svn commit: r334804 - in head/sys: kern modules/tcp modules/tcp/rack netinet netinet/tcp_stacks sys

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

svn commit: r334813 - head/sys/sys

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

svn commit: r334815 - head/sys/modules/tcp/rack

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/

svn commit: r335022 - head/sys/netinet

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

svn commit: r335106 - head/sys/netinet/tcp_stacks

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

svn commit: r335317 - head/sys/netinet

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

svn commit: r335361 - head/sys/netinet

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

svn commit: r335364 - head/sys/netinet/tcp_stacks

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:

svn commit: r335502 - head/sys/netinet

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

svn commit: r332770 - in head/sys: conf netinet netinet/tcp_stacks sys

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

svn commit: r332774 - head/sys/netinet

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:

svn commit: r333041 - head/sys/netinet

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

svn commit: r353156 - in head/sys: netinet sys

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

svn commit: r353490 - head/sys/netinet/tcp_stacks

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

svn commit: r354013 - head/sys/netinet/tcp_stacks

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

svn commit: r351328 - head/sys/netinet/tcp_stacks

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

svn commit: r345851 - head/sys/netinet

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

svn commit: r346094 - head/sys/netinet

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

svn commit: r351934 - in head/sys: netinet netinet/tcp_stacks sys

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

svn commit: r351951 - head/sys/netinet

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

svn commit: r352215 - head/sys/netinet

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

svn commit: r352657 - in head/sys: conf kern modules/tcp modules/tcp/bbr netinet netinet/tcp_stacks sys

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

svn commit: r352659 - head/sys/netinet

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

svn commit: r352660 - head/sys/netinet/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 ===

svn commit: r352661 - head/sys/netinet/tcp_stacks

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:

svn commit: r349893 - in head/sys: modules/tcp/rack netinet netinet/tcp_stacks sys

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

svn commit: r349907 - head/sys/netinet/tcp_stacks

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 ==

svn commit: r349908 - head/sys/netinet/tcp_stacks

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 =

svn commit: r349942 - head/sys/netinet/tcp_stacks

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 ==

svn commit: r349987 - in head/sys/netinet: . tcp_stacks

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

svn commit: r350501 - in head/sys: conf dev/cxgbe dev/mlx5/mlx5_en net netinet

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

svn commit: r350521 - head/sys/netinet

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 ==

svn commit: r350537 - head/sys/netinet

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 ===

svn commit: r350973 - head/sys/netinet/tcp_stacks

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

svn commit: r363725 - head/sys/netinet

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

svn commit: r355859 - in head/sys/netinet: . tcp_stacks

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

svn commit: r356414 - head/sys/netinet

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

svn commit: r356417 - in head/sys/netinet: . tcp_stacks

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

svn commit: r360385 - head/sys/netinet

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

svn commit: r360638 - head/sys/netinet

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

svn commit: r360639 - head/sys/netinet/tcp_stacks

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

svn commit: r360644 - head/sys/netinet/tcp_stacks

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

svn commit: r360776 - head/sys/netinet/tcp_stacks

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

svn commit: r360798 - head/sys/netinet/tcp_stacks

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

svn commit: r361080 - head/sys/netinet/tcp_stacks

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

svn commit: r361751 - in head/sys/netinet: . tcp_stacks

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

svn commit: r361752 - head/sys/netinet

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

svn commit: r361926 - in head/sys/netinet: . tcp_stacks

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

svn commit: r362113 - in head/sys/netinet: . tcp_stacks

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

svn commit: r362225 - head/sys/netinet/tcp_stacks

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

svn commit: r362234 - head/sys/netinet/tcp_stacks

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

svn commit: r357814 - head/sys/netinet/tcp_stacks

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

svn commit: r357815 - head/sys/netinet/tcp_stacks

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

svn commit: r357816 - head/sys/netinet

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.

svn commit: r357817 - head/sys/netinet

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

svn commit: r357818 - in head/sys/netinet: . cc

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

svn commit: r357823 - head/sys/netinet

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

svn commit: r358332 - in head/sys: net netinet

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

svn commit: r219057 - head/sys/netinet

Author: rrs Date: Sat Feb 26 15:23:46 2011 New Revision: 219057 URL: http://svn.freebsd.org/changeset/base/219057 Log: Improvements to CC modules: 1) Add four new points that allow you to get more information to cc algo's 2) Fix the case where user changes module on a existing TCB, in

svn commit: r219120 - head/sys/netinet

Author: rrs Date: Tue Mar 1 00:37:46 2011 New Revision: 219120 URL: http://svn.freebsd.org/changeset/base/219120 Log: Adds a new Congestion Control that helps reduce the RTT that a flow will build up in buffers in transit. It is a slight modification to RFC2581 but is more friendly i.e. l

svn commit: r219397 - head/sys/netinet

Author: rrs Date: Tue Mar 8 11:58:25 2011 New Revision: 219397 URL: http://svn.freebsd.org/changeset/base/219397 Log: Tunes and fixes the new DC-CC to seem to hit the right mix. Still may need some tweaks but it appears to almost not give away too much to an RFC2581 flow, but can really

Re: svn commit: r217592 - head/sys/netinet

> On Wednesday, January 19, 2011 2:07:16 pm Randall Stewart wrote: >> Author: rrs >> Date: Wed Jan 19 19:07:16 2011 >> New Revision: 217592 >> URL: http://svn.freebsd.org/changeset/base/217592 >> >> Log: >> Fix a bug where Multicast packets sent from a >

Re: svn commit: r217592 - head/sys/netinet

is correct. R On Mar 29, 2011, at 2:01 PM, John Baldwin wrote: > On Wednesday, January 19, 2011 2:07:16 pm Randall Stewart wrote: >> Author: rrs >> Date: Wed Jan 19 19:07:16 2011 >> New Revision: 217592 >> URL: http://svn.freebsd.org/changeset/base/217592 >> >

svn commit: r207441 - head/sys/mips/mips

Author: rrs Date: Fri Apr 30 17:12:20 2010 New Revision: 207441 URL: http://svn.freebsd.org/changeset/base/207441 Log: Bug in the memory mapping module. The wrong physaddr was being used in the macro (1 should be used not 2)... Obtained from:JC Modified: head/sys/mips/mips/pm

svn commit: r207924 - head/sys/netinet

Author: rrs Date: Tue May 11 17:02:29 2010 New Revision: 207924 URL: http://svn.freebsd.org/changeset/base/207924 Log: This fixes a bug with the one-2-one model socket when a user sets up a socket to a server sends data and closes the socket before the server has called accept(). It used t

svn commit: r207963 - head/sys/netinet

Author: rrs Date: Wed May 12 13:45:46 2010 New Revision: 207963 URL: http://svn.freebsd.org/changeset/base/207963 Log: This fixes PR-SCTP issues: - Slide the map at the proper place. - Mark the bits in the nr_array ONLY if there is no marking. - When generating a FWD-TSN we allow u

svn commit: r207983 - head/sys/netinet

Author: rrs Date: Wed May 12 18:00:15 2010 New Revision: 207983 URL: http://svn.freebsd.org/changeset/base/207983 Log: More PR-SCTP bugs: - Make sure that when you kick the streams you add correctly using a 16 bit unsigned. - Make sure when sending out you allow FWD-TSN to skip ove

svn commit: r207985 - head/sys/netinet

Author: rrs Date: Wed May 12 18:33:25 2010 New Revision: 207985 URL: http://svn.freebsd.org/changeset/base/207985 Log: Fix an old long time bug in generating a fwd-tsn. This would appear when greater than the size of mbuf TSN's would need to be skipped. MFC after:3 days Modified:

svn commit: r208160 - head/sys/netinet

Author: rrs Date: Sun May 16 17:03:56 2010 New Revision: 208160 URL: http://svn.freebsd.org/changeset/base/208160 Log: This adds back the Iterator to the sctp code base. We now properly have ONE thread that services all VNET's. Also we purge out the old timer based iterator code which had

svn commit: r208165 - in head/sys: kern mips/conf mips/include mips/mips mips/rmi mips/rmi/dev/xlr

Author: rrs Date: Sun May 16 19:43:48 2010 New Revision: 208165 URL: http://svn.freebsd.org/changeset/base/208165 Log: This pushes all of JC's patches that I have in place. I am now able to run 32 cores ok.. but I still will hang on buildworld with a NFS problem. I suspect I am missing a p

svn commit: r208249 - in head/sys/mips: include mips rmi sibyte

Author: rrs Date: Tue May 18 04:02:34 2010 New Revision: 208249 URL: http://svn.freebsd.org/changeset/base/208249 Log: Adds JC's cleanup patches that fix it so we call an platform dependant topo function as well as clean up all the XLR specific ifdefs around smp platform init. Obtaine

svn commit: r208250 - head/sys/mips/rmi

Author: rrs Date: Tue May 18 04:08:58 2010 New Revision: 208250 URL: http://svn.freebsd.org/changeset/base/208250 Log: Adds the file I forgot to add... that handles the mpwait.S for RMI Approved by: JC Added: head/sys/mips/rmi/mpwait.S (contents, props changed) Added: head/sys/mips

svn commit: r208852 - head/sys/netinet

Author: rrs Date: Sat Jun 5 21:17:23 2010 New Revision: 208852 URL: http://svn.freebsd.org/changeset/base/208852 Log: This fixes a bug in the close up of a socket that had un-accepted assoc's. Basically the assoc (and inp) would get stuck and never get cleaned up. MFC after:1 week

svn commit: r208853 - head/sys/netinet

Author: rrs Date: Sat Jun 5 21:20:28 2010 New Revision: 208853 URL: http://svn.freebsd.org/changeset/base/208853 Log: This does two changes: 1) Makes it so that the INVARIANT function validate nolocks is available anywhere. 2) Fixes a BUG where a close has been done on a collision sock

svn commit: r208854 - head/sys/netinet

Author: rrs Date: Sat Jun 5 21:22:58 2010 New Revision: 208854 URL: http://svn.freebsd.org/changeset/base/208854 Log: Use the proper increment macro when increasing the number on sent_queue_retran_cnt. MFC after:1 week Modified: head/sys/netinet/sctp_indata.c Modified: head/sys/n

svn commit: r208855 - head/sys/netinet

Author: rrs Date: Sat Jun 5 21:27:43 2010 New Revision: 208855 URL: http://svn.freebsd.org/changeset/base/208855 Log: This change does the following: 1) Fix the alignment of a comment. 2) Fix a BUG where we were NOT paying attention to the RESEND marking on retransmitting control

svn commit: r208856 - head/sys/netinet

Author: rrs Date: Sat Jun 5 21:33:16 2010 New Revision: 208856 URL: http://svn.freebsd.org/changeset/base/208856 Log: Spacing issues MFC after:1 Week Modified: head/sys/netinet/sctp_auth.c head/sys/netinet/sctp_bsd_addr.c Modified: head/sys/netinet/sctp_auth.c ===

svn commit: r208857 - head/sys/netinet

Author: rrs Date: Sat Jun 5 21:39:52 2010 New Revision: 208857 URL: http://svn.freebsd.org/changeset/base/208857 Log: Purge out a Windows def that somehow slipped past the scrubber. MFC after:1 Week Modified: head/sys/netinet/sctputil.c Modified: head/sys/netinet/sctputil.c =

Re: svn commit: r208857 - head/sys/netinet

en two worlds ;-) But of course we would have to get agreement from quite a few folks... but I am game ;-) R On Jun 5, 2010, at 5:44 PM, Luigi Rizzo wrote: On Sat, Jun 05, 2010 at 09:39:53PM +, Randall Stewart wrote: Author: rrs Date: Sat Jun 5 21:39:52 2010 New Revision: 208857 UR

svn commit: r208863 - head/sys/netinet

Author: rrs Date: Sun Jun 6 02:32:20 2010 New Revision: 208863 URL: http://svn.freebsd.org/changeset/base/208863 Log: Bruce's fix for some return's in error legs. MFC after:1 week Modified: head/sys/netinet/sctp_usrreq.c Modified: head/sys/netinet/sctp_usrreq.c ==

svn commit: r208864 - head/sys/netinet

Author: rrs Date: Sun Jun 6 02:33:46 2010 New Revision: 208864 URL: http://svn.freebsd.org/changeset/base/208864 Log: Hopefully this fixes a LOR by making so we only hold the iterator lock during updates to the iterators work. MFC after:1 week Modified: head/sys/netinet/sctp_pcb

svn commit: r208874 - head/sys/netinet

Author: rrs Date: Sun Jun 6 16:07:40 2010 New Revision: 208874 URL: http://svn.freebsd.org/changeset/base/208874 Log: Fix a bug in sctp_abort_assoc(). DON'T call the sctp_inpcb_free when the gone flag is set. You don't know what locks the caller has set and there is already a kill timer run

svn commit: r208875 - head/sys/netinet

Author: rrs Date: Sun Jun 6 16:09:12 2010 New Revision: 208875 URL: http://svn.freebsd.org/changeset/base/208875 Log: Fix a bug in the sctp_inpcb_free. Basically if the socket was setup to do an abortive close an association that was in the accept_queue could get stuck and never freed. Now

svn commit: r208876 - head/sys/netinet

Author: rrs Date: Sun Jun 6 16:11:16 2010 New Revision: 208876 URL: http://svn.freebsd.org/changeset/base/208876 Log: 1) Further enhance the INVARIANT lock validation (no locks) are held by checking the create and inp locks as well. 2) Fix a bug in that when a socket is closed an INIT

svn commit: r208878 - head/sys/netinet

Author: rrs Date: Sun Jun 6 19:24:32 2010 New Revision: 208878 URL: http://svn.freebsd.org/changeset/base/208878 Log: Ok, yet another bug in killing off all the hundreds of apitesters.. Basically we end up with attempting to destroy a lock thats contended on. A cookie echo arrives at the

svn commit: r208879 - head/sys/netinet

Author: rrs Date: Sun Jun 6 20:34:17 2010 New Revision: 208879 URL: http://svn.freebsd.org/changeset/base/208879 Log: 1) Optimize the cleanup and don't always depend on the timer. This is done by considering the locks we will destroy and if they are contended we consider it the s

svn commit: r208883 - head/sys/netinet

Author: rrs Date: Mon Jun 7 04:01:38 2010 New Revision: 208883 URL: http://svn.freebsd.org/changeset/base/208883 Log: Fix so we call socantrcvmore_locked so we don't see a race where we unlock to call the non-locked version and have the socket go away. MFC after:1 week Modified:

svn commit: r208891 - head/sys/netinet

Author: rrs Date: Mon Jun 7 11:33:20 2010 New Revision: 208891 URL: http://svn.freebsd.org/changeset/base/208891 Log: Opps... my bad.. we don't need a SOCK_UNLOCK() after calling socantrcvmore_locked() since it will unlock the lock for you. MFC after:1 week Modified: head/sys/ne

svn commit: r208897 - head/sys/netinet

Author: rrs Date: Mon Jun 7 18:29:10 2010 New Revision: 208897 URL: http://svn.freebsd.org/changeset/base/208897 Log: This fixes a BUG in the handling of the cum-ack calculation. We were only paying attention to the nr-mapping-array. Which seems to make sense on the surface, by definition t

Re: svn commit: r208876 - head/sys/netinet

Cool... I will move these over to that assert... And the invariant is ok since this function only is there when you compile invariant in. R On Jun 7, 2010, at 1:07 PM, Pawel Jakub Dawidek wrote: On Sun, Jun 06, 2010 at 04:11:17PM +, Randall Stewart wrote: Author: rrs Date: Sun Jun 6 16

svn commit: r208902 - head/sys/netinet

Author: rrs Date: Tue Jun 8 03:39:31 2010 New Revision: 208902 URL: http://svn.freebsd.org/changeset/base/208902 Log: 2 Bugs: 1) Only use both mapping arrays when NR sack is off. This way we can hold off moving the cumack (not the best but workable) when NR-sack is on. 2) We

svn commit: r208952 - head/sys/netinet

Author: rrs Date: Wed Jun 9 16:39:18 2010 New Revision: 208952 URL: http://svn.freebsd.org/changeset/base/208952 Log: BUG:Turns out we need to use both bit maps to calculate the cum-ack (we were not doing it for the NR-Sack case). With this fix NR-sack should now work correctly. MFC aft

svn commit: r208953 - head/sys/netinet

Author: rrs Date: Wed Jun 9 16:42:42 2010 New Revision: 208953 URL: http://svn.freebsd.org/changeset/base/208953 Log: Fix serveral bugs all having to do with freeing an sctp_inpcb: 1) Make sure not to remove the flag on the PCB until after the close() caller is back in control with the

svn commit: r208970 - head/sys/netinet

Author: rrs Date: Wed Jun 9 22:05:29 2010 New Revision: 208970 URL: http://svn.freebsd.org/changeset/base/208970 Log: Found by Michael. In cases where we run out of memory (no more inp space) we don't propely NULL the INP on return. Obtained from:tuexen MFC after:3 Days

svn commit: r209029 - head/sys/netinet

Author: rrs Date: Fri Jun 11 03:54:00 2010 New Revision: 209029 URL: http://svn.freebsd.org/changeset/base/209029 Log: 3 Fixes - a) There was a case where a ICMP message could cause us to return leaving a stuck lock on an stcb. b) The iterator needed some tweaks to fix its lock ord

svn commit: r209644 - head/sys/netinet

Author: rrs Date: Fri Jul 2 09:53:26 2010 New Revision: 209644 URL: http://svn.freebsd.org/changeset/base/209644 Log: Fix a bug that WILL cause a panic. Basically a read-lock is being called to check the vtag-timewait cache. Then in two cases (where a vtag is bad i.e. in the time-wait st

svn commit: r209663 - head/sys/netinet

Author: rrs Date: Sat Jul 3 14:03:31 2010 New Revision: 209663 URL: http://svn.freebsd.org/changeset/base/209663 Log: This fixes a crash in SCTP. It was possible to have a large number of packets queued to a crashing process. In a specific case you may get 2 ABORT's back (from say two pac

svn commit: r209760 - head/lib/libc/net

Author: rrs Date: Wed Jul 7 11:19:06 2010 New Revision: 209760 URL: http://svn.freebsd.org/changeset/base/209760 Log: If a user calls sctp_sendx() with a NULL sinfo we will crash. Instead we should provide our own temp structure to use internally. MFC after:1 month Modified: hea

svn commit: r210493 - head/sys/netinet

Author: rrs Date: Mon Jul 26 09:20:55 2010 New Revision: 210493 URL: http://svn.freebsd.org/changeset/base/210493 Log: When counting the number of chunks in the retransmission queue to validate the retran count, we need to include the chunks in the control send queue too. Otherwise the cou

svn commit: r210494 - head/sys/netinet

Author: rrs Date: Mon Jul 26 09:22:52 2010 New Revision: 210494 URL: http://svn.freebsd.org/changeset/base/210494 Log: Make sure that we report chunks if a socket still exists that were not sent. In either case carefully remove the data if it does not get taken by the reporting routines.

  1   2   3   >