In the last episode (May 07), Andre Guibert de Bruet said: > Here's the output of top while moving a large file from one disk to > another on my otherwise idle desktop machine: > > PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND > 26 root -68 -187 0K 12K WAIT 0 141:36 79.98% 79.98% irq7: dc0 ohci0 Is this line truncated? If not, your CPU load is caused by either your network card or your USB interface, not your file copy. > 24 root -64 -183 0K 12K WAIT 0 0:27 1.22% 1.22% irq15: ata1 > Now, I realize that copying a file from one disk to another requires cpu > cycles even with DMA enabled, but is 80% of cpu usage to be expected? > Also, wouldn't one expect process id 24 (irq15: ata1) to be using CPU > during such transfers? It is; I'd even call 1.22% cpu load a bit high, although I have no IDE systems to compare it against. All it's doing is DMA'ing data. -- Dan Nelson dnelson_at_allantgroup.comReceived on Wed May 07 2003 - 18:03:55 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:06 UTC