Skip to content

[SYCL] Propagate attributes from transitive calls to kernel #1878

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

Merged
merged 11 commits into from
Jul 29, 2020
Merged

[SYCL] Propagate attributes from transitive calls to kernel #1878

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
fd4591c
[SYCL] Fixing comment about propagation of attributes to kernel
schittir Jun 19, 2020
6b5d4a9
[SYCL] Fixing comment about propagation of attributes to kernel
schittir Jun 19, 2020
3857f54
[SYCL] Fixing comment about propagation of attributes to kernel
schittir Jun 19, 2020
ef2deea
Propagate attributes from transitive calls to kernel
schittir Jul 10, 2020
3426ef0
Correct failing lit tests
schittir Jul 10, 2020
acb7a65
Minor changes to address comments and resolve branch merge conflicts
schittir Jul 13, 2020
eb3a32e
Fix formatting issues reported by clang-format-check bot
schittir Jul 14, 2020
49c747c
Add diagnostics for conflicting attributes from transitive calls
schittir Jul 20, 2020
e3e79b1
Fix formatting
schittir Jul 21, 2020
882b3d0
Fix lit test failure caused by newer commits to repo
schittir Jul 24, 2020
1735df4
Fix the bug causing lit test failure
schittir Jul 27, 2020
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
Prev Previous commit
Next Next commit
Correct failing lit tests
  • Loading branch information
@schittir
schittir committed Jul 27, 2020
commit 3426ef0fc17cfb7d604da655fe41ef1f4fe11c2b
7 changes: 5 additions & 2 deletions clang/test/SemaSYCL/check-notdirect-attribute-propagation.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
// RUN: %clang_cc1 -std=c++11 -fsyntax-only -fsycl -fsycl-is-device -verify %s
// RUN: %clang %s -fsyntax-only -Xclang -ast-dump -fsycl-device-only | FileCheck %s

[[intelfpga::no_global_work_offset]] void not_direct() {} // expected-no-warning
[[intelfpga::no_global_work_offset]] void not_direct() {}

void func() { not_direct(); }

Expand All @@ -16,5 +16,8 @@ void parallel_for(Type lambda) {
}

void invoke_foo2() {
// CHECK-LABEL: FunctionDecl {{.*}} invoke_foo2 'void ()'
// CHECK: `-FunctionDecl {{.*}} _ZTSZ11invoke_foo2vE10KernelName 'void ()'
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would suggest {{.}}KernelName{{.}} Differences in name mangling on Windows will cause issues otherwise.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I agree with this thought, and changed it per your suggestion. However, please note that there are other lit tests following this name mangled pattern, with which I intended to be consistent.

// CHECK: -SYCLIntelNoGlobalWorkOffsetAttr {{.*}} Enabled
parallel_for<class KernelName>([]() {});
}
2 changes: 1 addition & 1 deletion clang/test/SemaSYCL/intel-max-global-work-dim.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ void foo() {

#else // __SYCL_DEVICE_ONLY__

[[intelfpga::max_global_work_dim(2)]] // expected-warning{{'max_global_work_dim' attribute ignored}}
[[intelfpga::max_global_work_dim(2)]]
void func_ignore() {}

struct FuncObj {
Expand Down
2 changes: 1 addition & 1 deletion clang/test/SemaSYCL/intel-max-work-group-size.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ void foo() {

#else // __SYCL_DEVICE_ONLY__

[[intelfpga::max_work_group_size(2, 2, 2)]] // expected-warning{{'max_work_group_size' attribute ignored}}
[[intelfpga::max_work_group_size(2, 2, 2)]]
void func_ignore() {}

struct FuncObj {
Expand Down
2 changes: 1 addition & 1 deletion clang/test/SemaSYCL/intel-restrict.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
// RUN: %clang_cc1 %s -fsyntax-only -fsycl -fsycl-is-device -triple spir64 -DCHECKDIAG -verify
// RUN: %clang_cc1 %s -fsyntax-only -ast-dump -fsycl -fsycl-is-device -triple spir64 | FileCheck %s

[[intel::kernel_args_restrict]] // expected-warning{{'kernel_args_restrict' attribute ignored}}
[[intel::kernel_args_restrict]]
void func_ignore() {}

struct FuncObj {
Expand Down
2 changes: 1 addition & 1 deletion clang/test/SemaSYCL/num_simd_work_items.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ void foo() {

#else // __SYCL_DEVICE_ONLY__

[[intelfpga::num_simd_work_items(2)]] // expected-warning{{'num_simd_work_items' attribute ignored}}
[[intelfpga::num_simd_work_items(2)]]
void func_ignore() {}

struct FuncObj {
Expand Down