1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489
|
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en-us" xml:lang="en-us">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"></meta>
<meta http-equiv="X-UA-Compatible" content="IE=edge"></meta>
<meta name="copyright" content="(C) Copyright 2005"></meta>
<meta name="DC.rights.owner" content="(C) Copyright 2005"></meta>
<meta name="DC.Type" content="concept"></meta>
<meta name="DC.Title" content="Nsight Compute CLI"></meta>
<meta name="abstract" content="The user manual for the NVIDIA Nsight Compute Command Line Interface."></meta>
<meta name="description" content="The user manual for the NVIDIA Nsight Compute Command Line Interface."></meta>
<meta name="DC.Coverage" content="Nsight Compute"></meta>
<meta name="DC.subject" content="Nsight Compute CLI"></meta>
<meta name="keywords" content="Nsight Compute CLI"></meta>
<meta name="DC.Format" content="XHTML"></meta>
<meta name="DC.Identifier" content="abstract"></meta>
<link rel="stylesheet" type="text/css" href="../common/formatting/commonltr.css"></link>
<link rel="stylesheet" type="text/css" href="../common/formatting/site.css"></link>
<title>Nsight Compute CLI :: Nsight Compute Documentation</title>
<!--[if lt IE 9]>
<script src="../common/formatting/html5shiv-printshiv.min.js"></script>
<![endif]-->
<script type="text/javascript" charset="utf-8" src="../common/scripts/tynt/tynt.js"></script>
-->
<script src="https://assets.adobedtm.com/5d4962a43b79/c1061d2c5e7b/launch-191c2462b890.min.js"></script>
<script type="text/javascript" charset="utf-8" src="../common/formatting/jquery.min.js"></script>
<script type="text/javascript" charset="utf-8" src="../common/formatting/jquery.ba-hashchange.min.js"></script>
<script type="text/javascript" charset="utf-8" src="../common/formatting/jquery.scrollintoview.min.js"></script>
<script type="text/javascript" src="../search/htmlFileList.js"></script>
<script type="text/javascript" src="../search/htmlFileInfoList.js"></script>
<script type="text/javascript" src="../search/nwSearchFnt.min.js"></script>
<script type="text/javascript" src="../search/stemmers/en_stemmer.min.js"></script>
<script type="text/javascript" src="../search/index-1.js"></script>
<script type="text/javascript" src="../search/index-2.js"></script>
<script type="text/javascript" src="../search/index-3.js"></script>
<link rel="canonical" href="https://developer.nvidia.com/NsightComputeCli/index.html"></link>
<link rel="stylesheet" type="text/css" href="../common/formatting/qwcode.highlight.css"></link>
</head>
<body>
<header id="header"><span id="company">NVIDIA</span><span id="site-title">Nsight Compute Documentation</span><form id="search" method="get" action="search">
<input type="text" name="search-text"></input><fieldset id="search-location">
<legend>Search In:</legend>
<label><input type="radio" name="search-type" value="site"></input>Entire Site</label>
<label><input type="radio" name="search-type" value="document"></input>Just This Document</label></fieldset>
<button type="reset">clear search</button>
<button id="submit" type="submit">search</button></form>
</header>
<div id="site-content">
<nav id="site-nav">
<div class="category closed"><a href="../index.html" title="The root of the site.">Nsight Compute
v2022.3.0</a></div>
<div class="category"><a href="index.html" title="Nsight Compute CLI">Nsight Compute CLI</a></div>
<ul>
<li>
<div class="section-link"><a href="#introduction">1. Introduction</a></div>
</li>
<li>
<div class="section-link"><a href="#quick-start">2. Quickstart</a></div>
</li>
<li>
<div class="section-link"><a href="#usage">3. Usage</a></div>
<ul>
<li>
<div class="section-link"><a href="#modes">3.1. Modes</a></div>
</li>
<li>
<div class="section-link"><a href="#multi-process">3.2. Multi-Process Support</a></div>
</li>
<li>
<div class="section-link"><a href="#output-pages">3.3. Output Pages</a></div>
</li>
<li>
<div class="section-link"><a href="#profile-import">3.4. Profile Import</a></div>
</li>
<li>
<div class="section-link"><a href="#metrics-and-units">3.5. Metrics and Units</a></div>
</li>
<li>
<div class="section-link"><a href="#nvtx-filtering">3.6. NVTX Filtering</a></div>
</li>
</ul>
</li>
<li>
<div class="section-link"><a href="#command-line-options">4. Command Line Options</a></div>
<ul>
<li>
<div class="section-link"><a href="#command-line-options-general">4.1. General</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-launch">4.2. Launch</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-attach">4.3. Attach</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-profile">4.4. Profile</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-sampling">4.5. Sampling</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-file">4.6. File</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-console-output">4.7. Console Output</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-response-file">4.8. Response File</a></div>
</li>
<li>
<div class="section-link"><a href="#command-line-options-file-macros">4.9. File Macros</a></div>
</li>
</ul>
</li>
<li>
<div class="section-link"><a href="#environment-variables">5. Environment Variables</a></div>
</li>
<li>
<div class="section-link"><a href="#nvprof-guide">6. Nvprof Transition Guide</a></div>
<ul>
<li>
<div class="section-link"><a href="#nvprof-trace">6.1. Trace</a></div>
</li>
<li>
<div class="section-link"><a href="#nvprof-metric-collection">6.2. Metric Collection</a></div>
</li>
<li>
<div class="section-link"><a href="#nvprof-metric-comparison">6.3. Metric Comparison</a></div>
</li>
<li>
<div class="section-link"><a href="#nvprof-event-comparison">6.4. Event Comparison</a></div>
</li>
<li>
<div class="section-link"><a href="#nvprof-filtering">6.5. Filtering</a></div>
</li>
<li>
<div class="section-link"><a href="#nvprof-output">6.6. Output</a></div>
</li>
<li>
<div class="section-link"><a href="#nvprof-launch-attach">6.7. Launch and Attach</a></div>
</li>
</ul>
</li>
</ul>
</nav>
<div id="resize-nav"></div>
<nav id="search-results">
<h2>Search Results</h2>
<ol></ol>
</nav>
<div id="contents-container">
<div id="breadcrumbs-container">
<div id="release-info">Nsight Compute CLI
(<a href="../pdf/NsightComputeCli.pdf">PDF</a>)
-
v2022.3.0
(<a href="https://developer.nvidia.com">older</a>)
-
Last updated August 24, 2022
-
<a href="mailto:devtools@nvidia.com?subject=Nsight Compute Documentation Feedback: Nsight Compute CLI">Send Feedback</a>
-
<span class="st_facebook"></span><span class="st_twitter"></span><span class="st_linkedin"></span><span class="st_reddit"></span><span class="st_slashdot"></span><span class="st_tumblr"></span><span class="st_sharethis"></span></div>
</div>
<article id="contents">
<div class="topic nested0" id="abstract"><a name="abstract" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#abstract" name="abstract" shape="rect">Nsight Compute CLI</a></h2>
<div class="body conbody">
<p class="shortdesc">The user manual for the NVIDIA Nsight Compute Command Line Interface.</p>
<p class="p">
NVIDIA Nsight Compute Command Line Interface (CLI) manual.
Information on workflows and options for the command line, including multi-process profiling and NVTX filtering.
Transitions guide for Nvprof.
</p>
</div>
</div>
<div class="topic concept nested0" id="introduction"><a name="introduction" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#introduction" name="introduction" shape="rect">1. Introduction</a></h2>
<div class="body conbody">
<p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> (<span class="keyword">ncu</span>) provides a non-interactive way
to profile applications from the command line.
It can print the results directly on the command line or store them in a report file.
It can also be used to simply launch the target application (see <a class="xref" href="index.html#command-line-options-general" shape="rect">General</a> for details)
and later attach with <span class="keyword">NVIDIA Nsight Compute</span> or another <span class="keyword">ncu</span> instance.
</p>
<p class="p">
For users migrating from nvprof to <span class="keyword">NVIDIA Nsight Compute</span>, please additionally see the <a class="xref" href="index.html#nvprof-guide" shape="rect">Nvprof Transition Guide</a>
for comparison of features and workflows.
</p>
</div>
</div>
<div class="topic concept nested0" id="quick-start"><a name="quick-start" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#quick-start" name="quick-start" shape="rect">2. Quickstart</a></h2>
<div class="body conbody">
<ol class="ol">
<li class="li"><strong class="ph b">Launch the target application with the command line profiler</strong><p class="p"> The command line profiler launches the target application, instruments the
target API, and collects profile results for the specified kernels. The CLI
executable is called <span class="keyword">ncu</span>. A shortcut with
this name is located in the base directory of the <span class="keyword">NVIDIA Nsight Compute</span> installation. The actual executable is located in
the folder <samp class="ph codeph">target\windows-desktop-win7-x64</samp> on Windows or
<samp class="ph codeph">target/linux-desktop-glibc_2_11_3-x64</samp> on Linux. By default,
<span class="keyword">NVIDIA Nsight Compute</span> is installed in
<samp class="ph codeph">/usr/local/cuda-<cuda-version>/NsightCompute-<version></samp>
on Linux and in <samp class="ph codeph">C:\Program Files\NVIDIA Corporation\Nsight Compute
<version></samp> on Windows.
</p>
<div class="p"> To collect the default set of data for all kernel launches in the target
application, launch:
<pre class="pre screen" xml:space="preserve">$ <span class="keyword">ncu</span> -o profile CuVectorAddMulti.exe</pre></div>
<div class="p"> The application runs in instrumented mode and for each kernel launch, a profile
result is created. The results are written by default to profile.nsight-cuprof.
Each output from the compute profiler starts with <samp class="ph codeph">==PROF==</samp> The
other lines are output from the application itself. For each profiled kernel,
the name of the kernel function and the progress of data collection is shown. To
collect all requested profile information, it may be required to replay the
kernels multiple times. The total number of replay passes per kernel is shown
after profiling has completed.
<pre class="pre screen" xml:space="preserve">
[Vector addition of 1144477 elements]
==PROF== Connected to process 5268
Copy input data from the host memory to the CUDA device
CUDA kernel launch A with 4471 blocks of 256 threads
==PROF== Profiling "vectorAdd_A" - 0: 0%....50%....100% - 46 passes
CUDA kernel launch B with 4471 blocks of 256 threads
==PROF== Profiling "vectorAdd_B" - 1: 0%....50%....100% - 46 passes
Copy output data from the CUDA device to the host memory
Done
==PROF== Disconnected from process 5268
==PROF== Report: profile.ncu-rep
</pre></div>
</li>
<li class="li"><strong class="ph b">Customizing data collection</strong><p class="p">
Options are available to specify for which kernels data should be collected.
<samp class="ph codeph">-c</samp> limits the number of kernel launches collected.
<samp class="ph codeph">-s</samp> skips the given number of kernels before data collection starts.
<samp class="ph codeph">-k</samp> allows you to filter the kernels by a regex match of their names.
<samp class="ph codeph">--kernel-id</samp> allows you to filter kernels by context, stream,
name and invocation, similar to nvprof.
</p>
<p class="p">
To limit what should be collected for each kernel launch, specify the exact *.section (files) by their identifier using <samp class="ph codeph">--section</samp>.
Each section file defines a set of metrics to be collected, grouped logically to solve a specific performance question.
By default, the sections associated with the default set are collected.
Use <samp class="ph codeph">--list-sets</samp> to see the list of currently available sets.
Use <samp class="ph codeph">--list-sections</samp> to see the list of currently available sections.
The default search directory and location of pre-defined section files is also called <samp class="ph codeph">sections/</samp>.
See the <a class="xref" href="../ProfilingGuide/index.html#sets-and-sections" shape="rect">Profiling Guide</a> for more details.
</p>
<p class="p">
Alternatively, you can collect a set of individual metrics using <samp class="ph codeph">--metrics</samp>.
The available metrics can be queried using <samp class="ph codeph">--query-metrics</samp>.
For an explanation of the naming conventions and structuring of metrics, see
<a class="xref" href="../ProfilingGuide/index.html#metrics-structure" shape="rect">Metrics Structure</a>.
</p>
<p class="p">
Most metrics in <span class="keyword">NVIDIA Nsight Compute</span> are named using a base name and various suffixes, e.g. <dfn class="term">sm__throughput.avg.pct_of_peak_sustained_elapsed</dfn>.
The base name is <dfn class="term">sm__throughput</dfn> and the suffix is <dfn class="term">avg.pct_of_peak_sustained_elapsed</dfn>.
This is because most metrics follow the same structure and have the same set of suffixes.
You need to pass the base or full name to <span class="keyword">NVIDIA Nsight Compute</span> when selecting a metric for profiling.
Use <samp class="ph codeph">--query-metrics-mode suffix --metrics <metrics list></samp> to see the full names for the chosen metrics.
</p>
<p class="p">
Some additional metrics do not follow this structured naming. They are documented in the
<a class="xref" href="../ProfilingGuide/index.html#metrics-reference" shape="rect">Metrics Reference</a>.
</p>
</li>
<li class="li"><strong class="ph b">Changing command line output</strong><p class="p">
By default, a temporary file is used to store profiling results, and data is printed to the command line.
To permanently store the profiler report, use <samp class="ph codeph">-o</samp> to specify the output filename.
</p>
<p class="p">
Besides storing results in a report file, the command line profiler can print
results using different pages. Those pages correspond to the respective pages in the UI’s report.
By default, the <a class="xref" href="index.html#output-pages" shape="rect">Details page</a> is printed, if no explicit output file is specified.
To select a different page or print in addition to storing in an explicit file,
use the <samp class="ph codeph">--page=<Page></samp> command. Currently, the following pages are supported: <samp class="ph codeph">details, raw, source</samp>.
</p>
<p class="p">
Use <samp class="ph codeph">--csv</samp> to make any output comma separated and easier to process further.
See <a class="xref" href="index.html#command-line-options-console-output" shape="rect">Console Output</a> for further options, e.g. summary views.
</p>
</li>
<li class="li"><strong class="ph b">Open the report in the UI</strong><p class="p">
The UI executable is called <span class="keyword">ncu-ui</span>. A shortcut with this name is located in the base directory
of the <span class="keyword">NVIDIA Nsight Compute</span> installation. The actual executable is located in the folder
<samp class="ph codeph">host\windows-desktop-win7-x64</samp> on Windows or <samp class="ph codeph">host/linux-desktop-glibc_2_11_3-x64</samp> on Linux.
In the UI window, close the <dfn class="term">Connection</dfn> dialog and open the report file through <dfn class="term">File > Open</dfn>,
by dragging the report file into <span class="keyword">NVIDIA Nsight Compute</span>.
</p>
<p class="p">
You can also specify the report file as a command line parameter to the executable, i.e. as
<samp class="ph codeph"><span class="keyword">ncu-ui</span> <MyReport.ncu-rep></samp>.
Alternatively, when using <span class="keyword">NVIDIA Nsight Compute CLI</span> on a platform with host support,
<samp class="ph codeph">--open-in-ui</samp> can be used directly with <span class="keyword">ncu</span>
to open a collected report in the user interface.
</p>
<p class="p">
The report opens in a new document window.
For more information about the report, see the <a class="xref" href="../NsightCompute/index.html#profiler-report" shape="rect">Profiler Report</a>
for collecting profile information through <span class="keyword">NVIDIA Nsight Compute</span>.
</p>
</li>
</ol>
</div>
</div>
<div class="topic concept nested0" id="usage"><a name="usage" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#usage" name="usage" shape="rect">3. Usage</a></h2>
<div class="body conbody"></div>
<div class="topic concept nested1" id="modes"><a name="modes" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#modes" name="modes" shape="rect">3.1. Modes</a></h3>
<div class="body conbody">
<p class="p">
Modes change the fundamental behavior of the command line profiler.
Depending on which mode is chosen, different <a class="xref" href="index.html#command-line-options" shape="rect">Command Line Options</a> become available.
For example, <a class="xref" href="index.html#command-line-options-launch" shape="rect">Launch</a> is invalid if the <dfn class="term">Attach</dfn> mode is selected.
</p>
<ul class="ul">
<li class="li">
<p class="p"><strong class="ph b">Launch-and-attach:</strong>
The target application is launched on the local system with the tool's injection libraries.
Depending on which profiling options are chosen, selected kernels in
the application are profiled and the results printed to the console or stored in a report file.
The tool exits once the target application finishes or crashes, and once all results are processed.
</p>
<p class="p">
This is the default, and the only mode that supports profiling of child processes on selected platforms.
</p>
</li>
<li class="li">
<p class="p"><strong class="ph b">Launch:</strong>
The target application is launched on the local system with the tool's injection libraries.
As soon as the first intercepted API call is reached (commonly <samp class="ph codeph">cuInit()</samp>), all application threads are suspended.
The application now expects a tool to attach for profiling.
You can attach using <span class="keyword">NVIDIA Nsight Compute</span> or using the command line profiler's <dfn class="term">Attach</dfn> mode.
</p>
</li>
<li class="li">
<p class="p"><strong class="ph b">Attach:</strong>
The tool tries to connect to a target application previously launched using <span class="keyword">NVIDIA Nsight Compute</span>
or using the command line profiler's <dfn class="term">Launch</dfn> mode.
The tool can attach to a target on the local system or using a remote connection.
</p>
</li>
</ul>
</div>
</div>
<div class="topic concept nested1" id="multi-process"><a name="multi-process" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#multi-process" name="multi-process" shape="rect">3.2. Multi-Process Support</a></h3>
<div class="body conbody">
<p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> supports profiling multi-process applications on the following platforms:
x86_64 Windows, x86_64 Linux, DRIVE OS Linux, DRIVE OS QNX, PowerPC.
See the <a class="xref" href="index.html#command-line-options-launch" shape="rect">Launch</a> options on how to enable this feature.
</p>
<p class="p"> On x86_64 Windows, <span class="keyword">NVIDIA Nsight Compute CLI</span> supports profiling 64-bit
processes launched from 32-bit applications by default . On x86_64 Linux, launching from
32-bit applications requires you to enable the <samp class="ph codeph">support-32bit</samp> option,
and the required 32-bit libraries must be installed on your system. On DRIVE OS Linux, DRIVE OS QNX
and PowerPC, tracking of 32-bit applications is not supported. <dfn class="term">Profiling</dfn> of 32-bit
processes is not supported on any platform.
</p>
<p class="p"><strong class="ph b">Profiling MPI applications is a special case of multi-process profiling.</strong><br clear="none"></br><span class="keyword">NVIDIA Nsight Compute CLI</span> can be used to profile applications launched with the <samp class="ph codeph">mpirun</samp> command.
</p>
<ul class="ul">
<li class="li">
<div class="p">
To profile all ranks on a node and store all the profiling data in a single report file:
<pre xml:space="preserve">
ncu --target-processes all -o <report-name> mpirun [mpi arguments] <app> [app arguments]
</pre></div>
</li>
<li class="li">
<div class="p">
To profile multi-node submissions, one instance of <span class="keyword">NVIDIA Nsight Compute CLI</span> can be used per node.
Ensure that you specify unique report files per rank.
<pre xml:space="preserve">
mpirun [mpi arguments] ncu -o report_%q{OMPI_COMM_WORLD_RANK} <app> [app arguments]
</pre></div>
</li>
<li class="li">
<div class="p">
To profile a single rank one can use a wrapper script. The following script (called "wrap.sh") profiles rank 0 only:
<pre xml:space="preserve">
#!/bin/bash
if [[ $OMPI_COMM_WORLD_RANK == 0 ]]; then
ncu -o report_${OMPI_COMM_WORLD_RANK} --target-processes all "$@"
else
"$@"
fi
</pre>
and then execute:
<pre xml:space="preserve">
mpirun [mpi arguments] ./wrap.sh <app> [app arguments]
</pre></div>
</li>
</ul>
</div>
</div>
<div class="topic concept nested1" id="output-pages"><a name="output-pages" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#output-pages" name="output-pages" shape="rect">3.3. Output Pages</a></h3>
<div class="body conbody">
<p class="p">
The command line profiler supports printing results to the console using various pages.
Each page has an equivalent in <span class="keyword">NVIDIA Nsight Compute</span>'s <dfn class="term">Profiler Report</dfn>.
In the command line profiler, they are slightly adapted to fit console output.
To select a page, use the <samp class="ph codeph">--page</samp> option.
By default, the details page is used.
Note that if <samp class="ph codeph">--page</samp> is not used but <samp class="ph codeph">--export</samp> is, no results will be printed to the console.
</p>
<ul class="ul">
<li class="li">
<p class="p"><strong class="ph b">Details:</strong>
This page represents <span class="keyword">NVIDIA Nsight Compute</span>'s <dfn class="term">Details</dfn> page.
For every profiled kernel launch, each collected is printed as section as a three-column table,
followed by any rule results applied to this section.
Rule results not associated with any section are printed after the kernel’s sections.
</p>
<p class="p">
The first section table column shows the metric name.
If the metric was given a label in the section, it is used instead.
The second column shows the metric unit, if available.
The third column shows the unit value.
Both metric unit and value are automatically adjusted to the most fitting order of magnitude.
By default, only metrics defined in section headers are shown.
This can be changed by passing the <samp class="ph codeph">--details-all</samp> option on the command line.
</p>
<p class="p">
Some metrics will show multiple values, separated by ";", e.g.
memory_l2_transactions_global Kbytes 240; 240; 240; 240; 240.
Those are instanced metrics, which have one value per represented instance.
An instance can be a streaming multiprocessor, an assembly source line, etc.
</p>
</li>
<li class="li">
<p class="p"><strong class="ph b">Raw:</strong>
This page represents <span class="keyword">NVIDIA Nsight Compute</span>'s <dfn class="term">Raw</dfn> page.
For every profiled kernel launch, each collected metric is printed as a three-column table.
Besides metrics from sections, this includes automatically collected metrics such as device
attributes and kernel launch information.
</p>
<p class="p">
The first column shows the metric name.
The second and third columns show the metric unit and value, respectively.
Both metric unit and value are automatically adjusted to the most fitting order of magnitude.
No unresolved regex:, group:, or breakdown: metrics are included.
</p>
</li>
</ul>
</div>
</div>
<div class="topic concept nested1" id="profile-import"><a name="profile-import" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#profile-import" name="profile-import" shape="rect">3.4. Profile Import</a></h3>
<div class="body conbody">
<p class="p">
Using the <samp class="ph codeph">--import</samp> option, saved reports can be imported into the command line profiler.
When using this flag, most other options are not available, except for certain result filterting options.
They are marked as such in the <a class="xref" href="index.html#command-line-options-profile" shape="rect">Profile options</a> table.
</p>
</div>
</div>
<div class="topic concept nested1" id="metrics-and-units"><a name="metrics-and-units" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#metrics-and-units" name="metrics-and-units" shape="rect">3.5. Metrics and Units</a></h3>
<div class="body conbody">
<p class="p">
When available and applicable, metrics are shown along with their unit.
This is to make it apparent if a metric represents cycles, threads, bytes/s, and so on.
</p>
<p class="p">
By default, units are scaled automatically so that metric values are shown with a reasonable order of magnitude.
Units are scaled using their SI-factors, i.e. byte-based units are scaled using a factor of 1000 and the prefixes K, M, G,
etc.
Time-based units are also scaled using a factor of 1000, with the prefixes n, u and m.
This scaling can be changed using a command line option, see <a class="xref" href="index.html#command-line-options-console-output" shape="rect">Console Output</a> options for details.
</p>
</div>
</div>
<div class="topic concept nested1" id="nvtx-filtering"><a name="nvtx-filtering" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvtx-filtering" name="nvtx-filtering" shape="rect">3.6. NVTX Filtering</a></h3>
<div class="body conbody">
<p class="p"><samp class="ph codeph">--nvtx-include <configuration> --nvtx-exclude
<configuration></samp><br clear="none"></br> These options are used to profile only those
kernels which satisfy the conditions mentioned in the configuration. Through these
options, you can choose which kernel falls into a specific range or collection of
ranges.
</p>
<p class="p"> You can use both options multiple times, mentioning all the
<samp class="ph codeph">--nvtx-include</samp> configurations followed by all
<samp class="ph codeph">--nvtx-exclude</samp> configurations. NVTX filtering requires
<samp class="ph codeph">--nvtx</samp> option.
</p>
<p class="p">
NVTX ranges are of two types: NvtxRangeStart/End and NvtxRangePush/Pop. The configuration syntax for both the types are briefly
described below.
<br clear="none"></br><br clear="none"></br><br clear="none"></br></p>
<ul class="ul">
<li class="li"><strong class="ph b">Start-End Ranges</strong><div class="tablenoborder"><a name="nvtx-filtering__nvtx-filtering-start-end-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="nvtx-filtering__nvtx-filtering-start-end-table" class="table" frame="border" border="1" rules="all">
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e631" rowspan="1" colspan="1">Quantifier</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e634" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e637" rowspan="1" colspan="1">Example</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e631" rowspan="1" colspan="1">,</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e634" rowspan="1" colspan="1">Delimiter between range
names
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e637" rowspan="1" colspan="1">
<p class="p">Range A,Range B</p>
<p class="p">Range B,Range A,Range C</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e631" rowspan="1" colspan="1">@</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e634" rowspan="1" colspan="1">Specify domain name. If not
mentioned, assuming <default domain>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e637" rowspan="1" colspan="1">
<p class="p">Domain A@Range A</p>
<p class="p">Domain B@Range B,Range Z</p>
</td>
</tr>
</tbody>
</table>
</div><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "Domain A@Range A" CuNvtx.exe
</pre><p class="p"> The kernels wrapped inside 'Range A' of 'Domain A' in the application are
profiled.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "Range A,Range B" CuNvtx.exe
</pre><p class="p"> The kernels wrapped inside both ranges, 'Range A' and 'Range B' of '<default
domain>' in the application are profiled.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "Range A" --nvtx-include "Range B" CuNvtx.exe
</pre><p class="p"> The kernels wrapped inside ranges, 'Range A' or 'Range B' of '<default
domain>' in the application are profiled.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-exclude "Range A" CuNvtx.exe
</pre><p class="p"> All the kernels in the application are profiled except those which are wrapped
inside 'Range A' of '<default domain>'.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "Range B"--nvtx-exclude "Range A" CuNvtx.exe
</pre><p class="p"> The kernels wrapped inside only 'Range B' and not 'Range A' of '<default
domain>' in the application are profiled.<br clear="none"></br></p>
</li>
<li class="li"><strong class="ph b">Push-Pop Ranges</strong><div class="tablenoborder"><a name="nvtx-filtering__nvtx-filtering-push-pop-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="nvtx-filtering__nvtx-filtering-push-pop-table" class="table" frame="border" border="1" rules="all">
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e737" rowspan="1" colspan="1">Quantifier</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e740" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e743" rowspan="1" colspan="1">Example</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e737" rowspan="1" colspan="1">/</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e740" rowspan="1" colspan="1">Delimiter between range names</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e743" rowspan="1" colspan="1">
<p class="p">Range A/Range B</p>
<p class="p">Range A/*/Range B</p>
<p class="p">Range A/</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e737" rowspan="1" colspan="1">[</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e740" rowspan="1" colspan="1">Range is at the bottom of the stack</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e743" rowspan="1" colspan="1">
<p class="p">[Range A</p>
<p class="p">[Range A/+/Range Z</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e737" rowspan="1" colspan="1">]</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e740" rowspan="1" colspan="1">Range is at the top of the stack</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e743" rowspan="1" colspan="1">
<p class="p">Range Z]</p>
<p class="p">Range C/*/Range Z]</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e737" rowspan="1" colspan="1">+</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e740" rowspan="1" colspan="1">Only one range between the two other ranges</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e743" rowspan="1" colspan="1">
<p class="p">Range B/+/Range D</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e737" rowspan="1" colspan="1">*</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e740" rowspan="1" colspan="1">Zero or more range(s) between the two other ranges</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e743" rowspan="1" colspan="1">
<p class="p">Range B/*/Range Z</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e737" rowspan="1" colspan="1">@</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e740" rowspan="1" colspan="1">Specify domain name. If not mentioned, assuming <default domain></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e743" rowspan="1" colspan="1">
<p class="p">Domain A@Range A</p>
<p class="p">Domain B@Range A/*/Range Z]</p>
</td>
</tr>
</tbody>
</table>
</div><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "Domain A@Range A/" CuNvtx.exe
</pre><p class="p">
The kernels wrapped inside 'Range A' of 'Domain A' in the application are profiled.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "[Range A" CuNvtx.exe
</pre><p class="p">
The kernels wrapped inside 'Range A' of '<default domain>' where 'Range A' is at the bottom of the stack in the application
are profiled.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "Range A/*/Range B" CuNvtx.exe
</pre><p class="p">
The kernels wrapped inside 'Range A' and 'Range B' of '<default domain>' with zero or many ranges between them in the application
are profiled.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-exclude "Range A/*/Range B" CuNvtx.exe
</pre><p class="p">
All the kernels in the application are profiled except those which are wrapped inside 'Range A' and 'Range B' of '<default
domain>' with zero or many ranges between them.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "Range A/" --nvtx-exclude "Range B]" CuNvtx.exe
</pre><p class="p">
The kernels wrapped inside only 'Range A' of '<default domain>' but not inside 'Range B' at the top of the stack in the application
are profiled.<br clear="none"></br></p>
</li>
<li class="li"><strong class="ph b">Regular Expression Support</strong><p class="p">
The configuration syntax for both the types NvtxRangeStart/End and NvtxRangePush/Pop is the same. Additionally, to use regular
expressions, follow the following syntax.
</p>
<ul class="ul">
<li class="li">
<p class="p">Provide prefix 'regex:' to treat nvtx config as regular expression.</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "<strong class="ph b">regex:</strong>Domain[A-Z]@Range[0-9]/" CuNvtx.exe
</pre><p class="p">
The kernels wrapped inside push/pop range with matching regex 'Range[0-9]' of domain with matching regex 'Domain[A-Z]' are
profiled.<br clear="none"></br></p>
</li>
<li class="li">
<p class="p">
Provide '/' as a prefix to "[" or "]" only for the range part of the config if "[" or "]" is at the start or at the end of
the range part, respectively. This is needed so that NCU can distinguish if "[" or "]" is part of the regex or represents
the top/bottom of the stack.
</p><pre class="pre screen" xml:space="preserve">
ncu --nvtx --nvtx-include "regex:[0-9]domainA@<strong class="ph b">/</strong>[0-9]rangeA,RangeC[0-9<strong class="ph b">/</strong>]" CuNvtx.exe
</pre><p class="p">
The kernels wrapped inside start/end ranges with matching regex '[0-9]rangeA' and 'RangeC[0-9]' of domain with matching regex
'[0-9]domainA' are profiled.<br clear="none"></br></p>
</li>
<li class="li">
<p class="p">
If any quantifier is part of the domain/range name, you need to use '\\' or '\' as a prefix. For the "$" quantifier, only
the '\\' prefix is valid.
</p>
</li>
</ul>
</li>
<li class="li"><strong class="ph b">Additional Information</strong><pre class="pre screen" xml:space="preserve">
--nvtx-include DomainA@RangeA,DomainB@RangeB //Not a valid config
</pre><p class="p"> In a single NVTX configuration, multiple ranges with regard to a single domain
can be specified. Mentioning ranges from different domains inside a single NVTX
config is not supported.
</p><pre class="pre screen" xml:space="preserve">
--nvtx-include "Range A\[i\]"
</pre><p class="p"> Quantifiers '@' ',' '[' ']' '/' '*' '+' can be used in range names using prefix
'\'. The kernels wrapped inside 'Range A[i]' of '<default domain>' in the
application are profiled.
</p><pre class="pre screen" xml:space="preserve">
--nvtx-include "Range A" //Start/End configuration
--nvtx-inlcude "Range A/" //Push/Pop configuration
--nvtx-inlcude "Range A]" //Push/Pop configuration
</pre><p class="p"> If the domain/range name contains '\', you need to provide '\\\\' in the config.</p>
<p class="p"> Do not use '\\\\' before any quantifier.</p>
<p class="p"> Including/Excluding only single range for Push/Pop configuration without
specifying stack frame position '[' or ']', use '/' quantifier at the end.
</p><pre class="pre screen" xml:space="preserve">
--nvtx-include "Range A/*/RangeB"
</pre><p class="p"> The order in which you mention Push/Pop configurations is important. In the
above example, 'Range A' should be below 'Range B' in the stack of ranges so
that the kernel is profiled.
</p>
<p class="p"> NVTX filtering honors cudaProfilerStart() and cudaProfilerStop(). There is no
support for ranges with no name.
</p>
</li>
</ul>
</div>
</div>
</div>
<div class="topic concept nested0" id="command-line-options"><a name="command-line-options" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#command-line-options" name="command-line-options" shape="rect">4. Command Line Options</a></h2>
<div class="body conbody">
<p class="p">
For long command line options, passing a unique initial substring can be sufficient.
</p>
</div>
<div class="topic concept nested1" id="command-line-options-general"><a name="command-line-options-general" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-general" name="command-line-options-general" shape="rect">4.1. General</a></h3>
<div class="body conbody">
<div class="tablenoborder"><a name="command-line-options-general__command-line-options-general-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-general__command-line-options-general-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 1. General Command Line Options</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1012" rowspan="1" colspan="1">Option</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1015" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1018" rowspan="1" colspan="1">Default</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1012" rowspan="1" colspan="1">h,help</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1015" rowspan="1" colspan="1">Show help message</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1018" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1012" rowspan="1" colspan="1">v,version</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1015" rowspan="1" colspan="1">Show version information</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1018" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1012" rowspan="1" colspan="1">mode</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1015" rowspan="1" colspan="1">
Select the mode of interaction with the target application
<ul class="ul">
<li class="li"><strong class="ph b">launch-and-attach:</strong> Launch the target application and immediately attach for profiling.
</li>
<li class="li"><strong class="ph b">launch:</strong> Launch the target application and suspend in the first intercepted API call, wait for tool to attach.
</li>
<li class="li"><strong class="ph b">attach:</strong> Attach to a previously launched application to which no other tool is attached.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1018" rowspan="1" colspan="1">launch-and-attach</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1012" rowspan="1" colspan="1">p,port</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1015" rowspan="1" colspan="1">Base port used for connecting to target applications for <samp class="ph codeph">--mode launch/attach</samp></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1018" rowspan="1" colspan="1">49152</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1012" rowspan="1" colspan="1">max-connections</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1015" rowspan="1" colspan="1">Maximum number of ports for connecting to target applications</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1018" rowspan="1" colspan="1">64</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-launch"><a name="command-line-options-launch" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-launch" name="command-line-options-launch" shape="rect">4.2. Launch</a></h3>
<div class="body conbody">
<div class="tablenoborder"><a name="command-line-options-launch__command-line-options-launch-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-launch__command-line-options-launch-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 2. Launch Command Line Options</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1135" rowspan="1" colspan="1">Option</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1138" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1141" rowspan="1" colspan="1">Default</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">check-exit-code</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">
Check the application exit code and print an error if it is different than 0.
If set, <samp class="ph codeph">--replay-mode application</samp> will stop after the first pass if the exit code is not 0.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1">yes</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">injection-path-64</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">
Override the default path for the injection libraries.
The injection libraries are used by the tools to intercept relevant APIs (like CUDA or NVTX).
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">call-stack</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">Enable CPU Call Stack collection.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1">false</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">nvtx</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">Enable NVTX support for tools.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1">false</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">target-processes</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">
Select the processes you want to profile. Available modes are:
<ul class="ul">
<li class="li"><strong class="ph b">application-only</strong>
Profile only the root application process.
</li>
<li class="li"><strong class="ph b">all</strong>
Profile the application and all its child processes.
</li>
</ul>
This option is only available for Linux and Windows targets.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1">
application-only
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">target-processes-filter</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">
Set the comma separated expressions to filter which processes are profiled.
<ul class="ul">
<li class="li"><samp class="ph codeph"><process name></samp> Set the process name to filter by. Only exactly matched processes are profiled.
</li>
<li class="li"><samp class="ph codeph">regex:<expression></samp> Set the regex to filter matching process name profiling.
On shells that recognize regular expression symbols as special characters (e.g. Linux bash),
the expression needs to be escaped with quotes, e.g. <samp class="ph codeph">--target-processes-filter regex:".*Process"</samp>.
</li>
</ul>
The executable name will be considered as process name to match.
If the process name or the provided expression match, the process will be profiled.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1">
<p class="p"><strong class="ph b">Examples</strong></p>
<p class="p"><samp class="ph codeph">--target-processes-filter MatrixMul</samp> Filter all processes having executable name exactly as "MatrixMul".
</p>
<p class="p"><samp class="ph codeph">--target-processes-filter regex:Matrix</samp>Filter all processes that include the string "Matrix" in their executable name, e.g. "MatrixMul" and "MatrixAdd".
</p>
<p class="p"><samp class="ph codeph">--target-processes-filter MatrixMul,MatrixAdd</samp>Filter all processes having executable name exactly as "MatrixMul" or "MatrixAdd".
</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">support-32bit</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">
Support profiling processes launched from 32-bit applications.
This option is only available on x86_64 Linux.
On Windows, tracking 32-bit applications is enabled by default.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1">no</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1135" rowspan="1" colspan="1">null-stdin</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1138" rowspan="1" colspan="1">
Launch the application with '/dev/null' as its standard input.
This avoids applications reading from standard input being stopped by
<samp class="ph codeph">SIGTTIN</samp> signals and hanging when running as backgrounded processes.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1141" rowspan="1" colspan="1">false</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-attach"><a name="command-line-options-attach" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-attach" name="command-line-options-attach" shape="rect">4.3. Attach</a></h3>
<div class="body conbody">
<div class="tablenoborder"><a name="command-line-options-attach__command-line-options-attach-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-attach__command-line-options-attach-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 3. Attach Command Line Options</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1331" rowspan="1" colspan="1">Option</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1334" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1337" rowspan="1" colspan="1">Default</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1331" rowspan="1" colspan="1">hostname</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1334" rowspan="1" colspan="1">
Set the hostname or IP address for connecting to the machine on which the target application is running.
When attaching to a local target application, use 127.0.0.1.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1337" rowspan="1" colspan="1">127.0.0.1</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-profile"><a name="command-line-options-profile" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-profile" name="command-line-options-profile" shape="rect">4.4. Profile</a></h3>
<div class="body conbody">
<div class="tablenoborder"><a name="command-line-options-profile__command-line-options-profile-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-profile__command-line-options-profile-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 4. Profile Command Line Options</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1386" rowspan="1" colspan="1">Option</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1389" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e1392" rowspan="1" colspan="1">Default/Examples</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">devices</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
List the GPU devices to enable profiling on, separated by comma.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
<p class="p">All devices</p>
<p class="p"><strong class="ph b">Examples</strong></p>
<p class="p"><samp class="ph codeph">--devices 0,2</samp></p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">filter-mode</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Set the filtering mode for kernel launches. Available modes:
<ul class="ul">
<li class="li"><strong class="ph b">global:</strong> Apply provided launch filters on kernel launches collectively.
</li>
<li class="li"><strong class="ph b">per-gpu:</strong> Apply provided launch filters on kernel launches separately on each device. Effective launch filters for this mode are <samp class="ph codeph">--launch-count</samp> and <samp class="ph codeph">--launch-skip</samp></li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">global</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">kernel-id</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
<p class="p">
Set the identifier to use for matching kernels.
If the kernel does not match the identifier, it will be ignored for profiling.
</p>
<p class="p">
The identifier must be of the following format:
<dfn class="term">context-id:stream-id:[name-operator:]kernel-name:invocation-nr</dfn></p>
<div class="p">
<ul class="ul">
<li class="li"><strong class="ph b">context-id</strong> is the CUDA context ID or regular expression to match the NVTX name.
</li>
<li class="li"><strong class="ph b">stream-id</strong> is the CUDA stream ID or regular expression to match the NVTX name.
</li>
<li class="li"><strong class="ph b">name-operator</strong> is an optional operator to <dfn class="term">kernel-name</dfn>. Currently, only <dfn class="term">regex</dfn> is the only supported operator.
</li>
<li class="li"><strong class="ph b">kernel-name</strong> is the expression to match the kernel name.
By default, this is a full, literal match to what is specified by <samp class="ph codeph">--kernel-name-base</samp>.
When specifying the optional <dfn class="term">regex</dfn> name operator, this is a partial regular expression match to what is specified by <samp class="ph codeph">--kernel-name-base</samp>.
</li>
<li class="li"><strong class="ph b">invocation-nr</strong> is the N’th invocation of this kernel function. Multiple invocations can also be specified using regular expressions.
</li>
</ul>
If the context/stream ID is a positive number, it will be strictly matched against the CUDA context/stream ID. Otherwise it
will be treated as a regular expression and matched against the context/stream name specified using the NVTX library.
</div><a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
<p class="p"><strong class="ph b">Examples</strong></p>
<p class="p"><samp class="ph codeph">--kernel-id ::foo:2</samp> For kernel "foo", match the second invocation.
</p>
<p class="p"><samp class="ph codeph">--kernel-id :::".*5|3"</samp> For all kernels, match the third invocation, and all for which the invocation number ends in "5".
</p>
<p class="p"><samp class="ph codeph">--kernel-id ::regex:^.*foo$:</samp> Match all kernels ending in "foo".
</p>
<p class="p"><samp class="ph codeph">--kernel-id ::regex:^(?!foo):</samp> Match all kernels except those starting with "foo".
Note that depending on your OS and shell, you might need to quote the expression,
e.g. using single quotes in Linux <dfn class="term">bash</dfn>: <samp class="ph codeph">--kernel-id ::regex:'^(?!foo)':</samp></p>
<p class="p"><samp class="ph codeph">--kernel-id 1:2::7</samp> Match all seventh kernel invocations on context 1, stream 2.
</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">k,kernel-name</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Set the expression to use when matching kernel names.
<ul class="ul">
<li class="li"><samp class="ph codeph"><kernel name></samp> Set the kernel name for an exact match.
</li>
<li class="li"><samp class="ph codeph">regex:<expression></samp> Set the regex to use for matching the kernel name.
On shells that recognize regular expression symbols as special characters (e.g. Linux bash),
the expression needs to be escaped with quotes, e.g. <samp class="ph codeph">--kernel-name regex:".*Foo"</samp>.
</li>
</ul>
If the kernel name or the provided expression do not match, it will be ignored for profiling.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
<p class="p"><strong class="ph b">Examples</strong></p>
<p class="p"><samp class="ph codeph">-k foo</samp> Match all kernels named exactly "foo".
</p>
<p class="p"><samp class="ph codeph">-k regex:foo</samp> Match all kernels that include the string "foo", e.g. "foo" and "fooBar".
</p>
<p class="p"><samp class="ph codeph">-k regex:"foo|bar"</samp> Match all kernels including the strings "foo" or "bar", e.g. "foo", "foobar", "_bar2".
</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">kernel-name-base</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Set the basis for <samp class="ph codeph">--kernel-name</samp>, and <samp class="ph codeph">--kernel-id</samp> kernel-name. Options are:
<ul class="ul">
<li class="li"><strong class="ph b">function:</strong> Function name without parameters, templates etc.
e.g. <samp class="ph codeph">dmatrixmul</samp></li>
<li class="li"><strong class="ph b">demangled:</strong> Demangled function name, including parameters, templates, etc.
e.g. <samp class="ph codeph">dmatrixmul(float*,int,int)</samp></li>
<li class="li"><strong class="ph b">mangled:</strong> Mangled function name.
e.g. <samp class="ph codeph">_Z10dmatrixmulPfiiS_iiS_</samp></li>
</ul><a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">function</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">c,launch-count</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">Limit the number of profiled kernel launches.
The count is only incremented for launches that match the kernel filters.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">s,launch-skip</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Set the number of kernel launches to skip before starting to profile kernels.
The number takes into account only launches that match the kernel filters.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">0</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">launch-skip-before-match</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Set the number of kernel launches to skip before starting to profile.
The count is incremented for all launches, regardless of the kernel filters.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">0</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">kill</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Terminate the target application when the requested --launch-count was profiled.
Allowed values:
<ul class="ul">
<li class="li">
on/off
</li>
<li class="li">
yes/no
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">no</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">replay-mode</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Mechanism used for replaying a kernel launch multiple times to collect all requested profiling data:
<ul class="ul">
<li class="li"><strong class="ph b">kernel:</strong> Replay individual kernel launches "transparently" during the execution of the application.
See <a class="xref" href="../ProfilingGuide/index.html#kernel-replay" shape="rect">Kernel Replay</a> for more details.
</li>
<li class="li"><strong class="ph b">application:</strong> Relaunch the entire application multiple times. Requires deterministic program execution.
See <a class="xref" href="../ProfilingGuide/index.html#application-replay" shape="rect">Application Replay</a> for more details.
</li>
<li class="li"><strong class="ph b">range:</strong> Replay ranges of CUDA API calls and kernel launches "transparently" during the execution of the application.
Ranges must be defined using <samp class="ph codeph">cu(da)ProfilerStart/Stop</samp> API pairs or <a class="xref" href="index.html#nvtx-filtering" shape="rect">NVTX expressions</a>.
See <a class="xref" href="../ProfilingGuide/index.html#range-replay" shape="rect">Range Replay</a> for more details.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">kernel</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">app-replay-buffer</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Application replay buffer location.
<ul class="ul">
<li class="li"><strong class="ph b">file:</strong>
Replay pass data is buffered in a temporary file. The report is created after profiling completed.
This mode is more scalable, as the amount of required memory does not scale with the number of profiled kernels.
</li>
<li class="li"><strong class="ph b">memory:</strong>
Replay pass data is buffered in memory, and the report is created while profiling.
This mode can result in better performance if the filesystem is slow,
but the amount of required memory scales with the number of profiled kernels.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">file</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">app-replay-match</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Application replay kernel matching strategy.
For all options, kernels are matched on a per-process and per-device (GPU) basis.
Below options are used to configure the applied strategy in more detail.
<ul class="ul">
<li class="li"><strong class="ph b">name:</strong> Kernels are matched in the following order:
1. (mangled) name,
2. order of execution
</li>
<li class="li"><strong class="ph b">grid:</strong> Kernels are matched in the following order:
1. (mangled) name,
2. CUDA grid/block size,
3. order of execution
</li>
<li class="li"><strong class="ph b">all:</strong> Kernels are matched in the following order:
1. (mangled) name,
2. CUDA grid/block size,
3. CUDA context ID,
4. CUDA stream ID,
5. order of execution
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">grid</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">app-replay-mode</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Application replay kernel matching mode:
<ul class="ul">
<li class="li"><strong class="ph b">strict:</strong> Requires all kernels to match across all replay passes.
</li>
<li class="li"><strong class="ph b">relaxed:</strong> Produces results only for kernels that could be matched across replay passes.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">strict</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">range-replay-options</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Range replay options, separated by comma.
Below options are supported:
<ul class="ul">
<li class="li"><strong class="ph b">enable-greedy-sync</strong><p class="p">
Insert ctx sync for applicable deferred APIs during capture.
</p>
</li>
<li class="li"><strong class="ph b">disable-host-restore</strong><p class="p">
Disable restoring device-written host allocations.
</p>
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">none</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">list-sets</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
List all section sets found in the searched section folders and exit.
For each set, the associated sections are shown, as well as the estimated number of metrics collected as part of this set.
This number can be used as an estimate of the relative profiling overhead per kernel launch of this set.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">set</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Identifier of section set to collect.
If not specified, the default set is collected.
The full set of sections can be collected with <samp class="ph codeph">--set full</samp>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
If no <samp class="ph codeph">--set</samp> option is given, the default set if collected.
If not specified and <samp class="ph codeph">--section</samp> or <samp class="ph codeph">--metrics</samp> are used, no sets are collected.
Use <samp class="ph codeph">--list-sets</samp> to see which set is the default.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">list-sections</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">List all sections found in the searched section folders and exit.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">section</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Add a section identifier to collect in one of the following ways:
<ul class="ul">
<li class="li"><samp class="ph codeph"><section identifier></samp> Set the section identifier for an exact match.
</li>
<li class="li"><samp class="ph codeph">regex:<expression></samp> Regular expression allows matching full section identifier.
For example, <samp class="ph codeph">.*Stats</samp>, matches all sections ending with 'Stats'.
On shells that recognize regular expression symbols as special characters (e.g. Linux bash),
the expression needs to be escaped with quotes, e.g. <samp class="ph codeph">--section "regex:.*Stats"</samp>.
</li>
</ul>
This option is ignored when used with <samp class="ph codeph">--import</samp> and <samp class="ph codeph">--page raw</samp> or <samp class="ph codeph">--page source</samp>.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
If no <samp class="ph codeph">--section</samp> options are given, the sections associated with the default set are collected.
If no sets are found, all sections are collected.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">section-folder</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Add a non-recursive search path for .section files.
Section files in this folder will be made available to the <samp class="ph codeph">--section</samp> option.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
If no <samp class="ph codeph">--section-folder</samp> options are given, the <samp class="ph codeph">sections</samp> folder is added by default.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">section-folder-recursive</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1"> Add a recursive search path for .section
files. Section files in this folder and all folders below will be made
available to the <samp class="ph codeph">--section</samp> option.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
If no <samp class="ph codeph">--section-folder</samp> options are given, the <samp class="ph codeph">sections</samp> folder is added by default.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">list-rules</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">List all rules found in the searched section folders and exit.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">apply-rules</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">Apply active and applicable rules to each profiling result.
Use <samp class="ph codeph">--rule</samp> to limit which rules to apply.
Allowed values:
<ul class="ul">
<li class="li">
on/off
</li>
<li class="li">
yes/no
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">yes</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">rule</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">Add a rule identifier to apply. Implies <samp class="ph codeph">--apply-rules yes</samp>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> If no <samp class="ph codeph">--rule</samp> options are
given, all applicable rules in the <samp class="ph codeph">sections</samp> folder are
applied.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">import-source</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">If available from -lineinfo, correlated CUDA source files are permanently imported into the report.
Allowed values:
<ul class="ul">
<li class="li">
on/off
</li>
<li class="li">
yes/no
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">no</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">list-metrics</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
List all metrics collected from active sections.
If the list of active sections is restricted using the <samp class="ph codeph">--section</samp> option,
only metrics from those sections will be listed.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">query-metrics</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Query available metrics for the devices on system.
Use <samp class="ph codeph">--devices</samp> and <samp class="ph codeph">--chips</samp> to filter which devices to query.
Note that by default, listed metric names need to be appended a valid suffix in order for them to become valid metrics.
See <samp class="ph codeph">--query-metrics-mode</samp> for how to get the list of valid suffixes, or check the
<a class="xref" href="../ProfilingGuide/index.html#metrics-structure" shape="rect">Kernel Profiling Guide</a>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">query-metrics-mode</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Set the mode for querying metrics. Implies <samp class="ph codeph">--query-metrics</samp>.
Available modes:
<ul class="ul">
<li class="li"><strong class="ph b">base:</strong> Only the base names of the metrics.
</li>
<li class="li"><strong class="ph b">suffix:</strong> Suffix names for the base metrics. This gives the list of all metrics derived from the base metrics. Use <samp class="ph codeph">--metrics</samp> to specify the base metrics to query.
</li>
<li class="li"><strong class="ph b">all:</strong> Full names for all metrics. This gives the list of all base metrics and their suffix metrics.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">base</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">metrics</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
<p class="p">
Specify all metrics to be profiled, separated
by comma. If no <samp class="ph codeph">--section</samp> options are given, only the
temporary section containing all metrics listed using this option is
collected. If <samp class="ph codeph">--section</samp> options are given in addition
to <samp class="ph codeph">--metrics</samp>, all metrics from those sections and from
<samp class="ph codeph">--metrics</samp> are collected.
</p>
<p class="p">
Names passed to this option support the following prefixes:
<samp class="ph codeph">regex:<expression></samp> expands to all metrics that partially match the expression. Enclose the regular expression in ^...$ to force a full match.
<samp class="ph codeph">group:<name></samp> lists all metrics of the metric group with that name. See section files for valid group names.
<samp class="ph codeph">breakdown:<metric></samp> expands to the input metrics of the high-level throughput metric. If the specified metric does not support a breakdown, no
metrics are added.
</p>
<p class="p">
If a metric requires a suffix to be valid, and neither <samp class="ph codeph">regex:</samp> nor <samp class="ph codeph">group:</samp>
are used, this option automatically expands the name to all available first-level sub-metrics.
</p>
<p class="p">
When importing a report <samp class="ph codeph">:group</samp> and <samp class="ph codeph">:breakdown</samp> are not supported.
</p><a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"></td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">list-chips</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">List all supported chips that can be used with <samp class="ph codeph">--chips</samp>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">chips</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">Specify the chips for querying metrics, separated by comma.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">
<p class="p"><strong class="ph b">Examples</strong></p>
<p class="p"><samp class="ph codeph">--chips gv100,tu102</samp></p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">profile-from-start</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Set if application should be profiled from its start. Allowed values:
<ul class="ul">
<li class="li">
on/off
</li>
<li class="li">
yes/no
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">yes</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">disable-profiler-start-stop</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Disable profiler start/stop. When enabled, <samp class="ph codeph">cu(da)ProfilerStart/Stop</samp> API calls are ignored.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">quiet</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Suppress all profiling output.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">cache-control</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Control the behavior of the GPU caches during profiling. Allowed values:
<ul class="ul">
<li class="li"><strong class="ph b">all:</strong> All GPU caches are flushed before each kernel replay iteration during profiling.
While metric values in the execution environment of the application might be slightly different without invalidating the caches,
this mode offers the most reproducible metric results across the replay passes and also across multiple runs of the target
application.
</li>
<li class="li"><strong class="ph b">none:</strong> No GPU caches are flushed during profiling. This can improve performance and better replicates the
application behavior if only a single kernel replay pass is necessary for metric collection.
However, some metric results will vary depending on prior GPU work, and between replay iterations.
This can lead to inconsistent and out-of-bounds metric values.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">all</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">clock-control</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Control the behavior of the GPU clocks during profiling. Allowed values:
<ul class="ul">
<li class="li"><strong class="ph b">base:</strong>
GPC and memory clocks are locked to their respective base frequency during profiling.
This has no impact on thermal throttling.
Note that actual clocks might still vary, depending on the level of driver support for this feature.
As an alternative, use <samp class="ph codeph">nvidia-smi</samp> to lock the clocks externally and set this
option to <samp class="ph codeph">none</samp>.
</li>
<li class="li"><strong class="ph b">none:</strong>
No GPC or memory frequencies are changed during profiling.
</li>
<li class="li"><strong class="ph b">reset:</strong>
Reset GPC and memory clocks for all or the selected devices and exit.
Use if a previous, killed execution of <span class="keyword">ncu</span>
left the GPU clocks in a locked state.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1">base</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">nvtx-include</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Adds an include statement to the <a class="xref" href="index.html#nvtx-filtering" shape="rect">NVTX filter</a>, which allows selecting kernels to profile based on NVTX ranges.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1386" rowspan="1" colspan="1">nvtx-exclude</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1389" rowspan="1" colspan="1">
Adds an exclude statement to the <a class="xref" href="index.html#nvtx-filtering" shape="rect">NVTX filter</a>, which allows selecting kernels to profile based on NVTX ranges.
<a class="xref" href="index.html#command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a></td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e1392" rowspan="1" colspan="1"> </td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-sampling"><a name="command-line-options-sampling" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-sampling" name="command-line-options-sampling" shape="rect">4.5. Sampling</a></h3>
<div class="body conbody">
<div class="tablenoborder"><a name="command-line-options-sampling__command-line-options-sampling-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-sampling__command-line-options-sampling-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 5. Sampling Command Line Options</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2477" rowspan="1" colspan="1">Option</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2480" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2483" rowspan="1" colspan="1">Default</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2477" rowspan="1" colspan="1">sampling-interval</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2480" rowspan="1" colspan="1">
Set the sampling period in the range of [0..31].
The actual frequency is 2 ^ (5 + value) cycles.
If set to 'auto', the profiler tries to automatically determine a high sampling frequency without skipping samples or overflowing
the output buffer.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2483" rowspan="1" colspan="1">auto</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2477" rowspan="1" colspan="1">sampling-max-passes</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2480" rowspan="1" colspan="1">
Set maximum number of passes used for sampling (see the <a class="xref" href="../ProfilingGuide/index.html#overhead" shape="rect">Kernel Profiling Guide</a> for more details on profiling overhead).
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2483" rowspan="1" colspan="1">5</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2477" rowspan="1" colspan="1">sampling-buffer-size</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2480" rowspan="1" colspan="1">Set the size of the device-sided allocation for samples in bytes.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2483" rowspan="1" colspan="1">32*1024*1024</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-file"><a name="command-line-options-file" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-file" name="command-line-options-file" shape="rect">4.6. File</a></h3>
<div class="body conbody">
<div class="tablenoborder"><a name="command-line-options-file__command-line-options-file-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-file__command-line-options-file-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 6. File Command Line Options</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2558" rowspan="1" colspan="1">Option</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2561" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2564" rowspan="1" colspan="1">Default</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2558" rowspan="1" colspan="1">log-file</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2561" rowspan="1" colspan="1">Send all tool output to the specified file, or
one of the standard channels. The file will be overwritten. If the file doesn't exist, a new one will be created."stdout"
as the whole file name indicates standard output channel (stdout). "stderr" as the whole file name indicates standard error
channel (stderr)."
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2564" rowspan="1" colspan="1">
If <samp class="ph codeph">--log-file</samp> is not set , profile results will be printed on the console.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2558" rowspan="1" colspan="1">o,export</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2561" rowspan="1" colspan="1">Set the output file for writing the profile report.
If not set, a temporary file will be used which is removed afterwards. The specified
name supports macro expansion. See <a class="xref" href="index.html#command-line-options-file-macros" shape="rect">File Macros</a> for more details.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2564" rowspan="1" colspan="1">
If <samp class="ph codeph">--export</samp> is set and no <samp class="ph codeph">--page</samp> option is given,
no profile results will be printed on the console.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2558" rowspan="1" colspan="1">f,force-overwrite</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2561" rowspan="1" colspan="1">Force overwriting all output files.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2564" rowspan="1" colspan="1">By default, the profiler won't overwrite existing output files and show an error instead.</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2558" rowspan="1" colspan="1">i,import</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2561" rowspan="1" colspan="1">Set the input file for reading the profile results.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2564" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2558" rowspan="1" colspan="1">open-in-ui</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2561" rowspan="1" colspan="1">Open report in UI instead of showing result on terminal. (Only available on host platforms)</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2564" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2558" rowspan="1" colspan="1">section-folder-restore</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2561" rowspan="1" colspan="1">Restores stock files to the default section folder or the folder specified by an accompanying <dfn class="term">--section-folder</dfn> option. If the operation will overwrite modified files then the <dfn class="term">--force-overwrite</dfn> option is required.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2564" rowspan="1" colspan="1"> </td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-console-output"><a name="command-line-options-console-output" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-console-output" name="command-line-options-console-output" shape="rect">4.7. Console Output</a></h3>
<div class="body conbody">
<div class="tablenoborder"><a name="command-line-options-console-output__command-line-options-console-output-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-console-output__command-line-options-console-output-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 7. Console Output Command Line Options</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2688" rowspan="1" colspan="1">Option</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2691" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e2694" rowspan="1" colspan="1">Default</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">csv</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">Use comma-separated values as console output. Implies --print-units base by default.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">page</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Select the report page to print console output for. Available pages are:
<ul class="ul">
<li class="li"><strong class="ph b">details</strong>
Show results grouped as sections, include rule results.
Some metrics that are collected by default (e.g. device attributes) are omitted
if not specified explicitly in any section or using <samp class="ph codeph">--metrics</samp>.
</li>
<li class="li"><strong class="ph b">raw</strong>
Show all collected metrics by kernel launch.
</li>
<li class="li"><strong class="ph b">source</strong>
Show source. See <samp class="ph codeph">--print-source</samp> to select the source view.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1"><strong class="ph b">details</strong>.
If no <samp class="ph codeph">--page</samp> option is given and <samp class="ph codeph">--export</samp> is set,
no results are printed to the console output.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-source</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Select the source view:
<ul class="ul">
<li class="li"><strong class="ph b">sass</strong>
Show SASS (assembly) instructions for each kernel launch.
</li>
<li class="li"><strong class="ph b">ptx</strong>
Show PTX source of every cubin whose at least one kernel is profiled.
</li>
<li class="li"><strong class="ph b">cuda</strong>
Show entire CUDA-C source file which has kernel code as per kernel launch.
CLI shows CUDA source only if file exists on the host machine.
</li>
<li class="li"><strong class="ph b">cuda,sass</strong>
Show SASS CUDA-C source correlation for each kernel launch.
CLI shows CUDA source only if file exists on the host machine.
</li>
</ul>
Metric correlation with source is available in <strong class="ph b">sass</strong>, and <strong class="ph b">cuda,sass</strong> source view.
Metrics specified with <samp class="ph codeph">--metrics</samp> and specified section file with <samp class="ph codeph">--section</samp> are correlated.
Consider restricting the number of selected metrics such that values fit into a single output row.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">sass</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">resolve-source-file</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Resolve CUDA source file in the <samp class="ph codeph">--page source</samp> output.
Provide comma separated files full path.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">details-all</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">Include all section metrics on details page.</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1"> </td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-units</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Select the mode for scaling of metric units. Available modes are:
<ul class="ul">
<li class="li"><strong class="ph b">auto</strong>
Show all metrics automatically scaled to the most fitting order of magnitude.
</li>
<li class="li"><strong class="ph b">base</strong>
Show all metrics in their base unit.
</li>
</ul>
Replaces deprecated option <samp class="ph codeph">--units</samp>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">auto</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-fp</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">Show all numeric metrics in the console output as floating point numbers. Replaces deprecated option <samp class="ph codeph">--fp</samp>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">false</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-kernel-base</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">Set the basis for kernel name output. See <samp class="ph codeph">--kernel-regex-base</samp> for options. Replaces deprecated option <samp class="ph codeph">--kernel-base</samp>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">demangled</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-metric-instances</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Set output mode for metrics with instance values:
<ul class="ul">
<li class="li"><strong class="ph b">none</strong>
Only show GPU aggregate value.
</li>
<li class="li"><strong class="ph b">values</strong>
Show GPU aggregate followed by all instance values.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">none</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-nvtx-rename</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Select how NVTX should be used for renaming:
<ul class="ul">
<li class="li"><strong class="ph b">none</strong>
Don't use NVTX for renaming.
</li>
<li class="li"><strong class="ph b">kernel</strong>
Rename kernels with the most recent enclosing NVTX push/pop range.
</li>
</ul>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">none</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-rule-details</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Print additional details for rule results, such as the triggering metrics.
Currently has no effect in CSV mode.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">false</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2688" rowspan="1" colspan="1">print-summary</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2691" rowspan="1" colspan="1">
Select the summary output mode. Available modes are:
<ul class="ul">
<li class="li"><strong class="ph b">none</strong>
No summary.
</li>
<li class="li"><strong class="ph b">per-gpu</strong>
Summary for each GPU.
</li>
<li class="li"><strong class="ph b">per-kernel</strong>
Summary for each kernel type.
</li>
<li class="li"><strong class="ph b">per-nvtx</strong>
Summary for each NVTX context.
</li>
</ul>
Replaces deprecated option <samp class="ph codeph">--summary</samp>.
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e2694" rowspan="1" colspan="1">none</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-response-file"><a name="command-line-options-response-file" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-response-file" name="command-line-options-response-file" shape="rect">4.8. Response File</a></h3>
<div class="body conbody">
<p class="p">
Response files can be specified by adding <samp class="ph codeph">@FileName</samp> to the command line.
The file name must immediately follow the <samp class="ph codeph">@</samp> character.
The content of each response file is inserted in place of the corresponding response file option.
</p>
</div>
</div>
<div class="topic concept nested1" id="command-line-options-file-macros"><a name="command-line-options-file-macros" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#command-line-options-file-macros" name="command-line-options-file-macros" shape="rect">4.9. File Macros</a></h3>
<div class="body conbody">
<p class="p">
The file name specified with option <samp class="ph codeph">-o</samp> or <samp class="ph codeph">--export</samp>
supports the following macro expansions. Occurrences of these macros in the report
file name are replaced by the corresponding character sequence. If not specified
otherwise, the macros cannot be used as part of the file path.
</p>
<div class="tablenoborder"><a name="command-line-options-file-macros__command-line-options-file-macro-expansions" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="command-line-options-file-macros__command-line-options-file-macro-expansions" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 8. Macro Expansions</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="20%" id="d54e3063" rowspan="1" colspan="1">Macro</th>
<th class="entry" valign="top" width="80%" id="d54e3066" rowspan="1" colspan="1">Description</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="20%" headers="d54e3063" rowspan="1" colspan="1">%h</td>
<td class="entry" valign="top" width="80%" headers="d54e3066" rowspan="1" colspan="1">
Expands to the host name of the machine on which the command line profiler is running.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="20%" headers="d54e3063" rowspan="1" colspan="1">%q{ENV_NAME}</td>
<td class="entry" valign="top" width="80%" headers="d54e3066" rowspan="1" colspan="1">
Expands to the content of the variable with the given name <samp class="ph codeph">ENV_NAME</samp> from the environment of the command line profiler.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="20%" headers="d54e3063" rowspan="1" colspan="1">%p</td>
<td class="entry" valign="top" width="80%" headers="d54e3066" rowspan="1" colspan="1">
Expands to the process ID of the command line profiler.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="20%" headers="d54e3063" rowspan="1" colspan="1">%i</td>
<td class="entry" valign="top" width="80%" headers="d54e3066" rowspan="1" colspan="1">
Expands to the lowest unused positive integer number that guarantees the resulting file name is not yet used.
This macro can only be used once in the output file name.
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="20%" headers="d54e3063" rowspan="1" colspan="1">%%</td>
<td class="entry" valign="top" width="80%" headers="d54e3066" rowspan="1" colspan="1">
Expands to a single <samp class="ph codeph">%</samp> character in the output file name.
This macro can be used in the file path and the file name.
</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
</div>
<div class="topic concept nested0" id="environment-variables"><a name="environment-variables" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#environment-variables" name="environment-variables" shape="rect">5. Environment Variables</a></h2>
<div class="body conbody">
<p class="p">
The following environment variables can be set before launching <span class="keyword">NVIDIA Nsight Compute CLI</span>, or the UI, respectively.
</p>
<div class="tablenoborder"><a name="environment-variables__environment-variables-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="environment-variables__environment-variables-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 9. Environment Variables</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="33.33333333333333%" id="d54e3160" rowspan="1" colspan="1">Name</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e3163" rowspan="1" colspan="1">Description</th>
<th class="entry" valign="top" width="33.33333333333333%" id="d54e3166" rowspan="1" colspan="1">Default/Values</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3160" rowspan="1" colspan="1">
NV_COMPUTE_PROFILER_DISABLE_STOCK_FILE_DEPLOYMENT
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3163" rowspan="1" colspan="1">
<p class="p">
Disable file deployment to the versioned <samp class="ph codeph">Sections</samp> directory, using section and rule files from the stock directory
within the <span class="keyword"></span> installation instead.
</p>
<p class="p">
By default, the versioned directory from the user's documents folder is used to ensure that
any potential user updates are taken into account.
</p>
<p class="p">
Only supported in the <span class="keyword">NVIDIA Nsight Compute CLI</span>.
</p>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3166" rowspan="1" colspan="1">
<p class="p">
Default: unset
</p>
<p class="p">
Set to "1" to disable deployment.
</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3160" rowspan="1" colspan="1">
NV_COMPUTE_PROFILER_LOCAL_CONNECTION_OVERRIDE
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3163" rowspan="1" colspan="1">
<p class="p">
Override the default local connection mechanism between frontend and profiled target processes.
The default mechanism is platform-dependent.
This should only be used if there are connection problems between frontend and target processes in a local launch.
</p>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3166" rowspan="1" colspan="1">
<p class="p">
Default: unset (use default mechanism)
</p>
<p class="p">
Set to "uds" to use Unix Domain Socket connections (available on Posix platforms, only).
Set to "tcp" to use TCP (available on all platforms).
Set to "named-pipes" to use Windows Named Pipes (available on Windows, only).
</p>
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3160" rowspan="1" colspan="1">
NV_COMPUTE_PROFILER_DISABLE_SW_PRE_PASS
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3163" rowspan="1" colspan="1">
<p class="p">
Disable the instruction-level software (SW) metric pre-pass.
When collectinh SW-patched metrics, such as <samp class="ph codeph">inst_executed</samp>,
the pre-pass is used to determine which functions are executed as part of the kernel and should be patched.
This requires a separat e replay pass and if only instruction-level SW metrics are to be collected, prevents single-pass data
collection.
Disabling the pre-pass can improve performance if memory save-and-restore is undesirable and application replay is not possible.
</p>
</td>
<td class="entry" valign="top" width="33.33333333333333%" headers="d54e3166" rowspan="1" colspan="1">
<p class="p">
Default: unset (use pre-pass when applicable)
</p>
<p class="p">
Set to "1" to disable pre-pass.
</p>
</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested0" id="nvprof-guide"><a name="nvprof-guide" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#nvprof-guide" name="nvprof-guide" shape="rect">6. Nvprof Transition Guide</a></h2>
<div class="body conbody">
<p class="p">
This guide provides tips for moving from nvprof to <span class="keyword">NVIDIA Nsight Compute CLI</span>.
<span class="keyword">NVIDIA Nsight Compute CLI</span> tries to provide as much feature and usage parity as possible with nvprof,
but some features are now covered by different tools and some command line options have changed their name or meaning.
</p>
</div>
<div class="topic concept nested1" id="nvprof-trace"><a name="nvprof-trace" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvprof-trace" name="nvprof-trace" shape="rect">6.1. Trace</a></h3>
<div class="body conbody">
<ul class="ul">
<li class="li"><strong class="ph b">GPU and API trace</strong><p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> does not support any form of tracing GPU or API activities.
This functionality is covered by <a class="xref" href="https://developer.nvidia.com/nsight-systems" target="_blank" shape="rect">NVIDIA Nsight Systems</a>.
</p>
</li>
</ul>
</div>
</div>
<div class="topic concept nested1" id="nvprof-metric-collection"><a name="nvprof-metric-collection" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvprof-metric-collection" name="nvprof-metric-collection" shape="rect">6.2. Metric Collection</a></h3>
<div class="body conbody">
<ul class="ul">
<li class="li"><strong class="ph b">Finding available metrics</strong><p class="p">
For nvprof, you can use <samp class="ph codeph">--query-metrics</samp> to see the list of metrics available for the current devices on your machine.
You can also use <samp class="ph codeph">--devices</samp> to filter which local devices to query.
For <span class="keyword">NVIDIA Nsight Compute CLI</span>, this functionality is the same.
However, in addition, you can combine <samp class="ph codeph">--query-metrics</samp> with <samp class="ph codeph">--chip [chipname]</samp> to query the
available metrics for any chip, not only the ones in your present CUDA devices.
</p>
<p class="p">
Note that metric names have changed between nvprof and <span class="keyword">NVIDIA Nsight Compute CLI</span>
and metric names also differ between chips after (and including) GV100 and those before.
See <a class="xref" href="index.html#nvprof-metric-comparison" shape="rect">Metric Comparison</a> for a comparison of nvprof and <span class="keyword">NVIDIA Nsight Compute</span> metric names.
</p>
<p class="p">
On Volta and newer GPUs, most metrics are named using a base name and various suffixes, e.g. <dfn class="term">sm__throughput.avg.pct_of_peak_sustained_elapsed</dfn>.
The base name is <dfn class="term">sm__throughput</dfn> and the suffix is <dfn class="term">avg.pct_of_peak_sustained_elapsed</dfn>.
This is because most metrics follow the same structure and have the same set of suffixes.
You need to pass the full name to <span class="keyword">NVIDIA Nsight Compute</span> when selecting a metric for profiling.
</p>
<p class="p">
To reduce the number of metrics shown for Volta and newer GPUs when using <samp class="ph codeph">--query-metrics</samp>, by default only the base names are shown.
Use <samp class="ph codeph">--query-metrics-mode suffix --metrics <metrics list></samp> to see the full names for the chosen metrics.
Use <samp class="ph codeph">--query-metrics-mode all</samp> to see all metrics with their full name directly.
</p>
</li>
<li class="li"><strong class="ph b">Selecting which metrics to collect</strong><p class="p">
In both nvprof and <span class="keyword">NVIDIA Nsight Compute CLI</span>, you can specify a comma-separated list of metric names
to the <samp class="ph codeph">--metrics</samp> option. While nvprof would allow you to collect either a list or all metrics,
in <span class="keyword">NVIDIA Nsight Compute CLI</span> you can use regular expressions to select a more fine-granular subset of all available metrics.
For example, you can use <samp class="ph codeph">--metrics "regex:.*"</samp> to collect all metrics, or <samp class="ph codeph">--metrics "regex:smsp__cycles_elapsed"</samp>
to collect all "smsp__cycles_elapsed" metrics.
</p>
</li>
<li class="li"><strong class="ph b">Selecting which events to collect</strong><p class="p">
You cannot collect any events in <span class="keyword">NVIDIA Nsight Compute CLI</span>.
</p>
</li>
<li class="li"><strong class="ph b">Selecting which section to collect</strong><p class="p">
In nvprof, you can either collect individual metrics or events, or a pre-configured set (all, analysis-metrics).
<span class="keyword">NVIDIA Nsight Compute CLI</span> adds the concept of a <dfn class="term">section</dfn>.
A section is a file that describes which metrics to collect for which GPU architecture, or architecture range.
Furthermore, it defines how those metrics will be shown in both the command line output or the user interface.
This includes structuring in tables, charts, histograms etc.
</p>
<p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> comes with a set of pre-defined sections, located in the <samp class="ph codeph">sections</samp>
directory. You can inspect, modify or extend those, as well as add new ones, e.g. to easily collect recurring metric sets.
Each section specifies a unique <dfn class="term">section identifier</dfn>, and there must not be two sections with the same identifier in the search path.
</p>
<p class="p">
By default, the sections associated with the default section set are collected.
You can select one or more individual sections using the <samp class="ph codeph">--section [section identifier]</samp> option one or more times.
If no <samp class="ph codeph">--section</samp> option is given, but <samp class="ph codeph">--metrics</samp> is used, no sections will be collected.
</p>
</li>
<li class="li"><strong class="ph b">Selecting which section set to collect</strong><p class="p">
In nvprof, you can either collect individual metrics or events, or a pre-configured set (all, analysis-metrics).
<span class="keyword">NVIDIA Nsight Compute CLI</span> adds the concept of <dfn class="term">section sets</dfn>.
A section set defines a group of sections to collect together, in order to achieve different profiling overheads,
depending on the required analysis level of detail.
</p>
<p class="p">
If no other options are selected, the default section set if collected.
You can select one or more sets using the <samp class="ph codeph">--set [set identifier]</samp> option one or more times.
If no <samp class="ph codeph">--set</samp> option is given, but <samp class="ph codeph">--section</samp> or <samp class="ph codeph">--metrics</samp> is used,
no sets will be collected.
</p>
</li>
</ul>
</div>
</div>
<div class="topic concept nested1" id="nvprof-metric-comparison"><a name="nvprof-metric-comparison" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvprof-metric-comparison" name="nvprof-metric-comparison" shape="rect">6.3. Metric Comparison</a></h3>
<div class="body conbody">
<p class="p"><span class="keyword">NVIDIA Nsight Compute</span> uses two groups of metrics, depending on which GPU
architecture is profiled. For nvprof metrics, the following table lists the equivalent
metrics in <span class="keyword">NVIDIA Nsight Compute</span>, if available. For a detailed explanation
of the structuring of PerfWorks metrics, see <a class="xref" href="../ProfilingGuide/index.html#metrics-structure" shape="rect">Metrics Structure</a>.
</p>
<p class="p">
Metrics starting with <dfn class="term">sm__</dfn> are collected per-SM.
Metrics starting with <dfn class="term">smsp__</dfn> are collected per-SM subpartition.
However, all corresponding nvprof events are collected per-SM, only.
Check the <a class="xref" href="../ProfilingGuide/index.html#metrics-guide" shape="rect">Metrics Guide</a> for more details on these terms.
</p>
<div class="tablenoborder"><a name="nvprof-metric-comparison__nvprof-metric-comparison-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="nvprof-metric-comparison__nvprof-metric-comparison-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 10. Metrics Mapping Table from CUPTI to PerfWorks</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="24%" id="d54e3529" rowspan="1" colspan="1">nvprof Metric</th>
<th class="entry" valign="top" width="76%" id="d54e3532" rowspan="1" colspan="1">PerfWorks Metric or Formula (>= SM 7.0)</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">achieved_occupancy</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">sm__warps_active.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">atomic_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_set_accesses_pipe_lsu_mem_global_op_atom.sum + l1tex__t_set_accesses_pipe_lsu_mem_global_op_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">atomic_transactions_per_request</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">(l1tex__t_sectors_pipe_lsu_mem_global_op_atom.sum + l1tex__t_sectors_pipe_lsu_mem_global_op_red.sum) / (l1tex__t_requests_pipe_lsu_mem_global_op_atom.sum
+ l1tex__t_requests_pipe_lsu_mem_global_op_red.sum)
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">branch_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_average_branch_targets_threads_uniform.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">cf_executed</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_pipe_cbu.sum + smsp__inst_executed_pipe_adu.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">cf_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">cf_issued</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">double_precision_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp64.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">dram_read_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">dram__bytes_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">dram_read_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">dram__bytes_read.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">dram_read_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">dram__sectors_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">dram_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">dram__throughput.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">dram_write_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">dram__bytes_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">dram_write_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">dram__bytes_write.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">dram_write_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">dram__sectors_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">eligible_warps_per_cycle</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warps_eligible.sum.per_cycle_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_dp</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_dadd_pred_on.sum + smsp__sass_thread_inst_executed_op_dmul_pred_on.sum + smsp__sass_thread_inst_executed_op_dfma_pred_on.sum
* 2
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_dp_add</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_dadd_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_dp_fma</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_dfma_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_dp_mul</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_dmul_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_hp</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_hadd_pred_on.sum + smsp__sass_thread_inst_executed_op_hmul_pred_on.sum + smsp__sass_thread_inst_executed_op_hfma_pred_on.sum
* 2
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_hp_add</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_hadd_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_hp_fma</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_hfma_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_hp_mul</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_hmul_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_sp</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_fadd_pred_on.sum + smsp__sass_thread_inst_executed_op_fmul_pred_on.sum + smsp__sass_thread_inst_executed_op_ffma_pred_on.sum
* 2
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_sp_add</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_fadd_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_sp_fma</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_ffma_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_sp_mul</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_fmul_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_count_sp_special</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_dp_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_ops_dadd_dmul_dfma_pred_on.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_hp_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_ops_hadd_hmul_hfma_pred_on.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">flop_sp_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_ops_fadd_fmul_ffma_pred_on.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gld_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_average_data_bytes_per_sector_mem_global_op_ld.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gld_requested_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gld_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_bytes_pipe_lsu_mem_global_op_ld.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gld_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_sectors_pipe_lsu_mem_global_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gld_transactions_per_request</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__average_t_sectors_per_request_pipe_lsu_mem_global_op_ld.ratio</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">global_atomic_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_lsu_mem_global_op_atom.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">global_hit_rate</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">(l1tex__t_sectors_pipe_lsu_mem_global_op_ld_lookup_hit.sum +
l1tex__t_sectors_pipe_lsu_mem_global_op_st_lookup_hit.sum +
l1tex__t_sectors_pipe_lsu_mem_global_op_red_lookup_hit.sum +
l1tex__t_sectors_pipe_lsu_mem_global_op_atom_lookup_hit.sum) /
(l1tex__t_sectors_pipe_lsu_mem_global_op_ld.sum +
l1tex__t_sectors_pipe_lsu_mem_global_op_st.sum +
l1tex__t_sectors_pipe_lsu_mem_global_op_red.sum +
l1tex__t_sectors_pipe_lsu_mem_global_op_atom.sum)
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">global_load_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_lsu_mem_global_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">global_reduction_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_lsu_mem_global_op_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">global_store_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_lsu_mem_global_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gst_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_average_data_bytes_per_sector_mem_global_op_st.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gst_requested_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gst_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_bytes_pipe_lsu_mem_global_op_st.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gst_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_sectors_pipe_lsu_mem_global_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">gst_transactions_per_request</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__average_t_sectors_per_request_pipe_lsu_mem_global_op_st.ratio</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">half_precision_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp16.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_bit_convert</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_conversion_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_compute_ld_st</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_memory_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_control</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_control_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_global_atomics</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_inst_executed_op_global_atom.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_global_loads</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_global_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_global_reductions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_global_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_global_stores</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_global_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_local_loads</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_local_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_local_stores</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_local_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_shared_atomics</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_shared_atom.sum + smsp__inst_executed_op_shared_atom_dot_alu.sum + smsp__inst_executed_op_shared_atom_dot_cas.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_shared_loads</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_shared_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_shared_stores</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_shared_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_surface_atomics</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_surface_atom.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_surface_loads</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_surface_ld.sum + smsp__inst_executed_op_shared_atom_dot_alu.sum + smsp__inst_executed_op_shared_atom_dot_cas.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_surface_reductions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_surface_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_surface_stores</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_surface_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_executed_tex_ops</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_op_texture.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_fp_16</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_fp16_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_fp_32</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_fp32_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_fp_64</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_fp64_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_integer</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_integer_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_inter_thread_communication</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_inter_thread_communication_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_issued</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_issued.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_misc</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_thread_inst_executed_op_misc_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_per_warp</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__average_inst_executed_per_warp.ratio</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">inst_replay_overhead</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">ipc</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed.avg.per_cycle_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">issue_slot_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__issue_active.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">issue_slots</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_issued.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">issued_ipc</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_issued.avg.per_cycle_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l1_sm_lg_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__lsu_writeback_active.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_atomic_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">2 * ( lts__t_sectors_op_atom.sum.per_second
+ lts__t_sectors_op_red.sum.per_second
)
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_atomic_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">2 * ( lts__t_sectors_op_atom.sum
+ lts__t_sectors_op_red.sum
)
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_global_atomic_store_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_bytes_equiv_l1sectormiss_pipe_lsu_mem_global_op_atom.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_global_load_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_bytes_equiv_l1sectormiss_pipe_lsu_mem_global_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_local_global_store_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_bytes_equiv_l1sectormiss_pipe_lsu_mem_local_op_st.sum + lts__t_bytes_equiv_l1sectormiss_pipe_lsu_mem_global_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_local_load_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_bytes_equiv_l1sectormiss_pipe_lsu_mem_local_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_read_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1"> lts__t_sectors_op_read.sum.per_second
+ lts__t_sectors_op_atom.sum.per_second
+ lts__t_sectors_op_red.sum.per_second
<a class="xref" href="index.html#nvprof-metric-comparison__l2-read-reduction-footnote" shape="rect"><sup>2</sup></a></td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_read_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1"> lts__t_sectors_op_read.sum
+ lts__t_sectors_op_atom.sum
+ lts__t_sectors_op_red.sum
<a class="xref" href="index.html#nvprof-metric-comparison__l2-read-reduction-footnote" shape="rect"><sup>2</sup></a></td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_surface_load_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_bytes_equiv_l1sectormiss_pipe_tex_mem_surface_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_surface_store_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_bytes_equiv_l1sectormiss_pipe_tex_mem_surface_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_tex_hit_rate</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sector_hit_rate.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_tex_read_hit_rate</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sector_op_read_hit_rate.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_tex_read_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_read.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_tex_read_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_tex_write_hit_rate</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sector_op_write_hit_rate.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_tex_write_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_write.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_tex_write_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_write_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1"> lts__t_sectors_op_write.sum.per_second
+ lts__t_sectors_op_atom.sum.per_second
+ lts__t_sectors_op_red.sum.per_second
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">l2_write_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1"> lts__t_sectors_op_write.sum
+ lts__t_sectors_op_atom.sum
+ lts__t_sectors_op_red.sum
</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">ldst_executed</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">ldst_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_pipe_lsu.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">ldst_issued</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_hit_rate</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_load_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_lsu_mem_local_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_load_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_bytes_pipe_lsu_mem_local_op_ld.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_load_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_sectors_pipe_lsu_mem_local_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_load_transactions_per_request</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__average_t_sectors_per_request_pipe_lsu_mem_local_op_ld.ratio</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_memory_overhead</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_store_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_lsu_mem_local_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_store_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_sectors_pipe_lsu_mem_local_op_st.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_store_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_sectors_pipe_lsu_mem_local_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">local_store_transactions_per_request</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__average_t_sectors_per_request_pipe_lsu_mem_local_op_st.ratio</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_data_receive_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_data_transmission_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_overhead_data_received</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">(nvlrx__bytes_data_protocol.sum / nvlrx__bytes.sum) * 100</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_overhead_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">(nvltx__bytes_data_protocol.sum / nvltx__bytes.sum) * 100</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_receive_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">nvlrx__bytes.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_total_data_received</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">nvlrx__bytes.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_total_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">nvltx__bytes.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_total_nratom_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_total_ratom_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_total_response_data_received</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_total_write_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_transmit_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">nvltx__bytes.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_user_data_received</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">nvlrx__bytes_data_user.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_user_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">nvltx__bytes_data_user.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_user_nratom_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_user_ratom_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_user_response_data_received</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">nvlink_user_write_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">pcie_total_data_received</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">pcie__read_bytes.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">pcie_total_data_transmitted</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">pcie__write_bytes.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__sass_average_data_bytes_per_wavefront_mem_shared.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_load_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__data_pipe_lsu_wavefronts_mem_shared_op_ld.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_load_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__data_pipe_lsu_wavefronts_mem_shared_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_load_transactions_per_request</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_store_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__data_pipe_lsu_wavefronts_mem_shared_op_st.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_store_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__data_pipe_lsu_wavefronts_mem_shared_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_store_transactions_per_request</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">shared_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__data_pipe_lsu_wavefronts_mem_shared.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">single_precision_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__pipe_fma_cycles_active.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sm_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__cycles_active.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sm_tex_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__texin_sm2tex_req_cycles_active.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">special_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_pipe_xu.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_constant_memory_dependency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_imc_miss_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_exec_dependency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_short_scoreboard_per_warp_active.pct + smsp__warp_issue_stalled_wait_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_inst_fetch</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_no_instruction_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_memory_dependency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_long_scoreboard_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_memory_throttle</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_drain_per_warp_active.pct + smsp__warp_issue_stalled_lg_throttle_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_not_selected</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_not_selected_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_other</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_dispatch_stall_per_warp_active.pct + smsp__warp_issue_stalled_misc_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_pipe_busy</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_math_pipe_throttle_per_warp_active.pct + smsp__warp_issue_stalled_mio_throttle_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_sleeping</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_sleeping_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_sync</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_barrier_per_warp_active.pct + smsp__warp_issue_stalled_membar_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">stall_texture</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__warp_issue_stalled_tex_throttle_per_warp_active.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">surface_atomic_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_tex_mem_surface_op_atom.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">surface_load_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_tex_mem_surface_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">surface_reduction_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_tex_mem_surface_op_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">surface_store_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_tex_mem_surface_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_read_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_read * 32</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_read_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_read.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_read_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_read_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_write_bytes</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_write * 32</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_write_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_write.sum.per_second</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_write_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">sysmem_write_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tensor_precision_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">sm__pipe_tensor_op_hmma_cycles_active.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tensor_precision_int_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">sm__pipe_tensor_op_imma_cycles_active.avg.pct_of_peak_sustained_active (SM 7.2+)</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tex_cache_hit_rate</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_sector_hit_rate.pct</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tex_cache_throughput</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tex_cache_transactions</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__lsu_writeback_active.avg.pct_of_peak_sustained_active + l1tex__tex_writeback_active.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tex_fu_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__inst_executed_pipe_tex.avg.pct_of_peak_sustained_active</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tex_sm_tex_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__f_tex2sm_cycles_active.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tex_sm_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">sm__mio2rf_writeback_active.avg.pct_of_peak_sustained_elapsed</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">tex_utilization</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">texture_load_requests</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">l1tex__t_requests_pipe_tex_mem_texture.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">warp_execution_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__thread_inst_executed_per_inst_executed.ratio</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="24%" headers="d54e3529" rowspan="1" colspan="1">warp_nonpred_execution_efficiency</td>
<td class="entry" valign="top" width="76%" headers="d54e3532" rowspan="1" colspan="1">smsp__thread_inst_executed_per_inst_executed.pct</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="nvprof-event-comparison"><a name="nvprof-event-comparison" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvprof-event-comparison" name="nvprof-event-comparison" shape="rect">6.4. Event Comparison</a></h3>
<div class="body conbody">
<p class="p">
For nvprof events, the following table lists the equivalent
metrics in <span class="keyword">NVIDIA Nsight Compute</span>, if available. For a detailed explanation
of the structuring of PerfWorks metrics, see <a class="xref" href="../ProfilingGuide/index.html#metrics-structure" shape="rect">Metrics Structure</a>.
</p>
<p class="p">
Metrics starting with <dfn class="term">sm__</dfn> are collected per-SM.
Metrics starting with <dfn class="term">smsp__</dfn> are collected per-SM subpartition.
However, all corresponding nvprof events are collected per-SM, only.
Check the <a class="xref" href="../ProfilingGuide/index.html#metrics-guide" shape="rect">Metrics Guide</a> for more details on these terms.
</p>
<div class="tablenoborder"><a name="nvprof-event-comparison__nvprof-event-comparison-table" shape="rect">
<!-- --></a><table cellpadding="4" cellspacing="0" summary="" id="nvprof-event-comparison__nvprof-event-comparison-table" class="table" frame="border" border="1" rules="all">
<caption><span class="tablecap">Table 11. Events Mapping Table from CUPTI Events to PerfWorks Metrics for Compute Capability >= 7.0</span></caption>
<thead class="thead" align="left">
<tr class="row">
<th class="entry" valign="top" width="30%" id="d54e5256" rowspan="1" colspan="1">nvprof Event</th>
<th class="entry" valign="top" id="d54e5259" rowspan="1" colspan="1">PerfWorks Metric or Formula (>= SM 7.0)</th>
</tr>
</thead>
<tbody class="tbody">
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">active_cycles</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__cycles_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">active_cycles_pm</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__cycles_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">active_cycles_sys</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sys__cycles_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">active_warps</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__warps_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">active_warps_pm</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__warps_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">atom_count</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_generic_atom_dot_alu.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">elapsed_cycles_pm</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__cycles_elapsed.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">elapsed_cycles_sm</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__cycles_elapsed.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">elapsed_cycles_sys</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sys__cycles_elapsed.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">fb_subp0_read_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">dram__sectors_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">fb_subp1_read_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">dram__sectors_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">fb_subp0_write_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">dram__sectors_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">fb_subp1_write_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">dram__sectors_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">global_atom_cas</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_generic_atom_dot_cas.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">gred_count</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_global_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__inst_executed.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fma_pipe_s0</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fma.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fma_pipe_s1</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fma.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fma_pipe_s2</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fma.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fma_pipe_s3</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fma.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp16_pipe_s0</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp16.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp16_pipe_s1</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp16.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp16_pipe_s2</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp16.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp16_pipe_s3</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp16.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp64_pipe_s0</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp64.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp64_pipe_s1</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp64.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp64_pipe_s2</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp64.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_executed_fp64_pipe_s3</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_pipe_fp64.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_issued1</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__inst_issued.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_read_sector_misses</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_read_lookup_miss.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_read_sector_misses</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_read_lookup_miss.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_read_sysmem_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_read_sysmem_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_read_tex_hit_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_read_lookup_hit.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_read_tex_hit_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_read_lookup_hit.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_read_tex_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_read_tex_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_read.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_total_read_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_read.sum + lts__t_sectors_op_atom.sum + lts__t_sectors_op_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_total_read_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_read.sum + lts__t_sectors_op_atom.sum + lts__t_sectors_op_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_total_write_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_write.sum + lts__t_sectors_op_atom.sum + lts__t_sectors_op_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_total_write_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_write.sum + lts__t_sectors_op_atom.sum + lts__t_sectors_op_red.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_write_sector_misses</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_write_lookup_miss.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_write_sector_misses</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_op_write_lookup_miss.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_write_sysmem_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_write_sysmem_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_aperture_sysmem_op_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_write_tex_hit_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_write_lookup_hit.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_write_tex_hit_sectors</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_write_lookup_hit.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp0_write_tex_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">l2_subp1_write_tex_sector_queries</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">lts__t_sectors_srcunit_tex_op_write.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">not_predicated_off_thread_inst_executed</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__thread_inst_executed_pred_on.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">pcie_rx_active_pulse</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">pcie_tx_active_pulse</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_00</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_01</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_02</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_03</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_04</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_05</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_06</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">prof_trigger_07</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">n/a</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">inst_issued0</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__issue_inst0.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">sm_cta_launched</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">sm__ctas_launched.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_load</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_shared_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_store</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_shared_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">generic_load</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_generic_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">generic_store</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_generic_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">global_load</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_global_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">global_store</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_global_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">local_load</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_local_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">local_store</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_local_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_atom</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_op_shared_atom.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_atom_cas</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__inst_executed_shared_atom_dot_cas.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_ld_bank_conflict</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_st_bank_conflict</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_ld_transactions</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">l1tex__data_pipe_lsu_wavefronts_mem_shared_op_ld.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">shared_st_transactions</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">l1tex__data_pipe_lsu_wavefronts_mem_shared_op_st.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">tensor_pipe_active_cycles_s0</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__pipe_tensor_cycles_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">tensor_pipe_active_cycles_s1</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__pipe_tensor_cycles_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">tensor_pipe_active_cycles_s2</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__pipe_tensor_cycles_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">tensor_pipe_active_cycles_s3</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__pipe_tensor_cycles_active.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">thread_inst_executed</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__thread_inst_executed.sum</td>
</tr>
<tr class="row">
<td class="entry" valign="top" width="30%" headers="d54e5256" rowspan="1" colspan="1">warps_launched</td>
<td class="entry" valign="top" headers="d54e5259" rowspan="1" colspan="1">smsp__warps_launched.sum</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
<div class="topic concept nested1" id="nvprof-filtering"><a name="nvprof-filtering" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvprof-filtering" name="nvprof-filtering" shape="rect">6.5. Filtering</a></h3>
<div class="body conbody">
<ul class="ul">
<li class="li"><strong class="ph b">Filtering by kernel name</strong><p class="p"> Both nvprof and <span class="keyword">NVIDIA Nsight Compute CLI</span> support filtering which
kernels' data should be collected. In nvprof, the option is
<samp class="ph codeph">--kernels</samp> and applies to following metric collection
options. In <span class="keyword">NVIDIA Nsight Compute CLI</span>, the option is named
<samp class="ph codeph">--kernel-regex</samp> and applies to the complete application
execution. In other words, <span class="keyword">NVIDIA Nsight Compute CLI</span> does not
currently support collecting different metrics for different kernels, unless
they execute on different GPU architectures.
</p>
</li>
<li class="li"><strong class="ph b">Filtering by kernel ID</strong><p class="p">
Nvprof allows users to specify which kernels to profile using a kernel ID description, using the same <samp class="ph codeph">--kernels</samp> option.
In <span class="keyword">NVIDIA Nsight Compute CLI</span>, the syntax for this kernel ID is identical, but the option is named
<samp class="ph codeph">--kernel-id</samp>.
</p>
</li>
<li class="li"><strong class="ph b">Filtering by device</strong><p class="p"> Both nvprof and <span class="keyword">NVIDIA Nsight Compute CLI</span> use
<samp class="ph codeph">--devices</samp> to filter the devices which to profile. In
contrast to nvprof, in <span class="keyword">NVIDIA Nsight Compute CLI</span> the option applies
globally, not only to following options.
</p>
</li>
</ul>
</div>
</div>
<div class="topic concept nested1" id="nvprof-output"><a name="nvprof-output" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvprof-output" name="nvprof-output" shape="rect">6.6. Output</a></h3>
<div class="body conbody">
<ul class="ul">
<li class="li"><strong class="ph b">API trace and summary</strong><p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> does not support any form of API-usage related output.
No API data is captured during profiling.
</p>
</li>
<li class="li"><strong class="ph b">Dependency analysis</strong><p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> does not support any dependency analysis.
No API data is captured during profiling.
</p>
</li>
<li class="li"><strong class="ph b">GPU trace</strong><p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> does not support any GPU trace output.
Due to kernel replay during profiling, kernel executions are serialized, and start and end timestamps do
not necessarily match those during application execution.
In addition, no records for memory activities are recorded.
</p>
</li>
<li class="li"><strong class="ph b">Print summary</strong><p class="p">
While nvprof has several command line options to specify which summary information to print,
<span class="keyword">NVIDIA Nsight Compute CLI</span> uses further arguments to the <samp class="ph codeph">--summary</samp> options.
Profiling data can be summarized <samp class="ph codeph">per-gpu</samp>, <samp class="ph codeph">per-kernel</samp> or <samp class="ph codeph">per-nvtx</samp> context.
</p>
</li>
<li class="li"><strong class="ph b">Kernel name demangling</strong><p class="p">
Nvprof allows users to decide between name demangling on or off using the <samp class="ph codeph">--demangling</samp> options.
<span class="keyword">NVIDIA Nsight Compute CLI</span> currently always demangles kernel names in the output.
In addition, the option <samp class="ph codeph">--kernel-regex-base</samp> can be used to decide which name format should be used
when matching kernel names during filtering.
</p>
</li>
<li class="li"><strong class="ph b">Pages</strong><p class="p">
Nvprof has no concept of output pages, all data is shown as a list or summarized.
<span class="keyword">NVIDIA Nsight Compute CLI</span> uses <dfn class="term">pages</dfn> to define how data should be structured and printed.
Those correspond to the report pages used in the GUI variant.
The option <samp class="ph codeph">--page</samp> can be used to select which page to show, and <samp class="ph codeph">details</samp> is selected by default.
All pages also support printing in CSV format for easier post-processing, using the <samp class="ph codeph">--csv</samp> option.
</p>
</li>
</ul>
</div>
</div>
<div class="topic concept nested1" id="nvprof-launch-attach"><a name="nvprof-launch-attach" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#nvprof-launch-attach" name="nvprof-launch-attach" shape="rect">6.7. Launch and Attach</a></h3>
<div class="body conbody">
<ul class="ul">
<li class="li"><strong class="ph b">Launching a process for profiling</strong><p class="p">
In nvprof, the application to profile is passed to the tool as a command line argument.
The application must be a local executable.
Alternatively, you can choose to use the tool in a <dfn class="term">daemon mode</dfn> and profile all applicable processes on the local machine (nvprof option <samp class="ph codeph">--profile-all-processes</samp>).
In nvprof, the decision to profile the complete process tree or only the root process is done via the <samp class="ph codeph">--profile-child-processes</samp> flag.
In <span class="keyword">NVIDIA Nsight Compute CLI</span>, the <samp class="ph codeph">--target-processes</samp> option is used for this.
</p>
<p class="p"><span class="keyword">NVIDIA Nsight Compute CLI</span> has several modes to determine which application to collect data for.
By default, the executable passed via the command line to the tool is started, connected to, and profiled.
This mode is called <samp class="ph codeph">launch-and-attach</samp>.
</p>
</li>
<li class="li"><strong class="ph b">Launching a process for attach</strong><p class="p">
In contrast to nvprof, you can choose to only launch a local executable. In this mode (<samp class="ph codeph">--mode launch</samp>),
the process is started, connected to, but then suspended at the first CUDA API call.
Subsequently, there is a third mode (<samp class="ph codeph">--mode attach</samp>) to attach to any process launched using the
aforementioned mode. In this case, all profiling and output options would be passed to the attaching instance of
<span class="keyword">NVIDIA Nsight Compute CLI</span>.
</p>
</li>
<li class="li"><strong class="ph b">Remote profiling</strong><p class="p"> Finally, using <samp class="ph codeph">launch</samp> and <samp class="ph codeph">attach</samp>, you can
connect to a launched process on a remote machine, which could even run a
different operating system than the local host. Use <samp class="ph codeph">--hostname</samp>
to select which remote host to connect to.
</p>
</li>
</ul>
</div>
</div>
</div>
<div class="topic concept nested0" id="notices-header"><a name="notices-header" shape="rect">
<!-- --></a><h2 class="title topictitle1"><a href="#notices-header" name="notices-header" shape="rect">Notices</a></h2>
<div class="topic reference nested1" id="notice"><a name="notice" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#notice" name="notice" shape="rect"></a></h3>
<div class="body refbody">
<div class="section">
<h3 class="title sectiontitle">Notice</h3>
<p class="p">ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND
SEPARATELY, "MATERIALS") ARE BEING PROVIDED "AS IS." NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE
WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS
FOR A PARTICULAR PURPOSE.
</p>
<p class="p">Information furnished is believed to be accurate and reliable. However, NVIDIA Corporation assumes no responsibility for the
consequences of use of such information or for any infringement of patents or other rights of third parties that may result
from its use. No license is granted by implication of otherwise under any patent rights of NVIDIA Corporation. Specifications
mentioned in this publication are subject to change without notice. This publication supersedes and replaces all other information
previously supplied. NVIDIA Corporation products are not authorized as critical components in life support devices or systems
without express written approval of NVIDIA Corporation.
</p>
</div>
</div>
</div>
<div class="topic reference nested1" id="trademarks"><a name="trademarks" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#trademarks" name="trademarks" shape="rect"></a></h3>
<div class="body refbody">
<div class="section">
<h3 class="title sectiontitle">Trademarks</h3>
<p class="p">NVIDIA and the NVIDIA logo are trademarks or registered trademarks of NVIDIA Corporation
in the U.S. and other countries. Other company and product names may be trademarks of
the respective companies with which they are associated.
</p>
</div>
</div>
</div>
<div class="topic reference nested1" id="copyright-past-to-present"><a name="copyright-past-to-present" shape="rect">
<!-- --></a><h3 class="title topictitle2"><a href="#copyright-past-to-present" name="copyright-past-to-present" shape="rect"></a></h3>
<div class="body refbody">
<div class="section">
<h3 class="title sectiontitle">Copyright</h3>
<p class="p">© <span class="ph">2018</span>-<span class="ph">2022</span> NVIDIA
Corporation and affiliates. All rights reserved.
</p>
<p class="p">This product includes software developed by the Syncro Soft SRL (http://www.sync.ro/).</p>
</div>
</div>
</div>
</div>
<div class="fn"><a id="command-line-options-profile__import-filtering-enabled-option" shape="rect"><sup>1</sup></a>
This filtering option is available when using <samp class="ph codeph">--import</samp>.
</div>
<div class="fn"><a id="nvprof-metric-comparison__l2-read-reduction-footnote" shape="rect"><sup>2</sup></a>
Sector reads from reductions are added here only for compatibility to the current definition of the metric in nvprof.
Reductions do not cause data to be communicated from L2 back to L1.
</div>
<hr id="contents-end"></hr>
</article>
</div>
</div>
<script language="JavaScript" type="text/javascript" charset="utf-8" src="../common/formatting/common.min.js"></script>
<script language="JavaScript" type="text/javascript" charset="utf-8" src="../common/scripts/google-analytics/google-analytics-write.js"></script>
<script language="JavaScript" type="text/javascript" charset="utf-8" src="../common/scripts/google-analytics/google-analytics-tracker.js"></script>
<script type="text/javascript">_satellite.pageBottom();</script>
<script type="text/javascript">var switchTo5x=true;</script><script type="text/javascript" src="http://w.sharethis.com/button/buttons.js"></script><script type="text/javascript">stLight.options({publisher: "998dc202-a267-4d8e-bce9-14debadb8d92", doNotHash: false, doNotCopy: false, hashAddressBar: false});</script></body>
</html>
|