nx2: get title 4 past the difficulty confirm

- MME: implement MME_DMA_READ / MME_DMA_READ_FIFOED (Turing macros read query
  results into their load FIFO; the conditional-render macro at 0x142 needs it).
- vi2 present waits up to 4 s for the frame fence (was 250 ms) so shader
  compile stalls don't let the game lap its equal-value gate semaphores.
- Native QMD: fire the third release slot at word 52, which NVN2 patches for
  queue-to-queue syncs (compute queue releasing a gate the 3D queue waits on).
- Compute class SET_REPORT_SEMAPHORE (0x6c0-0x6c3) releases.
- Ampere lowering: AOFFI, any LOP3 truth table, STG RZ, LDG 0x981, IABS,
  I2FP rounding, IMAD negated addend, SHFL 0x989 (all 77 title-4 shaders lower).
- Debug: EDEN_NX2_SEM_WATCH=<addr> traces acquires, releases, method
  arguments and compute macro emits for one semaphore block;
  EDEN_NX2_KEY_SCRIPT scripted key presses; per-submit logs moved to TRACE.

2 files changed, 11 insertions, 3 deletions

diff --git a/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp b/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp
index cf3948c321..2a144cc7ab 100644
--- a/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp
+++ b/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp
@@ -421,7 +421,7 @@ NvResult nvhost_gpu::SubmitGPFIFOImpl(IoctlSubmitGpfifo& params, Tegra::CommandL
 
     if (g_nx2_driver) {
         for (const auto& entry : entries.command_lists) {
-            LOG_DEBUG(Service_NVDRV, "nx2 GPFIFO entry raw={:#018x} addr={:#x} size={:#x}", entry.raw,
+            LOG_TRACE(Service_NVDRV, "nx2 GPFIFO entry raw={:#018x} addr={:#x} size={:#x}", entry.raw,
                      entry.addr.Value(), entry.size.Value());
         }
         static const bool check = std::getenv("EDEN_NX2_PB_CHECK") != nullptr;
@@ -471,7 +471,7 @@ NvResult nvhost_gpu::SubmitGPFIFOImpl(IoctlSubmitGpfifo& params, Tegra::CommandL
     }
 
     if (g_nx2_driver) {
-        LOG_DEBUG(Service_NVDRV,
+        LOG_TRACE(Service_NVDRV,
                  "nx2 SubmitGPFIFO entries={} flags={:#x} fence_out=({}, {}) guest_value={}",
                  params.num_entries, flags.raw, params.fence.id, params.fence.value,
                  system.Host1x().GetSyncpointManager().GetGuestSyncpointValue(params.fence.id));
diff --git a/src/core/hle/service/nx2_gsv.h b/src/core/hle/service/nx2_gsv.h
index 8073e0bd96..89f3a9fd50 100644
--- a/src/core/hle/service/nx2_gsv.h
+++ b/src/core/hle/service/nx2_gsv.h
@@ -580,17 +580,25 @@ private:
     // free, so pace the game here instead: wait (bounded, in case the GPU waits on the CPU) until
     // Eden's GPU thread has processed the frame. Otherwise the game laps the GPU thread and
     // rewrites command memory it hasn't read yet.
+    u32 m_wait_timeouts{};
+
     void WaitForFrame(u32 fence_id, u32 fence_value) {
         if (fence_id >= 192) {
             return;
         }
         auto& syncpoints = system.Host1x().GetSyncpointManager();
         const auto start = std::chrono::steady_clock::now();
-        const auto deadline = start + std::chrono::milliseconds{250};
+        // Long enough to cover shader/pipeline compile stalls: a game that gets more than a few
+        // command buffers ahead misses the equal-value gates its other channels acquire.
+        const auto deadline = start + std::chrono::milliseconds{4000};
         while (!syncpoints.IsReadyGuest(fence_id, fence_value) &&
                std::chrono::steady_clock::now() < deadline && system.IsPoweredOn()) {
             std::this_thread::sleep_for(std::chrono::microseconds{200});
         }
+        if (!syncpoints.IsReadyGuest(fence_id, fence_value) && m_wait_timeouts++ < 64) {
+            LOG_WARNING(Service, "nx2:vi2 present #{} fence ({}, {}) not reached in 4 s (now {})",
+                        m_presents, fence_id, fence_value, syncpoints.GetGuestSyncpointValue(fence_id));
+        }
         if (m_presents <= 8 || (m_presents % 600) == 0) {
             LOG_INFO(Service, "nx2:vi2 present #{} fence ({}, {}) now {}, waited {} us",
                      m_presents, fence_id, fence_value, syncpoints.GetGuestSyncpointValue(fence_id),