archive/rpcs3
1
0
Fork 0
mirror of https://github.com/RPCS3/rpcs3.git synced 2025-07-09 08:21:29 +12:00
Code Issues Projects Releases Packages Wiki Activity Actions

Initial Linux Aarch64 support

Browse source 
* Update asmjit dependency (aarch64 branch)
* Disable USE_DISCORD_RPC by default
* Dump some JIT objects in rpcs3 cache dir
* Add SIGILL handler for all platforms
* Fix resetting zeroing denormals in thread pool
* Refactor most v128:: utils into global gv_** functions
* Refactor PPU interpreter (incomplete), remove "precise"
* - Instruction specializations with multiple accuracy flags
* - Adjust calling convention for speed
* - Removed precise/fast setting, replaced with static
* - Started refactoring interpreters for building at runtime JIT
*   (I got tired of poor compiler optimizations)
* - Expose some accuracy settings (SAT, NJ, VNAN, FPCC)
* - Add exec_bytes PPU thread variable (akin to cycle count)
* PPU LLVM: fix VCTUXS+VCTSXS instruction NaN results
* SPU interpreter: remove "precise" for now (extremely non-portable)
* - As with PPU, settings changed to static/dynamic for interpreters.
* - Precise options will be implemented later
* Fix termination after fatal error dialog
This commit is contained in:
Nekotekina 2021-12-30 19:39:18 +03:00
parent d6aa834b5f
commit 580bd2b25e
89 changed files with 20360 additions and 5612 deletions
Download patch file Download diff file Expand all files Collapse all files

12
rpcs3/Emu/RSX/rsx_methods.cpp
View file

@ -42,8 +42,10 @@ namespace rsx
{
rsx->sync();
// Write ref+get atomically (get will be written again with the same value at command end)
vm::_ref<atomic_be_t<u64>>(rsx->dma_address + ::offset32(&RsxDmaControl::get)).store(u64{rsx->fifo_ctrl->get_pos()} << 32 | arg);
// Write ref+get (get will be written again with the same value at command end)
auto& dma = vm::_ref<RsxDmaControl>(rsx->dma_address);
dma.get.release(rsx->fifo_ctrl->get_pos());
dma.ref.store(arg);
}
void semaphore_acquire(thread* rsx, u32 /*reg*/, u32 arg)
@ -436,11 +438,11 @@ namespace rsx
if (rsx->m_graphics_state & rsx::pipeline_state::transform_constants_dirty)
{
// Minor optimization: don't compare values if we already know we need invalidation
copy_data_swap_u32(values, vm::base(rsx->fifo_ctrl->get_current_arg_ptr()), rcount);
copy_data_swap_u32(values, static_cast<u32*>(vm::base(rsx->fifo_ctrl->get_current_arg_ptr())), rcount);
}
else
{
if (copy_data_swap_u32_cmp(values, vm::base(rsx->fifo_ctrl->get_current_arg_ptr()), rcount))
if (copy_data_swap_u32_cmp(values, static_cast<u32*>(vm::base(rsx->fifo_ctrl->get_current_arg_ptr())), rcount))
{
// Transform constants invalidation is expensive (~8k bytes per update)
rsx->m_graphics_state |= rsx::pipeline_state::transform_constants_dirty;
@ -472,7 +474,7 @@ namespace rsx
rcount -= max - (max_vertex_program_instructions * 4);
}
copy_data_swap_u32(&rsx::method_registers.transform_program[load_pos * 4 + index % 4], vm::base(rsx->fifo_ctrl->get_current_arg_ptr()), rcount);
copy_data_swap_u32(&rsx::method_registers.transform_program[load_pos * 4 + index % 4], static_cast<u32*>(vm::base(rsx->fifo_ctrl->get_current_arg_ptr())), rcount);
rsx->m_graphics_state |= rsx::pipeline_state::vertex_program_ucode_dirty;
rsx::method_registers.transform_program_load_set(load_pos + ((rcount + index % 4) / 4));