ixgbe: Apply DMAC changes through an if-up reset request
Serialize cached DMA coalescing policy with initialization and schedule
its application through the admin task. The old running check
preceded if_init() acquiring the context lock, so an intervening down
operation could be followed by an unconditional initialization.
Use the deferred if-up request so restart permission is checked when the
task runs. Changes made while stopped or suspended remain cached for
the next initialization. Do not schedule a reset for an unchanged value.
MFC after: 2 weeks
Sponsored by: BBOX.io
enic: Report device enable failures to iflib
Do not publish a running interface after vnic_dev_enable_wait() fails.
Run the existing stop cleanup for the queues configured before the enable
request, then report initialization failure to iflib.
Mark those queues as needing cleanup before submitting the enable request
so the stopped-state shortcut cannot skip the unwind.
MFC after: 2 weeks
Sponsored by: BBOX.io
enic: Clean the completion ring matching each queue
The transmit stop loop cleaned the receive completion ring, and the
receive stop loop cleaned the transmit completion ring. Use enic_cq_wq()
for transmit queues and enic_cq_rq() for receive queues, matching the
queue indices used by the remaining ring bookkeeping.
MFC after: 2 weeks
Sponsored by: BBOX.io
math/gap: update 4.14.0 → 4.16.1
30+ port options were added for all compiled packages.
This is in addition to non-compiled packages that come with
the base GAP.
ChangeLog: https://github.com/gap-system/gap/blob/v4.16.1/CHANGES.md
ice: Honor iflib transmit completion batching
ice marked every transmitted packet RS and recorded every last
descriptor in its report-status queue. Hardware therefore wrote
descriptor status for every packet, and the driver traversed every
packet while reclaiming completed descriptors.
iflib marks selected packets with IPI_TX_INTR as completion
checkpoints. It forces a checkpoint as deferred work or ring pressure
grows. Retain EOP on every packet, but set RS and record the descriptor
only at those checkpoints.
DPDK uses the same sparse-RS design and defaults tx_rs_thresh to 32.
Let iflib choose the adaptive interval for FreeBSD. This reduces PCIe
and memory traffic while preserving bounded descriptor reclamation.
Validated on an E810-XXV in an A-B-A test with five matched
four-stream, TSO-disabled transmit runs per phase. Median throughput
was 9.413, 9.413, and 9.414 Gbps. Median whole-system CPU was 21.54%,
[11 lines not shown]
ice: Defer RDMA critical error notifications
ice_msix_admin() runs as an interrupt filter inside a critical section.
ice_rdma_notify_pe_intr() acquires the global RDMA sx and invokes the
client event handler, both of which require sleepable thread context. A
PE or HMC critical error could therefore panic under WITNESS or sleep
from interrupt context.
Accumulate OICR causes atomically in the interrupt filter and mark them
pending in the driver state. Deliver the notification from the iflib
admin task before processing reset events. This preserves the existing
ordering, lets an iRDMA-requested reset run in the same admin pass, and
coalesces causes from multiple interrupts.
MFC after: 2 weeks
Sponsored by: BBOX.io
Differential Revision: https://reviews.freebsd.org/D59340
ping: do a better job with what we received from the net, part 2
When we see a difference between the payload we sent and what we
received, we dump both but we were not prepared for the case when
the received payload is less than we sent. In this case we were trying
to dump more than needed.
Funny enough, 23 years ago I already fixed a similar issue here but
didn't pay attention to this small dumping loop.
Test written by jlduran.
Reviewed by: jlduran
MFC after: 1 month
Found with: Claude Code Sonnet 5
Differential Revision: https://reviews.freebsd.org/D59556
Differential Revision: https://reviews.freebsd.org/D59582
ping: do a better job checking what we receive from the net
- zero out a buffer for the incoming icmp message that
we are about to parse
- for ICMP_MASKREPLY and ICMP_TSTAMPREPLY responses check their length
and warn and reject them if they are truncated
Without these checks a part of stack allocated struct icmp icp could
be printed out which seems low severity since we already dropped root
privileges by the time icp is allocated.
Reviewed by: glebius
MFC after: 1 month
Found with: Claude Code Sonnet 5
Differential Revision: https://reviews.freebsd.org/D59555
iavf: Honor PF-negotiated RSS sizes
Use the key and lookup table lengths returned by GET_VF_RESOURCES when
configuring RSS through virtchnl, as DPDK does. The Windows E835 PF
advertises a 40-byte key and rejects our fixed 52-byte CONFIG_RSS_KEY
request, leaving receive traffic on queue zero.
Validate the negotiated lengths before constructing AdminQ messages and
publish the lookup table size to iflib. Preserve register-mode RSS
selection and its fixed hardware sizes. Use aligned, zero initialized
key storage so an RSS kernel's 40-byte key does not leave an uninitialized
tail when the PF requests 52 bytes.
Validation: normal and RSS enabled iavf module builds passed. On an E835
VF under Hyper-V Server 2025, repeated IPv4 and IPv6 receive tests used all
three configured guest RX queues in both transparent hn and non-transparent
lagg modes. The RSS key rejection disappeared, IPv4 transmit tests passed,
and no TX watchdog fired. Each traffic case used three runs of 16 streams.
[3 lines not shown]
