In D11370#249578, @mjoras wrote:Great. For my own curiosity, are your direct vlan handling patches posted anywhere?
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
Feed Advanced Search
Advanced Search
Advanced Search
Aug 16 2017
Aug 16 2017
Aug 15 2017
Aug 15 2017
I also did the same test with vlans created on top of lagg with 2x25G mellanox adapters.
I didn't see measurable performance drop there. It is able to forward 14Mpps with our RX direct vlan handling patch.
I have tested your patch in our test environment against forwarding performance.
[packet generator] -> [ switch ] -> [ix.10 -> ix.100]
Aug 9 2017
Aug 9 2017
Make user supplied data checks a bit stricter.
Add to if_enc(4) ability to capture packets via BPF after pfil processing.
Aug 3 2017
Aug 3 2017
I proposed this patch for the discussed problem:
https://lists.freebsd.org/pipermail/freebsd-net/2016-December/046650.html
Aug 1 2017
Aug 1 2017
What it we replace IPSEC with IPSEC_SUPPORT?
Then the profit will have those
- who doesn't use IPsec will not rebuild the kernel to avoid overhead
- who want to use IPsec, they can do kldload ipsec
- who want to use TCP-MD5, they can do kldload tcpmd5
Jul 31 2017
Jul 31 2017
Add inpcb pointer to struct ipsec_ctx_data and pass it to the pfil hook
Jul 26 2017
Jul 26 2017
Jul 25 2017
Jul 25 2017
Jul 21 2017
Jul 21 2017
Jul 19 2017
Jul 19 2017
Add HPE FlexFabric 10Gb 4-port 536FLR-T device id to the bxe(4) driver.
Jul 6 2017
Jul 6 2017
From a quick look, the iflib code does not bind irq to CPU cores. The old em/igb drivers did that and I guess, if you add bus_bind_intr() again, this will increase the performance.
The forwarding test is not rely on the described feature. It uses small packets that are each fits into one mbuf and there is no need to collapse or defrag them.
Jul 4 2017
Jul 4 2017
Please, describe in commit message your calculations.
Jul 3 2017
Jul 3 2017
Merge from stable/11 r320593:
Jun 30 2017
Jun 30 2017
In D11370#236225, @mav wrote:In D11370#236121, @matt.joras_gmail.com wrote:The reasoning for keeping the counter increment under the lock is to protect against the possibility of the vlan ifnet being freed while we are touching the counter, since the vlan ifnet can't be freed while we still have the read lock.
It looks odd to me that network stack does not protect against this. What if interface decide to go away earlier, just after entering vlan_transmit? I have subtle feeling that this may only hide the problem.
Jun 29 2017
Jun 29 2017
Fix IPv6 extension header parsing. The length field doesn't include the
Jun 20 2017
Jun 20 2017
Jun 13 2017
Jun 13 2017
Resurrect RTF_RNH_LOCKED flag and restore ability to call rtalloc1_fib()
Jun 12 2017
Jun 12 2017
ae closed D9352: merge projects/ipsec into head/ by committing rS313330: Merge projects/ipsec into head/..
Merge projects/ipsec into head/.
Jun 5 2017
Jun 5 2017
Jun 2 2017
Jun 2 2017
Jun 1 2017
Jun 1 2017
Build kdebug_secreplay() function only when IPSEC_DEBUG is defined.
May 30 2017
May 30 2017
May 29 2017
May 29 2017
Disable IPsec debugging code by default when IPSEC_DEBUG kernel option
May 24 2017
May 24 2017
May 23 2017
May 23 2017
Fix possible double releasing for SA and SP references.
Fix possible double releasing for SA reference.
May 17 2017
May 17 2017
Allow zero port specification in table entries with type flow.
Set M_BCAST and M_MCAST flags on mbuf sent via divert socket.
May 10 2017
May 10 2017
May 3 2017
May 3 2017
May 2 2017
May 2 2017
Add `ipfw table all destroy` support.
In parse_range() validate both range values instead of checking
Add sets support for ipfw table info/list/flush commands.
Properly initialize ipfw_range_tlv variable to fix possible EINVAL
Apr 28 2017
Apr 28 2017
In D10533#218288, @rgrimes wrote:Quick glance for 2 minutes says we should probably have a man page change with this?
Make refragmentation to be working.
Apr 27 2017
Apr 27 2017
Apr 26 2017
Apr 26 2017
Apr 25 2017
Apr 25 2017
Apr 21 2017
Apr 21 2017
Apr 18 2017
Apr 18 2017
Apr 17 2017
Apr 17 2017
Apr 14 2017
Apr 14 2017
Use address of specific union member instead of whole union address to
The rule field in the ipfw_dyn_rule structure is used as storage
Rework r316770 to make it protocol independent and general, like we
Apr 13 2017
Apr 13 2017
I'm not glebius@, but the patch looks correct.
Clear h/w csum flags on mbuf handled by UDP.
Add large replay widow support to setkey(8) and libipsec.
Apr 12 2017
Apr 12 2017
Apr 11 2017
Apr 11 2017
Inherit IPv6 checksum offloading flags to vlan interfaces.
Make sysctl identifiers for direct netisr queue unique.
Do not adjust interface MTU automatically. Leave this task to the system
Apr 9 2017
Apr 9 2017
Looks good to me. Also I want to note, that all
do read(); } while()
loops are affected to the problem described in rS303374. It would be nice to fix this problem too. :)
Apr 7 2017
Apr 7 2017