Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[HIP] fix op not support AMD GPU bug, the flag PADDLE_WITH_ROCM is in… #35394

Merged
merged 3 commits into from
Sep 7, 2021
Merged

[HIP] fix op not support AMD GPU bug, the flag PADDLE_WITH_ROCM is in… #35394

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
382b18f
[HIP] fix op not support AMD GPU bug, the flag PADDLE_WITH_ROCM is in…
xymyeah Sep 2, 2021
2c56ee1
[HIP] fix op not support AMD GPU bug, the flag PADDLE_WITH_ROCM is in…
xymyeah Sep 2, 2021
c0a90b6
[HIP] fix op not support AMD GPU bug
xymyeah Sep 3, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion paddle/fluid/operators/memcpy_d2h_op.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,7 +131,7 @@ REGISTER_OP_CPU_KERNEL_FUNCTOR(memcpy_d2h, float, ops::MemcpyD2HKernel, double,
ops::MemcpyD2HKernel, plat::float16,
ops::MemcpyD2HKernel);

#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_ROCM)
#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
REGISTER_OP_CUDA_KERNEL_FUNCTOR(memcpy_d2h, float, ops::MemcpyD2HKernel, double,
ops::MemcpyD2HKernel, int, ops::MemcpyD2HKernel,
int64_t, ops::MemcpyD2HKernel, bool,
Expand Down
2 changes: 1 addition & 1 deletion paddle/fluid/operators/memcpy_h2d_op.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,7 +131,7 @@ REGISTER_OP_CPU_KERNEL_FUNCTOR(memcpy_h2d, float, ops::MemcpyH2DKernel, double,
ops::MemcpyH2DKernel, plat::float16,
ops::MemcpyH2DKernel);

#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_ROCM)
#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
REGISTER_OP_CUDA_KERNEL_FUNCTOR(memcpy_h2d, float, ops::MemcpyH2DKernel, double,
ops::MemcpyH2DKernel, int, ops::MemcpyH2DKernel,
int64_t, ops::MemcpyH2DKernel, bool,
Expand Down
2 changes: 1 addition & 1 deletion paddle/fluid/operators/memcpy_op.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -141,7 +141,7 @@ REGISTER_OP_CPU_KERNEL_FUNCTOR(memcpy, float, ops::MemcpyKernel, double,
ops::MemcpyKernel, plat::float16,
ops::MemcpyKernel);

#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_ROCM)
#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
REGISTER_OP_CUDA_KERNEL_FUNCTOR(memcpy, float, ops::MemcpyKernel, double,
ops::MemcpyKernel, int, ops::MemcpyKernel,
int64_t, ops::MemcpyKernel, bool,
Expand Down
57 changes: 55 additions & 2 deletions python/paddle/fluid/tests/unittests/test_memcpy_op.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,7 @@ def get_prog(self):
})
return main_program, gpu_var, pinned_var

def test_gpu_cpoy_to_pinned(self):
def test_gpu_copy_to_pinned(self):
main_program, gpu_var, pinned_var = self.get_prog()
main_program.global_block().append_op(
type='memcpy',
Expand All @@ -79,7 +79,7 @@ def test_gpu_cpoy_to_pinned(self):
self.assertTrue(np.allclose(gpu_, pinned_))
self.assertTrue(np.allclose(pinned_, np.ones((10, 10))))

def test_pinned_cpoy_gpu(self):
def test_pinned_copy_gpu(self):
main_program, gpu_var, pinned_var = self.get_prog()
main_program.global_block().append_op(
type='memcpy',
Expand All @@ -94,6 +94,59 @@ def test_pinned_cpoy_gpu(self):
self.assertTrue(np.allclose(gpu_, pinned_))
self.assertTrue(np.allclose(gpu_, np.zeros((10, 10))))

def test_hip_copy_bool_value(self):
if core.is_compiled_with_rocm():
paddle.enable_static()
main_program = Program()
with program_guard(main_program):
pinned_var_name = "tensor@Pinned"
gpu_var_name = "tensor@GPU"
pinned_var = main_program.global_block().create_var(
name=pinned_var_name,
shape=[1],
dtype='bool',
persistable=False,
stop_gradient=True)
gpu_var = main_program.global_block().create_var(
name=gpu_var_name,
shape=[1],
dtype='bool',
persistable=False,
stop_gradient=True)
main_program.global_block().append_op(
type="fill_constant",
outputs={"Out": gpu_var_name},
attrs={
"shape": [1],
"dtype": gpu_var.dtype,
"value": False,
"place_type": 1
})
main_program.global_block().append_op(
type="fill_constant",
outputs={"Out": pinned_var_name},
attrs={
"shape": [1],
"dtype": gpu_var.dtype,
"value": True,
"place_type": 2
})

main_program.global_block().append_op(
type='memcpy',
inputs={'X': pinned_var},
outputs={'Out': gpu_var},
attrs={'dst_place_type': 1})
place = fluid.CUDAPlace(0)
exe = fluid.Executor(place)
gpu_, pinned_ = exe.run(main_program,
feed={},
fetch_list=[gpu_var.name, pinned_var.name])
expect_value = np.array([1]).astype('bool')
self.assertTrue(np.array_equal(gpu_, expect_value))
else:
pass


class TestMemcpyOPError(unittest.TestCase):
def get_prog(self):
Expand Down