Forum Replies Created

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • in reply to: Kernel Logging #14494
    iricah21
    Participant

    I’d be interested in this too. Even basic kernel-level logging would make troubleshooting a lot easier, especially when trying to figure out what’s happening without constantly attaching a debugger. If the source code is available under a suitable license, adding logging with printk() or the kernel’s pr_debug() macros should be possible, depending on how the module is structured. The main thing would be making the logs configurable so they don’t flood the kernel log during normal use!

    • This reply was modified 1 month, 2 weeks ago by iricah21.
    in reply to: VPM #14492
    iricah21
    Participant

    I noticed that too. I assumed “VPM” might just be a typo since the rest of the site consistently uses VPN. It would probably be worth correcting the button text just to avoid confusing someone who is downloading it for the first time. And yeah, the new site does look pretty nice!

    • This reply was modified 1 month, 2 weeks ago by iricah21.
    in reply to: shadowsocks support #14491
    iricah21
    Participant

    I agree with this. The UDP side of SOCKS5 is definitely the frustrating part, especially when you’re trying to find a public proxy that actually supports it properly. Localhost SOCKS support would be useful too, since it would make testing with a local proxy much easier.

    Having Shadowsocks support as well would give a lot more flexibility.

    in reply to: Speed issue #14490
    iricah21
    Participant

    If the speed drops immediately after installing the driver, even with no sample running, I’d first compare the NIC settings before and after installation. Windows Packet Filter works at the NDIS level, so the driver can affect things like NIC offloading even when nothing is actively filtering packets.

    I’d especially check RSS, Large Send Offload and the adapter’s MTU/Jumbo Frame settings, then run the same iperf3 test after uninstalling the filter to confirm the difference.

    The 10 Gb → 2.5 Gb drop is pretty large, so I’d also be interested in which NIC/driver you’re using and whether this is on physical hardware or a VM.
    If it’s a VM, I’d test the same package on a physical machine as well, virtual NIC implementations can behave quite differently with NDIS filters.

Viewing 4 posts - 1 through 4 (of 4 total)