https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=244356
--- Comment #72 from Maurizio <maurizio1...@gmail.com> --- (In reply to Olivier Certner from comment #52) The stick, a Kingston DataTraveler G4 32GB, works well if formatted as FAT32, bad if formatted as UFS and the performance will degrade over the time. I have written this script for testing the stick: #!/bin/sh set -x STICK_DEVICE=/dev/da0 FILE_ON_STICK=/media/5ec390d1148aa331/zero.bin COUNT=1024 tmux new-window "script /tmp/dd.log dd if=/dev/zero of=$FILE_ON_STICK bs=1M count=$COUNT; sleep 5; killall iostat" tmux new-window "script /tmp/iostast.log iostat -w 1 $STICK_DEVICE" set +x where FILE_ON_STICK depends on the stick mount point. I have run the script one time with good performance with the stick formatted as FAT32, the log output are: dd_fat32.log (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=214626 ) and iostat_fat32.log (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=214630). The dd instruction takes 49 sec. After formatting the stick as UFS I have runned the script 3 times, the dd instruction takes 57, 141 and 189 seconds. The logs are, in order: dd_ufs.log (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=214627 ), iostat_ufs.log (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=214631), dd_usf2.log (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=214628), iostat_ufs2.log (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=214633) dd_ufs3.log and iostat_ufs3.log (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=214634) Running the script with the stick formatted as UFS, the laptop used for the test, hangs a little. -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ freebsd-usb@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-usb To unsubscribe, send any mail to "freebsd-usb-unsubscr...@freebsd.org"
