Paddle

Author	SHA1	Message	Date
Chen Weihang	0b54d54fd8	Fix index overflow bug of the CUDA kernel loop increment (#25435 ) * fix softmax_with_cross_entropy cuda kernel overflow bug, test=develop * replace old macro & for condition, test=develop * polish details, test=develop	5 years ago
Chen Weihang	d1062d5278	Replace all errors thrown by LOG(FATAL) with PADDLE_THROW (#24759 ) * remove REPLACE_ENFORCE_GLOG compile option & add ci rule prohibit LOG(FATAL) using, test=develop * remove ci test case, test=develop * replace all LOG(FATAL) & polish message, test=develop * fix typo, test=develop * polish error info detail, test=develop	5 years ago
Wilber	7bc4b09500	add WITH_NCCL option for cmake. (#22384 ) cmake选项中添加了WITH_NCCL，显示指定是否编译NCCL的部分代码，WITH_NCCL默认打开，但如果WITH_GPU为OFF，则关闭WITH_NCCL 添加了PADDLE_WITH_NCCL定义单机单卡能够关闭NCCL编译，多卡的话需要默认打开NCCL，如果关闭NCCL，则只能使用单卡 Co-authored-by: 石晓伟 <39303645+Shixiaowei02@users.noreply.github.com>	5 years ago
hutuxian	a6b089c614	add macro to ban windows (#21422 ) remove nccl related code in windows	5 years ago
hutuxian	47a82e38e3	Support data_norm gpu kernel (#21325 ) * support data_norm_op run in CUDA * add two parameters sync_stats & summary_decay_rate * add UT	5 years ago