commit 931d3de6c1c903cfd47842bc5026a9294ac492b4
Author: Siu Chi Chan <siuchi.chan@amd.com>
Date: Mon Feb 13 10:54:06 2017 -0500
only force to libstdc++ if the g++ is older than version 5 since hcc already defaults to libstdc++ with newer g++
commit 1ef8d71aa788de7b9eead4906fe56186f06d7d3f
Author: scchan <siuchi.chan@amd.com>
Date: Sun Feb 12 14:44:11 2017 -0500
remove hardcoded -lc++ in tests
commit 5d99ef338eb3a66523cc9ddd139e86c6fd707b9c
Author: scchan <siuchi.chan@amd.com>
Date: Sun Feb 12 14:35:46 2017 -0500
force include libstdc++ headers and stdc++ only if g++ version < 5 since hcc uses libstdc++ by default if g++ > 5 is present
commit a2bc21b24d100feefe91cd3cb2271238bda0738a
Author: scchan <siuchi.chan@amd.com>
Date: Fri Feb 10 04:36:27 2017 -0500
use hcc-config to generate compiler and linker flags
Change-Id: I13a79629c0adfd75439a47d0488ff4fd619c55ba
(cherry picked from commit 866e744e40cf1378af8a679b54959bf5226c340f)
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/65
[Bug]
Access violation in cudaStructVar matcher.
StringRef name = structVar->getType()
->getAsStructureType()
->getDecl() // <--- Access Violation
->getNameAsString();
[Solution]
Add isStructureType() check before getting type as Struct.
[ToDo]
Find case-studies for that StructVar matcher with types other than Struct.
CUstream_st -> hipStream_t *
CUevent_st -> hipEvent_t * (fixed, was -> hipEvent_t)
Warning:
[HIPIFY] warning: the following reference is not handled: 'CUstream_st' [struct var ptr]. (thrust/examples/cuda/async_reduce.cu)
[Explanation]
Narrow cudaKernelCallExpr matcher to isExpansionInMainFile() - to the file being actually converted (without system includes).
But as for Thrust, there are *.inl files, which are nevertheless being expanded in main file.
These files contain cuda kernel launches, which are not converted successfully by hipify, for instance:
thrust/system/cuda/detail/detail/launch_closure.inl:98:23.
[ToDo]
File a bug on cudaKernelCallExpr matcher for conversion error (thrust\examples\cuda\<any cu example>).
1. Changed test to assert for same hipFunction values
2. Added better memory management for hipModule
Change-Id: I10d7aef13c215a2211e262f3c79017f26a17d9a7
1. Split hip_ir.ll to hip_hc.ll and hip_hc_gfx803.ll
a. hip_hc.ll contains arch generic ir implementations
b. hip_hc_gfx803.ll contains gfx803 (fiji, polaris) specific ir
2. HIPCC can now parse --amdgpu-target=*.
a. Usage: hipcc --amdgpu-target=gfx803 --amdgpu-target=gfx701
b. TODO: Convert to --amdgpu-target=gfx803,gfx701
3. With LLC in HCC able to generate native f16 isa, removed inline half asm math ops
4. Fixed threadfence and threadfence_block to use functions in rocdl
Change-Id: Ic9a9e3e04139b0d75d2c2a263c030ca77adc1019
Evgeny Mankov
Siu Chi Chan
Maneesh Gupta
Paul
Aditya Atluri
pensun
Rahul Garg
Ben Sander