if_re driver support for new hardware rev.

Hello forum.

I just purchased a Biostar N3150NH motherboard to replace an old Intel Atom-based board that was running as a small FreeBSD server. It is equipped with an onboard RealTek RTL8111H gigabit ethernet controller,

Disappointingly, the RealTek NIC doesn't work with 10.2-RELEASE. The kernel sees it during boot, but throws an error:

Code:
re0: <RealTek 8168/8111 B/C/CP/D/DP/E/F/G PCIe Gigabit Ethernet> port
 0xe000-0xe0ff mem 0x89204000-0x89204fff,0x89200000-0x89203fff irq 17
 at device 0.0 on pci2
re0: Using 1 MSI-X message
re0: turning off MSI enable bit.
re0: Chip rev. 0x54000000
re0: MAC rev. 0x00100000
re0: Unknown H/W revision: 0x54000000
device_attach: re0 attach returned 6

The re0 interface then doesn't show up in ifconfig(8), and pciconf(8) lists it as:

Code:
none2@pci0:2:0:0:  class=0x020000 card=0x23121565 chip=0x816810ec rev=0x15 hdr=0x00
    vendor     = 'Realtek Semiconductor Co., Ltd.'
    device     = 'RTL8111/8168B PCI Express Gigabit Ethernet controller'
    class      = network
    subclass   = ethernet

From looking through the 10.2 kernel source, it appears that only RTL8111 chips up to revision G are supported.

I had a look through the HEAD and STABLE branches in SVN and can see that preliminary support was added for the RTL8111H to HEAD, but isn't in STABLE yet. What sort of timescales are typically involved in changes to HEAD being merged to STABLE, then to RELENG? Is there any testing work that users can do to help increase the likelihood of this?

Thanks.
JM
 
There are very rarely merges from HEAD or STABLE to RELENG because releases are supposed to be feature locked. The standard response in this case is to wait for the merge to STABLE and then use STABLE instead of the RELENG branch.
 
  • Thanks
Reactions: jem
Once it's been added to 10.2-STABLE it'll be included in the 10.3-RELEASE (release date early next year).
 
  • Thanks
Reactions: jem
Back
Top