Hello, Since I installed a recent -CURRENT (from 16/08) on my laptop, I noted a big difference during "make clean" in a port directory. With 5.2.1, the make clean operation is transparent, I mean I don't notice any difference on system load or user interactivity. With -CURRENT, I see some little freezes, xmms skips etc. and a large swap is used... My laptop is a Dell Inspiron 4150 (P4 1.8GHz/256MB). Since it's difficult to explain a performance problem with words, I did some tests and used vmstat -w 1 as monitor. -CURRENT uses a GENERIC kernel without any debugging options. The 5.2.1 uses a custom kernel, it's just a GENERIC without some drivers (nothing particular that could explain the differences below). During the test I ran the same apps: X (X4.3 under 5.2.1 and Xorg under -CURRENT), window maker, gaim, firefox, xmms, and some rxvt with screen. Each time I did 2 "make clean" inside the same ports directory. After each test, I looked at swapinfo output. One thing to note is the fact the ports dir is already "clean". Here's what vmstat -w 1 gave me: 5.2.1 first make clean procs memory page disks faults cpu r b w avm fre flt re pi po fr sr ad0 fd0 in sy cs us sy id 0 0 0 191832 87252 0 0 0 0 0 0 0 0 552 0 1612 2 2 95 1 0 0 191832 87252 17 0 0 0 17 0 0 0 443 0 989 2 4 94 1 0 0 191832 87252 0 0 0 0 0 0 3 0 427 0 923 4 2 94 2 0 0 191832 87252 17 0 0 0 17 0 0 0 524 0 1438 5 2 93 3 0 0 191832 87252 0 0 0 0 0 0 0 0 566 0 1694 5 4 92 2 1 0 192420 86472 653 0 9 0 632 0 32 0 476 0 1197 5 4 91 2 3 0 203716 84808 4022 7 5 0 3816 0 57 0 619 0 2059 29 14 57 3 9 0 235544 81412 7382 0 0 0 6848 0 26 0 575 0 2360 67 18 15 3 13 0 256548 78756 8576 0 0 0 8357 0 39 0 618 0 2475 32 29 38 3 0 0 194356 83452 10573 0 0 0 12027 0 16 0 513 0 2159 67 23 9 5 1 0 195168 83212 8914 0 0 0 9161 0 1 0 533 0 2111 74 26 0 2 0 0 194400 83416 8094 0 0 0 8419 0 0 0 566 0 2155 75 25 0 3 0 0 193272 83532 4226 0 0 0 4394 0 10 0 539 0 2665 42 29 29 1 0 0 191848 84956 1212 0 0 0 1568 0 2 0 563 0 1578 5 5 89 2 0 0 191848 84956 0 0 0 0 0 0 0 0 522 0 1450 2 5 93 0 0 0 191848 84956 17 0 0 0 17 0 5 0 486 0 1193 3 2 95 4 0 0 191848 84956 0 0 0 0 0 0 0 0 420 0 950 5 1 95 2 0 0 191848 84956 17 0 0 0 17 0 0 0 419 0 942 3 2 95 3 0 0 191848 84956 0 0 0 0 0 0 0 0 418 0 947 5 1 95 2 0 0 191848 84956 19 0 0 0 17 0 0 0 460 0 1172 4 2 94 3 0 0 191848 84956 0 0 0 0 0 0 1 0 518 0 1350 5 2 92 swapinfo Device 1K-blocks Used Avail Capacity /dev/ad0s3b 524288 0 524288 0% It's easy to figure out when I launched the clean, the ad0 numbers seem normal. The swap is not used. 5.2.1 second make clean: procs memory page disks faults cpu r b w avm fre flt re pi po fr sr ad0 fd0 in sy cs us sy id 0 0 0 191880 83564 559 3 5 0 483 0 0 0 446 0 1133 8 4 88 3 0 0 191880 83564 21 0 0 0 21 0 0 0 506 0 1369 5 2 92 1 0 0 191880 83564 0 0 0 0 0 0 0 0 531 0 1584 5 5 90 2 0 0 191880 83564 17 0 0 0 17 0 0 0 464 0 1156 2 2 96 1 0 0 191880 83564 0 0 0 0 0 0 0 0 514 0 1410 5 7 88 2 0 0 193764 81684 1567 0 0 0 1097 0 0 0 502 0 1387 9 4 87 4 0 0 191880 83564 1550 0 0 0 2020 0 0 0 512 0 1613 11 6 83 2 0 0 191880 83564 3117 0 0 0 3121 0 4 0 556 0 1845 17 7 76 1 0 0 191880 83564 17 0 0 0 17 0 4 0 545 0 1516 5 5 90 2 0 0 191880 83564 0 0 0 0 0 0 0 0 519 0 1428 6 2 91 4 0 0 191880 83564 17 0 0 0 17 0 0 0 457 0 1132 4 3 93 2 0 0 191880 83564 0 0 0 0 0 0 0 0 417 0 903 4 2 94 2 0 0 191880 83564 17 0 0 0 17 0 0 0 418 0 920 2 2 96 1 0 0 191880 83564 0 0 0 0 0 0 0 0 415 0 925 6 2 91 2 0 0 191880 83564 22 0 0 0 17 0 0 0 422 0 881 4 2 95 3 0 0 191880 83588 0 0 0 0 0 0 0 0 419 0 934 6 2 92 swapinfo Device 1K-blocks Used Avail Capacity /dev/ad0s3b 524288 0 524288 0% The second time is of course less stressing for the system. Let's have a look to the same tests under -CURRENT: -CURRENT first make clean: procs memory page disk faults cpu r b w avm fre flt re pi po fr sr ad0 in sy cs us sy id 2 0 0 180764 94540 21 0 0 0 0 0 0 425 2410 970 3 2 95 3 0 0 180764 94540 0 0 0 0 0 0 0 423 2382 959 3 2 95 3 0 0 180764 94536 21 0 0 0 0 0 0 427 2330 958 5 4 92 3 0 0 180764 94520 0 0 0 0 0 0 1 430 2554 1002 2 3 95 1 0 0 180764 94492 25 0 0 0 0 0 4 427 2352 966 2 5 93 4 0 0 180764 94492 0 0 0 0 0 0 0 414 2245 916 5 2 92 2 0 0 180764 94492 66 0 0 0 49 0 8 431 2274 957 6 2 92 3 0 0 180764 94492 0 0 0 0 0 0 0 430 2552 1000 5 3 91 5 0 0 181516 93376 1655 0 14 0 1521 0 46 505 4757 1381 8 9 83 3 3 0 212432 65348 16076 0 2 0 9180 0 21 601 11847 2581 69 24 7 2 5 0 240964 38548 14398 0 0 0 7717 0 5 613 10569 2260 74 26 0 3 8 0 269996 41756 17186 128 0 113 14588 445649 130 676 9470 3049 36 55 9 1 13 0 327708 1260 18511 591 6 82 11808 64616 97 654 9364 2457 53 33 13 0 13 0 327812 1264 468 0 4 300 481 15972 304 765 2832 2641 5 11 84 0 15 0 327828 1692 63 0 12 574 117 1642 578 984 1173 3602 2 13 85 0 20 0 328652 1268 502 2 12 558 575 2068 565 967 1412 3501 4 7 89 2 12 0 328848 1520 103 4 9 208 112 8037 223 652 851 1819 0 5 95 6 13 0 357976 14092 4002 14 34 0 4059 5606 46 478 3454 1249 16 10 74 0 16 0 358164 1420 4624 5 8 343 4636 15189 346 763 2439 2763 12 10 78 2 48 0 357084 1332 163 18 11 495 189 1442 504 922 2187 3509 3 5 92 1 15 0 357976 1584 74 0 0 451 46 3063 448 843 2015 3215 3 10 87 0 18 0 361348 11372 1947 4 12 237 2011 10952 250 643 2554 2299 7 9 84 1 16 0 385016 1460 6868 25 4 372 6849 29816 369 759 3458 2877 15 13 72 2 15 0 385016 1408 1 0 1 239 16 3809 238 741 4355 2638 4 5 91 4 12 0 359612 36892 3766 8 22 3 10051 10065 33 492 3668 1683 18 12 71 4 9 0 327396 64440 14925 3 22 0 21775 0 27 495 7394 2169 52 24 24 2 2 0 212412 170504 18578 0 34 0 45009 0 44 511 8972 3098 60 35 5 5 0 0 185168 182860 13189 2362 236 0 13246 0 235 735 11780 4144 47 31 22 2 0 0 187028 182040 10937 0 0 0 10559 0 0 435 15867 1519 80 20 0 2 4 0 187028 180884 6240 7 68 0 6102 0 66 562 10787 2412 71 26 4 4 7 0 190820 178300 10918 38 121 0 10267 0 118 615 14704 2181 70 30 0 1 4 0 183012 179344 3611 25 105 0 3971 0 105 596 7862 2437 34 15 51 2 0 0 180992 178348 2040 33 104 0 1809 0 105 496 4083 1420 14 7 79 4 0 0 180992 178348 90 0 0 0 45 0 0 387 1873 800 5 3 91 2 0 0 180992 178348 21 0 0 0 0 0 0 386 1917 811 2 2 96 1 0 0 180992 178208 11 10 9 0 0 0 11 399 1900 843 2 1 98 1 0 0 180992 178208 21 0 0 0 4 0 34 426 1829 873 2 1 97 1 0 0 180992 178172 6 0 5 0 0 0 5 386 1904 808 2 3 95 1 0 0 180992 178172 21 0 0 0 0 0 0 389 1887 782 2 1 97 3 0 0 180992 178172 0 0 0 0 0 0 0 388 1879 784 2 1 97 1 0 0 180992 178172 21 0 0 0 0 0 0 385 1900 779 2 2 96 3 0 0 180992 178032 95 0 6 0 45 0 6 393 1904 809 4 5 91 1 0 0 180992 178032 21 0 0 0 0 0 0 392 1900 792 1 4 95 2 0 0 180992 177984 1 0 1 0 0 0 1 384 1910 785 0 4 96 1 0 0 180992 177984 21 0 0 0 0 0 0 388 1864 783 3 1 96 3 0 0 180992 177656 19 0 15 0 0 0 16 404 1843 843 2 2 96 swapinfo Device 1K-blocks Used Avail Capacity /dev/ad0s2b 786432 55052 731380 7% The system activity is completly different and swap has been used. -CURRENT second make clean: procs memory page disk faults cpu r b w avm fre flt re pi po fr sr ad0 in sy cs us sy id 1 1 0 179140 170460 943 11 6 10 904 1626 0 462 3669 1226 9 5 87 2 1 0 179140 170460 3 0 0 0 3 0 0 475 3180 1251 5 2 93 2 1 0 179140 170460 0 0 0 0 0 0 0 485 3623 1341 4 3 93 1 1 0 179140 170456 3 0 1 0 0 0 1 420 2112 1011 3 2 95 1 1 0 179140 170456 0 0 0 0 0 0 0 484 3236 1271 2 3 95 1 1 0 179140 170456 212 0 0 0 94 0 2 569 5013 1698 6 5 88 2 1 0 179140 170456 0 0 0 0 0 0 0 471 3232 1243 3 4 93 2 1 0 179140 170456 0 0 0 0 0 0 0 525 4487 1537 5 4 91 1 1 0 179140 170456 89 0 0 0 45 0 0 520 4153 1478 7 2 91 2 1 0 179140 170456 0 0 0 0 0 0 0 534 4383 1524 4 4 92 2 1 0 179140 170456 0 0 0 0 0 0 0 533 4157 1488 5 2 93 2 1 0 179140 170436 7653 0 1 0 7190 0 3 582 5933 2227 28 11 61 0 1 0 179140 170420 11 0 3 0 0 0 3 429 2244 1051 5 2 92 6 5 0 213272 142032 14636 0 1 0 7405 0 1 467 11598 1561 54 20 26 5 7 0 268484 95108 21043 0 7 0 9055 0 7 554 10820 1936 70 30 0 2 15 0 354884 23908 31746 0 1 0 13880 0 27 594 13591 2177 64 36 0 0 17 0 383164 1592 15294 87 5 296 9301 517506 295 848 7082 3048 25 60 15 2 17 0 383164 1492 23 0 12 93 43 159 103 696 4481 2093 4 6 90 1 17 0 383164 1272 31 0 19 316 77 567 333 890 4007 3085 4 4 92 0 17 0 383172 1288 170 4 6 512 193 1086 515 1031 2838 3763 1 12 88 2 21 0 383424 1264 199 0 4 273 187 1848 271 825 3142 2510 2 7 92 2 13 0 383424 1256 15 0 2 192 24 8269 199 761 2287 1954 1 4 95 2 26 0 354700 39580 802 36 72 0 7340 3806 65 585 2606 1642 9 10 81 1 23 0 356384 37724 631 31 76 0 332 0 77 676 5319 2010 11 5 84 2 23 0 356420 36680 677 26 68 0 507 0 68 531 3214 1465 5 7 88 1 18 0 298788 88612 8897 26 92 0 23041 0 92 621 6255 2903 32 22 47 1 12 0 268492 114544 522 2 93 0 7161 0 94 600 3441 1985 8 8 84 2 15 0 212260 166092 18069 6 88 0 31051 0 83 608 9027 3553 52 32 16 0 8 0 210052 158676 3326 1793 100 0 548 0 102 641 5113 2065 8 6 86 1 5 0 183512 183148 10438 16 80 0 15087 0 82 655 14062 2673 65 24 11 3 6 0 186136 181536 10343 4 97 0 9919 0 96 633 15278 2331 68 32 0 4 6 0 185256 180636 5051 32 106 0 4831 0 105 674 10696 2784 73 17 10 2 4 0 185128 179580 9768 10 108 0 9474 0 108 613 15113 2497 66 34 0 2 3 0 183536 177220 8809 183 134 0 8032 0 133 618 9339 2888 45 28 27 3 3 0 181024 177044 1067 24 159 0 996 0 157 692 5377 2428 12 6 81 2 1 0 180560 177076 5341 44 36 0 5037 0 37 612 5469 1950 26 9 65 3 1 0 179140 178500 3138 0 0 0 3288 0 0 525 4326 1526 12 4 84 3 1 0 179140 178376 7677 0 9 0 7190 0 9 584 5605 2403 25 18 57 0 1 0 179140 178376 0 0 0 0 0 0 0 489 3529 1349 2 5 94 3 1 0 179140 178340 92 0 2 0 45 0 2 542 4371 1550 7 5 88 1 1 0 180592 176824 2161 0 2 0 1657 0 2 480 3144 1249 9 5 85 1 1 0 179140 178280 5511 0 0 0 5533 0 6 556 5256 1756 19 9 72 3 1 0 179140 178280 0 0 0 0 0 0 0 553 4978 1659 3 6 91 1 3 0 179140 177408 61 8 46 0 0 0 45 566 4035 1691 5 5 91 4 1 0 180560 175512 3908 6 58 0 3243 0 90 643 4950 2062 14 9 78 0 3 0 179140 176708 3849 1 7 0 3951 0 6 538 4628 2059 18 8 75 % swapinfo Device 1K-blocks Used Avail Capacity /dev/ad0s2b 786432 60716 725716 8% The second make clean uses less resources but still more than 5.2.1, and swap used increased. That's all. It really seems something is not Ok with recent -CURRENT, I don't thing it's related to my hardware. Maybe someone could find why a such difference between 5.2.1 and -CURRENT. MarcReceived on Wed Aug 18 2004 - 18:05:35 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:07 UTC