Featured content is also available via:

Graphics

The Graphics channel is devoted to topics concerning Linux/BSD open-source graphics (GPU) hardware drivers, graphics rendering APIs, display servers, and related graphics/display projects.

Beignet

Beignet is an open-source OpenCL implementation for Intel graphics cores on Linux.

Cairo

Cairo is a vector graphics drawing API that offers multiple hardware and software-accelerated back-ends. Cairo is used by a variety of projects from X.Org to Mozilla and WebKit.

GLVND

The libglvnd project by NVIDIA is an attempt to create an OpenGL venture-neutral dispatch library for Linux systems. Advertised as the new Linux OpenGL ABI, GLVND is designed to make it easier for multiple GPU drivers to co-exist on the same system.

Mesa 3D Graphics Library

This sub-channel is devoted to everything Mesa, including Gallium3D drivers.

Pixman

Pixman is a pixel manipulation and rendering library.

POCL

POCL is the Portable Computing Language and most notably provides an open-source OpenCL library implementation.

Video Acceleration

Projects concerning video acceleration and video playback on graphics adapters are found within this sub-channel.

Wayland

Wayland and its related components are tracked here.

X.Org

The X.Org Server projects and other X.Org projects are tracked via this sub-channel.

Yesterday's Activity

iris: Cache assembly shaders in the on-disk shader cache
Mesa - Dylan Baker: This implements storing and retrieving iris_compiled_shader objects from the on-disk shader cache.

iris: Expose the disk cache to the state tracker as well
Mesa - Kenneth Graunke: This lets st/nir cache the NIR for shaders, based on the shader source string hash, allowing us to skip initial compiles altogether, and also letting us start from there should we need to recompile for NOS.

This Week's Activity

man: Add weston-bindings(7) describing desktop shell shortcuts
Weston - Manuel Stoeckl: The desktop-shell-specific bindings were identified from shell_add_bindings in desktop-shell/shell.c.

lima/gpir: support vector ssa load
Mesa - Qiang Yu: Some vector sysval can't be lowered to scaler, so need to break it to scaler in nir to gpir convertion.

panfrost: Improve fixed-function blending
Mesa - Alyssa Rosenzweig: This fixes a few miscellaneous issues with the fixed-function blending programming, though it is far from complete.

panfrost/nir: Add nir_lower_blend pass
Mesa - Alyssa Rosenzweig: This new lowering pass implements the OpenGL ES blend pipeline in shaders, applicable to hardware lacking full-featured blending hardware (including Midgard/Bifrost and vc4).

radv: add a workaround for Monster Hunter World and LLVM 7&8
Mesa - Samuel Pitoiset: The load/store optimizer pass doesn't handle WaW hazards correctly and this is the root cause of the reflection issue with Monster Hunter World.

svga: Add an environment variable to force coherent surface memory
Mesa - Thomas Hellstrom: The vmwgfx driver supports emulated coherent surface memory as of version 2.16.

winsys/radeon: implement ctx_query_reset_status by copying radeonsi
Mesa - Marek Olšák: To make it behave like amdgpu.

radeonsi: cull primitives with async compute for large draw calls
Mesa - Marek Olšák

winsys/amdgpu: add a parallel compute IB coupled with a gfx IB
Mesa - Marek Olšák

ac: add LLVM code for triangle culling
Mesa - Marek Olšák

This Month's Activity

intel/compiler: Implement TCS 8_PATCH mode and INTEL_DEBUG=tcs8
Mesa - Kenneth Graunke: Our tessellation control shaders can be dispatched in several modes. - SINGLE_PATCH (Gen7+) processes a single patch per thread, with each channel corresponding to a different patch vertex.

nir/algebraic: Commute 1-fsat(a) to fsat(1-a) for all non-fmul instructions
Mesa - Ian Romanick: The goal is to avoid having an extra MOV instruction to perform the saturate.

nir/algebraic: Push unary operations into source operands of fsat source
Mesa - Ian Romanick: Pushing a unary operation, like fneg, into the operation that generates its operand allows the fsat to be applied to the inner instruction instead of on a separate instruction that performs the unary operation.

nir/algebraic: Recognize open-coded fsat with modifiers
Mesa - Ian Romanick: This change also enables a later change (nir/algebraic: Replace 1-fsat(a) with fsat(1-a)) to affect more shaders. Almost all of the affected shaders are in Bioshock Infinite, and all of those shaders all require GLSL 4.10. All Intel platforms had similar results.

nir/algebraic: Eliminate a tautological compare
Mesa - Ian Romanick: The value-range tracking pass that is coming is not clever enough to know that the result of the ffma must be non-negative.

intel/fs: Allow cmod propagation to instructions with saturate modifier
Mesa - Ian Romanick: v2: Add unit tests.

intel/fs: Stop doing extra RA calls
Mesa - Jason Ekstrand: In the last phase of the schedule and RA loop, the RA call is redundant if we spill.

intel/fs/ra: Only add dest interference to sources that exist
Mesa - Jason Ekstrand

intel/fs/ra: Only add MRF hack interference if we're spilling
Mesa - Jason Ekstrand: The only use of the MRF hack these days is for spilling and there we don't need the precise MRF usage information.

intel/fs/ra: Spill without destroying the interference graph
Mesa - Jason Ekstrand: Instead of re-building the interference graph every time we spill, we modify it in place so we can avoid recalculating liveness and the whole O(n^2) interference graph building process.

Recent Releases

radv: consider MESA_VK_VERSION_OVERRIDE when setting the api version
Mesa - Eleni Maria Stea: Before setting the physical device API version, we should check if the MESA_VK_VERSION_OVERRIDE environment variable is set and take it into account.

Pre-release version bump to 0.38.4
Pixman - Matt Turner

simplify LLVM version string printing
Mesa - Eric Engestrom: Figure it out once in the build system, then just use that all over the place.

configure.ac/meson.build: bump version to 6.0.0 for the official release
Weston - Derek Foreman

configure.ac: bump to version 1.17.0 for the official release
Wayland - Derek Foreman

configure.ac/meson.build: bump to version 5.0.94 for the RC2 release
Weston - Derek Foreman

configure.ac: bump version to 1.16.93 for the RC1 release
Wayland - Derek Foreman

configure.ac: bump version to 5.0.93 for the RC1 release
Weston - Derek Foreman

Bump version for 19.0.0 release
xf86-video-ati - Michel Dänzer

configure.ac: bump to version 1.16.92 for the beta release
Wayland - Derek Foreman

Popular Activity In The Past Month

radv: Add support for driconf
Mesa - Bas Nieuwenhuizen

v3d: Re-add support for memory_barrier_shared
Mesa - Eric Anholt

nir: add rcp(w) lowering for gl_FragCoord
Mesa - Andreas Baierl

anv: add support for VK_EXT_memory_budget
Mesa - Eric Engestrom

radv: enable descriptor indexing capabilities
Mesa - Juan A. Suarez Romero

iris: Some tidying for preemption support
Mesa - Kenneth Graunke

radeonsi: add radeonsi_sync_compile option
Mesa - Nicolai Hähnle

radeonsi: remove dirty slot masks from scissor and viewport states
Mesa - Marek Olšák

iris: Replace buffer backing storage and rebind to update addresses
Mesa - Kenneth Graunke

radv: Add support for icd loader interface v4.
Mesa - Bas Nieuwenhuizen

panfrost: Adds Bifrost shader disassembler utility
Mesa - Ryan Houdek

iris: enable preemption support for gen10
Mesa - Mike Blumenkrantz

gallium: add ATC format support
Mesa - Jonathan Marek

anv: enable descriptor indexing capabilities
Mesa - Juan A. Suarez Romero

nir: add pass to lower fb reads
Mesa - Rob Clark

iris: Set XY Clipping correctly
Mesa - Kenneth Graunke

Legal Disclaimer & Disclosure | Contact Us
Copyright © 2010 - 2019 by Phoronix Media. All rights reserved.
All trademarks and copyrights used are properties of their respective owners.
Powered by the Phoronix Content Management System (PHXCMS-8).