60 Commits (73a6fa3ed0fe2bbbfe72c05f42faabccd3bbadb7)

Author	SHA1	Message	Date
Zhou Wei	4c0c55bba1	support Geforce RTX 30+ GPU (#31529)	4 years ago
Wojciech Uss	615d8a2264	Modify relu native implementation 2 (#30996) ... * Modify relu native implementation * fix GPU performance	4 years ago
Zhou Wei	c94a4b9468	Separate AVX and NO_AVX compilation, enhance installation error message (#30413)	4 years ago
Shang Zhizhou	b9e76a0103	detect tensorRT plugin fp16 in runtime (#27933) ... * remove -DSUPPORTS_CUDA_FP16 in cuda.cmake * comile with cuda9 * add some unittest * notest;test=coverage * add unittest for trt plugin swish && split * update ernie unittest * fix some error message * remove repeated judgement of CUDA version in mbEltwiseLayerNormOpConverter * fix comile errror when CUDA_ARCH_NAME < Pascal" * fix comile error * update unittest timeout * compile with cuda9 * update error msg * fix code style * add some comments * add define IF_CUDA_ARCH_SUPPORT_FP16 * rename IF_CUDA_ARCH_SUPPORT_FP16 to CUDA_ARCH_FP16_SUPPORTED	4 years ago
Zhou Wei	68c473e3e0	fix Automatic GPU detection failed on windows (#28148)	4 years ago
Zhou Wei	5d7000215a	fix dynamic_loader more safe and error message on windows (#28117)	4 years ago
Pei Yang	a5ef246cac	Optimize emb_eltwise_layernorm_plugin and support fp16 (#27128)	4 years ago
wangchaochaohu	c71d79b1d2	[cuda11 support] change the CMakeLists to support the cuda11 (#27124)	4 years ago
wangchaochaohu	9b7692b144	update gpu compute archs for cuda11 (#27039)	4 years ago
Zhou Wei	62bd7ba13c	specify cuda arch when dectected fail (#26420) ... specify cuda arch when dectected fail	5 years ago
Zhou Wei	1f74b94d3f	fix compile warning on windows MSVC, fix paddle_build.bat more safe (#25933) ... * Fixed compile warning about incorrect compile options,fix paddle_build.bat * fix paddle_build.bat to more safe	5 years ago
Chen Weihang	172d4ecb6c	remove WITH_DSO compile option (#25444)	5 years ago
T8T9	a73a4a8fe7	don't support cmake 3.12, 3.13, 3.14 (#25021)	5 years ago
Zhou Wei	3e04ed2227	fix bug in CUDA_NVCC_FALS and CMAKE_CUDA_FLAGS, and eliminate some warning,test=develop (#24982) ... fix bug in CUDA_NVCC_FALS and CMAKE_CUDA_FLAGS	5 years ago
T8T9	90d420b13c	add -DPADDLE_CUDA_BINVER (#24928) ... * add -DPADDLE_CUDA_BINVER. test=develop, test=win_gpu * nvcc will use add_compile_options, avoid using it if you don't want to pass arguments to nvcc. test=develop * test=develop, test=win_gpu	5 years ago
T8T9	211ef78c1e	Builtin cuda (#24904) ... * support CUDA using cmake built-in way (#24395) * support CUDA using cmake built-in way. test=develop * test=develop * cmake_minimum_required 3.10 * test=develop	5 years ago
Zhou Wei	80ec2fe71c	fix windows bug that compile .cu files use MSVC dynamic C runtime (#24729)	5 years ago
Shibo Tao	30efee339a	Revert "support CUDA using cmake built-in way (#24395). test=develop" (#24468) ... This reverts commit 068d3690c6.	5 years ago
Shibo Tao	068d3690c6	support CUDA using cmake built-in way (#24395) ... * support CUDA using cmake built-in way. test=develop * test=develop	5 years ago
Pei Yang	8c296dea75	fix compile error(cpuid.h not found) on nvidia jetson platforms. test=develop (#24329)	5 years ago
Guo Sheng	1fc6cc502a	Fix cusolver loader for Windows (#24157) ... * Fix cusolver loader for Windows in dynamic_loader.cc. test=develop * Fix missing CUSOLVER_ROUTINE_EACH_R1. test=gpu test=develop * Add unsupprot for cusolver on Windows temporarily. test=develop * Fix GetCusolverDsoHandle error message. test=develop	5 years ago
Zeng Jinle	d053dfd5fc	fix cuda arch detection (#24036)	5 years ago
Zhaolong Xing	c113302826	fix cuda9, volta, turing compile error (#23730)	5 years ago
Zhaolong Xing	430b0099c9	[Paddle-TRT]: Ernie Dynamic shape support. (#23138) ... * add dynamic plugin support. test=develop * change emb eltwise layernorm to math function test=develop * add emb eltwise layernorm test=develop * can run dynamic shape ernie test=develop * fix ci test=develop * add ut for trt ernie dynamic test=develop * refine dynamic shape c++ interface. test=develop * fix comments test=develop * fix comments test=develop	5 years ago
Zhaolong Xing	c5f0293cf3	NV jetson(nano, tx2, xavier) inference compile support (#21393) ... * add jeston compile support test=develop * refine the cmake test=develop	5 years ago
Tao Luo	d8e7d25274	make CUDA_ARCH_NAME default Auto (#21352) ... * make CUDA_ARCH_NAME default Auto test=develop * refine warning test=develop	5 years ago
Tao Luo	32a670badc	remove WITH_FAST_MATH option (#19149) ... test=develop	6 years ago
wopeizl	3d0e1204d6	add support for cuda9 on windows test=develop (#17594) ... * add support for cuda9 on windows test=develop * use different git address for cuda9 compatible on windows	6 years ago
Tao Luo	f52d372876	remove legacy EXTERNAL_LIBS variable ... test=develop	6 years ago
peizhilin	805d505f14	disable warnings for third parties ... test=develop	6 years ago
peizhilin	3a4110f960	fix ci broken randomly and disable some warnings ... test=develop	6 years ago
Tao Luo	2d529186f1	remove legacy CMAKE_CROSSCOMPILING option	6 years ago
chengduo	55a0672378	fix compute_75 of cuda_cmake (#15209) ... test=develop	6 years ago
chengduo	b1ea335f60	add sm_75 support (#15198) ... test=develop	6 years ago
peizhilin	1e7f83e60a	add cuda dso support for windows ... test=develop	6 years ago
peizhilin	1afa9492af	Recover the profiler	6 years ago
peizhilin	445fff24dc	add the bigobj option to NVCC compile ... fix code style	6 years ago
peizhilin	52f7644f53	Merge remote-tracking branch 'upstream/develop' into windows/build	6 years ago
peizhilin	4ffa92d4f0	Merge branch 'develop' into windows/build	6 years ago
Zhaolong Xing	ba8b5619a3	Revert "cherry picked windows patches."	6 years ago
peizhilin	3c439feadc	remove the duplicate flag	6 years ago
peizhilin	e3f7be959d	fix the debug flag for nvcc	6 years ago
peizhilin	130cdda65b	add gpu debug mode	6 years ago
peizhilin	1f12ba6192	gpu support, fix build issue: ... 1. Non utf-8 characters within comments of OPs may lead to protobuf fail to parse_from_string 2. comment out some ops which not supported on windows 3. cuda libs may not be correctly linked to target on windows	6 years ago
dzhwinter	bf2e4cb188	cleard. staged	6 years ago
dzhwinter	ebfe5a02b3	merge develop branch	6 years ago
wanghaoshuang	3ae9645084	compile in linux	6 years ago
Tao Luo	28889caea5	disable EIGEN_FAST_MATH and use_fast_math ... test=develop	6 years ago
dzhwinter	85f8dd1c77	debug version	7 years ago
dzhwinter	379b471ee2	squash commit	7 years ago