Age | Commit message (Collapse) | Author | Files | Lines |
|
This allow configuration for each transport protocol when combined,
e.g. HTTPS=HTTP+TLS.
Extended config of session endpoint config is now chunk of memory,
which allow seamless integration with VCL, and internal
representation is hidden behind APIs, which is better for
future improvements. By default chunk is allocated to 512B if the
app doesn't do so before.
Type: improvement
Change-Id: I323f19ec255eba31c58c06b8b83af45aab7f5bb1
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I6a0493834c5b60bdb42865112c453157fdfeb87c
|
|
Type: refactor
Change-Id: I5235bf3e9aff58af6ba2c14e8c6529c4fc9ec86c
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Quic does not seed random value, so if the plugin is loaded
separately RAND_bytes will fail.
Type: fix
Change-Id: If600cbde1fef30afb6316fc1a355261b008c3191
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
|
|
Session state cannot be updated after async notification event is
generated for app. Instead, make sure quic sessions that accept new
streams are switched to listening state only on accept.
Type: fix
Fixes: 0242d30
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I9663ccadbea99d555ad49e871f7dff897239dc84
|
|
Fifos need to be synchronously allocated once a transport like tcp
accepts a session. Since events are now delivered asynchronously,
proxy apps must explicitly register a cb function that manages
fifo allocation prior to being notified of connect event.
Type: fix
Fixes: 0242d30
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I7df973b7014e53e0766ea2bdc61e9871160bc18b
|
|
Move from synchronous flushing of io and ctrl events from transports to
applications to an async model via a new session_input input node that
runs in interrupt mode. Events are coalesced per application worker.
On the one hand, this helps by minimizing message queue locking churn.
And on the other, it opens the possibility for further optimizations of
event message generation, obviates need for rx rescheduling rpcs and is
a first step towards a fully async data/io rx path.
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Id6bebcb65fc9feef8aa02ddf1af6d9ba6f6745ce
|
|
Avoid explicit manipulation of session state and generate closing event
if need be.
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I07cda1db08a2673b27b496ea1371b0dfd8e6f98a
|
|
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ic11069c912a5e59bb3ea0e0c6de6cfcc879c5f4e
|
|
- load openssl legacy providers during quic init
when building with openssl 3.0 or greater
- re-enable quic 'make test' testcases on
ubuntu-22.04
Type: fix
Change-Id: Icfd429b6bc1bddf9f9937baa44cc47cd535ac5f2
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
|
|
Some upon apps(e.g. Nginx-quic) package it's several protocol buffers into a struct msg
which is a combination of gso_buffer and gso_size.
but if HostStack regardless the gso_size to the buffer and split the buffer with default mss,
that cause peer client failed on parsing the package.
Type: improvement
Signed-off-by: Dou Chao <chao.dou@intel.com>
Change-Id: I805eb642be826038ba96d1b85dad8ec0c0f6c459
Signed-off-by: Dou Chao <chao.dou@intel.com>
|
|
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ia03c3fe0ca669b319dec8decd503254d0a95e58b
|
|
Type: improvement
Change-Id: I3fbbda0378b72843ecd39a7e8592dedc9757793a
Signed-off-by: Damjan Marion <dmarion@me.com>
|
|
Type: fix
Fixes: 5b4b4c0
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: If4bd8f30cd23d862109cab665251ad89804b1734
|
|
if when the rx_fifo grows, svm_fifo_enqueue() return -4,
stream_data->app_rx_data_len += rlen type conversion occurs,
Finally,stream->recvstate.data_off calculation is wrong.
Type:fix
Signed-off-by: fanxb <fxb_mail@163.com>
Change-Id: Iae11f0c453f32d836f4148d70e3b121545a53a90
|
|
Type:fix
Signed-off-by: fanxb <fxb_mail@163.com>
Change-Id: I244f6ddd0a34e82ababd375646d3bb194602b5d5
|
|
Makes it similar to connects.
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I38c328670054e1a9ba4dc4ea8fe7519a5a09e8be
|
|
The vnet_crypto_key_add should only be called from the main thread.
This patch works around this limitation by allocating one key per worker
and updating it on the fly everytime we need to do a crypto operation.
This solution is far from ideal, but quicly has a strong assumption that
it can use a key immediately after determining it, so making the key
creation asynchronous is not a possibility.
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
Change-Id: I19fc5814195156003c36a73bb616738ba9d828f7
Type: fix
|
|
Type: fix
Signed-off-by: jxm <jiangxiaoming@outlook.com>
Change-Id: I6fb2620e7076e1e38a2ab85a70febe614b079e67
|
|
Only works with release binaries.
Type: test
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Iede88713e1dbc4d1c6ae62651ff9150f88953f4d
|
|
Use ctx timer wheel not current thread since connects are done on main
but the ctx is allocated on first worker.
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ibed371bed12970d0d1ae4572f36df0bb1fb70b0f
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I901980f1f8ae8eb51c3972537ccac80f620868d0
|
|
- quicly conn struct leaked on connection delete
- Occasionally connection delete is called twice
before connection is closed which caused a crash.
Type: fix
Change-Id: Ifaaaeda55f71d58c97fa4d6652bda60a3efd4b69
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
|
|
UDP sessions are allocated on first worker because main does not poll.
Consequently, quic should also allocate connection on first, to avoid
multiple migration events.
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ice229a42c399e87150e0f6dd76e2e06ab271556c
|
|
By convention, connects segment manager will be first. Therefore it will
be the one with the first segment wherein lies the app's message queue.
Saves us the trouble of allocating it on first connect, if app started
by listening, and we no longer need to track if it's assignable to a
listener or if it can be removed.
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Iba9a8ffaab618eeb41ec2144dcfee62d006dc7a2
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I277f2e9ae2b763f5833778406d11218211163e43
|
|
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I2992c66d11fe6adc96b525b0fde533e5ff58d7e4
|
|
A transport's custom_tx function must report the number of segments that
it enqueued.
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ie088c4215e6f17df639159820b8adf46b96fbb46
|
|
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I92f0c1f7c0f4696fa12071440a643aa703d6306f
|
|
Pass tls/quic crypto context using extended config instead of bloating
conect/listen messages.
Type: refactor
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I0bc637ae310e6c31ef1e16847501dcb81453ee94
|
|
Type: improvement
Change-Id: I1152e58d7bfcb3c4347147f87a834d45ad51cdfe
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Type: fix
Change-Id: I744cedb9c1b57945af5e83057e4759964fd2e104
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
|
|
- update quic plugin with new quicly/picotls API
- remove packet allocator
- remove crypto batching
- update picotls plugin
- add cli for quicly congestion control configuration
Type: feature
Change-Id: If76ef31c43b430eea2f7674539b2112aee0f351e
Signed-off-by: Mathias Raoul <mathias.raoul@gmail.com>
|
|
Type: refactor
Change-Id: Ie67dc579e88132ddb1ee4a34cb69f96920101772
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Type: fix
Change-Id: Id03f50c46d28c850865cc76692424d063a0c2cfb
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
|
|
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Iae5dbb8aaaf82d8e95c2ee8bbbe6844c9dd49f80
|
|
Properly close the stream sending side instead of resetting the entire
stream.
Change-Id: I2daca7e507daa7bf9cd6a27fa8828a9ea7d175f9
Type: fix
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
|
|
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ieb8bb9c6deb92479fdd3e045778fe5ae4782d1ea
|
|
Type: improvement
Change-Id: I9dd850a1ce85b0adb5136233f176117e0ee38817
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I840d43e79b1f826380bd56485441510e45bdfc7f
|
|
Type: feature
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I6aaaec20a2b6d4c6ddfbe659d9402acc1be2f7e2
|
|
Type: improvement
This can be used as alternative to udpc
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ic3f7efe6728b25d4a8a0b61ddb36de66b4672c4f
|
|
- updates the quicly version to 0.1.0-vpp
- adds workaround for quicly_send()/assert_consistency() failure
Type: feature
Change-Id: I4c7e0ffc720ad9a685b89046a83646d59febd6cd
Signed-off-by: MathiasRaoul <mathias.raoul@gmail.com>
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
|
|
Type: feature
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: If4dee6dba1ea942daa921d566b35cdecdda680ee
|
|
Type: fix
Change-Id: I95d3f8431b468cefc8777526dd3b988a299f0687
Signed-off-by: MathiasRaoul <mathias.raoul@gmail.com>
|
|
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>
|
|
- Implement our own quic packet allocator to allocate more memory at the end of the
packet to store crypto offloading related data
- 1RTT packets offloading encryption/decryption using vnet crypto
- Add cli to change max packet per key
Type: feature
Change-Id: I7557fd457d7ba492329d5d8ed192509cbd727f9c
Signed-off-by: MathiasRaoul <mathias.raoul@gmail.com>
|
|
Type: feature
Signed-off-by: MathiasRaoul <mathias.raoul@gmail.com>
Change-Id: I5452f8bbd0ff9e2a57f7bd7d134a8824efa5f30a
|
|
This prevents unnecessary calls to quic_send_packets.
Type: fix
Change-Id: I7abe509aa8b7b9d5a01c9876046cf0f4507a79cf
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
|
|
Type: fix
Ensure listeners for app transport protocols are added to lookup tables
using their session endpoints instead of their transport connections,
which can override the network connection id in the transport connection.
Change-Id: I56fa3666bb1422c0799fc7143cd099751ff6e2e6
Signed-off-by: Florin Coras <fcoras@cisco.com>
|