ampere-computing/clang.git - clang including Ampere Computing toolchain specific patches

Age	Commit message (Expand)	Author
2016-06-30	[CUDA] Give templated device functions internal linkage, templated kernels ex...	Justin Lebar
2016-05-02	[CUDA] Make sure device-side __global__ functions are always visible.	Artem Belevich
2015-09-23	[CUDA] __global__ functions should always be visible externally.	Artem Belevich
2015-09-22	[CUDA] Add implicit __attribute__((used)) to all __global__ functions.	Artem Belevich
2015-08-11	Revert "[CUDA] Add implicit __attribute__((used)) to all __global__ functions."	Daniel Jasper
2015-08-10	[CUDA] Add implicit __attribute__((used)) to all __global__ functions.	Artem Belevich
2014-12-15	IR: Make metadata typeless in assembly, clang side	Duncan P. N. Exon Smith
2014-04-28	Move all CUDA testing inputs to Inputs/ subdirectory inside the tests.	Eli Bendersky
2013-08-15	CHECK-LABEL-ify some code gen tests to improve diagnostic experience when tes...	Stephen Lin
2013-03-30	Use kernel metadata to differentiate between kernel and device	Justin Holewinski
2012-05-24	Replace PTX back-end with NVPTX back-end in all places where Clang cares	Justin Holewinski
2011-10-06	CUDA: add -fcuda-is-device flag	Peter Collingbourne
2011-10-06	CUDA: set proper calling conventions for PTX	Peter Collingbourne