hi, folks
I got bce0 attach returned 6 on my Freebsd 10.1-RELEASE-p8 server. It’s a quad
ports NIC.
any help?
bce0: mem 0xd400-0xd5ff
irq 18 at device 0.0 on pci6
bce0: /usr/src/sys/dev/bce/if_bce.c(1247): Management firmware enabled but not
running!
bce0: /usr/src/sys/dev/bce/if_bce.c(
I'm trying to programmatically create a VLAN in the kernel via ifioctl, but
I'm hitting a "copyin" in the ioctl path, and since the address I'm passing
in is a kernel address and not a user space address, the copyin is failing.
Calling the ioctl from user space is a non-starter at this point, and
On 03/24/15 at 01:51P, Adrian Chadd wrote:
> Hi,
>
> The main reason I didn't add it outside of RSS is that I didn't want
> to impact the behaviour that was there before. Before, it wasn't using
> the flowid - only the msix/queue id.
It'd break things if not all igb chips behave the same way in t
So, the 32 bit field that has the flowid in the rx descriptor can mean
multiple things - not just the flowid.
For igb it can also mean one of the RX checksums.
for ixgbe it can also mean RX checksum, or flowdirector flow
information, or a few other thngs.
-adrian
___
On 03/30/15 at 04:33P, Adrian Chadd wrote:
> So, the 32 bit field that has the flowid in the rx descriptor can mean
> multiple things - not just the flowid.
>
> For igb it can also mean one of the RX checksums.
I care about igb(4) right now and I350 specifically. I checked the spec
and section 7.
No, you're absolutely right. However, you (a) get flowid for L3/L4
V4/v6 but not other things, and (b) I was worried about changing the
default behaviour of the driver, and who knows what other weird crap
people may have done to their local driver. stuff like enabled
checksums instead, or hacked in
