update paper

Author: zuoyanzhang

content/publication/ICS_2019/cite.bib

@@ -1,14 +1,17 @@
-@inproceedings{swgemm,
-  location = {Bordeaux, France},
-  address = {New York, NY, USA},
-  author = {Tao, Xiaohan and Zhu, Yu and Wang, Boyang and Xu, Jinlong and Pang, Jianmin and Zhao, Jie},
-  booktitle = {Proceedings of the 51st International Conference on Parallel Processing},
-  isbn = {9781450397339},
-  numpages = {12},
-  pages = {23:1-23:12},
-  publisher = {Association for Computing Machinery},
-  series = {ICPP'22},
-  title = {Automatically Generating High-performance Matrix Multiplication Kernels on the Latest Sunway Processor},
-  doi = {https://doi.org/10.1145/3545008.3545031},
-  year = {2022}
+@inproceedings{10.1145/3330345.3331059,
+author = {Sun, Huihui and Fey, Florian and Zhao, Jie and Gorlatch, Sergei},
+title = {WCCV: improving the vectorization of IF-statements with warp-coherent conditions},
+year = {2019},
+isbn = {9781450360791},
+publisher = {Association for Computing Machinery},
+address = {New York, NY, USA},
+url = {https://doi.org/10.1145/3330345.3331059},
+doi = {10.1145/3330345.3331059},
+abstract = {When vectorizing programs for modern processors with SIMD extensions, IF-statements pose a challenge: existing vectorization approaches often introduce redundant computations or they resort to inefficient masked instructions.In this paper, we introduce a new notion of warp-coherence for conditions that exhibit coherent run-time behavior on different lanes of a vector register. We demonstrate that warp-coherent conditions appear frequently in practice. We present Warp-Coherent Condition Vectorization (WCCV) - an approach to detecting and optimizing IF-statements with warp-coherent conditions - to efficiently vectorize programs with IF-statements while avoiding the overhead of existing methods. WCCV detects warp-coherent conditions via the affine analysis of conditional boolean expressions and branch predication of IF-statements; the runtime code generated by WCCV avoids redundant computations and masked instructions. We employ auto-tuning to find the optimal benefit-overhead ratio for WCCV. We implement WCCV on top of Region Vectorizer (RV) - an LLVM-based vectorizing compiler, and we conduct experiments on the Rodinia benchmark suite, achieving a mean speedup of 1.14\texttimes{} over the original vectorized and optimized code, and speedup between 0.98\texttimes{} and 7.02\texttimes{} over the scalar code on Skylake with AVX512.},
+booktitle = {Proceedings of the ACM International Conference on Supercomputing},
+pages = {319–329},
+numpages = {11},
+keywords = {IF-statements, SPMD-on-SIMD, compiler optimization, vectorization, warp-coherence},
+location = {Phoenix, Arizona},
+series = {ICS '19}
 }