summaryrefslogtreecommitdiffstats
path: root/src/svm/svm_fifo.c
AgeCommit message (Collapse)AuthorFilesLines
2020-12-26svm: change fifo pointers to offsetsFlorin Coras1-107/+118
Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I113a630d6db4741f54e508f7e0ed77be139df0e4
2020-12-24svm: split fifo into private and shared structsFlorin Coras1-114/+127
Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Id8e77e8b2623be719fd43a95e181eaa5b7df2b6e
2020-12-21svm: honor max requested bytes in fifo segmentsFlorin Coras1-1/+1
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I00a6be53c4519937ac900560c669587921990821
2020-12-20tls: use fifo segments instead of chunksFlorin Coras1-0/+36
Type: refactor Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Id67cf8a3e1c5c9b4160689fde5de9ce7ed8a2ee3
2020-11-24svm: support for multi-segment enqueuesFlorin Coras1-0/+81
Type: feature Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I06c7022a6afbb146b23cbd3a430497ec9e8be73d
2020-11-19svm: init chunk rb tree indices in fifoFlorin Coras1-0/+2
Type: improvement Let fifo segment mainly deal with fifo and chunk allocations not initialization. Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I1f1eb5a1423ba52cb950ae771641fd7eeff4e76c
2020-10-21svm: fix ASAN annotations for external chunksBenoît Ganne1-0/+3
Chunks can be allocated from another process. We need to manually mark them as accessible for ASAN. Type: fix Change-Id: Ifbeef3346e9cee2c1231f80cbcf7f9673b5b54be Signed-off-by: Benoît Ganne <bganne@cisco.com>
2020-10-12vcl svm: segments improvementsFlorin Coras1-8/+19
Type: improvement Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I717c64666972bb4e440cb3d1180a5cb26ee25577
2020-09-29vcl svm: provide apps access to fifo chunksFlorin Coras1-29/+20
Type: feature Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I2191b8594b1e87ecc00f237316457db249f73603
2020-09-08svm session: document unsupported fifo deq combinationsFlorin Coras1-0/+4
Type: fix - Document that ooo dequeues with ooo lookups cannot be done in combination with in order dequeues. - Added assert to capture this scenario and de-initialized rbtrees for cut-through tx fifo Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Ic40d020b3f0391fcf022ea3c906b86121744144f
2020-08-06misc: harmonize namesDave Barach1-7/+7
Type: fix Signed-off-by: Dave Barach <dave@barachs.net> Change-Id: Ibad744788e200ce012ad88ff59c2c34920742454
2020-02-26tls session: fix unlistenFlorin Coras1-1/+0
Type: fix Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: Ie3bc31fc3df662e087f7931de6c274eb3608a2d8
2020-02-26svm: remove chunk add functionFlorin Coras1-35/+25
Type: refactor Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I606eabbe0f3f387529e77d2483f5d63db7bde88a
2020-02-25svm: minimal initial fifoFlorin Coras1-4/+3
Type: refactor Signed-off-by: Florin Coras <fcoras@cisco.com> Change-Id: I4ee46a6c3c53c58199c275e20702f7fd11b60d9a
2020-02-25svm: refactor fifoFlorin Coras1-574/+470
Type: refactor Switch from a wrapped byte space to a "continuous" one wherein fifo chunks are appended to the fifo as more data is enqueued and chunks are removed as data is dequeued. The fifo is still subject to a maximum size, i.e., maximum number of bytes that can be enqueued, so the max number of chunks associated to the fifo is also constrained. When enqueueing data, which must fit within the available free space, if not enough "supporting" chunk memory is available, the fifo asks the fifo segment for enough chunk memory to ensure that the write can succeed. To avoid allocating large amounts of small chunks due to small writes, if possible, the size of the chunks requested is lower capped by min_alloc. When dequeuing data, all the chunks that have been completely drained, i.e., head moved beyond the chunks’ end bytes, are unlinked from the fifo and returned to the fifo segment. The one exception to this is the last chunk which is never unlinked. Change-Id: I98c1dbd9135fb79650365c7e40c29238b96cd4ee Signed-off-by: Florin Coras <fcoras@cisco.com>
2020-02-25svm: refactor fifo chunk trackingFlorin Coras1-57/+281
Avoid tracking with rbtrees all of the chunks associated to a fifo. Instead, only track chunks when doing out-of-order operations (peek or ooo enqueue). Type: refactor Change-Id: I9f8bd266211746637d98e6a12ffc4b2d6346950a Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-10-16svm: immediate fifo growth even when wrappedRyujiro Shibuya1-8/+108
Type: fix - when the fifo is wrapped, and if applicable, insert a new chunk after the tail-chunk and rebuild the rb_tree. - make sure that this new algorithm can be applied only when the fifo is used by a single thread (master-thread of the fifo). Signed-off-by: Ryujiro Shibuya <ryujiro.shibuya@owmobility.com> Change-Id: I3fc187bc496ea537ca24381e4abc08d2906c9e03
2019-08-29svm: fix fifo hdr freelist allocationFlorin Coras1-1/+1
Type: fix Change-Id: Ia362ad821db1fd506e973e1844cc3ec74703cc17 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-08-05svm: fix fifo growth on dequeue with dropFlorin Coras1-0/+6
Type:fix Change-Id: I39c339abf1b51105ef1bcf3d6f0f4f6ded54f32d Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-07-03svm: fix multi-chunk fifo alloc and add more testsFlorin Coras1-6/+97
Type: fix - make sure that chunks and the rbtree are initialized if fifo segment allocates multiple chunks for the fifo. - ensure head/tail chunks are updated on all enqueue/dequeue events, including when dropping data. - more unit tests Also fixes dequeue drop updates of head chunk. Change-Id: I77f3550bc4e8b4e077f80ea87fe82b83ed013aeb Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-06-25svm: make sure chunk search does not return tnilFlorin Coras1-0/+2
Type:fix Change-Id: I8405bf8d93b4468c54f4f3c5dcd21ef91a6b1048 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-06-24session: full lock on session_send_evt_to_threadNathan Skrzypczak1-2/+2
Type: fix This was causing issues in QUIC when an app client & the protocol app compete for the worker msg_queue. Might not be ideal performance- wise. Change-Id: I629892253d5b5d968f31ad1d56f18463e143d6b4 Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
2019-05-07svm: decouple fifo and default chunk allocationFlorin Coras1-8/+20
Default chunk is no longer embedded into the fifo and on free is returned to its respective chunk list. Change-Id: Ifc5d214eaa6eca44356eb79dd75650fb8569113f Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-05-03session/svm: apis for fifo shrinkingFlorin Coras1-1/+4
Change-Id: Ie519683bb90aae6fb95f2a09e251cded1890ed41 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-05-03svm: basic support for fifo shrinkingFlorin Coras1-54/+208
As opposed to growing, this is not a bulk operation, instead dependent on how the producer/consumer advance head and tail, the fifo will shrink in one or multiple steps. Only once the fifo's nitems and size are reduced to their appropriate values, equal or larger to what was requested, can the fifo chunks be collected by the owner. Chunk collection must be done with the segment heap pushed. Change-Id: Iae407ccf48d85320aa3c1e0304df56c5972c88c1 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-05-03svm: store normalized head/tail for fifoFlorin Coras1-23/+21
If head/tail are stored as "absolute" values that are normalized to [0, fifo_size] interval, when fifo is shrunk/grown the consumer and producer have to independently update to the new fifo size and fix head and tail, respectively. If the head and tail are stored as normalized values, under the right conditions, they don't need to be fixed when fifo size changes. This reverts one of the changes in gerrit 18223. Change-Id: I55a908828afe90925cf7c20186a940b25e5805f9 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-30svm: more fifo refactor/cleanupFlorin Coras1-383/+385
Change-Id: Ie76c69641c8598164d0d00fd498018037258fd86 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-27svm: fix fifo tail/head/ooo logic for u32 wrapFlorin Coras1-2/+2
These were introduced with the switch to unbound tail/head size, so they only affect master. Added unit tests to avoid future surprises. Change-Id: I83b6c9efbe31d8092ba59b8e2ed46f4da97f35db Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-26svm: fifo segment support for chunk allocationFlorin Coras1-2/+38
Change-Id: Ie96706b4d8bcb32d2d5f065bc765f95f4e9369e7 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-25session: cleanup segment manager and fifo segmentFlorin Coras1-2/+2
Change-Id: I984f347fb465c0c405cef668d8690457e81788e2 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-24svm: fifo ooo reads/writes with multiple chunksFlorin Coras1-48/+153
Change-Id: If23a04623a7138c9f6c98ee9ecfa587396618a60 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-20svm: reorganize fifo march codeFlorin Coras1-169/+176
- make only the chunk copying (memcpy) code march aware - cleanup dependencies Change-Id: I369378264cacfcdaf0823353b957876554eaa17c Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-19svm: support addition of chunks to fifosFlorin Coras1-4/+6
Change-Id: Ia56cad89b85b7a99ab4bfb85318a45a71381fb53 Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-18svm: allow indirect fifo data chunksFlorin Coras1-134/+124
Fifos can use multiple memory chunks for simple read/write operations. Adding/removing chunks after assignment not yet supported. Change-Id: I2aceab6aea78059d74e0d3a9993c40d5196d077b Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-04-16svm_fifo rework to avoid contention on cursizeSirshak Das1-206/+229
Problems Addressed: - Contention of cursize by producer and consumer. - Reduce the no of modulo operations. Changes: - Synchronization between producer and consumer changed from cursize to head and tail indexes Implications: reduces the usable size of fifo by 1. - Using weaker memory ordering C++11 atomics to access head and tail based on producer and consumer role. - Head and tail indexes are unsigned 32 bit integers. Additions and subtraction on them are implicit 32 bit Modulo operation. - Adding weaker memory ordering variants of max_enq, max_deq, is_empty and is_full Using them appropriately in all places. Perfomance improvement (iperf3 via Hoststack): iperf3 Server: Marvell ThunderX2(AArch64) - iperf3 Client: Skylake(x86) ~6%(256 rxd/txd) - ~11%(2048 rxd/txd) Change-Id: I1d484e000e437430fdd5a819657d1c6b62443018 Signed-off-by: Sirshak Das <sirshak.das@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
2019-04-08host stack: update stale copyrightFlorin Coras1-1/+1
Change-Id: I33cd6e44d126c73c1f4c16b2041ea607b4d7f39f Signed-off-by: Florin Coras <fcoras@cisco.com>
2019-01-17vcl/session: replicate events for shared sessionsFlorin Coras1-0/+23
Change-Id: I7fb5402d4a530b5f2ffd9bb5787632099f4b4189 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-11-30session: segment handle in accept/connect notificationsFlorin Coras1-0/+1
Change-Id: I03884b6cde9d4c38ae13d1994fd8d37d44016ef0 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-11-28Use acquire/release ordering when accessing svm_fifo shared variable cursizeSirshak Das1-5/+5
Improves TCP iperf3 performance by ~3% on AArch64. Change-Id: I1e51bd8403ba45ec6af4c2f96b95e884c1ae0d67 Signed-off-by: Sirshak Das <sirshak.das@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Reviewed-by: Ola Liljedahl <ola.liljedahl@arm.com>
2018-11-14Remove c-11 memcpy checks from perf-critical codeDave Barach1-16/+17
Change-Id: Id4f37f5d4a03160572954a416efa1ef9b3d79ad1 Signed-off-by: Dave Barach <dave@barachs.net>
2018-10-25session/tcp: improve cliFlorin Coras1-10/+15
Change-Id: I91c9d040fc9b9b63f7109eeaac334c47fb1226cf Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-10-23c11 safe string handling supportDave Barach1-17/+18
Change-Id: Ied34720ca5a6e6e717eea4e86003e854031b6eab Signed-off-by: Dave Barach <dave@barachs.net>
2018-10-19vppinfra: add atomic macros for __sync builtinsSirshak Das1-5/+5
This is first part of addition of atomic macros with only macros for __sync builtins. - Based on earlier patch by Damjan (https://gerrit.fd.io/r/#/c/10729/) Additionally - clib_atomic_release macro added and used in the absence of any memory barrier. - clib_atomic_bool_cmp_and_swap added Change-Id: Ie4e48c1e184a652018d1d0d87c4be80ddd180a3b Original-patch-by: Damjan Marion <damarion@cisco.com> Signed-off-by: Sirshak Das <sirshak.das@arm.com> Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com> Reviewed-by: Ola Liljedahl <ola.liljedahl@arm.com> Reviewed-by: Steve Capper <steve.capper@arm.com>
2018-10-05session: fix local session tagging and cleanupFlorin Coras1-0/+1
Change-Id: I0d42a0c71fea7dd669fb1fe5ded7e6e944245c7d Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-09-24svm: march svm_fifo take 2Florin Coras1-71/+17
Change-Id: Ifa4fceef7edbe43d444790a624957db0817064de Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-09-18svm: better march support for svm fifoFlorin Coras1-96/+84
Change-Id: I6a4335654882a2ca66d3d465e35e350868242b8d Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-09-12vcl: add apis that expos fifo as bufferFlorin Coras1-0/+47
Change-Id: I4bd9c9f73499711e04b38d53daa5c917a4285bf5 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-07-27vcl: use events for epoll/select/read/writeFlorin Coras1-1/+1
Have vcl poll and wait on the event message queues as opposed to constantly polling the session fifos. This also adds event signaling to cut through sessions. On the downside, because we can't wait on multiple condvars, i.e., when we have multiple message queues because of cut-through registrations, we do timed waits. Change-Id: I29ade95dba449659fe46008bb1af502276a7c5fd Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-11tcp: cleanup connection/session fixesFlorin Coras1-0/+7
- Cleanup session state after last ack and avoid using a cleanup timer. - Change session cleanup to free the session as opposed to waiting for delete notify. - When in close-wait, postpone sending the fin on close until all outstanding data has been sent. - Don't flush rx fifo unless in closed state Change-Id: Ic2a4f0d5568b65c83f4b55b6c469a7b24b947f39 Signed-off-by: Florin Coras <fcoras@cisco.com>
2018-06-01session: improve disconnect handlingFlorin Coras1-1/+2
If the caller is the session owning thread or the main thread with a worker barrier sync (cli/api) add an event to the pending disconnects vector in the session node and entirely avoid using the event queue. Useful for bursts of disconnects (like an app detach). If disconnects come from a processes, be willing to retry enqueueing the disconnect to the event queue multiple times. Change-Id: Ieece1f1091b713f94c41c703b6e805bc8498816a Signed-off-by: Florin Coras <fcoras@cisco.com>