Page MenuHomeFreeBSD

Add flowid to siftr(4)
ClosedPublic

Authored by hiren on Mar 17 2015, 8:21 PM.
Tags
None
Referenced Files
Unknown Object (File)
Feb 14 2024, 11:34 PM
Unknown Object (File)
Jan 5 2024, 4:57 PM
Unknown Object (File)
Dec 20 2023, 12:04 AM
Unknown Object (File)
Sep 10 2023, 11:42 PM
Unknown Object (File)
Sep 6 2023, 7:54 PM
Unknown Object (File)
Sep 6 2023, 7:53 PM
Unknown Object (File)
Sep 6 2023, 7:51 PM
Unknown Object (File)
Sep 6 2023, 7:51 PM
Subscribers

Details

Summary

Adding connection flowid as one of the fields to siftr(4).

Diff Detail

Repository
rS FreeBSD src repository - subversion
Lint
Lint Skipped
Unit
Tests Skipped

Event Timeline

hiren retitled this revision from to Add flowid to siftr(4).
hiren updated this object.
hiren edited the test plan for this revision. (Show Details)
hiren added a reviewer: lstewart.
lstewart edited edge metadata.

Minor typo needs fixing.

sys/netinet/siftr.c
491 โ†—(On Diff #4265)

s/pkg/pkt ?

541 โ†—(On Diff #4265)

Ditto

This revision now requires changes to proceed.Mar 18 2015, 12:20 AM

Bah, this is what happens when you develop/test on one system and generate review from other with all the phps that phabric needs.

I am sorry, I should have been more careful. Updating the diffs in a bit.

hiren edited edge metadata.

Updating with lstewart's typo catch.

lstewart edited edge metadata.

Looks good, although you might want to document in the man page any caveats related to when the flow id might be 0 (I can't remember if there are any situations in which the flow id is not set?)

This revision is now accepted and ready to land.Mar 18 2015, 3:12 AM
In D2089#11, @lstewart wrote:

Looks good, although you might want to document in the man page any caveats related to when the flow id might be 0 (I can't remember if there are any situations in which the flow id is not set?)

Hum. Yeah, 0 could be a valid flowid or the default value when its not being set, afaik. I can add that to the manpage if that's what you prefer.

I didn't think 0 was a valid flow id, in which case I think it would be useful to document the cases in which a consumer of SIFTR data might legitimately expect to see a flowid of 0 (if it's even possible - it might be that a flowid is always set in which case you don't need to document anything).

In D2089#15, @lstewart wrote:

I didn't think 0 was a valid flow id, in which case I think it would be useful to document the cases in which a consumer of SIFTR data might legitimately expect to see a flowid of 0

igb for example uses plain cpu /msix/queue id as flowid in which case 0 is a valid flowid.

(if it's even possible - it might be that a flowid is always set in which case you don't need to document anything).

I just looked at what adrian did for netstat -R. On my laptop with a single queue em(4), all I am getting is 0 as flowid with netstat -R. And em(4) is not setting flowid at all in the drivers. So, in this case its a default value.

hiren edited edge metadata.

Update manpage with the caveat about flowid '0'.

This revision now requires review to proceed.Mar 18 2015, 8:51 PM
lstewart edited edge metadata.
This revision is now accepted and ready to land.Mar 18 2015, 11:09 PM
hiren updated this revision to Diff 4280.

Closed by commit rS280233 (authored by @hiren).