Page MenuHomeFreeBSD
Feed Advanced Search

Oct 27 2021

rrs requested review of D32693: tcp: Congestion control cleanup..
Oct 27 2021, 1:54 PM

Oct 26 2021

rrs added inline comments to D31716: sys: Nuke double-semicolons.
Oct 26 2021, 5:49 PM
rrs committed rG12752978d32b: tcp: The rack stack can incorrectly have an overflow when calculating a burst… (authored by rrs).
tcp: The rack stack can incorrectly have an overflow when calculating a burst…
Oct 26 2021, 5:20 PM
rrs closed D32668: tcp: The rack stack can incorrectly have an overflow when calculating a burst delay..
Oct 26 2021, 5:20 PM
rrs abandoned D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..
Oct 26 2021, 4:28 PM
rrs accepted D32655: Pass down some IP level setsockopt()s into TCP stack(s).

This method is fine, I will withdraw my fab in favor if this :)

Oct 26 2021, 4:28 PM
rrs added a reviewer for D32671: tcp: Rack might retransmit forever.: glebius.
Oct 26 2021, 2:48 PM
rrs updated the diff for D32668: tcp: The rack stack can incorrectly have an overflow when calculating a burst delay..

While where at it, there is a duplicate srtt that should be just one var at the higher scope. Lets fix
that and make the needed casts when comparing to slot. That way we never can get a
confused compiler :)

Oct 26 2021, 2:37 PM
rrs requested review of D32671: tcp: Rack might retransmit forever..
Oct 26 2021, 1:46 PM
rrs requested review of D32668: tcp: The rack stack can incorrectly have an overflow when calculating a burst delay..
Oct 26 2021, 11:49 AM

Oct 25 2021

rrs added a comment to D32649: Fix some NewReno handlers in case of being called from other CC modules.

Overall I think this is a good fix for an ugly problem. I.e. that other
CC algo's use newreno. I think we should re-look at this and have the
"reused" functions put into the kernel as just non-static functions that
can be used by CC modules. Pass in beta etc...

Oct 25 2021, 8:57 PM
rrs accepted D32649: Fix some NewReno handlers in case of being called from other CC modules.
Oct 25 2021, 8:37 PM
rrs updated the diff for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

Michael has shown me how to add the right magic to git to get full context yeah!

Oct 25 2021, 6:40 PM
rrs added a reviewer for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path.: jtl.
Oct 25 2021, 2:47 PM
rrs updated the diff for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

Update the diff with offline comments from Jonathan

Oct 25 2021, 2:47 PM
rrs added a comment to D32585: SMR protection for inpcbs.

I don't see any hpts changes in this review.. which is what I was looking for. I suppose
you have it in another review then??

Oct 25 2021, 11:53 AM
rrs added a comment to D32373: tcp: Add hystart-plus to cc_newreno and rack..

I haven't finished reviewing yet but some initial thoughts in addition to my inline comments:

  • Nit: please submit diffs with full context as it makes reviewing easier
  • You need to address how enabling Hystart++ when using cc_newreno with the default FreeBSD stack will work
  • The implementation pessimises the common case where newreno will be used without ABE or Hystart++ which is probably undesirable. The module's lack of a cb_init() hook implementation was very deliberate to avoid such pessimisation and also as noted inline, assumptions are made elsewhere that cc_newreno can always be used as an algo of last resort so we cannot allow init to ever fail.
Oct 25 2021, 10:01 AM

Oct 24 2021

rrs updated the diff for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

Opps one typo to fix

Oct 24 2021, 11:26 AM
rrs updated the diff for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

I believe this is the more generic approach or at least closer to what Gleb wants. The names
might be wrong but I think it gets at what he has been saying..

Oct 24 2021, 11:21 AM
rrs added a comment to D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

I'm in the process of writing more generic patch to cover this. I would ask not to commit the intermediate version.

Note that already existing IPV6_USE_MIN_MTU handling in tcp_ctloutput is pretty much the same problem.

For IPv4 options and rack, there is also IP_TTL socket option that (I believe) needs same handling as IP_TOS.

Oct 24 2021, 6:26 AM
rrs updated the diff for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

This remix's things so that we create a t_fb function that can be called
by the codepoint changes i.e. the socket option if the transport wants
it and this updates the rack fsb.

Oct 24 2021, 6:25 AM

Oct 23 2021

rrs added a reviewer for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path.: glebius.
Oct 23 2021, 9:45 AM
rrs updated the diff for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

Let's not mess in ip or ip6 but intercept this in TCP. This also makes it so
we don't try this on UDP.

Oct 23 2021, 9:45 AM

Oct 22 2021

rrs updated the diff for D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..

Fix my DHCP/DSCP typos :)

Oct 22 2021, 1:35 PM
rrs requested review of D32600: tcp: DSCP Codepoint not being propagated in the rack fast path..
Oct 22 2021, 1:29 PM
rrs committed rG4e4c84f8d101: tcp: Add hystart-plus to cc_newreno and rack. (authored by rrs).
tcp: Add hystart-plus to cc_newreno and rack.
Oct 22 2021, 11:12 AM
rrs closed D32373: tcp: Add hystart-plus to cc_newreno and rack..
Oct 22 2021, 11:12 AM

Oct 20 2021

rrs updated the diff for D32373: tcp: Add hystart-plus to cc_newreno and rack..

Address Michael's comments.

Oct 20 2021, 6:22 AM
rrs added inline comments to D32373: tcp: Add hystart-plus to cc_newreno and rack..
Oct 20 2021, 6:18 AM

Oct 8 2021

rrs requested review of D32373: tcp: Add hystart-plus to cc_newreno and rack..
Oct 8 2021, 4:03 PM
rrs closed D32158: tcp: Make dsack stats be aware of TLPs.
Oct 8 2021, 3:59 PM

Oct 1 2021

rrs committed rGa36230f75e0a: tcp: Make dsack stats available in netstat and also make sure its aware of… (authored by rrs).
tcp: Make dsack stats available in netstat and also make sure its aware of…
Oct 1 2021, 2:37 PM
rrs updated the diff for D32158: tcp: Make dsack stats be aware of TLPs.

This refactors so that we end up with the data in netstat instead
so that it can be used by others besides NF. I also add the hooks
into tcp_sack so the counters are maintained for the base stack, though
we don't track TLPs (or do TLPs for that matter). BBR has the netstats
removed and now uses the proper api. Note that it does do TLP but
as of yet does not track what dsack bytes were caused by TLPs.

Oct 1 2021, 1:59 PM

Sep 27 2021

rrs added a reviewer for D32158: tcp: Make dsack stats be aware of TLPs: tuexen.
Sep 27 2021, 4:40 PM
rrs requested review of D32158: tcp: Make dsack stats be aware of TLPs.
Sep 27 2021, 4:40 PM

Sep 23 2021

rrs closed D32082: tcp: Rack compressed ack path updates the recv window too easily.
Sep 23 2021, 3:45 PM
rrs committed rG1ca931a540cb: tcp: Rack compressed ack path updates the recv window too easily (authored by rrs).
tcp: Rack compressed ack path updates the recv window too easily
Sep 23 2021, 3:45 PM
rrs requested review of D32082: tcp: Rack compressed ack path updates the recv window too easily.
Sep 23 2021, 3:16 PM
rrs committed rGfd69939e79a6: tcp: Two bugs in rack one of which can lead to a panic. (authored by rrs).
tcp: Two bugs in rack one of which can lead to a panic.
Sep 23 2021, 2:58 PM
rrs closed D32062: tcp: Two bugs in rack one of which can lead to a panic..
Sep 23 2021, 2:58 PM

Sep 22 2021

rrs requested review of D32062: tcp: Two bugs in rack one of which can lead to a panic..
Sep 22 2021, 7:19 PM

Sep 14 2021

rrs accepted D31959: tcp: avoid useless calculations in tcp_account_for_send() when len is 0.
Sep 14 2021, 3:55 PM

Aug 17 2021

rrs committed rG5baf32c97a1d: tcp: Add support for DSACK based reordering window to rack. (authored by rrs).
tcp: Add support for DSACK based reordering window to rack.
Aug 17 2021, 8:31 PM
rrs closed D31506: tcp: Add support for DSACK based reordering window to rack..
Aug 17 2021, 8:31 PM

Aug 11 2021

rrs requested review of D31506: tcp: Add support for DSACK based reordering window to rack..
Aug 11 2021, 10:46 AM

Jul 19 2021

rrs accepted D31212: Fix issue with RACK/BBR when VIMAGE is enabled.
Jul 19 2021, 6:26 PM

Jul 16 2021

rrs added a comment to D31194: tcp: When rack or bbr get a pullup failure in the common code, don't free the NULL mbuf..

Committed

Jul 16 2021, 6:02 PM
rrs committed rGdb4d2d7222ee: tcp: When rack or bbr get a pullup failure in the common code, don't free the… (authored by rrs).
tcp: When rack or bbr get a pullup failure in the common code, don't free the…
Jul 16 2021, 6:02 PM
rrs closed D31194: tcp: When rack or bbr get a pullup failure in the common code, don't free the NULL mbuf..
Jul 16 2021, 6:02 PM
rrs committed rG1d171e5ab962: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it… (authored by rrs).
tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it…
Jul 16 2021, 10:09 AM
rrs closed D31195: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it parses..
Jul 16 2021, 10:09 AM

Jul 15 2021

rrs updated the diff for D31195: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it parses..

Correct the bug in the calculation as well as subtracting out the outer
header size from the m_len on the second inner header parse.

Jul 15 2021, 6:29 PM
rrs added inline comments to D31195: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it parses..
Jul 15 2021, 6:22 PM
rrs updated the diff for D31195: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it parses..

Add Hans predict false around the hopefully never exercised returns

Jul 15 2021, 5:32 PM
rrs added a comment to D31195: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it parses..

We can add predict false here, since the majority of times we would
never go those paths. That makes sense, but we need to have the
checks. If we don't then at some point a driver writer that is not
as good as you will send in packets in an invalid way and we
won't catch it.

Jul 15 2021, 5:28 PM
rrs added a comment to D31195: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it parses..

Ok just got through testing this, and it works great.. no problems as I saw before ;-)

Jul 15 2021, 4:53 PM
rrs requested review of D31195: tcp: Lro needs to validate that it does not go beyond the end of the mbuf as it parses..
Jul 15 2021, 4:01 PM
rrs requested review of D31194: tcp: When rack or bbr get a pullup failure in the common code, don't free the NULL mbuf..
Jul 15 2021, 3:09 PM
rrs abandoned D31189: tcp: LRO epoch issue as well a potential pullup issue..
Jul 15 2021, 3:03 PM
rrs added a comment to D31189: tcp: LRO epoch issue as well a potential pullup issue..

Michael and I have chatted on the transport sync. I am going to remove
this review, and he is going to separate out the EPOCH review on its own.

Jul 15 2021, 3:03 PM
rrs added inline comments to D31189: tcp: LRO epoch issue as well a potential pullup issue..
Jul 15 2021, 3:02 PM
rrs added a comment to D31189: tcp: LRO epoch issue as well a potential pullup issue..

Drew
Here is the bug Michael did the epoch thing over

Jul 15 2021, 2:00 PM
rrs added inline comments to D31189: tcp: LRO epoch issue as well a potential pullup issue..
Jul 15 2021, 1:58 PM
rrs added inline comments to D31189: tcp: LRO epoch issue as well a potential pullup issue..
Jul 15 2021, 1:35 PM
rrs added a comment to D31189: tcp: LRO epoch issue as well a potential pullup issue..

A couple of bits..

Jul 15 2021, 1:27 PM
rrs updated the diff for D31189: tcp: LRO epoch issue as well a potential pullup issue..

another extra nl

Jul 15 2021, 11:24 AM
rrs updated the diff for D31189: tcp: LRO epoch issue as well a potential pullup issue..

Spacing issues

Jul 15 2021, 11:22 AM
rrs requested review of D31189: tcp: LRO epoch issue as well a potential pullup issue..
Jul 15 2021, 11:17 AM

Jul 13 2021

rrs closed D31155: tcp: TCP_LRO getting bad checksums and sending it in to TCP incorrectly..
Jul 13 2021, 4:47 PM
rrs committed rGca1a7e1021a9: tcp: TCP_LRO getting bad checksums and sending it in to TCP incorrectly. (authored by rrs).
tcp: TCP_LRO getting bad checksums and sending it in to TCP incorrectly.
Jul 13 2021, 4:47 PM
rrs updated the diff for D31155: tcp: TCP_LRO getting bad checksums and sending it in to TCP incorrectly..

Update with Hans and Michael's suggestions.

Jul 13 2021, 12:04 PM
rrs added inline comments to D31155: tcp: TCP_LRO getting bad checksums and sending it in to TCP incorrectly..
Jul 13 2021, 11:57 AM

Jul 12 2021

rrs updated the diff for D31155: tcp: TCP_LRO getting bad checksums and sending it in to TCP incorrectly..

Opps forgot the tcp_subr.c and tcp_var.h changes too

Jul 12 2021, 7:01 PM
rrs requested review of D31155: tcp: TCP_LRO getting bad checksums and sending it in to TCP incorrectly..
Jul 12 2021, 6:58 PM

Jul 8 2021

rrs closed D31107: tcp: Fix 32 bit platform breakage.
Jul 8 2021, 12:17 PM
rrs committed rG7312e4e5cfc8: tcp: Fix 32 bit platform breakage (authored by rrs).
tcp: Fix 32 bit platform breakage
Jul 8 2021, 12:17 PM
rrs updated the summary of D31107: tcp: Fix 32 bit platform breakage.
Jul 8 2021, 11:10 AM
rrs requested review of D31107: tcp: Fix 32 bit platform breakage.
Jul 8 2021, 11:09 AM

Jul 7 2021

rrs accepted D31094: tcp: fix build for rack/bbr with various combos of INET and INET6 defined.
Jul 7 2021, 4:23 PM
rrs closed D31083: tcp: HPTS performance enhancements..
Jul 7 2021, 12:43 PM
rrs committed rGd7955cc0ffdf: tcp: HPTS performance enhancements (authored by rrs).
tcp: HPTS performance enhancements
Jul 7 2021, 11:23 AM

Jul 6 2021

rrs committed rGe834f9a44acc: tcp: Address goodput and TLP edge cases. (authored by rrs).
tcp: Address goodput and TLP edge cases.
Jul 6 2021, 7:27 PM
rrs closed D31076: tcp: Address goodput and TLP edge cases..
Jul 6 2021, 7:27 PM
rrs requested review of D31083: tcp: HPTS performance enhancements..
Jul 6 2021, 7:22 PM
rrs updated the diff for D31076: tcp: Address goodput and TLP edge cases..

Opps forgot to include the include file ;)

Jul 6 2021, 2:34 PM
rrs requested review of D31076: tcp: Address goodput and TLP edge cases..
Jul 6 2021, 2:20 PM
rrs closed D30895: tcp: Preparation for allowing hardware TLS to be able to kick a tcp connection that is retransmitting too much out of hardware and back to software..
Jul 6 2021, 2:13 PM
rrs accepted D30908: ktls: automatically disable inline hw (aka ifnet) ktls offload when TCP rexmits are high.
Jul 6 2021, 1:04 PM

Jun 25 2021

rrs committed rG9e4d9e4c4d79: tcp: Preparation for allowing hardware TLS to be able to kick a tcp connection… (authored by rrs).
tcp: Preparation for allowing hardware TLS to be able to kick a tcp connection…
Jun 25 2021, 1:33 PM
rrs added reviewers for D30895: tcp: Preparation for allowing hardware TLS to be able to kick a tcp connection that is retransmitting too much out of hardware and back to software.: tuexen, gallatin.
Jun 25 2021, 11:57 AM
rrs requested review of D30895: tcp: Preparation for allowing hardware TLS to be able to kick a tcp connection that is retransmitting too much out of hardware and back to software..
Jun 25 2021, 11:56 AM

Jun 24 2021

rrs closed D30885: tcp: Rack not being very friendly with V6:4 socket and having a connection from V4.
Jun 24 2021, 6:45 PM
rrs committed rG66aec14a5391: tcp: Rack not being very friendly with V6:4 socket and having a connection from… (authored by rrs).
tcp: Rack not being very friendly with V6:4 socket and having a connection from…
Jun 24 2021, 6:44 PM
rrs closed D30727: tcp: Missing mfree in rack and bbr..
Jun 24 2021, 6:41 PM
rrs closed D30704: tcp: Mbuf leak while holding a socket buffer lock .
Jun 24 2021, 6:41 PM
rrs added a reviewer for D30885: tcp: Rack not being very friendly with V6:4 socket and having a connection from V4: tuexen.
Jun 24 2021, 6:13 PM
rrs requested review of D30885: tcp: Rack not being very friendly with V6:4 socket and having a connection from V4.
Jun 24 2021, 6:13 PM

Jun 17 2021

rrs accepted D29441: tcp: Add PRR cwnd reduction for non-SACK loss.
Jun 17 2021, 3:15 PM
rrs accepted D30740: net.inet.tcp.tolerate_missing_ts=1.

I wound not advise putting names in the man page.

Jun 17 2021, 3:04 PM

Jun 11 2021

rrs accepted D30723: Remove debug output in RACK.
Jun 11 2021, 6:13 PM