6502/CLK
1
0
Fork 0
mirror of https://github.com/TomHarte/CLK.git synced 2024-09-27 18:55:48 +00:00
Code Issues Projects Releases Wiki Activity

Removes ONE_BIG_LOCK, having effectively neutered it anyway.

Browse Source 
Starts work on more explicit API usage validation. Maybe the issue isn't a race condition?
This commit is contained in:
Thomas Harte 2020-08-16 22:09:25 -04:00
parent 27e8a3a1b5
commit 246b474a25
2 changed files with 21 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
Outputs/ScanTargets/BufferingScanTarget.cpp
View File

@ -11,12 +11,6 @@
#include <cassert>
#include <cstring>
// If enabled, this uses the producer lock to cover both production and consumption
// rather than attempting to proceed lockfree. This is primarily for diagnostic purposes;
// it allows empirical exploration of whether the logical and memory barriers that are
// meant to mediate things between the read pointers and the submit pointers are functioning.
#define ONE_BIG_LOCK
#define TextureAddressGetY(v) uint16_t((v) >> 11)
#define TextureAddressGetX(v) uint16_t((v) & 0x7ff)
#define TextureSub(a, b) (((a) - (b)) & 0x3fffff)
@ -79,6 +73,7 @@ uint8_t *BufferingScanTarget::begin_data(size_t required_length, size_t required
}
// Everything checks out, note expectation of a future end_data and return the pointer.
assert(!data_is_allocated_);
data_is_allocated_ = true;
vended_write_area_pointer_ = write_pointers_.write_area = TextureAddress(aligned_start_x, output_y);
@ -150,12 +145,23 @@ Outputs::Display::ScanTarget::Scan *BufferingScanTarget::begin_scan() {
result->line = write_pointers_.line;
vended_scan_ = result;
#ifndef NDEBUG
assert(!scan_is_ongoing_);
scan_is_ongoing_ = true;
#endif
return &result->scan;
}
void BufferingScanTarget::end_scan() {
std::lock_guard lock_guard(producer_mutex_);
#ifndef NDEBUG
assert(scan_is_ongoing_);
scan_is_ongoing_ = false;
#endif
// Complete the scan only if one is afoot.
if(vended_scan_) {
vended_scan_->data_y = TextureAddressGetY(vended_write_area_pointer_);
@ -188,6 +194,10 @@ void BufferingScanTarget::announce(Event event, bool is_visible, const Outputs::
if(output_is_visible_ == is_visible) return;
output_is_visible_ = is_visible;
#ifndef NDEBUG
assert(!scan_is_ongoing_);
#endif
if(is_visible) {
const auto read_pointers = read_pointers_.load(std::memory_order::memory_order_relaxed);
@ -261,14 +271,7 @@ const Outputs::Display::Metrics &BufferingScanTarget::display_metrics() {
}
void BufferingScanTarget::set_write_area(uint8_t *base) {
// This is a bit of a hack. This call needs the producer mutex and should be
// safe to call from a @c perform block in order to support all potential consumers.
// But the temporary hack of ONE_BIG_LOCK then implies that either I need a recursive
// mutex, or I have to make a coupling assumption about my caller. I've done the latter,
// because ONE_BIG_LOCK is really really meant to be temporary. I hope.
#ifndef ONE_BIG_LOCK
std::lock_guard lock_guard(producer_mutex_);
#endif
write_area_ = base;
write_pointers_ = submit_pointers_ = read_pointers_ = PointerSet();
allocation_has_failed_ = true;
@ -325,10 +328,6 @@ void BufferingScanTarget::complete_output_area(const OutputArea &area) {
}
void BufferingScanTarget::perform(const std::function<void(void)> &function) {
#ifdef ONE_BIG_LOCK
std::lock_guard lock_guard(producer_mutex_);
#endif
while(is_updating_.test_and_set(std::memory_order_acquire));
function();
is_updating_.clear(std::memory_order_release);

5
Outputs/ScanTargets/BufferingScanTarget.hpp
View File

@ -250,6 +250,11 @@ class BufferingScanTarget: public Outputs::Display::ScanTarget {
// from a call to @c get_new_modals.
Modals modals_;
bool modals_are_dirty_ = false;
#ifndef NDEBUG
// Debug features; these amount to API validate.
bool scan_is_ongoing_ = false;
#endif
};