- 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>
New startup config command:
unix {
runtime-dir /run/vpp
}
Also, adds recursive mkdir funtion for use in deifferent places
like cli-config socket path and dpdk hugepage directory path.
Change-Id: I1446ceab9c220c25804e73a743a3ebb383450124
Signed-off-by: Damjan Marion <damarion@cisco.com>
- Make tcp output buffer allocation macro an inline function
- Use per ip version per thread tx frames for retransmits and timer
events
- Fix / parameterize tcp data structure preallocation
- Add a couple of gdb-callable show commands
- Fix local endpoint cleanup
Change-Id: I67b47b7570aa14cb4634b6fd93c57cd2eacbfa29
Signed-off-by: Florin Coras <fcoras@cisco.com>
Signed-off-by: Dave Barach <dave@barachs.net>
- api-segment prefix not used when unlinking shm files
- unlink root region on exit if no clients referenced
- stale reference to freed segment name
- don't add fake client to /db unless CLIB_DEBUG > 2
- turn off the gmond plugin
- clean up unused vars in vpp/api
Change-Id: I66451fcfd6ee64a12466c2d6c209050e3cdb74b7
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Signed-off-by: Dave Barach <dave@barachs.net>
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>
In multi-threaded model (e.g. 1 main and 1 worker threads),
after an ethernet interface is deleted (e.g. vhost-user interface),
'show runtime' command produces garbled output and sometimes
leads to vpp crash.
The reason is because vlib_node_rename() frees and reallocates node's
'n->name' vector, however the change is not propagated into copies
of the node on worker threads.
Change-Id: Ibf22422913b7f2df22f70f3b2fe8dafd34c1dd06
Signed-off-by: Igor Mikhailov (imichail) <imichail@cisco.com>
(cherry picked from commit 02989064e4)
- vnet_buffer_pool_create should probably be named
dpdk_buffer_pool_create since that is what it does.
- Its prototype should also be in a DPDK plugin header, not in
vlib/buffer_funcs.h, since the implementation is in the plugin
and nobody else should be calling it.
Change-Id: I7ba259afa4b888bc94f3ad257305e286b41e7370
Signed-off-by: Chris Luke <chrisy@flirble.org>
Currently, buffer index is calculated as a offset to the physmem
region shifted by log2_cacheline size.
When DPDK is used we "hack" physmem data with information taken from
dpdk mempool. This makes physmem code not usable with DPDK.
This change makes buffer memory start and size independent of physmem
basically allowing physmem to be used when DPDK plugin is loaded.
Change-Id: Ieb399d398f147583b9baab467152a352d58c9c31
Signed-off-by: Damjan Marion <damarion@cisco.com>
- PCI devices not properly discovered
- vlib_pci_bus_master_enable () not working
Change-Id: I7433ab1b19b890b8900635b43037b9a2017a1921
Signed-off-by: Damjan Marion <damarion@cisco.com>
Change https://gerrit.fd.io/r/#/c/7230/ added a Unix domain
CLI socket in the default startup.conf; however unless you
had previously run VPP with the DPDK plugin enabled the
directory that it is created in. /run/vpp, would not exist
and startup would fail. This directory is typically hosted
in a tmpfs ramdisk and is thus ephemeral.
This patch adds a function that attempts to mkdir VPP_RUN_DIR
and uses it in both the DPDK plugin and the CLI code if the
CLI socket is to be created in that directory.
Change-Id: Ibbf925819099dce2b5eb0fa238b9edca1036d6fd
Signed-off-by: Chris Luke <chrisy@flirble.org>
Spotted in the output of CLI command "show buffers", the name field
sometimes had trailing garbage, the hall sign of a string not being
terminated. In this case it was being inconsistently used as a cstring
or a vec.
- CLI printf needs %v to print the vec srring
- vlib_buffer_create_free_list_helper tried to use
clib_mem_is_heap_object() to detect a vec object, wheras it should
use clib_mem_is_vec()
Change-Id: Ib8b242a0c5a18924b8af7e8e1432784eebcf572c
Signed-off-by: Chris Luke <chrisy@flirble.org>
When stacking DPOs the VLIB graph is also updated to add the edge between the nodes, if this edge does not yet exist. This addition should be done with the workers stopped.
Change-Id: I327e4d7d26f0b23eb280f17e4619ff2093ff7940
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit c02bd03ddf)
Also allow group write as default for CLI socket connections.
Change-Id: I6af1f277f70581358cd9241bf0f5cb0752fe250f
Signed-off-by: Damjan Marion <damarion@cisco.com>
- refactor existing congestion control code (RFC 6582/5681). Handling of ack
feedback now consists of: ack parsing, cc event detection, event handling,
congestion control update
- extend sack scoreboard to support sack based retransmissions
- basic implementation of Eifel detection algorithm (RFC 3522) for
detecting spurious retransmissions
- actually initialize the per-thread frame freelist hash tables
- increase worker stack size to 2mb
- fix session queue node out-of-buffer handling
- ensure that the local buffer cache vec_len matches reality
- avoid 2x spurious event requeues when short of buffers
- count out-of-buffer events
- make the builtin server thread-safe
- fix bihash template threading issue: need to paint -1 across uninitialized
working_copy_length vector elements (via rebase from master)
Change-Id: I646cb9f1add9a67d08f4a87badbcb117980ebfc4
Signed-off-by: Florin Coras <fcoras@cisco.com>
Signed-off-by: Dave Barach <dbarach@cisco.com>
The main interior graph-node dispatch loop had a longstanding dangling
vector element reference:
for (i = 0; i < _vec_len (nm->pending_frames); i++)
cpu_time_now = dispatch_pending_node (vm, nm->pending_frames + i,
cpu_time_now);
Passing a pointer to a vector element (nm->pending_frames + i) has
considerable comedic potential if there's any chance that the vector
could expand.
dispatch_pending_node() calls dispatch_node(), and indirectly any
interior graph node dispatch function. If that node happens to expand
nm->pending_frames by filling in a new frame, nm->pending_frames can
expand.
After calling the node dispatch function, dispatch_node() does the
following:
nf = vec_elt_at_index (nm->next_frames, p->next_frame_index);
If nm->pending_frames expands during dispatch function execution, p is
a dangling reference to freed memory.
By luck, the TCP stack managed to allocate a fresh frame which
included "old-p," which caused p->next_frame_index to be filled with
the new-frame poison pattern 0xfefefefe.
This has been broken from day 1, summer 2007, first use of the
third-generation vector processing library.
Change-Id: Ideb6363bb060c4e8bf9b901882c318bd83853121
Signed-off-by: Dave Barach <dave@barachs.net>
Hitting tab:
- in the middle of a uniquely defined subcommand will expand the subcommand
- in the middle of a non-uniquely defined (or empty) subcommand will display
all possible subcommands, and possibly expand to the lowest common prefix
Change-Id: Ib858eefdb0353cd2c3aad472799d15cd537455a0
Signed-off-by: Yoann Desmouceaux <ydesmouc@cisco.com>
and add adaptive mode support to receive queue
- Migrate vhost to use device infra which does the interface/queue to worker
thread assignment.
- Retire vhost thread CLI and corresponding code which assigns interface/queue
to worker thread. set interface placement should be used instead to customize
the interface/queue to worker thread assignment.
- Retire vhost interrupt/polling option when creating vhost-user interface.
Instead, set interface rx-mode should be used.
- Add code in vnet_device_input_unassign_thread to change the node state
to interrupt if the last polling interface has left the worker thread for the
device of the corresponding interface/queue.
- Add adaptive mode support. The node state is set to interrupt initially.
When the scheduler detects a burst of traffic, it switches the input node to
polling. Then we inform the device that we don't need interrupt notification.
When the traffic subsides, the scheduler switches the input node back to
interrupt. Then we immediately tell the driver that we want interrupt
notification again.
- Remove some duplicate code in vlib/main.c
Change-Id: Id19bb1b9e50e6521c6464f470f5825c26924d3a8
Signed-off-by: Steven <sluong@cisco.com>
Off by default. Enable via cmdline "... vlib { elog-post-mortem-dump }
..."
Change-Id: I2056b9de9b37475f2bfeeb5404da838f1b42645a
Signed-off-by: Dave Barach <dave@barachs.net>
This happens mainly if plugin contains unresolved symbol.
Such situation typically leads on crash a bit later so it is better
to exit immediatelly and display meaningful error message.
Change-Id: I4abd9a9089a4863400bf609e8d3fd7cebab92913
Signed-off-by: Damjan Marion <damarion@cisco.com>
epoll was supposed to not sleep when timeout
is less than 1ms, but a typo made it not sleep
any time the requested timeout is lower than
1000 seconds (in practice, never...).
This patch replaces "1e3" with "1e-3", which
represents 1ms.
Change-Id: I731851b27a6bf6ab8e41586e017e94b962b09bf3
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
(cherry picked from commit ec06222ae1)
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>
Fix a decade-old ridiculous qsort function bug. Managed to subtract
floating-point numbers as if they were integers, leading to
manufactured time-paradoxes. That completely confuses g2, leading to
the summary disappearance of entire tracks' worth of data at high zoom
levels.
Add a manual alignment tweak parameter to elog_merge, users can
dial-out time paradoxes caused by NTP-grade clock synchronization.
The event-logger has a precision of O(100ns), whereas NTP
synchronization is O(1ms).
Change-Id: I69dedabaa314f69f9df74ec9ee66e21e6c87f703
Signed-off-by: Dave Barach <dave@barachs.net>
Worker main loop is now shared code with main thread
main loop so no need to export functions anymore.
Change-Id: I99ee2eee981c1b88ca31d20eabeb6c21d030a34d
Signed-off-by: Damjan Marion <damarion@cisco.com>
vhost currently supports only polling mode. This patch is to add
interrupt mode. When the interface is configured for interrupt
mode, our input node does not get called unless there is a packet
in the vring.
If a particular CPU has one interface configured for polling mode
and another in interrupt, the input node is set to polling for
that CPU.
This diffs also includes two crashes in vlib's dispatch_node. One is
included in https://gerrit.fd.io/r/#/c/5516. The other crash is in
the ASSERT. The ASSERT can become true when the caller of
dispatch_node is in a loop. The first call converted the node
to polling. The second call thereafter will hit the ASSERT.
Change-Id: If17b6d48b20d7d8605c6a161459828637173cd32
Signed-off-by: Steven <sluong@cisco.com>
dispatch_node may be invoked from vlib main or worker threads. The call to
ELOG_DATA in dispatch_node passes the parameter &vm->elog_main. It works fine
when dispatch_node is invoked from the main thread. It does bad thing when it
is invoked from the worker thread.
While we are at it, make two additional enhancements to the same area.
1. Use ELOG_TRACK_DATA instead of ELOG_DATA to enhance g2 viewer presentation.
2. Since ELOG_DATA is in the data path, it could get very chatty. Make the call
to ELOG_TRACK_DATA conditional compile.
Change-Id: I80ca0eea10bc1e5d0d5549f9844dd9a34dbb65a2
Signed-off-by: Steven <sluong@cisco.com>
