Page MenuHomeFreeBSD
Feed Advanced Search

Today

tuexen committed rS352438: Only allow a SCTP-AUTH shared key to be updated by the application.
Only allow a SCTP-AUTH shared key to be updated by the application
Tue, Sep 17, 9:46 AM

Yesterday

tuexen committed rS352386: Don't write to memory outside of the allocated array for SACK blocks..
Don't write to memory outside of the allocated array for SACK blocks.
Mon, Sep 16, 8:18 AM

Sun, Sep 15

tuexen committed rS352361: When the IP layer calls back into the SCTP layer to perform the SCTP.
When the IP layer calls back into the SCTP layer to perform the SCTP
Sun, Sep 15, 6:30 PM
tuexen added inline comments to D21664: Fix handling of Hop-by-Hop options over the loopback interface.
Sun, Sep 15, 6:03 PM
tuexen created D21666: Let the RACK stack take IP options into account when computing the space for user data in TCP segments.
Sun, Sep 15, 5:53 PM
tuexen created D21665: Don't send TCP segments when the IP header chain and the TCP header doesn't fit in a packet.
Sun, Sep 15, 3:36 PM
tuexen created D21664: Fix handling of Hop-by-Hop options over the loopback interface.
Sun, Sep 15, 2:37 PM

Fri, Sep 13

tuexen committed rS352284: MFC r352072:.
MFC r352072:
Fri, Sep 13, 8:15 AM

Thu, Sep 12

tuexen added a reviewer for D21616: Replace MD5 by SipHash as TCP keyed hash function: rscheff_gmx.at.
Thu, Sep 12, 2:29 PM
tuexen created D21616: Replace MD5 by SipHash as TCP keyed hash function.
Thu, Sep 12, 2:03 PM

Tue, Sep 10

tuexen accepted D21591: avoid unneeded call to arc4random() in syncache_add().
Tue, Sep 10, 9:12 PM

Mon, Sep 9

tuexen closed D21567: Improve DSACK computation.
Mon, Sep 9, 4:08 PM
tuexen committed rS352072: Only update SACK/DSACK lists when a non-empty segment was received..
Only update SACK/DSACK lists when a non-empty segment was received.
Mon, Sep 9, 4:08 PM
tuexen added a comment to D21567: Improve DSACK computation.

Checked this patch with various "pure FIN after hole" and "data FIN after hole" combinations; the FIN bit becomes part of the SACK (and DSACK) blocks, as it should be when considering that FIN is part of the sequence number space.

Mon, Sep 9, 2:25 PM
tuexen added a comment to D21567: Improve DSACK computation.

The patch has been updated to address Richard's comments.

Mon, Sep 9, 11:27 AM
tuexen updated the diff for D21567: Improve DSACK computation.

Address Richard's comments.

Mon, Sep 9, 11:26 AM
tuexen committed rS352057: MFC r349986:.
MFC r349986:
Mon, Sep 9, 11:20 AM
tuexen committed rS352055: The variable names in the description of the port number usage is.
The variable names in the description of the port number usage is
Mon, Sep 9, 10:45 AM
tuexen added a comment to D21567: Improve DSACK computation.
In D21567#469962, @rrs wrote:

I would like to see the Rack version of the code be identical to the tcp_input version.. i.e. it should check
if ((tlen > 0) && (tp->t_flags & TF_SACK_PERMITED))
The soon to be committed version of rack will allow a user to use the stack without sack.. so it is best to
put this in place now.

Mon, Sep 9, 9:55 AM
tuexen updated the diff for D21567: Improve DSACK computation.

Add a check to the RACK code whether the SACK extension is enabled. This was requested by rrs@.

Mon, Sep 9, 9:54 AM

Sun, Sep 8

tuexen created D21567: Improve DSACK computation.
Sun, Sep 8, 8:31 PM

Sat, Sep 7

tuexen committed rS352022: Improve DSACK support:.
Improve DSACK support:
Sat, Sep 7, 7:58 PM
tuexen committed rS352008: MFC r351655:.
MFC r351655:
Sat, Sep 7, 12:20 PM
tuexen committed rS352007: MFC r351654:.
MFC r351654:
Sat, Sep 7, 12:19 PM
tuexen committed rS352006: MFC r351641:.
MFC r351641:
Sat, Sep 7, 12:15 PM
tuexen committed rS352005: MFC r351638:.
MFC r351638:
Sat, Sep 7, 12:14 PM
tuexen committed rS352004: MFC r351511:.
MFC r351511:
Sat, Sep 7, 12:03 PM
tuexen committed rS352003: MFC r350745:.
MFC r350745:
Sat, Sep 7, 12:01 PM
tuexen committed rS352002: MFC r350627:.
MFC r350627:
Sat, Sep 7, 12:00 PM
tuexen committed rS352001: MFC r350626:.
MFC r350626:
Sat, Sep 7, 11:58 AM
tuexen committed rS352000: MFC r350625:.
MFC r350625:
Sat, Sep 7, 11:56 AM
tuexen committed rS351999: MFC r350520:.
MFC r350520:
Sat, Sep 7, 11:52 AM
tuexen committed rS351998: MFC r350508:.
MFC r350508:
Sat, Sep 7, 11:51 AM
tuexen committed rS351997: MFC r350488:.
MFC r350488:
Sat, Sep 7, 11:49 AM
tuexen committed rS351996: MFC r350487:.
MFC r350487:
Sat, Sep 7, 11:48 AM
tuexen committed rS351995: MFC r350404:.
MFC r350404:
Sat, Sep 7, 11:47 AM
tuexen committed rS351994: MFC r350403:.
MFC r350403:
Sat, Sep 7, 11:33 AM
tuexen committed rS351993: MFC r350265:.
MFC r350265:
Sat, Sep 7, 11:31 AM
tuexen committed rS351992: MFC r350254:.
MFC r350254:
Sat, Sep 7, 11:24 AM
tuexen committed rS351991: MFC r350248:.
MFC r350248:
Sat, Sep 7, 11:22 AM
tuexen committed rS351979: MFC r350216:.
MFC r350216:
Sat, Sep 7, 11:20 AM
tuexen committed rS351978: MFC 350028:.
MFC 350028:
Sat, Sep 7, 10:57 AM
tuexen committed rS351977: MFC 350027:.
MFC 350027:
Sat, Sep 7, 10:56 AM
tuexen committed rS351976: MFC r350026:.
MFC r350026:
Sat, Sep 7, 10:55 AM
tuexen committed rS351975: MFC r349999:.
MFC r349999:
Sat, Sep 7, 10:53 AM
tuexen committed rS351974: MFC r349998:.
MFC r349998:
Sat, Sep 7, 10:51 AM
tuexen committed rS351973: MFC r349989:.
MFC r349989:
Sat, Sep 7, 10:49 AM
tuexen committed rS351972: MFC r349986:.
MFC r349986:
Sat, Sep 7, 10:47 AM
tuexen committed rS351971: MFC r349968:.
MFC r349968:
Sat, Sep 7, 10:46 AM
tuexen committed rS351970: MFC r349228:.
MFC r349228:
Sat, Sep 7, 10:40 AM

Wed, Sep 4

tuexen committed rS351801: Fix the SACK block generation in the base TCP stack by bringing it in.
Fix the SACK block generation in the base TCP stack by bringing it in
Wed, Sep 4, 4:38 AM
tuexen closed D21513: Fix TCP SACK block generation in the base stack.
Wed, Sep 4, 4:38 AM

Tue, Sep 3

tuexen created D21513: Fix TCP SACK block generation in the base stack.
Tue, Sep 3, 8:30 PM
tuexen closed D21512: Fix two TCP Rack issues.
Tue, Sep 3, 7:48 PM
tuexen committed rS351782: Fix two TCP RACK issues:.
Fix two TCP RACK issues:
Tue, Sep 3, 7:48 PM
tuexen created D21512: Fix two TCP Rack issues.
Tue, Sep 3, 7:03 PM

Mon, Sep 2

tuexen closed D21038: DSACK patch to conform fully with RFC2883.
Mon, Sep 2, 7:04 PM
tuexen committed rS351725: This patch improves the DSACK handling to conform with RFC 2883..
This patch improves the DSACK handling to conform with RFC 2883.
Mon, Sep 2, 7:04 PM
tuexen accepted D21038: DSACK patch to conform fully with RFC2883.
Mon, Sep 2, 6:44 PM

Sun, Sep 1

tuexen committed rS351655: Fix initialization of top_fsn..
Fix initialization of top_fsn.
Sun, Sep 1, 10:39 AM
tuexen committed rS351654: Improve the handling of state cookie parameters in INIT-ACK chunks..
Improve the handling of state cookie parameters in INIT-ACK chunks.
Sun, Sep 1, 10:10 AM

Sat, Aug 31

tuexen committed rS351641: Improve function definition..
Improve function definition.
Sat, Aug 31, 1:13 PM
tuexen committed rS351638: Improve the handling of illegal sequence number combinations in received.
Improve the handling of illegal sequence number combinations in received
Sat, Aug 31, 8:19 AM

Mon, Aug 26

tuexen added a comment to D20347: efirt: Add efi_memory_attribute.

Is it possible to get this in head? It seems to be needed to run FreeBSD on Ampere systems...

Mon, Aug 26, 6:21 PM
tuexen closed D21416: Improve locking in tcp_ratelimit.c.
Mon, Aug 26, 4:24 PM
tuexen committed rS351512: Don't hold the rs_mtx lock while calling malloc()..
Don't hold the rs_mtx lock while calling malloc().
Mon, Aug 26, 4:24 PM
tuexen committed rS351511: Identify eMAG CPU used in Ampere Computing systems..
Identify eMAG CPU used in Ampere Computing systems.
Mon, Aug 26, 4:12 PM
tuexen closed D21314: Identify emag 8180 CPU.
Mon, Aug 26, 4:12 PM
tuexen created D21416: Improve locking in tcp_ratelimit.c.
Mon, Aug 26, 1:47 PM
tuexen added a comment to D21314: Identify emag 8180 CPU.

I tried a similar version of this change and my eMAG reports CPU 0: APM eMAG 8180 r3p2 affinity: 0 0
I used 'ampere' in the struct name and comments though.

I think not reporting APM and Ampere improves consistency. So I changed that. However, I left the apm in the structure names since it corresponds with the APM in the constant for the implementer constant and the implementer name. However, I added a comment to provide a hint that these CPUs are now managed by Ampere Computing.

Mon, Aug 26, 1:21 PM
D21314: Identify emag 8180 CPU now requires review to proceed.

Don't report APM and Ampere to be consistent with other entries. Also add comments to note that the CPU comes from Ampere even though using constants from Applied Micro.

Mon, Aug 26, 1:18 PM

Aug 18 2019

tuexen created D21314: Identify emag 8180 CPU.
Aug 18 2019, 9:35 AM

Aug 8 2019

tuexen committed rS350745: Fix a typo..
Fix a typo.
Aug 8 2019, 8:24 AM

Aug 6 2019

tuexen committed rS350627: Reference the RFC instead of the internet draft..
Reference the RFC instead of the internet draft.
Aug 6 2019, 11:26 AM
tuexen committed rS350626: Fix a locking issue in sctp_accept..
Fix a locking issue in sctp_accept.
Aug 6 2019, 10:29 AM
tuexen committed rS350625: Fix build issues for the userland stack on Raspbian..
Fix build issues for the userland stack on Raspbian.
Aug 6 2019, 8:33 AM

Aug 5 2019

tuexen committed rS350588: Improve consistency. No functional change..
Improve consistency. No functional change.
Aug 5 2019, 1:22 PM

Aug 1 2019

tuexen committed rS350520: Fix the reporting of multiple unknown parameters in an received INIT.
Fix the reporting of multiple unknown parameters in an received INIT
Aug 1 2019, 7:45 PM
tuexen committed rS350508: When responding with an ABORT to an INIT chunk containing a.
When responding with an ABORT to an INIT chunk containing a
Aug 1 2019, 5:36 PM
tuexen added inline comments to D21038: DSACK patch to conform fully with RFC2883.
Aug 1 2019, 2:01 PM

Jul 31 2019

tuexen committed rS350488: Small cleanup, no functional change intended..
Small cleanup, no functional change intended.
Jul 31 2019, 9:39 PM
tuexen committed rS350487: Consistently cleanup mbufs in case of other memory errors..
Consistently cleanup mbufs in case of other memory errors.
Jul 31 2019, 9:29 PM

Jul 29 2019

tuexen closed D20550: Fix bugs in DCTCP on TCP events.
Jul 29 2019, 9:20 AM
tuexen committed rS350404: When performing after_idle() or post_recovery(), don't disable the.
When performing after_idle() or post_recovery(), don't disable the
Jul 29 2019, 9:20 AM
tuexen added inline comments to D20550: Fix bugs in DCTCP on TCP events.
Jul 29 2019, 9:19 AM
D20550: Fix bugs in DCTCP on TCP events is now accepted and ready to land.
Jul 29 2019, 9:08 AM
tuexen committed rS350403: * Improve input validation of sysctl parameters for DCTPC..
* Improve input validation of sysctl parameters for DCTPC.
Jul 29 2019, 8:50 AM
Herald added a reviewer for D20549: fixing sysctl interface for DCTCP and types of dctcp variables: manpages.
Jul 29 2019, 8:50 AM

Jul 28 2019

tuexen added inline comments to D20549: fixing sysctl interface for DCTCP and types of dctcp variables.
Jul 28 2019, 6:15 PM
tuexen added a comment to D20549: fixing sysctl interface for DCTCP and types of dctcp variables.

Addressed all comments so far.
Question around the default for alpha still remains; the current default renders dctcp flows very unreactive on CE marks during the initial few windows. Linux is using MAX_ALPHA_VALUE as the inital, which IMHO would be the conservative approach...

Jul 28 2019, 6:14 PM
tuexen added inline comments to D20549: fixing sysctl interface for DCTCP and types of dctcp variables.
Jul 28 2019, 3:17 PM
tuexen added inline comments to D20549: fixing sysctl interface for DCTCP and types of dctcp variables.
Jul 28 2019, 2:30 PM

Jul 27 2019

tuexen added a reviewer for D20549: fixing sysctl interface for DCTCP and types of dctcp variables: bcr.

Add Benedict for man page change.

Jul 27 2019, 7:25 PM
tuexen added inline comments to D20549: fixing sysctl interface for DCTCP and types of dctcp variables.
Jul 27 2019, 7:24 PM
tuexen added a reviewer for D19329: address possible illegal SACK block after FIN when client keeps sending data: tuexen.
Jul 27 2019, 2:22 AM
tuexen added a comment to D20549: fixing sysctl interface for DCTCP and types of dctcp variables.

Please update also share/man/man4/cc_dctcp.4 to describe the valid ranges of the sysctl variables and the defaults. Please fix inconsistencies like the default for dctcp_shift_g given there and in the code.

Jul 27 2019, 1:30 AM

Jul 23 2019

tuexen added inline comments to D21038: DSACK patch to conform fully with RFC2883.
Jul 23 2019, 9:40 PM
tuexen added a comment to D21038: DSACK patch to conform fully with RFC2883.

The patch doesn't apply, since the indentation is based on spaces instead of tabs. Please fix.

Jul 23 2019, 9:34 PM
tuexen committed rS350265: Add a sysctl variable ts_offset_per_conn to change the computation.
Add a sysctl variable ts_offset_per_conn to change the computation
Jul 23 2019, 9:28 PM
tuexen closed D20980: Add sysctl to use per end point pair TCP timestamp offsets instead of per connection based offsets .
Jul 23 2019, 9:28 PM