master
Add attributes to functions pass (#539)
Update LLVM (#680)
Fix warning in GetArgKindForPodArgs (#699)
Fix local arg spec id assignment (#578)
Support global type-mangled pod args in push constants (#588)
Fix compilation for 32-bit systems (#717)
Lowers OpenCL builtin functions (#678)
Add support for OpenCL 2.0 atomic functions (#657)
Don't require inlining to use UBOs (#338)
A few fixes to SPIR-V 1.4 support (#748)
Update LLVM (#758)
Update LLVM (#752)
Remove unused variables (#569)
Fix generation of selection merges (#449)
Update LLVM (#737)
Re-implement clspv-opt as a standalone driver (#346)
Builtins refactor - phase3 (#598)
Update uses of LLVM (#601)
Add support for get_work_dim() and global offset (#514)
Update LLVM (#746)
Lower kernel parameters (#694)
Refactor the driver. (#236)
Update llvm (#563)
Update LLVM (#639)
Automate pod arg implementation decision (#575)
Ignore metadata types on LongVector support (#741)
Update llvm (#627)
Add an LLVM IR library for more builtin functions (#715)
Normalize constant global variables (#366)
Fix type for --max-ubo-size and --max-pushconstant-size options (#756)
Support SPIR-V 1.4 and 1.5 (#740)
Embedded reflection (#612)
Update deps (#523)
SPIR-V 1.4 fixes (#755)
Update LLVM dep (#441)
Add support for llvm.{smax,smin} intrinsics (#735)
Use distinct nodes for clspv.next_spec_constant_id (#696)
Add generic support to allocate spec ids (#562)
Add a pass to strip freeze instructions (#559)
Read write images (#643)
Update LLVM (#711)
Update LLVM (#628)
Unsampled image reads (#465)
Update llvm (#589)