Author: rrs
Date: Wed Sep 9 11:11:50 2020
New Revision: 365501
URL: https://svnweb.freebsd.org/changeset/base/365501
Log:
So it turns out that syzkaller hit another crash. It has to do with switching
stacks with a SENT_FIN outstanding. Both rack and bbr will only send a
FIN if all data is a
Author: rrs
Date: Fri Jul 31 10:03:32 2020
New Revision: 363725
URL: https://svnweb.freebsd.org/changeset/base/363725
Log:
The recent changes to move the ref count increment
back from the end of the function created an issue.
If one of the routines returns NULL during setup
we have inp's w
Author: rrs
Date: Tue Jun 16 18:16:45 2020
New Revision: 362234
URL: https://svnweb.freebsd.org/changeset/base/362234
Log:
iSo in doing final checks on OCA firmware with all the latest tweaks the
dup-ack checking
packet drill script was failing with a number of unexpected acks. So it turns
Author: rrs
Date: Tue Jun 16 12:26:23 2020
New Revision: 362225
URL: https://svnweb.freebsd.org/changeset/base/362225
Log:
So it turns out rack has a shortcoming in dup-ack counting. It counts the
dupacks but
then does not properly respond to them. This is because a few missing bits
are not
Author: rrs
Date: Fri Jun 12 19:56:19 2020
New Revision: 362113
URL: https://svnweb.freebsd.org/changeset/base/362113
Log:
So it turns out with the right window scaling you can get the code in all
stacks to
always want to do a window update, even when no data can be sent. Now in
cases where
Author: rrs
Date: Mon Jun 8 11:48:07 2020
New Revision: 361926
URL: https://svnweb.freebsd.org/changeset/base/361926
Log:
An important statistic in determining if a server process (or client) is
being delayed
is to know the time to first byte in and time to first byte out. Currently we
hav
Author: rrs
Date: Wed Jun 3 14:16:40 2020
New Revision: 361752
URL: https://svnweb.freebsd.org/changeset/base/361752
Log:
We should never allow either the broadcast or IN_ADDR_ANY to be
connected to or sent to. This was fond when working with Michael
Tuexen and Skyzaller. Skyzaller seems to
Author: rrs
Date: Wed Jun 3 14:07:31 2020
New Revision: 361751
URL: https://svnweb.freebsd.org/changeset/base/361751
Log:
This fixes a couple of skyzaller crashes. Most
of them have to do with TFO. Even the default stack
had one of the issues:
1) We need to make sure for rack that we d
Author: rrs
Date: Fri May 15 14:00:12 2020
New Revision: 361080
URL: https://svnweb.freebsd.org/changeset/base/361080
Log:
This fixes several skyzaller issues found with the
help of Michael Tuexen. There was some accounting
errors with TCPFO for bbr and also for both rack
and bbr there was
Author: rrs
Date: Thu May 7 20:29:38 2020
New Revision: 360798
URL: https://svnweb.freebsd.org/changeset/base/360798
Log:
When in the SYN-SENT state bbr and rack will not properly send an ACK but
instead start the D-ACK timer. This
causes so_reuseport_lb_test to fail since it slows down how
Author: rrs
Date: Thu May 7 10:46:02 2020
New Revision: 360776
URL: https://svnweb.freebsd.org/changeset/base/360776
Log:
NF has an internal option that changes the tcp_mcopy_m routine slightly (has
a few extra arguments). Recently that changed to only have one arg extra so
that two ifdefs
Author: rrs
Date: Mon May 4 23:02:58 2020
New Revision: 360644
URL: https://svnweb.freebsd.org/changeset/base/360644
Log:
This fixes two issues found by ankitrahej...@gmail.com
1) When BBR retransmits the syn it was messing up the snd_max
2) When we need to send a RST we might not send it w
Author: rrs
Date: Mon May 4 20:28:53 2020
New Revision: 360639
URL: https://svnweb.freebsd.org/changeset/base/360639
Log:
This commit brings things into sync with the advancements that
have been made in rack and adds a few fixes in BBR. This also
removes any possibility of incorrectly doing
Author: rrs
Date: Mon May 4 20:19:57 2020
New Revision: 360638
URL: https://svnweb.freebsd.org/changeset/base/360638
Log:
Adjust the fb to have a way to ask the underlying stack
if it can support the PRUS option (OOB). And then have
the new function call that to validate and give the
corr
Author: rrs
Date: Mon Apr 27 16:30:29 2020
New Revision: 360385
URL: https://svnweb.freebsd.org/changeset/base/360385
Log:
This change does a small prepratory step in getting the
latest rack and bbr in from the NF repo. When those come
in the OOB data handling will be fixed where Skyzaller c
Author: rrs
Date: Wed Feb 26 13:48:33 2020
New Revision: 358332
URL: https://svnweb.freebsd.org/changeset/base/358332
Log:
This commit expands tcp_ratelimit to be able to handle cards
like the mlx-c5 and c6 that require a "setup" routine before
the tcp_ratelimit code can declare and use a ra
Author: rrs
Date: Wed Feb 12 15:26:56 2020
New Revision: 357823
URL: https://svnweb.freebsd.org/changeset/base/357823
Log:
Lets get the real correct version.. gessh. I need
more coffee evidently.
Sponsored by: Netflix
Modified:
head/sys/netinet/tcp_ratelimit.c
Modified: head/sys/netin
Author: rrs
Date: Wed Feb 12 13:31:36 2020
New Revision: 357818
URL: https://svnweb.freebsd.org/changeset/base/357818
Log:
White space cleanup -- remove trailing tab's or spaces
from any line.
Sponsored by: Netflix Inc.
Modified:
head/sys/netinet/cc/cc_cdg.c
head/sys/netinet/cc/cc_dc
Author: rrs
Date: Wed Feb 12 13:07:09 2020
New Revision: 357817
URL: https://svnweb.freebsd.org/changeset/base/357817
Log:
Whitespace, remove from three files trailing white
space (leftover presents from emacs).
Sponsored by: Netflix Inc.
Modified:
head/sys/netinet/tcp_hpts.c
head/sy
Author: rrs
Date: Wed Feb 12 13:04:19 2020
New Revision: 357816
URL: https://svnweb.freebsd.org/changeset/base/357816
Log:
This small fix makes it so we properly follow
the RFC and only enable ECN when both the
CWR and ECT bits our set within the SYN packet.
Sponsored by: Netflix Inc.
Author: rrs
Date: Wed Feb 12 12:40:06 2020
New Revision: 357815
URL: https://svnweb.freebsd.org/changeset/base/357815
Log:
Remove all trailing white space from the BBR/Rack fold. Bits
left around by emacs (thanks emacs).
Modified:
head/sys/netinet/tcp_stacks/bbr.c
head/sys/netinet/tcp_sta
Author: rrs
Date: Wed Feb 12 12:36:55 2020
New Revision: 357814
URL: https://svnweb.freebsd.org/changeset/base/357814
Log:
Now that all of the stats framework is
in FreeBSD the bits that disabled stats
when netflix-stats is not defined is no longer
needed. Lets remove these bits so that we
Author: rrs
Date: Mon Jan 6 15:29:14 2020
New Revision: 356417
URL: https://svnweb.freebsd.org/changeset/base/356417
Log:
This catches rack up in the recent changes to ECN and
also commonizes the functions that both the freebsd and
rack stack uses.
Sponsored by:Netflix Inc
Differenti
Author: rrs
Date: Mon Jan 6 12:48:06 2020
New Revision: 356414
URL: https://svnweb.freebsd.org/changeset/base/356414
Log:
This change adds a small feature to the tcp logging code. Basically
a connection can now have a separate tag added to the id.
Obtained from:Lawrence Stewart
Author: rrs
Date: Tue Dec 17 16:08:07 2019
New Revision: 355859
URL: https://svnweb.freebsd.org/changeset/base/355859
Log:
This commit is a bit of a re-arrange of deck chairs. It
gets both rack and bbr ready for the completion of the STATs
framework in FreeBSD. For now if you don't have both
Author: rrs
Date: Thu Oct 24 05:54:30 2019
New Revision: 354013
URL: https://svnweb.freebsd.org/changeset/base/354013
Log:
Fix a small bug in bbr when running under a VM. Basically what
happens is we are more delayed in the pacer calling in so
we remove the stack from the pacer and recalcula
Author: rrs
Date: Mon Oct 14 13:10:29 2019
New Revision: 353490
URL: https://svnweb.freebsd.org/changeset/base/353490
Log:
if_hw_tsomaxsegsize needs to be initialized to zero, just
like in bbr.c and tcp_output.c
Modified:
head/sys/netinet/tcp_stacks/rack.c
Modified: head/sys/netinet/tcp_st
Author: rrs
Date: Sun Oct 6 22:29:02 2019
New Revision: 353156
URL: https://svnweb.freebsd.org/changeset/base/353156
Log:
Brad Davis identified a problem with the new LRO code, VLAN's
no longer worked. The problem was that the defines used the
same space as the VLAN id. This commit does thr
Author: rrs
Date: Tue Sep 24 20:36:43 2019
New Revision: 352661
URL: https://svnweb.freebsd.org/changeset/base/352661
Log:
lets put (void) in a couple of functions to keep older platforms that
are stuck with gcc happy (ppc). The changes are needed in both bbr and
rack.
Obtained from:
include ratelimit.. it costs little and makes
it so if you
do get a nic that supports rate limiting you will be able to take advantage of
it)
R
> On Sep 24, 2019, at 1:06 PM, Randall Stewart wrote:
>
> Right
>
> Thats because GENERIC does not add the optional TCP stacks.
>
&
Author: rrs
Date: Tue Sep 24 20:11:55 2019
New Revision: 352660
URL: https://svnweb.freebsd.org/changeset/base/352660
Log:
don't call in_ratelmit detach when RATELIMIT is not
compiled in the kernel.
Modified:
head/sys/netinet/tcp_stacks/bbr.c
Modified: head/sys/netinet/tcp_stacks/bbr.c
===
;
> On Tue, Sep 24, 2019 at 9:55 PM Randall Stewart wrote:
>>
>> 12.0R would not have BBR .. its only in head… hmm it could be a issue with
>> TCP_RATELIMIT not defined
>> though I did compile GENERIC without the extra stacks (and without rate
>> limit and hpts) an
Author: rrs
Date: Tue Sep 24 20:04:31 2019
New Revision: 352659
URL: https://svnweb.freebsd.org/changeset/base/352659
Log:
Fix the ifdefs in tcp_ratelimit.h. They were reversed so
that instead of functions only being inside the _KERNEL and
the absence of RATELIMIT causing us to have NULL/err
Ok I have found it
Its a reversal in an ifdef in tcp_ratelimit.h .. it supposed to be that if
its not define (RATELIMIT) the main interfaces return errors.. and the
ifdef kernel/ratelimit is reversed of what it should be.
Let me fix that :)
R
> On Sep 24, 2019, at 12:55 PM, Randall Stew
, 2019 at 9:29 PM O. Hartmann wrote:
>>
>> -BEGIN PGP SIGNED MESSAGE-
>> Hash: SHA256
>>
>> Am Tue, 24 Sep 2019 18:18:11 + (UTC)
>> Randall Stewart schrieb:
>>
>>> Author: rrs
>>> Date: Tue Sep 24 18:18:11 2019
>>>
this time it *should* have
built fine :(
R
> On Sep 24, 2019, at 12:28 PM, O. Hartmann wrote:
>
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA256
>
> Am Tue, 24 Sep 2019 18:18:11 +0000 (UTC)
> Randall Stewart schrieb:
>
>> Author: rrs
>> Date: Tue Sep 24 1
Y OF
+ * SUCH DAMAGE.
+ */
+
+/*
+ * Author: Randall Stewart
+ */
+#include
+__FBSDID("$FreeBSD$");
+#include
+#include
+#include
+#include
+
+void
+reset_time(struct time_filter *tf, uint32_t time_len)
+{
+ tf->cur_time_limit = time_len;
+}
+
+void
+reset_time_small(struct time_fi
Author: rrs
Date: Wed Sep 11 15:41:36 2019
New Revision: 352215
URL: https://svnweb.freebsd.org/changeset/base/352215
Log:
With the recent commit of ktls, we no longer have a
sb_tls_flags, its just the sb_flags. Also the ratelimit
code, now that the defintion is in sockbuf.h, does not
need
Author: rrs
Date: Fri Sep 6 18:29:48 2019
New Revision: 351951
URL: https://svnweb.freebsd.org/changeset/base/351951
Log:
This adds in the missing counter initialization which
I had forgotten to bring over.. opps.
Differential Revision: https://reviews.freebsd.org/D21127
Modified:
hea
Author: rrs
Date: Fri Sep 6 14:25:41 2019
New Revision: 351934
URL: https://svnweb.freebsd.org/changeset/base/351934
Log:
This adds the final tweaks to LRO that will now allow me
to add BBR. These changes make it so you can get an
array of timestamps instead of a compressed ack/data segment
Author: rrs
Date: Wed Apr 10 18:58:11 2019
New Revision: 346094
URL: https://svnweb.freebsd.org/changeset/base/346094
Log:
Fix a small bug in the tcp_log_id where the bucket
was unlocked and yet the bucket-unlock flag was not
changed to false. This can cause a panic if INVARIANTS
is on and
Author: rrs
Date: Wed Apr 3 19:35:07 2019
New Revision: 345851
URL: https://svnweb.freebsd.org/changeset/base/345851
Log:
Undo my previous erroneous commit changing the tcp_output kassert.
Hmm now the question is where did the tcp_log_id change go :o
Modified:
head/sys/netinet/tcp_output.c
Author: rrs
Date: Wed Aug 21 10:45:28 2019
New Revision: 351328
URL: https://svnweb.freebsd.org/changeset/base/351328
Log:
Fix an issue when TSO and Rack play together. Basically
an retransmission of the initial SYN (with data) would
cause us to strip the SYN and decrement/increase offset/le
Author: rrs
Date: Tue Aug 13 12:41:15 2019
New Revision: 350973
URL: https://svnweb.freebsd.org/changeset/base/350973
Log:
Place back in the dependency on HPTS via module depends versus
a fatal error in compiling. This was taken out by mistake
when I mis-merged from the 18q22p2 sources of ra
Author: rrs
Date: Fri Aug 2 11:17:07 2019
New Revision: 350537
URL: https://svnweb.freebsd.org/changeset/base/350537
Log:
Fix one more atomic for i86
Obtained from:mtue...@freebsd.org
Modified:
head/sys/netinet/tcp_ratelimit.c
Modified: head/sys/netinet/tcp_ratelimit.c
===
Author: rrs
Date: Thu Aug 1 20:26:27 2019
New Revision: 350521
URL: https://svnweb.freebsd.org/changeset/base/350521
Log:
Opps use fetchadd_u64 not long to keep old 32 bit platforms
happy.
Modified:
head/sys/netinet/tcp_ratelimit.c
Modified: head/sys/netinet/tcp_ratelimit.c
==
PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTER
Author: rrs
Date: Sun Jul 14 16:05:47 2019
New Revision: 349987
URL: https://svnweb.freebsd.org/changeset/base/349987
Log:
This is the second in a number of patches needed to
get BBRv1 into the tree. This fixes the DSACK bug but
is also needed by BBR. We have yet to go two more
one will be
Author: rrs
Date: Fri Jul 12 11:45:42 2019
New Revision: 349942
URL: https://svnweb.freebsd.org/changeset/base/349942
Log:
add back the comment around the pending DSACK fixes.
Modified:
head/sys/netinet/tcp_stacks/rack.c
Modified: head/sys/netinet/tcp_stacks/rack.c
==
opps.. that was a error on my part I will fix it :)
> On Jul 11, 2019, at 4:37 AM, Enji Cooper wrote:
>
>
>> On Jul 10, 2019, at 9:38 PM, Randall Stewart wrote:
>>
>> Author: rrs
>> Date: Thu Jul 11 04:38:33 2019
>> New Revision: 349907
>> U
on using M_PROTO12 for that
what do you think?
R
> On Jul 10, 2019, at 7:28 PM, John Baldwin wrote:
>
> On 7/10/19 1:40 PM, Randall Stewart wrote:
>> Author: rrs
>> Date: Wed Jul 10 20:40:39 2019
>> New Revision: 349893
>> URL: https://svnweb.freebsd.org
Author: rrs
Date: Thu Jul 11 04:40:58 2019
New Revision: 349908
URL: https://svnweb.freebsd.org/changeset/base/349908
Log:
Update to jhb's other suggestion, use #error when
we are missing HPTS.
Modified:
head/sys/netinet/tcp_stacks/rack.c
Modified: head/sys/netinet/tcp_stacks/rack.c
=
Author: rrs
Date: Thu Jul 11 04:38:33 2019
New Revision: 349907
URL: https://svnweb.freebsd.org/changeset/base/349907
Log:
Update copyright per JBH's suggestions.. thanks.
Modified:
head/sys/netinet/tcp_stacks/rack.c
Modified: head/sys/netinet/tcp_stacks/rack.c
==
Author: rrs
Date: Wed Jul 10 20:40:39 2019
New Revision: 349893
URL: https://svnweb.freebsd.org/changeset/base/349893
Log:
This commit updates rack to what is basically being used at NF as
well as sets in some of the groundwork for committing BBR. The
hpts system is updated as well as some o
Author: rrs
Date: Wed Apr 10 18:58:11 2019
New Revision: 346094
URL: https://svnweb.freebsd.org/changeset/base/346094
Log:
Fix a small bug in the tcp_log_id where the bucket
was unlocked and yet the bucket-unlock flag was not
changed to false. This can cause a panic if INVARIANTS
is on and
Author: rrs
Date: Wed Apr 3 19:35:07 2019
New Revision: 345851
URL: https://svnweb.freebsd.org/changeset/base/345851
Log:
Undo my previous erroneous commit changing the tcp_output kassert.
Hmm now the question is where did the tcp_log_id change go :o
Modified:
head/sys/netinet/tcp_output.c
Author: rrs
Date: Tue Mar 26 10:41:27 2019
New Revision: 345527
URL: https://svnweb.freebsd.org/changeset/base/345527
Log:
Fix a small bug in the tcp_log_id where the bucket
was unlocked and yet the bucket-unlock flag was not
changed to false. This can cause a panic if INVARIANTS
is on and
> On Feb 13, 2019, at 1:10 PM, John Baldwin wrote:
>
> On 2/13/19 10:03 AM, Randall Stewart wrote:
>> oh and one other thing..
>>
>> It was *not* a random IFP.. it was the IFP to the lagg.
>>
>> I.e. an alloc() was done to the lagg.. and the free
oh and one other thing..
It was *not* a random IFP.. it was the IFP to the lagg.
I.e. an alloc() was done to the lagg.. and the free was
done back to the same IFP (that provided the allocate).
R
> On Feb 13, 2019, at 1:02 PM, Randall Stewart wrote:
>
> I disagree. If you define an
provide a Free…
R
> On Feb 13, 2019, at 12:09 PM, John Baldwin wrote:
>
> On 2/13/19 6:57 AM, Randall Stewart wrote:
>> Author: rrs
>> Date: Wed Feb 13 14:57:59 2019
>> New Revision: 344099
>> URL: https://svnweb.freebsd.org/changeset/base/344099
>>
&
Author: rrs
Date: Wed Feb 13 14:57:59 2019
New Revision: 344099
URL: https://svnweb.freebsd.org/changeset/base/344099
Log:
This commit adds the missing release mechanism for the
ratelimiting code. The two modules (lagg and vlan) did have
allocation routines, and even though they are indirect
Author: rrs
Date: Mon Aug 20 12:43:18 2018
New Revision: 338102
URL: https://svnweb.freebsd.org/changeset/base/338102
Log:
This change represents a substantial restructure of the way we
reassembly inbound tcp segments. The old algorithm just blindly
dropped in segments without coalescing. Th
Author: rrs
Date: Wed Aug 8 13:36:49 2018
New Revision: 337455
URL: https://svnweb.freebsd.org/changeset/base/337455
Log:
Fix a small bug in rack where it will
end up sending the FIN twice.
Sponsored by: Netflix Inc.
Differential Revision:https://reviews.freebsd.org/D16604
Modifi
Author: rrs
Date: Mon Aug 6 09:22:07 2018
New Revision: 337375
URL: https://svnweb.freebsd.org/changeset/base/337375
Log:
This fixes a bug in Rack where we were
not properly using the correct value for
Delayed Ack.
Sponsored by: Netflix Inc.
Differential Revision: https://reviews.fre
Author: rrs
Date: Mon Jul 30 10:23:29 2018
New Revision: 336893
URL: https://svnweb.freebsd.org/changeset/base/336893
Log:
This fixes a hole where rack could end up
sending an invalid segment into the reassembly
queue. This would happen if you enabled the
data after close option.
Spon
Author: rrs
Date: Tue Jul 24 14:55:47 2018
New Revision: 336672
URL: https://svnweb.freebsd.org/changeset/base/336672
Log:
Delete the example tcp stack "fastpath" which
was only put in has an example.
Sponsored by: Netflix inc.
Differential Revision:https://reviews.freebsd.org/D
Author: rrs
Date: Wed Jul 18 22:49:53 2018
New Revision: 336465
URL: https://svnweb.freebsd.org/changeset/base/336465
Log:
Bump the ICMP echo limits to match the RFC
Reviewed by: tuexen
Sponsored by: Netflix Inc.
Differential Revision:https://reviews.freebsd.org/D16333
Author: rrs
Date: Thu Jun 21 21:03:58 2018
New Revision: 335502
URL: https://svnweb.freebsd.org/changeset/base/335502
Log:
This adds in an optimization so that we only walk one
time through the mbuf chain during copy and TSO limiting.
It is used by both Rack and now the FreeBSD stack.
Spon
Author: rrs
Date: Tue Jun 19 11:20:28 2018
New Revision: 335364
URL: https://svnweb.freebsd.org/changeset/base/335364
Log:
Make sure that the t_peakrate_thr is not compiled in
by default until NF can upstream it.
Reviewed by: and suggested lstewart
Sponsored by: Netflix Inc.
Modified:
Author: rrs
Date: Tue Jun 19 05:28:14 2018
New Revision: 335361
URL: https://svnweb.freebsd.org/changeset/base/335361
Log:
Move the tp set back to where it was before
we started playing with the VNET sets. This
way we have verified the INP settings before
we go to the trouble of de-referen
Author: rrs
Date: Mon Jun 18 14:10:12 2018
New Revision: 335317
URL: https://svnweb.freebsd.org/changeset/base/335317
Log:
Move to using the inp->vnet pointer has suggested by lstewart.
This is far better since the hpts system is using the inp
as its basis anyway. Unfortunately his comments
Author: rrs
Date: Thu Jun 14 03:27:42 2018
New Revision: 335106
URL: https://svnweb.freebsd.org/changeset/base/335106
Log:
This fixes several bugs that Larry Rosenman helped me find in
Rack with respect to its handling of TCP Fast Open. Several
fixes all related to TFO are included in this c
Author: rrs
Date: Tue Jun 12 23:54:08 2018
New Revision: 335022
URL: https://svnweb.freebsd.org/changeset/base/335022
Log:
This fixes missing VNET sets in the hpts system. Basically
without this and running vnets with a TCP stack that uses
some of the features is a recipe for panic (without
> On Jun 7, 2018, at 6:01 PM, hiren panchasara
> wrote:
>
> On 06/07/18 at 06:18P, Randall Stewart wrote:
>> Author: rrs
>> Date: Thu Jun 7 18:18:13 2018
>> New Revision: 334804
>> URL: https://svnweb.freebsd.org/changeset/base/334804
>>
>>
Author: rrs
Date: Thu Jun 7 20:57:12 2018
New Revision: 334815
URL: https://svnweb.freebsd.org/changeset/base/334815
Log:
Take out the stack alias inadvertantly added by my commit.
Reported by: Peter Lei
Modified:
head/sys/modules/tcp/rack/Makefile
Modified: head/sys/modules/tcp/rack/
Author: rrs
Date: Thu Jun 7 19:57:55 2018
New Revision: 334813
URL: https://svnweb.freebsd.org/changeset/base/334813
Log:
Fix build issue with const and volatile and the
myriad ways that the various compliers treat this. The
only safe prefetch appears to be for AMD. The other
compilers ei
Author: rrs
Date: Thu Jun 7 18:18:13 2018
New Revision: 334804
URL: https://svnweb.freebsd.org/changeset/base/334804
Log:
This commit brings in a new refactored TCP stack called Rack.
Rack includes the following features:
- A different SACK processing scheme (the old sack structures are no
Author: rrs
Date: Thu Apr 26 21:41:16 2018
New Revision: 333041
URL: https://svnweb.freebsd.org/changeset/base/333041
Log:
This change re-arranges the fields within the tcp-pcb so that
they are more in order of cache line use as one passes
through the tcp_input/output paths (non-errors most
Author: rrs
Date: Thu Apr 19 15:03:48 2018
New Revision: 332774
URL: https://svnweb.freebsd.org/changeset/base/332774
Log:
These two modules need the tcp_hpts.h file for
when the option is enabled (not sure how LINT/build-universe
missed this) opps.
Sponsored by: Netflix Inc
Modified:
Author: rrs
Date: Thu Apr 19 13:37:59 2018
New Revision: 332770
URL: https://svnweb.freebsd.org/changeset/base/332770
Log:
This commit brings in the TCP high precision timer system (tcp_hpts).
It is the forerunner/foundational work of bringing in both Rack and BBR
which use hpts for pacing o
Author: rrs
Date: Tue Aug 16 15:17:36 2016
New Revision: 304224
URL: https://svnweb.freebsd.org/changeset/base/304224
Log:
A few more wording tweaks as suggested (with some modifications
as well) by Ravi Pokala. Thanks for the comments :-)
Sponsored by: Netflix Inc.
Modified:
head/sys/net
Author: rrs
Date: Tue Aug 16 15:11:46 2016
New Revision: 304223
URL: https://svnweb.freebsd.org/changeset/base/304223
Log:
Here we update the modular tcp to be able to switch to an
alternate TCP stack in other then the closed state (pre-listen/connect).
The idea is that *if* that is support
> On Tue, Aug 16, 2016 at 12:40:56PM +, Randall Stewart wrote:
>
>> Author: rrs
>> Date: Tue Aug 16 12:40:56 2016
>> New Revision: 304218
>> URL: https://svnweb.freebsd.org/changeset/base/304218
>>
>> Log:
>> This cleans up the timer code in TCP
outside of wanting to kill the tcb.. but
I am not sure why one
would want to hold the INFO_RLOCK if one did not want to purge the tcb.
R
> On Aug 16, 2016, at 6:14 AM, Hans Petter Selasky wrote:
>
> On 08/16/16 15:01, Randall Stewart wrote:
>> Sure
>>
>> Let me add
Author: rrs
Date: Tue Aug 16 13:08:03 2016
New Revision: 304219
URL: https://svnweb.freebsd.org/changeset/base/304219
Log:
Comments describing how to properly use the new lock_add functions
and its respective companion.
Sponsored by: Netflix Inc.
Modified:
head/sys/netinet/tcp_timer.c
have to use its companion function to drop the reference).
> On Aug 16, 2016, at 5:58 AM, Hans Petter Selasky wrote:
>
> On 08/16/16 14:40, Randall Stewart wrote:
>> +int
>> +tcp_inpinfo_lock_add(struct inpcb *inp)
>> +{
>> +in_pcbref(inp);
>> +I
Author: rrs
Date: Tue Aug 16 12:40:56 2016
New Revision: 304218
URL: https://svnweb.freebsd.org/changeset/base/304218
Log:
This cleans up the timer code in TCP and also makes it so we do not
take the INFO lock *unless* we are really going to delete the TCB.
Differential Revision:D
Author: rrs
Date: Wed Jul 27 20:37:32 2016
New Revision: 303412
URL: https://svnweb.freebsd.org/changeset/base/303412
Log:
Remove myself from kern_timeout.c yeah!
Modified:
head/MAINTAINERS
Modified: head/MAINTAINERS
===
You are most welcome to backout anything you like.. as far as
I am concerned you own the code..
R
> On Jul 20, 2016, at 6:35 PM, Gleb Smirnoff wrote:
>
> Randall,
>
> I have just tested and r303037 brings the TCP panic back. I
> got two crashes during 2.5 hours.
>
&g
:53 AM, Gleb Smirnoff wrote:
>
> On Tue, Jul 19, 2016 at 06:31:19PM +, Randall Stewart wrote:
> R> Author: rrs
> R> Date: Tue Jul 19 18:31:19 2016
> R> New Revision: 303037
> R> URL: https://svnweb.freebsd.org/changeset/base/303037
> R>
> R> Log:
&g
Author: rrs
Date: Tue Jul 19 18:31:19 2016
New Revision: 303037
URL: https://svnweb.freebsd.org/changeset/base/303037
Log:
This reverts out Gleb's changes and adds three small
fixes that I think closes up the races Gleb was
looking for. This is running quite nicely in Netflix and
now no lo
4PM +0200, Randall Stewart wrote:
>> Glen:
>>
>> My changes work.. I have them running in NF in at least 1/2 dozen machines.
>>
>
> For how long? What are the uptimes on these machines?
>
> This is the blocker for 11.0-BETA2, and I don't want to see more
&
yet another mess in that
respect (TCP depends on ASYNC-drain).
I can commit what I have.. if you like.. or not.. I really don’t care (I hate
kern_timeout.c :-o)
R
> On Jul 19, 2016, at 2:25 PM, Glen Barber wrote:
>
> On Tue, Jul 19, 2016 at 01:43:16PM +0200, Randall Stewart wrote
.. I know it works (it ran
on a number of machines at NF last night.. and we will of course continue
testing
it as we finish our dev testing for the upcoming OCA software release).. For now
this is what will be going out into the OCA’s at least :-)
R
> On Jul 18, 2016, at 6:19 PM, Randall Stew
did not fix.
R
> On Jul 18, 2016, at 5:44 PM, Randall Stewart wrote:
>
> Gleb:
>
> This now leaks TCP-PCB’s since you have broken the return codes with all your
> fixes that used to be in here.
>
> It was
>
> return 1 — You stopped the callout
> return 0 —
> CC_UNLOCK(cc);
> - return (-1);
> + return (0);
> }
>
> c->c_iflags &= ~CALLOUT_PENDING;
>
Randall Stewart
r...@netflix.com
803-317-4952
___
svn-src-head@freebsd.o
Author: rrs
Date: Tue May 17 09:53:22 2016
New Revision: 300042
URL: https://svnweb.freebsd.org/changeset/base/300042
Log:
This small change adopts the excellent suggestion for using named
structures in the add of a new tcp-stack that came in late to me
via email after the last commit. It al
Author: rrs
Date: Thu Apr 28 15:53:10 2016
New Revision: 298747
URL: https://svnweb.freebsd.org/changeset/base/298747
Log:
Complete the UDP tunneling of ICMP msgs to those protocols
interested in having tunneled UDP and finding out about the
ICMP (tested by Michael Tuexen with SCTP.. soon to
Author: rrs
Date: Thu Apr 28 13:27:12 2016
New Revision: 298743
URL: https://svnweb.freebsd.org/changeset/base/298743
Log:
This cleans up the timers code in TCP to start using the new
async_drain functionality. This as been tested in NF as well as
by Verisign. Still to do in here is to remov
Author: rrs
Date: Thu Apr 7 09:34:41 2016
New Revision: 297663
URL: https://svnweb.freebsd.org/changeset/base/297663
Log:
A couple of minor changes that I missed that Michael had done, most noted
in these is the change to non-strict ordering for incoming data (this will
make pkt-drill test
1 - 100 of 272 matches
Mail list logo