On 2019/12/25 10:33, Randy Dunlap wrote: [...] >> +For sequential write zone files, the file size changes as data is appended at >> +the end of the file, similarly to any regular file system. >> + >> +# dd if=/dev/zero of=/mnt/seq/0 bs=4096 count=1 conv=notrunc oflag=direct >> +1+0 records in >> +1+0 records out >> +4096 bytes (4.1 kB, 4.0 KiB) copied, 1.05112 s, 3.9 kB/s > > why so slow? > >> + >> +# ls -l /mnt/seq/0 >> +-rw-r----- 1 root root 4096 Nov 25 13:23 /mnt/sdh/seq/0 > > I don't understand the "sdh/" here. Please explain for me (not necessarily > in the doc file). The drive I used for generating the example was /dev/sdh and it was in fact mounted under /mnt/sdh/ for the test, but I removed "sdh" from the pasted commands and results to make things simpler. I forgot to remove the drive name on this line. Fixed now. Thank you for the review. -- Damien Le Moal Western Digital Research
