This commit allow use od cooperative multitasking with multiple
descheduling reasons (i.e. event wait and suspend) inside the same
process node. In previus code remote node will wake up process node
by sending event evein if process node is waiting in
vlib_process_ssupend().
This change also allowed new vlib_process_yield() API which deschedules
current process and it puts it into the end of queue.
Change-Id: I846e5a99b4ea1809eb80895f6ffe0ef0b2fd21ae
Type: improvement
Signed-off-by: Damjan Marion <damarion@cisco.com>
This change allow a node registration A to name a next node B that does not
exist yet at registration time. When node B is effectively created,
vlib_node_main_lazy_next_update need to be called so that the vlib graph
is updated accordingly. To enable this behavior, node A needs to bear
the new VLIB_NODE_FLAG_ALLOW_LAZY_NEXT_NODES.
Change-Id: I561d3a0de19a0b7bd1045760a2ba8e27d27caa9a
Type: improvement
Signed-off-by: Mohammed Hawari <mohammed@hawari.fr>
Free node frames in worker mains on refork. Otherwise these frames are
never returned to free pool and it causes massive memory leaks if
performed under traffic load
Type: fix
Signed-off-by: Dmitry Valter <d-valter@yandex-team.ru>
Change-Id: I15cbf024a3f4b4082445fd5e5aaa10bfcf77f363
- support of in-place growth of vectors (if there is available space next to
existing alloc)
- drops the need for alloc_aligned_at_offset from memory allocator,
which allows easier swap to different memory allocator and reduces
malloc overhead
- rework of pool and vec macros to inline functions to improve debuggability
- fix alignment - in many cases macros were not using native alignment
of the particular datatype. Explicitly setting alignment with XXX_aligned()
versions of the macro is not needed anymore in > 99% of cases
- fix ASAN usage
- avoid use of vector of voids, this was root cause of several bugs
found in vec_* and pool_* function where sizeof() was used on voids
instead of real vector data type
- introduce minimal alignment which is currently 8 bytes, vectors will
be always aligned at least to that value (underlay allocator actually always
provide 16-byte aligned allocs)
Type: improvement
Change-Id: I20f4b081bb13bbf7bc0ace85cc4e301787f12fdf
Signed-off-by: Damjan Marion <damarion@cisco.com>
Align runtime_data to 8 bytes to avoid alignment issues with any
pointers in runtime_data located structures.
Type: fix
Signed-off-by: Dmitry Valter <d-valter@yandex-team.ru>
Change-Id: I9cb1b73595e654a4b15c45f87b43fa6cfbcb6e51
Don't switch nodes from interrupt to polling state unless adaptive mode
flag set. For starters, flag set only on interface input nodes
with no polling rx queue and at least one in adaptive mode.
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ica1c75f605ead82b7cf74c45c6a774461008f054
It allows default variant selection from startup.conf
Type: improvement
Change-Id: Idff95e12dd0c105dab7c905089548b05a6e974e0
Signed-off-by: Damjan Marion <damarion@cisco.com>
This adds a new data model for counters.
Specifying the errors severity and unit.
A later patch will update vpp_get_stats to take advantage of this.
Only the map plugin is updates as an example.
New .api language:
A new "counters" keyword to define counter sets.
counters map {
none {
severity info;
type counter64;
units "packets";
description "valid MAP packets";
};
bad_protocol {
severity error;
type counter64;
units "packets";
description "bad protocol";
};
};
Each counter has 4 keywords. severity, which is one of error, info or warn.
A type, which is one of counter64 or gauge64.
units, which is a text field using units from YANG.
paths {
"/err/ip4-map" "map";
"/err/ip6-map" "map";
"/err/ip4-t-map" "map";
"/err/ip6-t-map" "map";
};
A new paths keyword that maps the counter-set to a path in the stats segment KV store.
Updated VPP CLI to include severity so user can see error counter severity.
DBGvpp# show errors
Count Node Reason Severity
13 ethernet-input no error error
Type: feature
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: Ib2177543f49d4c3aef4d7fa72476cff2068f7771
Signed-off-by: Ole Troan <ot@cisco.com>
Callbacks for monitoring and performance measurement:
- Add new callback list type, with context
- Add callbacks for API, CLI, and barrier sync
- Modify node dispatch callback to pass plugin-specific context
- Modify perfmon plugin to keep PMC samples local to the plugin
- Include process nodes in dispatch callback
- Pass dispatch function return value to callback
Type: refactor
Signed-off-by: Tom Seidenberg <tseidenb@cisco.com>
Change-Id: I28b06c58490611e08d76ff5b01b2347ba2109b22
- add ability to pass data together with interrupt
- avoid locking for local interrupts (same thread)
Type: improvement
Change-Id: I73a2ab2e716bb887a1f02c87788ae83e329f9b40
Signed-off-by: Damjan Marion <damarion@cisco.com>
Instead of allocating stack from the main heap, this patch mmaps stack
memory together with guard page.
This aproach reduces main heap usage, and stack memory is prefaulted
on demand, so bigger process stacks will have zero impact on memory
usage as long as stack memory is not needed for real.
In addition, it fixes issue with systems which have bigger default page
size (observed with 65536).
Type: improvement
Change-Id: I593365c603d4702e428967d80fd425fdee2c4a21
Signed-off-by: Damjan Marion <damarion@cisco.com>
Support for startup node multi-arch variant selection through startup.conf.
This is to facilitate unit, functional testing and benchmarking of non-default
multi-arch variant node code path. Also added parameters to make test, to
specific using multi-arch variants in unit testing.
Type: improvement
Signed-off-by: Ray Kinsella <mdr@ashroe.eu>
Change-Id: I94fd332bb629683b7a7dd770ee9f615a9a424060
Credits to ray.kinsella@intel.com who spotted the issue and identified
root cause.
Type: fix
Change-Id: I4afe74c47769484309f6aebca2de56ad32c8041f
Signed-off-by: Damjan Marion <damarion@cisco.com>
Encoding the vpp node index into the vlib_error_t as a 10-bit quantity
limits us to 1K graph nodes. Unfortunately, a few nodes need 6 bit
per-node error codes. Only a very few nodes have so many counters.
It turns out that there are about 2K total error counters in the system,
which is (approximately) the maximum error heap index.
The current (index,code) encoding limits the number of interfaces to
around 250, since each interface has two associated graph nodes and we
have about 500 "normal, interior" graph node
This patch adds an error-index to node-index map, so we can store
error heap indices directly in the vlib_buffer_t.
Type: refactor
Change-Id: I28101cad3d8750819e27b8785fc0cf71ff54f79a
Signed-off-by: Dave Barach <dave@barachs.net>
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>
It is causing compilation sloness with gcc-7 so removing it
before it was originally planned.
So far macros are left in the tree so we can know which nodes to
convert to new multiarch code.
Change-Id: Idb14622ca61fdce1eba59723b20d98715b7971e6
Signed-off-by: Damjan Marion <damarion@cisco.com>
As a FUD reduction measure, this patch implements 2-way parallel
counter collection. Synthetic stat component counter pairs run at the
same time. Running two counters (of any kind) at the same time
naturally reduces the aggregate time required by an approximate
factor-of-2, depending on whether an even or odd number of stats have
been requested.
I don't completely buy the argument that computing synthetic stats
such as instructions-per-clock will be inaccurate if component counter
values are collected sequentially. Given uniform traffic pattern, it
must make no difference.
As the collection interval increases, the difference between serial
and parallel component counter collection will approach zero, see also
the Central Limit theorem.
Change-Id: I36ebdcf125e8882cca8a1929ec58f17fba1ad8f1
Signed-off-by: Dave Barach <dave@barachs.net>
VPP graph dispatch trace record description:
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Major Version | Minor Version | NStrings | ProtoHint |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Buffer index (big endian) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ VPP graph node name ... ... | NULL octet |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Buffer Metadata ... ... | NULL octet |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Buffer Opaque ... ... | NULL octet |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Buffer Opaque 2 ... ... | NULL octet |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| VPP ASCII packet trace (if NStrings > 4) | NULL octet |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Packet data (up to 16K) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Graph dispatch records comprise a version stamp, an indication of how
many NULL-terminated strings will follow the record header, and a
protocol hint.
The buffer index allows downstream consumers of these data to easily
filter/track single packets as they traverse the forwarding
graph. FWIW, the 32-bit buffer index is stored in big endian format.
As of this writing, major version = 1, minor version = 0. Nstrings
will be either 4 or 5.
Here is the current set of protocol hints:
typedef enum
{
VLIB_NODE_PROTO_HINT_NONE = 0,
VLIB_NODE_PROTO_HINT_ETHERNET,
VLIB_NODE_PROTO_HINT_IP4,
VLIB_NODE_PROTO_HINT_IP6,
VLIB_NODE_PROTO_HINT_TCP,
VLIB_NODE_PROTO_HINT_UDP,
VLIB_NODE_N_PROTO_HINTS,
} vlib_node_proto_hint_t;
Example: VLIB_NODE_PROTO_HINT_IP6 means that the first octet of packet
data SHOULD be 0x60, and should begin an ipv6 packet header.
Change-Id: Idf310bad80cc0e4207394c80f18db5f77c378741
Signed-off-by: Dave Barach <dave@barachs.net>
Those flags have local significance and they can be used for
sending hints to the next node.
Change-Id: Ic2596ee81c64cd16f96344365370e8fcdc465354
Signed-off-by: Damjan Marion <damarion@cisco.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>
- Remove frame handoff support machinery. We haven't used it in a long
time.
- Configuration support for the local endpoints bihash table
- Drop lookup failure packets in tcp46_syn_sent
Change-Id: Icd51e6785f74661c741e76fac23d21c4cc998d17
Signed-off-by: Dave Barach <dave@barachs.net>
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>
