Spotted in Debian's buildd logs: on ia64, clone(2) is not available:
the glibc wrapper is named __clone2() and it takes, additionally,
the size of the stack area passed by the caller.
Add a do_clone() wrapper handling the different cases, and also
taking care of pointing the child's stack in the middle of the
allocated area: on PA-RISC (hppa), handled by clone(), the stack
grows up, and on ia64 the stack grows down, but the register backing
store grows up -- and I think it might be actually used here.
Suggested-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
We recently corrected some errors handling the endianness of IPv4
addresses. These are very easy errors to make since although we mostly
store them in network endianness, we sometimes need to manipulate them in
host endianness.
To reduce the chances of making such mistakes again, change to always using
a (struct in_addr) instead of a bare in_addr_t or uint32_t to store network
endian addresses. This makes it harder to accidentally do arithmetic or
comparisons on such addresses as if they were host endian.
We introduce a number of IN4_IS_ADDR_*() helpers to make it easier to
directly work with struct in_addr values. This has the additional benefit
of making the IPv4 and IPv6 paths more visually similar.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
...instead of one fourth. On the main() -> conf() -> nl_sock_init()
call path, LTO from gcc 12 on (at least) x86_64 decides to inline...
everything: nl_sock_init() is effectively part of main(), after
commit 3e2eb4337b ("conf: Bind inbound ports with
CAP_NET_BIND_SERVICE before isolate_user()").
This means we exceed the maximum stack size, and we get SIGSEGV,
under any condition, at start time, as reported by Andrea on a recent
build for CentOS Stream 9.
The calculation of NS_FN_STACK_SIZE, which is the stack size we
reserve for clones, was previously obtained by dividing the maximum
stack size by two, to avoid an explicit check on architecture (on
PA-RISC, also known as hppa, the stack grows up, so we point the
clone to the middle of this area), and then further divided by two
to allow for any additional usage in the caller.
Well, if there are essentially no function calls anymore, this is
not enough. Divide it by eight, which is anyway much more than
possibly needed by any clone()d callee.
I think this is robust, so it's a fix in some sense. Strictly
speaking, though, we have no formal guarantees that this isn't
either too little or too much.
What we should do, eventually: check cloned() callees, there are just
thirteen of them at the moment. Note down any stack usage (they are
mostly small helpers), bonus points for an automated way at build
time, quadruple that or so, to allow for extreme clumsiness, and use
as NS_FN_STACK_SIZE. Perhaps introduce a specific condition for hppa.
Reported-by: Andrea Bolognani <abologna@redhat.com>
Fixes: 3e2eb4337b ("conf: Bind inbound ports with CAP_NET_BIND_SERVICE before isolate_user()")
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
In a few places we use the FWRITE() macro to open a file, replace it's
contents with a given string and close it again. There's no real
reason this needs to be a macro rather than just a function though.
Turn it into a function 'write_file()' and make some ancillary
cleanups while we're there:
- Add a return code so the caller can handle giving a useful error message
- Handle the case of short write()s (unlikely, but possible)
- Add O_TRUNC, to make sure we replace the existing contents entirely
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
While building with clang:
./util.h:176:24: warning: pragma diagnostic pop could not pop, no matching push [-Wunknown-pragmas]
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
If the user specifies an explicit loopback address for a port
binding, we're going to use that address for the 'tap' socket, and
the same exact address for the 'spliced' socket (because those are,
by definition, only bound to loopback addresses).
This means that the second binding will fail, and, unexpectedly, the
port is forwarded, but via tap device, which means the source address
in the namespace won't be a loopback address.
Make it explicit under which conditions we're creating which kind of
socket, by refactoring tcp_sock_init() into two separate functions
for IPv4 and IPv6 and gathering those conditions at the beginning.
Also, don't create spliced sockets if the user specifies explicitly
a non-loopback address, those are harmless but not desired either.
Fixes: 3c6ae62510 ("conf, tcp, udp: Allow address specification for forwarded ports")
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Since kernel version 5.7, commit c427bfec18f2 ("net: core: enable
SO_BINDTODEVICE for non-root users"), we can bind sockets to
interfaces, if they haven't been bound yet (as in bind()).
Introduce an optional interface specification for forwarded ports,
prefixed by %, that can be passed together with an address.
Reported use case: running local services that use ports we want
to have externally forwarded:
https://github.com/containers/podman/issues/14425
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Logging to file is going to add some further complexity that we don't
want to squeeze into util.c.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Commit c318ffcb4c ("udp: Ignore bogus -Wstringop-overread for
write() from gcc 12.1") uses a gcc pragma to ignore a bogus warning,
which started appearing on gcc 12.1 (aarch64 and x86_64) due to:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103483
...but gcc 12.2 has the same issue. Not just that: if LTO is enabled,
the pragma itself is ignored (this wasn't the case with gcc 12.1),
because of:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80922
Drop the pragma, and assign a frame (in the networking sense) pointer
from the offset of the Ethernet header in the buffer, then pass it to
write() and pcap(), so that gcc doesn't obsess anymore with the fact
that an Ethernet header is 14 bytes and we're sending more than that.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
We define our own ffsl() as a weak symbol, in case our C library doesn't
include it. On glibc systems which *do* include it, this causes a cppcheck
warning because unsurprisingly our version doesn't pick the same argument
names. Convert the suppression for this into an inline suppression.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
passt/pasta contains a number of routines designed to isolate passt from
the rest of the system for security. These are spread through util.c and
passt.c. Move them together into a new isolation.c file.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Currently the logic to work out what UID and GID we will run as is spread
across conf(). If --runas is specified it's handled in conf_runas(),
otherwise it's handled by check_root(), which depends on initialization of
the uid and gid variables by either conf() itself or conf_runas().
Make this clearer by putting all the UID and GID logic into a single
conf_ugid() function.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
check_root() both checks to see if we are root (in the init namespace),
and if we are drops to an unprivileged user. To make future cleanups
simpler, split the checking for root (now in check_root()) from the actual
dropping of privilege (now in drop_root()).
Note that this does slightly alter semantics. Previously we would only
setuid() if we were originally root (in the init namespace). Now we will
always setuid() and setgid(), though it won't actually change anything if
we weren't privileged to begin with. This also means that we will now
always attempt to switch to the user specified with --runas, even if we
aren't (init namespace) root to begin with. Obviously this will fail with
an error if we weren't privileged to start with. --help and the man page
are updated accordingly.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
c->uid and c->gid are first set in conf(), and last used in check_root()
itself called from conf(). Therefore these don't need to be fields in the
long lived context structure and can instead be locals in conf().
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The v4 and v6 fields of the context structure can be confusing, because
they change meaning part way through the code: Before conf_ip(), they are
booleans which indicate whether the -4 or -6 options have been given.
After conf_ip() they are DISABLED|ENABLED|PROBE enums which indicate
whether the IP version is available (which means both that it was allowed
on the command line and we were able to configure it). The PROBE variant
of the enum is only used locally within conf_ip() and since recent changes
there it no longer has a real purpose different from ENABLED.
Simplify this all by making the context fields always just a boolean
indicating the availability of the IP version. They both default to 1, but
can be set to 0 by either command line options or configuration failures.
We use some local variables in conf() for tracking the state of the command
line options on their own.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
[sbrivio: Minor coding style fix in conf.c]
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
On some systems, user and group "nobody" might not be available. The
new --runas option allows to override the default "nobody" choice if
started as root.
Now that we allow this, drop the initgroups() call that was used to
add any additional groups for the given user, as that might now
grant unnecessarily broad permissions. For instance, several
distributions have a "kvm" group to allow regular user access to
/dev/kvm, and we don't need that in passt or pasta.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
With current OpenSUSE Tumbleweed on aarch64 (gcc-12-1.3.aarch64) and
on x86_64 (gcc-12-1.4.x86_64), but curiously not on armv7hl
(gcc-12-1.3.armv7hl), gcc warns about using the _pointer_ to the
802.3 header to write the whole frame to the tap descriptor:
reading between 62 and 4294967357 bytes from a region of size 14
which is bogus:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103483
Probably declaring udp_sock_fill_data_v{4,6}() as noinline would
"fix" this, but that's on the data path, so I'd rather not. Use
a gcc pragma instead.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
This feature is available in slirp4netns but was missing in passt and
pasta.
Given that we don't do dynamic memory allocation, we need to bind
sockets while parsing port configuration. This means we need to
process all other options first, as they might affect addressing and
IP version support. It also implies a minor rework of how TCP and UDP
implementations bind sockets.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Implement a packet abstraction providing boundary and size checks
based on packet descriptors: packets stored in a buffer can be queued
into a pool (without storage of its own), and data can be retrieved
referring to an index in the pool, specifying offset and length.
Checks ensure data is not read outside the boundaries of buffer and
descriptors, and that packets added to a pool are within the buffer
range with valid offset and indices.
This implies a wider rework: usage of the "queueing" part of the
abstraction mostly affects tap_handler_{passt,pasta}() functions and
their callees, while the "fetching" part affects all the guest or tap
facing implementations: TCP, UDP, ICMP, ARP, NDP, DHCP and DHCPv6
handlers.
Suggested-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
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>
...those were used when epoll references used to be socket numbers,
they should have gone away a long time ago.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
--debug can be a bit too noisy, especially as single packets or
socket messages are logged: implement a new option, --trace,
implying --debug, that enables all debug messages.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
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>
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>
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>
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>
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>
SPDX tags don't replace license files. Some notices were missing and
some tags were not according to the SPDX specification, too.
Now reuse --lint from the REUSE tool (https://reuse.software/) passes.
Reported-by: Martin Hauke <mardnh@gmx.de>
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Replace libc functions that might dynamically allocate memory with own
implementations or wrappers.
Drop brk(2) from list of allowed syscalls in seccomp profile.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
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>
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>
Detecting bound ports at start-up time isn't terribly useful: do this
periodically instead, if configured.
This is only implemented for TCP at the moment, UDP is somewhat more
complicated: leave a TODO there.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
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>
Traffic with loopback source address will be forwarded to the direct
loopback connection in the namespace, and the tap interface is used
for the rest.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Provide an AVX2-based function using compiler intrinsics for
TCP/IP-style checksums. The load/unpack/add idea and implementation
is largely based on code from BESS (the Berkeley Extensible Software
Switch) licensed as 3-Clause BSD, with a number of modifications to
further decrease pipeline stalls and to minimise cache pollution.
This speeds up considerably data paths from sockets to tap
interfaces, decreasing overhead for checksum computation, with
16-64KiB packet buffers, from approximately 11% to 7%. The rest is
just syscalls at this point.
While at it, provide convenience targets in the Makefile for avx2,
avx2_debug, and debug targets -- these simply add target-specific
CFLAGS to the build.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Packets are received directly onto pre-cooked, static buffers
for IPv4 (with partial checksum pre-calculation) and IPv6 frames,
with pre-filled Ethernet addresses and, partially, IP headers,
and sent out from the same buffers with sendmmsg(), for both
passt and pasta (non-local traffic only) modes.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
PASTA (Pack A Subtle Tap Abstraction) provides quasi-native host
connectivity to an otherwise disconnected, unprivileged network
and user namespace, similarly to slirp4netns. Given that the
implementation is largely overlapping with PASST, no separate binary
is built: 'pasta' (and 'passt4netns' for clarity) both link to
'passt', and the mode of operation is selected depending on how the
binary is invoked. Usage example:
$ unshare -rUn
# echo $$
1871759
$ ./pasta 1871759 # From another terminal
# udhcpc -i pasta0 2>/dev/null
# ping -c1 pasta.pizza
PING pasta.pizza (64.190.62.111) 56(84) bytes of data.
64 bytes from 64.190.62.111 (64.190.62.111): icmp_seq=1 ttl=255 time=34.6 ms
--- pasta.pizza ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 34.575/34.575/34.575/0.000 ms
# ping -c1 spaghetti.pizza
PING spaghetti.pizza(2606:4700:3034::6815:147a (2606:4700:3034::6815:147a)) 56 data bytes
64 bytes from 2606:4700:3034::6815:147a (2606:4700:3034::6815:147a): icmp_seq=1 ttl=255 time=29.0 ms
--- spaghetti.pizza ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 28.967/28.967/28.967/0.000 ms
This entails a major rework, especially with regard to the storage of
tracked connections and to the semantics of epoll(7) references.
Indexing TCP and UDP bindings merely by socket proved to be
inflexible and unsuitable to handle different connection flows: pasta
also provides Layer-2 to Layer-2 socket mapping between init and a
separate namespace for local connections, using a pair of splice()
system calls for TCP, and a recvmmsg()/sendmmsg() pair for UDP local
bindings. For instance, building on the previous example:
# ip link set dev lo up
# iperf3 -s
$ iperf3 -c ::1 -Z -w 32M -l 1024k -P2 | tail -n4
[SUM] 0.00-10.00 sec 52.3 GBytes 44.9 Gbits/sec 283 sender
[SUM] 0.00-10.43 sec 52.3 GBytes 43.1 Gbits/sec receiver
iperf Done.
epoll(7) references now include a generic part in order to
demultiplex data to the relevant protocol handler, using 24
bits for the socket number, and an opaque portion reserved for
usage by the single protocol handlers, in order to track sockets
back to corresponding connections and bindings.
A number of fixes pertaining to TCP state machine and congestion
window handling are also included here.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
...sharing the same filesystem. Instead of a fixed path for the UNIX
domain socket, passt now uses a path with a counter, probing for
existing instances, and picking the first free one.
The demo script is updated accordingly -- it can now be started several
times to create multiple namespaces with an instance of passt each,
with addressing reflecting separate subnets, and NDP proxying between
them.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
