Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Slow rsync speed on USB3 device mounted via USB3 port #8

Open
priyadarshan opened this issue Jan 18, 2018 · 3 comments
Open

Slow rsync speed on USB3 device mounted via USB3 port #8

priyadarshan opened this issue Jan 18, 2018 · 3 comments

Comments

@priyadarshan
Copy link

Hello,
I am now testing hfsfuse, rsync'ing a large iso file, and I m getting a speed of about 3.8 MB/s:

# rsync -avP -stats /media/hd/FreeBSD-11-RELEASE-x64-USB.iso /tmp
FreeBSD-11-RELEASE-x64-USB.iso
     88,113,152   2%    3.81MB/s    0:18:14

I am also trying on a different server, still with USB 3 device, connected to USB 3 port, with same kind of speed.

I can see the drive as da0: 400.000MB/s transfers:

ugen0.5: <Seagate BUP Slim BK> at usbus0
umass0 numa-domain 0 on uhub1
umass0: <Seagate BUP Slim BK, class 0/0, rev 3.00/1.00, addr 4> on usbus0
umass0:  SCSI over Bulk-Only; quirks = 0x0100
umass0:10:0: Attached to scbus10
da0 at umass-sim0 bus 0 scbus10 target 0 lun 0
da0: <Seagate BUP Slim BK 0302> Fixed Direct Access SPC-4 SCSI device
da0: Serial Number NA7K63XX
da0: 400.000MB/s transfers
da0: 1907729MB (3907029167 512 byte sectors)
da0: quirks=0x2<NO_6_BYTE>

Is that a current limit of hfsfuse, or is there perhaps a switch to activate?

Thank you again.

@priyadarshan
Copy link
Author

I have played around increasing UBLIO_BLOCKSIZE and `UBLIO_ITEMS before mounting hfs+ device, with no changes.

@0x09
Copy link
Owner

0x09 commented Jan 19, 2018

The driver itself is fairly unoptimized; for example I get 23mb/s sustained read on a large file vs 182 mb/s using the native mac driver. I've done profiling to look for weak spots and may do some work on this, but addressing these involves much more intrusive changes to the upstream code. It's fast enough for day to day use so raw speed was not much of a priority, but I'll leave this open as it's a real issue worth improving upon.

@priyadarshan
Copy link
Author

Thank you for checking and clarification. Let me know if I can be of any help testing on FreeBSD.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants