construct/ircd/server.cc

// Matrix Construct
//
// Copyright (C) Matrix Construct Developers, Authors & Contributors
// Copyright (C) 2016-2018 Jason Volk <jason@zemos.net>
//
// Permission to use, copy, modify, and/or distribute this software for any
// purpose with or without fee is hereby granted, provided that the above
// copyright notice and this permission notice is present in all copies. The
// full license for this software is available in the LICENSE file.

namespace ircd::server
{
	// Internal state
	ctx::dock dock;     // internal semaphore

	// Internal util
	template<class F> static size_t accumulate_peers(F&&);
	template<class F> static size_t accumulate_links(F&&);
	template<class F> static size_t accumulate_tags(F&&);
	static string_view canonize(const hostport &); // TLS buffer

	// Internal control
	static decltype(ircd::server::peers)::iterator
	create(const net::hostport &, decltype(peers)::iterator &);
}

decltype(ircd::server::log)
ircd::server::log
{
	"server", 'S'
};

ircd::conf::item<ircd::seconds>
close_all_timeout
{
	{ "name",     "ircd.server.close_all_timeout" },
	{ "default",  2L                              },
};

//
// init
//

ircd::server::init::init()
noexcept
{
}

ircd::server::init::~init()
noexcept
{
	interrupt(), close(), wait();
	peers.clear();
	log::debug
	{
		log, "All server peers, connections, and requests are clear."
	};
}

void
ircd::server::init::wait()
{
	static const auto finished
	{
		[] { return !peer_unfinished(); }
	};

	while(!dock.wait_for(seconds(5), finished))
	{
		for(const auto &[name, peer] : peers)
			log::dwarning
			{
				log, "Waiting for peer %s tags:%zu links:%zu err:%b op[r:%b f:%b]",
				name,
				peer->tag_count(),
				peer->link_count(),
				peer->err_has(),
				peer->op_resolve,
				peer->op_fini,
			};

		log::warning
		{
			log, "Waiting for %zu tags on %zu links on %zu of %zu peers to close...",
			tag_count(),
			link_count(),
			peer_unfinished(),
			peer_count()
		};
	}
}

void
ircd::server::init::close()
{
	log::debug
	{
		log, "Closing all %zu peers",
		peer_count()
	};

	net::close_opts opts;
	opts.timeout = seconds(close_all_timeout);
	for(auto &peer : peers)
		peer.second->close(opts);
}

void
ircd::server::init::interrupt()
{
	log::debug
	{
		log, "Cancel %zu tags on %zu links on %zu peers",
		tag_count(),
		link_count(),
		peer_count()
	};

	for(auto &peer : peers)
		peer.second->cancel();
}

//
// ircd::server
//

ircd::server::peer &
ircd::server::get(const net::hostport &hostport)
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	auto it
	{
		peers.lower_bound(hostcanon)
	};

	if(it == peers.end() || it->first != hostcanon)
		it = create(hostport, it);

	return *it->second;
}

decltype(ircd::server::peers)::iterator
ircd::server::create(const net::hostport &hostport,
                     decltype(peers)::iterator &it)
{
	auto peer
	{
		std::make_unique<server::peer>(hostport)
	};

	log::debug
	{
		log, "peer(%p) for %s created; adding...",
		peer.get(),
		peer->hostcanon,
	};

	assert(bool(peer));
	assert(!empty(peer->hostcanon));
	const string_view key{peer->hostcanon};
	it = peers.emplace_hint(it, key, std::move(peer));

	assert(it->second->hostcanon.data() == it->first.data());
	it->second->resolve(it->second->open_opts.hostport);
	return it;
}

ircd::server::peer &
ircd::server::find(const net::hostport &hostport)
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	return *peers.at(hostcanon);
}

bool
ircd::server::exists(const net::hostport &hostport)
noexcept
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	return peers.find(hostcanon) != end(peers);
}

bool
ircd::server::errclear(const net::hostport &hostport)
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	const auto it
	{
		peers.find(hostcanon)
	};

	if(it == end(peers))
		return false;

	auto &peer
	{
		*it->second
	};

	return peer.err_clear();
}

bool
ircd::server::avail(const net::hostport &hostport)
noexcept
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	const auto it
	{
		peers.find(hostcanon)
	};

	return it != end(peers)?
		!it->second->err_has():
		false;
}

bool
ircd::server::linked(const net::hostport &hostport)
noexcept
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	const auto it
	{
		peers.find(hostcanon)
	};

	return it != end(peers)?
		it->second->link_count():
		false;
}

bool
ircd::server::errant(const net::hostport &hostport)
noexcept
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	const auto it
	{
		peers.find(hostcanon)
	};

	return it != end(peers)?
		it->second->err_has():
		false;
}

ircd::string_view
ircd::server::errmsg(const net::hostport &hostport)
noexcept
{
	const auto hostcanon
	{
		server::canonize(hostport)
	};

	const auto it
	{
		peers.find(hostcanon)
	};

	return it != end(peers)?
		it->second->err_msg():
		string_view{};
}

bool
ircd::server::for_each(const request::each_closure &closure)
{
	for(const auto &[name, peer] : peers)
		if(!for_each(*peer, closure))
			return false;

	return true;
}

bool
ircd::server::for_each(const peer &peer,
                       const request::each_closure &closure)
{
	for(const auto &link : peer.links)
		if(!for_each(link, closure))
			return false;

	return true;
}

bool
ircd::server::for_each(const link &link,
                       const request::each_closure &closure)
{
	for(const auto &tag : link.queue)
	{
		assert(link.peer);
		if(!tag.request)
			continue;

		if(!closure(*link.peer, link, *tag.request))
			return false;
	}

	return true;
}

size_t
ircd::server::peer_unfinished()
{
	return accumulate_peers([]
	(const auto &peer)
	{
		return !peer.finished();
	});
}

size_t
ircd::server::peer_count()
{
	return peers.size();
}

size_t
ircd::server::link_count()
{
	return accumulate_peers([]
	(const auto &peer)
	{
		return peer.link_count();
	});
}

size_t
ircd::server::tag_count()
{
	return accumulate_peers([]
	(const auto &peer)
	{
		return peer.tag_count();
	});
}

template<class F>
size_t
ircd::server::accumulate_tags(F&& closure)
{
	return accumulate_links([&closure]
	(const auto &link)
	{
		return link.accumulate_tags(std::forward<F>(closure));
	});
}

template<class F>
size_t
ircd::server::accumulate_links(F&& closure)
{
	return accumulate_peers([&closure]
	(const auto &peer)
	{
		return peer.accumulate_links(std::forward<F>(closure));
	});
}

template<class F>
size_t
ircd::server::accumulate_peers(F&& closure)
{
	return std::accumulate(begin(peers), end(peers), size_t(0), [&closure]
	(auto ret, const auto &pair)
	{
		const auto &peer{*pair.second};
		return ret += closure(peer);
	});
}

ircd::string_view
ircd::server::canonize(const net::hostport &hostport)
{
	thread_local char buf[512];
	return net::canonize(buf, hostport);
}

///////////////////////////////////////////////////////////////////////////////
//
// server/request.h
//

decltype(ircd::server::request::log)
ircd::server::request::log
{
	"server.request"
};

decltype(ircd::server::request::opts_default)
ircd::server::request::opts_default
{};

/// Canceling a request is tricky. This allows a robust way to let the user's
/// request go out of scope at virtually any time without disrupting the
/// pipeline and other requests.
[[GCC::stack_protect]]
bool
ircd::server::cancel(request &request)
{
	if(!request.tag)
		return false;

	if(request.tag->canceled())
		return false;

	if(request.tag->abandoned())
		return false;

	auto &tag
	{
		*request.tag
	};

	log::debug
	{
		request::log, "%s cancel commit:%d w:%zu hr:%zu cr:%zu",
		loghead(request),
		tag.committed(),
		tag.state.written,
		tag.state.head_read,
		tag.state.content_read
	};

	tag.set_exception<canceled>("Request canceled");

	// We got off easy... The link's write loop won't start an abandoned
	// request. All that has to be done is indicate a full cancellation
	// immediately and the user will know nothing was revealed to the remote.
	if(!tag.committed())
	{
		disassociate(request, tag);
		return true;
	}

	// Now things aren't so easy. More complicated logic happens inside...
	cancel(request, tag);
	return true;
}

[[GCC::stack_protect]]
void
ircd::server::submit(const hostport &hostport,
                     request &request)
{
	assert(request.tag == nullptr);
	if(unlikely(ircd::run::level != ircd::run::level::RUN))
		throw unavailable
		{
			"Unable to fulfill requests at this time."
		};

	auto &peer
	{
		server::get(hostport)
	};

	peer.submit(request);
}

ircd::string_view
ircd::server::loghead(const link &link,
                      const request &request)
{
	thread_local char buf[512];
	return loghead(buf, link, request);
}

ircd::string_view
ircd::server::loghead(const mutable_buffer &buf,
                      const link &link,
                      const request &request)
{
	return fmt::sprintf
	{
		buf, "%s %s",
		loghead(link),
		loghead(request)
	};
}

ircd::string_view
ircd::server::loghead(const request &request)
{
	thread_local char buf[512];
	return loghead(buf, request);
}

ircd::string_view
ircd::server::loghead(const mutable_buffer &buf,
                      const request &request)
try
{
	if(empty(request.out.head))
		return "<no head>";

	if(request.tag && request.tag->canceled())
		return "<canceled; out data is gone>";

	const http::request::head head
	{
		request.out.gethead(request)
	};

	if(!head.method || !head.path)
		return "<no head data>";

	return fmt::sprintf
	{
		buf, "tag:%lu %s %s",
		id(request),
		head.method,
		head.path
	};
}
catch(const std::exception &e)
{
	log::critical
	{
		log, "server::loghead(): %s", e.what()
	};

	return "<critical error>";
}

uint64_t
ircd::server::id(const request &request)
noexcept
{
	return request.tag?
		request.tag->state.id:
		0UL;
}

//
// server::in
//

ircd::http::response::head
ircd::server::in::gethead(const request &request)
try
{
	if(empty(request.in.head))
		return {};

	if(request.tag && request.tag->canceled())
		return {};

	parse::buffer pb{request.in.head};
	parse::capstan pc{pb, [](char *&read, char *stop)
	{
		read = stop;
	}};

	pc.read += size(request.in.head);
	return http::response::head{pc};
}
catch(const std::exception &e)
{
	log::critical
	{
		log, "server::in::gethead(): %s", e.what()
	};

	return {};
}

//
// server::out
//

ircd::http::request::head
ircd::server::out::gethead(const request &request)
try
{
	if(empty(request.out.head))
		return {};

	if(request.tag && request.tag->canceled())
		return {};

	parse::buffer pb{request.out.head};
	parse::capstan pc{pb, [](char *&read, char *stop)
	{
		read = stop;
	}};

	pc.read += size(request.out.head);
	return http::request::head{pc};
}
catch(const std::exception &e)
{
	log::critical
	{
		log, "server::out::gethead(): %s", e.what()
	};

	return {};
}

///////////////////////////////////////////////////////////////////////////////
//
// server/peer.h
//

decltype(ircd::server::peers)
ircd::server::peers
{};

decltype(ircd::server::peer::enable_ipv6)
ircd::server::peer::enable_ipv6
{
	{ "name",     "ircd.server.peer.enable_ipv6" },
	{ "default",  true                           }
};

decltype(ircd::server::peer::link_min_default)
ircd::server::peer::link_min_default
{
	{ "name",     "ircd.server.peer.link_min" },
	{ "default",  1L                          }
};

decltype(ircd::server::peer::link_max_default)
ircd::server::peer::link_max_default
{
	{ "name",     "ircd.server.peer.link_max" },
	{ "default",  4L                          }
};

decltype(ircd::server::peer::ids)
ircd::server::peer::ids;

//
// peer::peer
//

ircd::server::peer::peer(const net::hostport &hostport,
                         const net::open_opts &open_opts)
:hostcanon
{
	net::canonize(hostport)
}
,open_opts
{
	open_opts
}
{
	const net::hostport &canon
	{
		this->hostcanon
	};

	// Ensure references are to this class's members
	this->open_opts.hostport.host = host(canon);
	this->open_opts.hostport.service = service(canon);
	this->open_opts.hostport.port = port(hostport);

	// Send SNI for this name.
	this->open_opts.server_name = host(canon);

	// Cert verify this name.
	this->open_opts.common_name = host(canon);

	if(rfc3986::valid(std::nothrow, rfc3986::parser::ip_address, host(canon)))
		this->remote =
		{
			host(canon), port(hostport)
		};

	this->open_opts.ipport = this->remote;
}

ircd::server::peer::~peer()
noexcept
{
	assert(links.empty());
}

void
ircd::server::peer::close(const net::close_opts &opts)
{
	//assert(!op_fini);
	if(op_fini)
		return;

	op_fini = true;
	link *links[LINK_MAX];
	const auto end(pointers(this->links, links));
	for(link **link(links); link != end; ++link)
		(*link)->close(opts);

	if(finished())
		return handle_finished();
}

void
ircd::server::peer::cancel()
{
	for(auto &link : this->links)
		link.cancel_all(make_exception_ptr<canceled>
		(
			"Request was aborted due to interruption."
		));
}

bool
ircd::server::peer::err_clear()
{
	const bool ret(e);
	e.reset(nullptr);

	// only clear the fini flag if we're in runlevel run
	const bool fini(ircd::run::level != ircd::run::level::RUN);
	op_fini = false | fini;

	return ret;
}

template<class... A>
void
ircd::server::peer::err_set(A&&... args)
{
	this->e = std::make_unique<err>(std::forward<A>(args)...);
}

ircd::string_view
ircd::server::peer::err_msg()
const
{
	return bool(e)? what(e->eptr) : string_view{};
}

bool
ircd::server::peer::err_has()
const
{
	return bool(e);
}

decltype(ircd::server::peer::error_clear_default)
ircd::server::peer::error_clear_default
{
	{ "name",     "ircd.server.peer.error.clear_default" },
	{ "default",  305L                                   }
};

bool
ircd::server::peer::err_check()
{
	if(op_fini)
		return false;

	if(!err_has())
		return true;

	//TODO: The specific error type should be switched and finer
	//TODO: timeouts should be used depending on the error: i.e
	//TODO: NXDOMAIN vs. temporary conn timeout, etc.
	if(e->etime + seconds(error_clear_default) > now<system_point>())
		return false;

	err_clear();
	return true;
}

void
ircd::server::peer::submit(request &request)
try
{
	if(!err_check() || unlikely(ircd::run::level != ircd::run::level::RUN))
		throw unavailable
		{
			"Peer %s is unable to take any requests :%s",
			hostcanon,
			err_msg()
		};

	// Run a GC over the links and tags for this peer first.
	cleanup_canceled();

	// Select the best link for this request or create anew.
	link *const ret
	{
		link_get(request)
	};

	if(likely(ret))
	{
		ret->submit(request);
		return;
	}

	if(!request.tag)
		throw unavailable
		{
			"No link to peer %s available", hostcanon
		};
	else
		request.tag->set_exception<unavailable>
		(
			"No link to peer %s available", hostcanon
		);
}
catch(const std::exception &e)
{
	if(!request.tag)
		throw;

	const auto eptr(std::current_exception());
	const ctx::exception_handler eh;
	request.tag->set_exception(eptr);
}

/// Dispatch algorithm here; finds the best link to place this request on,
/// or creates a new link entirely. There are a number of factors: foremost
/// if any special needs are indicated,
//
ircd::server::link *
ircd::server::peer::link_get(const request &request)
{
	assert(request.opt);
	const auto &prio(request.opt->priority);

	if(links.empty())
		return &link_add(1);

	// Indicates that we can't add anymore links for this peer and the rest
	// of the algorithm should consider this.
	const bool links_maxed
	{
		links.size() >= link_max()
	};

	link *best{nullptr};
	for(auto &cand : links)
	{
		// Don't want a link that's shutting down or marked for exclusion
		if(cand.op_fini || cand.exclude)
			continue;

		if(!best)
		{
			best = &cand;
			continue;
		}

		// Indicates that the best candidate has its pipe saturated which can
		// be factored into the comparisons here.
		const bool best_maxed
		{
			best->tag_committed() >= best->tag_commit_max()
		};

		const bool cand_maxed
		{
			cand.tag_committed() >= cand.tag_commit_max()
		};

		if(best_maxed && !cand_maxed)
		{
			best = &cand;
			continue;
		}

		if(!best_maxed && cand_maxed)
			continue;

		// Candidates's queue has less or same backlog of unsent requests, but
		// now measure if candidate will take longer to process at least the
		// write-side of those requests.
		if(cand.write_remaining() > best->write_remaining())
			continue;

		// Candidate might be working through really large content; though
		// this is a very sketchy measurement right now since we only *might*
		// know about content-length for the *one* active tag occupying the
		// socket.
		if(cand.read_remaining() > best->read_remaining())
			continue;

		// Coarse distribution based on who has more work; this is weak, should
		// be replaced.
		if(cand.tag_count() > best->tag_count())
			continue;

		best = &cand;
	}

	// Even though the prio is set to the super special value we allow the
	// normal loop to first come up with a best link which already is open
	// rather than unconditionally opening a new connection.
	if(prio == std::numeric_limits<std::remove_reference<decltype(prio)>::type>::min())
	{
		if(!best)
			return &link_add(1);

		if(best->tag_committed())
			return &link_add(1);

		return best;
	}

	// When we've reached the max number of links we return the best.
	if(links_maxed)
		return best;

	// If no best was found or was nulled, we have room for another link.
	if(!best)
	{
		best = &link_add();
		return best;
	}

	// If the best has room in its pipe we give it a shot.
	if(best->tag_committed() < best->tag_commit_max())
		return best;

	// Otherwise create a new link.
	best = &link_add();
	return best;
}

ircd::server::link &
ircd::server::peer::link_add(const size_t &num)
{
	assert(!finished());

	if(e)
	{
		std::rethrow_exception(e->eptr);
		__builtin_unreachable();
	}

	assert(!op_fini);
	links.emplace_back(*this);
	auto &link{links.back()};

	if(remote)
		link.open(open_opts);

	return link;
}

void
ircd::server::peer::handle_open(link &link,
                                std::exception_ptr eptr)
{
	if(eptr)
	{
		// Mark the peer as errored if the first link connection failed.
		assert(!links.empty());
		if(std::addressof(link) == std::addressof(links.front()))
			err_set(eptr);

		thread_local char rembuf[64];
		log::derror
		{
			log, "%s [%s]: open :%s",
			loghead(link),
			string(rembuf, remote),
			what(eptr)
		};

		if(op_fini)
		{
			if(link.finished())
				handle_finished(link);

			return;
		}

		link.close(net::dc::RST);
		return;
	}
}

void
ircd::server::peer::handle_close(link &link,
                                 std::exception_ptr eptr)
{
	thread_local char rembuf[64];

	if(eptr)
		log::derror
		{
			log, "%s [%s]: close :%s",
			loghead(link),
			string(rembuf, remote),
			what(eptr)
		};

	if(link.finished())
		handle_finished(link);
}

void
ircd::server::peer::handle_error(link &link,
                                 std::exception_ptr eptr)
{
	assert(bool(eptr));
	link.cancel_committed(eptr);
	log::derror
	{
		log, "%s :%s",
		loghead(link),
		what(eptr)
	};

	link.close(net::dc::RST);
}

void
ircd::server::peer::handle_error(link &link,
                                 const std::system_error &e)
{
	using std::errc;
	using boost::asio::error::get_misc_category;
	thread_local char rembuf[64];

	const auto &ec{e.code()};
	if(system_category(ec)) switch(ec.value())
	{
		case 0:
			assert(0);
			break;

		default:
			break;
	}
	else if(ec == net::eof)
	{
		log::debug
		{
			log, "%s [%s]: %s",
			loghead(link),
			string(rembuf, remote),
			e.what()
		};

		link.close(net::dc::RST);
		return;
	}

	log::derror
	{
		log, "%s [%s]: %s",
		loghead(link),
		string(rembuf, remote),
		e.what()
	};

	link.cancel_committed(std::make_exception_ptr(e));
	link.close(net::dc::RST);
}

void
ircd::server::peer::handle_finished(link &link)
{
	assert(link.finished());
	del(link);

	if(finished())
		handle_finished();
}

/// This is where we're notified a tag has been completed either to start the
/// next request when the link has too many requests in flight or perhaps to
/// reschedule the queues in various links to diffuse the pending requests.
/// This can't throw because the link still has to remove this tag from its
/// queue.
void
ircd::server::peer::handle_tag_done(link &link,
                                    tag &tag)
noexcept try
{
	log::debug
	{
		log, "%s [%s] => [%u] done wt:%zu rt:%zu %zu more in queue",
		loghead(link),
		tag.request?
			loghead(*tag.request):
			"<no request>"_sv,
		uint(tag.state.status),
		tag.write_size(),
		tag.read_size(),
		link.tag_count() - 1
	};

	if(tag.request)
	{
		assert(link.peer);
		++tag_done;
		log::logf
		{
			request::log, uint(tag.state.status) >= 300? log::DERROR: log::DEBUG,
			"%s [%u] %s wt:%zu rt:%zu hr:%zu cr:%zu cl:%zu chunks:%zu",
			loghead(*tag.request),
			uint(tag.state.status),
			link.peer->hostcanon,
			tag.write_size(),
			tag.read_size(),
			tag.state.head_read,
			tag.state.content_read,
			tag.state.content_length,
			tag.request->in.chunks.size(),
		};
	}

	if(link.tag_committed() >= link.tag_commit_max())
		link.wait_writable();
}
catch(const std::exception &e)
{
	log::critical
	{
		log, "%s tag:%lu done; %s",
		loghead(link),
		tag.state.id,
		e.what()
	};
}

/// This is where we're notified a link has processed its queue and has no
/// more work. We can choose whether to close the link or keep it open and
/// reinstate the read poll; reschedule other work to this link, etc.
void
ircd::server::peer::handle_link_done(link &link)
{
	assert(link.tag_count() == 0);

	if(link_ready() > link_min())
	{
		link.close();
		return;
	}

	link.wait_readable();
}

/// This is called when a tag on a link receives an HTTP response head.
/// We can use this to learn information from the tag's request and the
/// response head etc.
void
ircd::server::peer::handle_head_recv(const link &link,
                                     const tag &tag,
                                     const http::response::head &head)
{
	// Learn the software version of the remote peer so we can shape
	// requests more effectively.
	if(!server_version && head.server)
	{
		thread_local char rembuf[64];
		server_version = std::string{head.server};
		log::debug
		{
			log, "%s learned %s is '%s'",
			loghead(link),
			string(rembuf, remote),
			server_version,
		};
	}
}

void
ircd::server::peer::disperse(link &link)
{
	disperse_uncommitted(link);
	link.cancel_committed(make_exception_ptr<canceled>
	(
		"Request was aborted; though it was partially completed"
	));

	assert(link.queue.empty());
}

void
ircd::server::peer::disperse_uncommitted(link &link)
{
	auto &queue(link.queue);
	auto it(begin(queue));
	while(it != end(queue)) try
	{
		auto &tag{*it};
		if(!tag.request || tag.committed())
		{
			++it;
			continue;
		}

		submit(*tag.request);
		it = queue.erase(it);
	}
	catch(const std::exception &e)
	{
		const auto &tag{*it};
		log::warning
		{
			log, "%s failed to resubmit tag:%lu :%s",
			loghead(link),
			tag.state.id,
			e.what()
		};

		it = queue.erase(it);
	}
}

void
ircd::server::peer::cleanup_canceled()
{
	for(auto &link : links)
		link.cleanup_canceled();
}

/// This *cannot* be called unless a link's socket is closed and its queue
/// is empty. It is usually only called by a disconnect handler because
/// the proper way to remove a link is asynchronously through link.close();
void
ircd::server::peer::del(link &link)
{
	assert(!link.tag_count());
	assert(!link.opened());
	const auto it(std::find_if(begin(links), end(links), [&link]
	(const auto &link_)
	{
		return &link_ == &link;
	}));

	assert(it != end(links));
	thread_local char rembuf[64];
	log::debug
	{
		log, "%s removing from peer(%p) %zu of %zu to %s",
		loghead(link),
		this,
		std::distance(begin(links), it),
		links.size(),
		string(rembuf, remote)
	};

	links.erase(it);
}

void
ircd::server::peer::resolve(const hostport &hostport)
{
	net::dns::opts opts;

	// Figure out the initial query type. Most of the time it's SRV.
	opts.qtype = net::service(hostport) && !net::port(hostport)?
			33:  // SRV
	peer::enable_ipv6 && net::enable_ipv6?
			28:  // AAAA
			 1;  // A

	// When the result comes back as nxdomain this tells the resolver to
	// not set eptr; instead it gives an empty set of results. We do this
	// with SRV/AAAA queries for seamless fallback.
	opts.nxdomain_exceptions =
		opts.qtype != 33 &&     // SRV
		opts.qtype != 28;       // AAAA

	resolve(hostport, opts);
}

void
ircd::server::peer::resolve(const hostport &hostport,
                            const net::dns::opts &opts)
try
{
	assert(!op_resolve);
	if(op_fini)
		return;

	const unwind_exceptional failure{[this]
	{
		op_resolve = false;
		err_set(std::current_exception());
		if(unlikely(ircd::run::level != ircd::run::level::RUN))
			op_fini = true;
	}};

	// Skip DNS resolution for IP literals
	if(rfc3986::valid(std::nothrow, rfc3986::parser::ip_address, host(hostport)))
	{
		this->remote = {host(hostport), port(hostport)};
		open_opts.ipport = this->remote;
		open_links();
		return;
	}

	if(unlikely(opts.qtype != 33 && opts.qtype != 28 && opts.qtype != 1))
		throw error
		{
			"Unsupported DNS question type '%u' for resolve", opts.qtype
		};

	auto handler
	{
		opts.qtype == 33?
			net::dns::callback(std::bind(&peer::handle_resolve_SRV, this, ph::_1, ph::_2)):
		opts.qtype == 28?
			net::dns::callback(std::bind(&peer::handle_resolve_AAAA, this, ph::_1, ph::_2)):
			net::dns::callback(std::bind(&peer::handle_resolve_A, this, ph::_1, ph::_2))
	};

	op_resolve = true;
	assert(ctx::current); // sorry, ircd::ctx required for now.
	net::dns::resolve(hostport, opts, std::move(handler));
}
catch(const std::exception &e)
{
	thread_local char buf[256];
	log::error
	{
		log, "peer(%p) DNS resolution for '%s' :%s",
		this,
		string(buf, hostport),
		e.what()
	};
}

void
ircd::server::peer::handle_resolve_SRV(const hostport &hp,
                                       const json::array &rrs)
try
{
	assert(op_resolve);
	op_resolve = false;

	if(unlikely(ircd::run::level != ircd::run::level::RUN))
		op_fini = true;

	if(unlikely(finished()))
		return handle_finished();

	if(op_fini)
		return;

	const json::object &rr
	{
		net::dns::random_choice(rrs)
	};

	if(net::dns::is_error(rr))
	{
		const json::string &error(rr.get("error"));
		err_set(make_exception_ptr<rfc1035::error>("%s", error));
		assert(this->e && this->e->eptr);
		std::rethrow_exception(this->e->eptr);
		__builtin_unreachable();
	}

	// Target for the next address record query.
	const hostport &target
	{
		rr.has("tgt")?
			rstrip(unquote(rr.at("tgt")), '.'):
			host(hp),

		rr.has("port")?
			rr.get<uint16_t>("port"):
		port(remote)?
			port(remote):
			port(hp)
	};

	// Save the port from the SRV record to a class member because it won't
	// get carried through the next A/AAAA query.
	port(remote) = port(target);
	port(open_opts.hostport) = port(target);

	// Setup the address record query off this SRV response.
	net::dns::opts opts;
	opts.qtype = bool(peer::enable_ipv6) && bool(net::enable_ipv6)? 28: 1;

	log::debug
	{
		log, "peer(%p) resolved %s SRV rrs:%zu resolving %s %s",
		this,
		hostcanon,
		rrs.size(),
		host(target),
		rfc1035::rqtype.at(opts.qtype)
	};

	resolve(target, opts);
}
catch(const std::exception &e)
{
	log::derror
	{
		log, "peer(%p) '%s' resolve SRV :%s",
		this,
		this->hostcanon,
		e.what()
	};

	err_set(std::current_exception());
	const ctx::exception_handler eh;
	close();
}

void
ircd::server::peer::handle_resolve_AAAA(const hostport &target,
                                        const json::array &rrs)
try
{
	assert(op_resolve);
	op_resolve = false;

	if(unlikely(ircd::run::level != ircd::run::level::RUN))
		op_fini = true;

	if(unlikely(finished()))
		return handle_finished();

	if(op_fini)
		return;

	if(net::dns::is_empty(rrs) || net::dns::is_error(rrs))
	{
		// Setup the address record query off this SRV response.
		net::dns::opts opts;
		opts.qtype = 1;

		log::debug
		{
			log, "peer(%p) resolved %s AAAA rrs:%zu resolving %s %s",
			this,
			hostcanon,
			rrs.size(),
			host(target),
			rfc1035::rqtype.at(opts.qtype)
		};

		resolve(target, opts);
		return;
	}

	const json::object &rr
	{
		net::dns::random_choice(rrs)
	};

	assert(!net::dns::is_error(rr));
	this->remote = net::ipport
	{
		unquote(rr.at("ip")), port(this->remote)?: port(target)
	};

	open_opts.ipport = this->remote;
	open_links();
}
catch(const std::exception &e)
{
	log::derror
	{
		log, "peer(%p) resolve '%s' AAAA: %s",
		this,
		this->hostcanon,
		e.what()
	};

	err_set(std::current_exception());
	const ctx::exception_handler eh;
	close();
}

void
ircd::server::peer::handle_resolve_A(const hostport &target,
                                     const json::array &rrs)
try
{
	const ctx::critical_assertion ca;
	assert(op_resolve);
	op_resolve = false;

	if(unlikely(ircd::run::level != ircd::run::level::RUN))
		op_fini = true;

	if(unlikely(finished()))
		return handle_finished();

	if(op_fini)
		return;

	if(net::dns::is_empty(rrs))
	{
		err_set(make_exception_ptr<unavailable>("Host has no address record."));
		assert(this->e && this->e->eptr);
		std::rethrow_exception(this->e->eptr);
		__builtin_unreachable();
	}

	const json::object &rr
	{
		net::dns::random_choice(rrs)
	};

	if(net::dns::is_error(rr))
	{
		const json::string &error(rr.get("error"));
		err_set(make_exception_ptr<rfc1035::error>("%s", error));
		assert(this->e && this->e->eptr);
		std::rethrow_exception(this->e->eptr);
		__builtin_unreachable();
	}

	// Save the results of the query to this object instance.
	this->remote = net::ipport
	{
		unquote(rr.at("ip")), port(this->remote)?: port(target)
	};

	open_opts.ipport = this->remote;
	open_links();
}
catch(const std::exception &e)
{
	log::derror
	{
		log, "peer(%p) '%s' resolve A :%s",
		this,
		this->hostcanon,
		e.what()
	};

	err_set(std::current_exception());
	const ctx::exception_handler eh;
	close();
}

void
ircd::server::peer::open_links()
try
{
	// The hostname in open_opts should still reference this object's string.
	assert(host(open_opts.hostport).data() == this->hostcanon.data());

	link *links[LINK_MAX];
	const auto end(pointers(this->links, links));
	for(link **link(links); link != end; ++link)
		(*link)->open(open_opts);
}
catch(const std::exception &e)
{
	log::derror
	{
		log, "peer(%p) open links :%s",
		this,
		e.what()
	};

	err_set(std::current_exception());
	const ctx::exception_handler eh;
	close();
}

void
ircd::server::peer::handle_finished()
{
	assert(finished());

	// Right now this is what the server:: ~init sequence needs
	// to wait for all links to close on IRCd shutdown.
	server::dock.notify_all();
}

size_t
ircd::server::peer::read_total()
const
{
	return read_bytes;
}

size_t
ircd::server::peer::write_total()
const
{
	return write_bytes;
}

size_t
ircd::server::peer::read_remaining()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.read_remaining();
	});
}

size_t
ircd::server::peer::read_completed()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.read_completed();
	});
}

size_t
ircd::server::peer::read_size()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.read_size();
	});
}

size_t
ircd::server::peer::write_remaining()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.write_remaining();
	});
}

size_t
ircd::server::peer::write_completed()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.write_completed();
	});
}

size_t
ircd::server::peer::write_size()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.write_size();
	});
}

size_t
ircd::server::peer::tag_uncommitted()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.tag_uncommitted();
	});
}

size_t
ircd::server::peer::tag_committed()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.tag_committed();
	});
}

size_t
ircd::server::peer::tag_count()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.tag_count();
	});
}

size_t
ircd::server::peer::link_tag_done()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.tag_done;
	});
}

size_t
ircd::server::peer::link_ready()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.ready();
	});
}

size_t
ircd::server::peer::link_busy()
const
{
	return accumulate_links([](const auto &link)
	{
		return link.busy();
	});
}

size_t
ircd::server::peer::link_count()
const
{
	return links.size();
}

size_t
ircd::server::peer::link_min()
const
{
	return link_min_default;
}

size_t
ircd::server::peer::link_max()
const
{
	return link_max_default;
}

bool
ircd::server::peer::finished()
const
{
	return links.empty() && !op_resolve && op_fini;
}

template<class F>
size_t
ircd::server::peer::accumulate_tags(F&& closure)
const
{
	return accumulate_links([&closure](const auto &link)
	{
		return link.accumulate([&closure](const auto &tag)
		{
			return closure(tag);
		});
	});
}

template<class F>
size_t
ircd::server::peer::accumulate_links(F&& closure)
const
{
	return std::accumulate(begin(links), end(links), size_t(0), [&closure]
	(auto ret, const auto &tag)
	{
		return ret += closure(tag);
	});
}

//
// peer::err
//

ircd::server::peer::err::err(const std::exception_ptr &eptr)
:eptr{eptr}
,etime{now<system_point>()}
{
}

ircd::server::peer::err::~err()
noexcept
{
}

///////////////////////////////////////////////////////////////////////////////
//
// server/link.h
//

decltype(ircd::server::link::tag_max_default)
ircd::server::link::tag_max_default
{
	{ "name",     "ircd.server.link.tag_max" },
	{ "default",  16384L                     },
};

decltype(ircd::server::link::tag_commit_max_default)
ircd::server::link::tag_commit_max_default
{
	{ "name",     "ircd.server.link.tag_commit_max" },
	{ "default",  3L                                }
};

decltype(ircd::server::link::ids)
ircd::server::link::ids;

ircd::string_view
ircd::server::loghead(const link &link)
{
	thread_local char buf[512];
	return loghead(buf, link);
}

ircd::string_view
ircd::server::loghead(const mutable_buffer &buf,
                      const link &link)
{
	thread_local char rembuf[2][64];

	const auto local
	{
		link.socket?
			string(rembuf[0], local_ipport(*link.socket)):
			string_view{}
	};

	const auto remote
	{
		link.socket?
			string(rembuf[1], remote_ipport(*link.socket)):
			string_view{}
	};

	return fmt::sprintf
	{
		buf, "socket:%lu local:%s remote:%s link:%lu peer:%lu",
		link.socket?
			link.socket->id : 0UL,
		local?
			local : "0.0.0.0"_sv,
		remote?
			remote : "0.0.0.0"_sv,
		link.id,
		link.peer?
			link.peer->id: 0UL,
	};
}

//
// link::link
//

ircd::server::link::link(server::peer &peer)
:peer{&peer}
{
}

ircd::server::link::~link()
noexcept
{
	assert(!busy());
	assert(!opened());
}

void
ircd::server::link::submit(request &request)
{
	assert(!request.tag || !request.tag->committed());

	const auto it
	{
		request.tag? queue.emplace(end(queue), std::move(*request.tag)):
		             queue.emplace(end(queue), request)
	};
/*
	log::debug
	{
		log, "tag(%p) submitted to link(%p) queue: %zu",
		&(*it),
		this,
		tag_count()
	};
*/
	if(ready())
		wait_writable();
}

void
ircd::server::link::cancel_all(std::exception_ptr eptr)
{
	for(auto it(begin(queue)); it != end(queue); it = queue.erase(it))
	{
		auto &tag{*it};
		if(!tag.request)
			continue;

		tag.set_exception(eptr);
	}
}

void
ircd::server::link::cancel_committed(std::exception_ptr eptr)
{
	for(auto it(begin(queue)); it != end(queue); it = queue.erase(it))
	{
		auto &tag{*it};
		if(!tag.request)
			continue;

		if(!tag.committed())
			break;

		tag.set_exception(eptr);
	}
}

void
ircd::server::link::cancel_uncommitted(std::exception_ptr eptr)
{
	auto it(begin(queue));
	while(it != end(queue))
	{
		auto &tag{*it};
		if(!tag.request || tag.committed())
		{
			++it;
			continue;
		}

		tag.set_exception(eptr);
		it = queue.erase(it);
	}
}

void
ircd::server::link::cleanup_canceled()
{
	size_t dead(0);
	for(auto it(begin(queue)); it != end(queue); )
	{
		const auto &tag{*it};
		if(tag.committed() || tag.request)
		{
			dead += tag.committed() && tag.canceled();
			++it;
			continue;
		}

		#if 0
		log::dwarning
		{
			log, "%s removing abandoned tag:%lu",
			loghead(*this),
			tag.state.id,
		};
		#endif

		it = queue.erase(it);
	}

	// If every committed tag in the pipe is canceled we can close this link
	// to quickly disperse any queued tags to another link or simply kill this
	// link if it's timing out.
	assert(dead <= tag_committed());
	if(dead && dead == tag_committed())
		if(close())
			log::dwarning
			{
				log, "%s closing link since all %zu committed tags are dead in the pipe",
				loghead(*this),
				dead,
			};
}

bool
ircd::server::link::open(const net::open_opts &open_opts)
{
	assert(ircd::run::level == ircd::run::level::RUN);

	if(op_init)
		return false;

	auto handler
	{
		std::bind(&link::handle_open, this, ph::_1)
	};

	op_init = true;
	op_open = true;
	const unwind_exceptional unhandled{[this]
	{
		op_init = false;
		op_open = false;
	}};

	socket = net::open(open_opts, std::move(handler));
	op_open = false;

	if(finished())
	{
		assert(peer);
		peer->handle_finished(*this);
		return false;
	}

	return true;
}

void
ircd::server::link::handle_open(std::exception_ptr eptr)
{
	assert(op_init);
	op_init = false;
	synack_ts = time<seconds>();

	if(!eptr && !op_fini)
		wait_writable();

	if(peer)
		peer->handle_open(*this, std::move(eptr));
}

bool
ircd::server::link::close(const net::close_opts &close_opts)
{
	if(op_fini)
		return false;

	op_fini = true;

	// Tell the peer to ditch everything in the queue; op_fini has been set so
	// the tags won't get assigned back to this link.
	if(tag_count() && peer)
		peer->disperse(*this);

	auto handler
	{
		std::bind(&link::handle_close, this, ph::_1)
	};

	if(!socket)
	{
		handler(std::exception_ptr{});
		return true;
	}

	net::close(*socket, close_opts, std::move(handler));
	return true;
}

void
ircd::server::link::handle_close(std::exception_ptr eptr)
{
	assert(op_fini);

	if(op_init)
	{
		assert(bool(eptr));
	}

	if(peer)
		peer->handle_close(*this, std::move(eptr));
}

void
ircd::server::link::wait_writable()
{
	if(op_write || unlikely(op_fini))
		return;

	auto handler
	{
		std::bind(&link::handle_writable, this, ph::_1)
	};

	assert(ready());
	op_write = true;
	const unwind_exceptional unhandled{[this]
	{
		op_write = false;
	}};

	net::wait(*socket, net::ready::WRITE, std::move(handler));
}

[[GCC::stack_protect]]
void
ircd::server::link::handle_writable(const error_code &ec)
noexcept try
{
	using std::errc;

	op_write = false;
	write_ts = time<seconds>();

	if(unlikely(finished()))
	{
		assert(peer);
		return peer->handle_finished(*this);
	}

	if(system_category(ec)) switch(ec.value())
	{
		case 0:
			handle_writable_success();
			return;

		case int(errc::operation_canceled):
			return;

		default:
			break;
	}

	throw std::system_error{ec};
}
catch(const std::system_error &e)
{
	assert(peer);
	peer->handle_error(*this, e);
}
catch(...)
{
	assert(peer);
	peer->handle_error(*this, std::current_exception());
}

void
ircd::server::link::handle_writable_success()
{
	assert(socket);
	auto it(begin(queue));
	while(it != end(queue))
	{
		auto &tag{*it};
		if((tag.abandoned() || tag.canceled()) && !tag.committed())
		{
			log::debug
			{
				log, "%s discarding canceled:%d abandoned:%d uncommitted tag %zu of %zu",
				loghead(*this),
				tag.canceled(),
				tag.abandoned(),
				tag_committed(),
				tag_count()
			};

			it = queue.erase(it);
			continue;
		}

		if(unlikely(tag.canceled() && tag.committed() && tag_committed() <= 1))
		{
			log::debug
			{
				log, "%s closing to interrupt canceled committed tag:%lu of %zu",
				loghead(*this),
				tag.state.id,
				tag_count()
			};

			close();
			break;
		}

		if(tag_committed() == 0)
			wait_readable();

		if(!process_write(tag))
		{
			wait_writable();
			break;
		}

		// Limits the amount of requests in the pipe.
		if(tag_committed() >= tag_commit_max())
			break;

		++it;
	}
}

bool
ircd::server::link::process_write(tag &tag)
{
	if(!tag.committed())
	{
		log::debug
		{
			log, "%s starting on tag:%lu %zu of %zu: wt:%zu [%s]",
			loghead(*this),
			tag.state.id,
			tag_committed(),
			tag_count(),
			tag.write_size(),
			tag.request?
				loghead(*tag.request):
				"<no attached request>"_sv
		};

		if(tag.request)
			log::debug
			{
				request::log, "%s wt:%zu on %s",
				loghead(*tag.request),
				tag.write_size(),
				loghead(*this),
			};
	}

	while(tag.write_remaining())
	{
		const const_buffer buffer
		{
			tag.make_write_buffer()
		};

		assert(!empty(buffer));
		const const_buffer written
		{
			process_write_next(buffer)
		};

		tag.wrote_buffer(written);
		assert(tag_committed() <= tag_commit_max());
		if(size(written) < size(buffer))
			return false;
	}

	return true;
}

ircd::const_buffer
ircd::server::link::process_write_next(const const_buffer &buffer)
{
	const size_t bytes
	{
		write_any(*socket, buffer)
	};

	assert(bytes <= size(buffer));
	const const_buffer written
	{
		buffer, bytes
	};

	assert(peer);
	peer->write_bytes += bytes;
	return written;
}

void
ircd::server::link::wait_readable()
{
	if(op_read || op_fini)
		return;

	assert(ready());
	op_read = true;
	const unwind_exceptional unhandled{[this]
	{
		op_read = false;
	}};

	auto handler
	{
		std::bind(&link::handle_readable, this, ph::_1)
	};

	net::wait(*socket, net::ready::READ, std::move(handler));
}

[[GCC::stack_protect]]
void
ircd::server::link::handle_readable(const error_code &ec)
noexcept try
{
	using std::errc;

	op_read = false;
	read_ts = time<seconds>();

	if(unlikely(finished()))
	{
		assert(peer);
		return peer->handle_finished(*this);
	}

	if(system_category(ec)) switch(ec.value())
	{
		case 0:
			handle_readable_success();
			return;

		case int(errc::operation_canceled):
			return;

		default:
			break;
	}

	throw std::system_error{ec};
}
catch(const std::system_error &e)
{
	assert(peer);
	peer->handle_error(*this, e);
}
catch(...)
{
	assert(peer);
	peer->handle_error(*this, std::current_exception());
}

/// Process as many read operations from as many tags as possible
void
ircd::server::link::handle_readable_success()
{
	assert(socket);
	if(!tag_committed())
	{
		discard_read();
		wait_readable();
		return;
	}

	// Data pointed to by overrun will remain intact between iterations
	// because this loop isn't executing in any ircd::ctx. Since the buffers
	// we're using are supplied by users in other ctxs they will continue to
	// exist as this loop continues to the next tags. There is one exception
	// case though: canceled requests have their buffers free'ed when the tag
	// is pop'ed from this link's queue, because the user is gone; the scratch
	// buffer is maintained between iterations in that case.
	unique_buffer<mutable_buffer> scratch;
	const_buffer overrun; do
	{
		if(!process_read(overrun, scratch))
		{
			wait_readable();
			return;
		}
	}
	while(!queue.empty());

	assert(peer);
	peer->handle_link_done(*this);
}

/// Process as many read operations for one tag as possible
bool
ircd::server::link::process_read(const_buffer &overrun,
                                 unique_buffer<mutable_buffer> &scratch)
try
{
	assert(peer);
	assert(!queue.empty());

	auto &tag
	{
		queue.front()
	};

	if(unlikely(!tag.committed()))
	{
		discard_read(); // Should stumble on a socket error.
		assert(empty(overrun)); // Tag hasn't sent its data yet, we shouldn't
		return false;
	}

	if(unlikely(tag.canceled() && tag_committed() <= 1))
	{
		log::debug
		{
			log, "%s closing to interrupt canceled committed tag:%lu of %zu",
			loghead(*this),
			tag.state.id,
			tag_count()
		};

		close();
		return false;
	}

	bool done{false}; do
	{
		overrun = process_read_next(overrun, tag, done);
	}
	while(!done && !empty(overrun));

	if(likely(!done))
	{
		// This branch represents a read of -EAGAIN.
		assert(empty(overrun));
		return done;
	}

	// Branch to handle overrun out of a cancelled tag which needs its data
	// copied to scratch before being pop'ed off the queue; fairly rare case.
	// If the tag is not in a canceled state, the overrun will point to valid
	// data for the next tag even after being popped off the queue.
	if(!empty(overrun) && tag.canceled())
	{
		// Copy into new buffer before trashing the old buffer in case each
		// tag being processed here is just windowing down on the same data
		// nagled together at the first tag.
		unique_buffer<mutable_buffer> _scratch(overrun);
		scratch = std::move(_scratch);
		overrun = scratch;
		assert(!empty(overrun));
		assert(!empty(scratch));
	}

	peer->handle_tag_done(*this, tag);
	assert(!queue.empty());
	queue.pop_front();
	++tag_done;
	return done;
}
catch(const buffer_overrun &e)
{
	assert(!queue.empty());
	queue.pop_front();
	throw;
}

/// Process one read operation for one tag
ircd::const_buffer
ircd::server::link::process_read_next(const const_buffer &underrun,
                                      tag &tag,
                                      bool &done)
try
{
	const mutable_buffer buffer
	{
		tag.make_read_buffer()
	};

	const size_t copied
	{
		copy(buffer, underrun)
	};

	const mutable_buffer remaining
	{
		buffer + copied
	};

	if(unlikely(empty(remaining)))
		throw buffer_overrun
		{
			"Buffer of %zu bytes is insufficient to receive the HTTP request.",
			size(buffer),
		};

	const const_buffer view
	{
		read(remaining)
	};

	const const_buffer overrun
	{
		tag.read_buffer(view, done, *this)
	};

	assert(done || empty(overrun));
	return overrun;
}
catch(const buffer_overrun &)
{
	tag.set_exception(std::current_exception());
	throw;
}

/// Read directly off the link's socket into buf
ircd::const_buffer
ircd::server::link::read(const mutable_buffer &buf)
{
	assert(socket);
	assert(!empty(buf));
	const size_t received
	{
		read_one(*socket, buf)
	};

	assert(peer);
	peer->read_bytes += received;

	assert(received <= size(buf));
	return const_buffer
	{
		buf, received
	};
}

void
ircd::server::link::discard_read()
{
	assert(socket);
	const size_t available
	{
		net::available(*socket)
	};

	const ssize_t has_pending
	{
		#if OPENSSL_VERSION_NUMBER >= 0x10100000L
			SSL_has_pending(socket->ssl.native_handle())
		#else
			-2L
		#endif
	};

	const ssize_t pending
	{
		SSL_pending(socket->ssl.native_handle())
	};

	const size_t discarded
	{
		discard_any(*socket, size_t(pending))
	};

	if(discarded)
	{
		log::dwarning
		{
			log, "%s q:%zu discarded:%zu pending:%zd has_pending:%zd available:%zd",
			loghead(*this),
			queue.size(),
			discarded,
			pending,
			has_pending,
			available,
		};

		assert(peer);
		peer->read_bytes += discarded;
		return;
	}

	throw std::system_error
	{
		net::eof
	};
}

size_t
ircd::server::link::tag_uncommitted()
const
{
	return tag_count() - tag_committed();
}

size_t
ircd::server::link::tag_committed()
const
{
	return accumulate_tags([](const auto &tag)
	{
		return tag.committed();
	});
}

size_t
ircd::server::link::tag_count()
const
{
	return queue.size();
}

size_t
ircd::server::link::read_total()
const
{
	return socket? socket->in.bytes : 0;
}

size_t
ircd::server::link::write_total()
const
{
	return socket? socket->out.bytes : 0;
}

size_t
ircd::server::link::read_remaining()
const
{
	return accumulate_tags([](const auto &tag)
	{
		return tag.read_remaining();
	});
}

size_t
ircd::server::link::read_completed()
const
{
	return accumulate_tags([](const auto &tag)
	{
		return tag.read_completed();
	});
}

size_t
ircd::server::link::read_size()
const
{
	return accumulate_tags([](const auto &tag)
	{
		return tag.read_size();
	});
}

size_t
ircd::server::link::write_remaining()
const
{
	return accumulate_tags([](const auto &tag)
	{
		return tag.write_remaining();
	});
}

size_t
ircd::server::link::write_completed()
const
{
	return accumulate_tags([](const auto &tag)
	{
		return tag.write_completed();
	});
}

size_t
ircd::server::link::write_size()
const
{
	return accumulate_tags([](const auto &tag)
	{
		return tag.write_size();
	});
}

bool
ircd::server::link::busy()
const
{
	return !queue.empty();
}

bool
ircd::server::link::ready()
const
{
	return opened() && !op_init && !op_fini;
}

bool
ircd::server::link::opened()
const noexcept
{
	return bool(socket) && net::opened(*socket);
}

bool
ircd::server::link::finished()
const
{
	if(!bool(socket))
		return true;

	return !opened() && op_fini && !op_init && !op_open && !op_write && !op_read;
}

size_t
ircd::server::link::tag_commit_max()
const
{
	return tag_commit_max_default;
}

size_t
ircd::server::link::tag_max()
const
{
	return tag_max_default;
}

template<class F>
size_t
ircd::server::link::accumulate_tags(F&& closure)
const
{
	return std::accumulate(begin(queue), end(queue), size_t(0), [&closure]
	(auto ret, const auto &tag)
	{
		return ret += closure(tag);
	});
}

///////////////////////////////////////////////////////////////////////////////
//
// server/tag.h
//

/// Monotonic counter for tags.
decltype(ircd::server::tag::state::ids)
ircd::server::tag::state::ids;

/// This is tricky. When a user cancels a request which has committed some
/// writes to the remote we have to continue to service it through to
/// completion without disrupting the linearity of the link's pipeline
/// and causing trouble with other requests. This all depends on what phase
/// the request is currently in.
///
/// In any case, the goal here is to swap out the user's request buffers
/// and replace them with cancellation buffers which will be transparent
/// to the link as it completes the request.
void
ircd::server::cancel(request &request,
                     tag &tag)
noexcept
{
	// Must have a fully associated request/tag which has committed some
	// data to the wire to enter this routine.
	assert(tag.committed());
	assert(!tag.canceled());
	assert(request.tag == &tag);
	assert(tag.request == &request);
	assert(!tag.cancellation);

	// The cancellation is a straightforward facsimile except in the case of
	// dynamic chunked encoding mode where we need to add additional scratch.
	assert(tag.state.head_read <= size(request.in.head));
	const size_t additional_scratch
	{
		tag.state.chunk_length != 0 && null(request.in.content)?
			tag.state.head_rem:
			0_KiB
	};

	assert(additional_scratch <= 64_KiB); // sanity
	const size_t cancellation_size
	{
		size(request.out) + size(request.in) + additional_scratch
	};

	// Disassociate the user's request and add our dummy request in its place.
	disassociate(request, tag);

	assert(tag.request == nullptr);
	tag.request = new server::request{};
	tag.request->tag = &tag;

	// Setup the cancellation buffers by mirroring the current state of the
	// user's buffers.

	assert(!tag.cancellation);
	assert(cancellation_size < 64_MiB); // sanity
	tag.cancellation = unique_buffer<mutable_buffer>
	{
		cancellation_size
	};

	char *ptr
	{
		data(tag.cancellation)
	};

	const mutable_buffer out_head{ptr, size(request.out.head)};
	tag.request->out.head = out_head;
	ptr += size(out_head);

	const mutable_buffer out_content{ptr, size(request.out.content)};
	tag.request->out.content = out_content;
	ptr += size(out_content);

	// The in_head buffer is a straightforward facsimile except in the case of
	// dynamic chunked encoding mode where we need to add additional scratch.
	const mutable_buffer in_head{ptr, size(request.in.head)};
	tag.request->in.head = in_head;
	ptr += size(in_head);

	const mutable_buffer in_scratch{ptr, additional_scratch};
	ptr += size(in_scratch);

	const mutable_buffer in_content{ptr, size(request.in.content)};
	// The nullity (btw that's a real word) of in.content has to be preserved
	// between the user's tag and the cancellation tag. This is important for
	// a dynamic chunked encoded response which has null in.content until done.
	if(!null(request.in.content))
	{
		tag.request->in.content = in_content;
		ptr += size(in_content);
	}
	else tag.request->in.content = request.in.content;

	assert(size_t(std::distance(data(tag.cancellation), ptr)) == cancellation_size);

	// If the head is not completely written we have to copy the remainder from where
	// the socket left off.
	if(tag.state.written < size(request.out.head))
	{
		const const_buffer src
		{
			request.out.head + tag.state.written
		};

		const mutable_buffer dst
		{
			out_head + tag.state.written
		};

		copy(dst, src);
	}

	// If the content is not completely written we have to copy the remainder from where
	// the socket left off.
	const size_t content_written
	{
		tag.state.written > size(request.out.head)? tag.state.written - size(request.out.head) : 0
	};

	if(content_written < size(request.out.content))
	{
		const const_buffer src
		{
			request.out.content + content_written
		};

		const mutable_buffer dst
		{
			out_content + content_written
		};

		copy(dst, src);
	}

	// If the head is not completely read we have to copy what's been received so far so
	// we can parse a coherent head.
	if(tag.state.head_read > 0 && tag.state.head_read < size(request.in.head))
	{
		const const_buffer src
		{
			request.in.head, tag.state.head_read
		};

		const mutable_buffer dst
		{
			in_head
		};

		copy(dst, src);
	}

	// If the chunk head (in dynamic mode) is not complete at this point we
	// copy that portion. This is where the scratch area after the real head
	// is used.
	if(tag.state.chunk_length != 0 && null(request.in.content))
	{
		assert(tag.state.content_read >= tag.state.chunk_read);
		const const_buffer src
		{
			data(request.in.head) + tag.state.head_read,
			tag.state.chunk_read
		};

		const mutable_buffer dst
		{
			in_scratch
		};

		copy(dst, src);
	}

	// Normally we have no reason to copy content, but there is one exception:
	// If the content is chunked encoding and the tag is in the phase of
	// receiving the chunk head we have to copy what's been received of that
	// head so far so the grammar can parse a coherent head to continue.
	if(tag.state.chunk_length == size_t(-1) && !null(request.in.content))
	{
		assert(tag.state.content_read >= tag.state.content_length);
		const const_buffer src
		{
			request.in.content + tag.state.content_length,
			tag.state.content_read - tag.state.content_length
		};

		const mutable_buffer dst
		{
			in_content + tag.state.content_length
		};

		copy(dst, src);
	}

	// Moving the dynamic buffer should have no real effect because the
	// cancellation buffer already took over for it. We could do it anyway
	// to prevent regressions but at the cost of maintaining twice the memory
	// allocated. For now it's commented to let it die with the user's req.
	//tag.request->in.dynamic = std::move(request.in.dynamic);

	// Moving the chunk vector is important to maintain the state of dynamic
	// chunk transfers through this cancel. There is no condition here for if
	// this is not a dynamic chunk transfer because it's trivial.
	tag.request->in.chunks = std::move(request.in.chunks);
}

void
ircd::server::associate(request &request,
                        tag &tag)
noexcept
{
	assert(request.tag == nullptr);
	assert(tag.request == nullptr);

	auto &future
	{
		static_cast<ctx::future<http::code> &>(request)
	};

	future = tag.p;
	request.tag = &tag;
	tag.request = &request;
}

void
ircd::server::associate(request &request,
                        tag &cur,
                        tag &&old)
noexcept
{
	assert(request.tag == &old);         // ctor moved
	assert(cur.request == &request);     // ctor moved
	assert(old.request == &request);     // ctor didn't trash old

	cur.request = &request;
	old.request = nullptr;
	request.tag = &cur;
}

void
ircd::server::associate(request &cur,
                        tag &tag,
                        request &&old)
noexcept
{
	assert(tag.request == &old);   // ctor already moved
	assert(cur.tag == &tag);       // ctor already moved
	assert(old.tag == &tag);       // ctor didn't trash old

	cur.tag = &tag;
	tag.request = &cur;
	old.tag = nullptr;
}

void
ircd::server::disassociate(request &request,
                           tag &tag)
noexcept
{
	assert(request.tag == &tag);
	assert(tag.request == &request);
	assert(tag.abandoned());

	request.tag = nullptr;
	tag.request = nullptr;

	// If the original request was canceled a new request was attached in its
	// place in addition to an cancellation buffer. The existence of this
	// cancellation buffer indicates that we must delete the request here.
	// This is a little hacky but it gets the job done.
	if(!!tag.cancellation)
		delete &request;
}

void
ircd::server::tag::wrote_buffer(const const_buffer &buffer)
{
	assert(request);
	const auto &req{*request};
	state.written += size(buffer);

	if(state.written <= size(req.out.head))
	{
		assert(data(buffer) >= begin(req.out.head));
		assert(data(buffer) < end(req.out.head));
	}
	else if(state.written <= size(req.out.head) + size(req.out.content))
	{
		assert(data(buffer) >= begin(req.out.content));
		assert(data(buffer) < end(req.out.content));
		assert(state.written <= write_size());

		// Invoke the user's optional progress callback; this function
		// should be marked noexcept and has no reason to throw yet.
		if(req.out.progress)
			req.out.progress(buffer, const_buffer{req.out.content, state.written});
	}
	else
	{
		assert(0);
	}
}

ircd::const_buffer
ircd::server::tag::make_write_buffer()
const
{
	assert(request);
	const auto &req{*request};

	return
		state.written < size(req.out.head)?
			make_write_head_buffer():

		state.written < size(req.out.head) + size(req.out.content)?
			make_write_content_buffer():

		const_buffer{};
}

ircd::const_buffer
ircd::server::tag::make_write_head_buffer()
const
{
	assert(request);
	const auto &req{*request};

	const const_buffer window
	{
		req.out.head + state.written
	};

	return window;
}

ircd::const_buffer
ircd::server::tag::make_write_content_buffer()
const
{
	assert(request);
	const auto &req{*request};

	assert(state.written >= size(req.out.head));
	const size_t content_offset
	{
		state.written - size(req.out.head)
	};

	const size_t remain
	{
		size(req.out.head) + size(req.out.content) - state.written
	};

	const const_buffer window
	{
		req.out.content + content_offset, remain
	};

	return window;
}

/// Called by the controller of the socket with a view of the data received by
/// the socket. The location and size of `buffer` is the same or smaller than
/// the buffer previously supplied by make_read_buffer().
///
/// Sometimes make_read_buffer() supplies a buffer that is too large, and some
/// data read off the socket does not belong to this tag. In that case, This
/// function returns a const_buffer viewing the portion of `buffer` which is
/// considered the "overrun," and the socket controller will copy that over to
/// the next tag.
///
/// The tag indicates it is entirely finished with receiving its data by
/// setting the value of `done` to true. Otherwise it is assumed false.
///
/// The link argument is not to be used to control/modify the link from the
/// tag; it's only a backreference to flash information to the link/peer
/// through specific callbacks so the peer can learn information.
///
[[GCC::stack_protect]]
ircd::const_buffer
ircd::server::tag::read_buffer(const const_buffer &buffer,
                               bool &done,
                               link &link)
{
	assert(!done);
	assert(request);
	const bool chunk_header_mode
	{
		state.chunk_length == size_t(-1)
	};

	const const_buffer ret
	{
		empty(buffer)?
			buffer:

		state.status == (http::code)0?
			read_head(buffer, done, link):

		chunk_header_mode && null(request->in.content)?
			read_chunk_dynamic_head(buffer, done):

		chunk_header_mode?
			read_chunk_head(buffer, done):

		state.chunk_length && null(request->in.content)?
			read_chunk_dynamic_content(buffer, done):

		state.chunk_length?
			read_chunk_content(buffer, done):

		read_content(buffer, done)
	};

	return ret;
}

namespace ircd::server
{
	static void content_completed(tag &, bool &done);
}

ircd::const_buffer
ircd::server::tag::read_head(const const_buffer &buffer,
                             bool &done,
                             link &link)
{
	assert(request);
	auto &req{*request};

	// Total useful bytes in head buffer from prior packets and the latest
	// packet; this may extend past this head.
	assert(overlap(req.in.head, buffer));
	const const_buffer candidate_head
	{
		req.in.head, state.head_read + size(buffer)
	};

	// informal search for head  terminator
	assert(size(candidate_head) <= size(req.in.head));
	const auto pos
	{
		string_view{candidate_head}.find(http::headers::terminator)
	};

	// No terminator found; account for what was received in this buffer
	// for the next call to make_head_buffer() preparing for the subsequent
	// invocation of this function with more data.
	if(pos == string_view::npos)
	{
		state.head_read += size(buffer);
		return {};
	}

	// This indicates how much head was just received from this buffer only,
	// including the terminator which is considered part of the dome.
	assert(pos + size(http::headers::terminator) >= state.head_read);
	const size_t addl_head_bytes
	{
		pos + size(http::headers::terminator) - state.head_read
	};

	// The received buffer may go past the end of the head.
	assert(size(buffer) >= addl_head_bytes);
	const size_t beyond_head_len
	{
		size(buffer) - addl_head_bytes
	};

	// The final update for the confirmed length of the head.
	state.head_read += addl_head_bytes;
	assert(state.head_read + beyond_head_len <= size(req.in.head));
	assert(state.head_read <= size(candidate_head));

	// Window on any data in the buffer after the head.
	const const_buffer beyond_head
	{
		req.in.head + state.head_read, beyond_head_len
	};

	// Before changing the user's head buffer, we branch for a feature that
	// allows the user to receive head and content into a single contiguous
	// buffer by assigning in.content = in.head.
	const bool contiguous
	{
		data(req.in.content) == data(req.in.head)
	};

	// Alternatively branch for a feature that allows dynamic allocation of
	// the content buffer if the user did not specify any buffer.
	const bool dynamic
	{
		!contiguous && empty(req.in.content)
	};

	// Resize the user's head buffer tight to the head; this is how we convey
	// the size of the dome back to the user.
	assert(state.head_read <= size(req.in.head));
	state.head_rem = size(req.in.head) - state.head_read;
	req.in.head = mutable_buffer
	{
		req.in.head, state.head_read
	};

	// Setup the capstan and mark the end of the tape
	parse::buffer pb{req.in.head};
	parse::capstan pc{pb};
	pc.read += size(req.in.head);

	// Play the tape through the formal grammar.
	const http::response::head head{pc};
	assert(pb.completed() == pb.size());
	assert(pb.completed() == state.head_read);
	state.status = http::status(head.status);
	state.content_length = head.content_length;

	// Proffer the HTTP head to the peer instance which owns the link working
	// this tag so it can learn from any header data.
	assert(link.peer);
	link.peer->handle_head_recv(link, *this, head);

	if(contiguous)
	{
		const auto content_max
		{
			std::max(ssize_t(size(req.in.content) - state.head_read), ssize_t(0))
		};

		//TODO: XXX data(req.in.head)
		req.in.content = mutable_buffer
		{
			data(req.in.head) + state.head_read, size_t(content_max)
		};
	}

	// Branch for starting chunked encoding. We feed it whatever we have from
	// beyond the head as whole or part (or none) of the first chunk. Similar
	// to the non-chunked routine below, beyond_head may include all of the
	// chunk content and then part of the next message too: read_chunk_head
	// will return anything beyond this message as overrun and indicate done.
	if(head.transfer_encoding == "chunked")
	{
		if(dynamic)
		{
			assert(req.opt);
			req.in.chunks.reserve(req.opt->chunks_reserve);
		}

		const const_buffer chunk
		{
			!dynamic?
				const_buffer{req.in.content, move(req.in.content, beyond_head)}:
				beyond_head
		};

		assert(!state.chunk_length);
		state.chunk_length = -1;
		const const_buffer overrun
		{
			!dynamic?
				read_chunk_head(chunk, done):
				read_chunk_dynamic_head(chunk, done)
		};

		assert(empty(overrun) || done == true);
		return overrun;
	}

	// If no branch taken the rest of this function expects a content length
	// to be known from the received head.
	if(head.transfer_encoding)
		throw error
		{
			"Unsupported transfer-encoding '%s'", head.transfer_encoding
		};

	if(dynamic)
	{
		assert(req.opt);
		const size_t alloc_size
		{
			std::min(state.content_length, req.opt->content_length_maxalloc)
		};

		req.in.dynamic = unique_buffer<mutable_buffer>{alloc_size};
		req.in.content = req.in.dynamic;
	}

	// Now we check how much content was received beyond the head
	const size_t &content_read
	{
		std::min(state.content_length, beyond_head_len)
	};

	// Now we know how much bleed into the next message was also received
	assert(beyond_head_len >= content_read);
	const size_t beyond_content_len
	{
		beyond_head_len - content_read
	};

	//TODO: XXX data(req.in.head)
	const const_buffer partial_content
	{
		data(req.in.head) + state.head_read, content_read
	};

	// Anything remaining is not our response and must be given back.
	const const_buffer overrun
	{
		beyond_head + size(partial_content), beyond_content_len
	};

	// Reduce the user's content buffer to the content-length. This is sort of
	// how we convey the content-length back to the user. The buffer size will
	// eventually reflect how much content was actually received; the user can
	// find the given content-length by parsing the header.
	req.in.content = mutable_buffer
	{
		req.in.content, state.content_length
	};

	// Any partial content was written to the head buffer by accident,
	// that may have to be copied over to the content buffer.
	if(!empty(partial_content) && !contiguous)
		copy(req.in.content, partial_content);

	// Invoke the read_content() routine which will increment this->content_read
	read_content(partial_content, done);
	assert(state.content_read == size(partial_content));
	assert(state.content_read == state.content_length || !done);

	return overrun;
}

ircd::const_buffer
ircd::server::tag::read_content(const const_buffer &buffer,
                                bool &done)
{
	assert(request);
	auto &req{*request};
	const auto &content{req.in.content};

	// The amount of remaining content for the response sequence
	assert(size(content) + content_overflow() >= state.content_read);
	assert(size(content) + content_overflow() == state.content_length);
	const size_t remaining
	{
		size(content) + content_overflow() - state.content_read
	};

	// The amount of content read in this buffer only.
	const size_t addl_content_read
	{
		std::min(size(buffer), remaining)
	};

	state.content_read += addl_content_read;
	assert(size(buffer) - addl_content_read == 0);
	assert(state.content_read <= size(content) + content_overflow());
	assert(state.content_read <= state.content_length);

	// Invoke the user's optional progress callback; this function
	// should be marked noexcept for the time being.
	if(req.in.progress)
		req.in.progress(buffer, const_buffer{content, state.content_read});

	// Finished with content
	if(state.content_read == size(content) + content_overflow())
		content_completed(*this, done);

	return {};
}

void
ircd::server::content_completed(tag &tag,
                                bool &done)
{
	assert(tag.request);
	auto &req{*tag.request};
	const auto &content{req.in.content};

	assert(!done);
	done = true;

	assert(req.opt);
	assert(tag.state.content_read == tag.state.content_length);
	if(tag.content_overflow() && !req.opt->truncate_content)
	{
		assert(tag.state.content_read > size(content));
		tag.set_exception<buffer_overrun>
		(
			"buffer of %zu bytes too small for content-length %zu bytes by %zu bytes",
			size(content),
			tag.state.content_length,
			tag.content_overflow()
		);
	}
	else tag.set_value(tag.state.status);
}

//
// chunked encoding into fixed-size buffers
//

namespace ircd::server
{
	static void chunk_content_completed(tag &, bool &done);
}

[[GCC::stack_protect]]
ircd::const_buffer
ircd::server::tag::read_chunk_head(const const_buffer &buffer,
                                   bool &done,
                                   const uint8_t recursion_level)
{
	assert(request);
	auto &req{*request};
	const auto &content{req.in.content};

	// Total useful bytes in content buffer at this time
	const size_t content_read_max
	{
		state.content_read + size(buffer)
	};

	// Candidate chunk head includes prior packets and the current; may
	// extend past the chunk head into other data.
	assert(content_read_max >= state.content_length);
	assert(content_read_max <= size(content));
	const const_buffer candidate_head
	{
		content + state.content_length, content_read_max - state.content_length
	};

	// informal search for head terminator
	assert(size(candidate_head) <= size(content));
	const auto pos
	{
		string_view{candidate_head}.find(http::line::terminator)
	};

	if(pos == string_view::npos)
	{
		state.content_read += size(buffer);
		return {};
	}

	// This indicates how much head was just received from this buffer only.
	assert(state.content_read >= state.content_length);
	assert(state.content_read - state.content_length <= pos + size(http::line::terminator));
	assert(size(candidate_head) - size(buffer) <= pos + size(http::line::terminator));
	assert(pos + size(http::line::terminator) <= size(candidate_head));
	assert(size(candidate_head) >= size(buffer));
	const size_t addl_head_bytes
	{
		pos + size(http::line::terminator) - (size(candidate_head) - size(buffer))
	};

	// The received buffer may go past the end of the head.
	assert(addl_head_bytes <= size(buffer));
	const size_t beyond_head_length
	{
		size(buffer) - addl_head_bytes
	};

	// The total head length is found from the end of the last chunk content
	state.content_read += addl_head_bytes;
	assert(state.content_read > state.content_length);
	const size_t head_length
	{
		state.content_read - state.content_length
	};

	// Window on any data in the buffer after the head.
	const const_buffer beyond_head
	{
		content + state.content_length + head_length, beyond_head_length
	};

	// Setup the capstan and mark the end of the tape
	parse::buffer pb
	{
		mutable_buffer
		{
			content + state.content_length, head_length
		}
	};
	parse::capstan pc{pb};
	pc.read += head_length;

	// Play the tape through the formal grammar.
	const http::response::chunk chunk{pc};
	state.chunk_length = chunk.size + size(http::line::terminator);

	// Now we check how much chunk was received beyond the head
	const auto &chunk_read
	{
		std::min(state.chunk_length, beyond_head_length)
	};

	// Now we know how much bleed into the next message was also received
	assert(beyond_head_length >= chunk_read);
	const size_t beyond_chunk_length
	{
		beyond_head_length - chunk_read
	};

	// Finally we erase the chunk head by replacing it with everything received
	// after it.
	const mutable_buffer target
	{
		content + state.content_length, beyond_head_length
	};

	assert(!empty(target) || !beyond_head);
	move(target, beyond_head);

	// Increment the content_length to now include this chunk
	state.content_length += state.chunk_length;

	// Adjust the content_read to erase the chunk head.
	state.content_read -= head_length;

	const const_buffer partial_chunk
	{
		target, chunk_read
	};

	const const_buffer overrun
	{
		target + chunk_read, beyond_chunk_length
	};

	assert(state.chunk_length >= 2);
	assert(state.chunk_length != size_t(-1));
	read_chunk_content(partial_chunk, done);

	if(done || empty(overrun))
		return overrun;

	// Prevent stack overflow from lots of tiny chunks nagled together.
	if(unlikely(recursion_level >= 32))
		throw error
		{
			"Chunking recursion limit exceeded"
		};

	return read_chunk_head(overrun, done, recursion_level + 1);
}

ircd::const_buffer
ircd::server::tag::read_chunk_content(const const_buffer &buffer,
                                      bool &done)
{
	assert(request);
	auto &req{*request};
	const auto &content{req.in.content};

	// The amount of remaining content for the response sequence
	const size_t remaining
	{
		content_remaining()
	};

	// The amount of content read in this buffer only.
	const size_t addl_content_read
	{
		std::min(size(buffer), remaining)
	};

	// Increment the read counters for this chunk and all chunks.
	state.chunk_read += addl_content_read;
	state.content_read += addl_content_read;
	assert(state.chunk_read <= state.content_read);

	// Invoke the user's optional progress callback; this function
	// should be marked noexcept for the time being.
	if(req.in.progress && !done)
		req.in.progress(buffer, const_buffer{content, state.content_read});

	// This branch is taken at the completion of a chunk. The size
	// all the buffers is rolled back to hide the terminator so it's
	// either ignored or overwritten so it doesn't leak to the user.
	if(state.content_read == state.content_length)
		chunk_content_completed(*this, done);

	// Not finished
	if(likely(state.content_read != state.content_length))
		return {};

	assert(state.chunk_read == state.chunk_length);
	assert(state.chunk_read <= state.content_read);
	assert(state.chunk_length != size_t(-1));
	assert(state.chunk_length != 0 || done);
	assert(state.chunk_read != 0 || done);
	state.chunk_length = size_t(-1);
	state.chunk_read = 0;
	return {};
}

void
ircd::server::chunk_content_completed(tag &tag,
                                      bool &done)
{
	assert(tag.request);
	auto &req{*tag.request};
	auto &state{tag.state};

	// Remove the terminator from the total length state.
	assert(state.content_length >= size(http::line::terminator));
	assert(state.content_read >= size(http::line::terminator));
	state.content_length -= size(http::line::terminator);
	state.content_read -= size(http::line::terminator);

	// Remove the terminator from the chunk length state.
	assert(state.chunk_length >= size(http::line::terminator));
	assert(state.chunk_read >= size(http::line::terminator));
	assert(state.chunk_read == state.chunk_length);
	state.chunk_length -= size(http::line::terminator);
	state.chunk_read -= size(http::line::terminator);

	// State sanity tests
	assert(state.content_length >= state.content_read);
	assert(state.content_length >= state.chunk_length);
	assert(state.content_length >= state.chunk_read);
	assert(state.content_read >= state.chunk_length);
	assert(state.content_read >= state.chunk_read);
	assert(state.chunk_length >= state.chunk_read);
	if(state.chunk_length > 0)
		return;

	assert(state.chunk_read == 0);
	req.in.content = mutable_buffer
	{
		req.in.content, state.content_length
	};

	assert(!done);
	done = true;
	tag.set_value(state.status);
}

//
// chunked encoding into dynamic memories
//

namespace ircd::server
{
	static void chunk_dynamic_contiguous_copy(struct tag::state &, request &);
	static void chunk_dynamic_content_completed(tag &, bool &done);
}

[[GCC::stack_protect]]
ircd::const_buffer
ircd::server::tag::read_chunk_dynamic_head(const const_buffer &buffer,
                                           bool &done,
                                           const uint8_t recursion_level)
{
	assert(request);
	auto &req{*request};
	assert(null(req.in.content)); // dynamic chunk mode
	assert(state.chunk_length == size_t(-1)); // chunk head mode

	// The primary HTTP head was placed in req.in.head. Before this function
	// was reached req.in.head was resized tight to that head. There is still
	// buffer remaining after that which we now use for chunk heads. We offset
	// to state.head_read and cannot use more than state.head_rem for chunk
	// head scratch. Chunk heads may overwrite each other to not run out of
	// head buffer while keeping the data buffers pure with chunk content.
	assert(size(req.in.head) >= state.head_read);
	const const_buffer chunk_head_scratch
	{
		data(req.in.head) + state.head_read, state.head_rem
	};

	assert(size(chunk_head_scratch) >= state.chunk_read + size(buffer));
	const const_buffer chunk_head_buffer
	{
		chunk_head_scratch, state.chunk_read + size(buffer)
	};

	// informal search for head terminator
	const auto pos
	{
		string_view{chunk_head_buffer}.find(http::line::terminator)
	};

	if(pos == string_view::npos)
	{
		state.chunk_read += size(buffer);
		state.content_read += size(buffer);
		assert(state.content_read == state.content_length + state.chunk_read);
		return {};
	}

	// This indicates how much head was just received from this buffer only.
	assert(pos + size(http::line::terminator) >= state.chunk_read);
	const size_t addl_head_bytes
	{
		pos + size(http::line::terminator) - state.chunk_read
	};

	// The received buffer may go past the end of the head.
	assert(addl_head_bytes >= size(http::line::terminator));
	assert(addl_head_bytes <= size(buffer));
	const size_t beyond_head_length
	{
		size(buffer) - addl_head_bytes
	};

	state.chunk_read += addl_head_bytes;
	state.content_read += addl_head_bytes;

	// Focus specifically on this chunk head and nothing after.
	assert(data(chunk_head_buffer) <= data(buffer));
	const const_buffer chunk_head
	{
		chunk_head_buffer, state.chunk_read
	};

	// Window on any data in the buffer after the head.
	const const_buffer beyond_head
	{
		chunk_head_buffer + size(chunk_head), beyond_head_length
	};

	state.chunk_read = 0;
	assert(state.content_read >= size(chunk_head));
	state.content_read -= size(chunk_head);

	// Setup the capstan and mark the end of the tape
	parse::buffer pb
	{
		mutable_buffer(const_cast<char *>(data(chunk_head)), size(chunk_head))
	};
	parse::capstan pc{pb};
	pc.read += size(chunk_head);

	// Play the tape through the formal grammar.
	const http::response::chunk chunk{pc};
	assert(state.chunk_length == size_t(-1));
	state.chunk_length = chunk.size + size(http::line::terminator);

	// Increment the content_length to now include this chunk
	state.content_length += state.chunk_length;
	assert(state.content_read <= state.content_length);

	// Allocate the chunk content on the vector.
	//TODO: maxalloc
	req.in.chunks.emplace_back(state.chunk_length);
	assert(size_chunks(req.in) == state.content_length);

	// Now we check how much chunk was received beyond the head
	// state.chunk_read is still 0 here because that's only incremented
	// in the content read function.
	assert(state.chunk_read == 0);
	const auto &chunk_read
	{
		std::min(state.chunk_length, beyond_head_length)
	};

	// Now we know how much bleed into the next message was also received
	assert(beyond_head_length >= chunk_read);
	const size_t beyond_chunk_length
	{
		beyond_head_length - chunk_read
	};

	const const_buffer partial_chunk
	{
		beyond_head, chunk_read
	};

	const size_t copied
	{
		copy(req.in.chunks.back(), partial_chunk)
	};

	const const_buffer overrun
	{
		beyond_head + chunk_read, beyond_chunk_length
	};

	assert(state.chunk_length >= 2);
	read_chunk_dynamic_content(partial_chunk, done);

	if(done || empty(overrun))
		return overrun;

	// Prevent stack overflow from lots of tiny chunks nagled together.
	if(unlikely(recursion_level >= 32))
		throw error
		{
			"Chunking recursion limit exceeded"
		};

	return read_chunk_dynamic_head(overrun, done, recursion_level + 1);
}

ircd::const_buffer
ircd::server::tag::read_chunk_dynamic_content(const const_buffer &buffer,
                                              bool &done)
{
	assert(request);
	auto &req{*request};

	assert(state.chunk_length != size_t(-1)); // content mode
	assert(null(req.in.content)); // dynamic chunk mode
	assert(!req.in.chunks.empty());
	const auto &chunk
	{
		req.in.chunks.back()
	};

	// The amount of remaining content for the response sequence
	assert(size(chunk) >= state.chunk_read);
	const size_t remaining
	{
		size(chunk) - state.chunk_read
	};

	// The amount of content read in this buffer only.
	const size_t addl_content_read
	{
		std::min(size(buffer), remaining)
	};

	// Increment the read counters for this chunk and all chunks.
	state.chunk_read += addl_content_read;
	state.content_read += addl_content_read;
	assert(state.chunk_read <= state.content_read);
	assert(state.chunk_read <= state.chunk_length);
	assert(state.content_length >= state.chunk_length);
	assert(state.content_length >= state.chunk_read);

	// Invoke the user's optional progress callback; this function
	// should be marked noexcept for the time being.
	if(req.in.progress && !done)
		req.in.progress(buffer, const_buffer{chunk, state.chunk_read});

	if(state.chunk_read == state.chunk_length)
		chunk_dynamic_content_completed(*this, done);

	assert(state.chunk_read <= state.chunk_length);
	if(likely(state.chunk_read != state.chunk_length))
		return {};

	assert(state.chunk_read <= state.content_read);
	assert(state.chunk_length != size_t(-1));
	assert(state.chunk_length != 0 || done);
	assert(state.chunk_read != 0 || done);
	state.chunk_length = size_t(-1);
	state.chunk_read = 0;
	return {};
}

void
ircd::server::chunk_dynamic_content_completed(tag &tag,
                                              bool &done)
{
	assert(tag.request);
	auto &req{*tag.request};
	auto &state{tag.state};
	assert(!req.in.chunks.empty());
	auto &chunk{req.in.chunks.back()};

	// Remove the terminator from the total length state.
	assert(state.content_length >= size(http::line::terminator));
	assert(state.content_read >= size(http::line::terminator));
	state.content_length -= size(http::line::terminator);
	state.content_read -= size(http::line::terminator);

	// Remove the terminator from the chunk length state.
	assert(state.chunk_length >= size(http::line::terminator));
	assert(state.chunk_read >= size(http::line::terminator));
	assert(state.chunk_read == state.chunk_length);
	state.chunk_length -= size(http::line::terminator);
	state.chunk_read -= size(http::line::terminator);

	// Remove the terminator from the end of the chunk
	std::get<1>(chunk) -= size(http::line::terminator);
	assert(size(chunk) == state.chunk_length);
	assert(std::get<0>(chunk) <= std::get<1>(chunk));

	// State sanity tests
	assert(state.content_length == size_chunks(req.in));
	assert(state.content_length >= state.chunk_length);
	assert(state.content_length >= state.chunk_read);
	assert(state.content_read >= state.chunk_length);
	assert(state.content_read >= state.chunk_read);
	assert(state.chunk_length >= state.chunk_read);
	assert(state.chunk_length != size_t(-1));
	if(state.chunk_length > 0)
		return;

	assert(state.chunk_read == 0);
	assert(req.opt);
	if(req.opt->contiguous_content && !req.in.chunks.empty())
		chunk_dynamic_contiguous_copy(state, req);

	assert(!done);
	done = true;
	tag.set_value(state.status);
}

void
ircd::server::chunk_dynamic_contiguous_copy(struct tag::state &state,
                                            request &req)
{
	assert(state.content_length == size_chunks(req.in));
	assert(req.in.chunks.size() >= 1);
	assert(empty(req.in.chunks.back()));
	req.in.chunks.pop_back();

	if(req.in.chunks.size() > 1)
	{
		req.in.dynamic = size_chunks(req.in);
		req.in.content = req.in.dynamic;

		size_t copied{0};
		for(const auto &buffer : req.in.chunks)
			copied += copy(req.in.content + copied, buffer);

		assert(copied == size(req.in.content));
		assert(copied == state.content_length);
	}
	else if(req.in.chunks.size() == 1)
	{
		req.in.dynamic = std::move(req.in.chunks.front());
		req.in.content = req.in.dynamic;
		assert(size(req.in.content) == state.content_length);
	}

	req.in.chunks.clear();
}

/// An idempotent operation that provides the location of where the socket
/// should place the next received data. The tag figures this out based on
/// whether it receiving HTTP head data or whether it is in content mode.
///
[[GCC::stack_protect]]
ircd::mutable_buffer
ircd::server::tag::make_read_buffer()
const
{
	const bool chunk_header_mode
	{
		state.chunk_length == size_t(-1)
	};

	const bool chunk_dynamic_header_mode
	{
		chunk_header_mode && null(request->in.content)
	};

	assert(request);
	assert(state.head_read <= size(request->in.head));
	assert(state.content_read <= state.content_length + state.chunk_read || chunk_header_mode);
	assert(state.content_read <= state.content_length || chunk_header_mode);
	const mutable_buffer ret
	{
		state.status == (http::code)0?
			make_read_head_buffer():

		chunk_dynamic_header_mode?
			make_read_chunk_dynamic_head_buffer():

		chunk_header_mode?
			make_read_chunk_head_buffer():

		state.chunk_length && null(request->in.content)?
			make_read_chunk_dynamic_content_buffer():

		state.chunk_length?
			make_read_chunk_content_buffer():

		state.content_read >= size(request->in.content)?
			make_read_discard_buffer():

		make_read_content_buffer()
	};

	return ret;
}

ircd::mutable_buffer
ircd::server::tag::make_read_head_buffer()
const
{
	assert(request);
	const auto &req{*request};
	const auto &head{req.in.head};
	const mutable_buffer buffer
	{
		head + state.head_read
	};

	if(unlikely(empty(buffer)))
		throw buffer_overrun
		{
			"Head buffer too small for HTTP; size:%zu head_read:%zu",
			size(req.in.head),
			state.head_read,
		};

	assert(size(buffer) <= size(head));
	assert(size(buffer) > 0);
	return buffer;
}

ircd::mutable_buffer
ircd::server::tag::make_read_content_buffer()
const
{
	assert(request);
	const auto &req{*request};
	const auto &content{req.in.content};
	const mutable_buffer buffer
	{
		content + state.content_read
	};

	if(unlikely(empty(buffer)))
		throw buffer_overrun
		{
			"Content buffer too small; size:%zu content_length:%zu content_read:%zu",
			size(content),
			state.content_length,
			state.content_read,
		};

	assert(!empty(buffer));
	return buffer;
}

/// The chunk head buffer starts after the last chunk ended and has a size of
/// the rest of the available content buffer (hopefully much less will be
/// needed). If only part of the chunk head was received previously this
/// function accounts for that by returning a buffer which starts at the
/// content_read offset (which is at the end of that previous read).
///
ircd::mutable_buffer
ircd::server::tag::make_read_chunk_head_buffer()
const
{
	assert(request);
	assert(state.chunk_length == size_t(-1));
	assert(state.content_read >= state.content_length);

	const auto &req{*request};
	const auto &content{req.in.content};

	assert(size(content) >= state.content_read);
	const mutable_buffer buffer
	{
		content + state.content_read
	};

	if(unlikely(empty(buffer)))
		throw buffer_overrun
		{
			"Content buffer too small to read next chunk header; size:%zu content_read:%zu",
			size(content),
			state.content_read,
		};

	assert(!empty(buffer));
	return buffer;
}

ircd::mutable_buffer
ircd::server::tag::make_read_chunk_content_buffer()
const
{
	assert(request);
	assert(state.chunk_length > 0);
	assert(state.content_read <= state.content_length);

	const auto &req{*request};
	const auto &content{req.in.content};

	assert(size(content) >= state.content_read);
	const size_t chunk_remaining
	{
		content_remaining()
	};

	assert(chunk_remaining <= state.chunk_length);
	assert(chunk_remaining == state.content_length - state.content_read);
	const mutable_buffer buffer
	{
		content + state.content_read, chunk_remaining
	};

	if(unlikely(empty(buffer)))
		throw buffer_overrun
		{
			"Chunk dynamic content buffer too small content[size:%zu read:%zu] chunk[size:%zu remain:%zu]",
			size(content),
			state.content_read,
			state.chunk_length,
			chunk_remaining,
		};

	assert(!empty(buffer));
	return buffer;
}

/// The dynamic chunk head buffer starts after the main head and has a size
/// of the remaining main head buffer. This area is overwritten for each
/// chunk head.
///
ircd::mutable_buffer
ircd::server::tag::make_read_chunk_dynamic_head_buffer()
const
{
	assert(request);
	const auto &req{*request};

	assert(state.chunk_length == size_t(-1));
	assert(null(req.in.content));
	assert(size(req.in.head) >= state.head_read);

	const mutable_buffer head_buffer
	{
		data(req.in.head) + state.head_read, state.head_rem
	};

	// The total offset in the head buffer is the message head plus the
	// amount of chunk head received so far, which is kept in chunk_read.
	const mutable_buffer buffer
	{
		head_buffer + state.chunk_read
	};

	if(unlikely(size(buffer) < 16))
		throw buffer_overrun
		{
			"Chunk dynamic head buffer too small size:%zu chunk_read:%zu head_read:%zu head_rem:%zu",
			size(buffer),
			state.chunk_read,
			state.head_read,
			state.head_rem,
		};

	assert(!empty(buffer));
	return buffer;
}

ircd::mutable_buffer
ircd::server::tag::make_read_chunk_dynamic_content_buffer()
const
{
	assert(request);
	const auto &req{*request};

	assert(state.chunk_length > 0);
	assert(state.content_read <= state.content_length);
	assert(null(req.in.content));
	assert(!req.in.chunks.empty());
	const auto &buffer
	{
		req.in.chunks.back()
	};

	assert(size(buffer) == state.chunk_length);
	assert(state.chunk_read <= size(buffer));
	const mutable_buffer ret
	{
		buffer + state.chunk_read
	};

	if(unlikely(empty(ret)))
		throw buffer_overrun
		{
			"Chunk dynamic content buffer too small size:%zu chunk_read:%zu chunk_length:%zu",
			size(buffer),
			state.chunk_read,
			state.chunk_length,
		};

	assert(!empty(ret));
	return ret;
}

ircd::mutable_buffer
ircd::server::tag::make_read_discard_buffer()
const
{
	assert(request);
	assert(content_overflow() > 0);
	assert(content_overflow() <= state.content_length);
	assert(state.content_read >= size(request->in.content));
	assert(state.content_length >= state.content_read);
	const size_t remaining
	{
		state.content_length - state.content_read
	};

	assert(remaining <= state.content_length);
	thread_local char buffer[512];
	const size_t buffer_max
	{
		std::min(remaining, sizeof(buffer))
	};

	const mutable_buffer ret
	{
		buffer, buffer_max
	};

	assert(!empty(ret));
	return ret;
}

size_t
ircd::server::tag::content_remaining()
const
{
	assert(state.content_length >= state.content_read);
	return state.content_length - state.content_read;
}

size_t
ircd::server::tag::content_overflow()
const
{
	assert(request);
	const auto &req{*request};
	const ssize_t diff(state.content_length - size(req.in.content));
	return std::max(diff, ssize_t(0));
}

template<class... args>
void
ircd::server::tag::set_value(args&&... a)
{
	if(abandoned())
		return;

	const http::code &code
	{
		std::forward<args>(a)...
	};

	assert(request->opt);
	if(request->opt->http_exceptions && code >= http::code(300))
	{
		const string_view content
		{
			data(request->in.content), size(request->in.content)
		};

		set_exception<http::error>(code, std::string{content});
		return;
	}

	p.set_value(code);
	assert(abandoned());
}

template<class E,
         class... args>
void
ircd::server::tag::set_exception(args&&... a)
try
{
	if(abandoned())
		return;

	throw E
	{
		std::forward<args>(a)...
	};
}
catch(const std::exception &e)
{
	set_exception(std::current_exception());
}

void
ircd::server::tag::set_exception(std::exception_ptr eptr)
{
	if(abandoned())
		return;

	p.set_exception(std::move(eptr));
	assert(abandoned());
}

bool
ircd::server::tag::abandoned()
const
{
	if(!p.valid())
		return true;

	assert(p.st);
	assert(is(p.state(), ctx::future_state::PENDING));
	return false;
}

bool
ircd::server::tag::canceled()
const
{
	return !!cancellation;
}

bool
ircd::server::tag::committed()
const
{
	return write_completed() > 0;
}

size_t
ircd::server::tag::read_remaining()
const
{
	return read_size() - read_completed();
}

size_t
ircd::server::tag::read_completed()
const
{
	return state.head_read + state.content_read;
}

size_t
ircd::server::tag::read_size()
const
{
	return state.head_read + state.content_length;
}

size_t
ircd::server::tag::write_remaining()
const
{
	return write_size() - write_completed();
}

size_t
ircd::server::tag::write_completed()
const
{
	return state.written;
}

size_t
ircd::server::tag::write_size()
const
{
	return request? size(request->out) : 0;
}