• Learned Something New About Ethernet Today

    From Lawrence D'Oliveiro@ldo@nz.invalid to comp.misc on Fri Jul 18 06:40:22 2025
    From Newsgroup: comp.misc

    Today I discovered that there are Ethernet interfaces that are not backward-compatible with older, slower speeds. Specifically, anything
    using the Intel X722 controller chip can do 1-gigabit and 10-gigabit, but
    not 100 megabits or 10 megabits per second. Connect it to a 10/100 switch
    port and ... silence.
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From not@not@telling.you.invalid (Computer Nerd Kev) to comp.misc on Sat Jul 19 10:55:05 2025
    From Newsgroup: comp.misc

    Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
    Today I discovered that there are Ethernet interfaces that are not backward-compatible with older, slower speeds. Specifically, anything
    using the Intel X722 controller chip can do 1-gigabit and 10-gigabit, but not 100 megabits or 10 megabits per second. Connect it to a 10/100 switch port and ... silence.

    A quick search suggests that controller is targeted at servers.
    Does it connect to any physical ports with connectors that are
    used for 10/100 Ethernet? I'd expect them only to be used with
    the higher speed Ethernet connectors used on servers, which I
    don't know of being used for 10/100 interfaces. Such servers
    might use a different controller chipset for any RJ45 Ethernet
    sockets.
    --
    __ __
    #_ < |\| |< _#
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Theo@theom+news@chiark.greenend.org.uk to comp.misc on Sat Jul 19 22:55:12 2025
    From Newsgroup: comp.misc

    Computer Nerd Kev <not@telling.you.invalid> wrote:
    Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
    Today I discovered that there are Ethernet interfaces that are not backward-compatible with older, slower speeds. Specifically, anything using the Intel X722 controller chip can do 1-gigabit and 10-gigabit, but not 100 megabits or 10 megabits per second. Connect it to a 10/100 switch port and ... silence.

    Yes, it simplifies the silicon design to throw out the 10 and 100Mbps parts which are likely never going to be used. The simplest design is single rate (10G) but 1G/10G dual rate is better than quad rate 10M/100M/1G/10G.

    A quick search suggests that controller is targeted at servers.
    Does it connect to any physical ports with connectors that are
    used for 10/100 Ethernet? I'd expect them only to be used with
    the higher speed Ethernet connectors used on servers, which I
    don't know of being used for 10/100 interfaces. Such servers
    might use a different controller chipset for any RJ45 Ethernet
    sockets.

    X722 appears to be targeting SFP+, so it's designed for optical transceivers
    or direct attach cables. It seems there are some versions that support
    BASE-T copper (RJ45 connectors) but almost mobody wants to hook those up to
    a 10/100 network.

    Theo
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Lawrence D'Oliveiro@ldo@nz.invalid to comp.misc on Sun Jul 20 00:24:20 2025
    From Newsgroup: comp.misc

    On 19 Jul 2025 10:55:05 +1000, Computer Nerd Kev wrote:

    A quick search suggests that controller is targeted at servers.

    Specifically in this case, a Lenovo server. Lots of CPU cores, lots of
    RAM, lots of network interfaces, lots of disks, lots of fans ...

    Only, Lenovo insists on making firmware updaters available only through
    stupid .exe files that only run under Windows. So we had to do a temporary Windows Server installation just to try applying that firmware update,
    only for it to fail for some reason we couldn’t fathom. But the option to extract the firmware to a .rom file worked.

    So we saved that to a USB stick, and then, we rebooted into the UEFI
    monitor, and it was able to load and apply that .rom file just fine.

    Why couldn’t Lenovo just make the firmware available in a simple .zip archive to begin with? That would have saved so much trouble. Not to
    mention, not encouraging the habit among users of downloading and running random .exe files from random websites ...
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From kludge@kludge@panix.com (Scott Dorsey) to comp.misc on Sat Jul 19 20:31:06 2025
    From Newsgroup: comp.misc

    Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
    Only, Lenovo insists on making firmware updaters available only through >stupid .exe files that only run under Windows. So we had to do a temporary >Windows Server installation just to try applying that firmware update,
    only for it to fail for some reason we couldn’t fathom. But the option to >extract the firmware to a .rom file worked.

    Do they have an MS-DOS or FreeDOS option available? That's what I
    use to run the HP server upgraderizers.
    --scott
    --
    "C'est un Nagra. C'est suisse, et tres, tres precis."
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Lawrence D'Oliveiro@ldo@nz.invalid to comp.misc on Sun Jul 20 03:10:31 2025
    From Newsgroup: comp.misc

    On Sat, 19 Jul 2025 20:31:06 -0400 (EDT), Scott Dorsey wrote:

    Do they have an MS-DOS or FreeDOS option available?

    I specifically suggested trying FreeDOS, as I understand that is a common thing to bundle with server hardware in lieu of a full-fat, full-price OS.

    Sadly, I’m pretty sure the executable was Windows only. It had a GUI, for one thing (not much of one, but still). I don’t think you can build
    Windows executables that can also work as DOS command-line-only
    executables ...
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Kerr-Mudd, John@admin@127.0.0.1 to comp.misc,alt.folklore.computers on Sun Jul 20 10:01:15 2025
    From Newsgroup: comp.misc

    On Sun, 20 Jul 2025 03:10:31 -0000 (UTC)
    Lawrence D'Oliveiro <ldo@nz.invalid> wrote:

    On Sat, 19 Jul 2025 20:31:06 -0400 (EDT), Scott Dorsey wrote:

    Do they have an MS-DOS or FreeDOS option available?

    I specifically suggested trying FreeDOS, as I understand that is a common thing to bundle with server hardware in lieu of a full-fat, full-price OS.

    Sadly, I’m pretty sure the executable was Windows only. It had a GUI, for one thing (not much of one, but still). I don’t think you can build Windows executables that can also work as DOS command-line-only
    executables ...

    There is/was a guy over in erm some other group alt.os.development ? who claim(s/ed) to have xcompiled stuff.

    DOS TUI was a thing for a bit - prior to Borland and Norton IBM went
    overboard at the start of OS/2 with a set of "standards" - Common User Interface? I think the only product that implemented it was PASF/PC.

    xpost to afc for better memories
    --
    Bah, and indeed Humbug.
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Oregonian Haruspex@no_email@invalid.invalid to comp.misc on Mon Jul 28 15:01:22 2025
    From Newsgroup: comp.misc

    Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
    On 19 Jul 2025 10:55:05 +1000, Computer Nerd Kev wrote:

    A quick search suggests that controller is targeted at servers.

    Specifically in this case, a Lenovo server. Lots of CPU cores, lots of
    RAM, lots of network interfaces, lots of disks, lots of fans ...

    Only, Lenovo insists on making firmware updaters available only through stupid .exe files that only run under Windows. So we had to do a temporary Windows Server installation just to try applying that firmware update,
    only for it to fail for some reason we couldn’t fathom. But the option to extract the firmware to a .rom file worked.

    So we saved that to a USB stick, and then, we rebooted into the UEFI monitor, and it was able to load and apply that .rom file just fine.

    Why couldn’t Lenovo just make the firmware available in a simple .zip archive to begin with? That would have saved so much trouble. Not to mention, not encouraging the habit among users of downloading and running random .exe files from random websites ...


    Lenovo also pushes firmware updates through that buggy Linux program that people use too. fwupd or something it’s called. I run the fish now.

    --- Synchronet 3.21a-Linux NewsLink 1.2