bulat-gpgpu-links.txt

https://devtalk.nvidia.com/default/topic/933827/cuda-programming-and-performance/fast-256-bin-histogram/
http://www.cse.uconn.edu/~zshi/course/cse5302/ref/chhugani08sorting.pdf
http://link.springer.com/chapter/10.1007/978-3-642-23397-5_16
http://arxiv.org/abs/1008.2849 Faster Radix Sort via Virtual Memory and Write-Combining Jan Wassenberg, Peter Sanders

https://devtalk.nvidia.com/default/topic/378826/cuda-programming-and-performance/my-speedy-sgemm/post/2703033/#2703033
https://devtalk.nvidia.com/default/topic/390366/cuda-programming-and-performance/instruction-latency/post/2768197/#2768197
https://devtalk.nvidia.com/default/topic/913832/cuda-programming-and-performance/sum-reduction-working-in-fermi-kepler-and-maxwell/
https://devtalk.nvidia.com/default/topic/776043/cuda-programming-and-performance/whats-new-in-maxwell-sm_52-gtx-9xx-/1
https://devtalk.nvidia.com/default/topic/690631/cuda-programming-and-performance/so-whats-new-about-maxwell-/post/4305310/#4305310
https://devtalk.nvidia.com/default/topic/878664/cuda-programming-and-performance/custom-memory-allocator-for-cuda-desired/post/4755097/#4755097
https://devtalk.nvidia.com/default/topic/695408/first-impressions-of-cuda-6-managed-memory/
https://devtalk.nvidia.com/default/topic/878455/cuda-programming-and-performance/gtx750ti-and-buffers-gt-1gb-on-win7/post/4672378/#4672378
https://devtalk.nvidia.com/default/topic/638031/ldg-versus-textures/
http://stackoverflow.com/questions/17004557/how-to-avoid-tlb-miss-and-high-global-memory-replay-overhead-in-cuda-gpus
https://devtalk.nvidia.com/default/topic/873995/global-memory-access-bottleneck/
http://www.techenablement.com/inside-nvidias-unified-memory-multi-gpu-limitations-and-the-need-for-a-cudamadvise-api-call/
https://parallel-computing.pro/index.php/9-cuda/43-openmp-4-0-on-nvidia-cuda-gpus
GPGPU-Sim 3.x Manual http://archive.is/krK9N
https://devtalk.nvidia.com/default/topic/928796/gpu-accelerated-libraries/moderngpu-2-0/
https://devtalk.nvidia.com/default/topic/844924/announcements/cudapad-and-its-source-code-are-now-available-for-download-/
CUB: http://on-demand.gputechconf.com/gtc/2015/video/S5617.html
     https://www.microway.com/hpc-tech-tips/cub-action-simple-examples-using-cub-template-library/
http://on-demand.gputechconf.com/gtc-express/2013/videos/understanding-parallel-graph-algorithms.mp4
http://on-demand.gputechconf.com/gtc/2013/webinar/essential-optimization-techniques-for-nvidia-kepler-and-fermi-architecture.mp4
https://devtalk.nvidia.com/default/topic/799429/cuda-programming-and-performance/possible-to-use-the-cuda-math-api-integer-intrinsics-to-find-the-nth-unset-bit-in-a-32-bit-int/post/4407256/#4407256
https://devtalk.nvidia.com/default/topic/804281/cuda-programming-and-performance/maxwell-integer-mul-mad-instruction-counts
https://devtalk.nvidia.com/default/topic/980740/cuda-programming-and-performance/xmad-meaning/
http://stackoverflow.com/questions/35566178/instruction-replay-in-cuda/35593124#35593124
https://devtalk.nvidia.com/default/topic/937736/cuda-programming-and-performance/saturated-16-bit-1-15-float-hack/post/4887190/#4887190
http://stackoverflow.com/questions/37732735/nvprof-option-for-bandwidth
https://devtalk.nvidia.com/default/topic/1006066/cuda-programming-and-performance/pascal-l1-cache
https://devtalk.nvidia.com/default/topic/1009766/cuda-programming-and-performance/single-gpu-core-vs-single-cpu-core
http://www.hardware.fr/articles/948-2/gp104-7-2-milliards-transistors-16-nm.html
    http://www.hardware.fr/articles/951-2/polaris-10-5-7-milliards-transistors-14-nm.html

the way i learned sass is
1. ptx manual: http://docs.nvidia.com/cuda/parallel-thread-execution/
2. http://docs.nvidia.com/cuda/cuda-binary-utilities/#instruction-set-ref
3. https://github.com/laanwj/decuda
4. read wiki of asfermi project: https://github.com/hyqneuron/asfermi/wiki
5. read manual of kepler sass: https://hpc.aliyun.com/doc/keplerAssemblerUserGuide
6. there is also maxas, but its docs doesn't describe commands

low-level benchmarks:
http://www.cs.berkeley.edu/~volkov/volkov10-GTC.pdf
http://www.stuffedcow.net/research/cudabmk Demystifying GPU Microarchitecture through Microbenchmarking
http://asg.ict.ac.cn/dgemm/microbenchs.tar.gz
http://repository.lib.ncsu.edu/ir/bitstream/1840.16/9585/1/etd.pdf
https://hal.inria.fr/file/index/docid/789958/filename/112_Lai.pdf
http://hgpu.org/?p=14541 Dissecting GPU Memory Hierarchy through Microbenchmarking
http://hgpu.org/?p=16616 Understanding Latency Hiding on GPUs by Vasily Volkov


a few books with low-level GPU details:
http://www.cudahandbook.com/
Shane Cook "CUDA Programming: A Developer’s Guide to Parallel Computing with GPUs"
Rob Farber "CUDA Application Design and Development"
David Kirk, Wen-mei Hwu "Programming Massively Parallel Processors"

Talks:
http://on-demand-gtc.gputechconf.com/gtc-quicklink/9BNvqKX
http://on-demand.gputechconf.com/gtc/2013/presentations/S3466-Programming-Guidelines-GPU-Architecture.pdf
http://on-demand.gputechconf.com/gtc/2016/presentation/s6807-angerer-dynamic-parallelism.pdf


AMD:
https://radeonopencompute.github.io/documentation.html
http://developer.amd.com/wordpress/media/2013/06/2620_final.pdf
http://developer.amd.com/tools-and-sdks/opencl-zone/amd-accelerated-parallel-processing-app-sdk/documentation/
http://developer.amd.com/wordpress/media/2013/07/AMD_Accelerated_Parallel_Processing_OpenCL_Programming_Guide-rev-2.7.pdf
https://forum.beyond3d.com/posts/1721467/
https://forum.beyond3d.com/threads/amd-southern-islands-7-series-speculation-rumour-thread.50220/page-22#post-1515943
https://github.com/SunsetQuest/Asm4GCN
https://realhet.wordpress.com/
http://x.pgy.hu/~worm/het/hp/GCN_Reference_Card.html

http://www.asmcommunity.net/forums/topic/?id=30544


Intel:
https://software.intel.com/en-us/articles/introduction-to-gen-assembly

ARM:
http://www.anandtech.com/show/10375/arm-unveils-bifrost-and-mali-g71/2