This commit contains large contributions from Giuseppe Lettieri and
Stefano Garzarella, is partly supported by grants from Verisign and Cisco,
and brings in the following:
- fix zerocopy monitor ports and introduce copying monitor ports
(the latter are lower performance but give access to all traffic
in parallel with the application)
- exclusive open mode, useful to implement solutions that recover
from crashes of the main netmap client (suggested by Patrick Kelsey)
- revised memory allocator in preparation for the 'passthrough mode'
(ptnetmap) recently presented at bsdcan. ptnetmap is described in
S. Garzarella, G. Lettieri, L. Rizzo;
Virtual device passthrough for high speed VM networking,
ACM/IEEE ANCS 2015, Oakland (CA) May 2015
http://info.iet.unipi.it/~luigi/research.html
- fix rx CRC handing on ixl
- add module dependencies for netmap when building drivers as modules
- minor simplifications to device-specific routines (*txsync, *rxsync)
- general code cleanup (remove unused variables, introduce macros
to access rings and remove duplicate code,
Applications do not need to be recompiled, unless of course
they want to use the new features (monitors and exclusive open).
Those willing to try this code on stable/10 can just update the
sys/dev/netmap/*, sys/net/netmap* with the version in HEAD
and apply the small patches to individual device drivers.
MFC after: 1 month
Sponsored by: (partly) Verisign, Cisco
up to 2 rx/tx queues for the 82574.
Program the 82574 to enable 5 msix vectors, assign 1 to each rx queue,
1 to each tx queue and 1 to the link handler.
Inspired by DragonFlyBSD, enable some RSS logic for handling tx queue
handling/processing.
Move multiqueue handler functions so that they line up better in a diff
review to if_igb.c
Always enqueue tx work to be done in em_mq_start, if unable to acquire
the TX lock, then this will be processed in the background later by the
taskqueue. Remove mbuf argument from em_start_mq_locked() as the work
is always enqueued. (stolen from igb)
Setup TARC, TXDCTL and RXDCTL registers for better performance and stability
in multiqueue and singlequeue implementations. Handle Intel errata 3 and
generic multiqueue behavior with the initialization of TARC(0) and TARC(1)
Bind interrupt threads to cpus in order. (stolen from igb)
Add 2 new DDB functions, one to display the queue(s) and their settings and
one to reset the adapter. Primarily used for debugging.
In the multiqueue configuration, bump RXD and TXD ring size to max for the
adapter (4096). Setup an RDTR of 64 and an RADV of 128 in multiqueue configuration
to cut down on the number of interrupts. RADV was arbitrarily set to 2x RDTR
and can be adjusted as needed.
Cleanup the display in top a bit to make it clearer where the taskqueue threads
are running and what they should be doing.
Ensure that both queues are processed by em_local_timer() by writing them both
to the IMS register to generate soft interrupts.
Ensure that an soft interrupt is generated when em_msix_link() is run so that
any races between assertion of the link/status interrupt and a rx/tx interrupt
are handled.
Document existing tuneables: hw.em.eee_setting, hw.em.msix, hw.em.smart_pwr_down, hw.em.sbp
Document use of hw.em.num_queues and the new kernel option EM_MULTIQUEUE
Thanks to Intel for their continued support of FreeBSD.
Reviewed by: erj jfv hiren gnn wblock
Obtained from: Intel Corporation
MFC after: 2 weeks
Relnotes: Yes
Sponsored by: Limelight Networks
Differential Revision: https://reviews.freebsd.org/D1994
applying them to em(4).
Rely on iterations through the local timer, and the tx queue state to
determine if an actual hang has occurred. Any time a descriptor is used
(packet sent), the tx queue is flagged as busy. Then when txeof runs, it
either clears the flag when all is clean, or resets it to 1 if ANY are
cleaned, if nothing is cleaned it increments the flag.
Local timer simply checks to see if busy ever reaches MAX (10, which
is compile time configurable), and then sets it as HUNG, at that point
there is one more timer cycle in which to have any cleans, if not a
watchdog reset will occur.
Differential Revision: https://reviews.freebsd.org/D2019
Submitted by: jfv
Reviewed by: hiren
Obtained from: Intel Corporation
MFC after: 2 weeks
Relnotes: Yes
Sponsored by: Limelight Networks
- Wrong integer type was specified.
- Wrong or missing "access" specifier. The "access" specifier
sometimes included the SYSCTL type, which it should not, except for
procedural SYSCTL nodes.
- Logical OR where binary OR was expected.
- Properly assert the "access" argument passed to all SYSCTL macros,
using the CTASSERT macro. This applies to both static- and dynamically
created SYSCTLs.
- Properly assert the the data type for both static and dynamic
SYSCTLs. In the case of static SYSCTLs we only assert that the data
pointed to by the SYSCTL data pointer has the correct size, hence
there is no easy way to assert types in the C language outside a
C-function.
- Rewrote some code which doesn't pass a constant "access" specifier
when creating dynamic SYSCTL nodes, which is now a requirement.
- Updated "EXAMPLES" section in SYSCTL manual page.
MFC after: 3 days
Sponsored by: Mellanox Technologies
- Do not ever set a counter to a value. For those counters
that we don't increment, but return directly from hardware
create cases in if_get_counter() method.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
and keep both converted to drvapi and non-converted drivers
compilable.
o Make if_t typedef to struct ifnet *.
o Remove shim functions.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Mostly bugfixes or features developed in the past 6 months,
so this is a 10.1 candidate.
Basically no user API changes (some bugfixes in sys/net/netmap_user.h).
In detail:
1. netmap support for virtio-net, including in netmap mode.
Under bhyve and with a netmap backend [2] we reach over 1Mpps
with standard APIs (e.g. libpcap), and 5-8 Mpps in netmap mode.
2. (kernel) add support for multiple memory allocators, so we can
better partition physical and virtual interfaces giving access
to separate users. The most visible effect is one additional
argument to the various kernel functions to compute buffer
addresses. All netmap-supported drivers are affected, but changes
are mechanical and trivial
3. (kernel) simplify the prototype for *txsync() and *rxsync()
driver methods. All netmap drivers affected, changes mostly mechanical.
4. add support for netmap-monitor ports. Think of it as a mirroring
port on a physical switch: a netmap monitor port replicates traffic
present on the main port. Restrictions apply. Drive carefully.
5. if_lem.c: support for various paravirtualization features,
experimental and disabled by default.
Most of these are described in our ANCS'13 paper [1].
Paravirtualized support in netmap mode is new, and beats the
numbers in the paper by a large factor (under qemu-kvm,
we measured gues-host throughput up to 10-12 Mpps).
A lot of refactoring and additional documentation in the files
in sys/dev/netmap, but apart from #2 and #3 above, almost nothing
of this stuff is visible to other kernel parts.
Example programs in tools/tools/netmap have been updated with bugfixes
and to support more of the existing features.
This is meant to go into 10.1 so we plan an MFC before the Aug.22 deadline.
A lot of this code has been contributed by my colleagues at UNIPI,
including Giuseppe Lettieri, Vincenzo Maffione, Stefano Garzarella.
MFC after: 3 days.
not done after the call to m_defrag(). This fixes a problem
where m_pullup() would prepend an mbuf to the list created
by m_defrag() making the chain greater than 32 again.
Tested by: rcarter@pinyon.org
Reviewed by: yongari, jfv
MFC after: 2 weeks
These changes prevent sysctl(8) from returning proper output,
such as:
1) no output from sysctl(8)
2) erroneously returning ENOMEM with tools like truss(1)
or uname(1)
truss: can not get etype: Cannot allocate memory
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.
Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.
MFC after: 2 weeks
Sponsored by: Mellanox Technologies
NULL to determine if bus_dmamap_unload() or bus_dmamem_free() should be
called. Instead, check the associated bus and virtual addresses.
- Don't clear static DMA maps to NULL.
Reviewed by: jfv
Most relevant features:
- netmap emulation on any NIC, even those without native netmap support.
On the ixgbe we have measured about 4Mpps/core/queue in this mode,
which is still a lot more than with sockets/bpf.
- seamless interconnection of VALE switch, NICs and host stack.
If you disable accelerations on your NIC (say em0)
ifconfig em0 -txcsum -txcsum
you can use the VALE switch to connect the NIC and the host stack:
vale-ctl -h valeXX:em0
allowing sharing the NIC with other netmap clients.
- THE USER API HAS SLIGHTLY CHANGED (head/cur/tail pointers
instead of pointers/count as before). This was unavoidable to support,
in the future, multiple threads operating on the same rings.
Netmap clients require very small source code changes to compile again.
On the plus side, the new API should be easier to understand
and the internals are a lot simpler.
The manual page has been updated extensively to reflect the current
features and give some examples.
This is the result of work of several people including Giuseppe Lettieri,
Vincenzo Maffione, Michio Honda and myself, and has been financially
supported by EU projects CHANGE and OPENLAB, from NetApp University
Research Fund, NEC, and of course the Universita` di Pisa.
The problems do not affect bouncing busdma in a visible way, but are
critical for the dmar backend.
- The bus_dmamap_create(9) is not documented to take BUS_DMA_NOWAIT flag.
- Unload descriptor map after receive.
- Do not reset descriptor map to NULL, bus_dmamap_load(9) requires
valid map, and also this leaks the map.
Reported and tested by: pho
Approved by: jfv
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
This includes the following:
- use separate memory regions for VALE ports
- locking fixes
- some simplifications in the NIC-specific routines
- performance improvements for the VALE switch
- some new features in the pkt-gen test program
- documentation updates
There are small API changes that require programs to be recompiled
(NETMAP_API has been bumped so you will detect old binaries at runtime).
In particular:
- struct netmap_slot now is 16 bytes to support an extra pointer,
which may save one data copy when using VALE ports or VMs;
- the struct netmap_if has two extra fields;
MFC after: 3 days
to this event, adding if_var.h to files that do need it. Also, include
all includes that now are included due to implicit pollution via if_var.h
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
command register. The lazy BAR allocation code in FreeBSD sometimes
disables this bit when it detects a range conflict, and will re-enable
it on demand when a driver allocates the BAR. Thus, the bit is no longer
a reliable indication of capability, and should not be checked. This
results in the elimination of a lot of code from drivers, and also gives
the opportunity to simplify a lot of drivers to use a helper API to set
the busmaster enable bit.
This changes fixes some recent reports of disk controllers and their
associated drives/enclosures disappearing during boot.
Submitted by: jhb
Reviewed by: jfv, marius, achadd, achim
MFC after: 1 day
the changes. Make sure that pci_alloc_msix() does give us the vectors
we need and fall back to MSI when it doesn't, also release any that
were allocated when insufficient.
MFC after: 3 days
ixgbe driver. As it was, when building them as a module INET
and INET6 are not defined. In these drivers it does not cause
a panic, however it does result in different behavior in the
ioctl routine when you are using a module vs static, and I
think the behavior should be the same.
MFC after: 3 days
irrespective of the setting of lem_rx_process_limit, while
giving a chance to the taskqueue scheduler to act after
each chunk.
This makes lem_rxeof similar to the one in if_em.c and if_igb.c .
if_lem.c and if_em.c: add a sysctl to manually configure the
'itr' moderation register.
Approved by: Jack Vogel
well as enabled when necessary. And simplify the checksum routine
itself, adding UDP bit to the test. Thanks to Kevin Lo for pointing
out the problems and code suggestions.
done in ixgbe, thanks to Mike Karels for this fix. When exiting
promiscuous mode MPE bit was being unconditionally cleared, this
should not be done if we are in MAX multicast groups.
- bear with me, there are lots of white space changes, I would not
do them, but I am a mere consumer of this stuff and if these drivers
are to stay in shape they need to be taken.
em driver changes: support for the new i217/i218 interfaces
igb driver changes:
- TX mq start has a quick turnaround to the stack
- Link/media handling improvement
- When link status changes happen the current flow control state
will now be displayed.
- A few white space/style changes.
lem driver changes:
- the shared code uncovered a bogus write to the RLPML register
(which does not exist in this hardware) in the vlan code,this
is removed.
of the newer drivers. The basic problem was
that the driver was pulling the mbuf off the
drbr ring and then when sending with xmit(), encounting
a full transmit ring. Thus the lower layer
xmit() function would return an error, and the
drivers would then append the data back on to the ring.
For TCP this is a horrible scenario sure to bring
on a fast-retransmit.
The fix is to use drbr_peek() to pull the data pointer
but not remove it from the ring. If it fails then
we either call the new drbr_putback or drbr_advance
method. Advance moves it forward (we do this sometimes
when the xmit() function frees the mbuf). When
we succeed we always call advance. The
putback will always copy the mbuf back to the top
of the ring. Note that the putback *cannot* be used
with a drbr_dequeue() only with drbr_peek(). We most
of the time, in putback, would not need to copy it
back since most likey the mbuf is still the same, but
sometimes xmit() functions will change the mbuf via
a pullup or other call. So the optimial case for
the single consumer is to always copy it back. If
we ever do a multiple_consumer (for lagg?) we
will need a test and atomic in the put back possibly
a seperate putback_mc() in the ring buf.
Reviewed by: jhb@freebsd.org, jlv@freebsd.org
tree used it incorrectly, which lead to inaccurate overrated
if_obytes accounting. The drbr(9) used to update ifnet stats on
drbr_enqueue(), which is not accurate since enqueuing doesn't
imply successful processing by driver. Dequeuing neither mean
that. Most drivers also called drbr_stats_update() which did
accounting again, leading to doubled if_obytes statistics. And
in case of severe transmitting, when a packet could be several
times enqueued and dequeued it could have been accounted several
times.
o Thus, make drbr(9) API thinner. Now drbr(9) merely chooses between
ALTQ queueing or buf_ring(9) queueing.
- It doesn't touch the buf_ring stats any more.
- It doesn't touch ifnet stats anymore.
- drbr_stats_update() no longer exists.
o buf_ring(9) handles its stats itself:
- It handles br_drops itself.
- br_prod_bytes stats are dropped. Rationale: no one ever
reads them but update of a common counter on every packet
negatively affects performance due to excessive cache
invalidation.
- buf_ring_enqueue_bytes() reduced to buf_ring_enqueue(), since
we no longer account bytes.
o Drivers handle their stats theirselves: if_obytes, if_omcasts.
o mlx4(4), igb(4), em(4), vxge(4), oce(4) and ixv(4) no longer
use drbr_stats_update(), and update ifnet stats theirselves.
o bxe(4) was the most correct driver, it didn't call
drbr_stats_update(), thus it was the only driver accurate under
moderate load. Now it also maintains stats itself.
o ixgbe(4) had already taken stats from hardware, so just
- drop software stats updating.
- take multicast packet count from hardware as well.
o mxge(4) just no longer needs NO_SLOW_STATS define.
o cxgb(4), cxgbe(4) need no change, since they obtain stats
from hardware.
Reviewed by: jfv, gnn
adapter->dropped_pkts instead of if_ierrors because if_ierrors is
overwritten by hw stats collection.
Submitted by: Andrew Boyer <aboyer@averesystems.com>
Reviewed by: Jack F Vogel <jfv@freebsd.org>
MFC after: 2 weeks
#defines. This also has the advantage that it makes the names more
compact, iand also allows us to correct the non-uniform naming of
the PCIM_LINK_* defines, making them all consistent amongst themselves.
This is a mostly mechanical rename:
s/PCIR_EXPRESS_/PCIER_/g
s/PCIM_EXP_/PCIEM_/g
s/PCIM_LINK_/PCIEM_LINK_/g
When this is MFC'd, #defines will be added for the old names to assist
out-of-tree drivers.
Discussed with: jhb
MFC after: 1 week
"m_getjcl:invalid cluster type" that occurred some
time back with the igb driver. This happens often when
booting over the net. I believe the NIC hardware is left
in a warm state when handed over to the driver, and a stray
RX interrupt happens earlier than the code is prepared for
it to happen. This change was verified to fix the problem,
its kind of a bandaid... but it is similar to what was done
in the igb code.
setting in the igb and em driver. This was necessitated by
a shared code change that I was given late in the game, a data
type changed from bool to int, in the last update I dealt with
it by a cast, but it was pointed out (thanks jhb) that there
was a potential problem with this. John suggested this safer
approach, and it is fine with me...
MFC after:2 days (to catch the 9.1 update)
shared code update and small changes in core required
Add support for new i210/i211 devices
Improve queue calculation based on mac type
MFC after:5 days
- Do not define the foo_start() methods or set if_start in the ifnet if
multiq transmit is enabled. Also, set if_transmit and if_qflush before
ether_ifattach rather than after when multiq transmit is enabled. This
helps to ensure that the drivers never try to mix different transmit
methods.
- Properly restart transmit during resume. igb(4) was not restarting it
at all, and em(4) was restarting even if the link was down and was
calling the wrong method if multiq transmit was enabled.
- Remove all the 'more' handling for transmit completions. Transmit
completion processing does not have a processing limit, so it always
runs to completion and never has more work to do when it returns.
Instead, the previous code was returning 'true' anytime there were
packets in the queue that weren't still in the process of being
transmitted. The effect was that the driver would continuously
reschedule a task to process TX completions in effect running at 100%
CPU polling the hardware until it finished transmitting all of the
packets in the ring. Now it will just wait for the next TX completion
interrupt.
- Restart packet transmission when the link becomes active.
- Fix the MSI-X queue interrupt handlers to restart packet transmission if
there are pending packets in the relevant software queue (IFQ or buf_ring)
after processing TX completions. This is the root cause for the OACTIVE
hangs as if the MSI-X queue handler drained all the pending packets from
the TX ring, nothing would ever restart it. As such, remove some
previously-added workarounds to reschedule a task to poll the TX ring
anytime OACTIVE was set.
Tested by: sbruno
Reviewed by: jfv
MFC after: 1 week
USERSPACE:
1. add support for devices with different number of rx and tx queues;
2. add better support for zero-copy operation, adding an extra field
to the netmap ring to indicate how many buffers we have already processed
but not yet released (with help from Eddie Kohler);
3. The two changes above unfortunately require an API change, so while
at it add a version field and some spares to the ioctl() argument
to help detect mismatches.
4. update the manual page for the two changes above;
5. update sample applications in tools/tools/netmap
KERNEL:
1. simplify the internal structures moving the global wait queues
to the 'struct netmap_adapter';
2. simplify the functions that map kring<->nic ring indexes
3. normalize device-specific code, helps mainteinance;
4. start exploring the impact of micro-optimizations (prefetch etc.)
in the ixgbe driver.
Use 'legacy' descriptors on the tx ring and prefetch slots gives
about 20% speedup at 900 MHz. Another 7-10% would come from removing
the explict calls to bus_dmamap* in the core (they are effectively
NOPs in this case, but it takes expensive load of the per-buffer
dma maps to figure out that they are all NULL.
Rx performance not investigated.
I am postponing the MFC so i can import a few more improvements
before merging.
Introduce some functions to map NIC ring indexes into netmap ring
indexes and vice versa. This way we can implement the bound
checks only in one place (and hopefully in a correct way).
On passing, make the code and comments more uniform across the
various drivers.
