{"id":1847,"date":"2024-02-29T18:57:00","date_gmt":"2024-02-29T18:57:00","guid":{"rendered":"https:\/\/mvermeulen.org\/perf\/?page_id=1847"},"modified":"2024-03-01T01:54:27","modified_gmt":"2024-03-01T01:54:27","slug":"compilebench","status":"publish","type":"page","link":"https:\/\/mvermeulen.org\/perf\/workloads\/phoronix\/compilebench\/","title":{"rendered":"compilebench"},"content":{"rendered":"\n<p>A test of file system operations related to what is used to maintain kernel trees. This has three subtests: compile, initial create and read compiled tree. These tests run quickly and look to be single-threaded.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1280\" height=\"960\" src=\"https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/systemtime-1.png\" alt=\"\" class=\"wp-image-1865\" srcset=\"https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/systemtime-1.png 1280w, https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/systemtime-1-1024x768.png 1024w, https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/systemtime-1-768x576.png 768w\" sizes=\"auto, (max-width: 1280px) 100vw, 1280px\" \/><\/figure>\n\n\n\n<p>Topdown profile shows frontend stalls as highest with relatively low backend stalls.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1280\" height=\"960\" src=\"https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/amdtopdown-1.png\" alt=\"\" class=\"wp-image-1866\" srcset=\"https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/amdtopdown-1.png 1280w, https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/amdtopdown-1-1024x768.png 1024w, https:\/\/mvermeulen.org\/perf\/wp-content\/uploads\/sites\/7\/2024\/03\/amdtopdown-1-768x576.png 768w\" sizes=\"auto, (max-width: 1280px) 100vw, 1280px\" \/><\/figure>\n\n\n\n<p>AMD metrics show little floating point, a high opcache miss rate and an 11% icache miss rate with 1\/5 of the instructions as branches. Frontend latency is high. There is a high amount of output blocks and almost no input blocks.<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>elapsed              89.996\non_cpu               0.044          # 0.71 \/ 16 cores\nutime                7.809\nstime                56.158\nnvcsw                2509           # 24.27%\nnivcsw               7829           # 75.73%\ninblock              8              # 0.09\/sec\nonblock              65538520       # 728234.48\/sec\ncpu-clock            63999083442    # 63.999 seconds\ntask-clock           64004384231    # 64.004 seconds\npage faults          178065         # 2782.075\/sec\ncontext switches     10580          # 165.301\/sec\ncpu migrations       394            # 6.156\/sec\nmajor page faults    2              # 0.031\/sec\nminor page faults    178063         # 2782.044\/sec\nalignment faults     0              # 0.000\/sec\nemulation faults     0              # 0.000\/sec\nbranches             63013567795    # 198.359 branches per 1000 inst\nbranch misses        5940612790     # 9.43% branch miss\nconditional          35303103205    # 111.130 conditional branches per 1000 inst\nindirect             1192983745     # 3.755 indirect branches per 1000 inst\ncpu-cycles           290958943660   # 0.20 GHz\ninstructions         317320090766   # 1.09 IPC\nslots                583966234098   #\nretiring             116929459720   # 20.0% (20.0%)\n-- ucode             571788094      #     0.1%\n-- fastpath          116357671626   #    19.9%\nfrontend             401304724510   # 68.7% (68.7%) high\n-- latency           336924894132   #    57.7%\n-- bandwidth         64379830378    #    11.0%\nbackend              62079412727    # 10.6% (10.6%) low\n-- cpu               13257356583    #     2.3%\n-- memory            48822056144    #     8.4%\nspeculation          3454155649     #  0.6% ( 0.6%) low\n-- branch mispredict 3447845685     #     0.6%\n-- pipeline restart  6309964        #     0.0%\nsmt-contention       198173313      #  0.0% ( 0.0%)\ncpu-cycles           289259190827   # 0.20 GHz\ninstructions         315675605246   # 1.09 IPC\ninstructions         105762868493   # 68.909 l2 access per 1000 inst\nl2 hit from l1       6720971259     # 6.82% l2 miss\nl2 miss from l1      241036077      #\nl2 hit from l2 pf    311085299      #\nl3 hit from l2 pf    88085211       #\nl3 miss from l2 pf   167888574      #\ninstructions         105756026935   # 12.301 float per 1000 inst\nfloat 512            74             # 0.000 AVX-512 per 1000 inst\nfloat 256            668            # 0.000 AVX-256 per 1000 inst\nfloat 128            1300940242     # 12.301 AVX-128 per 1000 inst\nfloat MMX            0              # 0.000 MMX per 1000 inst\nfloat scalar         21             # 0.000 scalar per 1000 inst\ninstructions         317305283501   #\nopcache              96197134370    # 303.169 opcache per 1000 inst\nopcache miss         57752090686    # 60.0% opcache miss rate\nl1 dTLB miss         978514157      # 3.084 L1 dTLB per 1000 inst\nl2 dTLB miss         13665626       # 0.043 L2 dTLB per 1000 inst\ninstructions         318170252018   #\nicache               124538457131   # 391.421 icache per 1000 inst\nicache miss          13704861327    # 11.0% icache miss rate\nl1 iTLB miss         112977480      # 0.355 L1 iTLB per 1000 inst\nl2 iTLB miss         0              # 0.000 L2 iTLB per 1000 inst\ntlb flush            24981          # 0.000 TLB flush per 1000 inst\n<\/code><\/pre>\n\n\n\n<p>Intel metrics show L2 as the largest part of memory bound aspects.<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>elapsed              103.155\non_cpu               0.025          # 0.41 \/ 16 cores\nutime                7.438\nstime                34.585\nnvcsw                6535           # 75.58%\nnivcsw               2112           # 24.42%\ninblock              1336           # 12.95\/sec\nonblock              65527288       # 635229.35\/sec\ncpu-clock            42023931539    # 42.024 seconds\ntask-clock           42032493432    # 42.032 seconds\npage faults          167919         # 3994.981\/sec\ncontext switches     8957           # 213.097\/sec\ncpu migrations       491            # 11.681\/sec\nmajor page faults    0              # 0.000\/sec\nminor page faults    167919         # 3994.981\/sec\nalignment faults     0              # 0.000\/sec\nemulation faults     0              # 0.000\/sec\nbranches             50807657927    # 172.266 branches per 1000 inst\nbranch misses        135894069      # 0.27% branch miss\nconditional          50807670919    # 172.266 conditional branches per 1000 inst\nindirect             1194939709     # 4.052 indirect branches per 1000 inst\nslots                773415976724   #\nretiring             331308877913   # 42.8% (42.8%)\n-- ucode             35167431992    #     4.5%\n-- fastpath          296141445921   #    38.3%\nfrontend             233006981567   # 30.1% (30.1%)\n-- latency           80228342919    #    10.4%\n-- bandwidth         152778638648   #    19.8%\nbackend              181914046624   # 23.5% (23.5%)\n-- cpu               69711483930    #     9.0%\n-- memory            112202562694   #    14.5%\nspeculation          39111851292    #  5.1% ( 5.1%)\n-- branch mispredict 32350542141    #     4.2%\n-- pipeline restart  6761309151     #     0.9%\nsmt-contention       0              #  0.0% ( 0.0%)\ncpu-cycles           129783140930   # 0.08 GHz\ninstructions         294812371436   # 2.27 IPC\nl2 access            16882455472    # 57.333 l2 access per 1000 inst\nl2 miss              2140286539     # 12.68% l2 miss\ncpu-cycles           129742170073   # 23.8% memory latency\nload stalls          25972698653    #  4.1% l1 bound\nl1 miss              20702278744    # 10.4% l2 bound\nl2 miss              7150731886     #  1.4% l3 bound\nl3 miss              5283474262     #  4.1% dram bound\nstore_stalls         4878082314     #  3.8% store bound\n<\/code><\/pre>\n\n\n\n<p>The process overview shows this driven from python with system time as the largest portion.<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>390 processes\n\t 68 clinfo                  17.19     8.64\n\t  3 python2                  6.44    55.32\n\t 38 vulkaninfo               1.14     1.80\n\t  6 glxinfo:gdrv0            0.13     0.07\n\t  6 glxinfo:gl0              0.13     0.07\n\t  4 vulkani:disk$0           0.12     0.19\n\t  6 php                      0.09     0.06\n\t  2 glxinfo                  0.07     0.03\n\t  2 glxinfo:cs0              0.07     0.03\n\t  2 glxinfo:disk$0           0.07     0.03\n\t  2 glxinfo:sh0              0.07     0.03\n\t  2 glxinfo:shlo0            0.07     0.03\n\t  2 llvmpipe-0               0.06     0.10\n\t  2 llvmpipe-1               0.06     0.10\n\t  2 llvmpipe-10              0.06     0.10\n\t  2 llvmpipe-11              0.06     0.10\n\t  2 llvmpipe-12              0.06     0.10\n\t  2 llvmpipe-13              0.06     0.10\n\t  2 llvmpipe-14              0.06     0.10\n\t  2 llvmpipe-15              0.06     0.10\n\t  2 llvmpipe-2               0.06     0.10\n\t  2 llvmpipe-3               0.06     0.10\n\t  2 llvmpipe-4               0.06     0.10\n\t  2 llvmpipe-5               0.06     0.10\n\t  2 llvmpipe-6               0.06     0.10\n\t  2 llvmpipe-7               0.06     0.10\n\t  2 llvmpipe-8               0.06     0.10\n\t  2 llvmpipe-9               0.06     0.10\n\t  6 clang                    0.05     0.07\n\t  3 rocminfo                 0.03     0.00\n\t  1 lspci                    0.01     0.02\n\t 12 sync                     0.00     0.13\n\t 96 sh                       0.00     0.00\n\t 12 gcc                      0.00     0.00\n\t 12 gsettings                0.00     0.00\n\t  9 stat                     0.00     0.00\n\t  8 systemd-detect-          0.00     0.00\n\t  6 llvm-link                0.00     0.00\n\t  5 phoronix-test-s          0.00     0.00\n\t  5 which                    0.00     0.00\n\t  3 compilebench             0.00     0.00\n\t  3 gmain                    0.00     0.00\n\t  3 mkdir                    0.00     0.00\n\t  3 rm                       0.00     0.00\n\t  2 lscpu                    0.00     0.00\n\t  2 uname                    0.00     0.00\n\t  2 xset                     0.00     0.00\n\t  1 cc                       0.00     0.00\n\t  1 date                     0.00     0.00\n\t  1 dconf worker             0.00     0.00\n\t  1 dirname                  0.00     0.00\n\t  1 dmesg                    0.00     0.00\n\t  1 dmidecode                0.00     0.00\n\t  1 grep                     0.00     0.00\n\t  1 ifconfig                 0.00     0.00\n\t  1 ip                       0.00     0.00\n\t  1 lsmod                    0.00     0.00\n\t  1 mktemp                   0.00     0.00\n\t  1 ps                       0.00     0.00\n\t  1 python                   0.00     0.00\n\t  1 python3                  0.00     0.00\n\t  1 qdbus                    0.00     0.00\n\t  1 readlink                 0.00     0.00\n\t  1 realpath                 0.00     0.00\n\t  1 sed                      0.00     0.00\n\t  1 sort                     0.00     0.00\n\t  1 stty                     0.00     0.00\n\t  1 systemctl                0.00     0.00\n\t  1 template.sh              0.00     0.00\n\t  1 wc                       0.00     0.00\n\t  1 xrandr                   0.00     0.00\n0 processes running\n47 maximum processes\n<\/code><\/pre>\n\n\n\n<p>Computation blocks look as follows<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>      12442) compilebench     cpu=9 start=6.64  finish=29.93\n        12443) rm               cpu=11 start=6.64  finish=6.64 \n        12444) mkdir            cpu=15 start=6.64  finish=6.64 \n        12445) which            cpu=12 start=6.64  finish=6.64 \n        12446) python2          cpu=1 start=6.64  finish=29.92\n          12448) sh               cpu=4 start=15.08 finish=16.91\n            12449) sync             cpu=5 start=15.09 finish=16.91\n          12450) sh               cpu=12 start=22.44 finish=23.04\n            12451) sync             cpu=5 start=22.44 finish=23.04\n          12452) sh               cpu=12 start=23.99 finish=24.03\n            12453) sync             cpu=5 start=23.99 finish=24.03\n          12454) sh               cpu=10 start=29.87 finish=29.92\n            12455) sync             cpu=3 start=29.87 finish=29.92\n\n<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"<p>A test of file system operations related to what is used to maintain kernel trees. This has three subtests: compile, initial create and read compiled tree. These tests run quickly and look to be single-threaded. Topdown profile shows frontend stalls <span class=\"excerpt-dots\">&hellip;<\/span> <a class=\"more-link\" href=\"https:\/\/mvermeulen.org\/perf\/workloads\/phoronix\/compilebench\/\"><span class=\"more-msg\">Continue reading &rarr;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":58,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-1847","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/1847","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/comments?post=1847"}],"version-history":[{"count":2,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/1847\/revisions"}],"predecessor-version":[{"id":1867,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/1847\/revisions\/1867"}],"up":[{"embeddable":true,"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/pages\/58"}],"wp:attachment":[{"href":"https:\/\/mvermeulen.org\/perf\/wp-json\/wp\/v2\/media?parent=1847"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}