fprintf(f, "tess_state: {default_outer_level = {%f, %f, %f, %f}, "
+                 "default_inner_level = {%f, %f}}\n",
+                 dctx->tess_default_levels[0],
+                 dctx->tess_default_levels[1],
+                 dctx->tess_default_levels[2],
+                 dctx->tess_default_levels[3],
+                 dctx->tess_default_levels[4],
+                 dctx->tess_default_levels[5]);
+
+      if (sh == PIPE_SHADER_FRAGMENT)
+         if (dctx->rs) {
+            unsigned num_viewports = dd_num_active_viewports(dctx);
+
+            if (dctx->rs->state.rs.clip_plane_enable)
+               DUMP(clip_state, &dctx->clip_state);
+
+            for (i = 0; i < num_viewports; i++)
+               DUMP_I(viewport_state, &dctx->viewports[i], i);
+
+            if (dctx->rs->state.rs.scissor)
+               for (i = 0; i < num_viewports; i++)
+                  DUMP_I(scissor_state, &dctx->scissors[i], i);
+
+            DUMP(rasterizer_state, &dctx->rs->state.rs);
+
+            if (dctx->rs->state.rs.poly_stipple_enable)
+               DUMP(poly_stipple, &dctx->polygon_stipple);
+            fprintf(f, "\n");
+         }
+
+      if (!dctx->shaders[sh])
+         continue;
+
+      fprintf(f, COLOR_SHADER "begin shader: %s" COLOR_RESET "\n", shader_str[sh]);
+      DUMP(shader_state, &dctx->shaders[sh]->state.shader);
+
+      for (i = 0; i < PIPE_MAX_CONSTANT_BUFFERS; i++)
+         if (dctx->constant_buffers[sh][i].buffer ||
+             dctx->constant_buffers[sh][i].user_buffer) {
+            DUMP_I(constant_buffer, &dctx->constant_buffers[sh][i], i);
+            if (dctx->constant_buffers[sh][i].buffer)
+               DUMP_M(resource, &dctx->constant_buffers[sh][i], buffer);
+         }
+
+      for (i = 0; i < PIPE_MAX_SAMPLERS; i++)
+         if (dctx->sampler_states[sh][i])
+            DUMP_I(sampler_state, &dctx->sampler_states[sh][i]->state.sampler, i);
+
+      for (i = 0; i < PIPE_MAX_SAMPLERS; i++)
+         if (dctx->sampler_views[sh][i]) {
+            DUMP_I(sampler_view, dctx->sampler_views[sh][i], i);
+            DUMP_M(resource, dctx->sampler_views[sh][i], texture);
+         }
+
+      /* TODO: print shader images */
+      /* TODO: print shader buffers */
+
+      fprintf(f, COLOR_SHADER "end shader: %s" COLOR_RESET "\n\n", shader_str[sh]);
+   }
+
+   if (dctx->dsa)
+      DUMP(depth_stencil_alpha_state, &dctx->dsa->state.dsa);
+   DUMP(stencil_ref, &dctx->stencil_ref);
+
+   if (dctx->blend)
+      DUMP(blend_state, &dctx->blend->state.blend);
+   DUMP(blend_color, &dctx->blend_color);
+
+   print_named_value(f, "min_samples", dctx->min_samples);
+   print_named_xvalue(f, "sample_mask", dctx->sample_mask);
+   fprintf(f, "\n");
+
+   DUMP(framebuffer_state, &dctx->framebuffer_state);
+   for (i = 0; i < dctx->framebuffer_state.nr_cbufs; i++)
+      if (dctx->framebuffer_state.cbufs[i]) {
+         fprintf(f, "  " COLOR_STATE "cbufs[%i]:" COLOR_RESET "\n    ", i);
+         DUMP(surface, dctx->framebuffer_state.cbufs[i]);
+         fprintf(f, "    ");
+         DUMP(resource, dctx->framebuffer_state.cbufs[i]->texture);
+      }
+   if (dctx->framebuffer_state.zsbuf) {
+      fprintf(f, "  " COLOR_STATE "zsbuf:" COLOR_RESET "\n    ");
+      DUMP(surface, dctx->framebuffer_state.zsbuf);
+      fprintf(f, "    ");
+      DUMP(resource, dctx->framebuffer_state.zsbuf->texture);
+   }
+   fprintf(f, "\n");
+}
+
+static void
+dd_dump_resource_copy_region(struct dd_context *dctx,
+                             struct call_resource_copy_region *info,
+                             FILE *f)
+{
+   fprintf(f, "%s:\n", __func__+8);
+   DUMP_M(resource, info, dst);
+   DUMP_M(uint, info, dst_level);
+   DUMP_M(uint, info, dstx);
+   DUMP_M(uint, info, dsty);
+   DUMP_M(uint, info, dstz);
+   DUMP_M(resource, info, src);
+   DUMP_M(uint, info, src_level);
+   DUMP_M(box, info, src_box);
+}
+
+static void
+dd_dump_blit(struct dd_context *dctx, struct pipe_blit_info *info, FILE *f)
+{
+   fprintf(f, "%s:\n", __func__+8);
+   DUMP_M(resource, info, dst.resource);
+   DUMP_M(uint, info, dst.level);
+   DUMP_M_ADDR(box, info, dst.box);
+   DUMP_M(format, info, dst.format);
+
+   DUMP_M(resource, info, src.resource);
+   DUMP_M(uint, info, src.level);
+   DUMP_M_ADDR(box, info, src.box);
+   DUMP_M(format, info, src.format);
+
+   DUMP_M(hex, info, mask);
+   DUMP_M(uint, info, filter);
+   DUMP_M(uint, info, scissor_enable);
+   DUMP_M_ADDR(scissor_state, info, scissor);
+   DUMP_M(uint, info, render_condition_enable);
+
+   if (info->render_condition_enable)
+      dd_dump_render_condition(dctx, f);
+}
+
+static void
+dd_dump_flush_resource(struct dd_context *dctx, struct pipe_resource *res,
+                       FILE *f)
+{
+   fprintf(f, "%s:\n", __func__+8);
+   DUMP(resource, res);
+}
+
+static void
+dd_dump_clear(struct dd_context *dctx, struct call_clear *info, FILE *f)
+{
+   fprintf(f, "%s:\n", __func__+8);
+   DUMP_M(uint, info, buffers);
+   DUMP_M(color_union, info, color);
+   DUMP_M(double, info, depth);
+   DUMP_M(hex, info, stencil);
+}
+
+static void
+dd_dump_clear_buffer(struct dd_context *dctx, struct call_clear_buffer *info,
+                     FILE *f)
+{
+   int i;
+   const char *value = (const char*)info->clear_value;
+
+   fprintf(f, "%s:\n", __func__+8);
+   DUMP_M(resource, info, res);
+   DUMP_M(uint, info, offset);
+   DUMP_M(uint, info, size);
+   DUMP_M(uint, info, clear_value_size);
+
+   fprintf(f, "  clear_value:");
+   for (i = 0; i < info->clear_value_size; i++)
+      fprintf(f, " %02x", value[i]);
+   fprintf(f, "\n");
+}
+
+static void
+dd_dump_clear_render_target(struct dd_context *dctx, FILE *f)
+{
+   fprintf(f, "%s:\n", __func__+8);
+   /* TODO */
+}
+
+static void
+dd_dump_clear_depth_stencil(struct dd_context *dctx, FILE *f)
+{
+   fprintf(f, "%s:\n", __func__+8);
+   /* TODO */
+}
+
+static void
+dd_dump_driver_state(struct dd_context *dctx, FILE *f, unsigned flags)
+{
+   if (dctx->pipe->dump_debug_state) {
+	   fprintf(f,"\n\n**************************************************"
+		     "***************************\n");
+	   fprintf(f, "Driver-specific state:\n\n");
+	   dctx->pipe->dump_debug_state(dctx->pipe, f, flags);
+   }
+}
+
+static void
+dd_dump_call(struct dd_context *dctx, struct dd_call *call, unsigned flags)
+{
+   FILE *f = dd_get_file_stream(dctx);
+
+   if (!f)
+      return;
+
+   switch (call->type) {
+   case CALL_DRAW_VBO:
+      dd_dump_draw_vbo(dctx, &call->info.draw_vbo, f);
+      break;
+   case CALL_RESOURCE_COPY_REGION:
+      dd_dump_resource_copy_region(dctx, &call->info.resource_copy_region, f);
+      break;
+   case CALL_BLIT:
+      dd_dump_blit(dctx, &call->info.blit, f);
+      break;
+   case CALL_FLUSH_RESOURCE:
+      dd_dump_flush_resource(dctx, call->info.flush_resource, f);
+      break;
+   case CALL_CLEAR:
+      dd_dump_clear(dctx, &call->info.clear, f);
+      break;
+   case CALL_CLEAR_BUFFER:
+      dd_dump_clear_buffer(dctx, &call->info.clear_buffer, f);
+      break;
+   case CALL_CLEAR_RENDER_TARGET:
+      dd_dump_clear_render_target(dctx, f);
+      break;
+   case CALL_CLEAR_DEPTH_STENCIL:
+      dd_dump_clear_depth_stencil(dctx, f);
+   }
+
+   dd_dump_driver_state(dctx, f, flags);
+   dd_close_file_stream(f);
+}
+
+static void
+dd_kill_process(void)
+{
+   sync();
+   fprintf(stderr, "dd: Aborting the process...\n");
+   fflush(stdout);
+   fflush(stderr);
+   abort();
+}
+
+static bool
+dd_flush_and_check_hang(struct dd_context *dctx,
+                        struct pipe_fence_handle **flush_fence,
+                        unsigned flush_flags)
+{
+   struct pipe_fence_handle *fence = NULL;
+   struct pipe_context *pipe = dctx->pipe;
+   struct pipe_screen *screen = pipe->screen;
+   uint64_t timeout_ms = dd_screen(dctx->base.screen)->timeout_ms;
+   bool idle;
+
+   assert(timeout_ms > 0);
+
+   pipe->flush(pipe, &fence, flush_flags);
+   if (flush_fence)
+      screen->fence_reference(screen, flush_fence, fence);
+   if (!fence)
+      return false;
+
+   idle = screen->fence_finish(screen, fence, timeout_ms * 1000000);
+   screen->fence_reference(screen, &fence, NULL);
+   if (!idle)
+      fprintf(stderr, "dd: GPU hang detected!\n");
+   return !idle;
+}
+
+static void
+dd_flush_and_handle_hang(struct dd_context *dctx,
+                         struct pipe_fence_handle **fence, unsigned flags,
+                         const char *cause)
+{
+   if (dd_flush_and_check_hang(dctx, fence, flags)) {
+      FILE *f = dd_get_file_stream(dctx);
+
+      if (f) {
+         fprintf(f, "dd: %s.\n", cause);
+         dd_dump_driver_state(dctx, f, PIPE_DEBUG_DEVICE_IS_HUNG);
+         dd_close_file_stream(f);
+      }
+
+      /* Terminate the process to prevent future hangs. */
+      dd_kill_process();
+   }
+}
+
+static void
+dd_context_flush(struct pipe_context *_pipe,
+                 struct pipe_fence_handle **fence, unsigned flags)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+
+   switch (dd_screen(dctx->base.screen)->mode) {
+   case DD_DETECT_HANGS:
+      dd_flush_and_handle_hang(dctx, fence, flags,
+                               "GPU hang detected in pipe->flush()");
+      break;
+   case DD_DUMP_ALL_CALLS:
+      pipe->flush(pipe, fence, flags);
+      break;
+   default:
+      assert(0);
+   }
+}
+
+static void
+dd_before_draw(struct dd_context *dctx)
+{
+   if (dd_screen(dctx->base.screen)->mode == DD_DETECT_HANGS &&
+       !dd_screen(dctx->base.screen)->no_flush)
+      dd_flush_and_handle_hang(dctx, NULL, 0,
+                               "GPU hang most likely caused by internal "
+                               "driver commands");
+}
+
+static void
+dd_after_draw(struct dd_context *dctx, struct dd_call *call)
+{
+   switch (dd_screen(dctx->base.screen)->mode) {
+   case DD_DETECT_HANGS:
+      if (!dd_screen(dctx->base.screen)->no_flush &&
+          dd_flush_and_check_hang(dctx, NULL, 0)) {
+         dd_dump_call(dctx, call, PIPE_DEBUG_DEVICE_IS_HUNG);
+
+         /* Terminate the process to prevent future hangs. */
+         dd_kill_process();
+      }
+      break;
+   case DD_DUMP_ALL_CALLS:
+      dd_dump_call(dctx, call, 0);
+      break;
+   default:
+      assert(0);
+   }
+}
+
+static void
+dd_context_draw_vbo(struct pipe_context *_pipe,
+                    const struct pipe_draw_info *info)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_DRAW_VBO;
+   call.info.draw_vbo = *info;
+
+   dd_before_draw(dctx);
+   pipe->draw_vbo(pipe, info);
+   dd_after_draw(dctx, &call);
+}
+
+static void
+dd_context_resource_copy_region(struct pipe_context *_pipe,
+                                struct pipe_resource *dst, unsigned dst_level,
+                                unsigned dstx, unsigned dsty, unsigned dstz,
+                                struct pipe_resource *src, unsigned src_level,
+                                const struct pipe_box *src_box)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_RESOURCE_COPY_REGION;
+   call.info.resource_copy_region.dst = dst;
+   call.info.resource_copy_region.dst_level = dst_level;
+   call.info.resource_copy_region.dstx = dstx;
+   call.info.resource_copy_region.dsty = dsty;
+   call.info.resource_copy_region.dstz = dstz;
+   call.info.resource_copy_region.src = src;
+   call.info.resource_copy_region.src_level = src_level;
+   call.info.resource_copy_region.src_box = src_box;
+
+   dd_before_draw(dctx);
+   pipe->resource_copy_region(pipe,
+                              dst, dst_level, dstx, dsty, dstz,
+                              src, src_level, src_box);
+   dd_after_draw(dctx, &call);
+}
+
+static void
+dd_context_blit(struct pipe_context *_pipe, const struct pipe_blit_info *info)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_BLIT;
+   call.info.blit = *info;
+
+   dd_before_draw(dctx);
+   pipe->blit(pipe, info);
+   dd_after_draw(dctx, &call);
+}
+
+static void
+dd_context_flush_resource(struct pipe_context *_pipe,
+                          struct pipe_resource *resource)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_FLUSH_RESOURCE;
+   call.info.flush_resource = resource;
+
+   dd_before_draw(dctx);
+   pipe->flush_resource(pipe, resource);
+   dd_after_draw(dctx, &call);
+}
+
+static void
+dd_context_clear(struct pipe_context *_pipe, unsigned buffers,
+                 const union pipe_color_union *color, double depth,
+                 unsigned stencil)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_CLEAR;
+   call.info.clear.buffers = buffers;
+   call.info.clear.color = color;
+   call.info.clear.depth = depth;
+   call.info.clear.stencil = stencil;
+
+   dd_before_draw(dctx);
+   pipe->clear(pipe, buffers, color, depth, stencil);
+   dd_after_draw(dctx, &call);
+}
+
+static void
+dd_context_clear_render_target(struct pipe_context *_pipe,
+                               struct pipe_surface *dst,
+                               const union pipe_color_union *color,
+                               unsigned dstx, unsigned dsty,
+                               unsigned width, unsigned height)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_CLEAR_RENDER_TARGET;
+
+   dd_before_draw(dctx);
+   pipe->clear_render_target(pipe, dst, color, dstx, dsty, width, height);
+   dd_after_draw(dctx, &call);
+}
+
+static void
+dd_context_clear_depth_stencil(struct pipe_context *_pipe,
+                               struct pipe_surface *dst, unsigned clear_flags,
+                               double depth, unsigned stencil, unsigned dstx,
+                               unsigned dsty, unsigned width, unsigned height)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_CLEAR_DEPTH_STENCIL;
+
+   dd_before_draw(dctx);
+   pipe->clear_depth_stencil(pipe, dst, clear_flags, depth, stencil,
+                             dstx, dsty, width, height);
+   dd_after_draw(dctx, &call);
+}
+
+static void
+dd_context_clear_buffer(struct pipe_context *_pipe, struct pipe_resource *res,
+                        unsigned offset, unsigned size,
+                        const void *clear_value, int clear_value_size)
+{
+   struct dd_context *dctx = dd_context(_pipe);
+   struct pipe_context *pipe = dctx->pipe;
+   struct dd_call call;
+
+   call.type = CALL_CLEAR_BUFFER;
+   call.info.clear_buffer.res = res;
+   call.info.clear_buffer.offset = offset;
+   call.info.clear_buffer.size = size;
+   call.info.clear_buffer.clear_value = clear_value;
+   call.info.clear_buffer.clear_value_size = clear_value_size;
+
+   dd_before_draw(dctx);
+   pipe->clear_buffer(pipe, res, offset, size, clear_value, clear_value_size);
+   dd_after_draw(dctx, &call);
+}
+
+void
+dd_init_draw_functions(struct dd_context *dctx)
+{
+   CTX_INIT(flush);
+   CTX_INIT(draw_vbo);
+   CTX_INIT(resource_copy_region);
+   CTX_INIT(blit);
+   CTX_INIT(clear);
+   CTX_INIT(clear_render_target);
+   CTX_INIT(clear_depth_stencil);
+   CTX_INIT(clear_buffer);
+   CTX_INIT(flush_resource);
+   /* launch_grid */
+}

src/gallium/drivers/ddebug/dd_pipe.h

@@ -0,0 +1,141 @@
+/**************************************************************************
+ *
+ * Copyright 2015 Advanced Micro Devices, Inc.
+ * Copyright 2008 VMware, Inc.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * on the rights to use, copy, modify, merge, publish, distribute, sub
+ * license, and/or sell copies of the Software, and to permit persons to whom
+ * the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHOR(S) AND/OR THEIR SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ **************************************************************************/
+
+#ifndef DD_H_
+#define DD_H_
+
+#include "pipe/p_context.h"
+#include "pipe/p_state.h"
+#include "pipe/p_screen.h"
+
+/* name of the directory in home */
+#define DD_DIR "ddebug_dumps"
+
+enum dd_mode {
+   DD_DETECT_HANGS,
+   DD_DUMP_ALL_CALLS
+};
+
+struct dd_screen
+{
+   struct pipe_screen base;
+   struct pipe_screen *screen;
+   unsigned timeout_ms;
+   enum dd_mode mode;
+   bool no_flush;
+};
+
+struct dd_query
+{
+   unsigned type;
+   struct pipe_query *query;
+};
+
+struct dd_state
+{
+   void *cso;
+
+   union {
+      struct pipe_blend_state blend;
+      struct pipe_depth_stencil_alpha_state dsa;
+      struct pipe_rasterizer_state rs;
+      struct pipe_sampler_state sampler;
+      struct {
+         struct pipe_vertex_element velems[PIPE_MAX_ATTRIBS];
+         unsigned count;
+      } velems;
+      struct pipe_shader_state shader;
+   } state;
+};
+
+struct dd_context
+{
+   struct pipe_context base;
+   struct pipe_context *pipe;
+
+   struct {
+      struct dd_query *query;
+      bool condition;
+      unsigned mode;
+   } render_cond;
+
+   struct pipe_index_buffer index_buffer;
+   struct pipe_vertex_buffer vertex_buffers[PIPE_MAX_ATTRIBS];
+
+   unsigned num_so_targets;
+   struct pipe_stream_output_target *so_targets[PIPE_MAX_SO_BUFFERS];
+   unsigned so_offsets[PIPE_MAX_SO_BUFFERS];
+
+   struct dd_state *shaders[PIPE_SHADER_TYPES];
+   struct pipe_constant_buffer constant_buffers[PIPE_SHADER_TYPES][PIPE_MAX_CONSTANT_BUFFERS];
+   struct pipe_sampler_view *sampler_views[PIPE_SHADER_TYPES][PIPE_MAX_SAMPLERS];
+   struct dd_state *sampler_states[PIPE_SHADER_TYPES][PIPE_MAX_SAMPLERS];
+   struct pipe_image_view *shader_images[PIPE_SHADER_TYPES][PIPE_MAX_SHADER_IMAGES];
+   struct pipe_shader_buffer shader_buffers[PIPE_SHADER_TYPES][PIPE_MAX_SHADER_BUFFERS];
+
+   struct dd_state *velems;
+   struct dd_state *rs;
+   struct dd_state *dsa;
+   struct dd_state *blend;
+
+   struct pipe_blend_color blend_color;
+   struct pipe_stencil_ref stencil_ref;
+   unsigned sample_mask;
+   unsigned min_samples;
+   struct pipe_clip_state clip_state;
+   struct pipe_framebuffer_state framebuffer_state;
+   struct pipe_poly_stipple polygon_stipple;
+   struct pipe_scissor_state scissors[PIPE_MAX_VIEWPORTS];
+   struct pipe_viewport_state viewports[PIPE_MAX_VIEWPORTS];
+   float tess_default_levels[6];
+};
+
+
+struct pipe_context *
+dd_context_create(struct dd_screen *dscreen, struct pipe_context *pipe);
+
+void
+dd_init_draw_functions(struct dd_context *dctx);
+
+
+static inline struct dd_context *
+dd_context(struct pipe_context *pipe)
+{
+   return (struct dd_context *)pipe;
+}
+
+static inline struct dd_screen *
+dd_screen(struct pipe_screen *screen)
+{
+   return (struct dd_screen*)screen;
+}
+
+
+#define CTX_INIT(_member) \
+   dctx->base._member = dctx->pipe->_member ? dd_context_##_member : NULL
+
+#endif /* DD_H_ */

src/gallium/drivers/ddebug/dd_public.h

@@ -0,0 +1,36 @@
+/**************************************************************************
+ *
+ * Copyright 2015 Advanced Micro Devices, Inc.
+ * Copyright 2010 VMware, Inc.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * on the rights to use, copy, modify, merge, publish, distribute, sub
+ * license, and/or sell copies of the Software, and to permit persons to whom
+ * the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHOR(S) AND/OR THEIR SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ **************************************************************************/
+
+#ifndef DD_PUBLIC_H_
+#define DD_PUBLIC_H_
+
+struct pipe_screen;
+
+struct pipe_screen *
+ddebug_screen_create(struct pipe_screen *screen);
+
+#endif /* DD_PUBLIC_H_ */

src/gallium/drivers/ddebug/dd_screen.c

@@ -0,0 +1,353 @@
+/**************************************************************************
+ *
+ * Copyright 2015 Advanced Micro Devices, Inc.
+ * Copyright 2008 VMware, Inc.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * on the rights to use, copy, modify, merge, publish, distribute, sub
+ * license, and/or sell copies of the Software, and to permit persons to whom
+ * the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHOR(S) AND/OR THEIR SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ **************************************************************************/
+
+#include "dd_pipe.h"
+#include "dd_public.h"
+#include "util/u_memory.h"
+#include <stdio.h>
+
+
+static const char *
+dd_screen_get_name(struct pipe_screen *_screen)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_name(screen);
+}
+
+static const char *
+dd_screen_get_vendor(struct pipe_screen *_screen)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_vendor(screen);
+}
+
+static const char *
+dd_screen_get_device_vendor(struct pipe_screen *_screen)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_device_vendor(screen);
+}
+
+static int
+dd_screen_get_param(struct pipe_screen *_screen,
+                    enum pipe_cap param)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_param(screen, param);
+}
+
+static float
+dd_screen_get_paramf(struct pipe_screen *_screen,
+                     enum pipe_capf param)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_paramf(screen, param);
+}
+
+static int
+dd_screen_get_shader_param(struct pipe_screen *_screen, unsigned shader,
+                           enum pipe_shader_cap param)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_shader_param(screen, shader, param);
+}
+
+static uint64_t
+dd_screen_get_timestamp(struct pipe_screen *_screen)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_timestamp(screen);
+}
+
+static struct pipe_context *
+dd_screen_context_create(struct pipe_screen *_screen, void *priv,
+                         unsigned flags)
+{
+   struct dd_screen *dscreen = dd_screen(_screen);
+   struct pipe_screen *screen = dscreen->screen;
+
+   flags |= PIPE_CONTEXT_DEBUG;
+
+   return dd_context_create(dscreen,
+                            screen->context_create(screen, priv, flags));
+}
+
+static boolean
+dd_screen_is_format_supported(struct pipe_screen *_screen,
+                              enum pipe_format format,
+                              enum pipe_texture_target target,
+                              unsigned sample_count,
+                              unsigned tex_usage)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->is_format_supported(screen, format, target, sample_count,
+                                      tex_usage);
+}
+
+static boolean
+dd_screen_can_create_resource(struct pipe_screen *_screen,
+                              const struct pipe_resource *templat)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->can_create_resource(screen, templat);
+}
+
+static void
+dd_screen_flush_frontbuffer(struct pipe_screen *_screen,
+                            struct pipe_resource *resource,
+                            unsigned level, unsigned layer,
+                            void *context_private,
+                            struct pipe_box *sub_box)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   screen->flush_frontbuffer(screen, resource, level, layer, context_private,
+                             sub_box);
+}
+
+static int
+dd_screen_get_driver_query_info(struct pipe_screen *_screen,
+                                unsigned index,
+                                struct pipe_driver_query_info *info)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_driver_query_info(screen, index, info);
+}
+
+static int
+dd_screen_get_driver_query_group_info(struct pipe_screen *_screen,
+                                      unsigned index,
+                                      struct pipe_driver_query_group_info *info)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->get_driver_query_group_info(screen, index, info);
+}
+
+
+/********************************************************************
+ * resource
+ */
+
+static struct pipe_resource *
+dd_screen_resource_create(struct pipe_screen *_screen,
+                          const struct pipe_resource *templat)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+   struct pipe_resource *res = screen->resource_create(screen, templat);
+
+   if (!res)
+      return NULL;
+   res->screen = _screen;
+   return res;
+}
+
+static struct pipe_resource *
+dd_screen_resource_from_handle(struct pipe_screen *_screen,
+                               const struct pipe_resource *templ,
+                               struct winsys_handle *handle)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+   struct pipe_resource *res =
+      screen->resource_from_handle(screen, templ, handle);
+
+   if (!res)
+      return NULL;
+   res->screen = _screen;
+   return res;
+}
+
+static struct pipe_resource *
+dd_screen_resource_from_user_memory(struct pipe_screen *_screen,
+                                    const struct pipe_resource *templ,
+                                    void *user_memory)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+   struct pipe_resource *res =
+      screen->resource_from_user_memory(screen, templ, user_memory);
+
+   if (!res)
+      return NULL;
+   res->screen = _screen;
+   return res;
+}
+
+static void
+dd_screen_resource_destroy(struct pipe_screen *_screen,
+                           struct pipe_resource *res)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   screen->resource_destroy(screen, res);
+}
+
+static boolean
+dd_screen_resource_get_handle(struct pipe_screen *_screen,
+                              struct pipe_resource *resource,
+                              struct winsys_handle *handle)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->resource_get_handle(screen, resource, handle);
+}
+
+
+/********************************************************************
+ * fence
+ */
+
+static void
+dd_screen_fence_reference(struct pipe_screen *_screen,
+                          struct pipe_fence_handle **pdst,
+                          struct pipe_fence_handle *src)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   screen->fence_reference(screen, pdst, src);
+}
+
+static boolean
+dd_screen_fence_finish(struct pipe_screen *_screen,
+                       struct pipe_fence_handle *fence,
+                       uint64_t timeout)
+{
+   struct pipe_screen *screen = dd_screen(_screen)->screen;
+
+   return screen->fence_finish(screen, fence, timeout);
+}
+
+
+/********************************************************************
+ * screen
+ */
+
+static void
+dd_screen_destroy(struct pipe_screen *_screen)
+{
+   struct dd_screen *dscreen = dd_screen(_screen);
+   struct pipe_screen *screen = dscreen->screen;
+
+   screen->destroy(screen);
+   FREE(dscreen);
+}
+
+struct pipe_screen *
+ddebug_screen_create(struct pipe_screen *screen)
+{
+   struct dd_screen *dscreen;
+   const char *option = debug_get_option("GALLIUM_DDEBUG", NULL);
+   bool dump_always = option && !strcmp(option, "always");
+   bool no_flush = option && strstr(option, "noflush");
+   bool help = option && !strcmp(option, "help");
+   unsigned timeout = 0;
+
+   if (help) {
+      puts("Gallium driver debugger");
+      puts("");
+      puts("Usage:");
+      puts("");
+      puts("  GALLIUM_DDEBUG=always");
+      puts("    Dump context and driver information after every draw call into");
+      puts("    $HOME/"DD_DIR"/.");
+      puts("");
+      puts("  GALLIUM_DDEBUG=[timeout in ms] noflush");
+      puts("    Flush and detect a device hang after every draw call based on the given");
+      puts("    fence timeout and dump context and driver information into");
+      puts("    $HOME/"DD_DIR"/ when a hang is detected.");
+      puts("    If 'noflush' is specified, only detect hangs in pipe->flush.");
+      puts("");
+      exit(0);
+   }
+
+   if (!option)
+      return screen;
+   if (!dump_always && sscanf(option, "%u", &timeout) != 1)
+      return screen;
+
+   dscreen = CALLOC_STRUCT(dd_screen);
+   if (!dscreen)
+      return NULL;
+
+#define SCR_INIT(_member) \
+   dscreen->base._member = screen->_member ? dd_screen_##_member : NULL
+
+   dscreen->base.destroy = dd_screen_destroy;
+   dscreen->base.get_name = dd_screen_get_name;
+   dscreen->base.get_vendor = dd_screen_get_vendor;
+   dscreen->base.get_device_vendor = dd_screen_get_device_vendor;
+   dscreen->base.get_param = dd_screen_get_param;
+   dscreen->base.get_paramf = dd_screen_get_paramf;
+   dscreen->base.get_shader_param = dd_screen_get_shader_param;
+   /* get_video_param */
+   /* get_compute_param */
+   SCR_INIT(get_timestamp);
+   dscreen->base.context_create = dd_screen_context_create;
+   dscreen->base.is_format_supported = dd_screen_is_format_supported;
+   /* is_video_format_supported */
+   SCR_INIT(can_create_resource);
+   dscreen->base.resource_create = dd_screen_resource_create;
+   dscreen->base.resource_from_handle = dd_screen_resource_from_handle;
+   SCR_INIT(resource_from_user_memory);
+   dscreen->base.resource_get_handle = dd_screen_resource_get_handle;
+   dscreen->base.resource_destroy = dd_screen_resource_destroy;
+   SCR_INIT(flush_frontbuffer);
+   SCR_INIT(fence_reference);
+   SCR_INIT(fence_finish);
+   SCR_INIT(get_driver_query_info);
+   SCR_INIT(get_driver_query_group_info);
+
+#undef SCR_INIT
+
+   dscreen->screen = screen;
+   dscreen->timeout_ms = timeout;
+   dscreen->mode = dump_always ? DD_DUMP_ALL_CALLS : DD_DETECT_HANGS;
+   dscreen->no_flush = no_flush;
+
+   switch (dscreen->mode) {
+   case DD_DUMP_ALL_CALLS:
+      fprintf(stderr, "Gallium debugger active. Logging all calls.\n");
+      break;
+   case DD_DETECT_HANGS:
+      fprintf(stderr, "Gallium debugger active. "
+              "The hang detection timout is %i ms.\n", timeout);
+      break;
+   default:
+      assert(0);
+   }
+
+   return &dscreen->base;
+}

src/gallium/drivers/freedreno/a2xx/fd2_context.c

@@ -86,7 +86,7 @@ static const uint8_t a20x_primtypes[PIPE_PRIM_MAX] = {
 };
 
 struct pipe_context *
-fd2_context_create(struct pipe_screen *pscreen, void *priv)
+fd2_context_create(struct pipe_screen *pscreen, void *priv, unsigned flags)
 {
 	struct fd_screen *screen = fd_screen(pscreen);
 	struct fd2_context *fd2_ctx = CALLOC_STRUCT(fd2_context);

src/gallium/drivers/freedreno/a2xx/fd2_context.h

@@ -47,6 +47,6 @@ fd2_context(struct fd_context *ctx)
 }
 
 struct pipe_context *
-fd2_context_create(struct pipe_screen *pscreen, void *priv);
+fd2_context_create(struct pipe_screen *pscreen, void *priv, unsigned flags);
 
 #endif /* FD2_CONTEXT_H_ */

src/gallium/drivers/freedreno/a3xx/a3xx.xml.h

@@ -680,6 +680,7 @@ static inline uint32_t REG_A3XX_CP_PROTECT_REG(uint32_t i0) { return 0x00000460
 #define A3XX_GRAS_CL_CLIP_CNTL_VP_CLIP_CODE_IGNORE		0x00080000
 #define A3XX_GRAS_CL_CLIP_CNTL_VP_XFORM_DISABLE			0x00100000
 #define A3XX_GRAS_CL_CLIP_CNTL_PERSP_DIVISION_DISABLE		0x00200000
+#define A3XX_GRAS_CL_CLIP_CNTL_ZERO_GB_SCALE_Z			0x00400000
 #define A3XX_GRAS_CL_CLIP_CNTL_ZCOORD				0x00800000
 #define A3XX_GRAS_CL_CLIP_CNTL_WCOORD				0x01000000
 #define A3XX_GRAS_CL_CLIP_CNTL_ZCLIP_DISABLE			0x02000000

src/gallium/drivers/freedreno/a3xx/fd3_context.c

@@ -98,7 +98,7 @@ static const uint8_t primtypes[PIPE_PRIM_MAX] = {
 };
 
 struct pipe_context *
-fd3_context_create(struct pipe_screen *pscreen, void *priv)
+fd3_context_create(struct pipe_screen *pscreen, void *priv, unsigned flags)
 {
 	struct fd_screen *screen = fd_screen(pscreen);
 	struct fd3_context *fd3_ctx = CALLOC_STRUCT(fd3_context);

src/gallium/drivers/freedreno/a3xx/fd3_context.h

@@ -119,6 +119,6 @@ fd3_context(struct fd_context *ctx)
 }
 
 struct pipe_context *
-fd3_context_create(struct pipe_screen *pscreen, void *priv);
+fd3_context_create(struct pipe_screen *pscreen, void *priv, unsigned flags);
 
 #endif /* FD3_CONTEXT_H_ */

src/gallium/drivers/freedreno/a3xx/fd3_emit.c

@@ -828,7 +828,11 @@ fd3_emit_restore(struct fd_context *ctx)
 	OUT_RING(ring, A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_STARTENTRY(0) |
 			A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_ENDENTRY(0));
 
-	fd3_emit_cache_flush(ctx, ring);
+	OUT_PKT0(ring, REG_A3XX_UCHE_CACHE_INVALIDATE0_REG, 2);
+	OUT_RING(ring, A3XX_UCHE_CACHE_INVALIDATE0_REG_ADDR(0));
+	OUT_RING(ring, A3XX_UCHE_CACHE_INVALIDATE1_REG_ADDR(0) |
+			A3XX_UCHE_CACHE_INVALIDATE1_REG_OPCODE(INVALIDATE) |
+			A3XX_UCHE_CACHE_INVALIDATE1_REG_ENTIRE_CACHE);
 
 	OUT_PKT0(ring, REG_A3XX_GRAS_CL_CLIP_CNTL, 1);
 	OUT_RING(ring, 0x00000000);                  /* GRAS_CL_CLIP_CNTL */

src/gallium/drivers/freedreno/a3xx/fd3_emit.h

@@ -90,15 +90,4 @@ void fd3_emit_restore(struct fd_context *ctx);
 
 void fd3_emit_init(struct pipe_context *pctx);
 
-static inline void
-fd3_emit_cache_flush(struct fd_context *ctx, struct fd_ringbuffer *ring)
-{
-	fd_wfi(ctx, ring);
-	OUT_PKT0(ring, REG_A3XX_UCHE_CACHE_INVALIDATE0_REG, 2);
-	OUT_RING(ring, A3XX_UCHE_CACHE_INVALIDATE0_REG_ADDR(0));
-	OUT_RING(ring, A3XX_UCHE_CACHE_INVALIDATE1_REG_ADDR(0) |
-			A3XX_UCHE_CACHE_INVALIDATE1_REG_OPCODE(INVALIDATE) |
-			A3XX_UCHE_CACHE_INVALIDATE1_REG_ENTIRE_CACHE);
-}
-
 #endif /* FD3_EMIT_H */

src/gallium/drivers/freedreno/a3xx/fd3_format.c

@@ -355,8 +355,6 @@ fd3_fs_output_format(enum pipe_format format)
 	case PIPE_FORMAT_R16G16_FLOAT:
 	case PIPE_FORMAT_R11G11B10_FLOAT:
 		return RB_R16G16B16A16_FLOAT;
-	case PIPE_FORMAT_L8_UNORM:
-		return RB_R8G8B8A8_UNORM;
 	default:
 		return fd3_pipe2color(format);
 	}

src/gallium/drivers/freedreno/a3xx/fd3_gmem.c

@@ -558,8 +558,6 @@ fd3_emit_tile_mem2gmem(struct fd_context *ctx, struct fd_tile *tile)
 	OUT_RING(ring, fui(x1));
 	OUT_RING(ring, fui(y1));
 
-	fd3_emit_cache_flush(ctx, ring);
-
 	for (i = 0; i < 4; i++) {
 		OUT_PKT0(ring, REG_A3XX_RB_MRT_CONTROL(i), 1);
 		OUT_RING(ring, A3XX_RB_MRT_CONTROL_ROP_CODE(ROP_COPY) |

src/gallium/drivers/freedreno/a3xx/fd3_rasterizer.c

@@ -65,7 +65,8 @@ fd3_rasterizer_state_create(struct pipe_context *pctx,
 	if (cso->multisample)
 		TODO
 */
-	so->gras_cl_clip_cntl = A3XX_GRAS_CL_CLIP_CNTL_IJ_PERSP_CENTER; /* ??? */
+	so->gras_cl_clip_cntl = A3XX_GRAS_CL_CLIP_CNTL_IJ_PERSP_CENTER /* ??? */ |
+		COND(cso->clip_halfz, A3XX_GRAS_CL_CLIP_CNTL_ZERO_GB_SCALE_Z);
 	so->gras_su_point_minmax =
 			A3XX_GRAS_SU_POINT_MINMAX_MIN(psize_min) |
 			A3XX_GRAS_SU_POINT_MINMAX_MAX(psize_max);

src/gallium/drivers/freedreno/a4xx/fd4_context.c

@@ -96,7 +96,7 @@ static const uint8_t primtypes[PIPE_PRIM_MAX] = {
 };
 
 struct pipe_context *
-fd4_context_create(struct pipe_screen *pscreen, void *priv)
+fd4_context_create(struct pipe_screen *pscreen, void *priv, unsigned flags)
 {
 	struct fd_screen *screen = fd_screen(pscreen);
 	struct fd4_context *fd4_ctx = CALLOC_STRUCT(fd4_context);

src/gallium/drivers/freedreno/a4xx/fd4_context.h

@@ -97,6 +97,6 @@ fd4_context(struct fd_context *ctx)
 }
 
 struct pipe_context *
-fd4_context_create(struct pipe_screen *pscreen, void *priv);
+fd4_context_create(struct pipe_screen *pscreen, void *priv, unsigned flags);
 
 #endif /* FD4_CONTEXT_H_ */

src/gallium/drivers/freedreno/freedreno_screen.c

@@ -191,6 +191,7 @@ fd_screen_get_param(struct pipe_screen *pscreen, enum pipe_cap param)
 		return 16383;
 
 	case PIPE_CAP_DEPTH_CLIP_DISABLE:
+	case PIPE_CAP_CLIP_HALFZ:
 	case PIPE_CAP_SEAMLESS_CUBE_MAP_PER_TEXTURE:
 		return is_a3xx(screen);
 
@@ -228,7 +229,6 @@ fd_screen_get_param(struct pipe_screen *pscreen, enum pipe_cap param)
 	case PIPE_CAP_TGSI_FS_FINE_DERIVATIVE:
 	case PIPE_CAP_CONDITIONAL_RENDER_INVERTED:
 	case PIPE_CAP_SAMPLER_VIEW_TARGET:
-	case PIPE_CAP_CLIP_HALFZ:
 	case PIPE_CAP_POLYGON_OFFSET_CLAMP:
 	case PIPE_CAP_MULTISAMPLE_Z_RESOLVE:
 	case PIPE_CAP_RESOURCE_FROM_USER_MEMORY:
@@ -407,8 +407,6 @@ fd_screen_get_shader_param(struct pipe_screen *pscreen, unsigned shader,
 		return 16;
 	case PIPE_SHADER_CAP_PREFERRED_IR:
 		return PIPE_SHADER_IR_TGSI;
-	case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-		return 32;
 	}
 	debug_printf("unknown shader param %d\n", param);
 	return 0;

src/gallium/drivers/freedreno/ir3/ir3_compiler_nir.c

@@ -2312,7 +2312,7 @@ emit_instructions(struct ir3_compile *ctx)
 	ctx->ir = ir3_create(ctx->compiler, ninputs, noutputs);
 
 	/* Create inputs in first block: */
-	ctx->block = get_block(ctx, fxn->start_block);
+	ctx->block = get_block(ctx, nir_start_block(fxn));
 	ctx->in_block = ctx->block;
 	list_addtail(&ctx->block->node, &ctx->ir->block_list);
 

src/gallium/drivers/freedreno/ir3/ir3_nir_lower_if_else.c

@@ -29,6 +29,7 @@
 
 #include "ir3_nir.h"
 #include "glsl/nir/nir_builder.h"
+#include "glsl/nir/nir_control_flow.h"
 
 /* Based on nir_opt_peephole_select, and hacked up to more aggressively
  * flatten anything that can be flattened
@@ -171,7 +172,7 @@ flatten_block(nir_builder *bld, nir_block *if_block, nir_block *prev_block,
 					(intr->intrinsic == nir_intrinsic_discard_if)) {
 				nir_ssa_def *discard_cond;
 
-				nir_builder_insert_after_instr(bld,
+				bld->cursor = nir_after_instr(
 						nir_block_last_instr(prev_block));
 
 				if (invert) {

src/gallium/drivers/i915/i915_context.c

@@ -155,7 +155,7 @@ static void i915_destroy(struct pipe_context *pipe)
 }
 
 struct pipe_context *
-i915_create_context(struct pipe_screen *screen, void *priv)
+i915_create_context(struct pipe_screen *screen, void *priv, unsigned flags)
 {
    struct i915_context *i915;
 

src/gallium/drivers/i915/i915_context.h

@@ -401,7 +401,7 @@ void i915_init_string_functions( struct i915_context *i915 );
  * i915_context.c
  */
 struct pipe_context *i915_create_context(struct pipe_screen *screen,
-					 void *priv);
+					 void *priv, unsigned flags);
 
 
 /***********************************************************************

src/gallium/drivers/i915/i915_screen.c

@@ -167,8 +167,6 @@ i915_get_shader_param(struct pipe_screen *screen, unsigned shader, enum pipe_sha
       case PIPE_SHADER_CAP_TGSI_FMA_SUPPORTED:
       case PIPE_SHADER_CAP_TGSI_ANY_INOUT_DECL_RANGE:
          return 0;
-      case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-         return 32;
       default:
          debug_printf("%s: Unknown cap %u.\n", __FUNCTION__, cap);
          return 0;

src/gallium/drivers/ilo/ilo_context.c

@@ -135,7 +135,7 @@ ilo_context_destroy(struct pipe_context *pipe)
 }
 
 static struct pipe_context *
-ilo_context_create(struct pipe_screen *screen, void *priv)
+ilo_context_create(struct pipe_screen *screen, void *priv, unsigned flags)
 {
    struct ilo_screen *is = ilo_screen(screen);
    struct ilo_context *ilo;

src/gallium/drivers/ilo/ilo_screen.c

@@ -138,8 +138,6 @@ ilo_get_shader_param(struct pipe_screen *screen, unsigned shader,
       return PIPE_SHADER_IR_TGSI;
    case PIPE_SHADER_CAP_TGSI_SQRT_SUPPORTED:
       return 1;
-   case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-      return 32;
 
    default:
       return 0;

src/gallium/drivers/llvmpipe/lp_context.c

@@ -128,7 +128,8 @@ llvmpipe_render_condition ( struct pipe_context *pipe,
 }
 
 struct pipe_context *
-llvmpipe_create_context( struct pipe_screen *screen, void *priv )
+llvmpipe_create_context(struct pipe_screen *screen, void *priv,
+                        unsigned flags)
 {
    struct llvmpipe_context *llvmpipe;
 

src/gallium/drivers/llvmpipe/lp_context.h

@@ -160,7 +160,8 @@ struct llvmpipe_context {
 
 
 struct pipe_context *
-llvmpipe_create_context( struct pipe_screen *screen, void *priv );
+llvmpipe_create_context(struct pipe_screen *screen, void *priv,
+                        unsigned flags);
 
 struct pipe_resource *
 llvmpipe_user_buffer_create(struct pipe_screen *screen,

src/gallium/drivers/noop/noop_pipe.c

@@ -260,7 +260,8 @@ static void noop_destroy_context(struct pipe_context *ctx)
 	FREE(ctx);
 }
 
-static struct pipe_context *noop_create_context(struct pipe_screen *screen, void *priv)
+static struct pipe_context *noop_create_context(struct pipe_screen *screen,
+                                                void *priv, unsigned flags)
 {
 	struct pipe_context *ctx = CALLOC_STRUCT(pipe_context);
 

src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nv50.cpp

@@ -884,7 +884,7 @@ CodeEmitterNV50::emitINTERP(const Instruction *i)
    defId(i->def(0), 2);
    srcAddr8(i->src(0), 16);
 
-   if (i->encSize != 8 && i->getInterpMode() == NV50_IR_INTERP_FLAT) {
+   if (i->getInterpMode() == NV50_IR_INTERP_FLAT) {
       code[0] |= 1 << 8;
    } else {
       if (i->op == OP_PINTERP) {
@@ -896,11 +896,10 @@ CodeEmitterNV50::emitINTERP(const Instruction *i)
    }
 
    if (i->encSize == 8) {
-      if (i->getInterpMode() == NV50_IR_INTERP_FLAT)
-         code[1] = 4 << 16;
-      else
-         code[1] = (code[0] & (3 << 24)) >> (24 - 16);
-      code[0] &= ~0x03000000;
+      code[1] =
+         (code[0] & (3 << 24)) >> (24 - 16) |
+         (code[0] & (1 <<  8)) << (18 -  8);
+      code[0] &= ~0x03000100;
       code[0] |= 1;
       emitFlagsRd(i);
    }

src/gallium/drivers/nouveau/codegen/nv50_ir_lowering_nv50.cpp

@@ -202,10 +202,7 @@ NV50LegalizePostRA::visit(Function *fn)
    Program *prog = fn->getProgram();
 
    r63 = new_LValue(fn, FILE_GPR);
-   if (prog->maxGPR < 63)
-      r63->reg.data.id = 63;
-   else
-      r63->reg.data.id = 127;
+   r63->reg.data.id = 63;
 
    // this is actually per-program, but we can do it all on visiting main()
    std::list<Instruction *> *outWrites =

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

@@ -2602,10 +2602,6 @@ NV50PostRaConstantFolding::visit(BasicBlock *bb)
              !isFloatType(i->dType))
             break;
 
-         if (i->getDef(0)->reg.data.id >= 64 ||
-             i->getSrc(0)->reg.data.id >= 64)
-            break;
-
          def = i->getSrc(1)->getInsn();
          if (def->op == OP_MOV && def->src(0).getFile() == FILE_IMMEDIATE) {
             vtmp = i->getSrc(1);

src/gallium/drivers/nouveau/codegen/nv50_ir_print.cpp

@@ -411,7 +411,7 @@ int ImmediateValue::print(char *buf, size_t size, DataType ty) const
    case TYPE_U64:
    case TYPE_S64:
    default:
-      PRINT("0x%016" PRIx64, reg.data.u64);
+      PRINT("0x%016"PRIx64, reg.data.u64);
       break;
    }
    return pos;

src/gallium/drivers/nouveau/codegen/nv50_ir_ra.cpp

@@ -25,24 +25,9 @@
 
 #include <stack>
 #include <limits>
-#if __cplusplus >= 201103L
-#include <unordered_map>
-#else
-#include <tr1/unordered_map>
-#endif
 
 namespace nv50_ir {
 
-#if __cplusplus >= 201103L
-using std::hash;
-using std::unordered_map;
-#elif !defined(ANDROID)
-using std::tr1::hash;
-using std::tr1::unordered_map;
-#else
-#error Android release before Lollipop is not supported!
-#endif
-
 #define MAX_REGISTER_FILE_SIZE 256
 
 class RegisterSet
@@ -237,7 +222,6 @@ private:
    private:
       virtual bool visit(BasicBlock *);
       inline bool needNewElseBlock(BasicBlock *b, BasicBlock *p);
-      inline void splitEdges(BasicBlock *b);
    };
 
    class ArgumentMovesPass : public Pass {
@@ -361,89 +345,6 @@ RegAlloc::PhiMovesPass::needNewElseBlock(BasicBlock *b, BasicBlock *p)
    return (n == 2);
 }
 
-struct PhiMapHash {
-   size_t operator()(const std::pair<Instruction *, BasicBlock *>& val) const {
-      return hash<Instruction*>()(val.first) * 31 +
-         hash<BasicBlock*>()(val.second);
-   }
-};
-
-typedef unordered_map<
-   std::pair<Instruction *, BasicBlock *>, Value *, PhiMapHash> PhiMap;
-
-// Critical edges need to be split up so that work can be inserted along
-// specific edge transitions. Unfortunately manipulating incident edges into a
-// BB invalidates all the PHI nodes since their sources are implicitly ordered
-// by incident edge order.
-//
-// TODO: Make it so that that is not the case, and PHI nodes store pointers to
-// the original BBs.
-void
-RegAlloc::PhiMovesPass::splitEdges(BasicBlock *bb)
-{
-   BasicBlock *pb, *pn;
-   Instruction *phi;
-   Graph::EdgeIterator ei;
-   std::stack<BasicBlock *> stack;
-   int j = 0;
-
-   for (ei = bb->cfg.incident(); !ei.end(); ei.next()) {
-      pb = BasicBlock::get(ei.getNode());
-      assert(pb);
-      if (needNewElseBlock(bb, pb))
-         stack.push(pb);
-   }
-
-   // No critical edges were found, no need to perform any work.
-   if (stack.empty())
-      return;
-
-   // We're about to, potentially, reorder the inbound edges. This means that
-   // we need to hold on to the (phi, bb) -> src mapping, and fix up the phi
-   // nodes after the graph has been modified.
-   PhiMap phis;
-
-   j = 0;
-   for (ei = bb->cfg.incident(); !ei.end(); ei.next(), j++) {
-      pb = BasicBlock::get(ei.getNode());
-      for (phi = bb->getPhi(); phi && phi->op == OP_PHI; phi = phi->next)
-         phis.insert(std::make_pair(std::make_pair(phi, pb), phi->getSrc(j)));
-   }
-
-   while (!stack.empty()) {
-      pb = stack.top();
-      pn = new BasicBlock(func);
-      stack.pop();
-
-      pb->cfg.detach(&bb->cfg);
-      pb->cfg.attach(&pn->cfg, Graph::Edge::TREE);
-      pn->cfg.attach(&bb->cfg, Graph::Edge::FORWARD);
-
-      assert(pb->getExit()->op != OP_CALL);
-      if (pb->getExit()->asFlow()->target.bb == bb)
-         pb->getExit()->asFlow()->target.bb = pn;
-
-      for (phi = bb->getPhi(); phi && phi->op == OP_PHI; phi = phi->next) {
-         PhiMap::iterator it = phis.find(std::make_pair(phi, pb));
-         assert(it != phis.end());
-         phis.insert(std::make_pair(std::make_pair(phi, pn), it->second));
-         phis.erase(it);
-      }
-   }
-
-   // Now go through and fix up all of the phi node sources.
-   j = 0;
-   for (ei = bb->cfg.incident(); !ei.end(); ei.next(), j++) {
-      pb = BasicBlock::get(ei.getNode());
-      for (phi = bb->getPhi(); phi && phi->op == OP_PHI; phi = phi->next) {
-         PhiMap::const_iterator it = phis.find(std::make_pair(phi, pb));
-         assert(it != phis.end());
-
-         phi->setSrc(j, it->second);
-      }
-   }
-}
-
 // For each operand of each PHI in b, generate a new value by inserting a MOV
 // at the end of the block it is coming from and replace the operand with its
 // result. This eliminates liveness conflicts and enables us to let values be
@@ -456,13 +357,34 @@ bool
 RegAlloc::PhiMovesPass::visit(BasicBlock *bb)
 {
    Instruction *phi, *mov;
+   BasicBlock *pb, *pn;
 
-   splitEdges(bb);
+   std::stack<BasicBlock *> stack;
+
+   for (Graph::EdgeIterator ei = bb->cfg.incident(); !ei.end(); ei.next()) {
+      pb = BasicBlock::get(ei.getNode());
+      assert(pb);
+      if (needNewElseBlock(bb, pb))
+         stack.push(pb);
+   }
+   while (!stack.empty()) {
+      pb = stack.top();
+      pn = new BasicBlock(func);
+      stack.pop();
+
+      pb->cfg.detach(&bb->cfg);
+      pb->cfg.attach(&pn->cfg, Graph::Edge::TREE);
+      pn->cfg.attach(&bb->cfg, Graph::Edge::FORWARD);
+
+      assert(pb->getExit()->op != OP_CALL);
+      if (pb->getExit()->asFlow()->target.bb == bb)
+         pb->getExit()->asFlow()->target.bb = pn;
+   }
 
    // insert MOVs (phi->src(j) should stem from j-th in-BB)
    int j = 0;
    for (Graph::EdgeIterator ei = bb->cfg.incident(); !ei.end(); ei.next()) {
-      BasicBlock *pb = BasicBlock::get(ei.getNode());
+      pb = BasicBlock::get(ei.getNode());
       if (!pb->isTerminated())
          pb->insertTail(new_FlowInstruction(func, OP_BRA, bb));
 

src/gallium/drivers/nouveau/nouveau_buffer.c

@@ -80,12 +80,7 @@ release_allocation(struct nouveau_mm_allocation **mm,
 inline void
 nouveau_buffer_release_gpu_storage(struct nv04_resource *buf)
 {
-   if (buf->fence && buf->fence->state < NOUVEAU_FENCE_STATE_FLUSHED) {
-      nouveau_fence_work(buf->fence, nouveau_fence_unref_bo, buf->bo);
-      buf->bo = NULL;
-   } else {
-      nouveau_bo_ref(NULL, &buf->bo);
-   }
+   nouveau_bo_ref(NULL, &buf->bo);
 
    if (buf->mm)
       release_allocation(&buf->mm, buf->fence);
@@ -211,8 +206,8 @@ nouveau_transfer_write(struct nouveau_context *nv, struct nouveau_transfer *tx,
       nv->copy_data(nv, buf->bo, buf->offset + base, buf->domain,
                     tx->bo, tx->offset + offset, NOUVEAU_BO_GART, size);
    else
-   if (nv->push_cb && can_cb)
-      nv->push_cb(nv, buf,
+   if ((buf->base.bind & PIPE_BIND_CONSTANT_BUFFER) && nv->push_cb && can_cb)
+      nv->push_cb(nv, buf->bo, buf->domain, buf->offset, buf->base.width0,
                   base, size / 4, (const uint32_t *)data);
    else
       nv->push_data(nv, buf->bo, buf->offset + base, buf->domain, size, data);
@@ -286,8 +281,7 @@ nouveau_buffer_transfer_del(struct nouveau_context *nv,
 {
    if (tx->map) {
       if (likely(tx->bo)) {
-         nouveau_fence_work(nv->screen->fence.current,
-                            nouveau_fence_unref_bo, tx->bo);
+         nouveau_bo_ref(NULL, &tx->bo);
          if (tx->mm)
             release_allocation(&tx->mm, nv->screen->fence.current);
       } else {
@@ -538,13 +532,8 @@ nouveau_buffer_transfer_unmap(struct pipe_context *pipe,
    struct nv04_resource *buf = nv04_resource(transfer->resource);
 
    if (tx->base.usage & PIPE_TRANSFER_WRITE) {
-      if (!(tx->base.usage & PIPE_TRANSFER_FLUSH_EXPLICIT)) {
-         if (tx->map)
-            nouveau_transfer_write(nv, tx, 0, tx->base.box.width);
-
-         util_range_add(&buf->valid_buffer_range,
-                        tx->base.box.x, tx->base.box.x + tx->base.box.width);
-      }
+      if (!(tx->base.usage & PIPE_TRANSFER_FLUSH_EXPLICIT) && tx->map)
+         nouveau_transfer_write(nv, tx, 0, tx->base.box.width);
 
       if (likely(buf->domain)) {
          const uint8_t bind = buf->base.bind;
@@ -552,6 +541,9 @@ nouveau_buffer_transfer_unmap(struct pipe_context *pipe,
          if (bind & (PIPE_BIND_VERTEX_BUFFER | PIPE_BIND_INDEX_BUFFER))
             nv->vbo_dirty = true;
       }
+
+      util_range_add(&buf->valid_buffer_range,
+                     tx->base.box.x, tx->base.box.x + tx->base.box.width);
    }
 
    if (!tx->bo && (tx->base.usage & PIPE_TRANSFER_WRITE))
@@ -788,7 +780,7 @@ nouveau_buffer_migrate(struct nouveau_context *nv,
       nv->copy_data(nv, buf->bo, buf->offset, new_domain,
                     bo, offset, old_domain, buf->base.width0);
 
-      nouveau_fence_work(screen->fence.current, nouveau_fence_unref_bo, bo);
+      nouveau_bo_ref(NULL, &bo);
       if (mm)
          release_allocation(&mm, screen->fence.current);
    } else

src/gallium/drivers/nouveau/nouveau_buffer.h

@@ -41,8 +41,6 @@ struct nv04_resource {
    uint8_t status;
    uint8_t domain;
 
-   uint16_t cb_bindings[6]; /* per-shader per-slot bindings */
-
    struct nouveau_fence *fence;
    struct nouveau_fence *fence_wr;
 

src/gallium/drivers/nouveau/nouveau_context.h

@@ -6,8 +6,6 @@
 
 #define NOUVEAU_MAX_SCRATCH_BUFS 4
 
-struct nv04_resource;
-
 struct nouveau_context {
    struct pipe_context pipe;
    struct nouveau_screen *screen;
@@ -25,7 +23,8 @@ struct nouveau_context {
                      unsigned, const void *);
    /* base, size refer to the whole constant buffer */
    void (*push_cb)(struct nouveau_context *,
-                   struct nv04_resource *,
+                   struct nouveau_bo *, unsigned domain,
+                   unsigned base, unsigned size,
                    unsigned offset, unsigned words, const uint32_t *);
 
    /* @return: @ref reduced by nr of references found in context */

src/gallium/drivers/nouveau/nouveau_fence.c

@@ -190,14 +190,8 @@ nouveau_fence_wait(struct nouveau_fence *fence)
    /* wtf, someone is waiting on a fence in flush_notify handler? */
    assert(fence->state != NOUVEAU_FENCE_STATE_EMITTING);
 
-   if (fence->state < NOUVEAU_FENCE_STATE_EMITTED) {
-      PUSH_SPACE(screen->pushbuf, 8);
-      /* The space allocation might trigger a flush, which could emit the
-       * current fence. So check again.
-       */
-      if (fence->state < NOUVEAU_FENCE_STATE_EMITTED)
-         nouveau_fence_emit(fence);
-   }
+   if (fence->state < NOUVEAU_FENCE_STATE_EMITTED)
+      nouveau_fence_emit(fence);
 
    if (fence->state < NOUVEAU_FENCE_STATE_FLUSHED)
       if (nouveau_pushbuf_kick(screen->pushbuf, screen->pushbuf->channel))
@@ -230,22 +224,10 @@ nouveau_fence_wait(struct nouveau_fence *fence)
 void
 nouveau_fence_next(struct nouveau_screen *screen)
 {
-   if (screen->fence.current->state < NOUVEAU_FENCE_STATE_EMITTING) {
-      if (screen->fence.current->ref > 1)
-         nouveau_fence_emit(screen->fence.current);
-      else
-         return;
-   }
+   if (screen->fence.current->state < NOUVEAU_FENCE_STATE_EMITTING)
+      nouveau_fence_emit(screen->fence.current);
 
    nouveau_fence_ref(NULL, &screen->fence.current);
 
    nouveau_fence_new(screen, &screen->fence.current, false);
 }
-
-void
-nouveau_fence_unref_bo(void *data)
-{
-   struct nouveau_bo *bo = data;
-
-   nouveau_bo_ref(NULL, &bo);
-}

src/gallium/drivers/nouveau/nouveau_fence.h

@@ -37,9 +37,6 @@ void nouveau_fence_next(struct nouveau_screen *);
 bool nouveau_fence_wait(struct nouveau_fence *);
 bool nouveau_fence_signalled(struct nouveau_fence *);
 
-void nouveau_fence_unref_bo(void *data); /* generic unref bo callback */
-
-
 static inline void
 nouveau_fence_ref(struct nouveau_fence *fence, struct nouveau_fence **ref)
 {

src/gallium/drivers/nouveau/nouveau_winsys.h

@@ -24,8 +24,6 @@ PUSH_AVAIL(struct nouveau_pushbuf *push)
 static inline bool
 PUSH_SPACE(struct nouveau_pushbuf *push, uint32_t size)
 {
-   /* Provide a buffer so that fences always have room to be emitted */
-   size += 8;
    if (PUSH_AVAIL(push) < size)
       return nouveau_pushbuf_space(push, size, 0, 0) == 0;
    return true;

src/gallium/drivers/nouveau/nv30/nv30_context.c

@@ -190,7 +190,7 @@ nv30_context_destroy(struct pipe_context *pipe)
    } while(0)
 
 struct pipe_context *
-nv30_context_create(struct pipe_screen *pscreen, void *priv)
+nv30_context_create(struct pipe_screen *pscreen, void *priv, unsigned ctxflags)
 {
    struct nv30_screen *screen = nv30_screen(pscreen);
    struct nv30_context *nv30 = CALLOC_STRUCT(nv30_context);

src/gallium/drivers/nouveau/nv30/nv30_context.h

@@ -132,7 +132,7 @@ nv30_context(struct pipe_context *pipe)
 }
 
 struct pipe_context *
-nv30_context_create(struct pipe_screen *pscreen, void *priv);
+nv30_context_create(struct pipe_screen *pscreen, void *priv, unsigned flags);
 
 void
 nv30_vbo_init(struct pipe_context *pipe);

src/gallium/drivers/nouveau/nv30/nv30_format.c

@@ -78,12 +78,12 @@ nv30_format_info_table[PIPE_FORMAT_COUNT] = {
    _(B4G4R4X4_UNORM      , S___),
    _(B4G4R4A4_UNORM      , S___),
    _(B5G6R5_UNORM        , SB__),
-   _(BGRX8888_UNORM      , SB__),
-   _(BGRX8888_SRGB       , S___),
-   _(BGRA8888_UNORM      , SB__),
-   _(BGRA8888_SRGB       , S___),
+   _(B8G8R8X8_UNORM      , SB__),
+   _(B8G8R8X8_SRGB       , S___),
+   _(B8G8R8A8_UNORM      , SB__),
+   _(B8G8R8A8_SRGB       , S___),
    _(R8G8B8A8_UNORM      , __V_),
-   _(RGBA8888_SNORM      , S___),
+   _(R8G8B8A8_SNORM      , S___),
    _(DXT1_RGB            , S___),
    _(DXT1_SRGB           , S___),
    _(DXT1_RGBA           , S___),
@@ -138,8 +138,8 @@ const struct nv30_format
 nv30_format_table[PIPE_FORMAT_COUNT] = {
    R_(B5G5R5X1_UNORM    , X1R5G5B5          ),
    R_(B5G6R5_UNORM      , R5G6B5            ),
-   R_(BGRX8888_UNORM    , X8R8G8B8          ),
-   R_(BGRA8888_UNORM    , A8R8G8B8          ),
+   R_(B8G8R8X8_UNORM    , X8R8G8B8          ),
+   R_(B8G8R8A8_UNORM    , A8R8G8B8          ),
    Z_(Z16_UNORM         , Z16               ),
    Z_(X8Z24_UNORM       , Z24S8             ),
    Z_(S8_UINT_Z24_UNORM , Z24S8             ),
@@ -223,11 +223,11 @@ nv30_texfmt_table[PIPE_FORMAT_COUNT] = {
    _(B4G4R4X4_UNORM    , A4R4G4B4, 0, C, C, C, 1, 2, 1, 0, x, NONE, ____),
    _(B4G4R4A4_UNORM    , A4R4G4B4, 0, C, C, C, C, 2, 1, 0, 3, NONE, ____),
    _(B5G6R5_UNORM      , R5G6B5  , 0, C, C, C, 1, 2, 1, 0, x, NONE, ____),
-   _(BGRX8888_UNORM    , A8R8G8B8, 0, C, C, C, 1, 2, 1, 0, x, NONE, ____),
-   _(BGRX8888_SRGB     , A8R8G8B8, 0, C, C, C, 1, 2, 1, 0, x, SRGB, ____),
-   _(BGRA8888_UNORM    , A8R8G8B8, 0, C, C, C, C, 2, 1, 0, 3, NONE, ____),
-   _(BGRA8888_SRGB     , A8R8G8B8, 0, C, C, C, C, 2, 1, 0, 3, SRGB, ____),
-   _(RGBA8888_SNORM    , A8R8G8B8, 0, C, C, C, C, 0, 1, 2, 3, NONE, SSSS),
+   _(B8G8R8X8_UNORM    , A8R8G8B8, 0, C, C, C, 1, 2, 1, 0, x, NONE, ____),
+   _(B8G8R8X8_SRGB     , A8R8G8B8, 0, C, C, C, 1, 2, 1, 0, x, SRGB, ____),
+   _(B8G8R8A8_UNORM    , A8R8G8B8, 0, C, C, C, C, 2, 1, 0, 3, NONE, ____),
+   _(B8G8R8A8_SRGB     , A8R8G8B8, 0, C, C, C, C, 2, 1, 0, 3, SRGB, ____),
+   _(R8G8B8A8_SNORM    , A8R8G8B8, 0, C, C, C, C, 0, 1, 2, 3, NONE, SSSS),
    _(DXT1_RGB          , DXT1    , 0, C, C, C, 1, 2, 1, 0, x, NONE, ____),
    _(DXT1_SRGB         , DXT1    , 0, C, C, C, 1, 2, 1, 0, x, SRGB, ____),
    _(DXT1_RGBA         , DXT1    , 0, C, C, C, C, 2, 1, 0, 3, NONE, ____),

src/gallium/drivers/nouveau/nv30/nv30_miptree.c

@@ -28,7 +28,6 @@
 #include "util/u_surface.h"
 
 #include "nv_m2mf.xml.h"
-#include "nv_object.xml.h"
 #include "nv30/nv30_screen.h"
 #include "nv30/nv30_context.h"
 #include "nv30/nv30_resource.h"
@@ -145,54 +144,21 @@ nv30_resource_copy_region(struct pipe_context *pipe,
    nv30_transfer_rect(nv30, NEAREST, &src, &dst);
 }
 
-static void
-nv30_resource_resolve(struct nv30_context *nv30,
-                      const struct pipe_blit_info *info)
+void
+nv30_resource_resolve(struct pipe_context *pipe,
+                      const struct pipe_resolve_info *info)
 {
-   struct nv30_miptree *src_mt = nv30_miptree(info->src.resource);
+#if 0
+   struct nv30_context *nv30 = nv30_context(pipe);
    struct nv30_rect src, dst;
-   unsigned x, x0, x1, y, y1, w, h;
 
-   define_rect(info->src.resource, 0, info->src.box.z, info->src.box.x,
-      info->src.box.y, info->src.box.width, info->src.box.height, &src);
-   define_rect(info->dst.resource, 0, info->dst.box.z, info->dst.box.x,
-      info->dst.box.y, info->dst.box.width, info->dst.box.height, &dst);
+   define_rect(info->src.res, 0, 0, info->src.x0, info->src.y0,
+               info->src.x1 - info->src.x0, info->src.y1 - info->src.y0, &src);
+   define_rect(info->dst.res, info->dst.level, 0, info->dst.x0, info->dst.y0,
+               info->dst.x1 - info->dst.x0, info->dst.y1 - info->dst.y0, &dst);
 
-   x0 = src.x0;
-   x1 = src.x1;
-   y1 = src.y1;
-
-   /* On nv3x we must use sifm which is restricted to 1024x1024 tiles */
-   for (y = src.y0; y < y1; y += h) {
-      h = y1 - y;
-      if (h > 1024)
-         h = 1024;
-
-      src.y0 = 0;
-      src.y1 = h;
-      src.h = h;
-
-      dst.y1 = dst.y0 + (h >> src_mt->ms_y);
-      dst.h = h >> src_mt->ms_y;
-
-      for (x = x0; x < x1; x += w) {
-         w = x1 - x;
-         if (w > 1024)
-            w = 1024;
-
-         src.offset = y * src.pitch + x * src.cpp;
-         src.x0 = 0;
-         src.x1 = w;
-         src.w = w;
-
-         dst.offset = (y >> src_mt->ms_y) * dst.pitch +
-                      (x >> src_mt->ms_x) * dst.cpp;
-         dst.x1 = dst.x0 + (w >> src_mt->ms_x);
-         dst.w = w >> src_mt->ms_x;
-
-         nv30_transfer_rect(nv30, BILINEAR, &src, &dst);
-      }
-   }
+   nv30_transfer_rect(nv30, BILINEAR, &src, &dst);
+#endif
 }
 
 void
@@ -206,7 +172,7 @@ nv30_blit(struct pipe_context *pipe,
        info.dst.resource->nr_samples <= 1 &&
        !util_format_is_depth_or_stencil(info.src.resource->format) &&
        !util_format_is_pure_integer(info.src.resource->format)) {
-      nv30_resource_resolve(nv30, blit_info);
+      debug_printf("nv30: color resolve unimplemented\n");
       return;
    }
 
@@ -339,15 +305,10 @@ nv30_miptree_transfer_unmap(struct pipe_context *pipe,
    struct nv30_context *nv30 = nv30_context(pipe);
    struct nv30_transfer *tx = nv30_transfer(ptx);
 
-   if (ptx->usage & PIPE_TRANSFER_WRITE) {
+   if (ptx->usage & PIPE_TRANSFER_WRITE)
       nv30_transfer_rect(nv30, NEAREST, &tx->tmp, &tx->img);
 
-      /* Allow the copies above to finish executing before freeing the source */
-      nouveau_fence_work(nv30->screen->base.fence.current,
-                         nouveau_fence_unref_bo, tx->tmp.bo);
-   } else {
-      nouveau_bo_ref(NULL, &tx->tmp.bo);
-   }
+   nouveau_bo_ref(NULL, &tx->tmp.bo);
    pipe_resource_reference(&ptx->resource, NULL);
    FREE(tx);
 }
@@ -401,7 +362,6 @@ nv30_miptree_create(struct pipe_screen *pscreen,
    blocksz = util_format_get_blocksize(pt->format);
 
    if ((pt->target == PIPE_TEXTURE_RECT) ||
-       (pt->bind & PIPE_BIND_SCANOUT) ||
        !util_is_power_of_two(pt->width0) ||
        !util_is_power_of_two(pt->height0) ||
        !util_is_power_of_two(pt->depth0) ||
@@ -409,14 +369,6 @@ nv30_miptree_create(struct pipe_screen *pscreen,
        util_format_is_float(pt->format) || mt->ms_mode) {
       mt->uniform_pitch = util_format_get_nblocksx(pt->format, w) * blocksz;
       mt->uniform_pitch = align(mt->uniform_pitch, 64);
-      if (pt->bind & PIPE_BIND_SCANOUT) {
-         struct nv30_screen *screen = nv30_screen(pscreen);
-         int pitch_align = MAX2(
-               screen->eng3d->oclass >= NV40_3D_CLASS ? 1024 : 256,
-               /* round_down_pow2(mt->uniform_pitch / 4) */
-               1 << (util_last_bit(mt->uniform_pitch / 4) - 1));
-         mt->uniform_pitch = align(mt->uniform_pitch, pitch_align);
-      }
    }
 
    if (!mt->uniform_pitch)

src/gallium/drivers/nouveau/nv30/nv30_resource.h

@@ -65,6 +65,9 @@ nv30_resource_copy_region(struct pipe_context *pipe,
                           struct pipe_resource *src, unsigned src_level,
                           const struct pipe_box *src_box);
 
+void
+nv30_resource_resolve(struct pipe_context *, const struct pipe_resolve_info *);
+
 void
 nv30_blit(struct pipe_context *pipe,
           const struct pipe_blit_info *blit_info);

src/gallium/drivers/nouveau/nv30/nv30_screen.c

@@ -261,8 +261,6 @@ nv30_screen_get_shader_param(struct pipe_screen *pscreen, unsigned shader,
       case PIPE_SHADER_CAP_TGSI_FMA_SUPPORTED:
       case PIPE_SHADER_CAP_TGSI_ANY_INOUT_DECL_RANGE:
          return 0;
-      case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-         return 32;
       default:
          debug_printf("unknown vertex shader param %d\n", param);
          return 0;
@@ -304,8 +302,6 @@ nv30_screen_get_shader_param(struct pipe_screen *pscreen, unsigned shader,
       case PIPE_SHADER_CAP_TGSI_FMA_SUPPORTED:
       case PIPE_SHADER_CAP_TGSI_ANY_INOUT_DECL_RANGE:
          return 0;
-      case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-         return 32;
       default:
          debug_printf("unknown fragment shader param %d\n", param);
          return 0;
@@ -323,9 +319,8 @@ nv30_screen_is_format_supported(struct pipe_screen *pscreen,
                                 unsigned sample_count,
                                 unsigned bindings)
 {
-   if (sample_count > nv30_screen(pscreen)->max_sample_count)
+   if (sample_count > 4)
       return false;
-
    if (!(0x00000017 & (1 << sample_count)))
       return false;
 
@@ -349,9 +344,7 @@ nv30_screen_fence_emit(struct pipe_screen *pscreen, uint32_t *sequence)
 
    *sequence = ++screen->base.fence.sequence;
 
-   assert(PUSH_AVAIL(push) + push->rsvd_kick >= 3);
-   PUSH_DATA (push, NV30_3D_FENCE_OFFSET |
-              (2 /* size */ << 18) | (7 /* subchan */ << 13));
+   BEGIN_NV04(push, NV30_3D(FENCE_OFFSET), 2);
    PUSH_DATA (push, 0);
    PUSH_DATA (push, *sequence);
 }
@@ -457,23 +450,6 @@ nv30_screen_create(struct nouveau_device *dev)
       return NULL;
    }
 
-   /*
-    * Some modern apps try to use msaa without keeping in mind the
-    * restrictions on videomem of older cards. Resulting in dmesg saying:
-    * [ 1197.850642] nouveau E[soffice.bin[3785]] fail ttm_validate
-    * [ 1197.850648] nouveau E[soffice.bin[3785]] validating bo list
-    * [ 1197.850654] nouveau E[soffice.bin[3785]] validate: -12
-    *
-    * Because we are running out of video memory, after which the program
-    * using the msaa visual freezes, and eventually the entire system freezes.
-    *
-    * To work around this we do not allow msaa visauls by default and allow
-    * the user to override this via NV30_MAX_MSAA.
-    */
-   screen->max_sample_count = debug_get_num_option("NV30_MAX_MSAA", 0);
-   if (screen->max_sample_count > 4)
-      screen->max_sample_count = 4;
-
    pscreen = &screen->base.base;
    pscreen->destroy = nv30_screen_destroy;
    pscreen->get_param = nv30_screen_get_param;

src/gallium/drivers/nouveau/nv30/nv30_screen.h

@@ -38,8 +38,6 @@ struct nv30_screen {
    /*XXX: nvfx state */
    struct nouveau_heap *vp_exec_heap;
    struct nouveau_heap *vp_data_heap;
-
-   unsigned max_sample_count;
 };
 
 static inline struct nv30_screen *

src/gallium/drivers/nouveau/nv30/nv30_transfer.c

@@ -371,7 +371,7 @@ nv30_transfer_rect_blit(XFER_ARGS)
 static bool
 nv30_transfer_sifm(XFER_ARGS)
 {
-   if (!src->pitch || src->w > 1024 || src->h > 1024 || src->w < 2 || src->h < 2)
+   if (!src->pitch || (src->w | src->h) > 1024 || src->w < 2 || src->h < 2)
       return false;
 
    if (src->d > 1 || dst->d > 1)
@@ -381,7 +381,7 @@ nv30_transfer_sifm(XFER_ARGS)
       return false;
 
    if (!dst->pitch) {
-      if (dst->w > 2048 || dst->h > 2048 || dst->w < 2 || dst->h < 2)
+      if ((dst->w | dst->h) > 2048 || dst->w < 2 || dst->h < 2)
          return false;
    } else {
       if (dst->domain != NOUVEAU_BO_VRAM)

src/gallium/drivers/nouveau/nv30/nv30_vbo.c

@@ -191,11 +191,7 @@ nv30_vbo_validate(struct nv30_context *nv30)
    if (!nv30->vertex || nv30->draw_flags)
       return;
 
-#ifdef PIPE_ARCH_BIG_ENDIAN
-   if (1) { /* Figure out where the buffers are getting messed up */
-#else
    if (unlikely(vertex->need_conversion)) {
-#endif
       nv30->vbo_fifo = ~0;
       nv30->vbo_user = 0;
    } else {

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

@@ -1,4 +1,3 @@
-#include <strings.h>
 #include "pipe/p_context.h"
 #include "pipe/p_defines.h"
 #include "pipe/p_state.h"

src/gallium/drivers/nouveau/nv50/nv50_context.c

@@ -199,13 +199,9 @@ nv50_invalidate_resource_storage(struct nouveau_context *ctx,
          }
       }
 
-      if (nv50->idxbuf.buffer == res) {
-         /* Just rebind to the bufctx as there is no separate dirty bit */
-         nouveau_bufctx_reset(nv50->bufctx_3d, NV50_BIND_INDEX);
-         BCTX_REFN(nv50->bufctx_3d, INDEX, nv04_resource(res), RD);
+      if (nv50->idxbuf.buffer == res)
          if (!--ref)
             return ref;
-      }
 
       for (s = 0; s < 3; ++s) {
       assert(nv50->num_textures[s] <= PIPE_MAX_SAMPLERS);
@@ -244,7 +240,7 @@ nv50_context_get_sample_position(struct pipe_context *, unsigned, unsigned,
                                  float *);
 
 struct pipe_context *
-nv50_create(struct pipe_screen *pscreen, void *priv)
+nv50_create(struct pipe_screen *pscreen, void *priv, unsigned ctxflags)
 {
    struct nv50_screen *screen = nv50_screen(pscreen);
    struct nv50_context *nv50;

src/gallium/drivers/nouveau/nv50/nv50_context.h

@@ -186,7 +186,7 @@ nv50_context_shader_stage(unsigned pipe)
 }
 
 /* nv50_context.c */
-struct pipe_context *nv50_create(struct pipe_screen *, void *);
+struct pipe_context *nv50_create(struct pipe_screen *, void *, unsigned flags);
 
 void nv50_bufctx_fence(struct nouveau_bufctx *, bool on_flush);
 
@@ -197,7 +197,7 @@ extern struct draw_stage *nv50_draw_render_stage(struct nv50_context *);
 
 /* nv50_query.c */
 void nv50_init_query_functions(struct nv50_context *);
-void nv50_query_pushbuf_submit(struct nouveau_pushbuf *, uint16_t method,
+void nv50_query_pushbuf_submit(struct nouveau_pushbuf *,
                                struct pipe_query *, unsigned result_offset);
 void nv84_query_fifo_wait(struct nouveau_pushbuf *, struct pipe_query *);
 void nva0_so_target_save_offset(struct pipe_context *,

src/gallium/drivers/nouveau/nv50/nv50_formats.c

@@ -203,10 +203,8 @@ const struct nv50_format nv50_format_table[PIPE_FORMAT_COUNT] =
    F3B(B5G6R5_UNORM, B5G6R5_UNORM, C2, C1, C0, xx, UNORM, 5_6_5, TD),
    C4B(B5G5R5A1_UNORM, BGR5_A1_UNORM, C2, C1, C0, C3, UNORM, 5_5_5_1, TD),
    F3B(B5G5R5X1_UNORM, BGR5_X1_UNORM, C2, C1, C0, xx, UNORM, 5_5_5_1, TD),
-#if NOUVEAU_DRIVER != 0xc0
    C4B(B4G4R4A4_UNORM, NONE, C2, C1, C0, C3, UNORM, 4_4_4_4, T),
    F3B(B4G4R4X4_UNORM, NONE, C2, C1, C0, xx, UNORM, 4_4_4_4, T),
-#endif
    F3B(R9G9B9E5_FLOAT, NONE, C0, C1, C2, xx, FLOAT, 9_9_9_E5, T),
 
    C4A(R10G10B10A2_UNORM, RGB10_A2_UNORM, C0, C1, C2, C3, UNORM, 10_10_10_2,

src/gallium/drivers/nouveau/nv50/nv50_miptree.c

@@ -163,10 +163,7 @@ nv50_miptree_destroy(struct pipe_screen *pscreen, struct pipe_resource *pt)
 {
    struct nv50_miptree *mt = nv50_miptree(pt);
 
-   if (mt->base.fence && mt->base.fence->state < NOUVEAU_FENCE_STATE_FLUSHED)
-      nouveau_fence_work(mt->base.fence, nouveau_fence_unref_bo, mt->base.bo);
-   else
-      nouveau_bo_ref(NULL, &mt->base.bo);
+   nouveau_bo_ref(NULL, &mt->base.bo);
 
    nouveau_fence_ref(NULL, &mt->base.fence);
    nouveau_fence_ref(NULL, &mt->base.fence_wr);

src/gallium/drivers/nouveau/nv50/nv50_program.c

@@ -66,7 +66,6 @@ nv50_vertprog_assign_slots(struct nv50_ir_prog_info *info)
       case TGSI_SEMANTIC_VERTEXID:
          prog->vp.attrs[2] |= NV50_3D_VP_GP_BUILTIN_ATTR_EN_VERTEX_ID;
          prog->vp.attrs[2] |= NV50_3D_VP_GP_BUILTIN_ATTR_EN_VERTEX_ID_DRAW_ARRAYS_ADD_START;
-         prog->vp.vertexid = 1;
          continue;
       default:
          break;

src/gallium/drivers/nouveau/nv50/nv50_program.h

@@ -76,7 +76,6 @@ struct nv50_program {
       ubyte psiz;        /* output slot of point size */
       ubyte bfc[2];      /* indices into varying for FFC (FP) or BFC (VP) */
       ubyte edgeflag;
-      ubyte vertexid;
       ubyte clpd[2];     /* output slot of clip distance[i]'s 1st component */
       ubyte clpd_nr;
    } vp;

src/gallium/drivers/nouveau/nv50/nv50_query.c

@@ -266,7 +266,6 @@ nv50_query_end(struct pipe_context *pipe, struct pipe_query *pq)
       nv50_query_get(push, q, 0, 0x1000f010);
       break;
    case NVA0_QUERY_STREAM_OUTPUT_BUFFER_OFFSET:
-      q->sequence++;
       nv50_query_get(push, q, 0, 0x0d005002 | (q->index << 5));
       break;
    case PIPE_QUERY_TIMESTAMP_DISJOINT:
@@ -452,18 +451,18 @@ nv50_render_condition(struct pipe_context *pipe,
 }
 
 void
-nv50_query_pushbuf_submit(struct nouveau_pushbuf *push, uint16_t method,
+nv50_query_pushbuf_submit(struct nouveau_pushbuf *push,
                           struct pipe_query *pq, unsigned result_offset)
 {
    struct nv50_query *q = nv50_query(pq);
 
-   nv50_query_update(q);
-   if (q->state != NV50_QUERY_STATE_READY)
-      nouveau_bo_wait(q->bo, NOUVEAU_BO_RD, push->client);
-   q->state = NV50_QUERY_STATE_READY;
+   /* XXX: does this exist ? */
+#define NV50_IB_ENTRY_1_NO_PREFETCH (0 << (31 - 8))
 
-   BEGIN_NV04(push, SUBC_3D(method), 1);
-   PUSH_DATA (push, q->data[result_offset / 4]);
+   PUSH_REFN(push, q->bo, NOUVEAU_BO_RD | NOUVEAU_BO_GART);
+   nouveau_pushbuf_space(push, 0, 0, 1);
+   nouveau_pushbuf_data(push, q->bo, q->offset + result_offset, 4 |
+                        NV50_IB_ENTRY_1_NO_PREFETCH);
 }
 
 void

src/gallium/drivers/nouveau/nv50/nv50_screen.c

@@ -100,7 +100,7 @@ nv50_screen_get_param(struct pipe_screen *pscreen, enum pipe_cap param)
    case PIPE_CAP_MAX_TEXEL_OFFSET:
       return 7;
    case PIPE_CAP_MAX_TEXTURE_BUFFER_SIZE:
-      return 128 * 1024 * 1024;
+      return 65536;
    case PIPE_CAP_GLSL_FEATURE_LEVEL:
       return 330;
    case PIPE_CAP_MAX_RENDER_TARGETS:
@@ -297,8 +297,6 @@ nv50_screen_get_shader_param(struct pipe_screen *pscreen, unsigned shader,
    case PIPE_SHADER_CAP_TGSI_FMA_SUPPORTED:
    case PIPE_SHADER_CAP_TGSI_ANY_INOUT_DECL_RANGE:
       return 0;
-   case PIPE_SHADER_CAP_MAX_UNROLL_ITERATIONS_HINT:
-      return 32;
    default:
       NOUVEAU_ERR("unknown PIPE_SHADER_CAP %d\n", param);
       return 0;
@@ -388,7 +386,6 @@ nv50_screen_fence_emit(struct pipe_screen *pscreen, u32 *sequence)
    /* we need to do it after possible flush in MARK_RING */
    *sequence = ++screen->base.fence.sequence;
 
-   assert(PUSH_AVAIL(push) + push->rsvd_kick >= 5);
    PUSH_DATA (push, NV50_FIFO_PKHDR(NV50_3D(QUERY_ADDRESS_HIGH), 4));
    PUSH_DATAh(push, screen->fence.bo->offset);
    PUSH_DATA (push, screen->fence.bo->offset);

src/gallium/drivers/nouveau/nv50/nv50_shader_state.c

@@ -65,9 +65,14 @@ nv50_constbufs_validate(struct nv50_context *nv50)
                PUSH_DATA (push, (b << 12) | (i << 8) | p | 1);
             }
             while (words) {
-               unsigned nr = MIN2(words, NV04_PFIFO_MAX_PACKET_LEN);
+               unsigned nr;
+
+               if (!PUSH_SPACE(push, 16))
+                  break;
+               nr = PUSH_AVAIL(push);
+               assert(nr >= 16);
+               nr = MIN2(MIN2(nr - 3, words), NV04_PFIFO_MAX_PACKET_LEN);
 
-               PUSH_SPACE(push, nr + 3);
                BEGIN_NV04(push, NV50_3D(CB_ADDR), 1);
                PUSH_DATA (push, (start << 8) | b);
                BEGIN_NI04(push, NV50_3D(CB_DATA(0)), nr);
@@ -636,12 +641,12 @@ nv50_stream_output_validate(struct nv50_context *nv50)
       PUSH_DATA (push, so->num_attribs[i]);
       if (n == 4) {
          PUSH_DATA(push, targ->pipe.buffer_size);
+
+         BEGIN_NV04(push, NVA0_3D(STRMOUT_OFFSET(i)), 1);
          if (!targ->clean) {
             assert(targ->pq);
-            nv50_query_pushbuf_submit(push, NVA0_3D_STRMOUT_OFFSET(i),
-                                      targ->pq, 0x4);
+            nv50_query_pushbuf_submit(push, targ->pq, 0x4);
          } else {
-            BEGIN_NV04(push, NVA0_3D(STRMOUT_OFFSET(i)), 1);
             PUSH_DATA(push, 0);
             targ->clean = false;
          }
@@ -650,7 +655,6 @@ nv50_stream_output_validate(struct nv50_context *nv50)
             (so->stride[i] * nv50->state.prim_size);
          prims = MIN2(prims, limit);
       }
-      targ->stride = so->stride[i];
       BCTX_REFN(nv50->bufctx_3d, SO, buf, WR);
    }
    if (prims != ~0) {

src/gallium/drivers/nouveau/nv50/nv50_state_validate.c

@@ -503,8 +503,7 @@ static struct state_validate {
     { nv50_validate_samplers,      NV50_NEW_SAMPLERS },
     { nv50_stream_output_validate, NV50_NEW_STRMOUT |
                                    NV50_NEW_VERTPROG | NV50_NEW_GMTYPROG },
-    { nv50_vertex_arrays_validate, NV50_NEW_VERTEX | NV50_NEW_ARRAYS |
-                                   NV50_NEW_VERTPROG },
+    { nv50_vertex_arrays_validate, NV50_NEW_VERTEX | NV50_NEW_ARRAYS },
     { nv50_validate_min_samples,   NV50_NEW_MIN_SAMPLES },
 };
 #define validate_list_len (sizeof(validate_list) / sizeof(validate_list[0]))

src/gallium/drivers/nouveau/nv50/nv50_tex.c

@@ -221,26 +221,6 @@ nv50_create_texture_view(struct pipe_context *pipe,
    return &view->pipe;
 }
 
-static void
-nv50_update_tic(struct nv50_context *nv50, struct nv50_tic_entry *tic,
-                struct nv04_resource *res)
-{
-   uint64_t address = res->address;
-   if (res->base.target != PIPE_BUFFER)
-      return;
-   address += tic->pipe.u.buf.first_element *
-      util_format_get_blocksize(tic->pipe.format);
-   if (tic->tic[1] == (uint32_t)address &&
-       (tic->tic[2] & 0xff) == address >> 32)
-      return;
-
-   nv50_screen_tic_unlock(nv50->screen, tic);
-   tic->id = -1;
-   tic->tic[1] = address;
-   tic->tic[2] &= 0xffffff00;
-   tic->tic[2] |= address >> 32;
-}
-
 static bool
 nv50_validate_tic(struct nv50_context *nv50, int s)
 {
@@ -260,7 +240,6 @@ nv50_validate_tic(struct nv50_context *nv50, int s)
          continue;
       }
       res = &nv50_miptree(tic->pipe.texture)->base;
-      nv50_update_tic(nv50, tic, res);
 
       if (tic->id < 0) {
          tic->id = nv50_screen_tic_alloc(nv50->screen, tic);

src/gallium/drivers/nouveau/nv50/nv50_transfer.c

@@ -187,7 +187,14 @@ nv50_sifc_linear_u8(struct nouveau_context *nv,
    PUSH_DATA (push, 0);
 
    while (count) {
-      unsigned nr = MIN2(count, NV04_PFIFO_MAX_PACKET_LEN);
+      unsigned nr;
+
+      if (!PUSH_SPACE(push, 16))
+         break;
+      nr = PUSH_AVAIL(push);
+      assert(nr >= 16);
+      nr = MIN2(count, nr - 1);
+      nr = MIN2(nr, NV04_PFIFO_MAX_PACKET_LEN);
 
       BEGIN_NI04(push, NV50_2D(SIFC_DATA), nr);
       PUSH_DATAp(push, src, nr);
@@ -358,14 +365,9 @@ nv50_miptree_transfer_unmap(struct pipe_context *pctx,
             tx->rect[0].base += mt->layer_stride;
          tx->rect[1].base += tx->nblocksy * tx->base.stride;
       }
-
-      /* Allow the copies above to finish executing before freeing the source */
-      nouveau_fence_work(nv50->screen->base.fence.current,
-                         nouveau_fence_unref_bo, tx->rect[1].bo);
-   } else {
-      nouveau_bo_ref(NULL, &tx->rect[1].bo);
    }
 
+   nouveau_bo_ref(NULL, &tx->rect[1].bo);
    pipe_resource_reference(&transfer->resource, NULL);
 
    FREE(tx);
@@ -388,9 +390,12 @@ nv50_cb_push(struct nouveau_context *nv,
    nouveau_pushbuf_validate(push);
 
    while (words) {
-      unsigned nr = MIN2(words, NV04_PFIFO_MAX_PACKET_LEN);
+      unsigned nr;
+
+      nr = PUSH_AVAIL(push);
+      nr = MIN2(nr - 7, words);
+      nr = MIN2(nr, NV04_PFIFO_MAX_PACKET_LEN - 1);
 
-      PUSH_SPACE(push, nr + 7);
       BEGIN_NV04(push, NV50_3D(CB_DEF_ADDRESS_HIGH), 3);
       PUSH_DATAh(push, bo->offset + base);
       PUSH_DATA (push, bo->offset + base);

src/gallium/drivers/nouveau/nv50/nv50_vbo.c

@@ -293,8 +293,7 @@ nv50_vertex_arrays_validate(struct nv50_context *nv50)
    uint64_t addrs[PIPE_MAX_ATTRIBS];
    uint32_t limits[PIPE_MAX_ATTRIBS];
    struct nouveau_pushbuf *push = nv50->base.pushbuf;
-   struct nv50_vertex_stateobj dummy = {};
-   struct nv50_vertex_stateobj *vertex = nv50->vertex ? nv50->vertex : &dummy;
+   struct nv50_vertex_stateobj *vertex = nv50->vertex;
    struct pipe_vertex_buffer *vb;
    struct nv50_vertex_element *ve;
    uint32_t mask;
@@ -302,14 +301,6 @@ nv50_vertex_arrays_validate(struct nv50_context *nv50)
    unsigned i;
    const unsigned n = MAX2(vertex->num_elements, nv50->state.num_vtxelts);
 
-   /* A vertexid is not generated for inline data uploads. Have to use a
-    * VBO. This check must come after the vertprog has been validated,
-    * otherwise vertexid may be unset.
-    */
-   assert(nv50->vertprog->translated);
-   if (nv50->vertprog->vp.vertexid)
-      nv50->vbo_push_hint = 0;
-
    if (unlikely(vertex->need_conversion))
       nv50->vbo_fifo = ~0;
    else
@@ -326,6 +317,7 @@ nv50_vertex_arrays_validate(struct nv50_context *nv50)
          if (buf && buf->status & NOUVEAU_BUFFER_STATUS_GPU_WRITING) {
             buf->status &= ~NOUVEAU_BUFFER_STATUS_GPU_WRITING;
             nv50->base.vbo_dirty = true;
+            break;
          }
       }
    }
@@ -744,8 +736,9 @@ nva0_draw_stream_output(struct nv50_context *nv50,
       BEGIN_NV04(push, NVA0_3D(DRAW_TFB_BASE), 1);
       PUSH_DATA (push, 0);
       BEGIN_NV04(push, NVA0_3D(DRAW_TFB_STRIDE), 1);
-      PUSH_DATA (push, so->stride);
-      nv50_query_pushbuf_submit(push, NVA0_3D_DRAW_TFB_BYTES, so->pq, 0x4);
+      PUSH_DATA (push, 0);
+      BEGIN_NV04(push, NVA0_3D(DRAW_TFB_BYTES), 1);
+      nv50_query_pushbuf_submit(push, so->pq, 0x4);
       BEGIN_NV04(push, NV50_3D(VERTEX_END_GL), 1);
       PUSH_DATA (push, 0);
 
@@ -768,7 +761,6 @@ nv50_draw_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
 {
    struct nv50_context *nv50 = nv50_context(pipe);
    struct nouveau_pushbuf *push = nv50->base.pushbuf;
-   bool tex_dirty = false;
    int i, s;
 
    /* NOTE: caller must ensure that (min_index + index_bias) is >= 0 */
@@ -798,9 +790,6 @@ nv50_draw_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
 
    push->kick_notify = nv50_draw_vbo_kick_notify;
 
-   /* TODO: Instead of iterating over all the buffer resources looking for
-    * coherent buffers, keep track of a context-wide count.
-    */
    for (s = 0; s < 3 && !nv50->cb_dirty; ++s) {
       uint32_t valid = nv50->constbuf_valid[s];
 
@@ -828,21 +817,6 @@ nv50_draw_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
       nv50->cb_dirty = false;
    }
 
-   for (s = 0; s < 3 && !tex_dirty; ++s) {
-      for (i = 0; i < nv50->num_textures[s] && !tex_dirty; ++i) {
-         if (!nv50->textures[s][i] ||
-             nv50->textures[s][i]->texture->target != PIPE_BUFFER)
-            continue;
-         if (nv50->textures[s][i]->texture->flags &
-             PIPE_RESOURCE_FLAG_MAP_COHERENT)
-            tex_dirty = true;
-      }
-   }
-   if (tex_dirty) {
-      BEGIN_NV04(push, NV50_3D(TEX_CACHE_CTL), 1);
-      PUSH_DATA (push, 0x20);
-   }
-
    if (nv50->vbo_fifo) {
       nv50_push_vbo(nv50, info);
       push->kick_notify = nv50_default_kick_notify;
@@ -864,6 +838,10 @@ nv50_draw_vbo(struct pipe_context *pipe, const struct pipe_draw_info *info)
          nv50->base.vbo_dirty = true;
    }
 
+   if (!nv50->base.vbo_dirty && nv50->idxbuf.buffer &&
+       nv50->idxbuf.buffer->flags & PIPE_RESOURCE_FLAG_MAP_COHERENT)
+      nv50->base.vbo_dirty = true;
+
    if (nv50->base.vbo_dirty) {
       BEGIN_NV04(push, NV50_3D(VERTEX_ARRAY_FLUSH), 1);
       PUSH_DATA (push, 0);

src/gallium/drivers/nouveau/nvc0/nvc0_context.c

@@ -262,7 +262,7 @@ nvc0_context_get_sample_position(struct pipe_context *, unsigned, unsigned,
                                  float *);
 
 struct pipe_context *
-nvc0_create(struct pipe_screen *pscreen, void *priv)
+nvc0_create(struct pipe_screen *pscreen, void *priv, unsigned ctxflags)
 {
    struct nvc0_screen *screen = nvc0_screen(pscreen);
    struct nvc0_context *nvc0;