MFC r354106:

cxgbe(4): Use correct FetchBurstMin values for T6.

Sponsored by: Chelsio Communications

MFC r345334:

cxgbe(4): Treat the viid as an opaque identifier.

Recent firmwares prefer to use a different format for viid internally
and this change allows them to do so.

Sponsored by: Chelsio Communications

MFC r343666, r343861-r343862, r343923, r343968, r345660, r345810

r343666:
cxgbe(4): Improved error reporting and diagnostics.

"slow" interrupt handler:
- Expand the list of INT_CAUSE registers known to the driver.
- Add decode information for many more bits but decouple it from the
rest of intr_info so that it is entirely optional.
- Call t4_fatal_err exactly once, and from the top level PL intr handler.

t4_fatal_err:
- Use t4_shutdown_adapter from the common code to stop the adapter.
- Stop servicing slow interrupts after the first fatal one.

Driver/firmware interaction:
- CH_DUMP_MBOX: note whether the mailbox being dumped is a command or a
reply or something else.
- Log the raw value of pcie_fw for some errors.
- Use correct log levels (debug vs. error).

Sponsored by: Chelsio Communications

r343861:
cxgbe(4): Auto-dump the device log on a mailbox timeout or when the
firmware reports an error in pcie_fw.

Sponsored by: Chelsio Communications

r343862:
cxgbe(4): Auto-dump the CIM block's logic analyzer on a TIMER0 interrupt.

Sponsored by: Chelsio Communications

r343923:
cxgbe(4): Delay the panic due to a fatal error by 30s.

This lets information logged by the interrupt handler reach the system
log before the system goes down.

r343968:
cxgbe(4): Ignore unused interrupts.

Sponsored by: Chelsio Communications

r345660:
cxgbe(4): Count and clear interrupts generated at the software's request.

An interrupt can be requested by setting the F_SWINT bit in PL_PF_CTL.

Sponsored by: Chelsio Communications

r345810:
cxgbe(4): Add a flag to indicate that bits in interrupt cause but not in
interrupt enable are not fatal.

The firmware sets up all the interrupt enables based on run time
configuration, which means the information in the enables is more
accurate than what's compiled into the driver. This change also allows
the fatal bits to be updated without any changes in the driver in some
cases.

Sponsored by: Chelsio Communications

MFC r342954:

cxgbe(4): Move some INTx specific code to a more appropriate place.

MFC r341654:

cxgbe(4): Get Linux cxgb4vf working in bhyve VMs with VFs passed
through.

cxgb4vf doesn't own the buffer size list but still expects the first two
entries to be 4K and some power of 2 respectively. The BSD cxgbe
doesn't care where its preferred buffer sizes are as long as they're in
the list somewhere, so just move its entries towards the end as a
workaround.

Sponsored by: Chelsio Communications

MFC r339626:

cxgbe(4): Use automatic cidx updates with ofld and ctrl queues.

The bits that explicitly request cidx updates do not work reliably with
all possible WRs that can be sent over the queue. The F_FW_WR_EQUIQ
requests that still remain may also have to be replaced with explicit
credit flush WRs in the future.

Sponsored by: Chelsio Communications

MFC r338218:

cxgbev(4): Updates to the VF driver to cope with recent ifmedia and
ctrlq changes in the base driver.

Sponsored by: Chelsio Communications

MFC r338156, r338158-r338161, r338166.

r338156:
cxgbe(4): Avoid overflow while calculating channel rate.

Reported by: Coverity (CID 1008352)

r338158:
cxgbe(4): Check the RO bit properly before disabling relaxed ordering.

Reported by: Coverity (CID 1384286)

r338159:
cxgbe(4): Make it clear that VI_INIT_DONE implies vi->ntxq > 0, and so
rc will never be returned uninitialized.

Reported by: Coverity (CID 1394884). This is a false positive though.

r338160:
cxgbe(4): Do not leak memory in case of errors during VI initialization.

Reported by: Coverity (CID 1392026)

r338161:
cxgbe/tom: Make sure 'matched' is always initialized before use.

Reported by: Coverity (CID 1390894)

r338166:
cxgbe(4): Be explicit about ignoring the return value of cmpset in some
cases.

Reported by: Coverity (CIDs 1009398, 1009400, 1009401, 1357325, 1394783). All false positives.

MFC r337659:

cxgbe(4): Move all control queues to the adapter.

There used to be one control queue per adapter (the mgmtq) that was
initialized during adapter init and one per port that was initialized
later during port init. This change moves all the control queues (one
per port/channel) to the adapter so that they are initialized during
adapter init and are available before any port is up. This allows the
driver to issue ctrlq work requests over any channel without having to
bring up any port.

MFC r337609:

cxgbe(4): Create two variants of service_iq, one for queues with
freelists and one for those without.

MFH: 3 weeks
Sponsored by: Chelsio Communications

MFC r337538, r337987

r337538:
cxgbe(4): Add support for high priority filters on T6+. They have their
own region in the TCAM starting with T6, unlike previous chips where
they were in the same region as normal filters.

These filters "hit" before anything else in the LE's lookup. The exact
order is:
a) High priority filters
b) TOE's active region (TCAM and/or hash)
c) Servers (TOE hw listeners)
d) Normal filters

Sponsored by: Chelsio Communications

r337987:
cxgbe(4): Adjust ntids to account for nhptids in the TOE case too.
This should have been part of r337538.

MFC r336718, r336720, r336734-r336735, r337398, r337439, and r337540.
These are all related to tx rate limiting in cxgbe.

r336718:
cxgbe(4): Validate only those parameters that are relevant to the
type of rate limiter being programmed. Skip the ones that are not
applicable.

Sponsored by: Chelsio Communications

r336720:
cxgbe(4): Remove useless code that crept in with r336718.

X-MFC With: 336718

r336734:
cxgbe(4): Better defaults for all cl-rl rate limiters.

Start in "class" instead of "flow" mode. This eliminates the need to
specify an MTU, which is not available that early anyway. It also
allows the user to manually configure ch-rl rate limiting after attach.
This used to fail because ch-rl isn't supported if cl-rl "flow" mode is
configured.

Set all traffic classes to 1Gbps during initialization. The goal is to
start off with _any_ valid configuration and 1Gbps works even for
gigabit cards.

Sponsored by: Chelsio Communications

r336735:
cxgbe(4): Consider rateunit before ratemode when displaying information
about a traffic class. This matches the order in which the firmware
evaluates unit and mode internally.

Sponsored by: Chelsio Communications

r337398:
cxgbe(4): Allow user-configured and driver-configured traffic classes to
be used simultaneously. Move sysctl_tc and sysctl_tc_params to
t4_sched.c while here.

Sponsored by: Chelsio Communications

r337439:
cxgbe(4): Allow the driver to specify a burst size when configuring a
traffic class for rate limiting.

Add experimental knobs that allow the user to specify a default pktsize
and burstsize for traffic classes associated with a port:

dev.<ifname>.<instance>.tc.pktsize
dev.<ifname>.<instance>.tc.burstsize

Sponsored by: Chelsio Communications

r337540:
cxgbe(4): Display pkt-size and burst-size in traffic class parameters.

MFC r334138:

r334138: cxgbe(4): Make FW4_ACK a shared CPL.

MFC r334136:

cxgbe(4): Slightly simpler needs_<foo> functions.

MFC r334132:

cxgbe(4): Make sure that the egress queue's cidx is updated periodically
when the driver is writing WRs using start_wrq_wr/commit_wrq_wr all the
time.

Sponsored by: Chelsio Communications

MFC r333153, r333394, r333442, r333472, r333620, r334058, r334447,
r334452, and r335684. These revisions added hashfilters, NAT offload,
and SMAC/DMAC swapping filters to cxgbe.

r333153:
cxgbe(4): Move all TCAM filter code into a separate file.

Sponsored by: Chelsio Communications

r333394:
cxgbe(4): Add support for hash filters.

These filters reside in the card's memory instead of its TCAM and can be
configured via a new "hashfilter" subcommand in cxgbetool. Hash and
normal TCAM filters can be used together. The hardware does an
exact-match of packet fields for hash filters, unlike the masked match
performed for TCAM filters. Any T5/T6 card with memory can support at
least half a million hash filters. The sample config file with the
driver configures 512K of these, it is possible to double this to 1
million+ in some cases.

The chip does an exact-match of fields of incoming datagrams with hash
filters and performs the action configured for the filter if it matches.
The fields to match are specified in a "filter mask" in the firmware
config file. The filter mask always includes the 5-tuple (sip, dip,
sport, dport, ipproto). It can, optionally, also include any subset of
the filter mode (see filterMode and filterMask in the firmware config
file).

For example:
filterMode = fragmentation, mpshittype, protocol, vlan, port, fcoe
filterMask = protocol, port, vlan

Exact values of the 5-tuple, the physical port, and VLAN tag would have
to be provided while setting up a hash filter with the chip
configuration above.

Hash filters support all actions supported by TCAM filters. A packet
that hits a hash filter can be dropped, let through (with optional
steering to a specific queue or RSS region), switched out of another
port (with optional L2 rewrite of DMAC, SMAC, VLAN tag), or get NAT'ed.
(Support for some of these will show up in the driver in a follow-up
commit very shortly).

Sponsored by: Chelsio Communications

r333442:
cxgbe(4): Determine whether the firmware supports the FILTER2 work
request, which can be used to configure hardware NAT and swapmac.

All firmwares released after Jan 2017 support this work request.

Sponsored by: Chelsio Communications

r333472:
cxgbe(4): Add fields to support configuration of hardware NAT and
swapmac (SMAC/DMAC switcheroo) from userspace.

Sponsored by: Chelsio Communications

r333620:
cxgbe(4): Filtering related features and fixes.

- Driver support for hardware NAT.
- Driver support for swapmac action.
- Validate a request to create a hashfilter against the filter mask.
- Add a hashfilter config file for T5.

Sponsored by: Chelsio Communications

r334058:
cxgbe(4): Only valid filters are expected to have a valid tid.

r334447:
cxgbe(4): Add code to deal with the chip's source MAC table (aka SMT).

Submitted by: Krishnamraju Eraparaju @ Chelsio
Sponsored by: Chelsio Communications

r334452:
cxgbe(4): Add support for SMAC-rewriting filters.

Submitted by: Krishnamraju Eraparaju @ Chelsio
Sponsored by: Chelsio Communications

r335684:
cxgbe(4): Do not leak the filters in the hashfilter table on module
unload.

Sponsored by: Chelsio Communications

Relnotes: Yes

MFC r333121, r333128

r333121:
cxgbe/t4_tom: Use appropriate macros instead of magic math while
constructing the atid of an active open work request.

Sponsored by: Chelsio Communications

r333128:
cxgbe(4): Convert ACT_OPEN_RPL to a shared CPL.

Reserve 3b in the 14b atid to identify the owner and use it to dispatch
the CPL. This allows all CPLs that use an atid to be used as shared
CPLs, although ACT_OPEN_RPL is the only one being converted in this
revision.

Sponsored by: Chelsio Communications

MFC r333114:

cxgbe(4): Use opaque cookies or tid range-checks to determine the
intended recipient of a CPL when it can't be determined solely from the
opcode. Retire the per-queue handlers for such CPLs in favor of the new
scheme.

Sponsored by: Chelsio Communications

MFC r317849 (partial), r332506, and r332787.

r317849 (partial, required by r332506):
cxgbe/t4_tom: Per-connection rate limiting for TCP sockets handled by
the TOE.

Sponsored by: Chelsio Communications

r332506:
cxgbe(4): Add support for Connection Offload Policy (aka COP).

COP allows fine-grained control on whether to offload a TCP connection
using t4_tom, and what settings to apply to a connection selected for
offload. t4_tom must still be loaded and IFCAP_TOE must still be
enabled for full TCP offload to take place on an interface. The
difference is that IFCAP_TOE used to be the only knob and would enable
TOE for all new connections on the inteface, but now the driver will
also consult the COP, if any, before offloading to the hardware TOE.

A policy is a plain text file with any number of rules, one per line.
Each rule has a "match" part consisting of a socket-type (L = listen,
A = active open, P = passive open, D = don't care) and a pcap-filter(7)
expression, and a "settings" part that specifies whether to offload the
connection or not and the parameters to use if so. The general format
of a rule is: [socket-type] expr => settings

Example. See cxgbetool(8) for more information.
[L] ip && port http => offload
[L] port 443 => !offload
[L] port ssh => offload
[P] src net 192.168/16 && dst port ssh => offload !nagle !timestamp cong newreno
[P] dst port ssh => offload !nagle ecn cong tahoe
[P] dst port http => offload
[A] dst port 443 => offload tls
[A] dst net 192.168/16 => offload !timestamp cong highspeed

The driver processes the rules for each new listen, active open, or
passive open and stops at the first match. There is an implicit rule at
the end of every policy that prohibits offload when no rule in the
policy matches:
[D] all => !offload

This is a reworked and expanded version of a patch submitted by
Krishnamraju Eraparaju @ Chelsio.

Sponsored by: Chelsio Communications

r332787:
cxgbe(4): Fix bugs in the handling of COP rules that match on VLAN tag.

Retrieve the tag from the correct ifnet and use the provided tag
(instead of hardcoded 0xffff, implying no tag) in the routines that
process offload policy.

Submitted by: Krishnamraju Eraparaju @ Chelsio
Sponsored by: Chelsio Communications

MFC 341098: Add read-only sysctls for all tunables in the cxgbe(4) driver.

MFC 340021: Assert that reclaim_tx_descs() is always making forward progress.

MFC r325840, r327811, and r329701.

r325840:
CXGBE: fix big-endian behaviour

The setbit/clearbit pair casts the bitfield pointer
to uint8_t* which effectively treats its contents as
little-endian variable. The ffs() function accepts int as
the parameter, which is big-endian. Use uint8_t here to
avoid mismatch, as we have only 4 doorbells.

Submitted by: Wojciech Macek <wma@freebsd.org>
Reviewed by: np
Obtained from: Semihalf
Sponsored by: QCM Technologies
Differential revision: https://reviews.freebsd.org/D13084

r327811:
CXGBE: fix get_filt to be endianness-aware

Unconditional 32-bit shift is not endianness-safe.
Modify the logic to work both on LE and BE.

Submitted by: Wojciech Macek <wma@freebsd.org>
Reviewed by: np
Obtained from: Semihalf
Sponsored by: IBM, QCM Technologies
Differential revision: https://reviews.freebsd.org/D13102

r329701:
CXGBE: implement prefetch on non-Intel architectures

Submitted by: Michal Stanek <mst@semihalf.com>
Obtained from: Semihalf
Reviewed by: np, pdk@semihalf.com
Sponsored by: IBM, QCM Technologies
Differential revision: https://reviews.freebsd.org/D14452

Revert r330897:

This was intended to be a non-functional change. It wasn't. The commit
message was thus wrong. In addition it broke arm, and merged crypto
related code.

Revert with prejudice.

This revert skips files touched in r316370 since that commit was since
MFCed. This revert also skips files that require $FreeBSD$ property
changes.

Thank you to those who helped me get out of this mess including but not
limited to gonzo, kevans, rgrimes.

Requested by: gjb (re)

Partial merge of the SPDX changes

These changes are incomplete but are making it difficult
to determine what other changes can/should be merged.

No objections from: pfg

MFC r319506, r319872, r321063, r321103, r321179, r321390, r321435,
r321582, r321671, r322014, r322034, r322055, r322123, r322167, r322425,
r322549, r322914, r322960, r322962, r322964, r322985, r322990, r323006,
r323026, r323041, r323069, r323078, r323343, r323514, r323520, r324296,
r324379, r324386, r324443, r324945, r325596, r325680, r325880,
r325883-r325884, r325961, r326026, r326042, r327062, r327093, r327332,
r327528, r328420, and r328423.

r319506:
cxgbe(4): Update the statistics for compound tx work requests once per
work request, not once per frame.

r319872:
cxgbe(4): Do not request an FEC setting that the port does not support.

r321063:
cxgbe(4): Various link/media related improvements.

- Deal with changes to port_type, and not just port_mod when a
transceiver is changed. This fixes hot swapping of transceivers of
different types (QSFP+ or QSA or QSFP28 in a QSFP28 port, SFP+ or
SFP28 in a SFP28 port, etc.).

- Always refresh media information for ifconfig if the port is down.
The firmware does not generate tranceiver-change interrupts unless at
least one VI is enabled on the physical port. Before this change
ifconfig diplayed potentially stale information for ports that were
administratively down.

- Always recalculate and reapply L1 config on a transceiver change.

- Display PAUSE settings in ifconfig. The driver sysctls for this
continue to work as well.

r321103:
cxgbe(4): New ioctls to flash bootrom and boot config to the card.

r321179:
cxgbe/t4_tom: Log more details about the newly ESTABLISHED tid to the
trace buffer.

r321390:
cxgbe(4): Install the firmware bundled with the driver to the card if it
doesn't seem to have one. This lets the driver recover automatically
from incomplete firmware upgrades (panic, reboot, power loss, etc. in
the middle of an upgrade).

r321435:
cxgbe(4): Display some more TOE parameters related to retransmission
and keepalive in the sysctl MIB. Provide tunables to change some of
these parameters. These are supposed to be setup by the firmware so
these tunables are for experimentation only.

r321582:
cxgbe(4): Some updates to the common code.

- Updated register ranges.
- Helper routines for access to TP registers.
- Updated routine to read flash parameters.

r321671:
cxgbe/iw_cxgbe: Log the end point's history and flags to the trace
buffer just before it's freed.

r322014:
cxgbe(4): Initial import of the "collect" component of Chelsio unified
debug (cudbg) code, hooked up to the main driver via an ioctl.

The ioctl can be used to collect the chip's internal state in a
compressed dump file. These dumps can be decoded with the "view"
component of cudbg.

r322034:
cxgbe(4): Always use the first and not the last virtual interface
associated with a port in begin_synchronized_op.

r322055:
cxgbe(4): Allow the TOE timer tunables to be set with microsecond
precision. These timers are already displayed in microseconds in the
sysctl MIB. Add variables to track these tunables while here.

r322123:
cxgbe(4): Avoid a NULL dereference that would occur during module unload
if there were problems earlier during attach.

r322167:
cxgbe(4): Add the T6 and T5 Unified Wire configuration files to the
kernel, just like for T4, when the driver is compiled into the kernel.

r322425:
cxgbe(4): Save the last reported link parameters and compare them with
the current state to determine whether to generate a link-state change
notification. This fixes a bug introduced in r321063 that caused the
driver to sometimes skip these notifications.

r322549:
cxgbe/t4_tom: Use correct name for the ISS-valid bit in options2.

r322914:
cxgbe(4): Dump the mailbox contents in the same format as CH_DUMP_MBOX.

r322960:
cxgbe(4): Verify that the driver accesses the firmware mailbox in a
thread-safe manner.

r322962:
cxgbe(4): Remove write only variable from t4_port_init.

r322964:
cxgbe(4): vi_mac_funcs should include the base Ethernet function. It is
already used in the driver as if it does.

r322985:
cxgbe(4): Maintain one ifmedia per physical port instead of one per
Virtual Interface (VI). All autonomous VIs that share a port share the
same media.

r322990:
cxgbe(4): Do not access the mailbox without appropriate locks while
creating hardware VIs.

This fixes a bad race on systems with hw.cxgbe.num_vis > 1.

r323006:
cxgbe(4): Update T6/T5/T4 firmwares to 1.16.59.0.

r323026:
cxgbe(4): Zero out the memory allocated for the debug dump.
cudbg_collect seems to expect it this way.

r323041:
cxgbe(4): Add two new debug flags -- one to allow manual firmware
install after full initialization, and another to disable the TCB
cache (T6+). The latter works as a tunable only.

Note that debug_flags are for debugging only and should not be set
normally.

r323069:
cxgbe/t4_tom: Add a knob to select the congestion control algorigthm
used by the TOE hardware for fully offloaded connections. The knob
affects new connections only.

r323078:
cxgbe/t4_tom: There may not be a tid to update if the connection isn't
established.

r323343:
cxgbe(4): Fix a couple of problems in the sge_wrq data path.

- start_wrq_wr must not drain the wr_list if there are incomplete_wrs
pending. This can happen when a t4_wrq_tx runs between two
start_wrq_wr.

- commit_wrq_wr must examine the cookie's pidx and ndesc with the
queue's lock held. Otherwise there is a bad race when incomplete WRs
are being completed and commit_wrq_wr for the WR that is ahead in the
queue updates the next incomplete WR's cookie's pidx/ndesc but the
commit_wrq_wr for the second one is using stale values that it read
without the lock.

r323514:
cxgbetool(8): mode must be specified when creating the dump file.

r323520:
cxgbe(4): Ignore capabilities that depend on TOE when the firmware
reports TOE is not available.

r324296:
cxgbe(4): Provide knobs to set the holdoff parameters of TOE rx queues
separately from NIC rx queues instead of using the same parameters for
both types of queues.

r324379:
cxgbetool(8): Do not create a large file devoid of useful content when
the dumpstate ioctl fails. Make the file world-readable while here.

r324386:
cxgbe(4): Update T6, T5, and T4 firmwares to 1.16.63.0.

r324443:
cxgbetool(8): Do not close uninitialized fd on malloc failure.

r324945:
cxgbe(4): Read the MPS buffer group map from the firmware as it could be
different from hardware defaults. The congestion channel map, which is
still fixed, needs to be tracked separately now. Change the congestion
setting for TOE rx queues to match the drivers on other OSes while here.

r325596:
cxgbe(4): Do not request settings not supported by the port.

r325680:
cxgbe(4): Excluce mdi from the check against port capabilities.

r325880:
cxgbe(4): Combine all _10g and _1g tunables and drop the suffix from
their names. The finer-grained knobs weren't practically useful.

r325883:
cxgbe(4): Sanitize t4_num_vis during MOD_LOAD like all other t4_*
tunables. Add num_vis to the intrs_and_queues structure as it affects
the number of interrupts requested and queues created. In future
cfg_itype_and_nqueues might lower it incrementally instead of going
straight to 1 when enough interrupts aren't available.

r325884:
cxgbe(4): Remove rsrv_noflowq from intrs_and_queues structure as it does
not influence or get affected by the number of interrupts or queues.

r325961:
cxgbe(4): Add core Vdd to the sysctl MIB.

r326026:
cxgbe(4): Add a custom board to the device id list.

r326042:
cxgbe(4): Fix unsafe mailbox access in cudbg.

r327062:
cxgbe(4): Read the MFG diags version from the VPD and make it available
in the sysctl MIB.

r327093:
cxgbe(4): Do not forward interrupts to queues with freelists. This
leaves the firmware event queue (fwq) as the only queue that can take
interrupts for others.

This simplifies cfg_itype_and_nqueues and queue allocation in the driver
at the cost of a little (never?) used configuration. It also allows
service_iq to be split into two specialized variants in the future.

r327332:
cxgbe(4): Reduce duplication by consolidating minor variations of the
same code into a single routine.

r327528:
cxgbe(4): Add a knob to enable/disable PCIe relaxed ordering. Disable it by
default when running on Intel CPUs.

r328420:
cxgbe(4): Do not display harmless warning in non-debug builds.

r328423:
cxgbe(4): Accept old names of a couple of tunables.

Sponsored by: Chelsio Communications

MFC r318014, r318091, r318125, and r318263.

r318014:
cxgbe(4): Fixes related to the knob that controls link autonegotiation.

- Do not leak the adapter lock in sysctl_autoneg.
- Accept only 0 or 1 as valid settings for autonegotiation.
- A fixed speed must be requested by the driver when autonegotiation is
disabled otherwise the firmware will reject the l1cfg command. Use
the top speed supported by the port for now.

r318091:
cxgbe(4): Do not assume that if_qflush is always followed by inteface-down.

r318125:
Adjust whitespace and fix a comment. No functional change.

r318263:
cxgbe(4): netmap-only interrupts for a VI do not have an associated rxq
or ofld_rxq and should be ignored by vi_intr_iq.

Sponsored by: Chelsio Communications

MFC r317702, r317847, r318307

r317702:
cxgbe(4): Support routines for Tx traffic scheduling.

- Create a new file, t4_sched.c, and move all of the code related to
traffic management from t4_main.c and t4_sge.c to this file.
- Track both Channel Rate Limiter (ch_rl) and Class Rate Limiter (cl_rl)
parameters in the PF driver.
- Initialize all the cl_rl limiters with somewhat arbitrary default
rates and provide routines to update them on the fly.
- Provide routines to reserve and release traffic classes.

r317847:
cxgbe(4): The Tx scheduler initialization either works or doesn't. It
doesn't need a refresh in either case.

r318307:
cxgbe(4): Avoid an out of bounds access when an attempt to unbind a tx
queue from a traffic class fails.

Sponsored by: Chelsio Communications

MFC r317041:

cxgbe: Add tunables to control the number of LRO entries and the number
of rx mbufs that should be presorted before LRO. There is no change in
default behavior.

Sponsored by: Chelsio Communications

MFC r316971:

cxgbe: Add a tunable to configure the SGE time scaler, which is
available starting with T6. The values in the timer holdoff registers
are multiplied by the scaling factor before use.

dev.<nexus>.<n>.holdoff_timers shows the final values of the
timers in microseconds.

Sponsored by: Chelsio Communications

MFC r311831 and r311832.

r311831:
cxgbe(4): The wraparound logic in start_wrq_wr() should not get involved
in work requests that end at the end of the descriptor ring, even though
the pidx wraps around to 0.

r311832:
cxgbe(4): Enable automatic cidx flush for all control queues.

MFC r309666, r310033, r310049, r310100, r310152, and r310807.

r309666:
cxgbe(4): unsigned short isn't large enough to store link speed (which
is in Mbps) for 100Gbps links.

r310033:
cxgbe(4): Retire t4_bus_space_read_8 and t4_bus_space_write_8.

r310049:
cxgbe(4): Fix the tid range shown for T6 cards in misc.tids.

r310100:
cxgbe(4): Deal with compressed error vectors.

r310152:
cxgbe(4): Fix typo in an unused macro.

r310807:
cxgbe(4): Updates to link configuration.

- Update struct link_settings and associated shared code.

- Add tunables to control FEC and autonegotiation. All ports inherit
these values as their initial settings.
hw.cxgbe.fec
hw.cxgbe.autoneg

- Add per-port sysctls to control FEC and autonegotiation. These can be
modified at any time.
dev.<port>.<n>.fec
dev.<port>.<n>.autoneg

MFC 307876:
cxgbe(4): Fix bug in the calculation of the number of physically
contiguous regions in an mbuf chain.

If the payload of an mbuf ends at a page boundary count_mbuf_nsegs would
incorrectly consider the next mbuf's payload physically contiguous based
solely on a KVA comparison.

MFC 306277:
cxgbe(4): Make the location/length of all descriptor rings available in
the sysctl MIB.

MFC 305695,305696,305699,305702,305703,305713,305715,305827,305852,305906,
305908,306062,306063,306137,306138,306206,306216,306273,306295,306301,
306465,309302:
Add support for adapters using the Terminator T6 ASIC.

305695:
cxgbe(4): Set up fl_starve_threshold2 accurately for T6.

305696:
cxgbe(4): Use correct macro for header length with T6 ASICs. This
affects the transmit of the VF driver only.

305699:
cxgbe(4): Update the pad_boundary calculation for T6, which has a
different range of boundaries.

305702:
cxgbe(4): Use smaller min/max bursts for fl descriptors with a T6.

305703:
cxgbe(4): Deal with the slightly different SGE_STAT_CFG in T6.

305713:
cxgbe(4): Add support for additional port types and link speeds.

305715:
cxgbe(4): Catch up with the rename of tlscaps -> cryptocaps. TLS is one
of the capabilities of the crypto engine in T6.

305827:
cxgbe(4): Use the interface's viid to calculate the PF/VF/VFValid fields
to use in tx work requests.

305852:
cxgbe(4): Attach to cards with the Terminator 6 ASIC. T6 cards will
come up as 't6nex' nexus devices with 'cc' ports hanging off them.

The T6 firmware and configuration files will be added as soon as they
are released. For now the driver will try to work with whatever
firmware and configuration is on the card's flash.

305906:
cxgbe/t4_tom: The SMAC entry for a VI is at a different location in the T6.

305908:
cxgbe/t4_tom: Update the active/passive open code to support T6. Data
path works as-is.

306062:
cxgbe(4): Show wcwr_stats for T6 cards.

306063:
cxgbe(4): Setup congestion response for T6 rx queues.

306137:
cxgbetool: Add T6 support to the SGE context decoder.

306138:
Fix typo.

306206:
cxgbe(4): Catch up with the different layout of WHOAMI in T6.

Note that the code moved below t4_prep_adapter() as part of this change
because now it needs a working chip_id().

306216:
cxgbe(4): Fix the output of the "tids" sysctl on T6.

306273:
cxgbe(4): Fix netmap with T6, which doesn't encapsulate SGE_EGR_UPDATE
message inside a FW_MSG. The base NIC already deals with updates in
either form.

306295:
cxgbe(4): Support SIOGIFXMEDIA so that ifconfig displays correct media
for 25Gbps and 100Gbps ports. This should have been part of r305713,
which is when the driver first started reporting extended media types.

306301:
cxgbe(4): Use the port's top speed to figure out whether it is "high
speed" or not (for the purpose of calculating the number of queues etc.)
This does the right thing for 25Gbps and 100Gbps ports.

306465:
cxgbe(4): Claim the T6 -DBG card.

309302:
cxgbe(4): Include firmware for T6 cards in the driver. Update all
firmwares to 1.16.12.0.

Sponsored by: Chelsio Communications

MFC 305652: cxgbe(4): Do not prescreen frames before attempting LRO.

MFC 302440,304873,305704,305985,306787,307531: Fixes for sysctls.

302440:
cxgbe(4): Add sysctl to display the RSS indirection table size for an
interface.

dev.cxl.<n>.rss_size
dev.vcxl.<n>.rss_size

304873:
cxgbe(4): Provide more details about the card in the sysctl MIB.

dev.t5nex.0.%desc: Chelsio T580-CR
dev.t5nex.0.hw_revision: 1
dev.t5nex.0.sn: PT13140042
dev.t5nex.0.pn: 110117150A0
dev.t5nex.0.ec: 0000000000000000
dev.t5nex.0.na: 0007432AF490
dev.t5nex.0.vpd_version: 3
dev.t5nex.0.scfg_version: 53255
dev.t5nex.0.bs_version: 1.1.0.0
dev.t5nex.0.er_version: 1.0.0.68
dev.t5nex.0.tp_version: 0.1.4.9
dev.t5nex.0.firmware_version: 1.16.2.0

305704:
cxgbe(4): Rename the debug_flags driver tunable/sysctl to dflags.
Tunables that end with _flags are special.

305985:
cxgbe(4): Fixes to wrq stats.

- Increment tx_wrs_copied in the correct place.
- Add tx_wrs_sspace to the sysctl MIB.

306787:
cxgbe(4): Fix whitespace in the pm_stats display.

307531:
cxgbe(4): Adjust whitespace to line up the column titles in cim_qcfg
with the values displayed.

Sponsored by: Chelsio Communications

MFC 303859,305851: Fix a typo and some whitespace nits.

MFC 303454: Mark spg_len and fl_pktshift static.

These variables are no longer exported to t4_netmap.c after r296478.

MFC 305548: Don't break out of the m_advance() loop if len drops to zero.

If a packet contains the Ethernet header (14 bytes) in the first mbuf
and the payload (IP + UDP + data) in the second mbuf, then the attempt
to fetch the l3hdr will return a NULL pointer. The first loop iteration
will drop len to zero and exit the loop without setting 'p'. However,
the desired data is at the start of the second mbuf, so the correct
behavior is to loop around and let the conditional set 'p' to m_data of
the next mbuf (and leave offset as 0).

MFC 303522,303647,303860,303880,304168,304169,304170,304479,304485,305549:
Chelsio T4/T5 VF driver.

303522:
Various fixes to the t4/5nex character device.

- Remove null open/close methods.
- Don't set d_flags to 0 explicitly.
- Remove t5_cdevsw as the .d_name member isn't really used and doesn't
warrant a separate cdevsw just for the name.
- Use ENOTTY as the error value for an unknown ioctl request.
- Use make_dev_s() to close race with setting si_drv1.

303647:
Store the offset of the KDOORBELL and GTS registers in the softc.

VF devices use a different register layout than PF devices. Storing
the offset in a value in the softc allows code to be shared between the
PF and VF drivers.

303860:
Reserve an adapter flag IS_VF to mark VF devices vs PF devices.

303880:
Track the base absolute ID of ingress and egress queues.

Use this to map an absolute queue ID to a logical queue ID in interrupt
handlers. For the regular cxgbe/cxl drivers this should be a no-op as
the base absolute ID should be zero. VF devices have a non-zero base
absolute ID and require this change. While here, export the absolute ID
of egress queues via a sysctl.

304168:
Make SGE parameter handling more VF-friendly.

Add fields to hold the SGE control register and free list buffer sizes to
the sge_params structure. Populate these new fields in
t4_init_sge_params() for PF devices and change t4_read_chip_settings() to
pull these values out of the params structure instead of reading
registers directly. This will permit t4_read_chip_settings() to be reused
for VF devices which cannot read SGE registers directly.

While here, move the call to t4_init_sge_params() to
get_params__post_init(). The VF driver will populate the SGE parameters
structure via a different method before calling t4_read_chip_settings().

304169:
Update mailbox writes to work with VF devices.

- Use alternate register locations for the data and control registers for
VFs.
- Do a dummy read to force the writes to the mailbox data registers to
post before the write to the control register on VFs.
- Do not check the PCI-e firmware register for errors on VFs.

304170:
Add support for register dumps on VF devices.

- Add handling of VF register sets to t4_get_regs_len() and t4_get_regs().
- While here, use t4_get_regs_len() in the ioctl handler for regdump
instead of inlining it.

304479:
Add structures for VF-specific adapter parameters.

While here, mark which parameters are PF-specific and which are
VF-specific.

304485:
Reorder sysctls so that nodes shared with the VF driver are added first.

This permits a single early return for VF devices in the routines that
add sysctl nodes.

305549:
Chelsio T4/T5 VF driver.

The cxgbev/cxlv driver supports Virtual Function devices for Chelsio
T4 and T4 adapters. The VF devices share most of their code with the
existing PF4 driver (cxgbe/cxl) and as such the VF device driver
currently depends on the PF4 driver.

Similar to the cxgbe/cxl drivers, the VF driver includes a t4vf/t5vf
PCI device driver that attaches to the VF device. It then creates
child cxgbev/cxlv devices representing ports assigned to the VF.
By default, the PF driver assigns a single port to each VF.

t4vf_hw.c contains VF-specific routines from the shared code used to
fetch VF-specific parameters from the firmware.

t4_vf.c contains the VF-specific PCI device driver and includes its
own attach routine.

VF devices are required to use a different firmware request when
transmitting packets (which in turn requires a different CPL message
to encapsulate messages). This alternate firmware request does not
permit chaining multiple packets in a single message, so each packet
results in a firmware request. In addition, the different CPL message
requires more detailed information when enabling hardware checksums,
so parse_pkt() on VF devices must examine L2 and L3 headers for all
packets (not just TSO packets) for VF devices. Finally, L2 checksums
on non-UDP/non-TCP packets do not work reliably (the firmware trashes
the IPv4 fragment field), so IPv4 checksums for such packets are
calculated in software.

Most of the other changes in the non-VF-specific code are to expose
various variables and functions private to the PF driver so that they
can be used by the VF driver.

Note that a limited subset of cxgbetool functions are supported on VF
devices including register dumps, scheduler classes, and clearing of
statistics. In addition, TOE is not supported on VF devices, only for
the PF interfaces.

Sponsored by: Chelsio Communications

MFC 303204: Install a handler for firmware work request error messages.

If a driver sends an malformed or disallowed work request, the firmware
responds with a work request error. Previously the driver treated this is
as an unexpected message and panicked. Now it decodes the error message
to aid in debugging.

Sponsored by: Chelsio Communications

Copy head@r302406 to stable/11 as part of the 11.0-RELEASE cycle.
Prune svn:mergeinfo from the new branch, as nothing has been merged
here.

Additional commits post-branch will follow.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

cxgbe(4): Changes to the CPL-handler registration mechanism and code
related to "shared" CPLs.

a) Combine t4_set_tcb_field and t4_set_tcb_field_rpl into a single
function. Allow callers to direct the response to any iq. Tidy up
set_ulp_mode_iscsi while there to use names from t4_tcb.h instead of
magic constants.

b) Remove all CPL handler tables from struct adapter. This reduces its
size by around 2KB. All handlers are now registered at MOD_LOAD instead
of attach or some kind of initialization/activation. The registration
functions do not need an adapter parameter any more.

c) Add per-iq handlers to deal with CPLs whose destination cannot be
determined solely from the opcode. There are 2 such CPLs in use right
now: SET_TCB_RPL and L2T_WRITE_RPL. The base driver continues to send
filter and L2T_WRITEs over the mgmtq and solicits the reply on fwq.
t4_tom (including the DDP code) now uses the port's ctrlq to send
L2T_WRITEs and SET_TCB_FIELDs and solicits the reply on an ofld_rxq.
fwq and ofld_rxq have different handlers that know what kind of tid to
expect in the reply. Update t4_write_l2e and callers to to support any
wrq/iq combination.

Approved by: re@ (kib@)
Sponsored by: Chelsio Communications

cxgbe(4): Merge netmap support from the ncxgbe/ncxl interfaces to the
vcxgbe/vcxl interfaces and retire the 'n' interfaces. The main
cxgbe/cxl interfaces and tunables related to them are not affected by
any of this and will continue to operate as usual.

The driver used to create an additional 'n' interface for every
cxgbe/cxl interface if "device netmap" was in the kernel. The 'n'
interface shared the wire with the main interface but was otherwise
autonomous (with its own MAC address, etc.). It did not have normal
tx/rx but had a specialized netmap-only data path. r291665 added
another set of virtual interfaces (the 'v' interfaces) to the driver.
These had normal tx/rx but no netmap support.

This revision consolidates the features of both the interfaces into the
'v' interface which now has a normal data path, TOE support, and native
netmap support. The 'v' interfaces need to be created explicitly with
the hw.cxgbe.num_vis tunable. This means "device netmap" will not
result in the automatic creation of any virtual interfaces.

The following tunables can be used to override the default number of
queues allocated for each 'v' interface. nofld* = 0 will disable TOE on
the virtual interface and nnm* = 0 to will disable native netmap
support.

# number of normal NIC queues
hw.cxgbe.ntxq_vi
hw.cxgbe.nrxq_vi

# number of TOE queues
hw.cxgbe.nofldtxq_vi
hw.cxgbe.nofldrxq_vi

# number of netmap queues
hw.cxgbe.nnmtxq_vi
hw.cxgbe.nnmrxq_vi

hw.cxgbe.nnm{t,r}xq{10,1}g tunables have been removed.

--- tl;dr version ---
The workflow for netmap on cxgbe starting with FreeBSD 11 is:
1) "device netmap" in the kernel config.
2) "hw.cxgbe.num_vis=2" in loader.conf. num_vis > 2 is ok too, you'll
end up with multiple autonomous netmap-capable interfaces for every
port.
3) "dmesg | grep vcxl | grep netmap" to verify that the interface has
netmap queues.
4) Use any of the 'v' interfaces for netmap. pkt-gen -i vcxl<n>... .
One major improvement is that the netmap interface has a normal data
path as expected.
5) Just ignore the cxl interfaces if you want to use netmap only. No
need to bring them up. The vcxl interfaces are completely independent
and everything should just work.
---------------------

Approved by: re@ (gjb@)
Relnotes: Yes
Sponsored by: Chelsio Communications

cxgbe(4): Add a sysctl to manage the binding of a txq to a traffic class.

Sponsored by: Chelsio Communications

sys/dev: minor spelling fixes.

Most affect comments, very few have user-visible effects.

cxgbe(4): Always dispatch all work requests that have been written to the
descriptor ring before leaving drain_wrq_wr_list.

tcp/lro: Use tcp_lro_flush_all in device drivers to avoid code duplication

And factor out tcp_lro_rx_done, which deduplicates the same logic with
netinet/tcp_lro.c

Reviewed by: gallatin (1st version), hps, zbb, np, Dexuan Cui <decui microsoft com>
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5725

cxgbe(4): Be consistent and call ETHER_BPF_MTAP before writing anything
to the descriptor ring no matter what path the frame takes within the
driver's tx.

cxgbe(4): Add a struct sge_params to store per-adapter SGE parameters.
Move the code that reads all the parameters to t4_init_sge_params in the
shared code. Use these per-adapter values instead of globals.

Sponsored by: Chelsio Communications

cxgbe(4): Very basic T6 awareness. This is part of ongoing work to
update to the latest internal shared code.

- Add a chip_params structure to keep track of hardware constants for
all generations of Terminators handled by cxgbe.
- Update t4_hw_pci_read_cfg4 to work with T6.
- Update the hardware debug sysctls (hidden within dev.<tNnex>.<n>.misc.*) to
work with T6. Most of the changes are in the decoders for the CIM
logic analyzer and the MPS TCAM.
- Acquire the regwin lock around indirect register accesses.

Obtained from: Chelsio Communications
Sponsored by: Chelsio Communications

Garbage collect unused arguments of m_init().

Add optimizing LRO wrapper:

- Add optimizing LRO wrapper which pre-sorts all incoming packets
according to the hash type and flowid. This prevents exhaustion of
the LRO entries due to too many connections at the same time.
Testing using a larger number of higher bandwidth TCP connections
showed that the incoming ACK packet aggregation rate increased from
~1.3:1 to almost 3:1. Another test showed that for a number of TCP
connections greater than 16 per hardware receive ring, where 8 TCP
connections was the LRO active entry limit, there was a significant
improvement in throughput due to being able to fully aggregate more
than 8 TCP stream. For very few very high bandwidth TCP streams, the
optimizing LRO wrapper will add CPU usage instead of reducing CPU
usage. This is expected. Network drivers which want to use the
optimizing LRO wrapper needs to call "tcp_lro_queue_mbuf()" instead
of "tcp_lro_rx()" and "tcp_lro_flush_all()" instead of
"tcp_lro_flush()". Further the LRO control structure must be
initialized using "tcp_lro_init_args()" passing a non-zero number
into the "lro_mbufs" argument.

- Make LRO statistics 64-bit. Previously 32-bit integers were used for
statistics which can be prone to wrap-around. Fix this while at it
and update all SYSCTL's which expose LRO statistics.

- Ensure all data is freed when destroying a LRO control structures,
especially leftover LRO entries.

- Reduce number of memory allocations needed when setting up a LRO
control structure by precomputing the total amount of memory needed.

- Add own memory allocation counter for LRO.

- Bump the FreeBSD version to force recompilation of all KLDs due to
change of the LRO control structure size.

Sponsored by: Mellanox Technologies
Reviewed by: gallatin, sbruno, rrs, gnn, transport
Tested by: Netflix
Differential Revision: https://reviews.freebsd.org/D4914

Add support for configuring additional virtual interfaces (VIs) on a port.

Each virtual interface has its own MAC address, queues, and statistics.
The dedicated netmap interfaces (ncxgbeX / ncxlX) were already implemented
as additional VIs on each port. This change allows additional non-netmap
interfaces to be configured on each port. Additional virtual interfaces
use the naming scheme vcxgbeX or vcxlX.

Additional VIs are enabled by setting the hw.cxgbe.num_vis tunable to a
value greater than 1 before loading the cxgbe(4) or cxl(4) driver.
NB: The first VI on each port is the "main" interface (cxgbeX or cxlX).

T4/T5 NICs provide a limited number of MAC addresses for each physical port.
As a result, a maximum of six VIs can be configured on each port (including
the "main" interface and the netmap interface when netmap is enabled).

One user-visible result is that when netmap is enabled, packets received
or transmitted via the netmap interface are no longer counted in the stats
for the "main" interface, but are not accounted to the netmap interface.

The netmap interfaces now also have a new-bus device and export various
information sysctl nodes via dev.n(cxgbe|cxl).X.

The cxgbetool 'clearstats' command clears the stats for all VIs on the
specified port along with the port's stats. There is currently no way to
clear the stats of an individual VI.

Reviewed by: np
MFC after: 1 month
Sponsored by: Chelsio

cxgbe(4): Add a new knob that controls the congestion response of netmap
rx queues. The default is to drop rather than backpressure.

This decouples the congestion settings of NIC and netmap rx queues.

MFC after: 3 days

cxgbe(4): Do not override the the global defaults for congestion drops.
The hw.cxgbe.cong_drop knob is not affected by this change because the
driver sets up congestion drop on a per-queue basis.

MFC after: 3 days

cxgbe: get_fl_payload returns a header mbuf when successful.

MFC after: 3 days

cxgbe: set minimum burst size when fetching freelist buffers to 128B.

MFC after: 3 days

cxgbe(4): provide the exact RSS hash type instead of a catch-all value
to the upper layers.

cxgbe(4): set up congestion management for netmap rx queues.

The hw.cxgbe.cong_drop knob controls the response of the chip when
netmap queues are congested.

cxgbe(4): there is no need to force an "unimplemented" panic needlessly.
The calls to free_nm_txq and free_nm_rxq are made just a few lines prior
to the panic.

cxgbe(4): major tx rework.

a) Front load as much work as possible in if_transmit, before any driver
lock or software queue has to get involved.

b) Replace buf_ring with a brand new mp_ring (multiproducer ring). This
is specifically for the tx multiqueue model where one of the if_transmit
producer threads becomes the consumer and other producers carry on as
usual. mp_ring is implemented as standalone code and it should be
possible to use it in any driver with tx multiqueue. It also has:
- the ability to enqueue/dequeue multiple items. This might become
significant if packet batching is ever implemented.
- an abdication mechanism to allow a thread to give up writing tx
descriptors and have another if_transmit thread take over. A thread
that's writing tx descriptors can end up doing so for an unbounded
time period if a) there are other if_transmit threads continuously
feeding the sofware queue, and b) the chip keeps up with whatever the
thread is throwing at it.
- accurate statistics about interesting events even when the stats come
at the expense of additional branches/conditional code.

The NIC txq lock is uncontested on the fast path at this point. I've
left it there for synchronization with the control events (interface
up/down, modload/unload).

c) Add support for "type 1" coalescing work request in the normal NIC tx
path. This work request is optimized for frames with a single item in
the DMA gather list. These are very common when forwarding packets.
Note that netmap tx in cxgbe already uses these "type 1" work requests.

d) Do not request automatic cidx updates every 32 descriptors. Instead,
request updates via bits in individual work requests (still every 32
descriptors approximately). Also, request an automatic final update
when the queue idles after activity. This means NIC tx reclaim is still
performed lazily but it will catch up quickly as soon as the queue
idles. This seems to be the best middle ground and I'll probably do
something similar for netmap tx as well.

e) Implement a faster tx path for WRQs (used by TOE tx and control
queues, _not_ by the normal NIC tx). Allow work requests to be written
directly to the hardware descriptor ring if room is available. I will
convert t4_tom and iw_cxgbe modules to this faster style gradually.

MFC after: 2 months

cxgbe(4): allow the driver to use rx buffers that do not end on a pack
boundary.

MFC after: 2 weeks

cxgbe(4): Allow for different pad and pack boundaries for different
adapters. Set the pack boundary for T5 cards to be the same as the
PCIe max payload size. The chip likes it this way.

In this revision the driver allocate rx buffers that align on both
boundaries. This is not a strict requirement and a followup commit
will switch the driver to a more relaxed allocation strategy.

MFC after: 2 weeks

Start process of removing the use of the deprecated "M_FLOWID" flag
from the FreeBSD network code. The flag is still kept around in the
"sys/mbuf.h" header file, but does no longer have any users. Instead
the "m_pkthdr.rsstype" field in the mbuf structure is now used to
decide the meaning of the "m_pkthdr.flowid" field. To modify the
"m_pkthdr.rsstype" field please use the existing "M_HASHTYPE_XXX"
macros as defined in the "sys/mbuf.h" header file.

This patch introduces new behaviour in the transmit direction.
Previously network drivers checked if "M_FLOWID" was set in "m_flags"
before using the "m_pkthdr.flowid" field. This check has now now been
replaced by checking if "M_HASHTYPE_GET(m)" is different from
"M_HASHTYPE_NONE". In the future more hashtypes will be added, for
example hashtypes for hardware dedicated flows.

"M_HASHTYPE_OPAQUE" indicates that the "m_pkthdr.flowid" value is
valid and has no particular type. This change removes the need for an
"if" statement in TCP transmit code checking for the presence of a
valid flowid value. The "if" statement mentioned above is now a direct
variable assignment which is then later checked by the respective
network drivers like before.

Additional notes:
- The SCTP code changes will be committed as a separate patch.
- Removal of the "M_FLOWID" flag will also be done separately.
- The FreeBSD version has been bumped.

MFC after: 1 month
Sponsored by: Mellanox Technologies

cxgbe(4): some optimizations in freelist handling.

MFC after: 2 weeks.

cxgbe(4): Fix an off by one error when looking for the BAR2 doorbell
address of an egress queue.

MFC after: 2 weeks

cxgbe(4): minor optimizations in ingress queue processing.

Reorganize struct sge_iq. Make the iq entry size a compile time
constant. While here, eliminate RX_FL_ESIZE and use EQ_ESIZE directly.

MFC after: 2 weeks

cxgbe(4): Keep track of the clusters that have to be freed by the
custom free routine (rxb_free) in the driver. Fail MOD_UNLOAD with
EBUSY if any such cluster has been handed up to the kernel but hasn't
been freed yet. This prevents a panic later when the cluster finally
needs to be freed but rxb_free is gone from the kernel.

MFC after: 1 week

Add missing newline to an error message.

MFC after: 3 days

Simplify r267600, there's no need to distinguish between allocated and
inlined mbufs.

MFC after: 1 week

All mbuf external free functions never fail, so let them be void.

Sponsored by: Nginx, Inc.

cxgbe(4): Update the bundled T4 and T5 firmwares to versions 1.11.27.0.

Obtained from: Chelsio
MFC after: 3 days

Consider the total number of descriptors available (and not just those
that are ready to be reclaimed) when deciding whether to resume tx after
a stall.

MFC after: 3 days

cxgbe(4): Fix bug in the fast rx buffer recycle path. In some cases rx
buffers were getting recycled when they should have been left alone.

MFC after: 3 days

cxgbe(4): Properly account for the freelist buffers used when returning
early from service_iq due to a budget restriction. This fixes a potential
rx hang when using INTx.

MFC after: 3 days

cxgbe(4): Fix a NULL dereference when the very first call to
get_scatter_segment() in get_fl_payload() fails. While here,
fix the code to adjust fl_bufs_used when a failure occurs for
any other scatter segment.

MFC after: 3 days

cxgbe(4): netmap support for Terminator 5 (T5) based 10G/40G cards.
Netmap gets its own hardware-assisted virtual interface and won't take
over or disrupt the "normal" interface in any way. You can use both
simultaneously.

For kernels with DEV_NETMAP, cxgbe(4) carves out an ncxl<N> interface
(note the 'n' prefix) in the hardware to accompany each cxl<N>
interface. These two ifnet's per port share the same wire but really
are separate interfaces in the hardware and software. Each gets its own
L2 MAC addresses (unicast and multicast), MTU, checksum caps, etc. You
should run netmap on the 'n' interfaces only, that's what they are for.

With this, pkt-gen is able to transmit > 45Mpps out of a single 40G port
of a T580 card. 2 port tx is at ~56Mpps total (28M + 28M) as of now.
Single port receive is at 33Mpps but this is very much a work in
progress. I expect it to be closer to 40Mpps once done. In any case
the current effort can already saturate multiple 10G ports of a T5 card
at the smallest legal packet size. T4 gear is totally untested.

trantor:~# ./pkt-gen -i ncxl0 -f tx -D 00:07:43:ab:cd:ef
881.952141 main [1621] interface is ncxl0
881.952250 extract_ip_range [275] range is 10.0.0.1:0 to 10.0.0.1:0
881.952253 extract_ip_range [275] range is 10.1.0.1:0 to 10.1.0.1:0
881.962540 main [1804] mapped 334980KB at 0x801dff000
Sending on netmap:ncxl0: 4 queues, 1 threads and 1 cpus.
10.0.0.1 -> 10.1.0.1 (00:00:00:00:00:00 -> 00:07:43:ab:cd:ef)
881.962562 main [1882] Sending 512 packets every 0.000000000 s
881.962563 main [1884] Wait 2 secs for phy reset
884.088516 main [1886] Ready...
884.088535 nm_open [457] overriding ifname ncxl0 ringid 0x0 flags 0x1
884.088607 sender_body [996] start
884.093246 sender_body [1064] drop copy
885.090435 main_thread [1418] 45206353 pps (45289533 pkts in 1001840 usec)
886.091600 main_thread [1418] 45322792 pps (45375593 pkts in 1001165 usec)
887.092435 main_thread [1418] 45313992 pps (45351784 pkts in 1000834 usec)
888.094434 main_thread [1418] 45315765 pps (45406397 pkts in 1002000 usec)
889.095434 main_thread [1418] 45333218 pps (45378551 pkts in 1001000 usec)
890.097434 main_thread [1418] 45315247 pps (45405877 pkts in 1002000 usec)
891.099434 main_thread [1418] 45326515 pps (45417168 pkts in 1002000 usec)
892.101434 main_thread [1418] 45333039 pps (45423705 pkts in 1002000 usec)
893.103434 main_thread [1418] 45324105 pps (45414708 pkts in 1001999 usec)
894.105434 main_thread [1418] 45318042 pps (45408723 pkts in 1002001 usec)
895.106434 main_thread [1418] 45332430 pps (45377762 pkts in 1001000 usec)
896.107434 main_thread [1418] 45338072 pps (45383410 pkts in 1001000 usec)
...

Relnotes: Yes
Sponsored by: Chelsio Communications.

cxgbe(4): significant rx rework.

- More flexible cluster size selection, including the ability to fall
back to a safe cluster size (PAGE_SIZE from zone_jumbop by default) in
case an allocation of a larger size fails.
- A single get_fl_payload() function that assembles the payload into an
mbuf chain for any kind of freelist. This replaces two variants: one
for freelists with buffer packing enabled and another for those without.
- Buffer packing with any sized cluster. It was limited to 4K clusters
only before this change.
- Enable buffer packing for TOE rx queues as well.
- Statistics and tunables to go with all these changes. The driver's
man page will be updated separately.

MFC after: 5 weeks

cxgbe(4): Use the rx channel map (instead of the tx channel map) as the
congestion channel map.

MFC after: 1 week

cxgbe(4): The T5 allows for a different freelist starvation threshold
for queues with buffer packing. Use the correct value to calculate a
freelist's low water mark.

MFC after: 1 week

Add an option to enable or disable the small RX packet copying that
is done to improve performance of small frames.

When doing RX packing, the RX copying isn't necessarily required.

Reviewed by: np

Read card capabilities after firmware initialization, instead of setting
them up as part of firmware initialization (which the driver gets to do
only if it's the master driver).

Read the range of tids available for the ETHOFLD functionality if it's
enabled.

New is_ftid() and is_etid() functions to test whether a tid falls within
the range of filter tids or ETHOFLD tids respectively.

MFC after: 2 weeks

cxgbe(4): save a copy of the RSS map for each port for the driver's use.

Disable an assertion that relies on some code[1] that isn't in HEAD yet.

[1] http://lists.freebsd.org/pipermail/freebsd-net/2013-August/036573.html

Include necessary headers that now are available due to pollution
via if_var.h.

Sponsored by: Netflix
Sponsored by: Nginx, Inc.

cxgbe(4): Store the log2 of the # of doorbells per BAR2 page for both
ingress and egress queues, and for both T4 and T5. These values are
used by the T4/T5 iWARP driver.

Fix kernel build on amd64 after r256118, since the machine/md_var.h
header is not implicitly included there. So include it explicitly.

Approved by: re (delphij)
Pointy hat to: dim
MFC after: 3 days
X-MFC-With: r256118

Remove redundant declaration of cpu_clflush_line_size in
sys/dev/cxgbe/t4_sge.c, to silence a gcc warning.

Approved by: re (gjb)
MFC after: 3 days

Fix the sysctl that displays whether buffer packing is enabled
or not.

Implement support for rx buffer packing. Enable it by default for T5
cards.

This is a T4 and T5 chip feature which lets the chip deliver multiple
Ethernet frames in a single buffer. This is more efficient within the
chip, in the driver, and reduces wastage of space in rx buffers.

- Always allocate rx buffers from the jumbop zone, no matter what the
MTU is. Do not use the normal cluster refcounting mechanism.
- Reserve space for an mbuf and a refcount in the cluster itself and let
the chip DMA multiple frames in the rest.
- Use the embedded mbuf for the first frame and allocate mbufs on the
fly for any additional frames delivered in the cluster. Each of these
mbufs has a reference on the underlying cluster.

Merge r254386 from user/np/cxl_tuning. Add an INET|INET6 check missing
in said revision.

r254386:
Flush inactive LRO entries periodically.

Add hooks in base cxgbe(4) for the iWARP upper-layer driver. Update a
couple of assertions in the TOE driver as well.

There is no need to hold the freelist lock around alloc/free of
software descriptors. This also silences WITNESS warnings when
the software descriptors are allocated with M_WAITOK.

MFC after: 1 week

Fix previous commit (r253873). "cong" has one bit per channel but the
congestion channel map has 1 nibble per channel. So bits wxyz need to
be blown up into 000w000x000y000z.

Set up congestion manager context properly for T5 based cards.

MFC after: 3 days (will check with re@)

Display SGE tunables in the sysctl tree.

dev.t5nex.0.fl_pktshift: payload DMA offset in rx buffer (bytes)
dev.t5nex.0.fl_pad: payload pad boundary (bytes)
dev.t5nex.0.spg_len: status page size (bytes)
dev.t5nex.0.cong_drop: congestion drop setting

Discussed with: scottl

Add support for packet-sniffing tracers to cxgbe(4). This works with
all T4 and T5 based cards and is useful for analyzing TSO, LRO, TOE, and
for general purpose monitoring without tapping any cxgbe or cxl ifnet
directly.

Tracers on the T4/T5 chips provide access to Ethernet frames exactly as
they were received from or transmitted on the wire. On transmit, a
tracer will capture a frame after TSO segmentation, hw VLAN tag
insertion, hw L3 & L4 checksum insertion, etc. It will also capture
frames generated by the TCP offload engine (TOE traffic is normally
invisible to the kernel). On receive, a tracer will capture a frame
before hw VLAN extraction, runt filtering, other badness filtering,
before the steering/drop/L2-rewrite filters or the TOE have had a go at
it, and of course before sw LRO in the driver.

There are 4 tracers on a chip. A tracer can trace only in one direction
(tx or rx). For now cxgbetool will set up tracers to capture the first
128B of every transmitted or received frame on a given port. This is a
small subset of what the hardware can do. A pseudo ifnet with the same
name as the nexus driver (t4nex0 or t5nex0) will be created for tracing.
The data delivered to this ifnet is an additional copy made inside the
chip. Normal delivery to cxgbe<n> or cxl<n> will be made as usual.

/* watch cxl0, which is the first port hanging off t5nex0. */
# cxgbetool t5nex0 tracer 0 tx0 (watch what cxl0 is transmitting)
# cxgbetool t5nex0 tracer 1 rx0 (watch what cxl0 is receiving)
# cxgbetool t5nex0 tracer list
# tcpdump -i t5nex0 <== all that cxl0 sees and puts on the wire

If you were doing TSO, a tcpdump on cxl0 may have shown you ~64K
"frames" with no L3/L4 checksum but this will show you the frames that
were actually transmitted.

/* all done */
# cxgbetool t5nex0 tracer 0 disable
# cxgbetool t5nex0 tracer 1 disable
# cxgbetool t5nex0 tracer list
# ifconfig t5nex0 destroy

- Make note of interface MTU change if the rx queues exist, and not just
when the interface is up.
- Add a tunable to control the TOE's rx coalesce feature (enabled by
default as it always has been). Consider the interface MTU or the
coalesce size when deciding which cluster zone to use to fill the
offload rx queue's free list. The tunable is:
dev.{t4nex,t5nex}.<N>.toe.rx_coalesce

MFC after: 1 day

Ring the egress queue's doorbell as soon as there are 8 or more
descriptors ready to be processed.

MFC after: 1 day

- Read all TP parameters in one place.
- Read the filter mode, calculate various shifts, and use them
properly during active open (in select_ntuple).

MFC after: 1 day

Cosmetic change (s/wrwc/wcwr/;s/WRWC/WCWR/).

MFC after: 3 days.

Auto-reduce the holdoff timers that are greater than the maximum value
allowed by the hardware.

MFC after: 3 days

cxgbe(4): Add support for Chelsio's Terminator 5 (aka T5) ASIC. This
includes support for the NIC and TOE features of the 40G, 10G, and
1G/100M cards based on the T5.

The ASIC is mostly backward compatible with the Terminator 4 so cxgbe(4)
has been updated instead of writing a brand new driver. T5 cards will
show up as cxl (short for cxlgb) ports attached to the t5nex bus driver.

Sponsored by: Chelsio

cxgbe(4): Ask the card's firmware to pad up tiny CPLs by encapsulating
them in a firmware message if it is able to do so. This works out
better for one of the FIFOs in the chip.

MFC after: 5 days

cxgbe(4): Assume that CSUM_TSO in the transmit path implies CSUM_IP and
CSUM_TCP too. They are all set explicitly by the kernel usually.

While here, fix an unrelated bug where hardware L4 checksum calculation
was accidentally disabled for some IPv6 packets.

Reported by: alfred@
MFC after: 3 days

Provide a statistic to track the number of drops in each of the port's
txq's buf_ring. The aggregate for all the queues of a port is already
provided in ifnet->if_snd.ifq_drops.

MFC after: 3 days.

Mechanically substitute flags from historic mbuf allocator with
malloc(9) flags in sys/dev.

Use global knob in the TP_PARA_REG3 register to disable congestion
drops if the user has chosen this behaviour.

MFC after: 3 days

There is no need to report the same error twice.

MFC after: 3 days

Remove unused item. cxgbe's rx queue's lock was removed a long time ago.

MFC after: 3 days

Initialize various DDP parameters in the main cxgbe(4) driver:

- Setup multiple DDP page sizes. When the driver attempts DDP it will
try to combine physically contiguous pages into regions of these sizes.

- Set the indicate size such that the payload carried in the indicate can
be copied in the header mbuf (and the 16K rx buffer can be recycled).

- Set DDP threshold to the max payload that the chip will coalesce and
deliver to the driver (this is ~16K by default, which is also why the
offload rx queue is backed by 16K buffers). If the chip is able to
coalesce up to the max it's allowed to, it's a good sign that the peer
is transmitting in bulk without any TCP PSH.

MFC after: 2 weeks

Allow for a different handler for each type of firmware message.

MFC after: 2 weeks

The size of the buffers in an Ethernet freelist has to be higher than the
interface's MTU. Initialize such freelists with correct values.

This wasn't a problem for common MTUs (1500 and 9000) as the buffers (2048
and 9216 in size) happened to have enough spare room. I ran into it when
playing around with unusual MTUs.

MFC after: 2 weeks

Convert some fixed parameters to tunables (with reasonable default
values).

- cong_drop specifies what to do on congestion: nothing, backpressure,
or drop.
- fl_pktshift specifies the padding before Ethernet payload.
- fl_pad specifies the boundary upto which to pad Ethernet payload.
- spg_len controls the length of the status page.

MFC after: 2 weeks

Fix a bug in code that calculates the number of the first interrupt
vector for a port. This affected the gigabit ports of T422 cards (the
ones with 2x10G ports and 2x1G ports).

MFC after: will check with re@

- Assign (don't OR) the CSUM_XXX bits to csum_flags in the rx checksum code.
- Fix TSO/TSO4 mixup.
- Add IFCAP_LINKSTATE to the available/enabled capabilities.

cxgbe(4): support for IPv6 TSO and LRO.

Submitted by: bz (this is a modified version of that patch)

cxgbe(4): support for IPv6 hardware checksumming (rx and tx).

Better way to determine the status page length and rx pad boundary.

Do not allocate extra vectors when adapter is not TOE
capable (or toecaps have been disallowed by the user).

+ one very minor unrelated cleanup in t4_sge.c

cxgbe(4): update to firmware interface 1.5.2.0; updates to shared code.

- Updated TOE support in the kernel.

- Stateful TCP offload drivers for Terminator 3 and 4 (T3 and T4) ASICs.
These are available as t3_tom and t4_tom modules that augment cxgb(4)
and cxgbe(4) respectively. The cxgb/cxgbe drivers continue to work as
usual with or without these extra features.

- iWARP driver for Terminator 3 ASIC (kernel verbs). T4 iWARP in the
works and will follow soon.

Build-tested with make universe.

30s overview
============
What interfaces support TCP offload? Look for TOE4 and/or TOE6 in the
capabilities of an interface:
# ifconfig -m | grep TOE

Enable/disable TCP offload on an interface (just like any other ifnet
capability):
# ifconfig cxgbe0 toe
# ifconfig cxgbe0 -toe

Which connections are offloaded? Look for toe4 and/or toe6 in the
output of netstat and sockstat:
# netstat -np tcp | grep toe
# sockstat -46c | grep toe

Reviewed by: bz, gnn
Sponsored by: Chelsio communications.
MFC after: ~3 months (after 9.1, and after ensuring MFC is feasible)

Change the default to not use packet counters to generate rx interrupts.
Rely solely on the timer based mechanism.

Update man page to reflect this change.

MFC after: 1 week

Many updates to cxgbe(4)

- Device configuration via plain text config file. Also able to operate
when not attached to the chip as the master driver.

- Generic "work request" queue that serves as the base for both ctrl and
ofld tx queues.

- Generic interrupt handler routine that can process any event on any
kind of ingress queue (via a dispatch table).

- A couple of new driver ioctls. cxgbetool can now install a firmware
to the card ("loadfw" command) and can read the card's memory
("memdump" and "tcb" commands).

- Lots of assorted information within dev.t4nex.X.misc.* This is
primarily for debugging and won't show up in sysctl -a.

- Code to manage the L2 tables on the chip.

- Updates to cxgbe(4) man page to go with the tunables that have changed.

- Updates to the shared code in common/

- Updates to the driver-firmware interface (now at fw 1.4.16.0)

MFC after: 1 month

Do not clobber the ingress queue's congestion setting.

MFC after: 1 month

- driver ioctl to get SGE context for any given queue.
- sysctls to display the context id, cidx, and pidx of all kinds of queues.

MFC after: 3 days

Allow lazy fill up of freelists.

MFC after: 3 days

Update to firmware interface 1.3.10

MFC after: 1 week

- Specialized ingress queues that take interrupts for other ingress
queues. Try to have a set of these per port when possible, fall back
to sharing a common pool between all ports otherwise.

- One control queue per port (used to be one per hardware channel).

- t4_eth_rx now handles Ethernet rx only.

- sysctls to display pidx/cidx for some queues.

MFC after: 1 week

- Enable per-channel congestion notification.
- Enable PCIe relaxed ordering for all egress queues and rx data buffers.

MFC after: 3 days

sysctl that displays the absolute queue id of an rxq.

T4 packet timestamps.

Reference code that shows how to get a packet's timestamp out of
cxgbe(4). Disabled by default because we don't have a standard way
today to pass this information up the stack.

The timestamp is 60 bits wide and each increment represents 1 tick of
the T4's core clock. As an example, the timestamp granularity is ~4.4ns
for this card:

# sysctl dev.t4nex.0.core_clock
dev.t4nex.0.core_clock: 228125

MFC after: 1 week

T4 packet filtering/steering.

- Enable 5-tuple and every-packet lookup.

- Setup the default filter mode to allow filtering/steering based on IP
protocol, ingress port, inner VLAN ID, IP frag, FCoE, and MPS match
type; all combined together. You can also filter based on MAC index,
Ethernet type, IP TOS/IPv6 Traffic Class, and outer VLAN ID but you'll
have to modify the default filter mode and exclude some of the
match-fields in it.

IPv4 and IPv6 SIP/DIP/SPORT/DPORT are always available in all filter
rules.

- Add driver ioctls to get/set the global filter mode.

- Add driver ioctls to program and delete hardware filters. A couple of
the "switch" actions that rewrite Ethernet and VLAN information and
switch the packet out of another port may not work as the L2 code is not
yet in place. Everything else, including all "drop" and "pass" rules
with RSS or absolute qid, should work.

Obtained from: Chelsio Communications

Always re-arm an iq's interrupt before leaving the handler.

MFC after: 1 week

Ring the freelist doorbell from within refill_fl. While here, fix a bug
that could have allowed the hardware pidx to reach the cidx even though
the freelist isn't empty. (Haven't actually seen this but it was there
waiting to happen..)

MFC after: 1 week

Use the correct free routine when destroying a control queue.

X-MFC after: r220873

- Move all Ethernet specific items from sge_eq to sge_txq. sge_eq is
now a suitable base for all kinds of egress queues.

- Add control queues (sge_ctrlq) and allocate one of these per hardware
channel. They can be used to program filters and steer traffic (and
more).

MFC after: 1 week

Fix a couple of bad races that can occur when a cxgbe interface is taken
down. The ingress queue lock was unused and has been removed as part of
these changes.

- An in-flight egress update from the SGE must be handled before the
queue that requested it is destroyed. Wait for the update to arrive.

- Interrupt handlers must stop processing rx events for a queue before
the queue is destroyed. Events that have not yet been processed
should be ignored once the queue disappears.

MFC after: 1 week

There is no need to request a tx credit flush if such a request is already
pending.

MFC after: 3 days

Do not over-allocate MSI interrupts for the case where each ingress
queue has its own interrupt. If the exact number that we need is not a
power of 2 and we're using MSI, then switch to interrupt multiplexing.

While here, replace the magic numbers with something more readable.

MFC after: 3 days

Fix an error while constructing the table that maps context id -> egress
queue.

MFC after: 1 day

cxgbe shouldn't directly know of the UMA zones where network buffers
come from.

MFC after: 1 week

There is no need to hold an ingress queue's lock while processing its
descriptors.

MFC after: 1 week

Calculate how many descriptors can be reclaimed before calling
reclaim_tx_descs

Tweaks for rx:

- everything related to LRO should be in #ifdef INET blocks
- reorder sge_iq's fields so that the most frequently used are all together
- pull all rx code into t4_intr_data directly
- let go of the ingress queue lock when passing up data
- refill the freelist only if it is short of at least 32 buffers

Store the ifnet rather than the port_info in each txq and rxq struct.

MFC after: 1 week

A txpkts work request should have a valid FID.

MFC after: 1 week

Resume tx immediately in response to an SGE egress update from the hardware.

MFC after: 1 week

cxgbe(4) - NIC driver for Chelsio T4 (Terminator 4) based 10Gb/1Gb adapters.

MFC after: 3 weeks