Remove cudaMemPrefetchAsync from managed allocator

Eager prefetching was filling HBM+EGM, causing subsequent
cudaMallocManaged calls to fail after model loading. On GH200
with EGM, pages should migrate on-demand via hardware page faults
over C2C NVLink. The cudaMemAdviseSetPreferredLocation(GPU) hint
is sufficient to prefer GPU placement with LPDDR fallback.

managed_alloc.cu

@@ -0,0 +1,89 @@
+// managed_alloc.cu - cudaMallocManaged allocator for PyTorch
+// Compile: nvcc -shared -o libmanaged_alloc.so managed_alloc.cu -Xcompiler -fPIC
+// Compatible with CUDA 13+ (uses cudaMemLocation API)
+//
+// Key design decisions for GH200 EGM:
+//   1. cudaMallocManaged → allocations can page-fault across HBM + EGM
+//   2. cudaMemAdviseSetPreferredLocation(GPU) → driver prefers keeping pages on GPU
+//   3. cudaMemAdviseSetAccessedBy(CPU) → CPU can access over C2C NVLink without
+//      triggering page migration back to system RAM (critical: prevents OOM)
+//   4. NO prefetching — pages migrate on-demand via hardware page faults.
+//      Eager prefetching fills HBM+EGM and causes subsequent allocations
+//      to fail. On-demand migration is the correct behavior for unified
+//      memory with HBM + LPDDR EGM.
+#include <cuda_runtime.h>
+#include <stdio.h>
+
+extern "C" {
+
+// PyTorch pluggable allocator signature: void*(size_t, int, cudaStream_t)
+void* managed_malloc(size_t size, int device, cudaStream_t stream) {
+  void* ptr = nullptr;
+
+  // Set the device before allocating
+  cudaError_t err = cudaSetDevice(device);
+  if (err != cudaSuccess) {
+    fprintf(stderr, "[managed_alloc] cudaSetDevice(%d) failed: %s\n",
+            device, cudaGetErrorString(err));
+    return nullptr;
+  }
+
+  // Use cudaMallocManaged - this is the key: allocations can page-fault
+  // across HBM and LPDDR on GH200 with EGM enabled
+  err = cudaMallocManaged(&ptr, size, cudaMemAttachGlobal);
+  if (err != cudaSuccess) {
+    fprintf(stderr, "[managed_alloc] cudaMallocManaged failed: %s "
+                    "(size=%zu bytes / %.2f GiB)\n",
+            cudaGetErrorString(err), size, (double)size / (1024.0*1024.0*1024.0));
+    return nullptr;
+  }
+
+  // CUDA 13+ uses cudaMemLocation struct instead of int for device
+  cudaMemLocation gpu_loc;
+  gpu_loc.type = cudaMemLocationTypeDevice;
+  gpu_loc.id = device;
+
+  // Advise: prefer GPU placement. On GH200 with EGM, the hardware will
+  // migrate pages as needed, but the driver tries to keep them on GPU.
+  cudaMemAdvise(ptr, size, cudaMemAdviseSetPreferredLocation, gpu_loc);
+
+  // Advise: CPU will access this memory too. On GH200, this sets up
+  // remote mapping over C2C NVLink so CPU can read/write without
+  // triggering page migration back to system RAM. This is CRITICAL
+  // to prevent OOM on EGM systems where most system RAM was carved
+  // out for the GPU.
+  cudaMemLocation cpu_loc;
+  cpu_loc.type = cudaMemLocationTypeHost;
+  cpu_loc.id = cudaCpuDeviceId;
+  cudaMemAdvise(ptr, size, cudaMemAdviseSetAccessedBy, cpu_loc);
+
+  // REMOVED: cudaMemPrefetchAsync — was causing allocation failures after
+  // model loading. Prefetching eagerly migrates ALL pages to GPU, filling
+  // up HBM+EGM. Once physical memory is consumed by prefetched pages, the
+  // next cudaMallocManaged call fails because the driver can't guarantee
+  // page-fault resolution for new allocations.
+  //
+  // On GH200 with EGM, the hardware handles page faults naturally via C2C
+  // NVLink. The cudaMemAdviseSetPreferredLocation(GPU) hint above tells
+  // the driver to prefer GPU placement, but allows fallback to LPDDR when
+  // HBM is full. That's exactly what we want — don't force it.
+  //
+  // Pages will migrate on-demand as they're accessed, which is the correct
+  // behavior for a unified memory system with 96 GiB HBM + 128+ GiB EGM.
+
+  return ptr;
+}
+
+// PyTorch pluggable allocator signature: void(void*, size_t, int, cudaStream_t)
+void managed_free(void* ptr, size_t size, int device, cudaStream_t stream) {
+  if (ptr != nullptr) {
+    // Sync the stream before freeing to avoid use-after-free with
+    // managed memory (in-flight page faults can race with deallocation).
+    if (stream != nullptr) {
+      cudaStreamSynchronize(stream);
+    }
+    cudaFree(ptr);
+  }
+}
+
+} // extern "C"