support for only one ARM CPU. It used to have x86 support, but it was broken
and I removed it. The legacy PMC interface will be removed from the kernel
too. Sent on tech-kern@ yesterday, thorpej was fine.
Because this is in a module, apparently, that means that netinet_in_offload.h
needs to get installed in /usr/include, so do that as well.
Feel free to fix this in a better way...
ip_undefer_csum -> in_undefer_cksum
in_delayed_cksum -> in_undefer_cksum_tcpudp
The two previous names were inconsistent and misleading.
Put the two functions into in_offload.c. Add comments to explain what
we're doing.
The same could be done for IPv6.
on affected type ids and revisions, so it's properly applied whenever
new types are added; taken with FreeBSD
contrary to FreeBSD, keep the IWM workaround for all XL revs - my XL rev. A3
seems to consistently have slightly (~2%) better Rx performance with the 0x04
value
code can't get enough buffers; fixes possible stall when the RXSTAT
interrupt arrives while no buffers are available
idea from OpenBSD if_msk.c rev. 1.131
as part of this remove the confusingly commented re-fill code from
msk_jfree() and instead just invoke the callout if needed; also
refill in msk_intr() only when running low instead of every RX intr;
this simplifies the code, and gives small RX performance boost on
my hardware (6%: ~38 MB/s -> ~40 MB/s)
tested on SK-9E22 (Yukon-2 XL rev. A3) by me, and 88E8071 (Yukon-2
Extreme rev. B0) by Jogn Halfpenny; unfortunately still doesn't
resolve the PR kern/53301 stall
it since we don't read the IPv4 header anymore.
If ip{6}_reass_packet fails, always free 'm', and always clear the nbuf.
We want to avoid the case where
'm' was reallocated
the nbuf pointer was not updated accordingly
the caller tried to use the nbuf pointer
This case doesn't happen right now, but the code is fragile, so strengthen
it.
function. In npf_reassembly, pass a simple boolean instead of a ** mbuf
pointer. Add a KASSERT for IPv4, we don't want (error && !m). Remove
the 'fastout' label, use 'out'.
Before my changes the nonsensical pointer ininitialization held, but
when I started introducing sanity checks the whole thing collapsed.
Need pullup-8.
nonaka