Page MenuHomeFreeBSD

New port: sysutils/ltfs: Reference implementation of LTFS for stand alone tape drive

Authored by kbowling on Jan 14 2019, 7:04 PM.
Referenced Files
Unknown Object (File)
Sat, May 6, 9:59 AM
Unknown Object (File)
Mar 31 2023, 10:02 PM
Unknown Object (File)
Mar 4 2023, 12:03 AM
Unknown Object (File)
Mar 1 2023, 5:46 PM
Unknown Object (File)
Feb 19 2023, 1:30 AM
Unknown Object (File)
Feb 13 2023, 4:51 PM
Unknown Object (File)
Feb 4 2023, 8:04 PM
Unknown Object (File)
Dec 26 2022, 8:07 AM



Add a new port for the FUSE based LTFS reference implementation

Diff Detail

rP FreeBSD ports repository
Lint Not Applicable
Tests Not Applicable

Event Timeline

Looks good. I assume you've made sure this installs correctly. I put a couple of suggestions in the description file, although the comment about MAXPHYS might also go into something that the user will see when he installs the package.

I think most folks won't think about the blocksize, but it will affect whether they can read tapes from other systems.

12 ↗(On Diff #52833)

Good catch. I have modified our local source to use DESTDIR there, but forgot to upstream the fix.

4 ↗(On Diff #52833)

I would add something like this:

This only works with IBM-branded LTO-5 or newer tape drives, although there is a file backend for simulation purposes.

You will only be able to read and write tapes with a blocksize <= MAXPHYS, which is 128KB by default. Put the following in your kernel config file and recompile/reinstall to allow reading and writing tapes up to the maximum LTFS blocksize of 1MB:

options MAXPHYS=(1024*1056)

This revision is now accepted and ready to land.Jan 14 2019, 8:06 PM
rpokala added inline comments.
2 ↗(On Diff #52833)


4 ↗(On Diff #52833)

@ken: why (1024*1056) rather than (1024*1024)?

linimon retitled this revision from sysutils/ltfs: New port to New port: sysutils/ltfs: Reference implementation of LTFS for stand alone tape drive.Jan 15 2019, 4:21 PM
4 ↗(On Diff #52833)

Just in case the buffer is misaligned. If that happens, you'll still be able to get the full 1MB I/O down to the drive if MAXPHYS is slightly larger than 1MB.

This revision was automatically updated to reflect the committed changes.