Age | Commit message (Collapse) | Author | Files | Lines |
|
This hash table intends to provide an alternative to the widely
used bihash table in places where either:
- Hash entry timeout is required
- The hash table data does not fit in CPU cache
Although the bihash table is very fast, each lookup requires
accessing two cache lines in a serialized fashion. It works fine
when the hash table is in cache, but hits a wall when it does not.
The 'flowhash' table uses a simplified design (at the cost of a
less good bucket auto-scaling) where each access only requires
a single memory lookup (in the absence of collision). The hash
table also uses a reduced number of registers.
In practice, a VPP node implementing a stateful feature would
typically:
- prefetch buffer metadata (in-cache)
- prefetch packet header (in-cache)
- compute hash & prefetch hash bucket (possibly in RAM)
- read/write key and value from bucket
Using this hash table, it is possible to pipeline accesses in a way
that does not exhaust CPU's line field buffers, even when the
requested value is located in RAM (i.e. not in cache).
Measurements showed it was possible to scale to tens of millions
of flows (with a full 5-tuple matching and 32B value, i.e. 1
cache line per flow) with no performance degradation when
the hash table grows to the point it doesn't fit in cache anymore.
I have used this table in a couple of non-open-sourced projects,
but think it might be useful to lb, nat, and possibly other VPP
subsystems.
More information in the .h file.
Change-Id: I2b13dde0eabd868b75da1cedbfca0bf74d705102
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
|
|
Change-Id: Ibc252d9ed595be955790ec1c97d8730e43ad89b2
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Taken from DPDK, also AVX2 variant updated to be in sync with DPDK
version.
Change-Id: I8a42e4141a5a1a8cfbee328b07bd0c9b38a9eb05
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I760b482b9de457bbb17de817db7079b57d3f5ec1
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
This is an all-purpose octet-vector memory hash, intended as a
thread-safe replacement for hash_create_mem / hash_create_string. All
u8 * key vectors are memorized by the hash table.
Change-Id: I22944daea8fda07dde8ba118a6529a6d382491f9
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I78215041588014e9e5c3599c60471ced610735bb
Signed-off-by: Klement Sekera <ksekera@cisco.com>
|
|
Change-Id: Iff33694fc42cc3bcc73cf1372339053a6365039c
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This will allow us to use this code in client libraries without vlib.
Change-Id: I8557b752496841ba588aa36b6082cbe2cd1867fe
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Simply call pool_init_fixed(...) before using the pool. Note that
fixed, preallocated pools live in individually-mmap'ed address
segments, except for the free element bitmap. A large fixed pool can
exceed 4gb.
Fix tcp buffer allocator leak, remove broken assert
Change-Id: I4421082e12a77c41c6e20f7747f3150dcd01fc26
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
According to Maciek, the easiest way to leverage the csit "performance
trend" job is to actually merge the patch once verified. Manual
testing indicates that the patch improves l2 path performance. Other
use-cases are TBD. It's possible that we'll need to back out the patch
depending on what happens.
Change-Id: Ic0a0363de35ef9be953ad7709c57c3936b73fd5a
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
This is preparation for new C API. Moving common stuff to separate
headers reduces dependency issues.
Change-Id: Ie7adb23398de72448e5eba6c1c1da4e1bc678725
Signed-off-by: Klement Sekera <ksekera@cisco.com>
|
|
prep work for s/timing_wheel/tw_timer/ in the vlib process model
Change-Id: I763f4968a8fce1764a3778b12def0afbd30086b1
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
32-bit code still can use crc32c instructions, but it operates
on 32 registers
Change-Id: I9bb6b0b59635d6ea6a753584676ebcf59c8f6584
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I86089e9bb604adfc260a111685001be1c897ce53
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I49e5ce0aae6e4ff634024387ceaf7dbc432a0351
Signed-off-by: Dave Barach <dave@barachs.net>
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: Ia25a8827ed94877e8fe6c0b2ff6d05c1568eb0e1
Signed-off-by: Gabriel Ganne <gabriel.ganne@enea.com>
|
|
Change-Id: I5499dd6b768425a56936afae50bd578620c83d30
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
- IPv[46] mfib tables with support for (*,G/m), (*,G) and (S,G) exact and longest prefix match
- Replication represented via a new replicate DPO.
- RPF configuration and data-plane checking
- data-plane signals sent to listening control planes.
The functions of multicast forwarding entries differ from their unicast conterparts, so we introduce a new mfib_table_t and mfib_entry_t objects. However, we re-use the fib_path_list to resolve and build the entry's output list. the fib_path_list provides the service to construct a replicate DPO for multicast.
'make tests' is added to with two new suites; TEST=mfib, this is invocation of the CLI command 'test mfib' which deals with many path add/remove, flag set/unset scenarios, TEST=ip-mcast, data-plane forwarding tests.
Updated applications to use the new MIFB functions;
- IPv6 NS/RA.
- DHCPv6
unit tests for these are undated accordingly.
Change-Id: I49ec37b01f1b170335a5697541c8fd30e6d3a961
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
- remove unused stuff
- add --quiet flag to libtool
- avoid building some tests programs when tests are not enabled
Change-Id: Ie34aeec1a598ad811256a00354f66cfddae9d0b9
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I7b51f88292e057c6443b12224486f2d0c9f8ae23
Signed-off-by: Damjan Marion <damarion@cisco.com>
|