Theoretical only, but we could leak an fd if we closed a conn before
the fd was sent. This doesn't happen in our current codebase because
we only hand fds to connectd, which only closes at shutdown.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
A fairly simple change: ccan/io will now call the underlying I/O
routines repeatedly until they indicate they are unfinished, *or* fail
with EAGAIN. This should make a significant difference to large
nodes, which currently spend far too much time calling poll() to
discover a single fd is still writable (mainly, for streaming gossip).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: connectd: now should use far less CPU on large nodes.
```
ccan/ccan/base64/base64.c:34:10: error: result of comparison of constant 255 with expression of type 'int8_t' (aka 'signed char') is always false [-Werror,-Wtautological-constant-out-of-range-compare]
if (ret == (char)0xff) {
~~~ ^ ~~~~~~~~~~
ccan/ccan/base64/base64.c:44:57: error: result of comparison of constant 255 with expression of type 'const signed char' is always true [-Werror,-Wtautological-constant-out-of-range-compare]
return (maps->decode_map[(const unsigned char)b64char] != (char)0xff);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~~
```
Reported-by: Christian Decker
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This adds:
1. ability to search for an option by name.
2. allowance to set our own bits when registering options.
3. show callbacks which can say "don't show", and variable length.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Most importantly, configurator used to use bitshifts on signed
integers which -fsanitize=undefined caught.
But also, tal played fast and loose with typing and aliases, which was
a signficant amount of rework.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
```
lightningd/jsonrpc.c: In function ‘destroy_json_command’:
lightningd/jsonrpc.c:1180:63: error: the comparison will always evaluate as ‘false’ for the address of ‘canary’ will never be NULL [-Werror=address]
lightningd/jsonrpc.c:108:53: note: ‘canary’ declared here
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Which includes not asserting in timer.c should time go backwards.
Fixes: #4401
Changelog-Fixed: lightningd: don't assert if time goes backwards temporarily.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Otherwise we get a configurator failure:
In file included from /usr/include/string.h:495,
from configuratortest.c:2:
In function ‘strncpy’,
inlined from ‘main’ at configuratortest.c:6:2:
/usr/include/x86_64-linux-gnu/bits/string_fortified.h:106:10: warning: ‘__builtin_strncpy’ specified bound 8 equals destination size [-Wstringop-truncation]
106 | return __builtin___strncpy_chk (__dest, __src, __len, __bos (__dest));
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I was wondering why TAGS was missing some functions, and finally
tracked it down: PRINTF_FMT() confuses etags if it's at the start
of a function, and it ignores the rest of the file.
So we put PRINTF_FMT at the end, but that doesn't work for
*definitions*, only *declarations*. So we remove it from definitions
and add gratuitous declarations in the few static places.1
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Dynamic plugins were keeping fds open; they should not have these
at all anyway, but worse, they interfere with operation because
we don't notice they're closed.
The symptom was that shutdown of the test_plugin_slowinit and
test_plugin_command was 30 seconds (10 seconds grace to kill each daemon).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Dumb programs which have a --daemon option call fork() early. This is
terrible UX since startup errors get lost: the program exits with
"success" immediately then you discover via the logs that it didn't
start at all.
However, forking late introduced a heap of problems with changing
pids. Instead, fork early but keep stderr and the parent around: if
we fail early on, the parent fails with us. We release our parent
with an explicit action just before the main loop.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
configurator failed under clang:
checking for #pragma omp and -fopenmp support... ccan/tools/configurator/configurator: Test for HAVE_OPENMP failed with 32512:
./configurator.out: error while loading shared libraries: libomp.so: cannot open shared object file: No such file or directory
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
These are generalized from our internal implementations.
The main difference is that 'struct json_escaped' is now 'struct
json_escape', so we replace that immediately.
The difference between lightningd's json-writing ringbuffer and the
more generic ccan/json_out is that the latter has a better API and
handles escaping transparently if something slips through (though
it does offer direct accessors so you can mess things up yourself!).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I decided to try a faster implementation, only to find our crc32c was
not correct! Ouch.
I removed the crc32c functions from ccan/crc, and added a new crc32c
module which has the Mark Adler x86-64-optimized variants.
We bump gossip_store version again, since csums have changed.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Brings in fixes for closing stderr in parent for pipecmd (oops!)
and configurator fix (which we don't need yet)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Note that this changes the order of arguments to pipecmd to match the
documentation, so we fix all the callers!
Also make configure re-run when configurator changes.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
