Mesa 17.1.6 Release Notes / August 7, 2017
==========================================

Mesa 17.1.6 is a bug fix release which fixes bugs found since the 17.1.5
release.

Mesa 17.1.6 implements the OpenGL 4.5 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
4.5. OpenGL 4.5 is **only** available if requested at context creation
because compatibility contexts are not supported.

SHA256 checksums
----------------

::

   971831bc1e748b3e8367eee6b9eb509bad2970e3c2f8520ad25f5caa12ca5491  mesa-17.1.6.tar.gz
   0686deadde1f126b20aa67e47e8c50502043eee4ecdf60d5009ffda3cebfee50  mesa-17.1.6.tar.xz

New features
------------

None

Bug fixes
---------

-  `Bug 97957 <https://bugs.freedesktop.org/show_bug.cgi?id=97957>`__ -
   Awful screen tearing in a separate X server with DRI3
-  `Bug 101683 <https://bugs.freedesktop.org/show_bug.cgi?id=101683>`__
   - Some games hang while loading when compositing is shut off or
   absent
-  `Bug 101867 <https://bugs.freedesktop.org/show_bug.cgi?id=101867>`__
   - Launch options window renders black in Feral Games in current Mesa
   trunk

Changes
-------

Andres Gomez (1):

-  docs: add sha256 checksums for 17.1.5

Bas Nieuwenhuizen (1):

-  radv: Don't underflow non-visible VRAM size.

Brian Paul (1):

-  svga: fix texture swizzle writemasking

Chad Versace (1):

-  anv/image: Fix VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT

Chris Wilson (1):

-  i965: Resolve framebuffers before signaling the fence

Connor Abbott (1):

-  nir: fix algebraic optimizations

Daniel Stone (1):

-  st/dri: Check get-handle return value in queryImage

Dave Airlie (5):

-  radv: fix non-0 based layer clears.
-  radv: fix buffer views on SI/CIK.
-  radv/ac: realign SI workaround with radeonsi.
-  radv/ac: port SI TC L1 write corruption fix.
-  radv: for stencil only set Z tile mode index to same value

Emil Velikov (23):

-  cherry-ignore: add "anv: Round u_vector element sizes to a power of
   two"
-  anv: advertise v6 of the wayland surface extension
-  radv: advertise v6 of the wayland surface extension
-  swrast: add dri2ConfigQueryExtension to the correct extension list
-  cherry-ignore: add "anv: Transition MCS buffers from the undefined
   layout"
-  swr: don't forget to link AVX/AVX2 against pthreads
-  cherry-ignore: add "i965: Fix offset addition in get_isl_surf"
-  cherry-ignore: add "i965: Fix = vs == in MCS aux usage assert."
-  cherry-ignore: add a couple of radeon commits
-  cherry-ignore: add "swr/rast: non-regex knob fallback code for gcc <
   4.9"
-  cherry-ignore: add "swr: fix transform feedback logic"
-  cherry-ignore: add a couple of radeonsi/gfx9 commits
-  cherry-ignore: ignore reverted st/mesa commit
-  cherry-ignore: add bindless textures fix
-  cherry-ignore: add "st/glsl_to_tgsi: fix getting the image type for
   array of structs"
-  cherry-ignore: add yet another bindless textures fix
-  bin/cherry-ignore: add radeonsi "fix of a fix"
-  travis: lower SWR requirement to GCC 4.8, aka std=c++11
-  i965: use strtol to convert the integer deviceID override
-  swr: remove unneeded fallback strcasecmp define
-  cherry-ignore: add a bunch more commits to the list
-  fixup! cherry-ignore: add a bunch more commits to the list
-  Update version to 17.1.6

Eric Anholt (1):

-  broadcom/vc4: Prefer blit via rendering to the software fallback.

Eric Engestrom (1):

-  configure: only install khrplatform.h if needed

Iago Toral Quiroga (2):

-  anv/cmd_buffer: fix off by one error in assertion
-  anv: only expose up to 28 vertex attributes

Ilia Mirkin (1):

-  nv50/ir: fix threads calculation for non-compute shaders

Jason Ekstrand (5):

-  anv/cmd_buffer: Properly handle render passes with 0 attachments
-  anv: Stop leaking the no_aux sampler surface state
-  anv/image: Add INPUT_ATTACHMENT to the list of required usages
-  nir/vars_to_ssa: Handle missing struct members in foreach_deref_node
-  spirv: Fix SpvImageFormatR16ui

Juan A. Suarez Romero (2):

-  anv/pipeline: use unsigned long long constant to check enable vertex
   inputs
-  anv/pipeline: do not use BITFIELD64_BIT()

Kenneth Graunke (1):

-  nir: Use nir_src_copy instead of direct assignments.

Lionel Landwerlin (1):

-  i965: perf: flush batchbuffers at the beginning of queries

Lucas Stach (1):

-  etnaviv: fix memory leak when BO allocation fails

Marek Olšák (2):

-  st/mesa: always unconditionally revalidate main framebuffer after
   SwapBuffers
-  gallium/radeon: make S_FIXED function signed and move it to shared
   code

Mark Thompson (1):

-  st/va: Fix scaling list ordering for H.265

Nicolai Hähnle (4):

-  radeonsi/gfx9: fix crash building monolithic merged ES-GS shader
-  radeonsi: fix detection of DRAW_INDIRECT_MULTI on SI
-  radeonsi/gfx9: reduce max threads per block to 1024 on gfx9+
-  gallium/radeon: fix ARB_query_buffer_object conversion to boolean

Thomas Hellstrom (2):

-  loader/dri3: Use dri3_find_back in loader_dri3_swap_buffers_msc
-  dri3: Wait for all pending swapbuffers to be scheduled before
   touching the front

Tim Rowley (3):

-  gallium/util: fix nondeterministic avx512 detection
-  swr/rast: quit using linux-specific gettid()
-  swr/rast: fix scons gen_knobs.h dependency

Timothy Arceri (1):

-  nir: fix nir_opt_copy_prop_vars() for arrays of arrays

Wladimir J. van der Laan (1):

-  etnaviv: Clear lbl_usage array correctly