Archive/rpcs3
0
0
Fork 0
mirror of https://github.com/RPCS3/rpcs3.git synced 2025-04-28 13:28:01 +03:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

IDM: Implement lock-free smart pointers (#16403)

Browse source 
Replaces `std::shared_pointer` with `stx::atomic_ptr` and `stx::shared_ptr`.

Notes to programmers:

* This pr kills the use of `dynamic_cast`, `std::dynamic_pointer_cast` and `std::weak_ptr` on IDM objects, possible replacement is to save the object ID on the base object, then use idm::check/get_unlocked to the destination type via the saved ID which may be null. Null pointer check is how you can tell type mismatch (as dynamic cast) or object destruction (as weak_ptr locking).
* Double-inheritance on IDM objects should be used with care, `stx::shared_ptr` does not support constant-evaluated pointer offsetting to parent/child type.
* `idm::check/get_unlocked` can now be used anywhere.

Misc fixes:
* Fixes some segfaults with RPCN with interaction with IDM.
* Fix deadlocks in access violation handler due locking recursion.
* Fixes race condition in process exit-spawn on memory containers read.
* Fix bug that theoretically can prevent RPCS3 from booting - fix `id_manager::typeinfo` comparison to compare members instead of `memcmp` which can fail spuriously on padding bytes.
* Ensure all IDM inherited types of base, either has `id_base` or `id_type` defined locally, this allows to make getters such as `idm::get_unlocked<lv2_socket, lv2_socket_raw>()` which were broken before. (requires save-states invalidation)
* Removes broken operator[] overload of `stx::shared_ptr` and `stx::single_ptr` for non-array types.
This commit is contained in:
Elad 2024-12-22 20:59:48 +02:00 • committed by GitHub
parent 385710672f
commit 575a245f8d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
124 changed files with 1145 additions and 1052 deletions
Download patch file Download diff file Expand all files Collapse all files

6
Utilities/Thread.cpp
View file

@ -1360,7 +1360,7 @@ bool handle_access_violation(u32 addr, bool is_writing, ucontext_t* context) noe
// check if address is RawSPU MMIO register
do if (addr - RAW_SPU_BASE_ADDR < (6 * RAW_SPU_OFFSET) && (addr % RAW_SPU_OFFSET) >= RAW_SPU_PROB_OFFSET)
{
auto thread = idm::get<named_thread<spu_thread>>(spu_thread::find_raw_spu((addr - RAW_SPU_BASE_ADDR) / RAW_SPU_OFFSET));
auto thread = idm::get_unlocked<named_thread<spu_thread>>(spu_thread::find_raw_spu((addr - RAW_SPU_BASE_ADDR) / RAW_SPU_OFFSET));
if (!thread)
{
@ -1548,7 +1548,7 @@ bool handle_access_violation(u32 addr, bool is_writing, ucontext_t* context) noe
}
}
if (auto pf_port = idm::get<lv2_obj, lv2_event_port>(pf_port_id); pf_port && pf_port->queue)
if (auto pf_port = idm::get_unlocked<lv2_obj, lv2_event_port>(pf_port_id); pf_port && pf_port->queue)
{
// We notify the game that a page fault occurred so it can rectify it.
// Note, for data3, were the memory readable AND we got a page fault, it must be due to a write violation since reads are allowed.
@ -2602,7 +2602,7 @@ bool thread_base::join(bool dtor) const
if (i >= 16 && !(i & (i - 1)) && timeout != atomic_wait_timeout::inf)
{
sig_log.error(u8"Thread [%s] is too sleepy. Waiting for it %.3fµs already!", *m_tname.load(), (utils::get_tsc() - stamp0) / (utils::get_tsc_freq() / 1000000.));
sig_log.error(u8"Thread [%s] is too sleepy. Waiting for it %.3fus already!", *m_tname.load(), (utils::get_tsc() - stamp0) / (utils::get_tsc_freq() / 1000000.));
}
}

2
Utilities/mutex.h
View file

@ -121,7 +121,7 @@ public:
void unlock_hle()
{
const u32 value = atomic_storage<u32>::fetch_add_hle_rel(m_value.raw(), 0u - c_one);
const u32 value = atomic_storage<u32>::fetch_add_hle_rel(m_value.raw(), ~c_one + 1);
if (value != c_one) [[unlikely]]
{