Attachment 14255 Details for Bug 32646 – mixbench HIP test for 7530U iGPU

mixbench HIP test for 7530U iGPU

mixbench-hip-7530U.txt (text/plain), 1.41 KB, created by christian barranco on 2024-01-06 23:57:46 CET

(hide)

Description:

Filename:

MIME Type:

Creator: christian barranco

Created: 2024-01-06 23:57:46 CET

Size: 1.41 KB

patch

obsolete

>Cuda error in file '/devel/rpmbuild/SRC-bkp/Mixbench/mixbench-0.04/mixbench-hip/mix_kernels_hip_ro.cpp' in line 89 : shared object initialization failed.
>mixbench-hip/read-only ()
>------------------------ Device specifications ------------------------
>Device:              AMD Radeon Graphics
>CUDA driver version: 60032.830
>GPU clock rate:      2000 MHz
>WarpSize:            64
>L2 cache size:       1024 KB
>Total global mem:    512 MB
>Total SPs:           896 (7 MPs x 128 SPs/MP)
>Compute throughput:  3584.00 GFlops (theoretical single precision FMAs)
>Memory bandwidth:    42.66 GB/sec
>-----------------------------------------------------------------------
>Total GPU memory 536870912, free 501219328
>Buffer size:          256MB
>Trade-off type:       compute with global memory (block strided)
>Elements per thread:  8
>Thread fusion degree: 1
>----------------------------------------------------------------------------- CSV data -------------------------------------------------------------------------------------------------------------------
>Experiment ID, Single Precision ops,,,,              Packed Single Precision ops,,,,       Double precision ops,,,,              Half precision ops,,,,                Integer operations,,, 
>Compute iters, Flops/byte, ex.time,  GFLOPS, GB/sec, Flops/byte, ex.time,  GFLOPS, GB/sec, Flops/byte, ex.time,  GFLOPS, GB/sec, Flops/byte, ex.time,  GFLOPS, GB/sec, Iops/byte, ex.time,   GIOPS, GB/sec

Cuda error in file '/devel/rpmbuild/SRC-bkp/Mixbench/mixbench-0.04/mixbench-hip/mix_kernels_hip_ro.cpp' in line 89 : shared object initialization failed.
mixbench-hip/read-only ()
------------------------ Device specifications ------------------------
Device:              AMD Radeon Graphics
CUDA driver version: 60032.830
GPU clock rate:      2000 MHz
WarpSize:            64
L2 cache size:       1024 KB
Total global mem:    512 MB
Total SPs:           896 (7 MPs x 128 SPs/MP)
Compute throughput:  3584.00 GFlops (theoretical single precision FMAs)
Memory bandwidth:    42.66 GB/sec
-----------------------------------------------------------------------
Total GPU memory 536870912, free 501219328
Buffer size:          256MB
Trade-off type:       compute with global memory (block strided)
Elements per thread:  8
Thread fusion degree: 1
----------------------------------------------------------------------------- CSV data -------------------------------------------------------------------------------------------------------------------
Experiment ID, Single Precision ops,,,,              Packed Single Precision ops,,,,       Double precision ops,,,,              Half precision ops,,,,                Integer operations,,, 
Compute iters, Flops/byte, ex.time,  GFLOPS, GB/sec, Flops/byte, ex.time,  GFLOPS, GB/sec, Flops/byte, ex.time,  GFLOPS, GB/sec, Flops/byte, ex.time,  GFLOPS, GB/sec, Iops/byte, ex.time,   GIOPS, GB/sec

Actions: View

Attachments on bug 32646: 14252 | 14253 | 14254 | 14255 | 14259