TCP PRR: Patch div/0 in tcp_prr_partialack

[rscheff]

Summary

With clearing of recover_fs in bc7ee8e5bc555, div/0
was observed while processing partial_acks.

Suspect that rewind of an erraneous RTO may be
causing this - with the above change, recover_fs
would no longer retained at the last calculated
value, and reset. But CC_RTO_ERR can reenable
IN_RECOVERY(), without setting this again.

Adding a safety net prior to the division in that
function, which I missed in D28114.

Adding line wraps in that PRR block.

[tuexen]

Maybe mm@ can test this?

[rscheff]

• retain recover_fs past recovery, in case of RTO rollback

[mm]

LGTM, both places where the division happens are now protected.
We are testing it now with public traffic (~ 8 hours), looks good so far.

[tuexen]

I would suggest to separate the whitspace changes and the functional changes into two commits.

[rscheff]

Adjust line length in tcp_prr_partialack, no functional change.

[rscheff]

• separate out whitespace changes

[mm]

LGTM, now both places with the division are protected.
As mentioned earlier, our live test runs fine for now (>12 hours).