Age | Commit message (Collapse) | Author | Files | Lines |
|
The fast path almost always has to deal with the real
pointers. Deriving the frame pointer from a frame_index requires a
load of the 32bit frame_index from memory, another 64bit load of the
heap base pointer and some calculations.
Lets store the full pointer instead and do a single 64bit load only.
This helps avoiding problems when the heap is grown and frames are
allocated below vm->heap_aligned_base.
Type: refactor
Change-Id: Ifa6e6e984aafe1e2755bff80f0a4dfcddee3623c
Signed-off-by: Andreas Schultz <andreas.schultz@travelping.com>
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
We have native implementation and we should not maintain both....
Change-Id: Ic09ebffda52cdc733b3cfeff06690e0d3cc08084
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
The function dpdk_ol_flags_extract should return u16
instead of u8.
Change-Id: Id0b08b04c93598818f9a2eee5a88733900320dfa
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
Change-Id: Idc3763c38f5aa638d4f290f4d4730577601d78b8
Signed-off-by: Simon Zhang <yuwei1.zhang@intel.com>
|
|
Change-Id: I65b1af5fa0cec4f9789f91f720d1396d06fa0206
Signed-off-by: Simon Zhang <yuwei1.zhang@intel.com>
|
|
Moved code to the ethernet input node, and the interface output
path(s). Since we no longer skip ethernet-input, there's no reason
for device drivers to know anything about pcap rx tracing, etc.
Change-Id: I08d32fb1b90cbee1bd4f609837d533e047b36fa4
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I01c4f5755d579282773ac227b0bc24f8ddbb2bd1
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
PKT_RX_FDIR is a bit flag, not a bit position.
Change-Id: Ib31ec9257e906b045522fa7c2b515b7b0c13bb32
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
|
|
Change-Id: I9a2e8ea2bf334dd8dabf3d25abbcc91087a43882
Signed-off-by: Dave Barach <dbarach@cisco.com>
|
|
Change-Id: I6aa030429c1740f7376e95daf82fce49efa6716b
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I998658ad7860b23425444e218ce2e1ec655b885a
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Storing buffer in local template seems to be better option....
Change-Id: I1a2fdd68cb956f99a5b36d2cd810fc623e089bcf
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I3555230ef7b7677d6187282fb0c0f02af216e0a0
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I4ec7750ef58363bd8966a16a2baeec6db18b7e9e
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Id4f37f5d4a03160572954a416efa1ef9b3d79ad1
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I085615fde1f966490f30ed5d32017b8b088cfd59
Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
|
|
This significantly reduces need for
...
in multiarch code. Simply constructor macros will jost create static unused
entry if CLIB_MARCH_VARIANT is defined and that will be optimized out by
compiler.
Change-Id: I17d1c4ac0c903adcfadaa4a07de1b854c7ab14ac
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Workaround for lack of driver interrupt support. Also quite handy for
home gateway, laptop/vagrant, other use-cases not requiring maximum
vectors/second for proper operation.
Change-Id: Ifc4b98112450664beef67b89ab8a6940a3bf24b5
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I737dad64bf6dd0743d36500d5cfa1cb1a6594b98
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
ip4 vxlan cli/api (using flow infra) to create flows and enable them on
different hardware (currently tested with i40e)
to offload a vxlan tunnel onto hw:
set flow-offload vxlan hw TwentyFiveGigabitEthernet3/0/0 rx vxlan_tunnel1
to remove offload:
set flow-offload vxlan hw TwentyFiveGigabitEthernet3/0/0 rx vxlan_tunnel1 del
TODO:ipv6 handling
Change-Id: I70e61f792ef8e3f007d03d7df70e97ea4725b101
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
Prior to the change, dpdk plugin assumes xd->device_index is
used both as index for internal dpdk_main->devices array
and DPDK port index to call into DPDK APIs.
However, when running on top of Failsafe PMDs,
DPDK port index range may no longer be contiguous (as noted:
http://dpdk.org/ml/archives/dev/2018-March/092375.html
for related changes in DPDK). Because this, dpdk plugin can
no longer iterate through all available DPDK ports
with a for 0->rte_eth_dev_count() loop and the assumption of
device_index no longer holds.
This is part of initial effort to enable vpp running over
dpdk on failsafe PMD in Microsoft Azure(3/4).
Change-Id: I416fd80f2d40e12e139f8f3492814da98343eae7
Signed-off-by: Rui Cai <rucai@microsoft.com>
|
|
Change-Id: Ibab5e27277f618ceb2d543b9d6a1a5f191e7d1db
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Should cost at most 1 clock per frame when not enabled.
Add "pcap rx trace..." debug CLI, refactored "pcap tx trace" debug CLI
to avoid duplicating code.
Change-Id: I19ac75d1cf94a6a24c98facbf0753381d37963ea
Signed-off-by: Dave Barach <dbarach@cisco.com>
|
|
Change-Id: I1042c0fe179b57a00ce99c8d62cb1bdbe24d9184
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ib3fcc3ceb7f315389bcdecbb7d9632540a5dd6ba
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ibea4a96bdec5e368301a03d8b11a0712fa0265e0
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I387b22427b3f322969bcf32fcfc189123c8ed6ae
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
Change-Id: I16557189aa4a763ec496cb4a45f6e12f2d46971f
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I2794384557c6272fe217269b14a9db09eda19220
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: If174d189de40e6f9ffae99997bba93a2519d9fda
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
dpdk-input was dropping packets with bad ip-checksum on l2 interfaces
Change-Id: Ife5b52766bb71e878b1da6e94ae7b8a1e59fc478
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
Packet trace initiated by dpdk-input node would occasionally
not fully function if next node is ip4-input.
Change packet trace saving order in the quad-loop so "sho trace"
will display trace in packet receive order.
Fold calling of vlib_trace_buffer() into dpdk_add_trace().
Change-Id: I9d7a9bf3b9391f95590e66150b26b0b15912d803
Signed-off-by: John Lo <loj@cisco.com>
|
|
Change-Id: Icd2184dec16d30cdcc689ca37c834b5df2f0a1a3
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
Change-Id: I7110436626352d45ffe0ca71fb88dea2c77ab639
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
dpdk_rx_next_from_packet_start is equivalent to dpdk_rx_next_from_etype and seems to have no side effects
Change-Id: I629dadfbfb35ad1f5c7198e7824883ae4bd6abaa
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
Change-Id: I0a6d1257e391c3b6f7da6498bd5f7d4c545d17e9
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: If581feca0d51d0420c971801aecdf9250c671b36
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Icaf7d7ad47284aea7a56e8006b69f45874d64202
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
For bonded interface in Active/Backup mode (mode 1), we need to
send a GARP/NA packet, if IP address is present, on slave link
state change to up or down to help with route convergence. The
callback from DPDK happens in a separate thread so we need to make
sure RPC call is used to signal the send_garp_na process in the
main thread. Also need to fix DPDK polling so the slave links are
not polled.
Change-Id: If5fd8ea2d28c54dd28726ac403ad366386ce9651
Signed-off-by: John Lo <loj@cisco.com>
|
|
To save space in the first cacheline following is changed:
- total_length_not_including_first_buffer moved to the 2nd cacheline.
This field is used only when VLIB_BUFFER_TOTAL_LENGTH_VALID and
VLIB_BUFFER_NEXT_PRESENT are both set.
- free_list_index is now stored in 4bits inside flags, which
allows up to 16 free lists. In case we need more we can store index
in the 2nd cachelin
Change-Id: Ic8521350819391af470d31d3fa1013e67ecb7681
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
For platforms which do not support u8x32, such as aarch64,
add an alternative implementation of 'dpdk_buffer_init_from_template'.
Change-Id: Ia7e8d0a5985fa5925e063ed6e890208c73e39933
Signed-off-by: Christophe Fontaine <christophe.fontaine@enea.com>
|
|
Change-Id: Id632ff1b30be808d9f270e2f77260391569fbda2
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This pathch deprecates "show dpdk placement" and
"set dpdk placement" CLI commands.
Change-Id: I4e052ec3e8b8e6c54b4816e1e689e5b7a24892db
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
1 - interface-DPO
Used in the Data-plane to change a packet's input interface
2 - MPLS multicast FIB entry
Same as a unicast entry but it links to a replicate not a load-balance DPO
3 - Multicast MPLS tunnel
Update MPLS tunnels to use a FIB path-list to describe the endpoint[s]. Use the path-list to generate the forwarding chain (DPOs) to link to .
4 - Resolve a path via a local label (of an mLDP LSP)
For IP multicast entries to use an LSP in the replication list, we need to decribe the 'resolve-via-label' where the label is that of a multicast LSP.
5 - MPLS disposition path sets RPF-ID
For a interface-less LSP (i.e. mLDP not RSVP-TE) at the tail of the LSP we still need to perform an RPF check. An MPLS disposition DPO performs the MPLS pop validation checks and sets the RPF-ID in the packet.
6 - RPF check with per-entry RPF-ID
An RPF-ID is used instead of a real interface SW if index in the case the IP traffic arrives from an LSP that does not have an associated interface.
Change-Id: Ib92e177be919147bafeb599729abf3d1abc2f4b3
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
This patch deprecates stack-based thread identification,
Also removes requirement that thread stacks are adjacent.
Finally, possibly annoying for some folks, it renames
all occurences of cpu_index and cpu_number with thread
index. Using word "cpu" is misleading here as thread can
be migrated ti different CPU, and also it is not related
to linux cpu index.
Change-Id: I68cdaf661e701d2336fc953dcb9978d10a70f7c1
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
As opposed to rubbish (typically) 14 octets past the ethertype.
Also fix buffer error code setup in dpdk-input node single loop.
Change-Id: Ide7c4097d3bb91e62749ed4e1d69a7b4b90225de
Signed-off-by: Dave Barach <dbarach@cisco.com>
Signed-off-by: John Lo <loj@cisco.com>
|
|
- fix issue caused by assumption that ethertype starts at 0
- intoduce buffer templates to speed-up vlib_buffer_t metadata
initialization
- avoid check for multiseg buffers inside loop if multiseg is disabled
- interleave prefetches to reduce load on L1 cache
Change-Id: I3b76e6d3e1e15ed28f01625edb7fbe9f38112e03
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I94e8737cad9222d24602db4ad03bf43c12c62b79
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: Ifac7d9134d03d79164ce6f06ae9413279bbaadb3
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Remove reliance on DPDK driver provided mbuf packet type in dpdk-input
node as some NIC driver provide misleading information. Now using
ethertype from the packet itself to determine packet type for next node.
Change-Id: Ie7b514a984f9382c29f1a1e3eb423d68f817c064
Signed-off-by: John Lo <loj@cisco.com>
|