Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commitb20b51c

Browse files
yushangdifacebook-github-bot
authored andcommitted
Fix too big to optimize in test, actually use O0 when aot_inductor.compile_wrapper_with_O0 is set (#148714)
Summary:1. Check against the "0" char instead2. We got the following error when using anything other than O0 flag: `error: Function ZN5torch12aot_inductorL22__check_inputs_outputsEPP16AtenTensorOpaqueS3 is too big to optimize [-Werror,-Wignored-optimization-argument]` So we use O0 flag in wrapper code when `aot_inductor.compile_wrapper_opt_level` is set to `O0`.Test Plan:``` buck run 'fbcode//mode/opt' fbcode//deeplearning/aot_inductor/cpu/test:ads_second_stage_dsnn_models_aoti_lowering_test -- -r AdsSecondStageDSNNModelsAOTILoweringTest```Reviewed By: desertfireDifferential Revision: D70670957
1 parent1e37e5b commitb20b51c

File tree

5 files changed

+8
-45
lines changed

5 files changed

+8
-45
lines changed

‎test/inductor/test_aot_inductor.py‎

Lines changed: 0 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -139,24 +139,6 @@ def forward(self, x, y):
139139
model,example_inputs,"AOTInductorModelRunMinimalArrayrefInterface(",1
140140
)
141141

142-
deftest_compile_wrapper_with_O0(self):
143-
classModel(torch.nn.Module):
144-
def__init__(self)->None:
145-
super().__init__()
146-
self.linear=torch.nn.Linear(10,10)
147-
148-
defforward(self,x,y):
149-
returnx+self.linear(y)
150-
151-
example_inputs= (
152-
torch.randn(10,10,device=self.device),
153-
torch.randn(10,10,device=self.device),
154-
)
155-
model=Model()
156-
withconfig.patch("aot_inductor.compile_wrapper_with_O0",True):
157-
self.check_model(model,example_inputs)
158-
self.code_check_count(model,example_inputs,"__attribute__((",2)
159-
160142
deftest_small_constant(self):
161143
classModel(torch.nn.Module):
162144
def__init__(self)->None:

‎torch/_inductor/codegen/cpp_wrapper_cpu.py‎

Lines changed: 2 additions & 12 deletions
Original file line numberDiff line numberDiff line change
@@ -407,7 +407,7 @@ def gen_check(handle_kind, idx, name, tensor):
407407
"""
408408
bool _check_aoti_runtime_check_inputs_env() {
409409
const static char* env_var_value = getenv("AOTI_RUNTIME_CHECK_INPUTS");
410-
const static bool result = env_var_value != nullptr && env_var_value[0] !=0;
410+
const static bool result = env_var_value != nullptr && env_var_value[0] !='0';
411411
return result;
412412
}
413413
@@ -461,17 +461,7 @@ def write_wrapper_decl(self):
461461
"""
462462
)
463463

464-
run_impl_proto=""
465-
ifconfig.aot_inductor.compile_wrapper_with_O0:
466-
run_impl_proto+="""
467-
#ifdef __clang__
468-
__attribute__((optnone))
469-
#else
470-
__attribute__((optimize("O0")))
471-
#endif
472-
"""
473-
474-
run_impl_proto+="""
464+
run_impl_proto="""
475465
void AOTInductorModel::run_impl(
476466
AtenTensorHandle*
477467
input_handles, // array of input AtenTensorHandle; handles

‎torch/_inductor/codegen/cpp_wrapper_cpu_array_ref.py‎

Lines changed: 1 addition & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -189,17 +189,7 @@ def write_wrapper_decl(self):
189189
"""
190190
)
191191

192-
run_impl_proto=""
193-
ifconfig.aot_inductor.compile_wrapper_with_O0:
194-
run_impl_proto+="""
195-
#ifdef __clang__
196-
__attribute__((optnone))
197-
#else
198-
__attribute__((optimize("O0")))
199-
#endif
200-
"""
201-
202-
run_impl_proto+="""
192+
run_impl_proto="""
203193
void AOTInductorModel::run_impl(
204194
AtenTensorHandle*
205195
input_handles, // array of input AtenTensorHandle; handles

‎torch/_inductor/config.py‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1168,9 +1168,9 @@ class aot_inductor:
11681168
debug_compile=os.environ.get("AOT_INDUCTOR_DEBUG_COMPILE","0")=="1"
11691169

11701170
# Annotate generated main wrapper function, i.e. AOTInductorModel::run_impl,
1171-
# toskipcpp compileroptimizations for faster compilation.
1172-
compile_wrapper_with_O0= (
1173-
os.environ.get("AOT_INDUCTOR_COMPILE_WRAPPER_WITH_O0","0")=="1"
1171+
# touse whichcpp compileroptimization level, default to O1
1172+
compile_wrapper_opt_level=os.environ.get(
1173+
"AOT_INDUCTOR_COMPILE_WRAPPER_OPT_LEVEL","O1"
11741174
)
11751175

11761176
# option for debug printing/saving for intermediate tensor values for aot inductor

‎torch/_inductor/cpp_builder.py‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -542,10 +542,11 @@ def _get_optimization_cflags(
542542
if_IS_WINDOWS:
543543
return ["O1"ifmin_optimizeelse"O2"]
544544
else:
545+
wrapper_opt_level=config.aot_inductor.compile_wrapper_opt_level
545546
cflags= (
546547
["O0","g"]
547548
ifconfig.aot_inductor.debug_compile
548-
else ["O1"ifmin_optimizeelse"O3","DNDEBUG"]
549+
else [wrapper_opt_levelifmin_optimizeelse"O3","DNDEBUG"]
549550
)
550551
cflags+=_get_ffast_math_flags()
551552
cflags.append("fno-finite-math-only")

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp