diff --git a/dpctl/_backend.pxd b/dpctl/_backend.pxd
index ab4af7d136..b4788ab493 100644
--- a/dpctl/_backend.pxd
+++ b/dpctl/_backend.pxd
@@ -227,8 +227,8 @@ cdef extern from "syclinterface/dpctl_sycl_device_selector_interface.h":
 cdef extern from "syclinterface/dpctl_sycl_event_interface.h":
     cdef DPCTLSyclEventRef DPCTLEvent_Create()
     cdef DPCTLSyclEventRef DPCTLEvent_Copy(const DPCTLSyclEventRef ERef)
-    cdef void DPCTLEvent_Wait(DPCTLSyclEventRef ERef)
-    cdef void DPCTLEvent_WaitAndThrow(DPCTLSyclEventRef ERef)
+    cdef void DPCTLEvent_Wait(DPCTLSyclEventRef ERef) nogil
+    cdef void DPCTLEvent_WaitAndThrow(DPCTLSyclEventRef ERef) nogil
     cdef void DPCTLEvent_Delete(DPCTLSyclEventRef ERef)
     cdef _event_status_type DPCTLEvent_GetCommandExecutionStatus(DPCTLSyclEventRef ERef)
     cdef _backend_type DPCTLEvent_GetBackend(DPCTLSyclEventRef ERef)
@@ -356,7 +356,7 @@ cdef extern from "syclinterface/dpctl_sycl_queue_interface.h":
         size_t NDims,
         const DPCTLSyclEventRef *DepEvents,
         size_t NDepEvents)
-    cdef void DPCTLQueue_Wait(const DPCTLSyclQueueRef QRef)
+    cdef void DPCTLQueue_Wait(const DPCTLSyclQueueRef QRef) nogil
     cdef DPCTLSyclEventRef DPCTLQueue_Memcpy(
         const DPCTLSyclQueueRef Q,
         void *Dest,
@@ -394,23 +394,25 @@ cdef extern from "syclinterface/dpctl_sycl_queue_manager.h":
 cdef extern from "syclinterface/dpctl_sycl_usm_interface.h":
     cdef DPCTLSyclUSMRef DPCTLmalloc_shared(
         size_t size,
-        DPCTLSyclQueueRef QRef)
+        DPCTLSyclQueueRef QRef) nogil
     cdef DPCTLSyclUSMRef DPCTLmalloc_host(
         size_t size,
-        DPCTLSyclQueueRef QRef)
-    cdef DPCTLSyclUSMRef DPCTLmalloc_device(size_t size, DPCTLSyclQueueRef QRef)
+        DPCTLSyclQueueRef QRef) nogil
+    cdef DPCTLSyclUSMRef DPCTLmalloc_device(
+        size_t size,
+        DPCTLSyclQueueRef QRef) nogil
     cdef DPCTLSyclUSMRef DPCTLaligned_alloc_shared(
         size_t alignment,
         size_t size,
-        DPCTLSyclQueueRef QRef)
+        DPCTLSyclQueueRef QRef) nogil
     cdef DPCTLSyclUSMRef DPCTLaligned_alloc_host(
         size_t alignment,
         size_t size,
-        DPCTLSyclQueueRef QRef)
+        DPCTLSyclQueueRef QRef) nogil
     cdef DPCTLSyclUSMRef DPCTLaligned_alloc_device(
         size_t alignment,
         size_t size,
-        DPCTLSyclQueueRef QRef)
+        DPCTLSyclQueueRef QRef) nogil
     cdef void DPCTLfree_with_queue(
         DPCTLSyclUSMRef MRef,
         DPCTLSyclQueueRef QRef)
diff --git a/dpctl/_sycl_event.pyx b/dpctl/_sycl_event.pyx
index 5dfb8ee10b..12f7c376b2 100644
--- a/dpctl/_sycl_event.pyx
+++ b/dpctl/_sycl_event.pyx
@@ -98,7 +98,7 @@ cdef class _SyclEvent:
 
     def __dealloc__(self):
         if (self._event_ref):
-            DPCTLEvent_Wait(self._event_ref)
+            with nogil: DPCTLEvent_Wait(self._event_ref)
             DPCTLEvent_Delete(self._event_ref)
         self._event_ref = NULL
         self.args = None
@@ -224,7 +224,7 @@ cdef class SyclEvent(_SyclEvent):
 
     @staticmethod
     cdef void _wait(SyclEvent event):
-        DPCTLEvent_WaitAndThrow(event._event_ref)
+        with nogil: DPCTLEvent_WaitAndThrow(event._event_ref)
 
     @staticmethod
     def wait_for(event):
@@ -370,4 +370,4 @@ cdef class SyclEvent(_SyclEvent):
 
     cpdef void wait(self):
         "Synchronously wait for completion of this event."
-        DPCTLEvent_Wait(self._event_ref)
+        with nogil: DPCTLEvent_Wait(self._event_ref)
diff --git a/dpctl/_sycl_queue.pyx b/dpctl/_sycl_queue.pyx
index 614b7b2ae8..05bd2f753b 100644
--- a/dpctl/_sycl_queue.pyx
+++ b/dpctl/_sycl_queue.pyx
@@ -824,7 +824,7 @@ cdef class SyclQueue(_SyclQueue):
         return SyclEvent._create(Eref, args)
 
     cpdef void wait(self):
-        DPCTLQueue_Wait(self._queue_ref)
+        with nogil: DPCTLQueue_Wait(self._queue_ref)
 
     cpdef memcpy(self, dest, src, size_t count):
         cdef void *c_dest
@@ -846,7 +846,7 @@ cdef class SyclQueue(_SyclQueue):
             raise RuntimeError(
                 "SyclQueue.memcpy operation encountered an error"
             )
-        DPCTLEvent_Wait(ERef)
+        with nogil: DPCTLEvent_Wait(ERef)
         DPCTLEvent_Delete(ERef)
 
     cpdef prefetch(self, mem, size_t count=0):
@@ -866,7 +866,7 @@ cdef class SyclQueue(_SyclQueue):
             raise RuntimeError(
                 "SyclQueue.prefetch encountered an error"
             )
-        DPCTLEvent_Wait(ERef)
+        with nogil: DPCTLEvent_Wait(ERef)
         DPCTLEvent_Delete(ERef)
 
     cpdef mem_advise(self, mem, size_t count, int advice):
@@ -886,7 +886,7 @@ cdef class SyclQueue(_SyclQueue):
             raise RuntimeError(
                 "SyclQueue.mem_advise operation encountered an error"
             )
-        DPCTLEvent_Wait(ERef)
+        with nogil: DPCTLEvent_Wait(ERef)
         DPCTLEvent_Delete(ERef)
 
     @property
diff --git a/dpctl/memory/_memory.pyx b/dpctl/memory/_memory.pyx
index 93f69ce548..741d72e22d 100644
--- a/dpctl/memory/_memory.pyx
+++ b/dpctl/memory/_memory.pyx
@@ -94,7 +94,7 @@ cdef void copy_via_host(void *dest_ptr, SyclQueue dest_queue,
         src_ptr,
         nbytes
     )
-    DPCTLEvent_Wait(E1Ref)
+    with nogil: DPCTLEvent_Wait(E1Ref)
 
     E2Ref = DPCTLQueue_Memcpy(
         dest_queue.get_queue_ref(),
@@ -102,7 +102,7 @@ cdef void copy_via_host(void *dest_ptr, SyclQueue dest_queue,
         <void *>&host_buf[0],
         nbytes
     )
-    DPCTLEvent_Wait(E2Ref)
+    with nogil: DPCTLEvent_Wait(E2Ref)
     DPCTLEvent_Delete(E1Ref)
     DPCTLEvent_Delete(E2Ref)
 
@@ -142,6 +142,7 @@ cdef class _Memory:
     cdef _cinit_alloc(self, Py_ssize_t alignment, Py_ssize_t nbytes,
                       bytes ptr_type, SyclQueue queue):
         cdef DPCTLSyclUSMRef p = NULL
+        cdef DPCTLSyclQueueRef QRef = NULL
 
         self._cinit_empty()
 
@@ -149,27 +150,28 @@ cdef class _Memory:
             if queue is None:
                 queue = dpctl.SyclQueue()
 
+            QRef = queue.get_queue_ref()
             if (ptr_type == b"shared"):
                 if alignment > 0:
-                    p = DPCTLaligned_alloc_shared(
-                        alignment, nbytes, queue.get_queue_ref()
+                    with nogil: p = DPCTLaligned_alloc_shared(
+                        alignment, nbytes, QRef
                     )
                 else:
-                    p = DPCTLmalloc_shared(nbytes, queue.get_queue_ref())
+                    with nogil: p = DPCTLmalloc_shared(nbytes, QRef)
             elif (ptr_type == b"host"):
                 if alignment > 0:
-                    p = DPCTLaligned_alloc_host(
-                        alignment, nbytes, queue.get_queue_ref()
+                    with nogil: p = DPCTLaligned_alloc_host(
+                        alignment, nbytes, QRef
                     )
                 else:
-                    p = DPCTLmalloc_host(nbytes, queue.get_queue_ref())
+                    with nogil: p = DPCTLmalloc_host(nbytes, QRef)
             elif (ptr_type == b"device"):
                 if (alignment > 0):
-                    p = DPCTLaligned_alloc_device(
-                        alignment, nbytes, queue.get_queue_ref()
+                    with nogil: p = DPCTLaligned_alloc_device(
+                        alignment, nbytes, QRef
                     )
                 else:
-                    p = DPCTLmalloc_device(nbytes, queue.get_queue_ref())
+                    with nogil: p = DPCTLmalloc_device(nbytes, QRef)
             else:
                 raise RuntimeError(
                     "Pointer type is unknown: {}".format(
@@ -398,7 +400,7 @@ cdef class _Memory:
             <void *>self.memory_ptr,  # source
             <size_t>self.nbytes
         )
-        DPCTLEvent_Wait(ERef)
+        with nogil: DPCTLEvent_Wait(ERef)
         DPCTLEvent_Delete(ERef)
 
         return obj
@@ -423,7 +425,7 @@ cdef class _Memory:
             <void *>&host_buf[0],     # source
             <size_t>buf_len
         )
-        DPCTLEvent_Wait(ERef)
+        with nogil: DPCTLEvent_Wait(ERef)
         DPCTLEvent_Delete(ERef)
 
     cpdef copy_from_device(self, object sycl_usm_ary):
@@ -465,7 +467,7 @@ cdef class _Memory:
                     <void *>src_buf.p,
                     <size_t>src_buf.nbytes
                 )
-                DPCTLEvent_Wait(ERef)
+                with nogil: DPCTLEvent_Wait(ERef)
                 DPCTLEvent_Delete(ERef)
             else:
                 copy_via_host(