aboutgitcodebugslistschat
Commit message (Collapse)AuthorAgeFilesLines
* passt: Align pkt_buf to PAGE_SIZE (start and size), try to fit in huge pagesStefano Brivio2021-09-274-3/+10
| | | | | | | | | If transparent huge pages are available, madvise() will do the trick. While at it, decrease EPOLL_EVENTS for the main loop from 10 to 8, for slightly better socket fairness. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* README: pasta mode, CI, performance, updated links, etc.Stefano Brivio2021-09-271-66/+185
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* Makefile: Visually separate CFLAGS from input files in resulting cc commandsStefano Brivio2021-09-271-2/+4
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* udp: Switch to new socket message after 32KiB instead of 64KiBStefano Brivio2021-09-271-2/+2
| | | | | | | For some reason, this measurably improves performance with qemu and virtio-net. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* udp: Decrease UDP_TAP_FRAMES to 16Stefano Brivio2021-09-271-1/+1
| | | | | | | Similarly to the decrease in TCP_TAP_FRAMES, this improves fairness, with a very small impact on performance. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Decrease TCP_TAP_FRAMES to 8Stefano Brivio2021-09-271-1/+1
| | | | | | This significantly improves fairness in serving concurrent connections. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Update comment about spliced connection statesStefano Brivio2021-09-271-1/+4
| | | | | | ...we now have SPLICE_FIN_{FROM,TO,BOTH} too. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Increase TAP_BUF_BYTES to 256 maximum-sized descriptorsStefano Brivio2021-09-191-1/+1
| | | | | | ...this actually improves performance. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* README: Source jsStefano Brivio2021-09-181-0/+8
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Don't reset connection from ESTABLISHED state on EPOLLHUPStefano Brivio2021-09-161-6/+1
| | | | | | | | | | | That might just mean we shut down the socket -- but we still have to go through the other states to ensure a orderly shutdown guest-side. While at it, drop the EPOLLHUP check for unhandled states: we should never hit that, but if we do, resetting the connection at that point is probably the wrong thing to do. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Mask EPOLLIN and EPOLLRDHUP after sending FINStefano Brivio2021-09-161-2/+14
| | | | | | | Now that we dropped EPOLLET, we'll keep getting EPOLLRDHUP, and possibly EPOLLIN, even if there's nothing to read anymore. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Break splice() loop once we've written everything that was readStefano Brivio2021-09-161-2/+7
| | | | | | That's a guarantee that we don't need to retry writing. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Don't set SPLICE_FIN_BOTH state on EPOLLHUPStefano Brivio2021-09-161-3/+1
| | | | | | | EPOLLHUP just means we shut down one side of the connection on *one* socket: remember, we have two sockets here. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Don't reset 'never_read' flag on write retriesStefano Brivio2021-09-161-2/+3
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Don't set TCP_CORK on spliced socketsStefano Brivio2021-09-161-6/+1
| | | | | | | | ...throughput isn't everything: this leads (of course) to horrible latency with small, sparse messages. As a consequence, there's no need to set TCP_NODELAY either. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Fix setting window from maximum ACK sequence in batchStefano Brivio2021-09-161-6/+2
| | | | | | | | | If we're at the first message in a batch, it's safe to get the window value from it, and there's no need to subtract anything for a comparison on that's not even done -- we'll override it later in any case if we find messages with a higher ACK sequence number. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Set pipe descriptor numbers to -1 after closingStefano Brivio2021-09-161-0/+4
| | | | | | ...so that we don't try to close them again, even if harmless. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta, tcp: Drop EPOLLET for spliced, established connectionsStefano Brivio2021-09-161-5/+5
| | | | | | | ...tcp_handler_splice() doesn't guarantee we read all the available data, the sending buffer might be full. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Read SO_SNDBUF unconditionallyStefano Brivio2021-09-161-17/+11
| | | | | | | | Checking it only when the cached value is smaller than the current window of the receiver is not enough: it might shrink further while the receiver window is growing. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta: Clean up FIN connection flags once a connection is deletedStefano Brivio2021-09-151-0/+1
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta: Set spliced connection flag in epoll reference on compactionStefano Brivio2021-09-151-2/+2
| | | | | | ...otherwise, we'll mix indices with non-spliced connections. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta: Clean up namespace processes on exit, reap zombies from clone()Stefano Brivio2021-09-151-9/+83
| | | | | | | | | | | | If pasta created the namespace, it's probably expected that processes started in the same namespace are terminated once pasta exits. Scan procfs namespace links for corresponding processes, send SIGQUIT and SIGKILL (after one second) if found. While at it, make the signal handler reap otherwise-zombies resulting from clone(). Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* checksum: Add checksum.hStefano Brivio2021-09-141-0/+5
| | | | | | I forgot to commit this. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Request retransmission with updated sequence also on partial write to ↵Stefano Brivio2021-09-141-11/+8
| | | | | | | | | | | socket If we couldn't write the whole batch of received packets to the socket, and we have missing segments, we still need to request their retransmission right away, otherwise it will take ages for the guest to figure out we're missing them. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: In ESTABLISHED state, acknowledge segments as they're sent to the socketStefano Brivio2021-09-141-6/+15
| | | | | | | | | ...instead of waiting for the remote peer to do that -- it's especially important in case we request retransmissions from the guest, but it also helps speeding up slow start. This should probably be a configurable behaviour in the future. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Properly time out ACK wait from tapStefano Brivio2021-09-141-3/+3
| | | | | | | | | Seen with iperf3: a control connection is established, no data flows for a while, all segments are acknowledged. The socket starts closing it, and we immediately time out because the last ACK from tap was one minute before that. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Don't mistake a FIN segment with no data for a Fast Retransmit requestStefano Brivio2021-09-141-1/+2
| | | | | | | It carries no data and usually duplicates the previous ACK sequence, but it's just a FIN. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Check errno on sendmmsg() failure, not just the return valueStefano Brivio2021-09-141-1/+1
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Make sure sending window is initialised before sending to tapStefano Brivio2021-09-141-1/+1
| | | | | | | Seen with iperf3: the first packet from socket (data connection) is 65520 bytes and doesn't fit in the window. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* qrap: Set x-txburst as temporary workaround for virtio-net TX stallStefano Brivio2021-09-091-1/+1
| | | | | | | | Flooding a virtio-net interface connected to a socket back-end results in a TX stall I'm still debugging. The stall goes away by setting a higher value for x-txburst (256 by default). Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* udp: Reset iov_base after sending partial message on sendmmsg() failureStefano Brivio2021-09-091-0/+2
| | | | | | | We set the length while processing messges, but the starting address is pre-initialised. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* udp: Fix comparison of seen IPv4 address for local connectionsStefano Brivio2021-09-091-1/+2
| | | | | | c->addr4_seen is stored in network order. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Fixes for closing states, spliced connections, out-of-order packets, etc.Stefano Brivio2021-09-091-207/+400
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fixes a number of issues found with some heavier testing with uperf and neper: - in most closing states, we can still accept data, check for EPOLLIN when appropriate - introduce a new state, ESTABLISHED_SOCK_FIN_SENT, to track the fact we already sent a FIN segment to the tap device, for proper sequence number bookkeeping - for pasta mode only: spliced connections also need tracking of (inferred) FIN segments and clean half-pipe shutdowns - streamline resetting epoll_wait bitmaps with a new function, tcp_tap_epoll_mask(), instead of repeating the logic all over the place - set EPOLLET for tap connections too, whenever we are waiting for EPOLLRDHUP or an event from the tap to proceed with data transfer, to avoid useless loops with EPOLLIN set - impose an additional limit on the sending window advertised to the guest, given by SO_SNDBUF: it makes no sense to completely fill the sending buffer and send a zero window: stop a bit before we hit that - handle *all* interrupted system calls as needed - simplify the logic for reordering of out-of-order segments received from tap: it's not a corner case, and the previous logic allowed for deadloops - fix comparison of seen IPv4 address when we get a new connection from a socket directed to the configured guest address Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tap: Fix calculation of number of tap scatter-gather IO messagesStefano Brivio2021-09-091-2/+3
| | | | | | Messages are typically smaller than ETH_MAX_MTU. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta: Set ping_group_range upon namespace creationStefano Brivio2021-09-091-0/+4
| | | | | | | ...this allows processes running as the only group available in the namespace to create ICMP Echo sockets. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Add epoll event indication and passt/pasta mode in socket debug messageStefano Brivio2021-09-091-1/+3
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* conf: Fix help message about default behaviour for UDP port forwardingStefano Brivio2021-09-091-6/+4
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* conf, dhcp, ndp: Fix message about default MTU, make NDP consistentStefano Brivio2021-09-093-4/+16
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* udp: Fix retry mechanism on partial sendmmsg()Stefano Brivio2021-09-091-3/+3
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* qrap: Drop debugging left-overs, enable timeout for connect() tooStefano Brivio2021-09-091-5/+3
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* conf: Introduce PASST_LEGACY_NO_OPTIONS ifdef for legacyStefano Brivio2021-09-091-0/+12
| | | | | | | | Before introducing options, the default behaviour in passt mode was to forward all ports, to run in foreground and to log to stderr. Make it a bit more convenient to restore that at build time. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp, udp: Restore usage of gateway for guest to connect to local hostStefano Brivio2021-09-012-6/+11
| | | | | | | This went lost in a recent rework: if the guest wants to connect directly to the host, it can use the address of the default gateway. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* Makefile: Make sure destination directories exist on installStefano Brivio2021-09-011-0/+1
| | | | | | Mostly theoretical, but convenient for testing. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* udp: Handle partial failure in sendmmsg() to UNIX domain socketStefano Brivio2021-09-011-20/+60
| | | | | | | | | | | | | | | Similarly to the handling introduced by commit "tcp: Proper error handling for sendmmsg() to UNIX domain socket" for TCP, we need to deal with partial sendmmsg() failures for UDP as well. Here, we can lose messages, but we need to make sure that the last message is delivered completely, otherwise qemu will fail to reassemble further packets. For UDP, this is somewhat complicated by the fact that one message might include multiple datagrams, and we need to respect message boundaries: go through headers, and calculate what we need to re-send, if anything. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* doc/demo: Also forward all UDP ports from namespaceStefano Brivio2021-09-011-1/+1
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* doc/demo: Explicitly run in foreground, drop pipe to catStefano Brivio2021-09-011-1/+1
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* dhcp: Send option 121 if the default gateway is not on the assigned subnetStefano Brivio2021-09-011-0/+11
| | | | | | | | This enables CirrOS with udhcpc to set up a route to a gateway that's not on the assigned subnet, together with: https://bugs.launchpad.net/cirros/+bug/1190372 Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* conf: Fix check for IPv6 DNS address being already setStefano Brivio2021-09-011-1/+1
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* arp: Don't resolve own, configured IPv4 addressStefano Brivio2021-09-011-0/+4
| | | | | | | DHCP clients might try to resolve the assigned address to check if it's already in use: don't resolve the configured IPv4 address. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* Makefile: Quick hack to build convenience Debian and RPM packagesStefano Brivio2021-09-011-1/+15
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>