llvm-project/llvm/lib
Marek Olsak 52b033b827 AMDGPU: enable 128-bit for local addr space under an option
Author: Samuel Pitoiset

ds_read_b128 and ds_write_b128 have been recently enabled
under the amdgpu-ds128 option because the performance benefit
is unclear.

Though, using 128-bit loads/stores for the local address space
appears to introduce regressions in tessellation shaders. Not
sure what is broken, but as ds_read_b128/ds_write_b128 are not
enabled by default, just introduce a global option and enable
128-bit only if requested (until it's fixed/used correctly).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105464
llvm-svn: 329591
2018-04-09 16:56:32 +00:00
..
Analysis [NFC] Loosen restriction on preheader to fix buildbot 2018-04-06 07:23:45 +00:00
AsmParser Add the ShadowCallStack attribute 2018-04-03 20:10:40 +00:00
BinaryFormat Fix build broken by r328090 2018-03-21 12:18:03 +00:00
Bitcode [Bitcode] Change std::sort to llvm::sort in response to r327219 2018-04-05 19:27:04 +00:00
CodeGen Support generic expansion of ordered vector reduction (PR36732) 2018-04-09 15:44:20 +00:00
DebugInfo [NFC] fix trivial typos in comments and error message 2018-04-09 04:37:53 +00:00
Demangle [demangler] Fix a bug in r328464 found by oss-fuzz. 2018-03-26 15:34:36 +00:00
ExecutionEngine [RuntimeDyld][PowerPC] Use global entry points for calls between sections. 2018-04-05 19:37:05 +00:00
FuzzMutate [llvm-opt-fuzzer] Add irce to the fuzzing options 2018-03-20 11:32:13 +00:00
Fuzzer [libFuzzer] Delete llvm/lib/Fuzzer 2017-10-16 20:48:19 +00:00
IR [X86] Merge some of the autoupgrade handling for masked intrinsics that just need to upgrade to an unmasked version plus a select. NFCI 2018-04-09 06:15:09 +00:00
IRReader LLParser: add an argument for overriding data layout and do not check alloca addr space 2018-01-30 22:32:39 +00:00
LTO Prevent data races in concurrent ThinLTO processes. 2018-03-30 21:35:42 +00:00
LineEditor
Linker [ThinLTO] Recommit of import global variables 2018-03-12 10:30:50 +00:00
MC [Support] Make line-number cache robust against access patterns. 2018-04-07 00:44:02 +00:00
Object [llvm-ar] Support multiple dashed options 2018-03-28 17:21:14 +00:00
ObjectYAML Reland "[DebugInfo] Support DWARF expressions in eh_frame" 2018-03-08 00:46:53 +00:00
Option [NFC] Replace iterators in PrintHelp with range-based for 2018-03-12 18:31:07 +00:00
Passes Re-land r329273: [Plugins] Add a slim plugin API to work together with the new PM 2018-04-05 15:04:13 +00:00
ProfileData Revert "Revert "[InstrProf] Support for external functions in text format."" 2018-03-22 21:26:52 +00:00
Support [NFC] fix trivial typos in comments and error message 2018-04-09 04:37:53 +00:00
TableGen [TableGen] Change std::sort to llvm::sort in response to r327219 2018-04-06 20:18:05 +00:00
Target AMDGPU: enable 128-bit for local addr space under an option 2018-04-09 16:56:32 +00:00
Testing [LLVMTestingSupport] Add explicit linkage to LLVMSupport 2018-04-08 06:49:17 +00:00
ToolDrivers [COFF] Keep the underscore on exported decorated stdcall functions in MSVC mode 2018-01-20 11:44:32 +00:00
Transforms Support generic expansion of ordered vector reduction (PR36732) 2018-04-09 15:44:20 +00:00
WindowsManifest Convert line endings of lib/WindowsManifest/CMakeLists.txt to unix. 2018-04-07 04:28:08 +00:00
XRay [XRay] clarify error messages when parsing broken traces 2018-01-30 13:41:34 +00:00
CMakeLists.txt
LLVMBuild.txt