AI/rocm-systems
2
0
Форкнуть 0
Код Задачи Запросы на слияние Действия Пакеты Проекты Релизы Вики Активность
1 512 коммитов 3 Ветки 0 Теги
620bbbc58348ee3f98c7f28a7b0b39e104f09d27
Граф коммитов

106 Коммитов

Автор SHA1 Сообщение Дата
Sun, Peng db4bedeafc revert workaround for square sample and update doc on GGL 
Change-Id: I731c68ca4111e7dc2e45bef51c4cad2c23fc81f8


[ROCm/clr commit: 14f8c7cf74]
 2017-03-21 10:26:09 -05:00
pensun 9c9a9bb330 Initial integration with Alex' Generic Grid Launch 
Change-Id: I559afb80e9e39ec0d119bb3bf3b85ef9e448caf6


[ROCm/clr commit: ad882222b0]
 2017-03-17 14:59:34 -05:00
Aditya Atluri b4b87b8786 Added default module launch api functionality 
1. As in hipModuleLaunchKernel(..., kernelParams, nullptr); works with this commit
    2. Added headers AMDGPUPTNote.h, AMDGPURuntimeMetadata.h to do code object meta data parsing
    3. Changed CMake to look at llvm link libraries
    4. HIP developer should set env variable LLVM_HOME to remove link errors
    5. HIP depends on installed LLVM (not source, not build)
    6. Added sample to test out the feature
    7. Right now HCC does not support embedding metadata in code object. Use clang opencl
    8. Changed HIPCC to read LLVM_HOME env var
    9. New argument to CMake should be given -DLLVM_HOME=<where llvm 5.0 is installed>

Change-Id: Iba38194aa872d97cc2c90a8e5ff746c48055c868


[ROCm/clr commit: 9645b0e0dc]
 2017-03-17 13:11:34 -05:00
Maneesh Gupta ee45f273fd 4_shfl and 5_2dshfl samples are unsupported on gfx701 
Change-Id: I81eb880350f25e89573ba14c62b549c6c43f8c91


[ROCm/clr commit: f91583b294]
 2017-03-14 15:56:18 +05:30
Ben Sander 49d7ea94f5 Refactor registered memory calls. 
[ROCm/clr commit: b7acb85fa8]
 2017-03-11 09:18:27 -06:00
Ben Sander e1c95b083d Add first step to a "registerd" mode in hipBusBandwidth. 
[ROCm/clr commit: f23b5a1f90]
 2017-03-11 09:18:27 -06:00
Rahul Garg 4dd856eb67 Fix for HCSWAP-128, make 5_2dshfl cookbook sample only for fiji 
Change-Id: I8869c28151bca1bd47a053a2808e93a801d16d00


[ROCm/clr commit: d48943699d]
 2017-03-10 10:29:52 +05:30
Aditya Atluri 974cb587a8 make 4_shfl cookbook sample only for fiji 
1. __shfl is not supported on hawaii gfx701

Change-Id: Iac09f5d30ee0674b8f58a6e74ec5c49b02be32ad


[ROCm/clr commit: 7f4b24886f]
 2017-03-09 08:52:50 -06:00
Aditya Atluri 106030f2c6 added new field to hipDeviceProp_t structure gcnArch. 
1. It is an integer containing gfx values 701, 801, 802, 803
2. On NV path, it is zero

Change-Id: I2b4c7f48981d0214d8c6b1905d2cc85b16203419


[ROCm/clr commit: 9f575721aa]
 2017-03-07 11:24:32 -06:00
Aditya Atluri 810afc2a36 removed hipblas samples as it is not yet supported 
Change-Id: I354b710e652ce0d0413d670530ceb8b70f4993d5


[ROCm/clr commit: 56744c0730]
 2017-02-17 08:51:02 -06:00
Rahul Garg 1b11b60e9d Command scripts for latency measurements 
Change-Id: I8c28765a09fb0358447367939de524b12699a317


[ROCm/clr commit: 73b56e690f]
 2017-02-07 15:03:46 +05:30
Ben Sander 48bd62db9a Doc update - describe debug techniques 
Also tweak sample to remove unneeded HIP_KERNEL_NAME.
Comment update


[ROCm/clr commit: ca1cef4e06]
 2017-01-19 12:40:45 -06:00
Rahul Garg 707c31913d Fixed hipcommander default execution for HCSWAP-106 
Change-Id: I9fbd10dfaeeb4928b2ec23ceed131b5200a658f9


[ROCm/clr commit: aa3f278475]
 2017-01-19 15:04:32 +05:30
Ben Sander 1a58ccc3b4 Print limits on CUDA devices 
[ROCm/clr commit: ab07d9b4a7]
 2016-12-16 08:55:11 -06:00
Ben Sander 8353b8ed72 remove TODO file 
[ROCm/clr commit: a30b36afdc]
 2016-12-15 14:42:52 -06:00
Sandeep Kumar 4d78e20eed Fixes in Makefile of couple of samples 
- modified Makefile for hipblas_saxpy to replaced hcblas.so with hipblas.so as part of HCSWAP-100
- Resolved missing separator issue in peer2peer cookbook Makefile

Change-Id: I678fea267eee1481f02da09379339ed78d3f95f2


[ROCm/clr commit: 8b0ce6e81b]
 2016-12-14 16:27:14 +05:30
Ben Sander 68f1a627e0 Add more debug info 
[ROCm/clr commit: a504df955e]
 2016-11-26 08:56:02 -06:00
Sandeep Kumar b8eda419a8 fix_format 
Change-Id: I34e265de434263a11654e5deba044c3f21e86578


[ROCm/clr commit: 1fe6890ba3]
 2016-11-18 14:34:14 +05:30
Sandeep Kumar dc599cf2b8 Add p2p for cookbook 
Change-Id: Id2e77ab31123ef95885d665efe34bc0d4596733a
(cherry picked from commit 6fbd0352713ca36e399b1ed4f17c486207a53875)


[ROCm/clr commit: 39e1b16d0b]
 2016-11-14 06:10:36 +05:30
Maneesh Gupta 99678b0000 hcc_dialects/Makefile: use clamp-config 
Change-Id: I86df82f75b75125825e22d0545209a19386d9936


[ROCm/clr commit: 052a580d5b]
 2016-11-10 11:31:50 +05:30
Ben Sander 445f888d97 Update gitignore for some common output files 
Change-Id: I9cd60f042af4dba07fe0fdbd2ee442936ff8c7bd


[ROCm/clr commit: 0e5cfed3eb]
 2016-11-06 04:26:15 -06:00
Ben Sander 43723d77cc Print non-peers too 
Change-Id: I2a6905edcdf144aa732ae3120c17780477f232ac


[ROCm/clr commit: 44aee4b61c]
 2016-11-04 06:34:07 -05:00
Ben Sander 4378a14789 Add initial hipProfileStart/Stop 
And modify sample to show how to use.
Still needs some work to understand interaction with CXL.

Change-Id: I2579824d2dd7863ea23874d34f0dabb3cb305d3e


[ROCm/clr commit: 18dbafe6e8]
 2016-10-27 23:09:32 -05:00
Ben Sander 04b77f0688 show how to use variety of HIP_PROFILE features 
Change-Id: I6edd66ac4c068b64e1dc3787d7f1f69ab3238469


[ROCm/clr commit: 08c08d6f9c]
 2016-10-27 23:09:32 -05:00
Ben Sander 9b2a6652c4 Add new hipdemangleatp and snapshot sample update for new functionality 
Change-Id: Ie19c683b2b0bdfeb0c3fcf89444c2e21b7f606e7


[ROCm/clr commit: 5b7da93e61]
 2016-10-27 23:09:31 -05:00
Ben Sander e0a3b2472c Rename HIP_ATP_MARKER and profiling vars 
HIP_PROFILE_API
HIP_DB_START_API
HIP_DB_STOP_API

Change-Id: I6c4da67212ff8217e6356a2622d4c6278a188c34


[ROCm/clr commit: f7c42efa3f]
 2016-10-27 23:09:31 -05:00
Sandeep Kumar e41aa1483a Add more apps to 2_Cookbook 
Change-Id: Iafe462df9726a32f450bd240a2de3eaa73a10057


[ROCm/clr commit: d025ed980b]
 2016-10-19 12:58:16 -04:00
Maneesh Gupta dbfd285aae samples: Updated copyright header 
Change-Id: I821f514ced5e34d492cb167b65d7273e26ed7b84


[ROCm/clr commit: 945a09c241]
 2016-10-15 23:05:04 +05:30
Ben Sander 5fb10e5742 Refactor module API test. 
- Add PASSED/FAIL indication.
- Set args using struct rather than void* array.

Change-Id: Ic924f88c49cc46979b12b7fef8650081e3b5f58c


[ROCm/clr commit: 4668a08783]
 2016-10-14 23:19:25 -05:00
Aditya Atluri 0a5761a1de indent correction for square.cu 
Change-Id: I2ca008e260b920ac3a503ad2a4bb28cd32300c98


[ROCm/clr commit: d8c0954e13]
 2016-10-12 19:26:59 -05:00
Aditya Atluri 18a7a5ce5d added copyright to module sample kernel file 
Change-Id: If57e0761df63c902e1677084ff85106ec49df5de


[ROCm/clr commit: cb734a5b1b]
 2016-10-12 19:23:48 -05:00
Maneesh Gupta 99d47a0850 clang-hipify -> hipify-clang in documentation 
Change-Id: I86ebc8112477db0d3e09f240beb3f9222d909ee6


[ROCm/clr commit: eb1fce0d77]
 2016-10-05 23:07:14 +05:30
Maneesh Gupta 19b9ab32b2 samples: Updated to use new hip include path 
Change-Id: I53a1385a17f13a997ea21d14315f15a3ad851dab


[ROCm/clr commit: ead394ec09]
 2016-10-04 22:21:10 +05:30
Ben Sander 2606f22c7f small typo fix 
Change-Id: I01906b330be8e6ec149bcdfe82def73e15931c89


[ROCm/clr commit: e413869271]
 2016-10-02 05:57:02 -05:00
sandeep kumar ae352f60d8 Add 2_Cookbook 
Change-Id: I10bbbd4bcb80a5900fe6af466c8f4c94ea5efe9a


[ROCm/clr commit: 67c7f0ab5a]
 2016-09-30 12:52:06 +05:30
Ben Sander d10d1425f3 Small tool, doc, sample enhancements. 
- Expand message when HIP version mismatch detected.
- Doc touchup.
- change sorting of hipBusBandwidth so byte results shown at top.
-

Change-Id: Ifb4e44a5fdfb65d59c4994b11e5f13385705f7e0


[ROCm/clr commit: 061bcc6a5c]
 2016-09-26 16:36:01 -05:00
Ben Sander 681f560fe1 Sample improvements. 
- Enable -O3 for hipDispatchLatency.
- Use nearly-null kernel to prevent it from being optimized away.
- Formatting for hipDispatchLatency.
- Formatting for hipInfo.


[ROCm/clr commit: e8beacbfb1]
 2016-09-22 13:05:47 -05:00
Aditya Atluri 891d5efd18 added vimrc for current project 
1. Added vimrc config file for HIP
2. Corrected square sample indent

Change-Id: I3e1d92403571148fe6825db6ad63ad925ae69519


[ROCm/clr commit: 8faf2ed7e3]
 2016-09-15 11:40:17 -05:00
Maneesh Gupta f963d347ff module_api/Makefile: Update as per newer hipgenisa.sh 
Change-Id: I479c74eae00d7521434f2740ce5930e326ea05cf


[ROCm/clr commit: 66232eb610]
 2016-09-06 17:47:10 +05:30
Maneesh Gupta de5701b678 module_api sample: Remove unnecessary platform checks 
Change-Id: I1d531264d51ff952a3a68d554672b6d293e23379


[ROCm/clr commit: 40a392800a]
 2016-09-04 21:25:14 +05:30
Rahul Garg d95642b85c Removed NVCC check for hipCtxXXX functions in module_api/runKernel.cpp 
Change-Id: I2bdd4fadf41063ec60626f1850e16f8307ebe6b5


[ROCm/clr commit: bb5f15e199]
 2016-09-04 20:37:29 +05:30
Maneesh Gupta cdfbacb907 module_api: HCC path no longer needs mangled kernel name 
Change-Id: I4c1cb218bfdd05c9fba57276167e3e4205b93614


[ROCm/clr commit: 34a3766ae3]
 2016-09-04 16:26:16 +05:30
Maneesh Gupta b3779207c3 module_api sample: no longer need EXTERN_C workaround 
Change-Id: Ida087d832df8e1f3620b38f920ec2853aad641c8


[ROCm/clr commit: 257774f8f0]
 2016-09-04 13:49:43 +05:30
Maneesh Gupta 7974137a91 module_api: workaround to use vcpy_kernel.cpp for NV path 
Change-Id: Ib4868bf02c64070e846c19427c39289609909466


[ROCm/clr commit: fa85974c47]
 2016-09-04 12:35:08 +05:30
pensun dd92a59939 For module_api sample, use vcpy_kernel.cu to generate ptx file for NV path. 
Change-Id: Id0033678834288c4eaa56b12e7d447119be99deb


[ROCm/clr commit: 793dc2bce0]
 2016-09-03 21:06:58 -05:00
Aditya Atluri d864a28183 corrected offline kernel compilation on hipcc path 
1. hipgenisa.sh now adds int main(){} during kernel compilation. User does not have to put it there
2. Renamed vcpy_isa.cpp to vcpy_kernel.cpp
3. Removed vcpy_isa.cu as the kernel code should be common for both paths
4. Changed Makefile and runkernel.cpp to work with above changes

Change-Id: I9f8c84706b44bb500bc493a68e959762b55a0142


[ROCm/clr commit: 8f536d8580]
 2016-09-02 13:17:17 -05:00
Ben Sander f3da27cd5a Fix double-lock of stream on hipModuleLaunchKernel 
Change-Id: I4ca164971c25f4eb8fbcca11d6258367bb3d2ab4


[ROCm/clr commit: 0fdb5bf517]
 2016-09-02 12:47:49 -05:00
Ben Sander f575ad7396 Fix av::copy in dialects to use capture-by-value 
Change-Id: Ibce1488a1326f66b92b4d5b351230666b691ed31


[ROCm/clr commit: 43e9932196]
 2016-09-02 09:46:59 -05:00
Ben Sander 61a5a53512 enable hc_am example in hcc_ddialects example 
Change-Id: Iec2f9eb05f95cb025c157fee8fd284aab844d1a2


[ROCm/clr commit: 27b3706a72]
 2016-09-02 09:46:59 -05:00
Aditya Atluri 8bbd449f9f Fixed offline kernel compilation 
1. Removed vcpy_isa.ptx as it should be generated during make
2. Made argument padding specific to hcc path
3. Renamed --gencodeobject to --genco
4. Changed Makefile to work on both nvcc and hcc path

Change-Id: Ifd053d541085d9ce4fd37bc21b07674786c7163e


[ROCm/clr commit: f22fda1291]
 2016-09-01 10:39:14 -05:00