diff --git a/openmp/libomptarget/include/omptargetplugin.h b/openmp/libomptarget/include/omptargetplugin.h
--- a/openmp/libomptarget/include/omptargetplugin.h
+++ b/openmp/libomptarget/include/omptargetplugin.h
@@ -14,7 +14,7 @@
#ifndef _OMPTARGETPLUGIN_H_
#define _OMPTARGETPLUGIN_H_
-#include <omptarget.h>
+#include "omptarget.h"
#ifdef __cplusplus
extern "C" {
diff --git a/openmp/libomptarget/src/device.cpp b/openmp/libomptarget/src/device.cpp
--- a/openmp/libomptarget/src/device.cpp
+++ b/openmp/libomptarget/src/device.cpp
@@ -497,12 +497,12 @@
int32_t DeviceTy::runRegion(void *TgtEntryPtr, void **TgtVarsPtr,
ptrdiff_t *TgtOffsets, int32_t TgtVarsSize,
AsyncInfoTy &AsyncInfo) {
- if (!RTL->run_region || !RTL->synchronize)
- return RTL->run_region(RTLDeviceID, TgtEntryPtr, TgtVarsPtr, TgtOffsets,
- TgtVarsSize);
+ if (!RTL->run_target_region || !RTL->synchronize)
+ return RTL->run_target_region(RTLDeviceID, TgtEntryPtr, TgtVarsPtr,
+ TgtOffsets, TgtVarsSize);
else
- return RTL->run_region_async(RTLDeviceID, TgtEntryPtr, TgtVarsPtr,
- TgtOffsets, TgtVarsSize, AsyncInfo);
+ return RTL->run_target_region_async(RTLDeviceID, TgtEntryPtr, TgtVarsPtr,
+ TgtOffsets, TgtVarsSize, AsyncInfo);
}
// Run region on device
@@ -519,14 +519,14 @@
int32_t NumTeams, int32_t ThreadLimit,
uint64_t LoopTripCount,
AsyncInfoTy &AsyncInfo) {
- if (!RTL->run_team_region_async || !RTL->synchronize)
- return RTL->run_team_region(RTLDeviceID, TgtEntryPtr, TgtVarsPtr,
- TgtOffsets, TgtVarsSize, NumTeams, ThreadLimit,
- LoopTripCount);
+ if (!RTL->run_target_team_region_async || !RTL->synchronize)
+ return RTL->run_target_team_region(RTLDeviceID, TgtEntryPtr, TgtVarsPtr,
+ TgtOffsets, TgtVarsSize, NumTeams,
+ ThreadLimit, LoopTripCount);
else
- return RTL->run_team_region_async(RTLDeviceID, TgtEntryPtr, TgtVarsPtr,
- TgtOffsets, TgtVarsSize, NumTeams,
- ThreadLimit, LoopTripCount, AsyncInfo);
+ return RTL->run_target_team_region_async(
+ RTLDeviceID, TgtEntryPtr, TgtVarsPtr, TgtOffsets, TgtVarsSize, NumTeams,
+ ThreadLimit, LoopTripCount, AsyncInfo);
}
// Whether data can be copied to DstDevice directly
diff --git a/openmp/libomptarget/src/rtl.h b/openmp/libomptarget/src/rtl.h
--- a/openmp/libomptarget/src/rtl.h
+++ b/openmp/libomptarget/src/rtl.h
@@ -13,7 +13,9 @@
#ifndef _OMPTARGET_RTL_H
#define _OMPTARGET_RTL_H
+#include "dlwrap.h"
#include "omptarget.h"
+#include "omptargetplugin.h"
#include <list>
#include <map>
#include <mutex>
@@ -25,43 +27,41 @@
struct __tgt_bin_desc;
struct RTLInfoTy {
- typedef int32_t(is_valid_binary_ty)(void *);
- typedef int32_t(is_data_exchangable_ty)(int32_t, int32_t);
- typedef int32_t(number_of_devices_ty)();
- typedef int32_t(init_device_ty)(int32_t);
+
+#ifdef RTL_MAKE_TYPEDEF
+#error "RTL_MAKE_TYPEDEF already defined"
+#endif
+#define RTL_MAKE_TYPEDEF(SYMBOL) \
+ using SYMBOL##_ty = dlwrap::trait<decltype(&__tgt_rtl_##SYMBOL)>::FunctionType
+ RTL_MAKE_TYPEDEF(is_valid_binary);
+ RTL_MAKE_TYPEDEF(is_data_exchangable);
+ RTL_MAKE_TYPEDEF(number_of_devices);
+ RTL_MAKE_TYPEDEF(init_device);
typedef __tgt_target_table *(load_binary_ty)(int32_t, void *);
- typedef void *(data_alloc_ty)(int32_t, int64_t, void *, int32_t);
- typedef int32_t(data_submit_ty)(int32_t, void *, void *, int64_t);
- typedef int32_t(data_submit_async_ty)(int32_t, void *, void *, int64_t,
- __tgt_async_info *);
- typedef int32_t(data_retrieve_ty)(int32_t, void *, void *, int64_t);
- typedef int32_t(data_retrieve_async_ty)(int32_t, void *, void *, int64_t,
- __tgt_async_info *);
- typedef int32_t(data_exchange_ty)(int32_t, void *, int32_t, void *, int64_t);
- typedef int32_t(data_exchange_async_ty)(int32_t, void *, int32_t, void *,
- int64_t, __tgt_async_info *);
- typedef int32_t(data_delete_ty)(int32_t, void *);
- typedef int32_t(run_region_ty)(int32_t, void *, void **, ptrdiff_t *,
- int32_t);
- typedef int32_t(run_region_async_ty)(int32_t, void *, void **, ptrdiff_t *,
- int32_t, __tgt_async_info *);
- typedef int32_t(run_team_region_ty)(int32_t, void *, void **, ptrdiff_t *,
- int32_t, int32_t, int32_t, uint64_t);
- typedef int32_t(run_team_region_async_ty)(int32_t, void *, void **,
- ptrdiff_t *, int32_t, int32_t,
- int32_t, uint64_t,
- __tgt_async_info *);
- typedef int64_t(init_requires_ty)(int64_t);
- typedef int32_t(synchronize_ty)(int32_t, __tgt_async_info *);
+ RTL_MAKE_TYPEDEF(data_alloc);
+ RTL_MAKE_TYPEDEF(data_submit);
+ RTL_MAKE_TYPEDEF(data_submit_async);
+ RTL_MAKE_TYPEDEF(data_retrieve);
+ RTL_MAKE_TYPEDEF(data_retrieve_async);
+ RTL_MAKE_TYPEDEF(data_exchange);
+ RTL_MAKE_TYPEDEF(data_exchange_async);
+ RTL_MAKE_TYPEDEF(data_delete);
+ RTL_MAKE_TYPEDEF(run_target_region);
+ RTL_MAKE_TYPEDEF(run_target_region_async);
+ RTL_MAKE_TYPEDEF(run_target_team_region);
+ RTL_MAKE_TYPEDEF(run_target_team_region_async);
+ RTL_MAKE_TYPEDEF(init_requires);
+ RTL_MAKE_TYPEDEF(synchronize);
typedef int32_t (*register_lib_ty)(__tgt_bin_desc *);
- typedef int32_t(supports_empty_images_ty)();
- typedef void(print_device_info_ty)(int32_t);
- typedef void(set_info_flag_ty)(uint32_t);
- typedef int32_t(create_event_ty)(int32_t, void **);
- typedef int32_t(record_event_ty)(int32_t, void *, __tgt_async_info *);
- typedef int32_t(wait_event_ty)(int32_t, void *, __tgt_async_info *);
- typedef int32_t(sync_event_ty)(int32_t, void *);
- typedef int32_t(destroy_event_ty)(int32_t, void *);
+ RTL_MAKE_TYPEDEF(supports_empty_images);
+ RTL_MAKE_TYPEDEF(print_device_info);
+ RTL_MAKE_TYPEDEF(set_info_flag);
+ RTL_MAKE_TYPEDEF(create_event);
+ RTL_MAKE_TYPEDEF(record_event);
+ RTL_MAKE_TYPEDEF(wait_event);
+ RTL_MAKE_TYPEDEF(sync_event);
+ RTL_MAKE_TYPEDEF(destroy_event);
+#undef RTL_MAKE_TYPEDEF
int32_t Idx = -1; // RTL index, index is the number of devices
// of other RTLs that were registered before,
@@ -76,35 +76,35 @@
#endif
// Functions implemented in the RTL.
- is_valid_binary_ty *is_valid_binary = nullptr;
- is_data_exchangable_ty *is_data_exchangable = nullptr;
- number_of_devices_ty *number_of_devices = nullptr;
- init_device_ty *init_device = nullptr;
+ is_valid_binary_ty is_valid_binary = nullptr;
+ is_data_exchangable_ty is_data_exchangable = nullptr;
+ number_of_devices_ty number_of_devices = nullptr;
+ init_device_ty init_device = nullptr;
load_binary_ty *load_binary = nullptr;
- data_alloc_ty *data_alloc = nullptr;
- data_submit_ty *data_submit = nullptr;
- data_submit_async_ty *data_submit_async = nullptr;
- data_retrieve_ty *data_retrieve = nullptr;
- data_retrieve_async_ty *data_retrieve_async = nullptr;
- data_exchange_ty *data_exchange = nullptr;
- data_exchange_async_ty *data_exchange_async = nullptr;
- data_delete_ty *data_delete = nullptr;
- run_region_ty *run_region = nullptr;
- run_region_async_ty *run_region_async = nullptr;
- run_team_region_ty *run_team_region = nullptr;
- run_team_region_async_ty *run_team_region_async = nullptr;
- init_requires_ty *init_requires = nullptr;
- synchronize_ty *synchronize = nullptr;
+ data_alloc_ty data_alloc = nullptr;
+ data_submit_ty data_submit = nullptr;
+ data_submit_async_ty data_submit_async = nullptr;
+ data_retrieve_ty data_retrieve = nullptr;
+ data_retrieve_async_ty data_retrieve_async = nullptr;
+ data_exchange_ty data_exchange = nullptr;
+ data_exchange_async_ty data_exchange_async = nullptr;
+ data_delete_ty data_delete = nullptr;
+ run_target_region_ty run_target_region = nullptr;
+ run_target_region_async_ty run_target_region_async = nullptr;
+ run_target_team_region_ty run_target_team_region = nullptr;
+ run_target_team_region_async_ty run_target_team_region_async = nullptr;
+ init_requires_ty init_requires = nullptr;
+ synchronize_ty synchronize = nullptr;
register_lib_ty register_lib = nullptr;
register_lib_ty unregister_lib = nullptr;
- supports_empty_images_ty *supports_empty_images = nullptr;
- set_info_flag_ty *set_info_flag = nullptr;
- print_device_info_ty *print_device_info = nullptr;
- create_event_ty *create_event = nullptr;
- record_event_ty *record_event = nullptr;
- wait_event_ty *wait_event = nullptr;
- sync_event_ty *sync_event = nullptr;
- destroy_event_ty *destroy_event = nullptr;
+ supports_empty_images_ty supports_empty_images = nullptr;
+ set_info_flag_ty set_info_flag = nullptr;
+ print_device_info_ty print_device_info = nullptr;
+ create_event_ty create_event = nullptr;
+ record_event_ty record_event = nullptr;
+ wait_event_ty wait_event = nullptr;
+ sync_event_ty sync_event = nullptr;
+ destroy_event_ty destroy_event = nullptr;
// Are there images associated with this RTL.
bool isUsed = false;
diff --git a/openmp/libomptarget/src/rtl.cpp b/openmp/libomptarget/src/rtl.cpp
--- a/openmp/libomptarget/src/rtl.cpp
+++ b/openmp/libomptarget/src/rtl.cpp
@@ -65,6 +65,15 @@
#endif
}
+template <typename T> static T CastDlsym(void *Handle, const char *Symbol) {
+ // dlsym returns void*, which can be cast to a function pointer in posix
+ // but not in c++, thus static_cast is an error and we need reinterpret
+ // or (c-style)
+ return reinterpret_cast<T>(dlsym(Handle, Symbol));
+}
+#define TypedDlsym(HANDLE, SYMBOL) \
+ CastDlsym<dlwrap::trait<decltype(&SYMBOL)>::FunctionType>(HANDLE, #SYMBOL)
+
void RTLsTy::LoadRTLs() {
// Parse environment variable OMP_TARGET_OFFLOAD (if set)
PM->TargetOffloadPolicy =
@@ -96,35 +105,30 @@
bool ValidPlugin = true;
- if (!(*((void **)&R.is_valid_binary) =
- dlsym(dynlib_handle, "__tgt_rtl_is_valid_binary")))
+ if (!(R.is_valid_binary =
+ TypedDlsym(dynlib_handle, __tgt_rtl_is_valid_binary)))
ValidPlugin = false;
- if (!(*((void **)&R.number_of_devices) =
- dlsym(dynlib_handle, "__tgt_rtl_number_of_devices")))
+ if (!(R.number_of_devices =
+ TypedDlsym(dynlib_handle, __tgt_rtl_number_of_devices)))
ValidPlugin = false;
- if (!(*((void **)&R.init_device) =
- dlsym(dynlib_handle, "__tgt_rtl_init_device")))
+ if (!(R.init_device = TypedDlsym(dynlib_handle, __tgt_rtl_init_device)))
ValidPlugin = false;
if (!(*((void **)&R.load_binary) =
dlsym(dynlib_handle, "__tgt_rtl_load_binary")))
ValidPlugin = false;
- if (!(*((void **)&R.data_alloc) =
- dlsym(dynlib_handle, "__tgt_rtl_data_alloc")))
+ if (!(R.data_alloc = TypedDlsym(dynlib_handle, __tgt_rtl_data_alloc)))
ValidPlugin = false;
- if (!(*((void **)&R.data_submit) =
- dlsym(dynlib_handle, "__tgt_rtl_data_submit")))
+ if (!(R.data_submit = TypedDlsym(dynlib_handle, __tgt_rtl_data_submit)))
ValidPlugin = false;
- if (!(*((void **)&R.data_retrieve) =
- dlsym(dynlib_handle, "__tgt_rtl_data_retrieve")))
+ if (!(R.data_retrieve = TypedDlsym(dynlib_handle, __tgt_rtl_data_retrieve)))
ValidPlugin = false;
- if (!(*((void **)&R.data_delete) =
- dlsym(dynlib_handle, "__tgt_rtl_data_delete")))
+ if (!(R.data_delete = TypedDlsym(dynlib_handle, __tgt_rtl_data_delete)))
ValidPlugin = false;
- if (!(*((void **)&R.run_region) =
- dlsym(dynlib_handle, "__tgt_rtl_run_target_region")))
+ if (!(R.run_target_region =
+ TypedDlsym(dynlib_handle, __tgt_rtl_run_target_region)))
ValidPlugin = false;
- if (!(*((void **)&R.run_team_region) =
- dlsym(dynlib_handle, "__tgt_rtl_run_target_team_region")))
+ if (!(R.run_target_team_region =
+ TypedDlsym(dynlib_handle, __tgt_rtl_run_target_team_region)))
ValidPlugin = false;
// Invalid plugin
@@ -152,41 +156,36 @@
R.NumberOfDevices);
// Optional functions
- *((void **)&R.init_requires) =
- dlsym(dynlib_handle, "__tgt_rtl_init_requires");
- *((void **)&R.data_submit_async) =
- dlsym(dynlib_handle, "__tgt_rtl_data_submit_async");
- *((void **)&R.data_retrieve_async) =
- dlsym(dynlib_handle, "__tgt_rtl_data_retrieve_async");
- *((void **)&R.run_region_async) =
- dlsym(dynlib_handle, "__tgt_rtl_run_target_region_async");
- *((void **)&R.run_team_region_async) =
- dlsym(dynlib_handle, "__tgt_rtl_run_target_team_region_async");
- *((void **)&R.synchronize) = dlsym(dynlib_handle, "__tgt_rtl_synchronize");
- *((void **)&R.data_exchange) =
- dlsym(dynlib_handle, "__tgt_rtl_data_exchange");
- *((void **)&R.data_exchange_async) =
- dlsym(dynlib_handle, "__tgt_rtl_data_exchange_async");
- *((void **)&R.is_data_exchangable) =
- dlsym(dynlib_handle, "__tgt_rtl_is_data_exchangable");
+ R.init_requires = TypedDlsym(dynlib_handle, __tgt_rtl_init_requires);
+ R.data_submit_async =
+ TypedDlsym(dynlib_handle, __tgt_rtl_data_submit_async);
+ R.data_retrieve_async =
+ TypedDlsym(dynlib_handle, __tgt_rtl_data_retrieve_async);
+ R.run_target_region_async =
+ TypedDlsym(dynlib_handle, __tgt_rtl_run_target_region_async);
+ R.run_target_team_region_async =
+ TypedDlsym(dynlib_handle, __tgt_rtl_run_target_team_region_async);
+ R.synchronize = TypedDlsym(dynlib_handle, __tgt_rtl_synchronize);
+ R.data_exchange = TypedDlsym(dynlib_handle, __tgt_rtl_data_exchange);
+ R.data_exchange_async =
+ TypedDlsym(dynlib_handle, __tgt_rtl_data_exchange_async);
+ R.is_data_exchangable =
+ TypedDlsym(dynlib_handle, __tgt_rtl_is_data_exchangable);
*((void **)&R.register_lib) =
dlsym(dynlib_handle, "__tgt_rtl_register_lib");
*((void **)&R.unregister_lib) =
dlsym(dynlib_handle, "__tgt_rtl_unregister_lib");
- *((void **)&R.supports_empty_images) =
- dlsym(dynlib_handle, "__tgt_rtl_supports_empty_images");
- *((void **)&R.set_info_flag) =
- dlsym(dynlib_handle, "__tgt_rtl_set_info_flag");
- *((void **)&R.print_device_info) =
- dlsym(dynlib_handle, "__tgt_rtl_print_device_info");
- *((void **)&R.create_event) =
- dlsym(dynlib_handle, "__tgt_rtl_create_event");
- *((void **)&R.record_event) =
- dlsym(dynlib_handle, "__tgt_rtl_record_event");
- *((void **)&R.wait_event) = dlsym(dynlib_handle, "__tgt_rtl_wait_event");
- *((void **)&R.sync_event) = dlsym(dynlib_handle, "__tgt_rtl_sync_event");
- *((void **)&R.destroy_event) =
- dlsym(dynlib_handle, "__tgt_rtl_destroy_event");
+
+ R.supports_empty_images =
+ TypedDlsym(dynlib_handle, __tgt_rtl_supports_empty_images);
+ R.set_info_flag = TypedDlsym(dynlib_handle, __tgt_rtl_set_info_flag);
+ R.print_device_info =
+ TypedDlsym(dynlib_handle, __tgt_rtl_print_device_info);
+ R.create_event = TypedDlsym(dynlib_handle, __tgt_rtl_create_event);
+ R.record_event = TypedDlsym(dynlib_handle, __tgt_rtl_record_event);
+ R.wait_event = TypedDlsym(dynlib_handle, __tgt_rtl_wait_event);
+ R.sync_event = TypedDlsym(dynlib_handle, __tgt_rtl_sync_event);
+ R.destroy_event = TypedDlsym(dynlib_handle, __tgt_rtl_destroy_event);
}
DP("RTLs loaded!\n");