If the cleanup function returns the appropriate type, use that to reset the
variable. For other functions (usually the foreign ones which return void), add
an explicit value to reset to.
This causes a bit of code churn, but I think it might be worth it. In a
following patch static destructors will be called from a fuzzer, and this
change allows them to be called multiple times. But I think such a change might
help with detecting unitialized code reuse too. We hit various bugs like this,
and things are more obvious when a pointer has been set to NULL.
I was worried whether this change increases text size, but it doesn't seem to:
-Dbuildtype=debug:
before "tree-wide: return NULL from freeing functions":
-rwxrwxr-x 1 zbyszek zbyszek 4117672 Feb 16 14:36 build/libsystemd.so.0.30.0*
-rwxrwxr-x 1 zbyszek zbyszek 4494520 Feb 16 15:06 build/systemd*
after "tree-wide: return NULL from freeing functions":
-rwxrwxr-x 1 zbyszek zbyszek 4117672 Feb 16 14:36 build/libsystemd.so.0.30.0*
-rwxrwxr-x 1 zbyszek zbyszek 4494576 Feb 16 15:10 build/systemd*
now:
-rwxrwxr-x 1 zbyszek zbyszek 4117672 Feb 16 14:36 build/libsystemd.so.0.30.0*
-rwxrwxr-x 1 zbyszek zbyszek 4494640 Feb 16 15:15 build/systemd*
-Dbuildtype=release:
before "tree-wide: return NULL from freeing functions":
-rwxrwxr-x 1 zbyszek zbyszek 5252256 Feb 14 14:47 build-rawhide/libsystemd.so.0.30.0*
-rwxrwxr-x 1 zbyszek zbyszek 1834184 Feb 16 15:09 build-rawhide/systemd*
after "tree-wide: return NULL from freeing functions":
-rwxrwxr-x 1 zbyszek zbyszek 5252256 Feb 14 14:47 build-rawhide/libsystemd.so.0.30.0*
-rwxrwxr-x 1 zbyszek zbyszek 1834184 Feb 16 15:10 build-rawhide/systemd*
now:
-rwxrwxr-x 1 zbyszek zbyszek 5252256 Feb 14 14:47 build-rawhide/libsystemd.so.0.30.0*
-rwxrwxr-x 1 zbyszek zbyszek 1834184 Feb 16 15:16 build-rawhide/systemd*
I would expect that the compiler would be able to elide the setting of a
variable if the variable is never used again. And this seems to be the case:
in optimized builds there is no change in size whatsoever. And the change in
size in unoptimized build is negligible.
Something strange is happening with size of libsystemd: it's bigger in
optimized builds. Something to figure out, but unrelated to this patch.
I started working on this because I wanted to change how
DEFINE_TRIVIAL_CLEANUP_FUNC is defined. Even independently of that change, it's
nice to make make things more consistent and predictable.
A previous commit made sure that when one of our own packets is looped
back to us, we ignore it. But let's go one step further, and refuse
operation if we notice the server we talk to is our own. This way we
won't generate unnecessary traffic and can return a cleaner error.
Fixes: #17413
It's rather convenient to be able to read all three types with this function.
Strictly speaking this change is not fully compatible, in case someone was
relying on sd_bus_message_read_strv() returning an error for anything except
"as", but I hope nobody was doing that.
Let's use uint32_t everywhere to maintain the seqno, since that's what
the kernel does. Prviously in the reply_callback logic we used 64bit,
for no apparent reason.
Using 32bit also provides us with the benefit that we can avoid using
uint64_hash_ops, and can use trivial_hash_ops instead for the reply
hashmap, so that we can store the seqno in the key pointer directly.
While we are at it, let's make sure we never run into serial collisions
internally (32bit is a lot, but not that much), and let's put a limit on
outstanding serials, to catch programming errors.
As suggested in https://github.com/systemd/systemd/pull/11484#issuecomment-775288617.
This does not touch anything exposed in src/systemd. Changing the defines there
would be a compatibility break.
Note that tests are broken after this commit. They will be fixed in the next one.
A hole was/is present after the booleans, so changing them to be one byte each
doesn't change the structure size (122 bits on amd64). If we add more stuff
later, it might make sense to turn some of those into bitfields again. For now,
let's take the easy route. EINVAL fits into type now.
Code size it minimally reduced:
-rwxrwxr-x 1 zbyszek zbyszek 4109792 Feb 10 14:00 build/libsystemd.so.0.30.0
-rwxrwxr-x 1 zbyszek zbyszek 4109712 Feb 10 14:01 build/libsystemd.so.0.30.0
By rearranging the fields, we can avoid one of the holes (on amd64).
By adding more space for .type, -EINVAL can be used as a value later on.
The structure is reduced from 96 to 88 bytes (on amd64).
Text size is also smaller:
-rwxrwxr-x 1 zbyszek zbyszek 4109832 Feb 9 19:50 build/libsystemd.so.0.30.0
-rwxrwxr-x 1 zbyszek zbyszek 4109792 Feb 9 19:51 build/libsystemd.so.0.30.0
This substantially beefs up the local DNS stub feature set in order to
allow local clients to do DNSSEC validation through the stub.
Previously we'd return NOTIMP if we'd get a DO or DO+CD lookup. With
this change we'll instead:
1. If we get DO+CD requests (i.e. DNSSEC with no local checking) we'll
proxy DNS queries and response mostly unmodified to/from upstream DNS
servers if possible (this is called "bypass" mode). We will patch in
new request IDs, (and patch them back out on reply), so that we can
sanely keep track of things. We'll also maintain a minimal local
cache for such lookups, always keeping the whole DNS packets in it
(if we reply from cache we'll patch the TTLs of all included RRs).
2. If we get DO requests without CD (i.e. DNSSEC with local checking)
we'll resolve and validate locally. In this mode we will not proxy
packets, but generate our own. We will however cache the combination
of answer RRs (along with their packet section assignments) we got
back in the cache, and use this information to generate reply packets
from the DNS stub.
In both cases: if we determine a lookup is to be answered from LLMNR or
mDNS we'll always revert to non-DNSSEC, non-proxy operation as before.
Answers will lack the DO bit then, since the data cannot be validated
via DNSSEC by the clients.
To make this logic more debuggable, this also adds query flags for
turning off RR sources. i.e. cache/network/zone/trust anchor/local
synthesis may now be disabled individually for each lookup.
The cache is substantially updated to make all this work: in addition to
caching simple RRs for lookup RR keys, we'll now cache the whole packets
and the whole combination of RRs, so that we can answer DO and DO+CD
replies sensibly according to the rules described above. This sounds
wasteful, but given that the
DnsResourceRecord/DnsResourceKey/DnsAnswer/DnsPacket
objects are all ref-counted and we try to merge references the actual
additional memory used should be limited (but this might be something to
optimize further later on).
To implement classic RR key lookups and new-style packet proxy lookups
(i.e. the ones necessary for DO+CD packet proxying, as described above)
DnsTransaction and DnsQuery objects now always maintain either a
DnsResourceKey/DnsQuestion as lookup key or a DnsPacket for "bypass"
mode.
Fixes: #4621#17218
When epoll_wait() returns number equivalent to the buffer size, there
may exist remaining events which may have higher priority. To make priority
sorting correctly, let's retrieve all events.
Even though many of those scripts are very simple, it is easier to include
the header than to try to say whether each of those files is trivial enough
not to require one.
While sd-bus already provides sd_bus_call() for calling a method
from a complete bus message object, We don't have an equivalent
function for replying from a method with a complete bus message
object.
Currently, we use sd_bus_send(call->bus, m, NULL) instead. Let's
add a shorthand for this pattern and name it sd_bus_reply().
Zbigniew Jędrzejewski-Szmek
Lennart Poettering
Yu Watanabe
Susant Sahani
Daan De Meyer
Dan Streetman