| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
|
| |
Also, some instances nni_aio are changed to nng_aio. We think we want to harmonize
some of these types going forward as it will reduce the need to include headers
hopefully letting us get away with just "defs.h" in more places.
|
| | |
|
| | |
|
| |
|
|
|
|
| |
We can retire the old approach that used separate allocations,
and all of the supporting code. This also gives us a more
natural signature for the end point initializations.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is a new transport API, which should make it easier for transports
to rely upon lifetime guarantees made by the common SP framework, thus
eliminating the need for transport specific reference counters, reap
lists, and similar.
The transport declares the size of the object in the ops vector (for
pipe, dialer, or listener), and the framework supplies one allocated
using the associated allocator.
For now these add the pipe object to the socket and endpoint using
linked linked lists. The plan is to transition those to reference
counts which should be lighter weight and free form locking issues.
The pipe teardown has been moved more fully to the reaper, to avoid
some of the deadlocks that can occur as nni_pipe_close can be called
from almost any context.
For now the old API is retained as well, but the intention is to convert
all the transports and then remove it.
|
| |
|
|
|
|
|
|
|
|
| |
This starts by using this for the nni_pipe, but we will use it
for the other primary objects as well. This should simplify
the tear down and hopefully eliminate some races.
It does mean that pipe destruction goes through an additional
context switch, for now at least. This shouldn't be on the hot
data path anyway.
|
| | |
|
| | |
|
| |
|
|
| |
the separation of nni_url and nng_url.
|
| |
|
|
|
|
|
| |
Also expose nng_url_sprintf() for users who need it.
This avoids some need to do dynamic memory on some things. Soon
the entirety of nng_url will be allocation free in the usual case.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
This eliminates several mutex operations done each time a pipe
is created or destroyed. For large scale systems this should
reduce overall pressure on the memory subsystem, and scale better
as many threads are coming and going.
This also reduces the overall size of nni_pipe -- on Linux by
36 bytes typically.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
| |
This should reduce the amount of copying, and the overall size
used by pipes and other objects quite a bit. (On my system, the
sizeof nni_pipe shrank by 400 bytes, for example.)
|
| |
|
|
|
|
|
|
|
|
|
| |
fixes #1288 id allocation can overallocate
fixes #1126 consider removing lock from idhash
This substantially refactors the id hash code, giving a cleaner API,
and eliminating a extra locking as well as some wasteful allocations.
The ZeroTier code has it's own copy, that is 64-bit friendly, as the
rest of the consumers need only a simpler 32-bit API.
|
| |
|
|
| |
fixes #1103 respondent could inline backtrace
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
fixes #698 Need TCP stats
fixes #699 Need IPC stats
fixes #701 Need TLS stats
This commit addresses a problem when negotiating using one of the stream
based negotiation APIs -- a slow or misbehaving peer can prevent well
behaved ones from establishing a connection. The fix is a fairly
significant change in how these transports link up, and it does rely
on the fact that the socket only has a single accept() or connect()
pending at a time (on a given endpoint that is).
While here, we have completely revamped the way transport statistics are
done, offering a standard API for collecting these statistics.
Unfortunately, this completely borks the statistics for inproc. As we
are planning to change the way inproc works soon, in order to provide
more control and work on performance fixes for the message queue, we feel
this is an acceptable trade-off. Furthermore, almost nobody uses inproc
for anything, and even fewer people are making use of the statistics
at this time.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This introduces new public APIs for obtaining statistics,
and adds some generic stats for dialers, listeners, pipes, and
sockets. Also added are stats for inproc and pairv1 protocol.
The other protocols and transports will have stats added
incrementally as time goes on.
A simple test program, and man pages are provided for this.
Start by looking at nng_stat(5).
Statistics does have some impact, and they can be disabled by
using the advanced NNG_ENABLE_STATS (setting it to OFF, it's
ON by default) if you need to build a minimized configuration.
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
fixes #599 nng_dial sync should not return until added to socket
This reintroduces the changes for the above fixes, building upon the
transport modifications that we have made to eliminate the separate
transport pipe start entry point. It also includes slightly reworked
code during start to put a hold on the pipe when it is created, which
we we drop at the end, hopefully fixing a use-after-free.
|
| |
|
|
|
| |
This changeset needs work. We are seeing errors described by
This reverts commit d7f7c896c0ede24249ef63b1e45b1878bf4bd473.
|
| |
|
|
|
|
|
|
|
|
| |
fixes #208 pipe start should occur before connect / accept
fixes #616 Race condition closing between header & body
This refactors the transports to handle their own connection
handshaking before passing the pipe to the socket. This
changes and simplifies the setup. This also fixes a rather
challenging race condition described by #616.
|
| | |
|
|
|
fixes #170 Make more use of reaper
This is a complete restructure/rethink of how child objects interact
with the socket. (This also backs out #576 as it turns out not to be
needed.) While 568 says reader/writer lock, for now we have settled
for a single writer lock. Its likely that this is sufficient.
Essentially we use the single socket lock to guard lists of the socket
children. We also use deferred deletion in the idhash to facilitate
teardown, which means endpoint closes are no longer synchronous.
We use the reaper to clean up objects when the reference count drops
to zero. We make a special exception for pipes, since they really
are not reference counted by their parents, and they are leaf objects
anyway.
We believe this addresses the main outstanding race conditions in
a much more correct and holistic way.
Note that endpoint shutdown is a little tricky, as it makes use of
atomic flags to guard against double entry, and against recursive
lock entry. This is something that would be nice to make a bit more
obvious, but what we have is safe, and the complexity is at least
confined to one place.
|
