AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,975 Commits 3 Branches 0 Tags
a00a5e3101cd5667518510a9ba5ed1e12dc481cc
Commit Graph

12 Commits

Author SHA1 Message Date
Aditya Atluri ba45e02007 added architecture specific macros 
1. Added __HIP_ARCH_GFXNUM__
2. Usage, -D__HIP_ARCH_GFX803__=1

Change-Id: I68b3a85d62cfab3a45d2b7a70cb3518ab2565236


[ROCm/clr commit: c6c90f6f5a]
 2017-02-08 19:45:32 -06:00
Aditya Atluri 76a58521f5 fixed hipcc for new compiler flags 
Change-Id: I49ec059be20ff26b7482c84d91ab7a43826c6a8d


[ROCm/clr commit: 400d0d4f78]
 2017-02-08 14:06:01 -06:00
Aditya Atluri 248b42d3e3 include arch specific ir on fallback path 
Change-Id: Ib04996aae2c21eb73ef2a9f6305915e0caccd704


[ROCm/clr commit: 26b5e57cfd]
 2017-02-08 12:19:06 -06:00
Aditya Atluri 3937b398e9 changed __global__ attribute 
1. Moved around tests and added them to HIT

Change-Id: I5d75280c42a5af852670ebabc7305ee56721ec7b


[ROCm/clr commit: f009f3533d]
 2017-02-03 10:53:36 -06:00
Aditya Atluri d9845446ef Added type conversion intrinsics 
1. Added all type conversion intrinsics
2. NO TESTS have been added. (Will add in next commit)
3. Sanatized code in hip_runtime.h
4. Added passed() to hipTestHalf to make it pass on HIT

Change-Id: I0987963c802fc7ff4d7e07d7b88d86da35da53c9


[ROCm/clr commit: d496576b55]
 2017-01-16 12:10:05 -06:00
Aditya Atluri d71fc0e60a added half2 log, log10, exp, exp10 math functions 
1. Enabled tests for log, log10, exp, exp10 half2
2. h2rint is still disabled.

Change-Id: I01f6002f6992259919893c524c526db5ee09473a


[ROCm/clr commit: 5c5f5c1ad1]
 2017-01-13 13:26:10 -06:00
Aditya Atluri 4756cf7c16 added half2 math operations 
1. They use SDWA + LLVM IR
2. Added these functions to test
3. Need to do exp, exp10, log, log10, rint

Change-Id: I06176acc6cb8bb054495310531777406a41b54e4


[ROCm/clr commit: eff68c989a]
 2017-01-13 12:27:11 -06:00
Aditya Atluri b2973b97e2 added math functions for half 
1. Added math functions for half precision
2. HRCP is not available due to device code linking errors, will be enabled once it is fixed
3. Added math functions to half test file

Change-Id: Ie317ce70ef518a4fc3f27142143d01e0327f5df3


[ROCm/clr commit: fe38e9652b]
 2017-01-13 12:05:29 -06:00
Aditya Atluri d3fe56550e added half2 cmp and conv, data movement device functions 
1. Added half2 comparision functions
2. Added conversion and data movement half apis

Change-Id: Ia33c0e957d9deb1f2b7a8fde8e22168f4d41b88b


[ROCm/clr commit: 646f566bbf]
 2017-01-13 10:56:07 -06:00
Aditya Atluri a52591d117 added comparision device functions for fp16 
1. Added comparision device functions
2. Added test to check correct isa getting generated

Change-Id: I16732f5a1438bdce145f7bfcecd28198e3cc4b79


[ROCm/clr commit: 89998d436f]
 2017-01-12 14:52:14 -06:00
Aditya Atluri d1f7c4e048 added packed math fp16 native device functions 
1. Added SDWA implementation inside IR file
2. Added device functions to header + used them in test

Change-Id: Ib4e059a58eee201cc82438689e3e9bc5f9d26653


[ROCm/clr commit: eeef055469]
 2017-01-12 14:10:51 -06:00
Aditya Atluri 5317ffa9ec Started adding native half math library support 
1. Removed HIP_EXPERIMENTAL env variable so that device code will be accessed from LLVM IR
2. Removed soft support from headers and moved to hip_fp16.cpp
3. Added LLVM IR + inline asm to hip_ir.ll
4. Added test for fp16
5. Added barriers for hcc 3.5 and hcc 4.0 for half support
a. Which means, hcc 4.0 can parse __fp16 but hcc 3.5 cant
b. HCC 4.0 code is implemented now, hcc 3.5 will be added later

Change-Id: Ic37859b2688ebb02e168bab643d1882bf4727952


[ROCm/clr commit: c286bf6f8a]
 2017-01-12 11:30:20 -06:00