Page MenuHomeFreeBSD

tuexen (Michael Tuexen)
User

Projects

User Details

User Since
Feb 4 2016, 4:45 PM (280 w, 6 d)

Recent Activity

Sat, Jun 19

tuexen added inline comments to D30740: net.inet.tcp.tolerate_missing_ts=1.
Sat, Jun 19, 10:38 AM
tuexen updated the diff for D30740: net.inet.tcp.tolerate_missing_ts=1.
Sat, Jun 19, 10:17 AM

Mon, Jun 14

tuexen committed R10:d0eaf95edcaf: tcp: Missing mfree in rack and bbr (authored by rrs).
tcp: Missing mfree in rack and bbr
Mon, Jun 14, 9:03 PM
tuexen committed R10:8ecbecdcfdb0: tcp: Mbuf leak while holding a socket buffer lock. (authored by rrs).
tcp: Mbuf leak while holding a socket buffer lock.
Mon, Jun 14, 9:00 PM
tuexen committed R10:2071c3fb0dcc: tcp: LRO timestamps have lost their previous precision (authored by rrs).
tcp: LRO timestamps have lost their previous precision
Mon, Jun 14, 8:52 PM
tuexen committed R10:82f75079f188: tcp: fix two bugs in new reno (authored by tuexen).
tcp: fix two bugs in new reno
Mon, Jun 14, 9:02 AM
tuexen committed R10:fce16041a86c: tcp: remove debug output from RACK (authored by tuexen).
tcp: remove debug output from RACK
Mon, Jun 14, 9:01 AM
tuexen committed R10:7a2030a10686: tcp: fix compilation of IPv4-only builds (authored by tuexen).
tcp: fix compilation of IPv4-only builds
Mon, Jun 14, 9:00 AM

Sat, Jun 12

tuexen added inline comments to D30740: net.inet.tcp.tolerate_missing_ts=1.
Sat, Jun 12, 2:44 PM

Fri, Jun 11

tuexen updated the diff for D30740: net.inet.tcp.tolerate_missing_ts=1.

Change man-page.

Fri, Jun 11, 7:51 PM
tuexen requested review of D30740: net.inet.tcp.tolerate_missing_ts=1.
Fri, Jun 11, 6:30 PM
tuexen closed D30723: Remove debug output in RACK.
Fri, Jun 11, 6:25 PM
tuexen committed R10:f1536bb53898: tcp: remove debug output from RACK (authored by tuexen).
tcp: remove debug output from RACK
Fri, Jun 11, 6:24 PM
tuexen accepted D30727: tcp: Missing mfree in rack and bbr..
Fri, Jun 11, 3:17 PM
tuexen requested review of D30723: Remove debug output in RACK.
Fri, Jun 11, 1:49 PM
tuexen closed D30721: New Reno fixes.
Fri, Jun 11, 1:42 PM
tuexen added a comment to D30721: New Reno fixes.

This was committed as fa3746be4203fc9a3414afb21d964eec8bad74f8

Fri, Jun 11, 1:41 PM
tuexen committed R10:fa3746be4203: tcp: fix two bugs in new reno (authored by tuexen).
tcp: fix two bugs in new reno
Fri, Jun 11, 1:41 PM
tuexen added a comment to D30721: New Reno fixes.
In D30721#690495, @rrs wrote:

One question, were you going to kill the printf?

Fri, Jun 11, 1:35 PM
tuexen added inline comments to D30721: New Reno fixes.
Fri, Jun 11, 11:47 AM
tuexen updated the diff for D30721: New Reno fixes.

Address comments.

Fri, Jun 11, 11:26 AM
tuexen added a comment to D30721: New Reno fixes.
In D30721#690480, @rrs wrote:

And I think Richard is right here, the conditions on setting beta_ecn should be removed as well.

I did not do Abe, so maybe this is a question for Grenville.. I am guessing that he
was wanting always to use the beta_ecn if Abe is in play.. but I dont know.

I wanted to have rack use

beta_ecn

When we do ECN and are pacing
and beta (set to other than 50) when we
are pacing.

Fri, Jun 11, 11:21 AM
tuexen added inline comments to D30721: New Reno fixes.
Fri, Jun 11, 9:38 AM
tuexen requested review of D30721: New Reno fixes.
Fri, Jun 11, 8:39 AM
tuexen committed R10:224cf7b35b9b: tcp: fix compilation of IPv4-only builds (authored by tuexen).
tcp: fix compilation of IPv4-only builds
Fri, Jun 11, 8:07 AM

Wed, Jun 9

tuexen added a comment to D30704: tcp: Mbuf leak while holding a socket buffer lock .

Are the pure whitespace changes intended?

Wed, Jun 9, 6:54 PM
tuexen committed R10:fc53b7269fed: tcp: A better fix for the previously attempted fix of the ack-war issue with… (authored by rrs).
tcp: A better fix for the previously attempted fix of the ack-war issue with…
Wed, Jun 9, 12:21 AM
tuexen committed R10:eb91abb4ba3d: tcp: When we have an out-of-order FIN we do want to strip off the FIN bit. (authored by rrs).
tcp: When we have an out-of-order FIN we do want to strip off the FIN bit.
Wed, Jun 9, 12:20 AM
tuexen committed R10:362f95f52897: tcp: Add a socket option to rack so we can test various changes to the slop… (authored by rrs).
tcp: Add a socket option to rack so we can test various changes to the slop…
Wed, Jun 9, 12:19 AM
tuexen committed R10:12e181b67273: tcp: Fix bugs related to the PUSH bit and rack and an ack war (authored by rrs).
tcp: Fix bugs related to the PUSH bit and rack and an ack war
Wed, Jun 9, 12:15 AM
tuexen committed R10:e99fa57b981e: tcp: Fix an issue with the PUSH bit as well as fill in the missing mtu change… (authored by rrs).
tcp: Fix an issue with the PUSH bit as well as fill in the missing mtu change…
Wed, Jun 9, 12:14 AM
tuexen committed R10:6264ff9bd9f0: tcp: Handle stack switch while processing socket options (authored by tuexen).
tcp: Handle stack switch while processing socket options
Wed, Jun 9, 12:11 AM
tuexen committed R10:56aeedd2fda4: tcp: Fix sending of TCP segments with IP level options (authored by tuexen).
tcp: Fix sending of TCP segments with IP level options
Wed, Jun 9, 12:09 AM
tuexen committed R10:3a3bba7df566: tcp: Incorrect KASSERT causes a panic in rack (authored by rrs).
tcp: Incorrect KASSERT causes a panic in rack
Wed, Jun 9, 12:07 AM
tuexen committed R10:821543e7834b: mod_cc: cross reference CC modules consistently (authored by tuexen).
mod_cc: cross reference CC modules consistently
Wed, Jun 9, 12:05 AM
tuexen committed R10:6170c93c03fa: tcp rack: improve initialisation of retransmit timeout (authored by tuexen).
tcp rack: improve initialisation of retransmit timeout
Wed, Jun 9, 12:03 AM
tuexen committed R10:ecfc25f05bf4: tcp: In rack, we must only convert restored rtt when the hostcache does restore… (authored by rrs).
tcp: In rack, we must only convert restored rtt when the hostcache does restore…
Wed, Jun 9, 12:03 AM
tuexen committed R10:87cf5dcc3335: tcp:Host cache and rack ending up with incorrect values. (authored by rrs).
tcp:Host cache and rack ending up with incorrect values.
Wed, Jun 9, 12:02 AM
tuexen committed R10:4651125ac6df: This takes Warners suggested approach to making it so that (authored by rrs).
This takes Warners suggested approach to making it so that
Wed, Jun 9, 12:01 AM
tuexen committed R10:f5d0badc7007: Fix a UDP tunneling issue with rack. Basically there are two (authored by rrs).
Fix a UDP tunneling issue with rack. Basically there are two
Wed, Jun 9, 12:00 AM

Tue, Jun 8

tuexen committed R10:a7619988ebc2: This brings into sync FreeBSD with the netflix versions of rack and bbr. (authored by rrs).
This brings into sync FreeBSD with the netflix versions of rack and bbr.
Tue, Jun 8, 11:58 PM
tuexen committed R10:39b87f051506: Fix LINT kernel builds after 1a714ff20419 . (authored by hselasky).
Fix LINT kernel builds after 1a714ff20419 .
Tue, Jun 8, 11:40 PM
tuexen accepted D30695: tcp: LRO timestamps have lost their previous precision.
Tue, Jun 8, 5:56 PM

Mon, Jun 7

tuexen committed R10:68d6663afbae: 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
Mon, Jun 7, 11:38 PM
tuexen committed R10:95b7e4e0febd: tcp: fix man page (authored by tuexen).
tcp: fix man page
Mon, Jun 7, 9:03 AM
tuexen committed R10:fa50e98328b4: mend (authored by tuexen).
mend
Mon, Jun 7, 9:02 AM
tuexen committed R10:1ceefa271e57: sctp: fix another locking bug in COOKIE handling (authored by tuexen).
sctp: fix another locking bug in COOKIE handling
Mon, Jun 7, 9:02 AM

Sun, Jun 6

tuexen committed R10:8421c612ee01: sctp: fix locking in case of error handling during a restart (authored by tuexen).
sctp: fix locking in case of error handling during a restart
Sun, Jun 6, 11:48 PM
tuexen committed R10:399798c97b03: sctp: improve consistency when handling chunks of wrong size (authored by tuexen).
sctp: improve consistency when handling chunks of wrong size
Sun, Jun 6, 11:46 PM
tuexen committed R10:a2fdddf83e8c: sctp: drop packet with SHUTDOWN-ACK chunks with wrong vtags (authored by tuexen).
sctp: drop packet with SHUTDOWN-ACK chunks with wrong vtags
Sun, Jun 6, 11:44 PM
tuexen committed R10:50948b57070f: sctp: improve address list scanning (authored by tuexen).
sctp: improve address list scanning
Sun, Jun 6, 11:43 PM
tuexen committed R10:405bee30bc93: sctp: improve restart handling (authored by tuexen).
sctp: improve restart handling
Sun, Jun 6, 11:41 PM
tuexen committed R10:bbcea8510964: Fix build after 5f2e1835054ee84f2e68ebc890d92716a91775b7. (authored by mav).
Fix build after 5f2e1835054ee84f2e68ebc890d92716a91775b7.
Sun, Jun 6, 11:40 PM
tuexen committed R10:7f1b88c36e06: sctp: improve error handling in INIT/INIT-ACK processing (authored by tuexen).
sctp: improve error handling in INIT/INIT-ACK processing
Sun, Jun 6, 11:39 PM
tuexen committed R10:1f663bc4baeb: sctp: update the vtag for INIT and INIT-ACK chunks (authored by tuexen).
sctp: update the vtag for INIT and INIT-ACK chunks
Sun, Jun 6, 11:36 PM
tuexen committed R10:4e71474ed97a: sctp: fix SCTP_PEER_ADDR_PARAMS socket option (authored by tuexen).
sctp: fix SCTP_PEER_ADDR_PARAMS socket option
Sun, Jun 6, 11:34 PM
tuexen committed R10:6b1dfc549caf: sctp: use RTO.Initial of 1 second as specified in RFC 4960bis (authored by tuexen).
sctp: use RTO.Initial of 1 second as specified in RFC 4960bis
Sun, Jun 6, 11:32 PM
tuexen committed R10:0cbc89fe44fd: sctp: improve consistency in handling chunks with wrong size (authored by tuexen).
sctp: improve consistency in handling chunks with wrong size
Sun, Jun 6, 11:31 PM
tuexen committed R10:54be3bf824e8: sctp: cleanup verification of INIT and INIT-ACK chunks (authored by tuexen).
sctp: cleanup verification of INIT and INIT-ACK chunks
Sun, Jun 6, 11:30 PM
tuexen committed R10:f7b539eb6d83: sctp: improve handling of illegal packets containing INIT chunks (authored by tuexen).
sctp: improve handling of illegal packets containing INIT chunks
Sun, Jun 6, 11:30 PM
tuexen committed R10:57e0431da113: sctp: small cleanup, no functional change (authored by tuexen).
sctp: small cleanup, no functional change
Sun, Jun 6, 11:27 PM
tuexen committed R10:abcf615834dd: sctp: fix handling of RTO.initial of 1 ms (authored by tuexen).
sctp: fix handling of RTO.initial of 1 ms
Sun, Jun 6, 11:25 PM
tuexen committed R10:15077bc4b91f: tcp: improve handling of SYN segments in SYN-SENT state (authored by tuexen).
tcp: improve handling of SYN segments in SYN-SENT state
Sun, Jun 6, 11:23 PM

Thu, Jun 3

tuexen accepted D30627: tcp: a more proper fix for the previous ack-wark.
Thu, Jun 3, 10:25 PM

Mon, May 31

tuexen added a comment to D30539: tcp, udp: Permit binding with AF_UNSPEC if the address is INADDR_ANY.

Hi Mark,

I'd personally like the approach D29519 . Is it desirable to have a dedicated sysctl variable to flag the compatible mode? The default value could be set to true.

I'm not sure. This is really a quite minor modification to D29519, I'd personally prefer not to add yet another sysctl for it. Maybe we should make this behaviour conditional on COMPAT_FREEBSD13 instead.

Mon, May 31, 12:53 PM

Fri, May 28

tuexen accepted D30539: tcp, udp: Permit binding with AF_UNSPEC if the address is INADDR_ANY.

Hmm. What about the corresponding case for IPv6 sockets? ttcp would still fail, since it also provides a sin_len of 0. I was expecting some fallout...

Do you know of any applications that depend on that? ttcp is IPv4-only.

Fri, May 28, 11:48 PM
tuexen added a comment to D30539: tcp, udp: Permit binding with AF_UNSPEC if the address is INADDR_ANY.

Hmm. What about the corresponding case for IPv6 sockets? ttcp would still fail, since it also provides a sin_len of 0. I was expecting some fallout...

Fri, May 28, 11:25 PM

Thu, May 27

tuexen accepted D30497: tcp: When we have an out-of-order FIN we do want to strip off the FIN bit..
Thu, May 27, 2:46 PM

Wed, May 26

tuexen accepted D30470: tcp: Use local CC data only in correct context.
Wed, May 26, 4:22 PM

Tue, May 25

tuexen accepted D30451: Bugs and problems even in the base stack..
Tue, May 25, 4:05 PM
tuexen updated the summary of D30451: Bugs and problems even in the base stack..
Tue, May 25, 3:37 PM

May 24 2021

tuexen closed D30395: Handle stack switch while processing stack specific socket options.
May 24 2021, 7:50 PM
tuexen accepted D30395: Handle stack switch while processing stack specific socket options.

Committed in https://cgit.FreeBSD.org/src/commit/?id=500eb6dd80404ea512e31a8f795c73cb802c9c64.

May 24 2021, 7:50 PM
tuexen closed D30358: Fix sending of TCP segments when IP level options are specified.

Commited in https://cgit.FreeBSD.org/src/commit/?id=500eb6dd80404ea512e31a8f795c73cb802c9c64.

May 24 2021, 7:49 PM
tuexen closed D30402: Fix a RACK socket buffer lock issue.
May 24 2021, 7:10 PM
tuexen committed R10:9bbd1a8fcb13: tcp: fix a RACK socket buffer lock issue (authored by tuexen).
tcp: fix a RACK socket buffer lock issue
May 24 2021, 7:08 PM
tuexen accepted D30413: clone rsm needs to do the right thing with the PUSH bit.
May 24 2021, 6:41 PM
tuexen accepted D30249: Create a way to measure the impact of changing timer_slop.
May 24 2021, 1:30 PM

May 23 2021

tuexen accepted D30407: tcp: Need to take network epoch when calling tfb_tcp_fb_fini.
May 23 2021, 8:01 PM
tuexen added a reviewer for D30402: Fix a RACK socket buffer lock issue: gallatin.
May 23 2021, 6:36 PM
tuexen requested review of D30402: Fix a RACK socket buffer lock issue.
May 23 2021, 12:58 PM

May 22 2021

tuexen committed R10:8923ce630492: tcp: Handle stack switch while processing socket options (authored by tuexen).
tcp: Handle stack switch while processing socket options
May 22 2021, 12:40 PM

May 21 2021

tuexen requested review of D30395: Handle stack switch while processing stack specific socket options.
May 21 2021, 10:50 PM
tuexen accepted D30390: rack: honor prior socket buffer lock when doing the upcall.
May 21 2021, 9:54 PM
tuexen added inline comments to D30358: Fix sending of TCP segments when IP level options are specified.
May 21 2021, 8:35 AM
tuexen committed R10:500eb6dd8040: tcp: Fix sending of TCP segments with IP level options (authored by tuexen).
tcp: Fix sending of TCP segments with IP level options
May 21 2021, 8:32 AM

May 20 2021

tuexen accepted D30349: socket: Fix error handling for pru_send(PRUS_NOTREADY).
May 20 2021, 10:13 PM
tuexen accepted D29690: [tcp] Keep socket buffer locked until upcall.
May 20 2021, 2:16 PM
tuexen added a comment to D30358: Fix sending of TCP segments when IP level options are specified.

Where are the ip options accounted for, then? Is the mbuf already adjusted to no longer contain ipoptions by that time (not familiar with IP level processing).

May 20 2021, 11:03 AM
tuexen requested review of D30358: Fix sending of TCP segments when IP level options are specified.
May 20 2021, 10:42 AM

May 19 2021

tuexen accepted D30349: socket: Fix error handling for pru_send(PRUS_NOTREADY).
May 19 2021, 11:42 PM
tuexen accepted D30249: Create a way to measure the impact of changing timer_slop.
May 19 2021, 7:26 AM

May 13 2021

tuexen accepted D30249: Create a way to measure the impact of changing timer_slop.
May 13 2021, 6:41 PM
tuexen accepted D30241: syzcaller found an interesting Invariant crash.
May 13 2021, 11:04 AM
tuexen retitled D30241: syzcaller found an interesting Invariant crash from Skyzall found an interesting Invariant crash to syzcaller found an interesting Invariant crash.
May 13 2021, 11:04 AM
tuexen committed R10:8ea5eeb913b9: mod_cc: cross reference CC modules consistently (authored by tuexen).
mod_cc: cross reference CC modules consistently
May 13 2021, 10:16 AM
tuexen closed D30240: Reference CC modules consistently.
May 13 2021, 10:16 AM
tuexen requested review of D30240: Reference CC modules consistently.
May 13 2021, 9:08 AM

May 12 2021

tuexen committed R10:eec6aed5b8c8: sctp: fix another locking bug in COOKIE handling (authored by tuexen).
sctp: fix another locking bug in COOKIE handling
May 12 2021, 9:16 PM
tuexen closed D30230: Fix retransmission timer initialisation for TCP RACK.
May 12 2021, 4:13 PM