AMD Ryzen 9 7900 12‐Core Processor - ssvb/tinymembench GitHub Wiki

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                                     :  19262.9 MB/s (0.1%)
 C copy backwards (32 byte blocks)                    :  18240.2 MB/s
 C copy backwards (64 byte blocks)                    :  18245.9 MB/s (0.3%)
 C copy                                               :  19290.8 MB/s
 C copy prefetched (32 bytes step)                    :  17812.3 MB/s
 C copy prefetched (64 bytes step)                    :  17934.6 MB/s
 C 2-pass copy                                        :  19547.3 MB/s (0.4%)
 C 2-pass copy prefetched (32 bytes step)             :  15875.7 MB/s (0.6%)
 C 2-pass copy prefetched (64 bytes step)             :  16191.4 MB/s (0.4%)
 C fill                                               :  67926.8 MB/s (0.2%)
 C fill (shuffle within 16 byte blocks)               :  68009.8 MB/s (2.2%)
 C fill (shuffle within 32 byte blocks)               :  66311.0 MB/s (0.2%)
 C fill (shuffle within 64 byte blocks)               :  57736.6 MB/s
 ---
 standard memcpy                                      :  25923.1 MB/s (0.2%)
 standard memset                                      :  28699.6 MB/s
 ---
 MOVSB copy                                           :  28258.5 MB/s
 MOVSD copy                                           :  28262.4 MB/s
 SSE2 copy                                            :  21805.2 MB/s (0.1%)
 SSE2 nontemporal copy                                :  27560.2 MB/s (0.1%)
 SSE2 copy prefetched (32 bytes step)                 :  23404.9 MB/s (4.7%)
 SSE2 copy prefetched (64 bytes step)                 :  21981.9 MB/s (0.5%)
 SSE2 nontemporal copy prefetched (32 bytes step)     :  27615.3 MB/s (0.1%)
 SSE2 nontemporal copy prefetched (64 bytes step)     :  28018.4 MB/s
 SSE2 2-pass copy                                     :  26161.0 MB/s (0.4%)
 SSE2 2-pass copy prefetched (32 bytes step)          :  23012.8 MB/s (0.4%)
 SSE2 2-pass copy prefetched (64 bytes step)          :  19859.2 MB/s (1.5%)
 SSE2 2-pass nontemporal copy                         :   4606.7 MB/s
 SSE2 fill                                            :  70485.6 MB/s (0.2%)
 SSE2 nontemporal fill                                :  28707.4 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, [MADV_NOHUGEPAGE]
      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 :    0.9 ns          /     1.3 ns 
    131072 :    1.4 ns          /     1.7 ns 
    262144 :    1.6 ns          /     1.8 ns 
    524288 :    2.4 ns          /     2.8 ns 
   1048576 :    3.6 ns          /     4.6 ns 
   2097152 :    6.6 ns          /     8.3 ns 
   4194304 :    8.5 ns          /     9.8 ns 
   8388608 :    9.5 ns          /    10.3 ns 
  16777216 :   11.2 ns          /    12.2 ns 
  33554432 :   22.0 ns          /    30.1 ns 
  67108864 :   59.0 ns          /    83.7 ns 

block size : single random read / dual random read, [MADV_HUGEPAGE]
      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 :    0.9 ns          /     1.3 ns 
    131072 :    1.4 ns          /     1.7 ns 
    262144 :    1.6 ns          /     1.8 ns 
    524288 :    1.8 ns          /     1.9 ns 
   1048576 :    1.9 ns          /     1.9 ns 
   2097152 :    5.4 ns          /     7.0 ns 
   4194304 :    7.1 ns          /     8.4 ns 
   8388608 :    8.0 ns          /     8.8 ns 
  16777216 :    8.4 ns          /     9.0 ns 
  33554432 :    8.8 ns          /     9.5 ns 
  67108864 :   52.1 ns          /    76.7 ns