This fixes two issues:
- values for mutiple header fields with the same name
passed as arrays were converted to string;
- the type of field value wasn't preserved as required
by specification.
The nxt_murmur_hash2() generated 4-byte hash that was stored in uintptr_t,
which was 8 bytes long on 64-bit systems. At each iteration, it took the
previous key and hashed it again.
The problem was that it took only the first 4 bytes of the key, and these
4 bytes were always zero on 64-bit big-endian system. That resulted in
equal keys at each iteration.
The bug was discovered on IBM/S390x.
The nxt_conf_map_object() function used nxt_int_t for NXT_CONF_MAP_INT, which
was 8 bytes long on 64-bit systems.
But the nxt_port_main_start_worker_handler() used it to map into the int field
of the nxt_common_app_conf_t structure, which was 4 bytes. As the result, on
a 64-bit big-endian system all the meaningful module type numbers were assigned
into the gap above the "type" field.
The bug was discovered on IBM/S390x.
The issue was caused by misplacement of allocated blocks in rbtree due
to broken comparison function if the distance between two allocations
did not fit into intptr_t. As the result, nxt_mp_free() could have
failed to find the allocation.
In particular, it was mostly observed when Unit was compiled with
musl C library on 32-bits systems.
This closes#118 issue on GitHub.
- Fixed handling of the "options" parameter in Socket() constructor;
- Now the connect() method returns "this";
- Deduplicated the address() method;
- Added missing "callback" argument to the end() method;
- Now the destroy() method returns "this";
- Added "timeout" argument type check in the setTimeout() method.
epoll changes are committed to the kernel before epoll_wait() or
on changes array overflow. In the latter case if there are errors
epoll_wait() timeout was not set to zero.
This commit is related to #173 issue on GitHub.
Thanks to 洪志道 (Hong Zhi Dao).
Timers that don't require maximum precision (most of them, actually) can be
triggered earlier or later within the bias interval.
To reduce wakeups by timers, the expire function now triggers not only all
timers that fall within the elapsed time, but also those whose bias falls
within this interval.
Delete/insert operation complexity for a red-black tree is O(log n),
where n is the total number of tree elements.
If all delete operations are performed before all insert operations,
the average number of tree elements during an operation will be
lower than in the mixed-operations case.
To optimize rbtree operations, all changes are stored in array and later
processed in batches.
The previous implementation of this mechanics had a number of design flaws.
Each change was saved in a new array entry; until the changes were applied,
the timer remained in an intermediate state (NXT_TIMER_CHANGING).
This intermediate state didn't allow to identify if time was going to be
disabled or enabled. However, the nxt_conn_io_read() function relied on
this information; as a result, in some cases the read timeout wasn't set.
Also, the nxt_timer_delete() function did not reliably track whether a timer
was added to the work queue. It checked the NXT_TIMER_ENQUEUED state of
a timer, but this state could be reset to NXT_TIMER_DISABLED by a
nxt_timer_disable() call or another nxt_timer_delete() call.
Now, instead of keeping the whole history of the timer's changes, the new
implementation updates the timer state immediately, and only one operation
is added to the array to add or delete timer in the rbtree according
to its final state.
This issue was introduced in libunit commit (e0f0cd7d244a). All port
sockets in application should be in blocking mode whereas Unit itself
operates non-blocking sockets.
Having non-blocking sockets in application may cause send error during
intensive response packets generation.
See https://mailman.nginx.org/pipermail/unit/2018-October/000080.html.
