summaryrefslogtreecommitdiffstats
path: root/src/vlib/unix/input.c
AgeCommit message (Collapse)AuthorFilesLines
2021-01-21interface: rx queue infra rework, part oneDamjan Marion1-2/+2
Type: improvement Change-Id: I4008cadfd5141f921afbdc09a3ebcd1dcf88eb29 Signed-off-by: Damjan Marion <damarion@cisco.com>
2020-07-01vlib: wake up workers if interrupts are postedBenoît Ganne1-1/+2
Type: fix Change-Id: If8dbbcb46193fd057fe8d704058609a3a8787d6c Signed-off-by: Benoît Ganne <bganne@cisco.com>
2019-06-11vlib: avoid retrieving freed file in epollFlorin Coras1-6/+7
Type:fix Change-Id: Id7f4f6e2a2f844085f511a33aa1db3968f5d97bb Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-05-16init / exit function orderingDave Barach1-2/+7
The vlib init function subsystem now supports a mix of procedural and formally-specified ordering constraints. We should eliminate procedural knowledge wherever possible. The following schemes are *roughly* equivalent: static clib_error_t *init_runs_first (vlib_main_t *vm) { clib_error_t *error; ... do some stuff... if ((error = vlib_call_init_function (init_runs_next))) return error; ... } VLIB_INIT_FUNCTION (init_runs_first); and static clib_error_t *init_runs_first (vlib_main_t *vm) { ... do some stuff... } VLIB_INIT_FUNCTION (init_runs_first) = { .runs_before = VLIB_INITS("init_runs_next"), }; The first form will [most likely] call "init_runs_next" on the spot. The second form means that "init_runs_first" runs before "init_runs_next," possibly much earlier in the sequence. Please DO NOT construct sets of init functions where A before B actually means A *right before* B. It's not necessary - simply combine A and B - and it leads to hugely annoying debugging exercises when trying to switch from ad-hoc procedural ordering constraints to formal ordering constraints. Change-Id: I5e4353503bf43b4acb11a45fb33c79a5ade8426c Signed-off-by: Dave Barach <dave@barachs.net>
2019-04-24Clean up multi-thread barrier-sync hold-down timerDave Barach1-3/+27
Main thread: don't bother with the barrier sync hold-down timer if none of the worker threads are busy. Worker threads: avoid epoll_pwait (10ms timeout) when the control-plane has been active in the last half-second. Change-Id: I82008d09968c65e2a4af0ebb7887389992e60603 Signed-off-by: Dave Barach <dave@barachs.net>
2019-04-22vlib epoll: handle file removal on EPOLLINFlorin Coras1-0/+2
Change-Id: I7a3526a8fdf17afb8cc2225bdfbd57f661680992 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-18vlib epoll: protect against clib file pool expansionFlorin Coras1-1/+3
Change-Id: I320e7c2fdacb3056bc448c73fec08d9e2978ee5e Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-18Fix memory corruption faulting [VPP-1639]Artem Belov1-3/+3
File pool may be reallocated on epoll events processing. *f* pointer shows to already freed address and corrupting memory chunk on clib_file_t property change. Change-Id: I751bddce27325452862b939c1a3eec2ccd9b71bb Signed-off-by: Artem Belov <artem.belov@xored.com>
2019-01-28Less intrusive way to capture numa_node and cpu_id changesDamjan Marion1-2/+10
Change-Id: I3a33fb81f31ed473811e9e7a6197b81135913865 Signed-off-by: Damjan Marion <damarion@cisco.com>
2018-09-09vlib: crash in linux_epoll_input_inline for accessing free file index [VPP-1412]Steven1-2/+31
Under rare scenario, epoll may still post an event to VPP although the file descriptor is already deleted via epoll_ctl (EPOLL_CTL_DEL) and the file descriptor is close. VPP tries to access the free file index entry and crash. The fix is to throw away the events which the file descriptor is already deleted. Change-Id: Ieca3a1873aecb28630c3abc42c40341f27c2faa7 Signed-off-by: Steven <sluong@cisco.com> (cherry picked from commit aec7297ba012e1fe4bbf85cdaec8e810aa476cea)
2018-06-21configurable per-dispatch-cycle sleepDave Barach1-2/+20
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>
2018-05-04Harmonize vec/pool_get_aligned object sizes and alignment requestsDave Barach1-0/+1
Object sizes must evenly divide alignment requests, or vice versa. Otherwise, only the first object will be aligned as requested. Three choices: add CLIB_CACHE_LINE_ALIGN_MARK(align_me) at the end of structures, manually pad to an even divisor or multiple of the alignment request, or use plain vectors/pools. static assert for enforcement. Change-Id: I41aa6ff1a58267301d32aaf4b9cd24678ac1c147 Signed-off-by: Dave Barach <dbarach@cisco.com>
2018-02-06vlib: epoll on worker threadsDamjan Marion1-33/+102
This patch teaches worer threads to sleep and to be waken up by kernel if there is activity on file desctiptors assigned to that thread. It also adds counters to epoll file descriptors and new debug cli 'show unix file'. Change-Id: Iaf67869f4aa88ff5b0a08982e1c08474013107c4 Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-10-03Repair vlib API socket serverDave Barach1-1/+1
- Teach vpp_api_test to send/receive API messages over sockets - Add memfd-based shared memory - Add api messages to create memfd-based shared memory segments - vpp_api_test supports both socket and shared memory segment connections - vpp_api_test pivot from socket to shared memory API messaging - add socket client support to libvlibclient.so - dead client reaper sends ping messages, container-friendly - dead client reaper falls back to kill (<pid>, 0) live checking if e.g. a python app goes silent for tens of seconds - handle ping messages in python client support code - teach show api ring about pairwise shared-memory segments - fix ip probing of already resolved destinations (VPP-998) We'll need this work to implement proper host-stack client isolation Change-Id: Ic23b65f75c854d0393d9a2e9d6b122a9551be769 Signed-off-by: Dave Barach <dave@barachs.net> Signed-off-by: Dave Wallace <dwallacelf@gmail.com> Signed-off-by: Florin Coras <fcoras@cisco.com>
2017-09-09move unix_file_* code to vppinfraDamjan Marion1-6/+7
This will allow us to use this code in client libraries without vlib. Change-Id: I8557b752496841ba588aa36b6082cbe2cd1867fe Signed-off-by: Damjan Marion <damarion@cisco.com>
2017-06-28switch vlib process model to tw_timer_template timer implDave Barach1-36/+25
Change-Id: I36bb47faea55a6fea7af7ee58d87d8f6dd28f93d Signed-off-by: Dave Barach <dave@barachs.net>
2017-04-11Fix typo in minimal epoll polling timePierre Pfister1-1/+1
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 ec06222ae189fe8d84b63410130fff04bf446573)
2017-03-02Clean up binary api message handler registration issuesDave Barach1-7/+24
Removed a fair number of "BUG" message handlers, due to conflicts with actual message handlers in api_format.c. Vpp itself had no business receiving certain messages, up to the point where we started building in relevant code from vpp_api_test. Eliminated all but one duplicate registration complaint. That one needs attention from the vxlan team since the duplicated handlers have diverged. Change-Id: Iafce5429d2f906270643b4ea5f0130e20beb4d1d Signed-off-by: Dave Barach <dave@barachs.net>
2016-12-28Reorganize source tree to use single autotools instanceDamjan Marion1-0/+265
Change-Id: I7b51f88292e057c6443b12224486f2d0c9f8ae23 Signed-off-by: Damjan Marion <damarion@cisco.com>