On Sunday 13 Apr 2014 01:01:53 Nikos Chantziaras wrote: > On 12/04/14 18:53, Peter Humphrey wrote: --->8 > > Has anyone got =net-misc/r8168-8.028.00::gentoo to compile? > > [...] > > Can anyone advise me how to proceed here? I've been using the in-kernel > > R8169 module with no problems until I installed Snort, but now I have to > > keep issuing "ethtool -K eth0 gro off" every minute > > I had similar problems until I discovered that the in-kernel R8169 > driver actually tries to load firmware from disk. That firmware wasn't > there. If that is your problem, then this: > > dmesg | grep -i firmware > > would print something like: > > r8169 [...]: unable to load firmware patch rtl_nic/rtl8168e-3.fw
Here's a snippet from dmesg: [ 0.053557] ACPI: bus type PCI registered [ 0.054231] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem 0xe0000000-0xefffffff] (base 0xe0000000) [ 0.054475] PCI: MMCONFIG at [mem 0xe0000000-0xefffffff] reserved in E820 [ 0.055024] PCI: Using MMCONFIG for extended config space [ 0.055196] PCI: Using configuration type 1 for base access [ 0.082221] bio: create slab <bio-0> at 0 [ 0.083189] ACPI: Added _OSI(Module Device) [ 0.083367] ACPI: Added _OSI(Processor Device) [ 0.083531] ACPI: Added _OSI(3.0 _SCP Extensions) [ 0.084031] ACPI: Added _OSI(Processor Aggregator Device) [ 0.087597] ACPI: EC: Look up EC in DSDT [ 0.092258] ACPI: Executed 1 blocks of module-level executable AML code [ 0.103159] [Firmware Bug]: ACPI: BIOS _OSI(Linux) query ignored and another: [ 5.232622] r8169 Gigabit Ethernet driver 2.3LK-NAPI loaded [ 5.233183] r8169 0000:01:00.0: irq 45 for MSI/MSI-X [ 5.233963] r8169 0000:01:00.0 eth0: RTL8168d/8111d at 0xf8170000, 70:71:bc:94:ee:71, XID 081000c0 IRQ 45 [ 5.233988] r8169 0000:01:00.0 eth0: jumbo features [frames: 9200 bytes, tx checksumming: ko] [ 5.291801] intel_rng: Firmware space is locked read-only. If you can't or intel_rng: don't want to disable this in firmware setup, and if intel_rng: you are certain that your system has a functional intel_rng: RNG, try using the 'no_fwh_detect' option. I suppose I can assume a functional RNG - or can I? I'll try adding intel_rng=no_fwh_detect to my kernel command line next time I reboot. Later: I've booted with that intel_rng command and I'm watching to see how it goes. > Solution to the problem was fairly simple. I just emerged this package: > > sys-kernel/linux-firmware Yes, I have that package installed, and the network has been as solid as yours. I just have this problem with the card insisting on doing generic receive offload and thereby confusing Snort. Later still: no improvement. So thank you all for your help, but pro tem I'm going to abandon Snort as being just too much trouble. Maybe an outfit like Alan McK's could afford the time but it's too heavy for home use. Pity. I'll have to look around for a lighter IDS. -- Regards Peter
