Also started teaching it about toxcore's alloc/dealloc functions in
hopes of it catching some errors (it doesn't seem to be very good at
this, but maybe better than nothing?).
If the `recvbuf` network function returns 0 all the time, that means
there is never any data available on the TCP socket. This change makes
it so there is a random amount of data available on the TCP socket.
This invalidates the bootstrap fuzzer corpus.
This is the "server-side" part of the new friend finding system,
allowing DHT nodes to store small amounts of data and permit searching
for it. A forwarding (proxying) mechanism allows this to be used by TCP
clients, and deals with non-transitivity in the network.
The idea here is to have a `Network` object that contains functions for
network operations and an optional userdata object that can manage those
network operations. This allows e.g. a fuzzer to replace the network
functions with no-ops or fuzzer inputs, reducing the need for `#ifdef`s.
These were found by the new stronger type check in cimple. The one
bugfix is in `crypto_sha512_cmp`, which used to think `crypto_verify_32`
returns bool while actually it's -1/0/1.
Also added a whole bunch of logging that I needed while debugging the
issue. The solution in the end is that bootstrap needs to resolve IPs,
and getaddrinfo fails in the browser. Most of the time we bootstrap
against IPs anyway, so trying to parse as IP address first will shortcut
that.
We have a more portable wrapper that is now also thread-safe. Also
stopped using sprintf in the one place we used it. This doesn't really
help much, but it allows us to forbid sprintf globally.
This flag isn't helpful for the most common case of aggregate
initialisation, namely `{0}`. We don't want to be writing `{{{0}}}` with
some brittle number of braces.
Currently only `crypto_core_mem.c` needs this. We should try not to
depend on configure'd values. Also note: config.h is only created and
used in the autotools build. In CMake, we pass `-D` flags directly.
We now either log errors or properly propagate them. In the case of
ping_send_request we make the function return void because it doesn't
seem to matter whether or not it succeeds.
In the "no proxy" case, the `IP_Port` in the default "no proxy" default
proxy info is uninitialised. It is never used for any decisions in the
code, but it is copied in memory, making it a potential crash on systems
with trap representations of ints.
Reduced by, e.g.:
* `file_transfer_test`: 33% of the `clock_gettime` calls.
* `tox_many_test`: 53% of the `clock_gettime` calls.
Other tests will see similar improvements. Real world applications will
be closer to 40-50% improvement, since tox_many_test has 100 nodes, while
file_transfer_test has 2 nodes.
* Moved PAIR to toxav, where it's used (but really this should die).
* Replace most MIN calls with typed `min_*` calls. Didn't replace the
ones where the desired semantics are unclear. Moved the MIN macro to
the one place where it's still used.
* Avoid assignments in `while` loops. Instead, factored out the loop body
into a separate `bool`-returning function.
* Use named types for callbacks (`_cb` types).
* Avoid assignments in `if` conditions.
* Removed `MAKE_REALLOC` and expanded its two calls. We can't have
templates in C, and this fake templating is ugly and hard to analyse
and debug (it expands on a single line).
* Moved epoll system include to the .c file, out of the .h file.
* Avoid assignments in expressions (`a = b = c;`).
* Avoid multiple declarators per struct member declaration.
* Fix naming inconsistencies.
* Replace `net_to_host` macro with function.
It turns out, `unix_time` is also monotonic, and is used as such, so I've
renamed the new functions to `mono_time_*`.
2018-07-08:
```
00:01 <@irungentoo> the idea used to be that the unix_time() function
could go backward in time but I think I might have started using it like
if it could not after I changed it so that it would never go back in time
```
