blob: a52e464fad21d3575b8356c3c4a9f7e4ffb36e5c [file] [log] [blame]
// Copyright 2018 The Fuchsia Authors
//
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file or at
// https://opensource.org/licenses/MIT
#include <lib/cmdline.h>
#include <lib/counters.h>
#include <trace.h>
#include <kernel/thread.h>
#include <lk/init.h>
#include <object/pager_dispatcher.h>
#include <object/thread_dispatcher.h>
#include <vm/page_source.h>
#define LOCAL_TRACE 0
KCOUNTER(dispatcher_pager_create_count, "dispatcher.pager.create")
KCOUNTER(dispatcher_pager_destroy_count, "dispatcher.pager.destroy")
KCOUNTER(dispatcher_pager_overtime_wait_count, "dispatcher.pager.overtime_waits")
KCOUNTER(dispatcher_pager_total_request_count, "dispatcher.pager.total_requests")
KCOUNTER(dispatcher_pager_succeeded_request_count, "dispatcher.pager.succeeded_requests")
KCOUNTER(dispatcher_pager_failed_request_count, "dispatcher.pager.failed_requests")
KCOUNTER(dispatcher_pager_timed_out_request_count, "dispatcher.pager.timed_out_requests")
// Log warnings every |pager_overtime_wait_seconds| a thread is blocked waiting on a page
// request. If the thread has been waiting for |pager_overtime_timeout_seconds|, return an
// error instead of waiting indefinitely.
static constexpr uint64_t kDefaultPagerOvertimeWaitSeconds = 20;
static constexpr uint64_t kDefaultPagerOvertimeTimeoutSeconds = 300;
static uint64_t pager_overtime_wait_seconds = kDefaultPagerOvertimeWaitSeconds;
static uint64_t pager_overtime_timeout_seconds = kDefaultPagerOvertimeTimeoutSeconds;
zx_status_t PagerDispatcher::Create(KernelHandle<PagerDispatcher>* handle, zx_rights_t* rights) {
fbl::AllocChecker ac;
KernelHandle new_handle(fbl::AdoptRef(new (&ac) PagerDispatcher()));
if (!ac.check()) {
return ZX_ERR_NO_MEMORY;
}
*rights = default_rights();
*handle = ktl::move(new_handle);
return ZX_OK;
}
PagerDispatcher::PagerDispatcher() : SoloDispatcher() {
kcounter_add(dispatcher_pager_create_count, 1);
}
PagerDispatcher::~PagerDispatcher() {
DEBUG_ASSERT(srcs_.is_empty());
kcounter_add(dispatcher_pager_destroy_count, 1);
}
zx_status_t PagerDispatcher::CreateSource(fbl::RefPtr<PortDispatcher> port, uint64_t key,
fbl::RefPtr<PageSource>* src_out) {
fbl::AllocChecker ac;
auto src = fbl::AdoptRef(new (&ac) PagerSource(this, ktl::move(port), key));
if (!ac.check()) {
return ZX_ERR_NO_MEMORY;
}
Guard<Mutex> guard{&list_mtx_};
srcs_.push_front(src);
*src_out = ktl::move(src);
return ZX_OK;
}
fbl::RefPtr<PagerSource> PagerDispatcher::ReleaseSource(PagerSource* src) {
Guard<Mutex> guard{&list_mtx_};
return src->InContainer() ? srcs_.erase(*src) : nullptr;
}
void PagerDispatcher::on_zero_handles() {
Guard<Mutex> guard{&list_mtx_};
while (!srcs_.is_empty()) {
fbl::RefPtr<PagerSource> src = srcs_.pop_front();
// Call unlocked to prevent a double-lock if PagerDispatcher::ReleaseSource is called,
// and to preserve the lock order that PagerSource locks are acquired before the
// list lock.
guard.CallUnlocked([&src]() mutable {
src->Close();
src->OnDispatcherClosed();
});
}
}
zx_status_t PagerDispatcher::RangeOp(uint32_t op, fbl::RefPtr<VmObject> vmo, uint64_t offset,
uint64_t length, uint64_t data) {
switch (op) {
case ZX_PAGER_OP_FAIL: {
auto signed_data = static_cast<int64_t>(data);
if (signed_data < INT32_MIN || signed_data > INT32_MAX) {
return ZX_ERR_INVALID_ARGS;
}
auto error_status = static_cast<zx_status_t>(data);
if (!PageSource::IsValidFailureCode(error_status)) {
return ZX_ERR_INVALID_ARGS;
}
return vmo->FailPageRequests(offset, length, error_status);
}
default:
return ZX_ERR_NOT_SUPPORTED;
}
}
PagerSource::PagerSource(PagerDispatcher* dispatcher, fbl::RefPtr<PortDispatcher> port,
uint64_t key)
: PageSource(), pager_(dispatcher), port_(ktl::move(port)), key_(key) {
LTRACEF("%p key %lx\n", this, key_);
}
PagerSource::~PagerSource() {
LTRACEF("%p\n", this);
DEBUG_ASSERT(closed_);
DEBUG_ASSERT(!complete_pending_);
}
void PagerSource::GetPageAsync(page_request_t* request) {
Guard<Mutex> guard{&mtx_};
ASSERT(!closed_);
QueueMessageLocked(request);
}
void PagerSource::QueueMessageLocked(page_request_t* request) {
if (packet_busy_) {
list_add_tail(&pending_requests_, &request->provider_node);
return;
}
packet_busy_ = true;
active_request_ = request;
uint64_t offset, length;
uint16_t cmd;
if (request != &complete_request_) {
cmd = ZX_PAGER_VMO_READ;
offset = request->offset;
length = request->length;
// The vm subsystem should guarantee this
uint64_t unused;
DEBUG_ASSERT(!add_overflow(offset, length, &unused));
// Trace flow events require an enclosing duration.
VM_KTRACE_DURATION(1, "page_request_queue", offset, length);
VM_KTRACE_FLOW_BEGIN(1, "page_request_queue", reinterpret_cast<uintptr_t>(&packet_));
} else {
offset = length = 0;
cmd = ZX_PAGER_VMO_COMPLETE;
}
zx_port_packet_t packet = {};
packet.key = key_;
packet.type = ZX_PKT_TYPE_PAGE_REQUEST;
packet.page_request.command = cmd;
packet.page_request.offset = offset;
packet.page_request.length = length;
packet_.packet = packet;
// We can treat ZX_ERR_BAD_HANDLE as if the packet was queued
// but the pager service never responds.
// TODO: Bypass the port's max queued packet count to prevent ZX_ERR_SHOULD_WAIT
ASSERT(port_->Queue(&packet_, ZX_SIGNAL_NONE) != ZX_ERR_SHOULD_WAIT);
}
void PagerSource::ClearAsyncRequest(page_request_t* request) {
Guard<Mutex> guard{&mtx_};
ASSERT(!closed_);
if (request == active_request_) {
if (request != &complete_request_) {
// Trace flow events require an enclosing duration.
VM_KTRACE_DURATION(1, "page_request_queue", active_request_->offset, active_request_->length);
VM_KTRACE_FLOW_END(1, "page_request_queue", reinterpret_cast<uintptr_t>(&packet_));
}
// Condition on whether or not we actually cancel the packet, to make sure
// we don't race with a call to PagerSource::Free.
if (port_->CancelQueued(&packet_)) {
OnPacketFreedLocked();
}
} else if (list_in_list(&request->provider_node)) {
list_delete(&request->provider_node);
}
}
void PagerSource::SwapRequest(page_request_t* old, page_request_t* new_req) {
Guard<Mutex> guard{&mtx_};
ASSERT(!closed_);
if (list_in_list(&old->provider_node)) {
list_replace_node(&old->provider_node, &new_req->provider_node);
} else if (old == active_request_) {
active_request_ = new_req;
}
}
void PagerSource::OnDetach() {
Guard<Mutex> guard{&mtx_};
ASSERT(!closed_);
complete_pending_ = true;
QueueMessageLocked(&complete_request_);
}
void PagerSource::OnClose() {
fbl::RefPtr<PagerSource> self;
Guard<Mutex> guard{&mtx_};
ASSERT(!closed_);
closed_ = true;
if (!complete_pending_) {
// We know PagerDispatcher::on_zero_handles hasn't been invoked, since that would
// have already closed this pager source.
self = pager_->ReleaseSource(this);
} // else this is released in PagerSource::Free
}
void PagerSource::OnDispatcherClosed() {
// The pager dispatcher's reference to this object is the only one we completely control. Now
// that it's gone, we need to make sure that port_ doesn't end up with an invalid pointer
// to packet_ if all external RefPtrs to this object go away.
Guard<Mutex> guard{&mtx_};
if (complete_pending_) {
if (port_->CancelQueued(&packet_)) {
// We successfully cancelled the message, so we don't have to worry about
// PagerSource::Free being called.
complete_pending_ = false;
} else {
// If we failed to cancel the message, then there is a pending call to
// PagerSource::Free. We need to make sure the object isn't deleted too early,
// so have it keep a reference to itself, which PagerSource::Free will then
// clean up.
self_ref_ = fbl::RefPtr(this);
}
} else {
// Either the complete message had already been dispatched when this object was closed or
// PagerSource::Free was called between this object being closed and this method taking the
// lock. In either case, the port no longer has a reference and cleanup is already done.
}
}
void PagerSource::Free(PortPacket* packet) {
fbl::RefPtr<PagerSource> self;
Guard<Mutex> guard{&mtx_};
if (active_request_ != &complete_request_) {
// Trace flow events require an enclosing duration.
VM_KTRACE_DURATION(1, "page_request_queue", active_request_->offset, active_request_->length);
VM_KTRACE_FLOW_END(1, "page_request_queue", reinterpret_cast<uintptr_t>(packet));
OnPacketFreedLocked();
} else {
complete_pending_ = false;
if (closed_) {
// If the source is closed, we need to do delayed cleanup. If the dispatcher
// has already been torn down, then there's a self-reference we need to clean
// up. Otherwise, clean up the dispatcher's reference to us.
self = ktl::move(self_ref_);
if (!self) {
self = pager_->ReleaseSource(this);
}
}
}
}
void PagerSource::OnPacketFreedLocked() {
packet_busy_ = false;
active_request_ = nullptr;
if (!list_is_empty(&pending_requests_)) {
QueueMessageLocked(list_remove_head_type(&pending_requests_, page_request, provider_node));
}
}
zx_status_t PagerSource::WaitOnEvent(Event* event) {
ThreadDispatcher::AutoBlocked by(ThreadDispatcher::Blocked::PAGER);
kcounter_add(dispatcher_pager_total_request_count, 1);
uint32_t waited = 0;
// declare a lambda to calculate our deadline to avoid an excessively large statement in our
// loop condition.
auto make_deadline = []() {
if (pager_overtime_wait_seconds == 0) {
return Deadline::infinite();
} else {
return Deadline::after(ZX_SEC(pager_overtime_wait_seconds));
}
};
zx_status_t result;
while ((result = event->Wait(make_deadline())) == ZX_ERR_TIMED_OUT) {
waited++;
// We might trigger this loop multiple times as we exceed multiples of the overtime counter, but
// we only want to count each unique overtime event in the kcounter.
if (waited == 1) {
dispatcher_pager_overtime_wait_count.Add(1);
}
// Error out if we've been waiting for longer than the specified timeout, to allow the rest of
// the system to make progress (if possible).
if (pager_overtime_timeout_seconds > 0 &&
waited * pager_overtime_wait_seconds >= pager_overtime_timeout_seconds) {
printf("ERROR Pager source %p has been blocked for %" PRIu64
" seconds. Page request timed out.\n",
this, pager_overtime_timeout_seconds);
dump_thread(Thread::Current::Get(), false);
kcounter_add(dispatcher_pager_timed_out_request_count, 1);
return ZX_ERR_TIMED_OUT;
}
// Determine whether we have any requests that have not yet been received off of the port.
bool active;
{
Guard<Mutex> guard{&mtx_};
active = !!active_request_;
}
printf("WARNING pager source %p has been blocked for %" PRIu64
" seconds with%s message waiting on port.\n",
this, waited * pager_overtime_wait_seconds, active ? "" : " no");
// Dump out the rest of the state of the oustanding requests.
Dump();
}
if (result == ZX_OK) {
kcounter_add(dispatcher_pager_succeeded_request_count, 1);
} else {
// Only counts failures that are *not* pager timeouts. Timeouts are tracked with
// dispatcher_pager_timed_out_request_count, which is updated above when we
// return early with ZX_ERR_TIMED_OUT.
kcounter_add(dispatcher_pager_failed_request_count, 1);
}
return result;
}
static void pager_init_func(uint level) {
pager_overtime_wait_seconds = gCmdline.GetUInt64("kernel.userpager.overtime_wait_seconds",
kDefaultPagerOvertimeWaitSeconds);
pager_overtime_timeout_seconds = gCmdline.GetUInt64("kernel.userpager.overtime_timeout_seconds",
kDefaultPagerOvertimeTimeoutSeconds);
}
LK_INIT_HOOK(pager_init, &pager_init_func, LK_INIT_LEVEL_LAST)