Performance diagnostics and small wins for title 1

- eden-cli: SDL called SDL_AppIterate back to back, keeping a host core at
  100% polling input; run the main callbacks at 240 Hz.
- EDEN_NX2_IDLE_SPIN_US: optional spin before an idle core blocks (measured:
  no gain on title 1, so off by default).
- EDEN_NX2_SVC_TRACE=<first>:<last>: trace SVCs (with thread ids) for a range
  of presents only.
- MME DMA reads flush GPU-written memory first (never triggered on title 1,
  kept for correctness); render-enable override histogram instead of the
  first-400 log.

Measured on title 1's title screen (RTX 4070 Laptop): GPU ~20% busy, the six
emulated cores ~0.9 host cores in total, Eden's GPU thread 45-70% busy and
largely waiting: on host fences for semaphore acquires (GPU accuracy High
delays every release to host completion), on ASTC upload Finish(), and in a
1 ms poll sleep. gpu_accuracy=Low: ~12.5 -> ~16 fps.

1 files changed, 17 insertions, 5 deletions

diff --git a/src/core/hle/kernel/svc.cpp b/src/core/hle/kernel/svc.cpp
index e458a5057d..53d964ecbc 100644
--- a/src/core/hle/kernel/svc.cpp
+++ b/src/core/hle/kernel/svc.cpp
@@ -7,6 +7,7 @@
 // DO NOT MODIFY IT MANUALLY
 
 #include <cstdlib>
+#include <cstring>
 #include <type_traits>
 
 #include "core/arm/arm_interface.h"
@@ -2506,10 +2507,20 @@ void Call(Core::System& system, u32 imm) {
     LOG_TRACE(Kernel_SVC, "{} [0]={:#x} [1]={:#x} [2]={:#x} [3]={:#x} [4]={:#x} [5]={:#x} [6]={:#x}",
         imm, GetArg64(args, 0), GetArg64(args, 1), GetArg64(args, 2),
         GetArg64(args, 3), GetArg64(args, 4), GetArg64(args, 5), GetArg64(args, 6));
-    // Set EDEN_NX2_SVC_TRACE=1 to log every SVC with its arguments and results.
-    static const bool svc_trace = std::getenv("EDEN_NX2_SVC_TRACE") != nullptr;
+    // Set EDEN_NX2_SVC_TRACE=1 to log every SVC with its arguments and results, or
+    // EDEN_NX2_SVC_TRACE=<first>:<last> for presents first..last only (a frame-time profile).
+    static const std::pair<u64, u64> trace_range = [] {
+        const char* text = std::getenv("EDEN_NX2_SVC_TRACE");
+        if (text == nullptr) return std::pair<u64, u64>{1, 0};
+        const char* colon = std::strchr(text, ':');
+        if (colon == nullptr) return std::pair<u64, u64>{0, ~u64{}};
+        return std::pair<u64, u64>{std::strtoull(text, nullptr, 10), std::strtoull(colon + 1, nullptr, 10)};
+    }();
+    const u64 presents = Core::g_nx2_present_count.load(std::memory_order_relaxed);
+    const bool svc_trace = presents >= trace_range.first && presents <= trace_range.second;
     if (svc_trace) {
-        LOG_INFO(Kernel_SVC, "svc {:#04x} in  [0]={:#x} [1]={:#x} [2]={:#x} [3]={:#x}", imm,
+        LOG_INFO(Kernel_SVC, "svc {:#04x} in  t{} [0]={:#x} [1]={:#x} [2]={:#x} [3]={:#x}", imm,
+                 GetCurrentThread(kernel).GetThreadId(),
                  GetArg64(args, 0), GetArg64(args, 1), GetArg64(args, 2), GetArg64(args, 3));
     }
     if (imm == 0x80) {
@@ -2546,8 +2557,9 @@ void Call(Core::System& system, u32 imm) {
         Call32(system, imm, args);
     }
     if (svc_trace) {
-        LOG_INFO(Kernel_SVC, "svc {:#04x} out [0]={:#x} [1]={:#x} [2]={:#x}", imm,
-                 GetArg64(args, 0), GetArg64(args, 1), GetArg64(args, 2));
+        LOG_INFO(Kernel_SVC, "svc {:#04x} out t{} [0]={:#x} [1]={:#x} [2]={:#x}", imm,
+                 GetCurrentThread(kernel).GetThreadId(), GetArg64(args, 0), GetArg64(args, 1),
+                 GetArg64(args, 2));
     }
     //kernel.ExitSVCProfile();
     kernel.CurrentPhysicalCore().LoadSvcArguments(process, args);