maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[OpenMP] Part 2 of At present, amdgpu plugin merges both asynchronous 

and synchronous kernel launch implementations into a single
synchronous version.  This patch prepares the plugin for asynchronous
implementation by:

    Privatizing actual kernel launch code (valid in both cases) into
    an anonymous namespace base function (submitted at D115267)

    - Separating the control flow path of asynchronous and synchronous
      kernel launch functions** (this diff)

Reviewed By: JonChesterfield

Differential Revision: https://reviews.llvm.org/D115273
This commit is contained in:
Carlo Bertolli 2021-12-10 19:10:53 +00:00
parent 2d1761bffc
commit 28309c5436
1 changed files with 24 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
openmp/libomptarget/plugins/amdgpu/src/rtl.cpp
View File

@ -1042,10 +1042,10 @@ static uint64_t acquire_available_packet_id(hsa_queue_t *queue) {
return packet_id; return packet_id;
} }
int32_t __tgt_rtl_run_target_team_region_locked( int32_t runRegionLocked(int32_t device_id, void *tgt_entry_ptr, void **tgt_args,
int32_t device_id, void *tgt_entry_ptr, void **tgt_args, ptrdiff_t *tgt_offsets, int32_t arg_num,
ptrdiff_t *tgt_offsets, int32_t arg_num, int32_t num_teams, int32_t num_teams, int32_t thread_limit,
int32_t thread_limit, uint64_t loop_tripcount) { uint64_t loop_tripcount) {
// Set the context we are using // Set the context we are using
// update thread limit content in gpu memory if un-initialized or specified // update thread limit content in gpu memory if un-initialized or specified
// from host // from host
@ -2218,9 +2218,9 @@ int32_t __tgt_rtl_run_target_team_region(int32_t device_id, void *tgt_entry_ptr,
uint64_t loop_tripcount) { uint64_t loop_tripcount) {
DeviceInfo.load_run_lock.lock_shared(); DeviceInfo.load_run_lock.lock_shared();
int32_t res = __tgt_rtl_run_target_team_region_locked( int32_t res =
device_id, tgt_entry_ptr, tgt_args, tgt_offsets, arg_num, num_teams, runRegionLocked(device_id, tgt_entry_ptr, tgt_args, tgt_offsets, arg_num,
thread_limit, loop_tripcount); num_teams, thread_limit, loop_tripcount);
DeviceInfo.load_run_lock.unlock_shared(); DeviceInfo.load_run_lock.unlock_shared();
return res; return res;
@ -2238,6 +2238,20 @@ int32_t __tgt_rtl_run_target_region(int32_t device_id, void *tgt_entry_ptr,
thread_limit, 0); thread_limit, 0);
} }
int32_t __tgt_rtl_run_target_team_region_async(
int32_t device_id, void *tgt_entry_ptr, void **tgt_args,
ptrdiff_t *tgt_offsets, int32_t arg_num, int32_t num_teams,
int32_t thread_limit, uint64_t loop_tripcount) {
DeviceInfo.load_run_lock.lock_shared();
int32_t res =
runRegionLocked(device_id, tgt_entry_ptr, tgt_args, tgt_offsets, arg_num,
num_teams, thread_limit, loop_tripcount);
DeviceInfo.load_run_lock.unlock_shared();
return res;
}
int32_t __tgt_rtl_run_target_region_async(int32_t device_id, int32_t __tgt_rtl_run_target_region_async(int32_t device_id,
void *tgt_entry_ptr, void **tgt_args, void *tgt_entry_ptr, void **tgt_args,
ptrdiff_t *tgt_offsets, ptrdiff_t *tgt_offsets,
@ -2250,9 +2264,9 @@ int32_t __tgt_rtl_run_target_region_async(int32_t device_id,
// fix thread num // fix thread num
int32_t team_num = 1; int32_t team_num = 1;
int32_t thread_limit = 0; // use default int32_t thread_limit = 0; // use default
return __tgt_rtl_run_target_team_region(device_id, tgt_entry_ptr, tgt_args, return __tgt_rtl_run_target_team_region_async(device_id, tgt_entry_ptr,
tgt_offsets, arg_num, team_num, tgt_args, tgt_offsets, arg_num,
thread_limit, 0); team_num, thread_limit, 0);
} }
int32_t __tgt_rtl_synchronize(int32_t device_id, __tgt_async_info *AsyncInfo) { int32_t __tgt_rtl_synchronize(int32_t device_id, __tgt_async_info *AsyncInfo) {