Use vm_guest == VM_GUEST_HV is not enough to determine whether FreeBSD
is running on Hyper-V or not. What a mess.
Reported by: smokehydration tutanota com
Sponsored by: Microsoft OSTC
Suggested by: jhb
Reviewed by: Dexuan Cui <decui microsoft com>, Jun Su <junsu microsoft com>
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5850
First of all sema_post() can't be called w/ spinlock, and the channel
message queue processing is not on hot code path, i.e. spinlock is not
necessary.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: sephe, Dexuan Cui <decui microsoft com>
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5812
Since atomic_thread_fence_seq_cst() will become compiler fence on UP kernel.
Reviewed by: kib, Dexuan Cui <decui microsoft com>
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5852
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
The i8254 simulation in Hyper-V is kinda broken and is not available
in Generation 2 Hyper-V VMs, so Hyper-V timer must be registered early
enough so that it can be used to do the TSC freq calibration.
This fixes the notorious warning like this:
calcru: runtime went backwards from 50 usec to 25 usec for pid 0 (kernel)
Submitted by: Dexuan Cui <decui microsoft com>
Reviewed by: kib, sephe
Tested by: kib, sephe
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5778
Using one taskqueue does not work, since the EOM MSR must be written
on the msg's owner CPU.
Noticed by: Jun Su <junsu microsoft com>
Discussed with: Jun Su <junsu microsoft com>, Dexuan Cui <decui microsoft com>
MFC after: 1 week
Sponsored by: Microsoft OSTC
Most often, hv_vmbus_post_message() doesn't fail. However, it fails
intermittently when GPADLs of large shared memory is to be established
with the host, e.g. on the hn(4) attach path: a GPADL of 15MB sendbuf
is created, for which lots of messages will be flooded to the host.
The host side tries to throttle the message rate by returning
HV_STATUS_INSUFFICIENT_BUFFERS.
Before this commit, we do several retries for failed messages, but the
delay between each retry is pretty/too low, which will cause sporadic
message posting failure. We now use large delay (>=1ms) between each
retry to fix the message posting failure.
Submitted by: Dexuan Cui <decui microsoft com>
Reviewed by: sephe
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5715
This mainly used to improve ACK timeliness when multiple RX rings
are enabled.
This value gives the best performance in both Azure and Hyper-V
environment, w/ both 10Ge and 40Ge using non-{INVARIANTS,WITNESS}
kernel.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5691
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: Dexuan Cui <decui microsoft com>, sephe
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5215
This gets rid of the per-cpu SWIs.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: Dexuan Cui <decui microsoft com>, sephe
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5215
Using the same message slot as the other types of the messages has
the side effect that the event timer message could be deferred to
the swi threads to run (lacking of trapframe and the original code
didn't even handle that, so the event timer was actually broken).
As of this commit we use an independent message slot for event timer,
so that we could handle all of event timer messages in the interrupt
handler directly. Note, the message slot for event timer is still
bind to the same interrupt vector as the other types of messages.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: sephe
Discussed with: Jun Su <junsu microsoft com>, Dexuan Cui <decui microsoft com>
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5696
Submitted by: Ju Sun <junsu microsoft com>
Reviewed by: Dexuan Cui <decui microsoft com>, sephe
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5651
So that functions shared w/ attach path could use if_printf().
While I'm here, remove unnecessary if_dunit and if_dname assignment.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5576
Each channel contains one RX ring and one TX ring. And we
try to distribute the channels to different evenly.
Note: Currently we don't have enough information to extract
the RSS type and RSS hash value from the received packets.
This greatly improves the TX/RX performance for 8 virtual CPU
Hyper-V over 10Ge: it can max out 10Ge for TCP when multiple
RX/TX rings are enabled.
This almost doubles the TX/RX performance for locally connected
Hyper-Vs: was 6Gbps w/ 128 TCP streams, now 11Gbps w/ multiple
RX/TX rings enabled.
It is not enabled by default; it will be switched on after more
tests.
Collaborated with: Hongjiang Zhang <honzhan microsoft com>
MFC after: 2 week
Sponsored by: Microsoft OSTC
And since the host may not being able to allocate the # of rings
requested by us, save the # of rings allocated by the host in the
ring_inuse counters; use ring_inuse counters for run time operation.
This paves the way for the upcoming vRSS support.
MFC after: 1 week
Sponsored by: Microsoft OSTC
And use it for cpu0 assignment; it does not sound right to assume that
cpu0 maps to vcpu0. And this factored out function will be exposed to
drivers, if driver specific CPU binding is needed, e.g. hn(4).
Move default cpu select after saving channel offer message. This makes
sure that all useful information of the channel has been setup.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5504
The renamed function create a sysctl tree for channel, and many
non-statistics nodes exists, so don't claim it only adds sysctl
nodes for statistics.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5503
taskqueue_enqueue() was changed to support both fast and non-fast
taskqueues 10 years ago in r154167. It has been a compat shim ever
since. It's time for the compat shim to go.
Submitted by: Howard Su <howard0su@gmail.com>
Reviewed by: sephe
Differential Revision: https://reviews.freebsd.org/D5131
So that the host could dispatch the TX done back to this TX ring's
owner channel
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5498
It would serve as a debug tool, if the shared buffer ring's indices
stopped updating.
Submitted by: HongJiang Zhang <honzhan microsoft com>
Reviewed by: sephe, Jun Su <junsu microsoft com>
Modified by: sephe
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5402
sfence only makes sure about the store-store order, which is not
sufficient here. Use atomic_thread_fence_seq_cst() as suggested
jhb and kib (a locked op in the nutshell, which should have the
Reviewed by: jhb, kib, Jun Su <junsu microsoft com>
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5436
Chimney sending buffer still needs conversion, which will be done
along with the upcoming vRSS support.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5457
This fixes the TX/RX ring selection for TX/RX done.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5454
This is preamble to associate the TX/RX rings to their channel.
While I'm here, revoke unused netvsc_recv_rollup.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5453
This is the preamble to pass channel back to hn(4) upon TX/RX done.
Reviewed by: Hongjiang Zhang <honzhan microsoft com>
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5452
And unregister hv_device only for primary channels, who own the hv_device.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5451
Split heartbeat, shutdown and timesync out of utils code
and name them properly.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: adrian, sephe, Hongjiang Zhang <honzhan microsoft com>
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5216
IF_PREPEND promises out-of-order packet sending when the TX desc list
is depleted. It was overlooked and copied blindly when the transmission
path was partially rewritten.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5386
It will be shared w/ the upcoming ifnet.if_transmit method
implementation.
No functional change.
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5385
need to include it explicitly when <vm/vm_param.h> is already included.
Suggested by: alc
Reviewed by: alc
Differential Revision: https://reviews.freebsd.org/D5379
It is only used in hv_netvsc_drv_freebsd.c; and rename it to hn_tx_done()
mainly to reserve "xmit" for ifnet.if_transmit implement.
While I'm here, remove unapplied comment.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5345
Preamble to implement the ifnet.if_transmit method.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5344
Tested on Windows Server 2012.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5317
It will be used to help tracking host side transmission ring selection
issue; and it will be turned on by default, once we have concrete result.
Reviewed by: adrian, Jun Su <junsu microsoft com>
Approved by: adrian (mento)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5316
So one spinlock is avoided, which would be potentially dangerous for
virtual machine, if the spinlock holder was scheduled out by the host,
as noted by royger.
Old spinlock based txdesc list is still kept around, so we could have
a safe fallback.
No performance regression nor improvement is observed.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5290
This paves the way for upcoming vRSS stuffs and eases more code cleanup.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5283
Performance stays same; so no need to use fast taskqueue here.
Suggested by: royger
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5282
This also eases experiment on the non-fast taskqueue.
Reviewed by: adrian, Jun Su <junsu microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5276
This paves the way for upcoming vRSS stuffs and eases more code cleanup.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5275
And use SYSCTL+CTLFLAG_RDTUN for them.
Suggested by: adrian
Reviewed by: adrian, Hongjiang Zhang <honzhan microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5274
This one gives the best performance so far.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5273
It is off by default. This eases further experimenting on this driver.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5272
Set TCP ACK append limit to 1, i.e. aggregate 2 ACKs at most. Aggregating
anything more than 2 hurts TCP sending performance in hyperv. This
significantly improves the TCP sending performance when the number of
concurrent connetion is low (2~8). And it greatly stabilizes the TCP
sending performance in other cases.
Set TCP data segments aggregation length limit to 37500. Without this
limitation, hn(4) could aggregate ~45 TCP data segments for each
connection (even at 64 or more connections) before dispatching them to
socket code; large aggregation slows down ACK sending and eventually
hurts/destabilizes TCP reception performance. This setting stabilizes
and improves TCP reception performance for >4 concurrent connections
significantly.
Make them sysctls so they could be adjusted.
Reviewed by: adrian, gallatin (previous version), hselasky (previous version)
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5185
We will eventually convert them to use busdma.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: adrian, sephe, Dexuan Cui <decui microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5087
And convert rndis non-hot path spinlock to mutex.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: adrian, sephe
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5081
HyperV code was ported from Linux. There is an implementation of
work queue called hv_work_queue. In FreeBSD, taskqueue could be
used for the same purpose. Convert all the consumer of hv_work_queue
to use taskqueue, and remove work queue implementation.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: adrian, Hongjiang Zhang <honzhan microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4963
It is off by default. This eases more experiment on hn(4).
Reviewed by: adrian, Hongjiang Zhang <honzhan microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5175
This significantly increases LRO aggregation ratio when there are
large amount of connections (improves reception performance a lot).
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5167
hn(4) only has one RX ring currently, so default 8 LRO entries
are too small.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5166
We lost half of the chimney sending space, because we mis-used
ffs() on a 64 bits mask, where ffsl() should be used.
While I'm here:
- Use system atomic operation instead.
- Stringent chimney sending index assertion.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5159
It will be shared w/ upcoming ifnet.if_transmit implementaion.
No functional changes.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5158
So that:
- TCP/IP stack will not do unnecessary IP header checksum for TSO
packets.
- Reduce guest load for non-TSO IP packets.
Reviewed by: adrian
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5099
- For non-TSO offloading, we don't need to access mbuf to know
which csum offloading is requested, we can just use the
CSUM_{IP,TCP,UDP} in the csum_flags.
- For TSO offloading, we still can depend on CSUM_{TSO4,TSO6}
in the csum_flags to tell whether the TSO packet is an IPv4
TSO packet or an IPv6 TSO packet.
This streamlines csum offloading handling (remove the two goto)
and allows us the nuke the unnecessary get_transport_proto_type().
Reviewed by: adrian, Hongjiang Zhang <honzhan microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5098
- Record csum features in softc, so we don't need to duplicate the
logic from attach path to ioctl path.
- Protect if_capenable and if_hwassist changes by main lock.
- Prefer turn on/off bits in if_hwassist explicitly instead of using
XOR.
Reviewed by: adrian, Hongjiang Zhang <honzhan microsoft com>
Approved by: adrian (mentor)
MFC after: 1 week
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5085
- Use taskqueue instead of swi for event handling.
- Scan the interrupt flags in filter
- Disable ringbuffer interrupt mask in filter to ensure no unnecessary
interrupts.
Submitted by: Jun Su <junsu microsoft com>
Reviewed by: adrian, sephe, Dexuan <decui microsoft com>
Approved by: adrian (mentor)
MFC after: 2 weeks
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4920
- Avoid main lock contention by trylock for if_start, if that fails,
schedule TX taskqueue for if_start
- Don't do direct sending if the packet to be sent is large, e.g.
TSO packet.
This change gives me stable 9.1Gbps TCP sending performance w/ TSO
over a 10Gbe directly connected network (the performance fluctuated
between 4Gbps and 9Gbps before this commit). It also improves non-
TSO TCP sending performance a lot.
Reviewed by: adrian, royger
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5074
The page information array could contain up to 32 elements (i.e. 512B).
And on network side w/ TSO, 11+ (176B+) elements, i.e. ~44K TSO packet,
in the page information array is quite common.
This saves us some cpu cycles.
Reviewed by: adrian, delphij
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4992
According to all available information, VMSWITCH always does the
TCP segment checksum verification before sending the segment to
guest.
Reviewed by: adrian, delphij, Hongjiang Zhang <honzhan microsoft com>
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4991
All used fields are setup one by one, so there is no need to zero
out this large struct.
While I'm here, move the stack variable near its usage.
Reviewed by: adrian, delphij, Jun Su <junsu microsoft com>
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4978
While I'm here, move stack variables near their usage.
Reviewed by: adrian, delphij, Jun Su <junsu microsoft com>
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4977
- Avoid unnecessary malloc/free on transmission path.
- busdma(9)-fy transmission path.
- Properly handle IFF_DRV_OACTIVE. This should fix the network
stalls reported by many.
- Properly setup TSO parameters.
- Properly handle bpf(4) tapping. This 5 times the performance
during TCP sending test, when there is one bpf(4) attached.
- Allow size of chimney sending be tuned on a running system.
Default value still needs more test to determine.
Reviewed by: adrian, delphij
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4972
Windows 10 and Window 2016 will return all zero inquiry data for
non-existing slots. If we dispatched them, then a lot of useless
(0 sized) disks would be created. So we verify the returned inquiry
data (valid type, non-empty vendor/product/revision etc.), before
further dispatching.
Minor white space cleanup and wording fix.
Submitted by: Hongjiang Zhang <honzhan microsoft com>
Reviewed by: adrian, sephe, Jun Su <junsu microsoft com>
Approved by: adrian (mentor)
Modified by: sephe
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4928
Vmbus event handler will need to find the channel by its relative
id, when software interrupt for event happens. The original lookup
searches the channel list, which is not very efficient. We now
create a table indexed by the channel relative id to speed up
the channel lookup.
Submitted by: Hongjiang Zhang <honzhan microsoft com>
Reviewed by: delphij, adrain, sephe, Dexuan Cui <decui microsoft com>
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4802
- 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
If the NVSP protocol version is not greater than NVSP_PROTOCOL_VERSION_2,
then the recv buffer size is 15MB, otherwise the buffer size is 16MB.
Submitted by: Hongjiang Zhang <honzhan microsoft com>
Reviewed by: royger, Dexuan Cui <decui microsoft com>, adrian
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4814
Submitted by: Howard Su <howard0su gmail com>
Reviewed by: royger, Dexuan Cui <decui microsoft com>, adrian
Approved by: adrian (mentor)
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D4693
