Page MenuHomeFreeBSD

rrs (Randall Stewart)
User

Projects

User Details

User Since
Jan 22 2015, 5:22 AM (246 w, 4 d)

Recent Activity

Today

rrs committed rS353490: if_hw_tsomaxsegsize needs to be initialized to zero, just.
if_hw_tsomaxsegsize needs to be initialized to zero, just
Mon, Oct 14, 1:10 PM

Sun, Oct 6

rrs committed rS353156: Brad Davis identified a problem with the new LRO code, VLAN's.
Brad Davis identified a problem with the new LRO code, VLAN's
Sun, Oct 6, 10:29 PM

Sat, Sep 28

rrs added a comment to D21669: mtod macro: Drop the type argument.

Well I am going to voice a different view. I do *not* see this as a nice cleanup. I see it
has a lessening of information.

Sat, Sep 28, 9:27 AM
rrs accepted D21825: Fix locking for some RACK specific socket options.
Sat, Sep 28, 9:21 AM

Tue, Sep 24

rrs committed rS352661: lets put (void) in a couple of functions to keep older platforms that.
lets put (void) in a couple of functions to keep older platforms that
Tue, Sep 24, 8:36 PM
rrs committed rS352660: don't call in_ratelmit detach when RATELIMIT is not.
don't call in_ratelmit detach when RATELIMIT is not
Tue, Sep 24, 8:12 PM
rrs committed rS352659: Fix the ifdefs in tcp_ratelimit.h. They were reversed so.
Fix the ifdefs in tcp_ratelimit.h. They were reversed so
Tue, Sep 24, 8:04 PM
rrs committed rS352657: This commit adds BBR (Bottleneck Bandwidth and RTT) congestion control. This.
This commit adds BBR (Bottleneck Bandwidth and RTT) congestion control. This
Tue, Sep 24, 6:18 PM

Mon, Sep 23

rrs accepted D21582: Ok after several years here is BBR..
Mon, Sep 23, 8:41 PM

Fri, Sep 20

rrs accepted D21644: During SYN floods, fallback exclusively to SYN cookies for a small period.
Fri, Sep 20, 1:15 PM

Tue, Sep 17

rrs updated the diff for D21582: Ok after several years here is BBR..

This updates the patch set, the ktls.h information was incorrect. I also add the fixes
that Michael is doing on the Rack side to BBR as well (check for invalid length in non-tso case).

Tue, Sep 17, 2:47 PM
rrs accepted D21527: Add SIOCGIFDOWNREASON.
Tue, Sep 17, 12:59 PM
rrs accepted D21644: During SYN floods, fallback exclusively to SYN cookies for a small period.
Tue, Sep 17, 12:55 PM
rrs accepted D21666: Let the RACK stack take IP options into account when computing the space for user data in TCP segments.
Tue, Sep 17, 12:48 PM

Sep 11 2019

rrs committed rS352215: With the recent commit of ktls, we no longer have a.
With the recent commit of ktls, we no longer have a
Sep 11 2019, 3:41 PM

Sep 10 2019

rrs accepted D21591: avoid unneeded call to arc4random() in syncache_add().
Sep 10 2019, 7:48 PM
rrs created D21582: Ok after several years here is BBR..
Sep 10 2019, 11:41 AM

Sep 9 2019

rrs accepted D21567: Improve DSACK computation.
Sep 9 2019, 3:16 PM
rrs requested changes to D21567: Improve DSACK computation.

I would like to see the Rack version of the code be identical to the tcp_input version.. i.e. it should check

Sep 9 2019, 9:44 AM

Sep 6 2019

rrs closed D21127: Final prep patch for BBR.
Sep 6 2019, 6:30 PM
rrs committed rS351951: This adds in the missing counter initialization which.
This adds in the missing counter initialization which
Sep 6 2019, 6:29 PM
rrs committed rS351934: This adds the final tweaks to LRO that will now allow me.
This adds the final tweaks to LRO that will now allow me
Sep 6 2019, 2:25 PM
rrs closed D21127: Final prep patch for BBR.
Sep 6 2019, 2:25 PM

Sep 3 2019

rrs accepted D21513: Fix TCP SACK block generation in the base stack.
Sep 3 2019, 9:35 PM
rrs accepted D21512: Fix two TCP Rack issues.
Sep 3 2019, 7:07 PM
rrs updated the diff for D21127: Final prep patch for BBR.

First version was missing a (void) in the registration function.. opps

Sep 3 2019, 12:06 PM
rrs updated the diff for D21127: Final prep patch for BBR.

This adds a simple registration mechanism that stacks can use as they successfully register/deregister to
make the LRO code aware if any stack wants to use mbuf-queueing per Drew's suggestion.

Sep 3 2019, 11:55 AM
rrs added inline comments to D21127: Final prep patch for BBR.
Sep 3 2019, 11:38 AM

Aug 26 2019

D21416: Improve locking in tcp_ratelimit.c is now accepted and ready to land.
Aug 26 2019, 2:45 PM

Aug 21 2019

D21277: Add kernel-side support for in-kernel TLS. is now accepted and ready to land.
Aug 21 2019, 10:56 AM
rrs committed rS351328: Fix an issue when TSO and Rack play together. Basically.
Fix an issue when TSO and Rack play together. Basically
Aug 21 2019, 10:45 AM

Aug 15 2019

rrs added a comment to D21127: Final prep patch for BBR.

One more update, mainly comments but we also fix a couple of things

  1. The timestamp provided via the ctf common functions should be the real time not something from the mbuf, we let the transports look at those flags (which both BBR and the latest Rack do).
  2. We also for LRO add a gating of the number of acks (currently set to infinity) that can also cause a wakeup.
Aug 15 2019, 1:40 PM

Aug 13 2019

rrs added a comment to D21127: Final prep patch for BBR.

The LRO patch was missing data and length limits for input processing via
MBUF_QUEUE methods. This update adds that and an elaborate set of
comments to the rack_bbr_common.c module describing in detail what
a transport designer using MBUF_QUEUEING needs to contemplate.

Aug 13 2019, 12:48 PM
rrs committed rS350973: Place back in the dependency on HPTS via module depends versus.
Place back in the dependency on HPTS via module depends versus
Aug 13 2019, 12:41 PM

Aug 2 2019

rrs committed rS350537: Fix one more atomic for i86.
Fix one more atomic for i86
Aug 2 2019, 11:17 AM

Aug 1 2019

rrs committed rS350521: Opps use fetchadd_u64 not long to keep old 32 bit platforms.
Opps use fetchadd_u64 not long to keep old 32 bit platforms
Aug 1 2019, 8:26 PM
rrs added reviewers for D21127: Final prep patch for BBR: gallatin, tuexen, transport.
Aug 1 2019, 2:34 PM
rrs created D21127: Final prep patch for BBR.
Aug 1 2019, 2:33 PM
Herald added 1 blocking reviewer(s) for D20953: Third patchset in the set of patches to bring BBRv1 into the FreeBSD tree: transport.
Aug 1 2019, 2:17 PM
rrs committed rS350501: This adds the third step in getting BBR into the tree. BBR and.
This adds the third step in getting BBR into the tree. BBR and
Aug 1 2019, 2:17 PM
rrs added inline comments to D20953: Third patchset in the set of patches to bring BBRv1 into the FreeBSD tree.
Aug 1 2019, 12:24 PM
rrs added inline comments to D20953: Third patchset in the set of patches to bring BBRv1 into the FreeBSD tree.
Aug 1 2019, 12:15 PM

Jul 29 2019

rrs added inline comments to D18811: nuke sack_newdata.
Jul 29 2019, 6:16 PM
D21038: DSACK patch to conform fully with RFC2883 now requires changes to proceed.

Please validate the rack_bbr_common.c has (or does not have) the extra bit you added in the early part.

Jul 29 2019, 6:13 PM
rrs added a comment to D20953: Third patchset in the set of patches to bring BBRv1 into the FreeBSD tree.

I plan on commiting this August 1st unless I hear screams.. its been pending since July 15th....

Jul 29 2019, 6:07 PM

Jul 21 2019

rrs added a comment to D20953: Third patchset in the set of patches to bring BBRv1 into the FreeBSD tree.

Update to address Michael's comments about copyright (missing new magic declarations at the top) and
also address an issue where if we have an interface that marks it to "not supporting" we properly
disable the rate limiting for that interface and do not allow the user to enable it (since it does
not support it).

Jul 21 2019, 4:13 PM

Jul 18 2019

D20980: Add sysctl to use per end point pair TCP timestamp offsets instead of per connection based offsets is now accepted and ready to land.
Jul 18 2019, 11:26 AM

Jul 16 2019

D20971: Fix compilation on platforms using gcc is now accepted and ready to land.
Jul 16 2019, 4:38 PM

Jul 15 2019

rrs added a reviewer for D20953: Third patchset in the set of patches to bring BBRv1 into the FreeBSD tree: hselasky.
Jul 15 2019, 12:09 PM
rrs created D20953: Third patchset in the set of patches to bring BBRv1 into the FreeBSD tree.
Jul 15 2019, 12:08 PM

Jul 14 2019

rrs committed rS349987: This is the second in a number of patches needed to.
This is the second in a number of patches needed to
Jul 14 2019, 4:06 PM
rrs closed D20908: Fixing the DSACK code to properly clean up after sending a DSACK..
Jul 14 2019, 4:06 PM

Jul 12 2019

rrs committed rS349942: add back the comment around the pending DSACK fixes..
add back the comment around the pending DSACK fixes.
Jul 12 2019, 11:45 AM

Jul 11 2019

rrs committed rS349908: Update to jhb's other suggestion, use #error when.
Update to jhb's other suggestion, use #error when
Jul 11 2019, 4:41 AM
rrs committed rS349907: Update copyright per JBH's suggestions.. thanks..
Update copyright per JBH's suggestions.. thanks.
Jul 11 2019, 4:38 AM

Jul 10 2019

rrs created D20908: Fixing the DSACK code to properly clean up after sending a DSACK..
Jul 10 2019, 9:13 PM
rrs abandoned D19878: Updated Rack Stack.
Jul 10 2019, 8:41 PM
rrs committed rS349893: This commit updates rack to what is basically being used at NF as.
This commit updates rack to what is basically being used at NF as
Jul 10 2019, 8:41 PM
rrs closed D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.
Jul 10 2019, 8:41 PM
rrs added a comment to D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.

Add in the hopeless exit to tcp_hpts and also get the corrected (with DSACK patch) rack_18q22p2

Jul 10 2019, 3:20 PM
rrs added a comment to D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.

Update to include some fixes Michael and I have been working on in a slow slow machine as well
as bring the Rack stack up to the latest production (with some small changes to make it compile).

Jul 10 2019, 11:00 AM

Jul 3 2019

rrs added a comment to D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.

And we can't loose the commented out opt_kern_tls until the kern tls hits the tree

Jul 3 2019, 8:12 PM
rrs added a comment to D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.

It helps if one remembers to bring the changes of the .h file too ... opps

Jul 3 2019, 8:09 PM
rrs added a comment to D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.

Address phabricator nits

Jul 3 2019, 1:08 PM
D20834: First step in bring hpts and infrastructure up for bbr v1 update part1 now requires review to proceed.

In my detailed testing of hpts I found we are *not* doing the right thing as to calculating
the time for sleep. The code was starting at where the prev_slot was and moving
through counting the slots until it found a non-empty slot.

Jul 3 2019, 12:59 PM

Jul 2 2019

rrs added a comment to D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.

So interestingly those includes are not needed.. not even sure the tls one is need (hwtls should be in the tree soon).. however more
interesting is why my freebsd head build works.. not sure why since it does not have hwtls in the mix.. need to go check whats going
on in my conf.

Jul 2 2019, 1:01 PM
rrs updated the diff for D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.
Jul 2 2019, 1:00 PM
rrs updated the diff for D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.
Jul 2 2019, 12:50 PM
rrs created D20834: First step in bring hpts and infrastructure up for bbr v1 update part1.
Jul 2 2019, 10:32 AM

Jun 28 2019

D20616: Add a new external mbuf type that holds multiple unmapped pages. is now accepted and ready to land.
Jun 28 2019, 3:43 PM

Jun 5 2019

rrs added a comment to D20110: Fix infinite loop on older hardware.

sorry wrong revision.. the trouble is in the DSACK commit which is r347382... thats what we are backing out of our sync... (off by
one error.. must be a programmer :-) )

Jun 5 2019, 6:51 PM
rrs added a comment to D20110: Fix infinite loop on older hardware.

There are for more problems then this. And its not explicitly with HPTS but it becomes the canary in the coal mine.
The basic problem is in r347381. This cause a huge slow down to the system at least with INVARIANTs. Hpts in
the form you are dealing with does handle wheel wrap well. We are backing out that change in our upstream
merge. And there will be changes I will be committing to fix wheel_wrap in hpts.

Jun 5 2019, 6:47 PM
rrs accepted D20372: Fix TCP RACK after adding DSACK support.
Jun 5 2019, 9:58 AM

May 9 2019

rrs accepted D20134: Bind TCP HPTS (pacer) threads to NUMA domains.
May 9 2019, 4:50 PM

Apr 10 2019

rrs created D19878: Updated Rack Stack.
Apr 10 2019, 7:48 PM
rrs committed rS346094: Fix a small bug in the tcp_log_id where the bucket.
Fix a small bug in the tcp_log_id where the bucket
Apr 10 2019, 6:58 PM

Apr 3 2019

rrs committed rS345851: Undo my previous erroneous commit changing the tcp_output kassert..
Undo my previous erroneous commit changing the tcp_output kassert.
Apr 3 2019, 7:35 PM

Mar 27 2019

rrs accepted D19334: Receiver Side DSACK (RFC2883) implementation.
Mar 27 2019, 4:09 PM

Mar 26 2019

rrs committed rS345527: Fix a small bug in the tcp_log_id where the bucket.
Fix a small bug in the tcp_log_id where the bucket
Mar 26 2019, 10:41 AM

Mar 23 2019

rrs accepted D19687: Fix KASSERT in tcp_output().
Mar 23 2019, 9:52 AM

Mar 19 2019

rrs abandoned D19611: Incorrect KASSERT.
Mar 19 2019, 5:36 PM

Mar 18 2019

rrs accepted D19355: Make RTO.Initial for TCP controllable via sysctl.
Mar 18 2019, 1:01 PM

Mar 17 2019

rrs updated the diff for D19611: Incorrect KASSERT.
Mar 17 2019, 1:22 PM
rrs created D19611: Incorrect KASSERT.
Mar 17 2019, 1:15 PM

Feb 21 2019

rrs accepted D18960: Patch to MFC TCP reassembly rewrite to stable/11.
Feb 21 2019, 12:21 PM

Feb 20 2019

rrs accepted D19254: Avoid dereferencing a NULL pointer when TCP_REASS_LOGGING is enabled.
Feb 20 2019, 8:31 PM
rrs accepted D18974: Use exponential backoff also for SYN segments.
Feb 20 2019, 12:08 PM

Feb 19 2019

rrs accepted D19152: Various cleanups to the management of multiple TCP stacks..
Feb 19 2019, 11:22 AM

Feb 13 2019

rrs created D19179: Crashing in a race with SOCK_DGRAM and unix domain sockets.
Feb 13 2019, 3:04 PM
rrs committed rS344099: This commit adds the missing release mechanism for the.
This commit adds the missing release mechanism for the
Feb 13 2019, 2:58 PM
rrs closed D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.
Feb 13 2019, 2:58 PM

Feb 11 2019

rrs updated the diff for D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.

get the update in and fix the pesky extra line.

Feb 11 2019, 12:37 PM

Feb 4 2019

rrs accepted D19071: Get TCP CDG CC working if net.inet.tcp.cc.cdg.smoothing_factor = 0 .
Feb 4 2019, 4:25 PM
rrs added inline comments to D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.
Feb 4 2019, 12:11 PM

Feb 1 2019

rrs added inline comments to D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.
Feb 1 2019, 4:05 PM
rrs updated the diff for D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.

Update to hopefully get a better view and not have the stray line in the diff

Feb 1 2019, 4:04 PM
rrs added a comment to D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.

Interesting my svn copy does not have that extra chunk.. let me re-update the diff and use -x -up

Feb 1 2019, 4:03 PM
rrs added inline comments to D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.
Feb 1 2019, 4:01 PM
rrs updated the diff for D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.

This updates the diff to Fix Hann's point on vlans, both the
vlan and lagg are now balanced.

Feb 1 2019, 12:31 PM
rrs added a comment to D19032: Balance the if_snd_tag_alloc with an appropriate if_snd_tag_free.

Ahh good catch I did not know vlan had an allocator.. I will add it to the patch-set

Feb 1 2019, 12:18 PM