CLBLAS(1)

✨ CLBLAS(1) ✨

okay bestie, welcome to the clBLAS-client manual page 💖 this is the User Commands corner of the CLBLAS section and honestly? it's giving "let me test literally every option for the OpenCL BLAS library" energy and I'm obsessed 🦋

🌸 NAME 🌸

clBLAS-client — a whole program whose entire job is testing the different options for the OpenCL BLAS library. no cap, that's it, that's the gig 💅

🎀 DESCRIPTION 🎀

🫶 clBLAS client command line options:

okay so these are all the flags you can throw at this thing. read them carefully, bestie, because they are literally your entire control panel ✨

• -h, [ --help ] — prints this exact help message. the "help me help you" of it all 😭
• -g, [ --gpu ] — forces the instantiation of an OpenCL GPU device. GPU main character energy 👑
• -c, [ --cpu ] — forces the instantiation of an OpenCL CPU device. yes she can do CPU too, versatile queen 🎀
• -a, [ --all ] — forces the instantiation of all OpenCL devices. greedy? no. thorough 💅
• --useimages — uses an image-based kernel ✨
• -m, [ --sizem ] arg (=128) — number of rows in A and C. default is 128
• -n, [ --sizen ] arg (=128) — number of columns in B and C. default is 128
• -k, [ --sizek ] arg (=128) — number of columns in A and rows in B. default is 128
• --lda arg (=0) — first dimension of A in memory. if set to 0, lda will default to M (when transposeA is "no transpose") or K (otherwise) 🧚
• --ldb arg (=0) — first dimension of B in memory. if set to 0, ldb will default to K (when transposeB is "no transpose") or N (otherwise) 🧚
• --ldc arg (=0) — first dimension of C in memory. if set to 0, ldc will default to M 🌈
• --offA arg (=0) — offset of the matrix A in the memory object 🍓
• --offBX arg (=0) — offset of the matrix B or vector X in the memory object 🍓
• --offCY arg (=0) — offset of the matrix C or vector Y in the memory object 🍓
• --alpha arg (=1) — specifies the scalar alpha ✨
• --beta arg (=1) — specifies the scalar beta ✨
• -o, [ --order ] arg (=0) — 0 = row major, 1 = column major 📐
• --transposeA arg (=0) — 0 = no transpose, 1 = transpose, 2 = conjugate transpose 🔄
• --transposeB arg (=0) — 0 = no transpose, 1 = transpose, 2 = conjugate transpose 🔄
• -f, [ --function ] arg (=gemm) — the BLAS function to test. options: gemm, trsm, trmm, gemv, symv, syrk, syr2k 👑
• -r, [ --precision ] arg (=s) — options: s, d, c, z 💜
• --side arg (=0) — 0 = left, 1 = right. only used with [list of function families] 🌸
• --uplo arg (=0) — 0 = upper, 1 = lower. only used with [list of function families] 🌸
• --diag arg (=0) — 0 = unit diagonal, 1 = non unit diagonal. only used with [list of function families] 🌸
• -p, [ --profile ] arg (=20) — times and reports the kernel speed (default: profiling off). the "how fast is she" check ⏱️🔥
• --roundtrip arg (=noroundtrip) — includes the time of OpenCL memory allocation and transportation. options: roundtrip, noroundtrip (default) 🚗💨
• --memalloc arg (=default) — sets the memory allocation flags for OpenCL. no effect if roundtrip time is not measured ⚠️ options: default (default), alloc_host_ptr, use_host_ptr, copy_host_ptr, use_persistent_mem_amd, rect_mem 🧠

💖 COPYRIGHT 💖

copyright © 2015 Ghislain Antony Vaillant 🫶

this manual page was written for the Debian system 🎀

permission is granted to copy, distribute and/or modify this document under the terms of the BSD 2-Clause license. slay, open source icon behavior ✨

💅 and that's the whole vibe check, bestie — go forth and benchmark responsibly 🦋

clBLAS client API version: 2.4.0 🌸 April 2015 🌸 CLBLAS(1) — periodt, no cap, she's iconic and she's done 👑💖