sbc-bench v0.9.9 Radxa ROCK 5B (Tue, 06 Dec 2022 22:08:57 +0100) Distributor ID: Ubuntu Description: Ubuntu 22.04.1 LTS Release: 22.04 Codename: jammy Armbian info: Rock 5B, rockchip-rk3588, rockchip-rk3588, 22.11.1, https://github.com/armbian/build /usr/bin/gcc (Ubuntu 11.3.0-1ubuntu1~22.04) 11.3.0 Uptime: 22:08:57 up 3:12, 3 users, load average: 0.27, 0.30, 0.18, 37.9°C, 9.37V, 152958635 Linux 5.10.72-rockchip-rk3588 (rock-5b) 12/06/22 _aarch64_ (8 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 0.23 0.00 0.12 0.00 0.00 99.65 Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd mmcblk1 1.29 45.34 48.55 0.00 524234 561374 0 mtdblock0 0.00 0.03 0.00 0.00 336 0 0 zram0 0.05 0.20 0.00 0.00 2264 4 0 zram1 0.06 0.04 0.58 0.00 408 6688 0 total used free shared buff/cache available Mem: 7.5Gi 198Mi 7.2Gi 42Mi 145Mi 7.2Gi Swap: 3.8Gi 0B 3.8Gi Filename Type Size Used Priority /dev/zram0 partition 3937420 0 5 ########################################################################## Checking cpufreq OPP for cpu0-cpu3 (Cortex-A55): Cpufreq OPP: 1800 Measured: 1822 (1822.123/1822.083/1822.003) (+1.2%) Cpufreq OPP: 1608 Measured: 1638 (1638.797/1638.797/1638.675) (+1.9%) Cpufreq OPP: 1416 Measured: 1416 (1416.840/1416.810/1416.658) Cpufreq OPP: 1200 Measured: 1230 (1230.543/1230.457/1230.429) (+2.5%) Cpufreq OPP: 1008 Measured: 1060 (1060.093/1059.965/1059.944) (+5.2%) Cpufreq OPP: 816 Measured: 846 (846.154/846.154/846.048) (+3.7%) Cpufreq OPP: 600 Measured: 591 (591.249/591.236/591.211) (-1.5%) Cpufreq OPP: 408 Measured: 393 (393.364/393.364/393.355) (-3.7%) Checking cpufreq OPP for cpu4-cpu5 (Cortex-A76): Cpufreq OPP: 2400 Measured: 2344 (2344.492/2344.492/2344.492) (-2.3%) Cpufreq OPP: 2208 Measured: 2181 (2181.413/2181.321/2181.321) (-1.2%) Cpufreq OPP: 2016 Measured: 2012 (2012.429/2012.429/2012.331) Cpufreq OPP: 1800 Measured: 1812 (1812.651/1812.612/1812.572) Cpufreq OPP: 1608 Measured: 1621 (1621.753/1621.673/1621.554) Cpufreq OPP: 1416 Measured: 1434 (1434.205/1434.205/1434.174) (+1.3%) Cpufreq OPP: 1200 Measured: 1252 (1252.248/1252.218/1252.189) (+4.3%) Cpufreq OPP: 1008 Measured: 1053 (1053.348/1053.296/1053.165) (+4.5%) Cpufreq OPP: 816 Measured: 845 (845.499/845.478/845.415) (+3.6%) Cpufreq OPP: 600 Measured: 592 (592.958/592.932/592.880) (-1.3%) Cpufreq OPP: 408 Measured: 394 (395.022/394.968/394.950) (-3.4%) Checking cpufreq OPP for cpu6-cpu7 (Cortex-A76): Cpufreq OPP: 2400 Measured: 2343 (2343.216/2343.216/2343.162) (-2.4%) Cpufreq OPP: 2208 Measured: 2180 (2180.953/2180.907/2180.907) (-1.3%) Cpufreq OPP: 2016 Measured: 2012 (2012.870/2012.870/2012.772) Cpufreq OPP: 1800 Measured: 1814 (1814.323/1814.283/1814.044) Cpufreq OPP: 1608 Measured: 1623 (1623.306/1623.306/1623.186) Cpufreq OPP: 1416 Measured: 1433 (1433.490/1433.304/1433.179) (+1.2%) Cpufreq OPP: 1200 Measured: 1254 (1254.476/1254.357/1254.327) (+4.5%) Cpufreq OPP: 1008 Measured: 1055 (1055.766/1055.740/1055.687) (+4.7%) Cpufreq OPP: 816 Measured: 846 (846.070/846.048/845.985) (+3.7%) Cpufreq OPP: 600 Measured: 592 (592.945/592.945/592.906) (-1.3%) Cpufreq OPP: 408 Measured: 394 (394.995/394.986/394.959) (-3.4%) ########################################################################## Hardware sensors: gpu_thermal-virtual-0 temp1: +37.0 C littlecore_thermal-virtual-0 temp1: +37.9 C bigcore0_thermal-virtual-0 temp1: +37.0 C tcpm_source_psy_4_0022-i2c-4-22 in0: 9.00 V (min = +9.00 V, max = +9.00 V) curr1: 1.67 A (max = +1.67 A) npu_thermal-virtual-0 temp1: +37.0 C center_thermal-virtual-0 temp1: +36.1 C bigcore1_thermal-virtual-0 temp1: +37.0 C soc_thermal-virtual-0 temp1: +37.9 C (crit = +115.0 C) ########################################################################## Executing benchmark on cpu0 (Cortex-A55): tinymembench v0.4.9 (simple benchmark for memory throughput and latency) ========================================================================== == Memory bandwidth tests == == == == Note 1: 1MB = 1000000 bytes == == Note 2: Results for 'copy' tests show how many bytes can be == == copied per second (adding together read and writen == == bytes would have provided twice higher numbers) == == Note 3: 2-pass copy means that we are using a small temporary buffer == == to first fetch data into it, and only then write it to the == == destination (source -> L1 cache, L1 cache -> destination) == == Note 4: If sample standard deviation exceeds 0.1%, it is shown in == == brackets == ========================================================================== C copy backwards : 3202.2 MB/s (0.5%) C copy backwards (32 byte blocks) : 3180.7 MB/s C copy backwards (64 byte blocks) : 3204.2 MB/s C copy : 5741.6 MB/s (1.8%) C copy prefetched (32 bytes step) : 2261.8 MB/s (4.0%) C copy prefetched (64 bytes step) : 5555.9 MB/s (1.0%) C 2-pass copy : 2769.6 MB/s C 2-pass copy prefetched (32 bytes step) : 1671.1 MB/s (1.1%) C 2-pass copy prefetched (64 bytes step) : 2675.8 MB/s C fill : 12543.7 MB/s (0.1%) C fill (shuffle within 16 byte blocks) : 12539.3 MB/s C fill (shuffle within 32 byte blocks) : 12537.1 MB/s C fill (shuffle within 64 byte blocks) : 12243.7 MB/s --- standard memcpy : 5861.6 MB/s standard memset : 21075.3 MB/s --- NEON LDP/STP copy : 5183.1 MB/s NEON LDP/STP copy pldl2strm (32 bytes step) : 1820.4 MB/s (11.0%) NEON LDP/STP copy pldl2strm (64 bytes step) : 3086.5 MB/s NEON LDP/STP copy pldl1keep (32 bytes step) : 2091.9 MB/s NEON LDP/STP copy pldl1keep (64 bytes step) : 4941.1 MB/s (0.3%) NEON LD1/ST1 copy : 4925.4 MB/s NEON STP fill : 20803.2 MB/s NEON STNP fill : 13372.4 MB/s (3.0%) ARM LDP/STP copy : 5086.0 MB/s ARM STP fill : 20786.4 MB/s (0.4%) ARM STNP fill : 13350.1 MB/s (2.2%) ========================================================================== == Framebuffer read tests. == == == == Many ARM devices use a part of the system memory as the framebuffer, == == typically mapped as uncached but with write-combining enabled. == == Writes to such framebuffers are quite fast, but reads are much == == slower and very sensitive to the alignment and the selection of == == CPU instructions which are used for accessing memory. == == == == Many x86 systems allocate the framebuffer in the GPU memory, == == accessible for the CPU via a relatively slow PCI-E bus. Moreover, == == PCI-E is asymmetric and handles reads a lot worse than writes. == == == == If uncached framebuffer reads are reasonably fast (at least 100 MB/s == == or preferably >300 MB/s), then using the shadow framebuffer layer == == is not necessary in Xorg DDX drivers, resulting in a nice overall == == performance improvement. For example, the xf86-video-fbturbo DDX == == uses this trick. == ========================================================================== NEON LDP/STP copy (from framebuffer) : 186.2 MB/s (9.0%) NEON LDP/STP 2-pass copy (from framebuffer) : 130.8 MB/s NEON LD1/ST1 copy (from framebuffer) : 34.6 MB/s NEON LD1/ST1 2-pass copy (from framebuffer) : 34.3 MB/s ARM LDP/STP copy (from framebuffer) : 69.0 MB/s ARM LDP/STP 2-pass copy (from framebuffer) : 67.5 MB/s ========================================================================== == Memory latency test == == == == Average time is measured for random memory accesses in the buffers == == of different sizes. The larger is the buffer, the more significant == == are relative contributions of TLB, L1/L2 cache misses and SDRAM == == accesses. For extremely large buffer sizes we are expecting to see == == page table walk with several requests to SDRAM for almost every == == memory access (though 64MiB is not nearly large enough to experience == == this effect to its fullest). == == == == Note 1: All the numbers are representing extra time, which needs to == == be added to L1 cache latency. The cycle timings for L1 cache == == latency can be usually found in the processor documentation. == == Note 2: Dual random read means that we are simultaneously performing == == two independent memory accesses at a time. In the case if == == the memory subsystem can't handle multiple outstanding == == requests, dual random read has the same timings as two == == single reads performed one after another. == ========================================================================== block size : single random read / dual random read 1024 : 0.0 ns / 0.0 ns 2048 : 0.0 ns / 0.0 ns 4096 : 0.0 ns / 0.0 ns 8192 : 0.0 ns / 0.0 ns 16384 : 0.1 ns / 0.1 ns 32768 : 0.6 ns / 1.0 ns 65536 : 1.5 ns / 2.6 ns 131072 : 2.7 ns / 4.3 ns 262144 : 8.0 ns / 11.8 ns 524288 : 11.7 ns / 15.1 ns 1048576 : 13.8 ns / 16.1 ns 2097152 : 15.8 ns / 18.5 ns 4194304 : 43.9 ns / 66.8 ns 8388608 : 93.5 ns / 149.2 ns 16777216 : 123.2 ns / 147.4 ns 33554432 : 250.3 ns / 312.1 ns 67108864 : 128.2 ns / 156.4 ns Executing benchmark on cpu4 (Cortex-A76): tinymembench v0.4.9 (simple benchmark for memory throughput and latency) ========================================================================== == Memory bandwidth tests == == == == Note 1: 1MB = 1000000 bytes == == Note 2: Results for 'copy' tests show how many bytes can be == == copied per second (adding together read and writen == == bytes would have provided twice higher numbers) == == Note 3: 2-pass copy means that we are using a small temporary buffer == == to first fetch data into it, and only then write it to the == == destination (source -> L1 cache, L1 cache -> destination) == == Note 4: If sample standard deviation exceeds 0.1%, it is shown in == == brackets == ========================================================================== C copy backwards : 9245.8 MB/s C copy backwards (32 byte blocks) : 9194.0 MB/s C copy backwards (64 byte blocks) : 9189.8 MB/s C copy : 9406.0 MB/s (0.3%) C copy prefetched (32 bytes step) : 9569.8 MB/s C copy prefetched (64 bytes step) : 9595.9 MB/s C 2-pass copy : 4775.4 MB/s (0.1%) C 2-pass copy prefetched (32 bytes step) : 7160.6 MB/s C 2-pass copy prefetched (64 bytes step) : 7421.6 MB/s C fill : 24687.4 MB/s (0.7%) C fill (shuffle within 16 byte blocks) : 24695.8 MB/s (1.0%) C fill (shuffle within 32 byte blocks) : 24753.1 MB/s (0.4%) C fill (shuffle within 64 byte blocks) : 24675.9 MB/s (0.8%) --- standard memcpy : 9580.9 MB/s standard memset : 24751.4 MB/s (0.7%) --- NEON LDP/STP copy : 9600.4 MB/s NEON LDP/STP copy pldl2strm (32 bytes step) : 9637.7 MB/s NEON LDP/STP copy pldl2strm (64 bytes step) : 9656.2 MB/s NEON LDP/STP copy pldl1keep (32 bytes step) : 9677.7 MB/s NEON LDP/STP copy pldl1keep (64 bytes step) : 9673.5 MB/s NEON LD1/ST1 copy : 9499.7 MB/s NEON STP fill : 24730.4 MB/s (0.6%) NEON STNP fill : 24747.7 MB/s (0.3%) ARM LDP/STP copy : 9580.3 MB/s ARM STP fill : 24533.1 MB/s (0.4%) ARM STNP fill : 24710.7 MB/s (0.3%) ========================================================================== == Framebuffer read tests. == == == == Many ARM devices use a part of the system memory as the framebuffer, == == typically mapped as uncached but with write-combining enabled. == == Writes to such framebuffers are quite fast, but reads are much == == slower and very sensitive to the alignment and the selection of == == CPU instructions which are used for accessing memory. == == == == Many x86 systems allocate the framebuffer in the GPU memory, == == accessible for the CPU via a relatively slow PCI-E bus. Moreover, == == PCI-E is asymmetric and handles reads a lot worse than writes. == == == == If uncached framebuffer reads are reasonably fast (at least 100 MB/s == == or preferably >300 MB/s), then using the shadow framebuffer layer == == is not necessary in Xorg DDX drivers, resulting in a nice overall == == performance improvement. For example, the xf86-video-fbturbo DDX == == uses this trick. == ========================================================================== NEON LDP/STP copy (from framebuffer) : 1738.2 MB/s (12.5%) NEON LDP/STP 2-pass copy (from framebuffer) : 985.3 MB/s NEON LD1/ST1 copy (from framebuffer) : 1211.8 MB/s NEON LD1/ST1 2-pass copy (from framebuffer) : 1007.4 MB/s ARM LDP/STP copy (from framebuffer) : 1196.0 MB/s ARM LDP/STP 2-pass copy (from framebuffer) : 1004.4 MB/s ========================================================================== == Memory latency test == == == == Average time is measured for random memory accesses in the buffers == == of different sizes. The larger is the buffer, the more significant == == are relative contributions of TLB, L1/L2 cache misses and SDRAM == == accesses. For extremely large buffer sizes we are expecting to see == == page table walk with several requests to SDRAM for almost every == == memory access (though 64MiB is not nearly large enough to experience == == this effect to its fullest). == == == == Note 1: All the numbers are representing extra time, which needs to == == be added to L1 cache latency. The cycle timings for L1 cache == == latency can be usually found in the processor documentation. == == Note 2: Dual random read means that we are simultaneously performing == == two independent memory accesses at a time. In the case if == == the memory subsystem can't handle multiple outstanding == == requests, dual random read has the same timings as two == == single reads performed one after another. == ========================================================================== block size : single random read / dual random read 1024 : 0.0 ns / 0.0 ns 2048 : 0.0 ns / 0.0 ns 4096 : 0.0 ns / 0.0 ns 8192 : 0.0 ns / 0.0 ns 16384 : 0.0 ns / 0.0 ns 32768 : 0.0 ns / 0.0 ns 65536 : 0.0 ns / 0.0 ns 131072 : 1.1 ns / 1.5 ns 262144 : 2.2 ns / 2.8 ns 524288 : 4.5 ns / 5.8 ns 1048576 : 10.1 ns / 13.0 ns 2097152 : 13.9 ns / 15.6 ns 4194304 : 68.7 ns / 112.1 ns 8388608 : 165.6 ns / 241.0 ns 16777216 : 104.6 ns / 133.8 ns 33554432 : 116.5 ns / 139.6 ns 67108864 : 124.0 ns / 144.3 ns Executing benchmark on cpu6 (Cortex-A76): tinymembench v0.4.9 (simple benchmark for memory throughput and latency) ========================================================================== == Memory bandwidth tests == == == == Note 1: 1MB = 1000000 bytes == == Note 2: Results for 'copy' tests show how many bytes can be == == copied per second (adding together read and writen == == bytes would have provided twice higher numbers) == == Note 3: 2-pass copy means that we are using a small temporary buffer == == to first fetch data into it, and only then write it to the == == destination (source -> L1 cache, L1 cache -> destination) == == Note 4: If sample standard deviation exceeds 0.1%, it is shown in == == brackets == ========================================================================== C copy backwards : 9265.7 MB/s C copy backwards (32 byte blocks) : 9207.6 MB/s C copy backwards (64 byte blocks) : 9210.3 MB/s C copy : 9359.4 MB/s C copy prefetched (32 bytes step) : 9539.0 MB/s C copy prefetched (64 bytes step) : 9564.7 MB/s (0.2%) C 2-pass copy : 4764.9 MB/s C 2-pass copy prefetched (32 bytes step) : 7081.3 MB/s C 2-pass copy prefetched (64 bytes step) : 7361.5 MB/s C fill : 24439.7 MB/s (0.9%) C fill (shuffle within 16 byte blocks) : 24558.2 MB/s (0.4%) C fill (shuffle within 32 byte blocks) : 24429.7 MB/s C fill (shuffle within 64 byte blocks) : 24532.1 MB/s (0.3%) --- standard memcpy : 9538.2 MB/s standard memset : 24443.0 MB/s (0.8%) --- NEON LDP/STP copy : 9556.9 MB/s NEON LDP/STP copy pldl2strm (32 bytes step) : 9612.9 MB/s NEON LDP/STP copy pldl2strm (64 bytes step) : 9631.6 MB/s NEON LDP/STP copy pldl1keep (32 bytes step) : 9659.0 MB/s (0.2%) NEON LDP/STP copy pldl1keep (64 bytes step) : 9653.4 MB/s NEON LD1/ST1 copy : 9475.0 MB/s NEON STP fill : 24576.4 MB/s (0.9%) NEON STNP fill : 24445.3 MB/s (0.4%) ARM LDP/STP copy : 9535.9 MB/s ARM STP fill : 24555.4 MB/s (0.5%) ARM STNP fill : 24512.2 MB/s (0.5%) ========================================================================== == Framebuffer read tests. == == == == Many ARM devices use a part of the system memory as the framebuffer, == == typically mapped as uncached but with write-combining enabled. == == Writes to such framebuffers are quite fast, but reads are much == == slower and very sensitive to the alignment and the selection of == == CPU instructions which are used for accessing memory. == == == == Many x86 systems allocate the framebuffer in the GPU memory, == == accessible for the CPU via a relatively slow PCI-E bus. Moreover, == == PCI-E is asymmetric and handles reads a lot worse than writes. == == == == If uncached framebuffer reads are reasonably fast (at least 100 MB/s == == or preferably >300 MB/s), then using the shadow framebuffer layer == == is not necessary in Xorg DDX drivers, resulting in a nice overall == == performance improvement. For example, the xf86-video-fbturbo DDX == == uses this trick. == ========================================================================== NEON LDP/STP copy (from framebuffer) : 1735.0 MB/s NEON LDP/STP 2-pass copy (from framebuffer) : 1492.4 MB/s (12.1%) NEON LD1/ST1 copy (from framebuffer) : 1214.7 MB/s NEON LD1/ST1 2-pass copy (from framebuffer) : 1025.4 MB/s ARM LDP/STP copy (from framebuffer) : 1193.9 MB/s ARM LDP/STP 2-pass copy (from framebuffer) : 1024.8 MB/s ========================================================================== == Memory latency test == == == == Average time is measured for random memory accesses in the buffers == == of different sizes. The larger is the buffer, the more significant == == are relative contributions of TLB, L1/L2 cache misses and SDRAM == == accesses. For extremely large buffer sizes we are expecting to see == == page table walk with several requests to SDRAM for almost every == == memory access (though 64MiB is not nearly large enough to experience == == this effect to its fullest). == == == == Note 1: All the numbers are representing extra time, which needs to == == be added to L1 cache latency. The cycle timings for L1 cache == == latency can be usually found in the processor documentation. == == Note 2: Dual random read means that we are simultaneously performing == == two independent memory accesses at a time. In the case if == == the memory subsystem can't handle multiple outstanding == == requests, dual random read has the same timings as two == == single reads performed one after another. == ========================================================================== block size : single random read / dual random read 1024 : 0.0 ns / 0.0 ns 2048 : 0.0 ns / 0.0 ns 4096 : 0.0 ns / 0.0 ns 8192 : 0.0 ns / 0.0 ns 16384 : 0.0 ns / 0.0 ns 32768 : 0.0 ns / 0.0 ns 65536 : 0.0 ns / 0.0 ns 131072 : 1.1 ns / 1.5 ns 262144 : 2.2 ns / 2.8 ns 524288 : 4.1 ns / 5.3 ns 1048576 : 9.9 ns / 13.0 ns 2097152 : 13.4 ns / 15.7 ns 4194304 : 36.5 ns / 55.1 ns 8388608 : 80.4 ns / 112.8 ns 16777216 : 219.3 ns / 288.2 ns 33554432 : 247.3 ns / 306.2 ns 67108864 : 262.1 ns / 311.8 ns ########################################################################## Executing ramlat on cpu0 (Cortex-A55), results in ns: size: 1x32 2x32 1x64 2x64 1xPTR 2xPTR 4xPTR 8xPTR 4k: 1.651 1.651 1.651 1.651 1.101 1.651 2.236 4.507 8k: 1.651 1.651 1.650 1.651 1.101 1.651 2.236 4.507 16k: 1.666 1.651 1.664 1.651 1.108 1.651 2.236 4.506 32k: 1.672 1.653 1.676 1.653 1.115 1.653 2.240 4.512 64k: 9.397 10.94 9.396 10.93 9.618 10.96 16.03 29.20 128k: 13.72 14.79 13.71 14.77 14.24 14.76 21.85 41.45 256k: 15.88 16.35 15.88 16.35 15.22 16.39 25.54 49.48 512k: 16.71 16.82 16.64 16.82 15.97 17.03 26.58 52.83 1024k: 16.92 16.91 16.75 16.91 16.16 17.10 27.72 52.98 2048k: 21.68 24.12 21.11 23.97 20.38 24.34 39.16 76.59 4096k: 151.5 192.6 166.6 187.7 149.9 185.7 283.1 542.0 8192k: 266.0 252.5 200.3 234.7 196.8 236.8 415.3 819.7 16384k: 275.6 285.8 248.8 264.4 239.1 265.7 459.1 838.2 Executing ramlat on cpu4 (Cortex-A76), results in ns: size: 1x32 2x32 1x64 2x64 1xPTR 2xPTR 4xPTR 8xPTR 4k: 1.712 1.712 1.712 1.712 1.712 1.712 1.712 3.258 8k: 1.712 1.712 1.712 1.712 1.712 1.712 1.712 3.336 16k: 1.712 1.712 1.712 1.712 1.712 1.712 1.712 3.336 32k: 1.712 1.712 1.712 1.712 1.712 1.712 1.712 3.338 64k: 1.713 1.712 1.713 1.712 1.713 1.713 1.713 3.339 128k: 5.143 5.139 5.136 5.139 5.136 5.676 7.185 12.97 256k: 6.045 6.114 6.055 6.132 6.047 6.033 7.536 12.97 512k: 9.826 9.327 9.735 9.325 9.723 9.833 11.48 17.62 1024k: 17.95 17.49 17.71 17.48 17.70 17.59 19.44 28.83 2048k: 25.63 22.69 24.59 22.65 24.39 23.35 27.22 42.27 4096k: 123.6 107.3 122.4 106.5 121.8 105.3 101.1 68.71 8192k: 118.9 87.77 101.6 86.73 101.2 87.96 116.8 141.7 16384k: 195.0 235.6 241.3 236.5 240.5 218.4 143.7 129.0 Executing ramlat on cpu6 (Cortex-A76), results in ns: size: 1x32 2x32 1x64 2x64 1xPTR 2xPTR 4xPTR 8xPTR 4k: 1.713 1.713 1.713 1.713 1.713 1.713 1.714 3.258 8k: 1.713 1.713 1.713 1.713 1.713 1.713 1.713 3.338 16k: 1.713 1.713 1.713 1.713 1.713 1.713 1.713 3.338 32k: 1.713 1.713 1.713 1.713 1.713 1.713 1.713 3.340 64k: 1.714 1.713 1.714 1.713 1.714 1.714 1.714 3.341 128k: 5.144 5.139 5.138 5.140 5.138 5.674 7.208 12.97 256k: 7.151 7.218 7.053 7.219 7.028 7.360 8.523 14.27 512k: 9.953 9.571 9.912 9.566 9.874 10.13 11.56 17.70 1024k: 17.92 17.48 17.72 17.47 17.66 17.83 19.67 28.89 2048k: 22.31 20.79 21.67 20.79 21.64 21.36 24.54 38.13 4096k: 125.5 107.9 124.9 107.6 123.3 107.9 101.5 70.26 8192k: 119.5 87.24 100.2 86.53 99.35 87.86 93.70 97.85 16384k: 131.1 115.1 121.2 111.6 122.0 147.9 154.2 162.8 ########################################################################## Executing benchmark on each cluster individually OpenSSL 3.0.2, built on 15 Mar 2022 (Library: OpenSSL 3.0.2 15 Mar 2022) type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes aes-128-cbc 161383.67k 477860.16k 935484.42k 1236525.06k 1363058.69k 1373650.94k (Cortex-A55) aes-128-cbc 661648.33k 1317811.18k 1690648.32k 1811161.09k 1858052.10k 1862718.81k (Cortex-A76) aes-128-cbc 653606.50k 1304631.21k 1688533.08k 1809831.59k 1856678.57k 1861479.08k (Cortex-A76) aes-192-cbc 153751.49k 425465.19k 761016.75k 950681.94k 1025952.43k 1032088.23k (Cortex-A55) aes-192-cbc 615589.00k 1153821.53k 1432311.47k 1507698.69k 1549727.06k 1552596.99k (Cortex-A76) aes-192-cbc 615469.00k 1151637.78k 1430890.33k 1505556.82k 1548151.47k 1551329.96k (Cortex-A76) aes-256-cbc 149009.40k 391002.26k 656184.06k 794822.31k 846686.89k 850875.73k (Cortex-A55) aes-256-cbc 582298.33k 1026549.01k 1241051.14k 1304572.93k 1328704.17k 1331074.39k (Cortex-A76) aes-256-cbc 593542.07k 1024435.50k 1238821.80k 1303066.97k 1327658.33k 1330113.19k (Cortex-A76) ########################################################################## Executing benchmark single-threaded on cpu0 (Cortex-A55) 7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,8 CPUs LE) LE CPU Freq: - - - - - - - - 2048000000 RAM size: 7690 MB, # CPU hardware threads: 8 RAM usage: 435 MB, # Benchmark threads: 1 Compressing | Decompressing Dict Speed Usage R/U Rating | Speed Usage R/U Rating KiB/s % MIPS MIPS | KiB/s % MIPS MIPS 22: 1118 99 1094 1088 | 21148 98 1845 1806 23: 982 100 1001 1001 | 21159 100 1832 1832 24: 978 99 1060 1052 | 20614 100 1810 1810 25: 915 100 1045 1045 | 19867 100 1768 1768 ---------------------------------- | ------------------------------ Avr: 100 1050 1047 | 99 1814 1804 Tot: 100 1432 1425 Executing benchmark single-threaded on cpu4 (Cortex-A76) 7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,8 CPUs LE) LE CPU Freq: - - - - - - - - - RAM size: 7690 MB, # CPU hardware threads: 8 RAM usage: 435 MB, # Benchmark threads: 1 Compressing | Decompressing Dict Speed Usage R/U Rating | Speed Usage R/U Rating KiB/s % MIPS MIPS | KiB/s % MIPS MIPS 22: 2054 100 1999 1999 | 38000 100 3245 3245 23: 1879 100 1915 1915 | 37182 100 3219 3218 24: 1782 100 1917 1917 | 35843 100 3147 3147 25: 1675 100 1913 1913 | 34176 100 3042 3042 ---------------------------------- | ------------------------------ Avr: 100 1936 1936 | 100 3163 3163 Tot: 100 2550 2549 Executing benchmark single-threaded on cpu6 (Cortex-A76) 7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,8 CPUs LE) LE CPU Freq: - - - 64000000 - - - - - RAM size: 7690 MB, # CPU hardware threads: 8 RAM usage: 435 MB, # Benchmark threads: 1 Compressing | Decompressing Dict Speed Usage R/U Rating | Speed Usage R/U Rating KiB/s % MIPS MIPS | KiB/s % MIPS MIPS 22: 2050 100 1995 1995 | 38205 100 3262 3262 23: 1881 100 1917 1917 | 37152 100 3216 3216 24: 1780 100 1914 1914 | 35806 100 3143 3143 25: 1678 100 1916 1916 | 34211 100 3045 3045 ---------------------------------- | ------------------------------ Avr: 100 1936 1936 | 100 3167 3167 Tot: 100 2551 2551 ########################################################################## Executing benchmark 3 times multi-threaded on CPUs 0-7 7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,8 CPUs LE) LE CPU Freq: - 64000000 - - - 256000000 - - - RAM size: 7690 MB, # CPU hardware threads: 8 RAM usage: 1765 MB, # Benchmark threads: 8 Compressing | Decompressing Dict Speed Usage R/U Rating | Speed Usage R/U Rating KiB/s % MIPS MIPS | KiB/s % MIPS MIPS 22: 14585 742 1912 14189 | 206010 679 2588 17572 23: 13773 729 1926 14034 | 200107 680 2546 17317 24: 13184 754 1880 14176 | 193942 681 2500 17022 25: 12478 761 1873 14248 | 186819 679 2449 16626 ---------------------------------- | ------------------------------ Avr: 746 1898 14162 | 680 2521 17134 Tot: 713 2209 15648 7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,8 CPUs LE) LE CPU Freq: - - - - - - - - - RAM size: 7690 MB, # CPU hardware threads: 8 RAM usage: 1765 MB, # Benchmark threads: 8 Compressing | Decompressing Dict Speed Usage R/U Rating | Speed Usage R/U Rating KiB/s % MIPS MIPS | KiB/s % MIPS MIPS 22: 14672 753 1895 14274 | 204069 677 2573 17406 23: 13807 744 1892 14068 | 199209 680 2536 17239 24: 13416 775 1861 14425 | 193112 680 2493 16949 25: 12412 778 1823 14172 | 186625 679 2445 16609 ---------------------------------- | ------------------------------ Avr: 762 1868 14235 | 679 2512 17051 Tot: 721 2190 15643 7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,8 CPUs LE) LE CPU Freq: - - - - - - - - - RAM size: 7690 MB, # CPU hardware threads: 8 RAM usage: 1765 MB, # Benchmark threads: 8 Compressing | Decompressing Dict Speed Usage R/U Rating | Speed Usage R/U Rating KiB/s % MIPS MIPS | KiB/s % MIPS MIPS 22: 14644 741 1922 14246 | 204560 679 2569 17448 23: 13552 727 1900 13808 | 199081 680 2533 17228 24: 13063 742 1892 14046 | 193545 681 2493 16987 25: 12445 771 1842 14209 | 187078 681 2443 16649 ---------------------------------- | ------------------------------ Avr: 745 1889 14077 | 681 2510 17078 Tot: 713 2199 15578 Compression: 14162,14235,14077 Decompression: 17134,17051,17078 Total: 15648,15643,15578 ########################################################################## Testing maximum cpufreq again, still under full load. System health now: Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp DC(V) 22:40:17: 2400/1800MHz 6.76 91% 1% 89% 0% 0% 0% 66.5°C 9.31 Checking cpufreq OPP for cpu0-cpu3 (Cortex-A55): Cpufreq OPP: 1800 Measured: 1803 (1803.592/1803.553/1802.963) Checking cpufreq OPP for cpu4-cpu5 (Cortex-A76): Cpufreq OPP: 2400 Measured: 2315 (2315.588/2315.121/2314.861) (-3.5%) Checking cpufreq OPP for cpu6-cpu7 (Cortex-A76): Cpufreq OPP: 2400 Measured: 2316 (2316.990/2316.938/2316.574) (-3.5%) ########################################################################## Hardware sensors: gpu_thermal-virtual-0 temp1: +54.5 C littlecore_thermal-virtual-0 temp1: +55.5 C bigcore0_thermal-virtual-0 temp1: +55.5 C tcpm_source_psy_4_0022-i2c-4-22 in0: 9.00 V (min = +9.00 V, max = +9.00 V) curr1: 1.67 A (max = +1.67 A) npu_thermal-virtual-0 temp1: +54.5 C center_thermal-virtual-0 temp1: +54.5 C bigcore1_thermal-virtual-0 temp1: +55.5 C soc_thermal-virtual-0 temp1: +55.5 C (crit = +115.0 C) ########################################################################## Transitions since last boot (13453520ms ago): /sys/devices/platform/dmc/devfreq/dmc: From : To : 528000000106800000015600000002112000000 time(ms) * 528000000: 0 0 0 218 12361740 1068000000: 58 0 0 16 213303 1560000000: 3 17 0 3 84273 2112000000: 158 57 23 0 791323 Total transition : 553 ########################################################################## Thermal source: /sys/devices/virtual/thermal/thermal_zone0/ (soc-thermal) System health while running tinymembench: Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp DC(V) 22:10:41: 2400/1800MHz 0.95 0% 0% 0% 0% 0% 0% 37.9°C 9.34 22:12:41: 2400/1800MHz 1.07 12% 0% 12% 0% 0% 0% 44.4°C 9.39 22:14:41: 2400/1800MHz 1.01 12% 0% 12% 0% 0% 0% 40.7°C 9.35 22:16:41: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 39.8°C 9.39 22:18:41: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 51.8°C 9.28 22:20:41: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.37 22:22:42: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.39 22:24:42: 2400/1800MHz 1.03 12% 0% 12% 0% 0% 0% 55.5°C 9.33 22:26:42: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 46.2°C 9.36 22:28:42: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.39 System health while running ramlat: Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp DC(V) 22:29:50: 2400/1800MHz 1.00 1% 0% 1% 0% 0% 0% 43.5°C 9.32 22:29:59: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.39 22:30:08: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.33 22:30:17: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.34 22:30:26: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.37 22:30:35: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.27 22:30:44: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.33 22:30:54: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.32 22:31:03: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.30 System health while running OpenSSL benchmark: Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp DC(V) 22:31:04: 2400/1800MHz 1.00 1% 0% 1% 0% 0% 0% 44.4°C 9.33 22:31:20: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.34 22:31:36: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.33 22:31:52: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.30 22:32:08: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.39 22:32:24: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.40 22:32:40: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.32 22:32:56: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.39 22:33:12: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.32 22:33:28: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.31 22:33:44: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.34 System health while running 7-zip single core benchmark: Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp DC(V) 22:33:46: 2400/1800MHz 1.00 1% 0% 1% 0% 0% 0% 44.4°C 9.40 22:33:59: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.33 22:34:12: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.40 22:34:25: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.38 22:34:38: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.36 22:34:51: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.34 22:35:04: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.35 22:35:17: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 42.5°C 9.39 22:35:30: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 41.6°C 9.35 22:35:43: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.31 22:35:56: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.36 22:36:09: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.34 22:36:23: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 43.5°C 9.37 22:36:36: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.32 22:36:49: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.34 22:37:02: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.36 22:37:15: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.35 22:37:28: 2400/1800MHz 1.00 12% 0% 12% 0% 0% 0% 44.4°C 9.30 System health while running 7-zip multi core benchmark: Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp DC(V) 22:37:37: 2400/1800MHz 1.00 1% 0% 1% 0% 0% 0% 45.3°C 9.38 22:37:47: 2400/1800MHz 2.08 90% 0% 90% 0% 0% 0% 55.5°C 9.32 22:37:57: 2400/1800MHz 3.21 86% 0% 85% 0% 0% 0% 58.2°C 9.30 22:38:09: 2400/1800MHz 3.19 85% 1% 84% 0% 0% 0% 61.0°C 9.33 22:38:20: 2400/1800MHz 4.33 80% 1% 79% 0% 0% 0% 61.0°C 9.34 22:38:30: 2400/1800MHz 4.57 88% 0% 87% 0% 0% 0% 61.0°C 9.33 22:38:40: 2400/1800MHz 4.66 91% 0% 90% 0% 0% 0% 62.8°C 9.37 22:38:50: 2400/1800MHz 5.25 87% 0% 86% 0% 0% 0% 63.8°C 9.24 22:39:03: 2400/1800MHz 5.82 84% 1% 83% 0% 0% 0% 66.5°C 9.28 22:39:13: 2400/1800MHz 6.02 81% 1% 80% 0% 0% 0% 64.7°C 9.33 22:39:23: 2400/1800MHz 6.62 91% 1% 90% 0% 0% 0% 64.7°C 9.32 22:39:33: 2400/1800MHz 6.39 91% 0% 90% 0% 0% 0% 66.5°C 9.28 22:39:43: 2400/1800MHz 6.34 85% 0% 85% 0% 0% 0% 68.4°C 9.32 22:39:57: 2400/1800MHz 6.52 82% 1% 81% 0% 0% 0% 68.4°C 9.33 22:40:07: 2400/1800MHz 6.53 81% 1% 80% 0% 0% 0% 67.5°C 9.34 22:40:17: 2400/1800MHz 6.76 91% 1% 89% 0% 0% 0% 66.5°C 9.31 ########################################################################## dmesg output while running the benchmarks: [11680.469203] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [11765.108375] dwhdmi-rockchip fde80000.hdmi: use tmds mode [11809.707272] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [11895.908371] dwhdmi-rockchip fde80000.hdmi: use tmds mode [11898.108380] dwhdmi-rockchip fde80000.hdmi: use tmds mode [11929.835311] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [11938.700385] dwhdmi-rockchip fde80000.hdmi: use tmds mode [11940.044367] dwhdmi-rockchip fde80000.hdmi: use tmds mode [11940.540389] dwhdmi-rockchip fde80000.hdmi: use tmds mode [11941.780389] dwhdmi-rockchip fde80000.hdmi: use tmds mode [12049.963775] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [12149.636814] dwhdmi-rockchip fde80000.hdmi: use tmds mode [12151.220823] dwhdmi-rockchip fde80000.hdmi: use tmds mode [12170.092208] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [12290.220781] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [12410.348901] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [12530.473073] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [12659.694239] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [12779.823119] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [12855.092382] dwhdmi-rockchip fde80000.hdmi: use tmds mode [12899.951337] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [13020.079133] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [13062.212371] dwhdmi-rockchip fde80000.hdmi: use tmds mode [13072.116364] dwhdmi-rockchip fde80000.hdmi: use tmds mode [13074.036358] dwhdmi-rockchip fde80000.hdmi: use tmds mode [13074.628381] dwhdmi-rockchip fde80000.hdmi: use tmds mode [13096.756371] dwhdmi-rockchip fde80000.hdmi: use tmds mode [13097.580529] dwhdmi-rockchip fde80000.hdmi: use tmds mode [13140.207523] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [13260.337472] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc [13380.462244] r8125 0004:41:00.0 enP4p65s0: rss get rxnfc ########################################################################## Linux 5.10.72-rockchip-rk3588 (rock-5b) 12/06/22 _aarch64_ (8 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 2.83 0.00 0.13 0.00 0.00 97.04 Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd mmcblk1 1.24 50.66 41.89 0.00 681510 563574 0 mtdblock0 0.00 0.02 0.00 0.00 336 0 0 zram0 0.04 0.17 0.00 0.00 2264 4 0 zram1 0.06 0.03 0.52 0.00 416 7008 0 total used free shared buff/cache available Mem: 7.5Gi 195Mi 7.0Gi 43Mi 301Mi 7.2Gi Swap: 3.8Gi 0B 3.8Gi Filename Type Size Used Priority /dev/zram0 partition 3937420 0 5 CPU sysfs topology (clusters, cpufreq members, clockspeeds) cpufreq min max CPU cluster policy speed speed core type 0 0 0 408 1800 Cortex-A55 / r2p0 1 0 0 408 1800 Cortex-A55 / r2p0 2 0 0 408 1800 Cortex-A55 / r2p0 3 0 0 408 1800 Cortex-A55 / r2p0 4 1 4 408 2400 Cortex-A76 / r4p0 5 1 4 408 2400 Cortex-A76 / r4p0 6 2 6 408 2400 Cortex-A76 / r4p0 7 2 6 408 2400 Cortex-A76 / r4p0 Architecture: aarch64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 8 On-line CPU(s) list: 0-7 Vendor ID: ARM Model name: Cortex-A55 Model: 0 Thread(s) per core: 1 Core(s) per socket: 4 Socket(s): 1 Stepping: r2p0 CPU max MHz: 1800.0000 CPU min MHz: 408.0000 BogoMIPS: 48.00 Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm lrcpc dcpop asimddp Model name: Cortex-A76 Model: 0 Thread(s) per core: 1 Core(s) per socket: 2 Socket(s): 2 Stepping: r4p0 CPU max MHz: 2400.0000 CPU min MHz: 408.0000 BogoMIPS: 48.00 Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm lrcpc dcpop asimddp L1d cache: 384 KiB (8 instances) L1i cache: 384 KiB (8 instances) L2 cache: 2.5 MiB (8 instances) L3 cache: 3 MiB (1 instance) Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; __user pointer sanitization Vulnerability Spectre v2: Not affected Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected SoC guess: Rockchip RK3588/RK3588s (35880000) DMC gov: dmc_ondemand (upthreshold: 40) DT compat: radxa,rock-5b rockchip,rk3588 Compiler: /usr/bin/gcc (Ubuntu 11.3.0-1ubuntu1~22.04) 11.3.0 / aarch64-linux-gnu Userland: arm64 Kernel: 5.10.72-rockchip-rk3588/aarch64 CONFIG_HZ=300 CONFIG_HZ_300=y CONFIG_PREEMPT_NOTIFIERS=y CONFIG_PREEMPT_VOLUNTARY=y raid6: neonx8 gen() 5983 MB/s raid6: neonx8 xor() 4607 MB/s raid6: neonx4 gen() 5992 MB/s raid6: neonx4 xor() 4683 MB/s raid6: neonx2 gen() 5835 MB/s raid6: neonx2 xor() 4378 MB/s raid6: neonx1 gen() 4758 MB/s raid6: neonx1 xor() 3423 MB/s raid6: int64x8 gen() 1474 MB/s raid6: int64x8 xor() 985 MB/s raid6: int64x4 gen() 1925 MB/s raid6: int64x4 xor() 1072 MB/s raid6: int64x2 gen() 2660 MB/s raid6: int64x2 xor() 1453 MB/s raid6: int64x1 gen() 2191 MB/s raid6: int64x1 xor() 1074 MB/s raid6: using algorithm neonx4 gen() 5992 MB/s raid6: .... xor() 4683 MB/s, rmw enabled raid6: using neon recovery algorithm xor: measuring software checksum speed xor: using function: arm64_neon (10689 MB/sec) cpu cpu0: pvtm=1521 cpu cpu0: pvtm-volt-sel=5 cpu cpu4: pvtm=1782 cpu cpu4: pvtm-volt-sel=7 cpu cpu6: pvtm=1784 cpu cpu6: pvtm-volt-sel=7 cpu0/index0: 32K, level: 1, type: Data cpu0/index1: 32K, level: 1, type: Instruction cpu0/index2: 128K, level: 2, type: Unified cpu0/index3: 3072K, level: 3, type: Unified cpu1/index0: 32K, level: 1, type: Data cpu1/index1: 32K, level: 1, type: Instruction cpu1/index2: 128K, level: 2, type: Unified cpu1/index3: 3072K, level: 3, type: Unified cpu2/index0: 32K, level: 1, type: Data cpu2/index1: 32K, level: 1, type: Instruction cpu2/index2: 128K, level: 2, type: Unified cpu2/index3: 3072K, level: 3, type: Unified cpu3/index0: 32K, level: 1, type: Data cpu3/index1: 32K, level: 1, type: Instruction cpu3/index2: 128K, level: 2, type: Unified cpu3/index3: 3072K, level: 3, type: Unified cpu4/index0: 64K, level: 1, type: Data cpu4/index1: 64K, level: 1, type: Instruction cpu4/index2: 512K, level: 2, type: Unified cpu4/index3: 3072K, level: 3, type: Unified cpu5/index0: 64K, level: 1, type: Data cpu5/index1: 64K, level: 1, type: Instruction cpu5/index2: 512K, level: 2, type: Unified cpu5/index3: 3072K, level: 3, type: Unified cpu6/index0: 64K, level: 1, type: Data cpu6/index1: 64K, level: 1, type: Instruction cpu6/index2: 512K, level: 2, type: Unified cpu6/index3: 3072K, level: 3, type: Unified cpu7/index0: 64K, level: 1, type: Data cpu7/index1: 64K, level: 1, type: Instruction cpu7/index2: 512K, level: 2, type: Unified cpu7/index3: 3072K, level: 3, type: Unified ########################################################################## vdd_cpu_big0_s0: 1000 mV (1050 mV max) vdd_cpu_big1_s0: 1000 mV (1050 mV max) vdd_npu_s0: 788 mV (950 mV max) cluster0-opp-table: 408 MHz 675.0 mV (00ff ffff) 600 MHz 675.0 mV (00ff ffff) 816 MHz 675.0 mV (00ff ffff) 1008 MHz 675.0 mV (00ff ffff) 1200 MHz 712.5 mV (00ff ffff) 1416 MHz 762.5 mV (00ff ffff) 1608 MHz 850.0 mV (00ff ffff) 1800 MHz 950.0 mV (00ff ffff) cluster1-opp-table: 408 MHz 675.0 mV (00ff ffff) 600 MHz 675.0 mV (00ff ffff) 816 MHz 675.0 mV (00ff ffff) 1008 MHz 675.0 mV (00ff ffff) 1200 MHz 675.0 mV (00ff ffff) 1416 MHz 725.0 mV (00ff ffff) 1608 MHz 762.5 mV (00ff ffff) 1800 MHz 850.0 mV (00ff ffff) 2016 MHz 925.0 mV (00ff ffff) 2208 MHz 987.5 mV (00ff ffff) 2256 MHz 1000.0 mV (00ff 0000) 2304 MHz 1000.0 mV (00ff 0000) 2352 MHz 1000.0 mV (00ff 0000) 2400 MHz 1000.0 mV (00ff ffff) cluster2-opp-table: 408 MHz 675.0 mV (00ff ffff) 600 MHz 675.0 mV (00ff ffff) 816 MHz 675.0 mV (00ff ffff) 1008 MHz 675.0 mV (00ff ffff) 1200 MHz 675.0 mV (00ff ffff) 1416 MHz 725.0 mV (00ff ffff) 1608 MHz 762.5 mV (00ff ffff) 1800 MHz 850.0 mV (00ff ffff) 2016 MHz 925.0 mV (00ff ffff) 2208 MHz 987.5 mV (00ff ffff) 2256 MHz 1000.0 mV (00ff 0000) 2304 MHz 1000.0 mV (00ff 0000) 2352 MHz 1000.0 mV (00ff 0000) 2400 MHz 1000.0 mV (00ff ffff) dmc-opp-table: 528 MHz 675.0 mV 1068 MHz 725.0 mV 1560 MHz 800.0 mV 2750 MHz 875.0 mV gpu-opp-table: 300 MHz 675.0 mV 400 MHz 675.0 mV 500 MHz 675.0 mV 600 MHz 675.0 mV 700 MHz 700.0 mV 800 MHz 750.0 mV 900 MHz 800.0 mV 1000 MHz 850.0 mV npu-opp-table: 300 MHz 700.0 mV 400 MHz 700.0 mV 500 MHz 700.0 mV 600 MHz 700.0 mV 700 MHz 700.0 mV 800 MHz 750.0 mV 900 MHz 800.0 mV 1000 MHz 850.0 mV | Radxa ROCK 5B | 2400/1800 MHz | 5.10 | Armbian 22.11.1 Jammy arm64 | 15620 | 2551 | 1331070 | 9580 | 24750 | - | RetroPIE