Page MenuHomeFreeBSD

rrs (Randall Stewart)
User

Projects

User Details

User Since
Jan 22 2015, 5:22 AM (318 w, 26 m)

Recent Activity

Thu, Feb 18

rrs requested changes to D28736: DO NOT MERGE: This seems to fix https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=253589.

This will cause a HUGE performance impact! sbspace() is called often and
twiddling the extra locks is not useful. You are just looking (reading) two
variables or even one variable in the sbused case. You do not need a lock for this!

Thu, Feb 18, 3:59 PM
rrs committed R10:e13e4fa6c4eb: fix Navdeeps LINT_NOINET error. (authored by rrs).
fix Navdeeps LINT_NOINET error.
Thu, Feb 18, 12:52 PM
rrs added a comment to D28761: tcp_lro_log: mark __unused to avoid unused function warning.

If we put the wrap
#ifdef TCPHPTS

Thu, Feb 18, 6:40 AM
rrs committed R10:0a4f851074a3: Fix another pesky missing #ifdef TCPHPTS (authored by rrs).
Fix another pesky missing #ifdef TCPHPTS
Thu, Feb 18, 6:28 AM

Wed, Feb 17

rrs committed R10:ab4fad4be144: Add ifdef TCPHPTS around build_ack_entry and do_bpf_and_csum to avoid (authored by rrs).
Add ifdef TCPHPTS around build_ack_entry and do_bpf_and_csum to avoid
Wed, Feb 17, 5:51 PM
rrs committed R10:69a34e8d0235: Update the LRO processing code so that we can support (authored by rrs).
Update the LRO processing code so that we can support
Wed, Feb 17, 3:41 PM
rrs closed D28374: Compressed ack enhancements to LRO.
Wed, Feb 17, 3:41 PM

Thu, Feb 4

rrs accepted D28374: Compressed ack enhancements to LRO.
Thu, Feb 4, 5:21 PM

Thu, Jan 28

rrs committed R10:1a714ff20419: This pulls over all the changes that are in the netflix (authored by rrs).
This pulls over all the changes that are in the netflix
Thu, Jan 28, 4:54 PM
rrs added a comment to D28357: Update the tcp_ratelimit code to finally work right :).

So Hans:

Thu, Jan 28, 12:04 PM

Wed, Jan 27

rrs committed R10:24a8f6d36996: When we are about to send down to the driver layer (authored by rrs).
When we are about to send down to the driver layer
Wed, Jan 27, 6:55 PM
rrs closed D28377: m_nextpkt pointer not NULL'ed before sending.
Wed, Jan 27, 6:55 PM
rrs requested review of D28377: m_nextpkt pointer not NULL'ed before sending.
Wed, Jan 27, 6:36 PM
rrs updated the diff for D28357: Update the tcp_ratelimit code to finally work right :).

Lets go ahead and not put the netflix_stats directive in front of the sysctl's.
Turns out we already had these defined if RATELIMIT is defined so lets
just add the two new ones.

Wed, Jan 27, 5:39 PM
rrs updated the diff for D28374: Compressed ack enhancements to LRO.

Looks like I forgot to make sure we get the counters properly initialized in tcp_subr.c opps.

Wed, Jan 27, 5:31 PM
rrs requested review of D28374: Compressed ack enhancements to LRO.
Wed, Jan 27, 5:18 PM
rrs updated the diff for D28357: Update the tcp_ratelimit code to finally work right :).

Take out the mlx bits so that Hans can commit them separately

Wed, Jan 27, 4:58 PM
rrs added a comment to D25215: Remove TCP option alignment padding.

Why are you removing the NOP Pad chunks? IP options *must* be a multiple of 4 which is what the pad does, taking
it out is going to make the options incorrect I would think??

Wed, Jan 27, 3:47 PM
rrs added a comment to D28357: Update the tcp_ratelimit code to finally work right :).

Thats fine if you want to do it that way but note that this patch adds the new
function if_next_snd_tag() to if_var.h. You won't be able to add that function unless
we commit this one (without the mlx changes) first.

Wed, Jan 27, 3:45 PM

Tue, Jan 26

rrs requested review of D28357: Update the tcp_ratelimit code to finally work right :).
Tue, Jan 26, 5:36 PM

Jan 14 2021

rrs accepted D28114: Address panic with PRR due to missed initialization of recover_fs.
Jan 14 2021, 2:17 PM

Dec 13 2020

rrs committed R9:2ec2bfae9a8b: Adds my key :-D (authored by rrs).
Adds my key :-D
Dec 13 2020, 6:09 PM
rrs committed R9:33c21e682a34: This commit finished the addition of the "first commit" (authored by rrs).
This commit finished the addition of the "first commit"
Dec 13 2020, 6:09 PM
rrs committed R9:30fd504f52c4: Ok, here is my missing commit... which I forgot :-0 (authored by rrs).
Ok, here is my missing commit... which I forgot :-0
Dec 13 2020, 6:08 PM
rrs committed R9:a9d00ec694e6: Approved by: gnn (authored by rrs).
Approved by: gnn
Dec 13 2020, 6:06 PM
rrs committed R9:b701c4c83aa0: __FreeBSDversion 700028 (authored by rrs).
__FreeBSDversion 700028
Dec 13 2020, 6:06 PM

Dec 3 2020

rrs accepted D27464: Fix hung TCP sessions on shutdown.

Not exactly the best, but I think it fixes the immediate problem. From a long term perspective we may want to
look for a better solution :)

Dec 3 2020, 8:03 PM

Nov 17 2020

rrs accepted D27188: LACP: When suppressing distributing, return ENOBUFS rather than ENETDOWN to preserve TCP conns.
Nov 17 2020, 8:56 PM

Nov 9 2020

rrs accepted D27148: Improve RFC 7323 Compliance.
Nov 9 2020, 7:17 PM

Oct 22 2020

rrs accepted D26808: TCP Cubic: improve reaction to (and rollback from) RTO.
Oct 22 2020, 2:30 PM
rrs accepted D26807: move cwnd and ssthresh updates into individual congestion control module.
Oct 22 2020, 2:27 PM
rrs accepted D24237: Fix erroneous "DSACK" during loss recovery.
Oct 22 2020, 2:22 PM

Oct 8 2020

rrs accepted D26409: Add IP(V6)_VLAN_PCP to adjust 802.1 priority per-flow..
Oct 8 2020, 2:42 PM

Sep 25 2020

rrs requested changes to D26409: Add IP(V6)_VLAN_PCP to adjust 802.1 priority per-flow..
Sep 25 2020, 7:19 PM

Sep 9 2020

rrs committed rS365501: So it turns out that syzkaller hit another crash. It has to do with switching.
So it turns out that syzkaller hit another crash. It has to do with switching
Sep 9 2020, 11:12 AM
rrs closed D26269: Another syzkaller crash.
Sep 9 2020, 11:12 AM

Sep 1 2020

rrs updated the diff for D26269: Another syzkaller crash.

Fix the typo's in the comment that Michael pointed out!

Sep 1 2020, 12:14 PM
rrs updated the summary of D26269: Another syzkaller crash.
Sep 1 2020, 12:10 PM
rrs requested review of D26269: Another syzkaller crash.
Sep 1 2020, 12:09 PM

Aug 25 2020

rrs accepted D25085: netinet6: In tcp_lro_rx_ipv6(), check for IPv6 flow ID..
Aug 25 2020, 1:32 PM · network

Aug 24 2020

rrs accepted D26120: Don't apply L in congestion avoidance.
Aug 24 2020, 6:08 PM

Aug 21 2020

rrs closed D24902: Time to First byte in and out.
Aug 21 2020, 2:59 AM

Aug 18 2020

rrs accepted D26103: TCP: remove special treatment for hardware (ifnet) TLS.
Aug 18 2020, 3:31 PM

Aug 13 2020

rrs accepted D24237: Fix erroneous "DSACK" during loss recovery.
Aug 13 2020, 4:44 PM

Aug 10 2020

rrs accepted D25909: Fix handling of TTL/Hop Limit and TOS/Traffic Class in the TCP SYN Cache.
Aug 10 2020, 6:34 PM

Jul 31 2020

rrs committed rS363725: The recent changes to move the ref count increment.
The recent changes to move the ref count increment
Jul 31 2020, 10:04 AM
rrs closed D25903: Leaks of inpcb.
Jul 31 2020, 10:04 AM
rrs added a comment to D25903: Leaks of inpcb.

First of all the tricky thing to remember about inp locks is that

Jul 31 2020, 10:00 AM
rrs requested review of D25903: Leaks of inpcb.
Jul 31 2020, 6:03 AM

Jul 16 2020

rrs accepted D25401: (Re)allow 0.0.0.0 to be used as a destination address in connect() for TCP.
Jul 16 2020, 4:33 PM

Jul 8 2020

rrs accepted D25584: Install extra tcp stack headers.
Jul 8 2020, 8:01 PM

Jun 23 2020

rrs closed D24736: Remove un-needed NF copy args.
Jun 23 2020, 3:26 AM
rrs updated the diff for D24826: Hung sessions.

Opps used the wrong goto label...

Jun 23 2020, 3:23 AM
rrs updated the diff for D24826: Hung sessions.

Looks like BBR has the same issue, and we need to strip trailing emacs presents too :)

Jun 23 2020, 3:18 AM
rrs requested review of D24736: Remove un-needed NF copy args.
Jun 23 2020, 3:17 AM

Jun 18 2020

rrs accepted D25016: Make After-Idle congestion control work correctly for transactional sessions..
Jun 18 2020, 5:27 PM

Jun 16 2020

rrs committed rS362234: iSo in doing final checks on OCA firmware with all the latest tweaks the dup….
iSo in doing final checks on OCA firmware with all the latest tweaks the dup…
Jun 16 2020, 6:17 PM
rrs closed D25298: More rwnd issues with rack and bbr.
Jun 16 2020, 6:17 PM
rrs requested review of D25298: More rwnd issues with rack and bbr.
Jun 16 2020, 4:28 PM
rrs closed D25294: Missing pieces to make rack properly respond to 3 dup-acks.
Jun 16 2020, 12:26 PM
rrs committed rS362225: So it turns out rack has a shortcoming in dup-ack counting. It counts the….
So it turns out rack has a shortcoming in dup-ack counting. It counts the…
Jun 16 2020, 12:26 PM
rrs closed D25247: Incorrect rwnd needs update decision.
Jun 16 2020, 12:18 PM
rrs requested review of D25294: Missing pieces to make rack properly respond to 3 dup-acks.
Jun 16 2020, 11:48 AM

Jun 12 2020

rrs committed rS362113: So it turns out with the right window scaling you can get the code in all….
So it turns out with the right window scaling you can get the code in all…
Jun 12 2020, 7:56 PM
rrs requested review of D25247: Incorrect rwnd needs update decision.
Jun 12 2020, 7:31 PM

Jun 8 2020

rrs committed rS361926: An important statistic in determining if a server process (or client) is being….
An important statistic in determining if a server process (or client) is being…
Jun 8 2020, 11:48 AM
rrs added a comment to D24902: Time to First byte in and out.

Wouldn't sbintime_t be easier and cheaper to work with?

Jun 8 2020, 11:36 AM

Jun 4 2020

rrs requested changes to D25016: Make After-Idle congestion control work correctly for transactional sessions..

I don't mind you tracking t_rtseq here, but please do not change the idle reduction in rack. I will go dig
up the right variable and change this to use that. Your are welcome to maintain t_rtseq but please don't change
the rack behavior here.

Jun 4 2020, 5:34 PM

Jun 3 2020

rrs closed D24852: Invalid addresses in sending or connecting.
Jun 3 2020, 2:16 PM
rrs committed rS361752: We should never allow either the broadcast or IN_ADDR_ANY to be.
We should never allow either the broadcast or IN_ADDR_ANY to be
Jun 3 2020, 2:16 PM
rrs closed D25000: Skyzaller crash found by Michael Tuexen.
Jun 3 2020, 2:13 PM
rrs committed rS361751: This fixes a couple of skyzaller crashes. Most.
This fixes a couple of skyzaller crashes. Most
Jun 3 2020, 2:07 PM
rrs accepted D25115: Restrict enabling TCP-FASTOPEN to end-points in CLOSED or LISTEN state.

It would not be EAGAIN/EBUSY since you can't do it again and expect it to succeed. I think
EINVAL is ok here.

Jun 3 2020, 11:21 AM

May 29 2020

rrs updated the diff for D25000: Skyzaller crash found by Michael Tuexen.

Ok turns out skyzaller has found yet another crash having to do with TFO for all three stacks. Lets update
the patch to fix that too.

May 29 2020, 1:36 PM

May 25 2020

rrs requested review of D25000: Skyzaller crash found by Michael Tuexen.
May 25 2020, 1:44 PM
rrs added a comment to D24902: Time to First byte in and out.

Has discussed on the transport call, I will be updating these to have a uint64_t and a nano-second basis for the values (including t_starttime I think since this
is what you compare these against) :)

May 25 2020, 11:39 AM

May 21 2020

rrs added inline comments to D24237: Fix erroneous "DSACK" during loss recovery.
May 21 2020, 3:37 PM
rrs accepted D24817: DCTCP: update alpha only once after loss recovery..
May 21 2020, 2:20 PM
rrs accepted D23364: Send CWR only on new data, as per sec. 6.1.2 of RFC3168.
May 21 2020, 2:16 PM

May 20 2020

rrs added a comment to D24902: Time to First byte in and out.

I don't think that holding just the delta is what you really want. You want to be able
to tell when the first byte came in and the first byte went out. The delta only gives
one a partial view of what happened. And we are not living in 1984 where 64k of
memory is a lot. Four bytes to get more detailed information is I think worth it.

May 20 2020, 11:22 AM

May 19 2020

rrs updated the diff for D24852: Invalid addresses in sending or connecting.

Note that we don't want to commit these for a while since skyzaller uses these addresses but it
is something we should do.

May 19 2020, 12:01 PM
rrs added inline comments to D24852: Invalid addresses in sending or connecting.
May 19 2020, 11:55 AM
rrs updated the diff for D24852: Invalid addresses in sending or connecting.

Hmm I somehow got the wrong patch here.

May 19 2020, 11:54 AM

May 18 2020

rrs requested review of D24902: Time to First byte in and out.
May 18 2020, 8:00 PM
rrs accepted D24781: Allow TCP to reuse local port with different destinations.

Yeah the spelling error check is rather dumb :)

May 18 2020, 6:29 AM

May 15 2020

rrs closed D24852: Invalid addresses in sending or connecting.
May 15 2020, 2:00 PM
rrs committed rS361080: This fixes several skyzaller issues found with the.
This fixes several skyzaller issues found with the
May 15 2020, 2:00 PM
rrs accepted D24831: Allow only IPv4 addresses in sendto() for TCP on AF_INET sockets.
May 15 2020, 1:57 PM
rrs requested review of D24852: Invalid addresses in sending or connecting.
May 15 2020, 12:41 PM
rrs updated the diff for D24826: Hung sessions.

We should not ever do GP measurements until we are established

May 15 2020, 12:39 PM
rrs updated the diff for D24826: Hung sessions.

One little other niggle that I noticed, we should not do a goto again if the SYN bit is set ...

May 15 2020, 12:01 PM
rrs updated the diff for D24826: Hung sessions.

And even yet another issue thanks to skyzaller in bbr.. the rc_lost when doing TCP-FO was incorrect in rxt timers.

May 15 2020, 11:13 AM
rrs requested review of D24826: Hung sessions.
May 15 2020, 5:24 AM
rrs updated the diff for D24826: Hung sessions.

Let's also fix the accounting panic that skyzaller is finding here too. When we are in a front
state and have a SYN outstanding that should count in the ctf_outstanding size. Also if
we have sent a FIN same thing.

May 15 2020, 4:49 AM

May 14 2020

rrs added inline comments to D24830: sendfile_test: Finish to implement linear probing to find a free port .
May 14 2020, 11:17 AM

May 12 2020

rrs accepted D24821: IPv6: Fix a panic in the nd6 code with unmapped mbufs..
May 12 2020, 1:51 PM
rrs accepted D24822: IPv6: sync IP_NO_SND_TAG_RL support from IPv4.
May 12 2020, 1:49 PM

May 7 2020

rrs closed D24747: Responding properly to SYN-ACK.
May 7 2020, 8:30 PM
rrs committed rS360798: When in the SYN-SENT state bbr and rack will not properly send an ACK but….
When in the SYN-SENT state bbr and rack will not properly send an ACK but…
May 7 2020, 8:30 PM
rrs added inline comments to D23371: Retain only mutually supported TCP options after simultaneous SYN.
May 7 2020, 3:01 PM