diff options
| -rw-r--r-- | README.md | 4 | ||||
| -rw-r--r-- | scripts/graphs/common.r (renamed from test/performance/graphs/common.r) | 0 | ||||
| -rw-r--r-- | scripts/graphs/xaxpy.r (renamed from test/performance/graphs/xaxpy.r) | 0 | ||||
| -rwxr-xr-x | scripts/graphs/xgemm.r (renamed from test/performance/graphs/xgemm.r) | 0 | ||||
| -rw-r--r-- | scripts/graphs/xgemv.r (renamed from test/performance/graphs/xgemv.r) | 0 | ||||
| -rw-r--r-- | scripts/graphs/xsymm.r (renamed from test/performance/graphs/xsymm.r) | 0 | ||||
| -rw-r--r-- | scripts/graphs/xsyr2k.r (renamed from test/performance/graphs/xsyr2k.r) | 0 | ||||
| -rw-r--r-- | scripts/graphs/xsyrk.r (renamed from test/performance/graphs/xsyrk.r) | 0 | ||||
| -rw-r--r-- | scripts/graphs/xtrmm.r (renamed from test/performance/graphs/xtrmm.r) | 0 |
9 files changed, 2 insertions, 2 deletions
@@ -169,9 +169,9 @@ To test the performance of CLBlast and compare optionally against clBLAS or a CP The performance tests come in the form of client executables named `clblast_client_xxxxx`, in which `xxxxx` is the name of a routine (e.g. `xgemm`). These clients take a bunch of configuration options and directly run CLBlast in a head-to-head performance test against optionally clBLAS and/or a CPU BLAS library. You can use the command-line options `-clblas 1` or `-cblas 1` to select a library to test against. -The folder `doc/performance` contains some PDF files with performance results on tested devices. Performance is compared in this case against a tuned version of the clBLAS library. These graphs can be generated automatically on your own device. First, compile CLBlast with the clients enabled. Then, make sure your installation of the reference clBLAS is performance-tuned by running the `tune` executable. Finally, run one of the graph-scripts found in `test/performance/graphs` using R. For example, to generate the Xgemm PDF on device 1 of platform 0: +The folder `doc/performance` contains some PDF files with performance results on tested devices. Performance is compared in this case against a tuned version of the clBLAS library. These graphs can be generated automatically on your own device. First, compile CLBlast with the clients enabled. Then, make sure your installation of the reference clBLAS is performance-tuned by running the `tune` executable. Finally, run one of the graph-scripts found in `scripts/graphs` using R. For example, to generate the Xgemm PDF on device 1 of platform 0 from the `build` subdirectory: - Rscript path/to/test/performance/graphs/xgemm.r 0 1 + Rscript ../scripts/graphs/xgemm.r 0 1 Note that the CLBlast library provides pre-tuned parameter-values for some devices only: if your device is not among these, then out-of-the-box performance might be poor. See above under `Using the tuners` to find out how to tune for your device. diff --git a/test/performance/graphs/common.r b/scripts/graphs/common.r index cd68cf26..cd68cf26 100644 --- a/test/performance/graphs/common.r +++ b/scripts/graphs/common.r diff --git a/test/performance/graphs/xaxpy.r b/scripts/graphs/xaxpy.r index 187590aa..187590aa 100644 --- a/test/performance/graphs/xaxpy.r +++ b/scripts/graphs/xaxpy.r diff --git a/test/performance/graphs/xgemm.r b/scripts/graphs/xgemm.r index e758f460..e758f460 100755 --- a/test/performance/graphs/xgemm.r +++ b/scripts/graphs/xgemm.r diff --git a/test/performance/graphs/xgemv.r b/scripts/graphs/xgemv.r index 9a8040f7..9a8040f7 100644 --- a/test/performance/graphs/xgemv.r +++ b/scripts/graphs/xgemv.r diff --git a/test/performance/graphs/xsymm.r b/scripts/graphs/xsymm.r index a65bb16f..a65bb16f 100644 --- a/test/performance/graphs/xsymm.r +++ b/scripts/graphs/xsymm.r diff --git a/test/performance/graphs/xsyr2k.r b/scripts/graphs/xsyr2k.r index 4b2dd4a0..4b2dd4a0 100644 --- a/test/performance/graphs/xsyr2k.r +++ b/scripts/graphs/xsyr2k.r diff --git a/test/performance/graphs/xsyrk.r b/scripts/graphs/xsyrk.r index 4ab46c9f..4ab46c9f 100644 --- a/test/performance/graphs/xsyrk.r +++ b/scripts/graphs/xsyrk.r diff --git a/test/performance/graphs/xtrmm.r b/scripts/graphs/xtrmm.r index c2faaa8b..c2faaa8b 100644 --- a/test/performance/graphs/xtrmm.r +++ b/scripts/graphs/xtrmm.r |