Netgate Discussion Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Search
    • Register
    • Login

    Intel 82574L NIC - Enable segmentation and large receive offload?

    Scheduled Pinned Locked Moved Hardware
    8 Posts 5 Posters 2.3k Views
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • S
      sofakng
      last edited by

      I'm using a Supermicro X7SPA-HF-O motherboard which has 2x Intel 82574L NICs.

      The documentation says it supports checksum offload and TCP segmentation offloading (TSO v2) but I don't see anything about large receive offload support (LRO).

      Should I enable these features in pfsense?  Is there any benefit (or harm?) for a home network?

      1 Reply Last reply Reply Quote 0
      • E
        extide
        last edited by

        I have been running 2 82574's on my pfSense box for a few years now, with all 3 hardware offloads ENABLED. Has been fine!

        1 Reply Last reply Reply Quote 0
        • K
          Keljian
          last edited by

          Try them, run a test? - really on a home network a reboot is not such a big deal.

          1 Reply Last reply Reply Quote 0
          • S
            sofakng
            last edited by

            Thanks for the info.

            I didn't want to try it without knowing because I read it could cause performance problems and I didn't want random or strange problems that might of been hard to track down.

            1 Reply Last reply Reply Quote 0
            • H
              Harvy66
              last edited by

              Shouldn't these features only help when fragmenting or coalescing TCP data packets across different MTU boundaries? If this is the case, then the feature shouldn't help much for Internet related traffic as 1500mtu is extremely common.

              This is kind of a question and statement in one. I'm curious how wrong I may be.

              1 Reply Last reply Reply Quote 0
              • E
                extide
                last edited by

                The TCP checksum offload can actually be reasonably beneficial, especially at higher bandwidths.

                1 Reply Last reply Reply Quote 0
                • R
                  razzfazz
                  last edited by

                  @extide:

                  The TCP checksum offload can actually be reasonably beneficial, especially at higher bandwidths.

                  For endpoints, sure, but how would it help for a router, where traffic is just passing through?

                  EDIT: I guess NAT would in the general case require the TCP checksum to be updated.

                  1 Reply Last reply Reply Quote 0
                  • E
                    extide
                    last edited by

                    TCP Checksum is verified for every packet afaik

                    1 Reply Last reply Reply Quote 0
                    • First post
                      Last post
                    Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.