| 1 | Mesa 17.1.5 Release Notes / July 14, 2017
|
|---|
| 2 | =========================================
|
|---|
| 3 |
|
|---|
| 4 | Mesa 17.1.5 is a bug fix release which fixes bugs found since the 17.1.4
|
|---|
| 5 | release.
|
|---|
| 6 |
|
|---|
| 7 | Mesa 17.1.5 implements the OpenGL 4.5 API, but the version reported by
|
|---|
| 8 | glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
|
|---|
| 9 | glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being
|
|---|
| 10 | used. Some drivers don't support all the features required in OpenGL
|
|---|
| 11 | 4.5. OpenGL 4.5 is **only** available if requested at context creation
|
|---|
| 12 | because compatibility contexts are not supported.
|
|---|
| 13 |
|
|---|
| 14 | SHA256 checksums
|
|---|
| 15 | ----------------
|
|---|
| 16 |
|
|---|
| 17 | ::
|
|---|
| 18 |
|
|---|
| 19 | 7e3eeee8f9c28052796eb18133c2be12c38ba34864cc496382a2fa20c29b0317 mesa-17.1.5.tar.gz
|
|---|
| 20 | 378516b171712687aace4c7ea8b37c85895231d7a6d61e1e27362cf6034fded9 mesa-17.1.5.tar.xz
|
|---|
| 21 |
|
|---|
| 22 | New features
|
|---|
| 23 | ------------
|
|---|
| 24 |
|
|---|
| 25 | None
|
|---|
| 26 |
|
|---|
| 27 | Bug fixes
|
|---|
| 28 | ---------
|
|---|
| 29 |
|
|---|
| 30 | - `Bug 100242 <https://bugs.freedesktop.org/show_bug.cgi?id=100242>`__
|
|---|
| 31 | - radeon buffer allocation failure during startup of Factorio
|
|---|
| 32 | - `Bug 101657 <https://bugs.freedesktop.org/show_bug.cgi?id=101657>`__
|
|---|
| 33 | - strtod.c:32:10: fatal error: xlocale.h: No such file or directory
|
|---|
| 34 | - `Bug 101666 <https://bugs.freedesktop.org/show_bug.cgi?id=101666>`__
|
|---|
| 35 | - bitfieldExtract is marked as a built-in function on OpenGL ES 3.0,
|
|---|
| 36 | but was added in OpenGL ES 3.1
|
|---|
| 37 | - `Bug 101703 <https://bugs.freedesktop.org/show_bug.cgi?id=101703>`__
|
|---|
| 38 | - No stencil buffer allocated when requested by GLUT
|
|---|
| 39 |
|
|---|
| 40 | Changes
|
|---|
| 41 | -------
|
|---|
| 42 |
|
|---|
| 43 | Aaron Watry (1):
|
|---|
| 44 |
|
|---|
| 45 | - radeon/winsys: Limit max allocation size to 70% of VRAM
|
|---|
| 46 |
|
|---|
| 47 | Aleksander Morgado (2):
|
|---|
| 48 |
|
|---|
| 49 | - etnaviv: fix refcnt initialization in etna_screen
|
|---|
| 50 | - etnaviv: don't dereference etna_resource pointer if allocation fails
|
|---|
| 51 |
|
|---|
| 52 | Alex Smith (2):
|
|---|
| 53 |
|
|---|
| 54 | - ac/nir: Use correct LLVM intrinsics for atomic ops on imageBuffers
|
|---|
| 55 | - ac/nir: Fix ordering of parameters for image atomic cmpswap
|
|---|
| 56 | intrinsics
|
|---|
| 57 |
|
|---|
| 58 | Andres Gomez (3):
|
|---|
| 59 |
|
|---|
| 60 | - docs: add sha256 checksums for 17.1.4
|
|---|
| 61 | - cherry-ignore: i965: Fix anisotropic filtering for mag filter
|
|---|
| 62 | - Update version to 17.1.5
|
|---|
| 63 |
|
|---|
| 64 | Anuj Phogat (2):
|
|---|
| 65 |
|
|---|
| 66 | - intel/isl: Use uint64_t to store total surface size
|
|---|
| 67 | - intel/isl: Add the maximum surface size limit
|
|---|
| 68 |
|
|---|
| 69 | Brian Paul (3):
|
|---|
| 70 |
|
|---|
| 71 | - draw: check for line_width != 1.0f in validate_pipeline()
|
|---|
| 72 | - svga: clamp device line width to at least 1 to fix HWv8 line
|
|---|
| 73 | stippling
|
|---|
| 74 | - svga: fix PIPE_CAP_MAX_TEXTURE_BUFFER_SIZE value
|
|---|
| 75 |
|
|---|
| 76 | Bruce Cherniak (1):
|
|---|
| 77 |
|
|---|
| 78 | - swr: Limit memory held by defer deleted resources.
|
|---|
| 79 |
|
|---|
| 80 | Chandu Babu N (1):
|
|---|
| 81 |
|
|---|
| 82 | - st/va: Fix leak in VAAPI subpictures
|
|---|
| 83 |
|
|---|
| 84 | Charmaine Lee (1):
|
|---|
| 85 |
|
|---|
| 86 | - svga: fixed surface size to include array size
|
|---|
| 87 |
|
|---|
| 88 | Connor Abbott (2):
|
|---|
| 89 |
|
|---|
| 90 | - spirv: fix OpBitcast when the src and dst bitsize are different (v3)
|
|---|
| 91 | - ac/nir: implement 64-bit packing and unpacking
|
|---|
| 92 |
|
|---|
| 93 | Iago Toral Quiroga (1):
|
|---|
| 94 |
|
|---|
| 95 | - glsl: gl_Max{Vertex,Fragment}UniformComponents exist in all desktop
|
|---|
| 96 | GL versions
|
|---|
| 97 |
|
|---|
| 98 | Ilia Mirkin (1):
|
|---|
| 99 |
|
|---|
| 100 | - glsl: check if any of the named builtins are available first
|
|---|
| 101 |
|
|---|
| 102 | James Legg (2):
|
|---|
| 103 |
|
|---|
| 104 | - ac/nir: Make intrinsic_name buffer long enough
|
|---|
| 105 | - spirv: Fix reaching unreachable for compare exchange on images
|
|---|
| 106 |
|
|---|
| 107 | Faith Ekstrand (1):
|
|---|
| 108 |
|
|---|
| 109 | - nir/spirv: Use the type from the deref for atomics
|
|---|
| 110 |
|
|---|
| 111 | Juan A. Suarez Romero (1):
|
|---|
| 112 |
|
|---|
| 113 | - glsl: do not call link_xfb_stride_layout_qualifiers() for fragment
|
|---|
| 114 | shaders
|
|---|
| 115 |
|
|---|
| 116 | Kenneth Graunke (2):
|
|---|
| 117 |
|
|---|
| 118 | - i965: Use true AA line distance on G45/Ironlake.
|
|---|
| 119 | - i965: Always set AALINEDISTANCE_TRUE on Sandybridge.
|
|---|
| 120 |
|
|---|
| 121 | Lucas Stach (1):
|
|---|
| 122 |
|
|---|
| 123 | - etnaviv: fix shader miscompilation with more than 16 labels
|
|---|
| 124 |
|
|---|
| 125 | Marek Olšák (1):
|
|---|
| 126 |
|
|---|
| 127 | - gallium/radeon: fix a possible crash for buffer exports
|
|---|
| 128 |
|
|---|
| 129 | Neha Bhende (1):
|
|---|
| 130 |
|
|---|
| 131 | - svga: loop over box.depth for ReadBack_image on each slice
|
|---|
| 132 |
|
|---|
| 133 | Nicolai Hähnle (1):
|
|---|
| 134 |
|
|---|
| 135 | - winsys/radeon: only call pb_slabs_reclaim when slabs are actually
|
|---|
| 136 | used
|
|---|
| 137 |
|
|---|
| 138 | Olivier Lauffenburger (1):
|
|---|
| 139 |
|
|---|
| 140 | - st/wgl: improve selection of pixel format
|
|---|
| 141 |
|
|---|
| 142 | Philipp Zabel (1):
|
|---|
| 143 |
|
|---|
| 144 | - st/mesa: release EGLImage on EGLImageTarget\* error
|
|---|
| 145 |
|
|---|
| 146 | Plamena Manolova (1):
|
|---|
| 147 |
|
|---|
| 148 | - mesa/main: Move NULL pointer check.
|
|---|
| 149 |
|
|---|
| 150 | Tim Rowley (2):
|
|---|
| 151 |
|
|---|
| 152 | - swr/rast: \_mm*_undefined_\* implementations for gcc<4.9
|
|---|
| 153 | - swr/rast: Correctly allocate SWR_STATS memory as cacheline aligned
|
|---|
| 154 |
|
|---|
| 155 | Tomasz Figa (1):
|
|---|
| 156 |
|
|---|
| 157 | - intel: common: Fix link failure with standalone Android build
|
|---|
| 158 |
|
|---|
| 159 | Vinson Lee (1):
|
|---|
| 160 |
|
|---|
| 161 | - scons: Check for xlocale.h before defining HAVE_XLOCALE_H.
|
|---|
