Skip to content

Instantly share code, notes, and snippets.

@cyring
Created May 21, 2023 08:11
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save cyring/be73a8afee6983bf39fb96c287f646d7 to your computer and use it in GitHub Desktop.
Save cyring/be73a8afee6983bf39fb96c287f646d7 to your computer and use it in GitHub Desktop.
XMRig
XMRig/6.19.2 gcc/12.2.1
@cyring
Copy link
Author

cyring commented May 21, 2023

Processor                                  [AMD Ryzen 9 3950X 16-Core Processor]
|- Architecture                                                   [Zen2/Matisse]
|- Vendor ID                                                      [AuthenticAMD]
|- Firmware                                                         [ 46.72.0-2]
|- Microcode                                                        [0x08701030]
|- Signature                                                           [  8F_71]
|- Stepping                                                            [      0]
|- Online CPU                                                          [ 32/ 32]
|- Base Clock                                                          [100.007]
|- Frequency            (MHz)                      Ratio                        
                 Min   2200.81                    <  22 >                       
                 Max   3501.29                    <  35 >                       
|- Factory                                                             [100.000]
                       3500                       [  35 ]                       
|- Performance                                                                  
   |- P-State                                                                   
                 TGT   3501.29                    <  35 >                       
   |- CPPC                                                                      
                 Min    500.18                    <   5 >                       
                 Max   4701.73                    <  47 >                       
                 TGT   4701.73                    <  47 >                       
|- Turbo Boost                                                         [ UNLOCK]
                 XFR   4801.77                    [  48 ]                       
                 CPB   4701.73                    [  47 ]                       
                  1C   2801.03                    <  28 >                       
                  2C   2200.81                    <  22 >                       
|- Uncore                                                              [   LOCK]
                 CLK   1800.66                    [  18 ]                       
                 MEM   5502.03                    [  55 ]                       

Technologies                                                                    
|- Instruction Cache Unit                                                       
   |- L1 IP Prefetcher                                          L1 HW IP   < ON>
|- Data Cache Unit                                                              
   |- L1 Prefetcher                                                L1 HW   < ON>
   |- L2 Prefetcher                                                L2 HW   <OFF>
|- System Management Mode                                       SMM-Lock   [ ON]
|- Simultaneous Multithreading                                       SMT   [ ON]
|- PowerNow!                                                         CnQ   [ ON]
|- Core C-States                                                     CCx   [ ON]
|- Core Performance Boost                                            CPB   < ON>
|- Watchdog Timer                                                    WDT   < ON>
|- Virtualization                                                    SVM   [ ON]
   |- I/O MMU                                                      AMD-V   [ ON]
   |- Version                                                     [         0.1]
   |- Hypervisor                                                           [OFF]
   |- Vendor ID                                                   [         N/A]

                              Zen UMC  [1440]                              
Controller #0                                                Dual Channel  
 Bus Rate  1833 MHz       Bus Speed 1833 MHz           DDR4 Speed 3667 MT/s
                                                                           
 Cha   CL  RCDr RCDw  RP  RAS   RC  RRDs RRDl FAW  WTRs WTRl  WR  clRR clWW
  #0   16   16   16   16   36   52    4    9   44    5   14   26    5    5 
  #1   16   16   16   16   36   52    4    9   44    5   14   26    5    5 
      CWL  RTP RdWr WrRd scWW sdWW ddWW scRR sdRR ddRR drRR drWW drWR drRRD
  #0   14   14   10    1    1    7    6    1    5    4    0    0    0    0 
  #1   14   14   11    1    1    7    6    1    5    4    0    0    0    0 
      REFI RFC1 RFC2 RFC4 RCPB RPPB  BGS:Alt  Ban  Page  CKE  CMD  GDM  ECC
  #0 14312  631  469  289   0    0   OFF  ON  R1W1   0    0   1T    ON   0 
  #1 14312  631  469  289   0    0   OFF  ON  R1W1   0    0   1T    ON   0 
      MRD:PDA   MOD:PDA  WRMPR STAG PDM RDDATA WRD  WRL  RDL  XS   XP CPDED
  #0    8  19    28  28    24  215 0:F:0   11   2    9   26  661   11    4 
  #1    8  19    28  28    24  215 0:F:0   11   2    9   28  661   11    4 
                                                                           
 DIMM Geometry for channel #0                                              
      Slot Bank Rank     Rows   Columns    Memory Size (MB)                
       #0                                                                  
       #1    16    2     65536      1024          16384   F4-3600C16-16GTZN
 DIMM Geometry for channel #1                                              
      Slot Bank Rank     Rows   Columns    Memory Size (MB)                
       #0                                                                  
       #1    16    2     65536      1024          16384   F4-3600C16-16GTZN

@cyring
Copy link
Author

cyring commented May 21, 2023

(13318.8 h/s)

$ xmrig --bench=1M --config=/etc/xmrig/xmrig.conf --randomx-wrmsr=-1 --randomx-no-rdmsr --cpu-no-yield
 * ABOUT        XMRig/6.19.2 gcc/12.2.1
 * LIBS         libuv/1.44.2 OpenSSL/3.0.8 hwloc/2.9.1
 * HUGE PAGES   supported
 * 1GB PAGES    disabled
 * CPU          AMD Ryzen 9 3950X 16-Core Processor (1) 64-bit AES
                L2:8.0 MB L3:64.0 MB 16C/32T NUMA:1
 * MEMORY       1.2/31.3 GB (4%)
 * DONATE       0%
 * ASSEMBLY     auto:ryzen
 * POOL #1      benchmark algo auto
 * COMMANDS     hashrate, pause, resume, results, connection
 * OPENCL       disabled
 * CUDA         disabled
[2023-05-21 10:07:38.593]  bench    start benchmark hashes 1M algo rx/0
[2023-05-21 10:07:38.593]  cpu      use argon2 implementation AVX2
[2023-05-21 10:07:38.593]  randomx  init dataset algo rx/0 (32 threads) seed 0000000000000000...
[2023-05-21 10:07:38.593]  randomx  allocated 2336 MB (2080+256) huge pages 0% 0/1168 +JIT (0 ms)
[2023-05-21 10:07:39.700]  randomx  dataset ready (1107 ms)
[2023-05-21 10:07:39.700]  cpu      use profile  rx  (32 threads) scratchpad 2048 KB
[2023-05-21 10:07:39.707]  cpu      READY threads 32/32 (32) huge pages 0% 0/32 memory 65536 KB (7 ms)
[2023-05-21 10:08:39.714]  miner    speed 10s/60s/15m 13317.0 n/a n/a H/s max 13330.6 H/s
[2023-05-21 10:08:39.714]  bench    79.94% 799369/1000000 (60.012s)
[2023-05-21 10:08:54.783]  bench    benchmark finished in 75.082 seconds (13318.8 h/s) hash sum = 898B6E0431C28A6B
[2023-05-21 10:08:54.783]  bench    press Ctrl+C to exit
[2023-05-21 10:08:55.216]  cpu      stopped (3 ms)
[2023-05-21 10:09:19.097]  signal   Ctrl+C received, exiting

@cyring
Copy link
Author

cyring commented May 21, 2023

(16780.2 h/s)

|- Collaborative Processor Performance Control                  CPPC       [N/A]
$ cat /proc/cmdline 
hugepagesz=1G hugepages=3

$ xmrig --bench=1M --config=/etc/xmrig/xmrig.conf --randomx-wrmsr=-1 --randomx-no-rdmsr --huge-pages-jit --randomx-1gb-pages --cpu-memory-pool=-1 --cpu-no-yield
 * ABOUT        XMRig/6.19.2 gcc/12.2.1
 * LIBS         libuv/1.44.2 OpenSSL/3.0.8 hwloc/2.9.1
 * HUGE PAGES   supported
 * 1GB PAGES    supported
 * CPU          AMD Ryzen 9 3950X 16-Core Processor (1) 64-bit AES
                L2:8.0 MB L3:64.0 MB 16C/32T NUMA:1
 * MEMORY       4.2/31.3 GB (13%)
 * DONATE       0%
 * ASSEMBLY     auto:ryzen
 * POOL #1      benchmark algo auto
 * COMMANDS     hashrate, pause, resume, results, connection
 * OPENCL       disabled
 * CUDA         disabled
[2023-05-21 10:57:39.146]  bench    start benchmark hashes 1M algo rx/0
[2023-05-21 10:57:39.146]  cpu      use argon2 implementation AVX2
[2023-05-21 10:57:39.146]  randomx  init dataset algo rx/0 (32 threads) seed 0000000000000000...
[2023-05-21 10:57:39.358]  randomx  allocated 3072 MB (2080+256) huge pages 100% 3/3 +JIT (213 ms)
[2023-05-21 10:57:40.419]  randomx  dataset ready (1061 ms)
[2023-05-21 10:57:40.419]  cpu      use profile  rx  (32 threads) scratchpad 2048 KB
[2023-05-21 10:57:40.426]  cpu      READY threads 32/32 (32) huge pages 0% 0/32 memory 65536 KB (6 ms)
[2023-05-21 10:58:40.015]  bench    benchmark finished in 59.594 seconds (16780.2 h/s) hash sum = 898B6E0431C28A6B
[2023-05-21 10:58:40.015]  bench    press Ctrl+C to exit
[2023-05-21 10:58:40.458]  miner    speed 10s/60s/15m 15988.6 n/a n/a H/s max 16795.0 H/s
[2023-05-21 10:58:40.458]  bench      inf% 1000000/0 (60.037s)
[2023-05-21 10:58:40.458]  cpu      stopped (1 ms)

@cyring
Copy link
Author

cyring commented May 21, 2023

(16896.7 h/s)

$ cat /proc/cmdline
hugepagesz=1G hugepages=4

$ cat /etc/sysctl.d/99-sysctl.conf
vm.nr_hugepages=32

$ xmrig --bench=1M --config=/etc/xmrig/xmrig.conf --randomx-wrmsr=-1 --randomx-no-rdmsr --huge-pages-jit --randomx-1gb-pages --cpu-memory-pool=-1 --cpu-no-yield
 * ABOUT        XMRig/6.19.2 gcc/12.2.1
 * LIBS         libuv/1.44.2 OpenSSL/3.0.8 hwloc/2.9.1
 * HUGE PAGES   supported
 * 1GB PAGES    supported
 * CPU          AMD Ryzen 9 3950X 16-Core Processor (1) 64-bit AES
                L2:8.0 MB L3:64.0 MB 16C/32T NUMA:1
 * MEMORY       5.3/31.3 GB (17%)
 * DONATE       0%
 * ASSEMBLY     auto:ryzen
 * POOL #1      benchmark algo auto
 * COMMANDS     hashrate, pause, resume, results, connection
 * OPENCL       disabled
 * CUDA         disabled
[2023-05-21 11:20:50.123]  bench    start benchmark hashes 1M algo rx/0
[2023-05-21 11:20:50.123]  cpu      use argon2 implementation AVX2
[2023-05-21 11:20:50.123]  randomx  init dataset algo rx/0 (32 threads) seed 0000000000000000...
[2023-05-21 11:20:50.336]  randomx  allocated 3072 MB (2080+256) huge pages 100% 3/3 +JIT (213 ms)
[2023-05-21 11:20:51.395]  randomx  dataset ready (1059 ms)
[2023-05-21 11:20:51.395]  cpu      use profile  rx  (32 threads) scratchpad 2048 KB
[2023-05-21 11:20:51.404]  cpu      READY threads 32/32 (32) huge pages 97% 31/32 memory 65536 KB (9 ms)
[2023-05-21 11:21:50.580]  bench    benchmark finished in 59.183 seconds (16896.7 h/s) hash sum = 898B6E0431C28A6B
[2023-05-21 11:21:50.580]  bench    press Ctrl+C to exit
[2023-05-21 11:21:50.944]  cpu      stopped (1 ms)

@cyring
Copy link
Author

cyring commented May 21, 2023

(16931.4 h/s)

$ cat /proc/cmdline 
hugepagesz=1G hugepages=3
$ sysctl vm.nr_hugepages
vm.nr_hugepages = 32
$ cat /sys/devices/system/node/node0/hugepages/hugepages-1048576kB/nr_hugepages 
3
$ cat /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages 
32

$ xmrig --bench=1M --config=/etc/xmrig/xmrig.conf --randomx-wrmsr=-1 --randomx-no-rdmsr --huge-pages-jit --randomx-1gb-pages --hugepage-size=-1 --cpu-memory-pool=-1 --cpu-no-yield
 * ABOUT        XMRig/6.19.2 gcc/12.2.1
 * LIBS         libuv/1.44.2 OpenSSL/3.0.8 hwloc/2.9.1
 * HUGE PAGES   supported
 * 1GB PAGES    supported
 * CPU          AMD Ryzen 9 3950X 16-Core Processor (1) 64-bit AES
                L2:23.5 MB L3:560.0 MB 16C/32T NUMA:1
 * MEMORY       4.6/31.3 GB (15%)
 * DONATE       0%
 * ASSEMBLY     auto:ryzen
 * POOL #1      benchmark algo auto
 * COMMANDS     hashrate, pause, resume, results, connection
 * OPENCL       disabled
 * CUDA         disabled
[2023-05-21 18:47:39.372]  bench    start benchmark hashes 1M algo rx/0
[2023-05-21 18:47:39.372]  cpu      use argon2 implementation AVX2
[2023-05-21 18:47:39.372]  randomx  init dataset algo rx/0 (32 threads) seed 0000000000000000...
[2023-05-21 18:47:39.587]  randomx  allocated 3072 MB (2080+256) huge pages 100% 3/3 +JIT (216 ms)
[2023-05-21 18:47:40.655]  randomx  dataset ready (1067 ms)
[2023-05-21 18:47:40.655]  cpu      use profile  rx  (32 threads) scratchpad 2048 KB
[2023-05-21 18:47:40.676]  cpu      READY threads 32/32 (32) huge pages 97% 31/32 memory 65536 KB (21 ms)
[2023-05-21 18:48:39.719]  bench    benchmark finished in 59.062 seconds (16931.4 h/s) hash sum = 898B6E0431C28A6B
[2023-05-21 18:48:39.719]  bench    press Ctrl+C to exit
[2023-05-21 18:48:40.171]  cpu      stopped (1 ms)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment