aboutgitcodebugslistschat
path: root/tcp.c
Commit message (Collapse)AuthorAgeFilesLines
* tcp: Refactor to use events instead of states, split out spliced implementationStefano Brivio2022-03-281-1495/+851
| | | | | | | | | | | | | | | | | | | | | Using events and flags instead of states makes the implementation much more straightforward: actions are mostly centered on events that occurred on the connection rather than states. An example is given by the ESTABLISHED_SOCK_FIN_SENT and FIN_WAIT_1_SOCK_FIN abominations: we don't actually care about which side started closing the connection to handle closing of connection halves. Split out the spliced implementation, as it has very little in common with the "regular" TCP path. Refactor things here and there to improve clarity. Add helpers to trace where resets and flag settings come from. No functional changes intended. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* seccomp, tcp: Add fcntl64 to pasta syscalls for armv6l, armv7lStefano Brivio2022-02-281-1/+1
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp, udp: Receive batching doesn't pay off when writing single frames to tapStefano Brivio2022-02-211-16/+20
| | | | | | | | | | | | | | | | In pasta mode, when we get data from sockets and write it as single frames to the tap device, we batch receive operations considerably, and then (conceptually) split the data in many smaller writes. It looked like an obvious choice, but performance is actually better if we receive data in many small frame-sized recvmsg()/recvmmsg(). The syscall overhead with the previous behaviour, observed by perf, comes predominantly from write operations, but receiving data in shorter chunks probably improves cache locality by a considerable amount. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt, pasta: Namespace-based sandboxing, defer seccomp policy applicationStefano Brivio2022-02-211-8/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To reach (at least) a conceptually equivalent security level as implemented by --enable-sandbox in slirp4netns, we need to create a new mount namespace and pivot_root() into a new (empty) mountpoint, so that passt and pasta can't access any filesystem resource after initialisation. While at it, also detach IPC, PID (only for passt, to prevent vulnerabilities based on the knowledge of a target PID), and UTS namespaces. With this approach, if we apply the seccomp filters right after the configuration step, the number of allowed syscalls grows further. To prevent this, defer the application of seccomp policies after the initialisation phase, before the main loop, that's where we expect bad things to happen, potentially. This way, we get back to 22 allowed syscalls for passt and 34 for pasta, on x86_64. While at it, move #syscalls notes to specific code paths wherever it conceptually makes sense. We have to open all the file handles we'll ever need before sandboxing: - the packet capture file can only be opened once, drop instance numbers from the default path and use the (pre-sandbox) PID instead - /proc/net/tcp{,v6} and /proc/net/udp{,v6}, for automatic detection of bound ports in pasta mode, are now opened only once, before sandboxing, and their handles are stored in the execution context - the UNIX domain socket for passt is also bound only once, before sandboxing: to reject clients after the first one, instead of closing the listening socket, keep it open, accept and immediately discard new connection if we already have a valid one Clarify the (unchanged) behaviour for --netns-only in the man page. To actually make passt and pasta processes run in a separate PID namespace, we need to unshare(CLONE_NEWPID) before forking to background (if configured to do so). Introduce a small daemon() implementation, __daemon(), that additionally saves the PID file before forking. While running in foreground, the process itself can't move to a new PID namespace (a process can't change the notion of its own PID): mention that in the man page. For some reason, fork() in a detached PID namespace causes SIGTERM and SIGQUIT to be ignored, even if the handler is still reported as SIG_DFL: add a signal handler that just exits. We can now drop most of the pasta_child_handler() implementation, that took care of terminating all processes running in the same namespace, if pasta started a shell: the shell itself is now the init process in that namespace, and all children will terminate once the init process exits. Issuing 'echo $$' in a detached PID namespace won't return the actual namespace PID as seen from the init namespace: adapt demo and test setup scripts to reflect that. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Address new clang-tidy warnings from LLVM 13.0.1Stefano Brivio2022-01-301-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | clang-tidy from LLVM 13.0.1 reports some new warnings from these checkers: - altera-unroll-loops, altera-id-dependent-backward-branch: ignore for the moment being, add a TODO item - bugprone-easily-swappable-parameters: ignore, nothing to do about those - readability-function-cognitive-complexity: ignore for the moment being, add a TODO item - altera-struct-pack-align: ignore, alignment is forced in protocol headers - concurrency-mt-unsafe: ignore for the moment being, add a TODO item Fix bugprone-implicit-widening-of-multiplication-result warnings, though, that's doable and they seem to make sense. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt, tap: Daemonise once socket is ready without waiting for connectionStefano Brivio2022-01-281-1/+1
| | | | | | | | | | | | | The existing behaviour is not really practical: an automated agent in charge of starting both qemu and passt would need to fork itself to start passt, because passt won't fork to background until qemu connects, and the agent needs to unblock to start qemu. Instead of waiting for a connection to daemonise, do it right away as soon as a socket is available: that can be considered an initialised state already. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* seccomp: Add a number of alternate and per-arch syscallsStefano Brivio2022-01-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Depending on the C library, but not necessarily in all the functions we use, statx() might be used instead of stat(), getdents() instead of getdents64(), readlinkat() instead of readlink(), openat() instead of open(). On aarch64, it's clone() and not fork(), and dup3() instead of dup2() -- just allow the existing alternative instead of dealing with per-arch selections. Since glibc commit 9a7565403758 ("posix: Consolidate fork implementation"), we need to allow set_robust_list() for fork()/clone(), even in a single-threaded context. On some architectures, epoll_pwait() is provided instead of epoll_wait(), but never both. Same with newfstat() and fstat(), sigreturn() and rt_sigreturn(), getdents64() and getdents(), readlink() and readlinkat(), unlink() and unlinkat(), whereas pipe() might not be available, but pipe2() always is, exclusively or not. Seen on Fedora 34: newfstatat() is used on top of fstat(). syslog() is an actual system call on some glibc/arch combinations, instead of a connect()/send() implementation. On ppc64 and ppc64le, _llseek(), recv(), send() and getuid() are used. For ppc64 only: ugetrlimit() for the getrlimit() implementation, plus sigreturn() and fcntl64(). On s390x, additionally, we need to allow socketcall() (on top of socket()), and sigreturn() also for passt (not just for pasta). Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Don't round down MSS to >= 64KiB page size, but clamp it in any caseStefano Brivio2022-01-261-3/+3
| | | | | | | On some architectures, the page size is bigger than the maximum size of an Ethernet frame. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp, udp, util: Fixes for bitmap handling on big-endian, castsStefano Brivio2022-01-261-1/+1
| | | | | | | | Bitmap manipulating functions would otherwise refer to inconsistent sets of bits on big-endian architectures. While at it, fix up a couple of casts. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp, netlink, HAS{BYTES_ACKED,MIN_RTT,GETRANDOM} and NETLINK_GET_STRICT_CHKStefano Brivio2022-01-261-0/+36
| | | | | | | | | | | | | | | | tcpi_bytes_acked and tcpi_min_rtt are only available on recent kernel versions: provide fall-back paths (incurring some grade of performance penalty). Support for getrandom() was introduced in Linux 3.17 and glibc 2.25: provide an alternate mechanism for that as well, reading from /dev/random. Also check if NETLINK_GET_STRICT_CHK is defined before using it: it's not strictly needed, we'll filter out irrelevant results from netlink anyway. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Drop <linux/ipv6.h> include, carry own ipv6hdr and opt_hdr definitionsStefano Brivio2022-01-261-1/+0
| | | | | | | This is the only remaining Linux-specific include -- drop it to avoid clang-tidy warnings and to make code more portable. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tap, tcp: Fix two comparisons with different signedness reported by gcc 7Stefano Brivio2022-01-261-1/+1
| | | | | | For some reason, those are not reported by recent versions of gcc. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Cover all usages of tcpi_snd_wnd with HAS_SND_WNDStefano Brivio2022-01-261-0/+2
| | | | | | ...I forgot two of them. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Restore source address to network endianness before using it for hash tableStefano Brivio2021-10-211-2/+3
| | | | | | | | | This was actually fine "on the wire", but it's inconsistent with the way we hash other addresses/protocols and also ends up with a wrong endianness in captures in case we replace the address with our default gateway. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Add cppcheck target, test, and address resulting warningsStefano Brivio2021-10-211-53/+56
| | | | | | | ...mostly false positives, but a number of very relevant ones too, in tcp_get_sndbuf(), tcp_conn_from_tap(), and siphash PREAMBLE(). Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Fix build with gcc 7, use std=c99, enable some more Clang checkersStefano Brivio2021-10-211-82/+90
| | | | | | | | | | | | | | Unions and structs, you all have names now. Take the chance to enable bugprone-reserved-identifier, cert-dcl37-c, and cert-dcl51-cpp checkers in clang-tidy. Provide a ffsl() weak declaration using gcc built-in. Start reordering includes, but that's not enough for the llvm-include-order checker yet. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* Makefile, tcp: Don't try to use tcpi_snd_wnd from tcp_info on pre-5.3 kernelsStefano Brivio2021-10-211-1/+8
| | | | | | | Detect missing tcpi_snd_wnd in struct tcp_info at build time, otherwise build fails with a pre-5.3 linux/tcp.h header. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* ndp, dhcpv6, tcp, udp: Always use link-local as source if gateway isn'tStefano Brivio2021-10-201-2/+10
| | | | | | | | | | | | This shouldn't happen on any sane configuration, but I just met an example of that: the default IPv6 gateway on the host is configured with a global unicast address, we use that as source for RA, DHCPv6 replies, and the guest ignores it. Same later on if we talk TCP or UDP and the guest has no idea where that address comes from. Use our link-local address in case the gateway address is global. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Add clang-tidy Makefile target and test, take care of warningsStefano Brivio2021-10-201-30/+32
| | | | | | | Most are just about style and form, but a few were actually serious mistakes (NDP-related). Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Address warnings from Clang's scan-buildStefano Brivio2021-10-201-16/+21
| | | | | | All false positives so far. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Address gcc 11 warningsStefano Brivio2021-10-201-11/+25
| | | | | | | | | A mix of unchecked return values, a missing permission mask for open(2) with O_CREAT, and some false positives from -Wstringop-overflow and -Wmaybe-uninitialized. Reported-by: Martin Hauke <mardnh@gmx.de> Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Fix for non-blocking splice() on older kernelsStefano Brivio2021-10-191-2/+2
| | | | | | | | For some reason, on 4.19, splice() doesn't honour SOCK_NONBLOCK from accept4() while reading from a TCP socket. Pass SPLICE_F_NONBLOCK explicitly in all cases. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Fix ACK reporting on older kernels (no tcp.kernel_snd_wnd case)Stefano Brivio2021-10-191-3/+4
| | | | | | | | If the window isn't updated on !c->tcp.kernel_snd_wnd, we still have to send ACKs if the ACK sequence was updated, or if an error occurred while querying TCP_INFO. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Arm tcp_data_noack on insufficient window too, don't reset if ACK ↵Stefano Brivio2021-10-161-2/+4
| | | | | | | | | | doesn't match ...and while at it, reverse the operands in the window equality comparison to detect the need for fast re-transmit: it's easier to read this way. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: ...and so I got a socket called zeroStefano Brivio2021-10-151-35/+44
| | | | | | | | I thought I'd get away with it, but no, after some clean-ups, I finally got a socket with number 0. Fix up all the convenient, yet botched assumptions. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Bump TCP_TAP_FRAMES back to 256Stefano Brivio2021-10-151-1/+1
| | | | | | With a batched sendmsg(), this is now beneficial. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Get rid of iov with cached MSS, drop sendmmsg(), add deferred flushStefano Brivio2021-10-151-139/+130
| | | | | | | | | | | | | | | Caching iov_len for messages from socket doesn't actually decrease overhead by the tiniest bit, and added a lot of complexity. Drop that. Also drop the sendmmsg(), we don't need to send multiple messages with TCP, as long as we make sure no messages with a length descriptor are sent partially, qemu is fine with it. Just like it's done for segments without data (flags), also defer the sendmsg() for sending data segments, to improve batching. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Clamp MSS depending on IP version, properly derive buffer sizesStefano Brivio2021-10-151-17/+19
| | | | | | | It makes no sense to include an IPv6 header in the calculation for clamping MSS on IPv4. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Explicitly align IP headers in tcp4_l2_{,flags}buf_t also in non-AVX2 buildStefano Brivio2021-10-141-14/+12
| | | | | | | Otherwise, tcp4_l2_flags_buf_t is not consistent with tcp4_l2_buf_t and header fields get all mixed up in tcp_l2_buf_fill_headers(). Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* conf, tcp, udp: Add --no-map-gw to disable mapping gateway address to hostStefano Brivio2021-10-141-2/+3
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt, pasta: Add seccomp supportStefano Brivio2021-10-141-0/+11
| | | | | | | | | | | | | | | | | | List of allowed syscalls comes from comments in the form: #syscalls <list> for syscalls needed both in passt and pasta mode, and: #syscalls:pasta <list> #syscalls:passt <list> for syscalls specifically needed in pasta or passt mode only. seccomp.sh builds a list of BPF statements from those comments, prefixed by a binary search tree to keep lookup fast. While at it, clean up a bit the Makefile using wildcards. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Decrease pool size for pipes to 16Stefano Brivio2021-10-141-1/+1
| | | | | | | This should be a reasonable balance between quick connection establishment and a fast start-up. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* conf, tap: Split netlink and pasta functions, allow interface configurationStefano Brivio2021-10-141-1/+1
| | | | | | | | | | Move netlink routines to their own file, and use netlink to configure or fetch all the information we need, except for the TUNSETIFF ioctl. Move pasta-specific functions to their own file as well, add parameters and calls to configure the tap interface in the namespace. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* pasta: Allow specifying paths and names of namespacesGiuseppe Scrivano2021-10-071-5/+5
| | | | | | | | | | | | | | | | | | Based on a patch from Giuseppe Scrivano, this adds the ability to: - specify paths and names of target namespaces to join, instead of a PID, also for user namespaces, with --userns - request to join or create a network namespace only, without entering or creating a user namespace, with --netns-only - specify the base directory for netns mountpoints, with --nsrun-dir Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com> [sbrivio: reworked logic to actually join the given namespaces when they're not created, implemented --netns-only and --nsrun-dir, updated pasta demo script and man page] Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Check if timestamp is passed also while sending FIN to tap/guestStefano Brivio2021-10-051-1/+1
| | | | | | | ...it's probably possible that we might need to reset a connection together with a FIN segment. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Drop EPOLLOUT for connections being established earlierStefano Brivio2021-10-051-3/+3
| | | | | | | | That's the first thing we have to do, before sending SYN, ACK: if tcp_send_to_tap() fails, we'll get a lot of useless events otherwise. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* passt: Shrink binary size by dropping static initialisersStefano Brivio2021-10-051-44/+49
| | | | | | ...from 11MiB to 155KiB for 'make avx2', 95KiB with -Os and stripped. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Simplify ACK-sending conditions in tcp_data_from_tap()Stefano Brivio2021-10-051-20/+9
| | | | | | | | Now that we have a proper function checking when and how to send ACKs and window updates, we don't need to duplicate this logic in tcp_data_from_tap(). Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Always probe SO_SNDBUF, second attemptStefano Brivio2021-10-051-3/+0
| | | | | | I fell for this already: the sending buffer might shrink later! Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Defer and coalesce all segments with no data (flags) to handlerStefano Brivio2021-10-051-142/+376
| | | | | | | | | | ...using pre-cooked buffers, just like we do with other segments. While at it, remove some code duplication by having separate functions for updating ACK sequence and window, and for filling in buffer headers. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Increase LOW_RTT_THRESHOLD to 10usStefano Brivio2021-10-051-1/+1
| | | | | | Sometimes we can get up to 6-7us minimum RTT for local connections too. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Reduce size of socket poolsStefano Brivio2021-10-051-2/+2
| | | | | | | A large pool helps marginally with CRR latency, but has detrimental effects on TCP memory pressure. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Increase TCP_TAP_FRAMES once moreStefano Brivio2021-10-051-1/+1
| | | | | | | With an increased sending buffer size for the AF_UNIX socket, we can get slightly lower overhead. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp, tap: Turn tcp_probe_mem() into sock_probe_mem(), use for AF_UNIX socket tooStefano Brivio2021-10-051-35/+3
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Fix botched timeout comparisonStefano Brivio2021-10-041-2/+2
| | | | Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Check pending ACK every two thirds of window, not every halfStefano Brivio2021-10-041-1/+1
| | | | | | | ...to spare some syscalls. If it's not enough, the timer will take care of it. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Don't set ACK flag while merely updating window valueStefano Brivio2021-10-041-15/+19
| | | | | | The receiver might take this as a duplicate ACK othewise. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Set TCP_TAP_FRAMES back to 32Stefano Brivio2021-10-041-1/+1
| | | | | | | Now that we fixed the issue with small receiving buffers, we can safely increase this back and get slightly lower syscall overhead. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Probe net.core.{r,w}mem_max, don't set SO_{RCV,SND}BUF if lowStefano Brivio2021-10-041-16/+22
| | | | | | | | | | | | | | | If net.core.rmem_max and net.core.wmem_max sysctls have low values, we can get bigger buffers by not trying to set them high -- the kernel would lock their values to what we get. Try, instead, to get bigger buffers by queueing as much as possible, and if maximum values in tcp_wmem and tcp_rmem are bigger than this, that will work. While at it, drop QUICKACK option for non-spliced sockets, I set that earlier by mistake. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
* tcp: Check if connection is local or low RTT was seen before using large MSSStefano Brivio2021-10-041-0/+59
| | | | | | | | If the connection is local or the RTT was comparable to the time it takes to queue a batch of messages, we can safely use a large MSS regardless of the sending buffer, but otherwise not. Signed-off-by: Stefano Brivio <sbrivio@redhat.com>