[Lldb-commits] [lldb] [lldb] Save the edited line before clearing it in Editline::PrintAsync (PR #84154)
DavidSpickett wrote: This failed on our AArch64 bot but due to zombie processes hanging about from other runs, not anything to do with this change specifically. https://github.com/llvm/llvm-project/pull/84154 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
DavidSpickett wrote:
Would it be even safer to error here?
https://github.com/llvm/llvm-project/pull/84860
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Update test past bdbad0d07bb600301cb324e87a6be37ca4af591a (PR #84889)
https://github.com/dklimkin created
https://github.com/llvm/llvm-project/pull/84889
None
>From 04886b73fc9ca7b4a6adc52f5306cb6addc85e5c Mon Sep 17 00:00:00 2001
From: Danial Klimkin
Date: Tue, 12 Mar 2024 11:02:32 +0100
Subject: [PATCH] Update test past bdbad0d07bb600301cb324e87a6be37ca4af591a
---
.../data-formatter/builtin-formats/TestBuiltinFormats.py| 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
a/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
b/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
index 8c3bdabeaac1ba..4d6f44db0195b4 100644
---
a/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
+++
b/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
@@ -308,5 +308,5 @@ def test_pointer(self):
@no_debug_info_test
def test_instruction(self):
self.assertIn(
-" addq 0xa(%rdi), %r8\n", self.getFormatted("instruction",
"0x0a47034c")
+"= addq 0xa(%rdi), %r8\n", self.getFormatted("instruction",
"0x0a47034c")
)
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Update test past bdbad0d07bb600301cb324e87a6be37ca4af591a (PR #84889)
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Danial Klimkin (dklimkin)
Changes
---
Full diff: https://github.com/llvm/llvm-project/pull/84889.diff
1 Files Affected:
- (modified)
lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
(+1-1)
``diff
diff --git
a/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
b/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
index 8c3bdabeaac1ba..4d6f44db0195b4 100644
---
a/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
+++
b/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
@@ -308,5 +308,5 @@ def test_pointer(self):
@no_debug_info_test
def test_instruction(self):
self.assertIn(
-" addq 0xa(%rdi), %r8\n", self.getFormatted("instruction",
"0x0a47034c")
+"= addq 0xa(%rdi), %r8\n", self.getFormatted("instruction",
"0x0a47034c")
)
``
https://github.com/llvm/llvm-project/pull/84889
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Update test past bdbad0d07bb600301cb324e87a6be37ca4af591a (PR #84889)
github-actions[bot] wrote: ⚠️ We detected that you are using a GitHub private e-mail address to contribute to the repo. Please turn off [Keep my email addresses private](https://github.com/settings/emails) setting in your account. See [LLVM Discourse](https://discourse.llvm.org/t/hidden-emails-on-github-should-we-do-something-about-it) for more information. https://github.com/llvm/llvm-project/pull/84889 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
https://github.com/Michael137 created
https://github.com/llvm/llvm-project/pull/84890
None
>From ecb8e2744628207dd0ec72272428e3f57e8d68bb Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:04:12 +
Subject: [PATCH] [lldb][test] TestExprCompletion.py: add tests for completion
of reserved identifiers
---
.../expression/completion/TestExprCompletion.py| 14 ++
.../API/commands/expression/completion/main.cpp| 11 +++
2 files changed, 25 insertions(+)
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index c6a1e3c0f42275..d322b9c323d1ef 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,6 +246,20 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
+self.complete_from_to(
+"expr myVec.__f", "expr myVec.__func()"
+)
+self.complete_from_to(
+"expr myVec._F", "expr myVec._Func()"
+)
+self.complete_from_to(
+"expr myVec.__m", "expr myVec.__mem"
+)
+self.complete_from_to(
+"expr myVec._M", "expr myVec._Mem"
+)
+
+
def test_expr_completion_with_descriptions(self):
self.build()
self.main_source = "main.cpp"
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index 908bebbebff568..b87f9626c93283 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/completion/main.cpp
@@ -1,3 +1,11 @@
+class MyVec {
+int __mem;
+int _Mem;
+public:
+void __func() {}
+void _Func() {}
+};
+
namespace LongNamespaceName { class NestedClass { long m; }; }
// Defined in other.cpp, we only have a forward declaration here.
@@ -31,5 +39,8 @@ int main()
some_expr.FooNumbersBar1();
Expr::StaticMemberMethodBar();
ForwardDecl *fwd_decl_ptr = &fwd_decl;
+MyVec myVec;
+myVec.__func();
+myVec._Func();
return 0; // Break here
}
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
https://github.com/Michael137 edited https://github.com/llvm/llvm-project/pull/84890 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Michael Buch (Michael137)
Changes
Depends on https://github.com/llvm/llvm-project/pull/84891
---
Full diff: https://github.com/llvm/llvm-project/pull/84890.diff
2 Files Affected:
- (modified) lldb/test/API/commands/expression/completion/TestExprCompletion.py
(+14)
- (modified) lldb/test/API/commands/expression/completion/main.cpp (+11)
``diff
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index c6a1e3c0f42275..d322b9c323d1ef 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,6 +246,20 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
+self.complete_from_to(
+"expr myVec.__f", "expr myVec.__func()"
+)
+self.complete_from_to(
+"expr myVec._F", "expr myVec._Func()"
+)
+self.complete_from_to(
+"expr myVec.__m", "expr myVec.__mem"
+)
+self.complete_from_to(
+"expr myVec._M", "expr myVec._Mem"
+)
+
+
def test_expr_completion_with_descriptions(self):
self.build()
self.main_source = "main.cpp"
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index 908bebbebff568..b87f9626c93283 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/completion/main.cpp
@@ -1,3 +1,11 @@
+class MyVec {
+int __mem;
+int _Mem;
+public:
+void __func() {}
+void _Func() {}
+};
+
namespace LongNamespaceName { class NestedClass { long m; }; }
// Defined in other.cpp, we only have a forward declaration here.
@@ -31,5 +39,8 @@ int main()
some_expr.FooNumbersBar1();
Expr::StaticMemberMethodBar();
ForwardDecl *fwd_decl_ptr = &fwd_decl;
+MyVec myVec;
+myVec.__func();
+myVec._Func();
return 0; // Break here
}
``
https://github.com/llvm/llvm-project/pull/84890
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
github-actions[bot] wrote:
:warning: Python code formatter, darker found issues in your code. :warning:
You can test this locally with the following command:
``bash
darker --check --diff -r
6c39fa9e9f198498ff7cf9646081437a0fc0882a...ecb8e2744628207dd0ec72272428e3f57e8d68bb
lldb/test/API/commands/expression/completion/TestExprCompletion.py
``
View the diff from darker here.
``diff
--- TestExprCompletion.py 2024-03-12 10:04:12.00 +
+++ TestExprCompletion.py 2024-03-12 10:08:45.666282 +
@@ -244,23 +244,14 @@
)
self.complete_from_to(
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
-self.complete_from_to(
-"expr myVec.__f", "expr myVec.__func()"
-)
-self.complete_from_to(
-"expr myVec._F", "expr myVec._Func()"
-)
-self.complete_from_to(
-"expr myVec.__m", "expr myVec.__mem"
-)
-self.complete_from_to(
-"expr myVec._M", "expr myVec._Mem"
-)
-
+self.complete_from_to("expr myVec.__f", "expr myVec.__func()")
+self.complete_from_to("expr myVec._F", "expr myVec._Func()")
+self.complete_from_to("expr myVec.__m", "expr myVec.__mem")
+self.complete_from_to("expr myVec._M", "expr myVec._Mem")
def test_expr_completion_with_descriptions(self):
self.build()
self.main_source = "main.cpp"
self.main_source_spec = lldb.SBFileSpec(self.main_source)
``
https://github.com/llvm/llvm-project/pull/84890
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
github-actions[bot] wrote:
:warning: C/C++ code formatter, clang-format found issues in your code.
:warning:
You can test this locally with the following command:
``bash
git-clang-format --diff 6c39fa9e9f198498ff7cf9646081437a0fc0882a
ecb8e2744628207dd0ec72272428e3f57e8d68bb --
lldb/test/API/commands/expression/completion/main.cpp
``
View the diff from clang-format here.
``diff
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index b87f9626c9..ecbb226318 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/completion/main.cpp
@@ -1,9 +1,10 @@
class MyVec {
-int __mem;
-int _Mem;
+ int __mem;
+ int _Mem;
+
public:
-void __func() {}
-void _Func() {}
+ void __func() {}
+ void _Func() {}
};
namespace LongNamespaceName { class NestedClass { long m; }; }
``
https://github.com/llvm/llvm-project/pull/84890
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
https://github.com/Michael137 updated
https://github.com/llvm/llvm-project/pull/84890
>From ecb8e2744628207dd0ec72272428e3f57e8d68bb Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:04:12 +
Subject: [PATCH 1/2] [lldb][test] TestExprCompletion.py: add tests for
completion of reserved identifiers
---
.../expression/completion/TestExprCompletion.py| 14 ++
.../API/commands/expression/completion/main.cpp| 11 +++
2 files changed, 25 insertions(+)
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index c6a1e3c0f42275..d322b9c323d1ef 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,6 +246,20 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
+self.complete_from_to(
+"expr myVec.__f", "expr myVec.__func()"
+)
+self.complete_from_to(
+"expr myVec._F", "expr myVec._Func()"
+)
+self.complete_from_to(
+"expr myVec.__m", "expr myVec.__mem"
+)
+self.complete_from_to(
+"expr myVec._M", "expr myVec._Mem"
+)
+
+
def test_expr_completion_with_descriptions(self):
self.build()
self.main_source = "main.cpp"
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index 908bebbebff568..b87f9626c93283 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/completion/main.cpp
@@ -1,3 +1,11 @@
+class MyVec {
+int __mem;
+int _Mem;
+public:
+void __func() {}
+void _Func() {}
+};
+
namespace LongNamespaceName { class NestedClass { long m; }; }
// Defined in other.cpp, we only have a forward declaration here.
@@ -31,5 +39,8 @@ int main()
some_expr.FooNumbersBar1();
Expr::StaticMemberMethodBar();
ForwardDecl *fwd_decl_ptr = &fwd_decl;
+MyVec myVec;
+myVec.__func();
+myVec._Func();
return 0; // Break here
}
>From 68364e2c6e20fbe6b57672d2cb4f07d6fcf3f34e Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:11:59 +
Subject: [PATCH 2/2] fixup! python format
---
.../expression/completion/TestExprCompletion.py | 17 -
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index d322b9c323d1ef..022b9436ee8ea6 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,19 +246,10 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
-self.complete_from_to(
-"expr myVec.__f", "expr myVec.__func()"
-)
-self.complete_from_to(
-"expr myVec._F", "expr myVec._Func()"
-)
-self.complete_from_to(
-"expr myVec.__m", "expr myVec.__mem"
-)
-self.complete_from_to(
-"expr myVec._M", "expr myVec._Mem"
-)
-
+self.complete_from_to("expr myVec.__f", "expr myVec.__func()")
+self.complete_from_to("expr myVec._F", "expr myVec._Func()")
+self.complete_from_to("expr myVec.__m", "expr myVec.__mem")
+self.complete_from_to("expr myVec._M", "expr myVec._Mem")
def test_expr_completion_with_descriptions(self):
self.build()
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Update test past bdbad0d07bb600301cb324e87a6be37ca4af591a (PR #84889)
https://github.com/dklimkin closed https://github.com/llvm/llvm-project/pull/84889 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] ce1fd92 - Update test past bdbad0d07bb600301cb324e87a6be37ca4af591a (#84889)
Author: Danial Klimkin
Date: 2024-03-12T11:19:48+01:00
New Revision: ce1fd9281707c2163728085d126ff83041e1db51
URL:
https://github.com/llvm/llvm-project/commit/ce1fd9281707c2163728085d126ff83041e1db51
DIFF:
https://github.com/llvm/llvm-project/commit/ce1fd9281707c2163728085d126ff83041e1db51.diff
LOG: Update test past bdbad0d07bb600301cb324e87a6be37ca4af591a (#84889)
Added:
Modified:
lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
Removed:
diff --git
a/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
b/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
index 8c3bdabeaac1ba..4d6f44db0195b4 100644
---
a/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
+++
b/lldb/test/API/functionalities/data-formatter/builtin-formats/TestBuiltinFormats.py
@@ -308,5 +308,5 @@ def test_pointer(self):
@no_debug_info_test
def test_instruction(self):
self.assertIn(
-" addq 0xa(%rdi), %r8\n", self.getFormatted("instruction",
"0x0a47034c")
+"= addq 0xa(%rdi), %r8\n", self.getFormatted("instruction",
"0x0a47034c")
)
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
https://github.com/Michael137 updated
https://github.com/llvm/llvm-project/pull/84890
>From ecb8e2744628207dd0ec72272428e3f57e8d68bb Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:04:12 +
Subject: [PATCH 1/3] [lldb][test] TestExprCompletion.py: add tests for
completion of reserved identifiers
---
.../expression/completion/TestExprCompletion.py| 14 ++
.../API/commands/expression/completion/main.cpp| 11 +++
2 files changed, 25 insertions(+)
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index c6a1e3c0f42275..d322b9c323d1ef 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,6 +246,20 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
+self.complete_from_to(
+"expr myVec.__f", "expr myVec.__func()"
+)
+self.complete_from_to(
+"expr myVec._F", "expr myVec._Func()"
+)
+self.complete_from_to(
+"expr myVec.__m", "expr myVec.__mem"
+)
+self.complete_from_to(
+"expr myVec._M", "expr myVec._Mem"
+)
+
+
def test_expr_completion_with_descriptions(self):
self.build()
self.main_source = "main.cpp"
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index 908bebbebff568..b87f9626c93283 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/completion/main.cpp
@@ -1,3 +1,11 @@
+class MyVec {
+int __mem;
+int _Mem;
+public:
+void __func() {}
+void _Func() {}
+};
+
namespace LongNamespaceName { class NestedClass { long m; }; }
// Defined in other.cpp, we only have a forward declaration here.
@@ -31,5 +39,8 @@ int main()
some_expr.FooNumbersBar1();
Expr::StaticMemberMethodBar();
ForwardDecl *fwd_decl_ptr = &fwd_decl;
+MyVec myVec;
+myVec.__func();
+myVec._Func();
return 0; // Break here
}
>From 68364e2c6e20fbe6b57672d2cb4f07d6fcf3f34e Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:11:59 +
Subject: [PATCH 2/3] fixup! python format
---
.../expression/completion/TestExprCompletion.py | 17 -
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index d322b9c323d1ef..022b9436ee8ea6 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,19 +246,10 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
-self.complete_from_to(
-"expr myVec.__f", "expr myVec.__func()"
-)
-self.complete_from_to(
-"expr myVec._F", "expr myVec._Func()"
-)
-self.complete_from_to(
-"expr myVec.__m", "expr myVec.__mem"
-)
-self.complete_from_to(
-"expr myVec._M", "expr myVec._Mem"
-)
-
+self.complete_from_to("expr myVec.__f", "expr myVec.__func()")
+self.complete_from_to("expr myVec._F", "expr myVec._Func()")
+self.complete_from_to("expr myVec.__m", "expr myVec.__mem")
+self.complete_from_to("expr myVec._M", "expr myVec._Mem")
def test_expr_completion_with_descriptions(self):
self.build()
>From 2961ca8068b2a90999e694ce40f53da80684e5b7 Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:21:20 +
Subject: [PATCH 3/3] fixup! move type into header
---
lldb/test/API/commands/expression/completion/Makefile | 1 +
lldb/test/API/commands/expression/completion/main.cpp | 8 +---
.../API/commands/expression/completion/sys/reserved.h | 7 +++
3 files changed, 9 insertions(+), 7 deletions(-)
create mode 100644 lldb/test/API/commands/expression/completion/sys/reserved.h
diff --git a/lldb/test/API/commands/expression/completion/Makefile
b/lldb/test/API/commands/expression/completion/Makefile
index 020dce7c31d11d..9882622b2189ea 100644
--- a/lldb/test/API/commands/expression/completion/Makefile
+++ b/lldb/test/API/commands/expression/completion/Makefile
@@ -1,3 +1,4 @@
CXX_SOURCES := main.cpp other.cpp
+CXXFLAGS += -isystem $(SRCDIR)/sys
include Makefile.rules
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index b87f9626c93283..5e03805a7a4d5c 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/c
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
https://github.com/Michael137 updated
https://github.com/llvm/llvm-project/pull/84890
>From ecb8e2744628207dd0ec72272428e3f57e8d68bb Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:04:12 +
Subject: [PATCH 1/4] [lldb][test] TestExprCompletion.py: add tests for
completion of reserved identifiers
---
.../expression/completion/TestExprCompletion.py| 14 ++
.../API/commands/expression/completion/main.cpp| 11 +++
2 files changed, 25 insertions(+)
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index c6a1e3c0f42275..d322b9c323d1ef 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,6 +246,20 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
+self.complete_from_to(
+"expr myVec.__f", "expr myVec.__func()"
+)
+self.complete_from_to(
+"expr myVec._F", "expr myVec._Func()"
+)
+self.complete_from_to(
+"expr myVec.__m", "expr myVec.__mem"
+)
+self.complete_from_to(
+"expr myVec._M", "expr myVec._Mem"
+)
+
+
def test_expr_completion_with_descriptions(self):
self.build()
self.main_source = "main.cpp"
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index 908bebbebff568..b87f9626c93283 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/completion/main.cpp
@@ -1,3 +1,11 @@
+class MyVec {
+int __mem;
+int _Mem;
+public:
+void __func() {}
+void _Func() {}
+};
+
namespace LongNamespaceName { class NestedClass { long m; }; }
// Defined in other.cpp, we only have a forward declaration here.
@@ -31,5 +39,8 @@ int main()
some_expr.FooNumbersBar1();
Expr::StaticMemberMethodBar();
ForwardDecl *fwd_decl_ptr = &fwd_decl;
+MyVec myVec;
+myVec.__func();
+myVec._Func();
return 0; // Break here
}
>From 68364e2c6e20fbe6b57672d2cb4f07d6fcf3f34e Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:11:59 +
Subject: [PATCH 2/4] fixup! python format
---
.../expression/completion/TestExprCompletion.py | 17 -
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index d322b9c323d1ef..022b9436ee8ea6 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,19 +246,10 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
-self.complete_from_to(
-"expr myVec.__f", "expr myVec.__func()"
-)
-self.complete_from_to(
-"expr myVec._F", "expr myVec._Func()"
-)
-self.complete_from_to(
-"expr myVec.__m", "expr myVec.__mem"
-)
-self.complete_from_to(
-"expr myVec._M", "expr myVec._Mem"
-)
-
+self.complete_from_to("expr myVec.__f", "expr myVec.__func()")
+self.complete_from_to("expr myVec._F", "expr myVec._Func()")
+self.complete_from_to("expr myVec.__m", "expr myVec.__mem")
+self.complete_from_to("expr myVec._M", "expr myVec._Mem")
def test_expr_completion_with_descriptions(self):
self.build()
>From 2961ca8068b2a90999e694ce40f53da80684e5b7 Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Tue, 12 Mar 2024 10:21:20 +
Subject: [PATCH 3/4] fixup! move type into header
---
lldb/test/API/commands/expression/completion/Makefile | 1 +
lldb/test/API/commands/expression/completion/main.cpp | 8 +---
.../API/commands/expression/completion/sys/reserved.h | 7 +++
3 files changed, 9 insertions(+), 7 deletions(-)
create mode 100644 lldb/test/API/commands/expression/completion/sys/reserved.h
diff --git a/lldb/test/API/commands/expression/completion/Makefile
b/lldb/test/API/commands/expression/completion/Makefile
index 020dce7c31d11d..9882622b2189ea 100644
--- a/lldb/test/API/commands/expression/completion/Makefile
+++ b/lldb/test/API/commands/expression/completion/Makefile
@@ -1,3 +1,4 @@
CXX_SOURCES := main.cpp other.cpp
+CXXFLAGS += -isystem $(SRCDIR)/sys
include Makefile.rules
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index b87f9626c93283..5e03805a7a4d5c 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/c
[Lldb-commits] [lldb] [lldb] Add frame recognizer for __builtin_verbose_trap (PR #80368)
@@ -0,0 +1,98 @@
+#include "lldb/Target/VerboseTrapFrameRecognizer.h"
+
+#include "lldb/Core/Module.h"
+#include "lldb/Symbol/Function.h"
+#include "lldb/Symbol/SymbolContext.h"
+#include "lldb/Target/Process.h"
+#include "lldb/Target/Target.h"
+
+#include "lldb/Utility/LLDBLog.h"
+#include "lldb/Utility/Log.h"
+
+using namespace llvm;
+using namespace lldb;
+using namespace lldb_private;
+
+VerboseTrapRecognizedStackFrame::VerboseTrapRecognizedStackFrame(
+StackFrameSP most_relevant_frame_sp, std::string stop_desc)
+: m_most_relevant_frame(most_relevant_frame_sp) {
+ m_stop_desc = std::move(stop_desc);
+}
+
+lldb::RecognizedStackFrameSP
+VerboseTrapFrameRecognizer::RecognizeFrame(lldb::StackFrameSP frame_sp) {
+ if (frame_sp->GetFrameIndex())
+return {};
+
+ ThreadSP thread_sp = frame_sp->GetThread();
+ ProcessSP process_sp = thread_sp->GetProcess();
+
+ StackFrameSP most_relevant_frame_sp = thread_sp->GetStackFrameAtIndex(1);
+
+ if (!most_relevant_frame_sp) {
+Log *log = GetLog(LLDBLog::Unwind);
+LLDB_LOG(
+log,
+"Failed to find most relevant frame: Hit unwinding bound (1 frame)!");
+return {};
+ }
+
+ SymbolContext sc = frame_sp->GetSymbolContext(eSymbolContextEverything);
+
+ if (!sc.block)
+return {};
+
+ // The runtime error is set as the function name in the inlined function info
+ // of frame #0 by the compiler
+ const InlineFunctionInfo *inline_info = nullptr;
+ Block *inline_block = sc.block->GetContainingInlinedBlock();
+
+ if (!inline_block)
+return {};
+
+ inline_info = sc.block->GetInlinedFunctionInfo();
+
+ if (!inline_info)
+return {};
+
+ auto error_message = inline_info->GetName().GetString();
+ if (error_message.empty())
+return {};
+
+ // Replaces "__llvm_verbose_trap: " with "Runtime Error: "
+ auto space_position = error_message.find(" ");
+ if (space_position == std::string::npos) {
+Log *log = GetLog(LLDBLog::Unwind);
+LLDB_LOGF(log,
+ "Unexpected function name format. Expected ': "
+ "' but got: '%s'.",
+ error_message.c_str());
+
+return {};
+ }
+
+ error_message.replace(0, space_position, "Runtime Error:");
+
+ return lldb::RecognizedStackFrameSP(new VerboseTrapRecognizedStackFrame(
+ most_relevant_frame_sp, std::move(error_message)));
+}
+
+lldb::StackFrameSP VerboseTrapRecognizedStackFrame::GetMostRelevantFrame() {
+ return m_most_relevant_frame;
+}
+
+namespace lldb_private {
+
+void RegisterVerboseTrapFrameRecognizer(Process &process) {
+ RegularExpressionSP module_regex_sp = nullptr;
+ RegularExpressionSP symbol_regex_sp(
+ new RegularExpression("^__llvm_verbose_trap: "));
Michael137 wrote:
Yup that would be nice
I'll leave it to @ahatanak to do it in
https://github.com/llvm/llvm-project/pull/79230 and then use it from LLDB
https://github.com/llvm/llvm-project/pull/80368
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Add frame recognizer for __builtin_verbose_trap (PR #80368)
@@ -0,0 +1,9 @@
+# RUN: %clang_host -g -O0 %S/Inputs/verbose_trap.cpp -o %t.out
+# RUN: %lldb -b -s %s %t.out | FileCheck %s
+run
+# CHECK: thread #{{.*}}stop reason = Runtime Error: Function is not implemented
Michael137 wrote:
How about `Runtime trap: ...`?
https://github.com/llvm/llvm-project/pull/80368
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Add frame recognizer for __builtin_verbose_trap (PR #80368)
https://github.com/Michael137 updated
https://github.com/llvm/llvm-project/pull/80368
>From 4ffcb261af05b2a68781cf353d8e45bf921f8cb7 Mon Sep 17 00:00:00 2001
From: Michael Buch
Date: Thu, 25 Jan 2024 11:05:02 +
Subject: [PATCH 1/6] [lldb] Add frame recognizer for __builtin_verbose_trap
This patch adds a frame recognizer for Clang's
`__builtin_verbose_trap`, which behaves like a
`__builtin_trap`, but emits a failure-reason string
into debug-info in order for debuggers to display
it to a user.
The frame recognizer triggers when we encounter
a frame with a function name that begins with
`__llvm_verbose_trap`, which is the magic prefix
Clang emits into debug-info for verbose traps.
Once such frame is encountered we display the
frame function name as the `Stop Reason` and display
that frame to the user.
Example output:
```
(lldb) run
warning: a.out was compiled with optimization - stepping may behave oddly;
variables may not be available.
Process 35942 launched: 'a.out' (arm64)
Process 35942 stopped
* thread #1, queue = 'com.apple.main-thread', stop reason =
__llvm_verbose_trap: Function is not implemented
frame #1: 0x00013fa4 a.out`main [inlined]
Dummy::func(this=) at verbose_trap.cpp:3:5 [opt]
1struct Dummy {
2 void func() {
-> 3__builtin_verbose_trap("Function is not implemented");
4 }
5};
6
7int main() {
(lldb) bt
* thread #1, queue = 'com.apple.main-thread', stop reason =
__llvm_verbose_trap: Function is not implemented
frame #0: 0x00013fa4 a.out`main [inlined] __llvm_verbose_trap:
Function is not implemented at verbose_trap.cpp:0 [opt]
* frame #1: 0x00013fa4 a.out`main [inlined]
Dummy::func(this=) at verbose_trap.cpp:3:5 [opt]
frame #2: 0x00013fa4 a.out`main at verbose_trap.cpp:8:13 [opt]
frame #3: 0x000189d518b4 dyld`start + 1988
```
---
.../lldb/Target/VerboseTrapFrameRecognizer.h | 39 +
lldb/source/Target/CMakeLists.txt | 1 +
lldb/source/Target/Process.cpp| 2 +
.../Target/VerboseTrapFrameRecognizer.cpp | 85 +++
.../Shell/Recognizer/Inputs/verbose_trap.cpp | 8 ++
lldb/test/Shell/Recognizer/verbose_trap.test | 9 ++
6 files changed, 144 insertions(+)
create mode 100644 lldb/include/lldb/Target/VerboseTrapFrameRecognizer.h
create mode 100644 lldb/source/Target/VerboseTrapFrameRecognizer.cpp
create mode 100644 lldb/test/Shell/Recognizer/Inputs/verbose_trap.cpp
create mode 100644 lldb/test/Shell/Recognizer/verbose_trap.test
diff --git a/lldb/include/lldb/Target/VerboseTrapFrameRecognizer.h
b/lldb/include/lldb/Target/VerboseTrapFrameRecognizer.h
new file mode 100644
index 00..7e045760a28be6
--- /dev/null
+++ b/lldb/include/lldb/Target/VerboseTrapFrameRecognizer.h
@@ -0,0 +1,39 @@
+#ifndef LLDB_TARGET_VERBOSETRAPFRAMERECOGNIZER_H
+#define LLDB_TARGET_VERBOSETRAPFRAMERECOGNIZER_H
+
+#include "lldb/Target/StackFrameRecognizer.h"
+
+namespace lldb_private {
+
+void RegisterVerboseTrapFrameRecognizer(Process &process);
+
+/// Holds the stack frame that caused the Verbose trap and the inlined stop
+/// reason message.
+class VerboseTrapRecognizedStackFrame : public RecognizedStackFrame {
+public:
+ VerboseTrapRecognizedStackFrame(lldb::StackFrameSP most_relevant_frame_sp,
+ std::string stop_desc);
+
+ lldb::StackFrameSP GetMostRelevantFrame() override;
+
+private:
+ lldb::StackFrameSP m_most_relevant_frame;
+};
+
+/// When a thread stops, it checks the current frame contains a
+/// Verbose Trap diagnostic. If so, it returns a \a
+/// VerboseTrapRecognizedStackFrame holding the diagnostic a stop reason
+/// description with and the parent frame as the most relavant frame.
+class VerboseTrapFrameRecognizer : public StackFrameRecognizer {
+public:
+ std::string GetName() override {
+return "Verbose Trap StackFrame Recognizer";
+ }
+
+ lldb::RecognizedStackFrameSP
+ RecognizeFrame(lldb::StackFrameSP frame) override;
+};
+
+} // namespace lldb_private
+
+#endif // LLDB_TARGET_VERBOSETRAPFRAMERECOGNIZER_H
diff --git a/lldb/source/Target/CMakeLists.txt
b/lldb/source/Target/CMakeLists.txt
index cf4818eae3eb8b..8186ccbea27d42 100644
--- a/lldb/source/Target/CMakeLists.txt
+++ b/lldb/source/Target/CMakeLists.txt
@@ -78,6 +78,7 @@ add_lldb_library(lldbTarget
UnixSignals.cpp
UnwindAssembly.cpp
UnwindLLDB.cpp
+ VerboseTrapFrameRecognizer.cpp
LINK_LIBS
lldbBreakpoint
diff --git a/lldb/source/Target/Process.cpp b/lldb/source/Target/Process.cpp
index 6d58873b54a3ad..cd9b48248e4bd9 100644
--- a/lldb/source/Target/Process.cpp
+++ b/lldb/source/Target/Process.cpp
@@ -63,6 +63,7 @@
#include "lldb/Target/ThreadPlanCallFunction.h"
#include "lldb/Target/ThreadPlanStack.h"
#include "lldb/Target/UnixSignals.h"
+#include "lldb/Target/VerboseTrapFrameRecognizer.h"
#include "lldb/Utility/Event.h"
#include "lldb/Utility/LLDBLog.h"
#include "lldb/Utility/Log.h"
@@ -497,6
[Lldb-commits] [lldb] [lldb] Add frame recognizer for __builtin_verbose_trap (PR #80368)
@@ -0,0 +1,39 @@ +#ifndef LLDB_TARGET_VERBOSETRAPFRAMERECOGNIZER_H Michael137 wrote: done https://github.com/llvm/llvm-project/pull/80368 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
https://github.com/rupprecht updated
https://github.com/llvm/llvm-project/pull/84860
>From b5e04eb49c89c8c654305c6ba5db5e2f237bccec Mon Sep 17 00:00:00 2001
From: Jordan Rupprecht
Date: Mon, 11 Mar 2024 11:23:59 -0700
Subject: [PATCH 1/3] [lldb][test] Add `pexpect` category for tests that
`import pexpect`
- Add pexpect category
- Replace skipIfWindows with this
- Make categories apply to test classes too
---
.../Python/lldbsuite/test/decorators.py | 4
lldb/packages/Python/lldbsuite/test/dotest.py | 20 +++
.../Python/lldbsuite/test/lldbpexpect.py | 2 +-
.../Python/lldbsuite/test/test_categories.py | 1 +
.../Python/lldbsuite/test/test_result.py | 10 +-
.../expression/TestExpressionCmd.py | 5 +
.../expression/TestRepeatedExprs.py | 5 +
.../TestFrameVariableResponse.py | 5 +
.../benchmarks/startup/TestStartupDelays.py | 5 +
.../benchmarks/stepping/TestSteppingSpeed.py | 5 +
.../TestCompileRunToBreakpointTurnaround.py | 5 +
.../API/terminal/TestSTTYBeforeAndAfter.py| 2 +-
12 files changed, 34 insertions(+), 35 deletions(-)
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index b691f82b90652c..8e13aa6a13882f 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -409,10 +409,6 @@ def add_test_categories(cat):
cat = test_categories.validate(cat, True)
def impl(func):
-if isinstance(func, type) and issubclass(func, unittest.TestCase):
-raise Exception(
-"@add_test_categories can only be used to decorate a test
method"
-)
try:
if hasattr(func, "categories"):
cat.extend(func.categories)
diff --git a/lldb/packages/Python/lldbsuite/test/dotest.py
b/lldb/packages/Python/lldbsuite/test/dotest.py
index 291d7bad5c0897..268c02e6b49dde 100644
--- a/lldb/packages/Python/lldbsuite/test/dotest.py
+++ b/lldb/packages/Python/lldbsuite/test/dotest.py
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
+)
+
+
def run_suite():
# On MacOS X, check to make sure that domain for com.apple.DebugSymbols
defaults
# does not exist before proceeding to running the test suite.
@@ -1013,6 +1032,7 @@ def run_suite():
checkDebugServerSupport()
checkObjcSupport()
checkForkVForkSupport()
+checkPexpectSupport()
skipped_categories_list = ", ".join(configuration.skip_categories)
print(
diff --git a/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
b/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
index 9d216d90307473..998a080565b6b3 100644
--- a/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
+++ b/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
@@ -10,7 +10,7 @@
@skipIfRemote
-@skipIfWindows # llvm.org/pr22274: need a pexpect replacement for windows
+@add_test_categories(["pexpect"])
class PExpectTest(TestBase):
NO_DEBUG_INFO_TESTCASE = True
PROMPT = "(lldb) "
diff --git a/lldb/packages/Python/lldbsuite/test/test_categories.py
b/lldb/packages/Python/lldbsuite/test/test_categories.py
index 3f8de175e29df3..036bda9c957d11 100644
--- a/lldb/packages/Python/lldbsuite/test/test_categories.py
+++ b/lldb/packages/Python/lldbsuite/test/test_categories.py
@@ -33,6 +33,7 @@
"lldb-server": "Tests related to lldb-server",
"lldb-dap": "Tests for the Debug Adaptor Protocol with lldb-dap",
"llgs": "Tests for the gdb-server functionality of lldb-server",
+"pexpect": "Tests requiring the pexpect library to be available",
"objc": "Tests related to the Objective-C programming language support",
"pyapi": "Tests related to the Python API",
"std-module": "Tests related to importing the std module",
diff --git a/lldb/packages/Python/lldbsuite/test/test_result.py
b/lldb/packages/Python/lldbsuite/test/test_result.py
index 20365f53a67541..cda67ae27d4674 100644
--- a/lldb/packages/Python/lldbsuite/test/test_result.py
+++ b/lldb/packages/Python/lldbsuite/test/test_result.py
@@ -148,9 +148,11 @@ def getCategoriesForTest(self, test):
Gets all the categories for the currently running test meth
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
rupprecht wrote:
For a pexpect test, it would fail either way, but raising an error here would
be more clear about the failure.
For a non-pexpect test, this will unnecessarily fail. `ninja check-lldb-api`
will have ~1k test failures instead of ~30 failures.
Updated to `raise` instead here.
https://github.com/llvm/llvm-project/pull/84860
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
rupprecht wrote:
`checkPexpectSupport()` will be called for every test, just like the other
check* methods above.
This `try` block here will only run if a non-Windows platform is configured to
run `pexpect` tests, either because:
* The test is running w/ various `--skip-category=foo --skip-category=bar`, and
`pexpect` is not one of the skipped categories, or
* The test is running w/ `--category=pexpect` (this is unusual)
I don't think there's a way to only run this check if the test has a certain
category. We don't know what the category is, because the test file isn't
loaded, and loading it would mean potentially evaluating the `import pexpect`
statement that will fail.
The whole `elif` block is just to provide a better message if `pexpect` is not
installed. If I remove it, then running a `pexpect` test will fail like so:
```
$ bin/lldb-dotest -p TestSTTYBeforeAndAfter.py
...
lldb version 19.0.0git (https://github.com/llvm/llvm-project.git revision
b5e04eb49c89c8c654305c6ba5db5e2f237bccec)
clang revision b5e04eb49c89c8c654305c6ba5db5e2f237bccec
llvm revision b5e04eb49c89c8c654305c6ba5db5e2f237bccec
Skipping the following test categories: ['dsym', 'gmodules', 'debugserver',
'objc']
FAIL: LLDB (/home/rupprecht/src/llvm-build/dev/bin/clang-x86_64) ::
test_stty_dash_a_before_and_afetr_invoking_lldb_command
(TestSTTYBeforeAndAfter.TestSTTYBeforeAndAfter.test_stty_dash_a_before_and_afetr_invoking_lldb_command)
==
ERROR: test_stty_dash_a_before_and_afetr_invoking_lldb_command
(TestSTTYBeforeAndAfter.TestSTTYBeforeAndAfter.test_stty_dash_a_before_and_afetr_invoking_lldb_command)
Test that 'stty -a' displays the same output before and after running the
lldb command.
--
Traceback (most recent call last):
File
"/home/rupprecht/src/llvm-project/lldb/packages/Python/lldbsuite/test/decorators.py",
line 446, in wrapper
return func(self, *args, **kwargs)
^^^
File
"/home/rupprecht/src/llvm-project/lldb/test/API/terminal/TestSTTYBeforeAndAfter.py",
line 26, in test_stty_dash_a_before_and_afetr_invoking_lldb_command
import pexpect
ModuleNotFoundError: No module named 'pexpect'
Config=x86_64-/home/rupprecht/src/llvm-build/dev/bin/clang
--
Ran 1 test in 0.011s
FAILED (errors=1)
```
I suppose the error raised by the `elif` block, as written, doesn't provide
much benefit on top of that -- it's clear that the issue is trying to run a
pexpect test w/o pexpect being installed. So maybe I should replace it with
something more actionable, e.g. to suggest configuring cmake w/
`-DLLDB_TEST_USER_ARGS=--skip-category=pexpect`. Or just remove it altogether?
https://github.com/llvm/llvm-project/pull/84860
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] [debugserver] Handle interrupted reads correctly (PR #84872)
https://github.com/JDevlieghere approved this pull request. Makes sense. LGTM. https://github.com/llvm/llvm-project/pull/84872 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
DavidSpickett wrote:
I understand now.
Yeah adding something actionable would be a good improvement, tell them to
install pexpect or to skip the category.
https://github.com/llvm/llvm-project/pull/84860
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
JDevlieghere wrote:
So what would happens without this code? Wouldn't we try to import pexpect
anyway in the relevant test? In other words, do we even need this? It seems
that printing a warning for non-pexpect either adds noise to the already
verbose test output or will be invisible if the test passes.
If the error message is poor enough that we would want to print something more
actionable, could we do it from the `PExpectTest` base class?
https://github.com/llvm/llvm-project/pull/84860
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] 87dc068 - [lldb] [debugserver] Handle interrupted reads correctly (#84872)
Author: Jason Molenda
Date: 2024-03-12T09:03:28-07:00
New Revision: 87dc068280aaddc98acb7865ae3df1d248f4a170
URL:
https://github.com/llvm/llvm-project/commit/87dc068280aaddc98acb7865ae3df1d248f4a170
DIFF:
https://github.com/llvm/llvm-project/commit/87dc068280aaddc98acb7865ae3df1d248f4a170.diff
LOG: [lldb] [debugserver] Handle interrupted reads correctly (#84872)
The first half of this patch is a long-standing annoyance, if I attach
to debugserver with lldb while it is waiting for an lldb connection, the
syscall is interrupted and it doesn't retry, debugserver exits
immediately.
The second half is a request from another tool that is communicating
with debugserver, that we retry reads on our sockets in the same way. I
haven't dug in to the details of how they're communicating that this is
necessary, but the best I've been able to find reading the POSIX API
docs, this is fine.
rdar://117113298
Added:
Modified:
lldb/tools/debugserver/source/RNBSocket.cpp
Removed:
diff --git a/lldb/tools/debugserver/source/RNBSocket.cpp
b/lldb/tools/debugserver/source/RNBSocket.cpp
index 1282ea221625a0..fc55dbf2e1f1b0 100644
--- a/lldb/tools/debugserver/source/RNBSocket.cpp
+++ b/lldb/tools/debugserver/source/RNBSocket.cpp
@@ -120,8 +120,13 @@ rnb_err_t RNBSocket::Listen(const char *listen_host,
uint16_t port,
while (!accept_connection) {
struct kevent event_list[4];
-int num_events =
-kevent(queue_id, events.data(), events.size(), event_list, 4, NULL);
+int num_events;
+do {
+ errno = 0;
+ num_events =
+ kevent(queue_id, events.data(), events.size(), event_list, 4, NULL);
+} while (num_events == -1 &&
+ (errno == EAGAIN || errno == EWOULDBLOCK || errno == EINTR));
if (num_events < 0) {
err.SetError(errno, DNBError::MachKernel);
@@ -291,7 +296,12 @@ rnb_err_t RNBSocket::Read(std::string &p) {
// DNBLogThreadedIf(LOG_RNB_COMM, "%8u RNBSocket::%s calling read()",
// (uint32_t)m_timer.ElapsedMicroSeconds(true), __FUNCTION__);
DNBError err;
- ssize_t bytesread = read(m_fd, buf, sizeof(buf));
+ ssize_t bytesread;
+ do {
+errno = 0;
+bytesread = read(m_fd, buf, sizeof(buf));
+ } while (bytesread == -1 &&
+ (errno == EAGAIN || errno == EWOULDBLOCK || errno == EINTR));
if (bytesread <= 0)
err.SetError(errno, DNBError::POSIX);
else
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] [debugserver] Handle interrupted reads correctly (PR #84872)
https://github.com/jasonmolenda closed https://github.com/llvm/llvm-project/pull/84872 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] 93503aa - Fix MSVC build issues (#84362)
Author: Hiroshi Yamauchi
Date: 2024-03-12T10:26:44-07:00
New Revision: 93503aafcdc66837ecf220243aaa530c05c35895
URL:
https://github.com/llvm/llvm-project/commit/93503aafcdc66837ecf220243aaa530c05c35895
DIFF:
https://github.com/llvm/llvm-project/commit/93503aafcdc66837ecf220243aaa530c05c35895.diff
LOG: Fix MSVC build issues (#84362)
MSVC fails when there is ambiguity (multiple options) around implicit
type conversion operators.
Make ConstString's conversion operator to string_view explicit to avoid
ambiguity with one to StringRef and remove an unused local variable that
MSVC also fails on.
Added:
Modified:
lldb/include/lldb/Utility/ConstString.h
lldb/source/Core/Mangled.cpp
lldb/unittests/SymbolFile/PDB/SymbolFilePDBTests.cpp
Removed:
diff --git a/lldb/include/lldb/Utility/ConstString.h
b/lldb/include/lldb/Utility/ConstString.h
index 470a554ca04869..f7f7ec7605eba4 100644
--- a/lldb/include/lldb/Utility/ConstString.h
+++ b/lldb/include/lldb/Utility/ConstString.h
@@ -168,8 +168,8 @@ class ConstString {
// Implicitly convert \class ConstString instances to \class StringRef.
operator llvm::StringRef() const { return GetStringRef(); }
- // Implicitly convert \class ConstString instances to \class
std::string_view.
- operator std::string_view() const {
+ // Explicitly convert \class ConstString instances to \class
std::string_view.
+ explicit operator std::string_view() const {
return std::string_view(m_string, GetLength());
}
diff --git a/lldb/source/Core/Mangled.cpp b/lldb/source/Core/Mangled.cpp
index 23ae3913093faf..b167c51fdce247 100644
--- a/lldb/source/Core/Mangled.cpp
+++ b/lldb/source/Core/Mangled.cpp
@@ -125,7 +125,7 @@ void Mangled::SetValue(ConstString name) {
}
// Local helpers for
diff erent demangling implementations.
-static char *GetMSVCDemangledStr(std::string_view M) {
+static char *GetMSVCDemangledStr(llvm::StringRef M) {
char *demangled_cstr = llvm::microsoftDemangle(
M, nullptr, nullptr,
llvm::MSDemangleFlags(
@@ -169,27 +169,29 @@ static char *GetItaniumDemangledStr(const char *M) {
return demangled_cstr;
}
-static char *GetRustV0DemangledStr(std::string_view M) {
+static char *GetRustV0DemangledStr(llvm::StringRef M) {
char *demangled_cstr = llvm::rustDemangle(M);
if (Log *log = GetLog(LLDBLog::Demangle)) {
if (demangled_cstr && demangled_cstr[0])
LLDB_LOG(log, "demangled rustv0: {0} -> \"{1}\"", M, demangled_cstr);
else
- LLDB_LOG(log, "demangled rustv0: {0} -> error: failed to demangle", M);
+ LLDB_LOG(log, "demangled rustv0: {0} -> error: failed to demangle",
+ static_cast(M));
}
return demangled_cstr;
}
-static char *GetDLangDemangledStr(std::string_view M) {
+static char *GetDLangDemangledStr(llvm::StringRef M) {
char *demangled_cstr = llvm::dlangDemangle(M);
if (Log *log = GetLog(LLDBLog::Demangle)) {
if (demangled_cstr && demangled_cstr[0])
LLDB_LOG(log, "demangled dlang: {0} -> \"{1}\"", M, demangled_cstr);
else
- LLDB_LOG(log, "demangled dlang: {0} -> error: failed to demangle", M);
+ LLDB_LOG(log, "demangled dlang: {0} -> error: failed to demangle",
+ static_cast(M));
}
return demangled_cstr;
diff --git a/lldb/unittests/SymbolFile/PDB/SymbolFilePDBTests.cpp
b/lldb/unittests/SymbolFile/PDB/SymbolFilePDBTests.cpp
index 6a2ea8c4a41b1c..f237dd63ab1cce 100644
--- a/lldb/unittests/SymbolFile/PDB/SymbolFilePDBTests.cpp
+++ b/lldb/unittests/SymbolFile/PDB/SymbolFilePDBTests.cpp
@@ -527,7 +527,6 @@ TEST_F(SymbolFilePDBTests, TestTypedefs) {
SymbolFilePDB *symfile =
static_cast(module->GetSymbolFile());
llvm::pdb::IPDBSession &session = symfile->GetPDBSession();
- TypeMap results;
const char *TypedefsToCheck[] = {"ClassTypedef", "NSClassTypedef",
"FuncPointerTypedef",
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Fix MSVC build issues (PR #84362)
https://github.com/hjyamauchi closed https://github.com/llvm/llvm-project/pull/84362 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -0,0 +1,88 @@
+//===-- Alarm.h -*- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#ifndef LLDB_HOST_ALARM_H
+#define LLDB_HOST_ALARM_H
+
+#include "lldb/Host/HostThread.h"
+#include "lldb/lldb-types.h"
+#include "llvm/Support/Chrono.h"
+
+namespace lldb_private {
+
+class Alarm {
+public:
+ using Handle = uint64_t;
+ using Callback = std::function;
+ using TimePoint = llvm::sys::TimePoint<>;
+ using Duration = std::chrono::milliseconds;
+
+ Alarm(Duration timeout, bool run_callback_on_exit = false);
+ ~Alarm();
+
+ Handle Create(Callback callback);
adrian-prantl wrote:
Maybe in the toplevel doxygen comment explain what the expected flow is, ie,
that one Alarm can deal with multiple handles and what the expected behavior is?
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -0,0 +1,88 @@
+//===-- Alarm.h -*- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#ifndef LLDB_HOST_ALARM_H
+#define LLDB_HOST_ALARM_H
+
+#include "lldb/Host/HostThread.h"
+#include "lldb/lldb-types.h"
+#include "llvm/Support/Chrono.h"
+
+namespace lldb_private {
+
+class Alarm {
adrian-prantl wrote:
Can you add Doxygen comment explaining what this is to be used for?
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -146,19 +147,31 @@ class ProgressManager {
void Increment(const Progress::ProgressData &);
void Decrement(const Progress::ProgressData &);
+ static void Initialize();
+ static void Terminate();
+ static bool Enabled();
static ProgressManager &Instance();
-private:
+protected:
enum class EventType {
Begin,
End,
};
static void ReportProgress(const Progress::ProgressData &progress_data,
EventType type);
- llvm::StringMap>
- m_progress_category_map;
- std::mutex m_progress_map_mutex;
+ static std::optional &InstanceImpl();
+
+ void Expire(llvm::StringRef key);
adrian-prantl wrote:
Can you add a Doxygen comment?
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -146,19 +147,31 @@ class ProgressManager {
void Increment(const Progress::ProgressData &);
void Decrement(const Progress::ProgressData &);
+ static void Initialize();
+ static void Terminate();
+ static bool Enabled();
static ProgressManager &Instance();
-private:
+protected:
enum class EventType {
Begin,
End,
};
static void ReportProgress(const Progress::ProgressData &progress_data,
EventType type);
- llvm::StringMap>
- m_progress_category_map;
- std::mutex m_progress_map_mutex;
+ static std::optional &InstanceImpl();
+
+ void Expire(llvm::StringRef key);
+ struct Entry {
adrian-prantl wrote:
same here
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
https://github.com/adrian-prantl approved this pull request. https://github.com/llvm/llvm-project/pull/84890 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -75,45 +81,86 @@ void Progress::ReportProgress() {
}
}
-ProgressManager::ProgressManager() : m_progress_category_map() {}
+ProgressManager::ProgressManager()
+: m_alarm(std::chrono::milliseconds(100)), m_entries() {}
ProgressManager::~ProgressManager() {}
+void ProgressManager::Initialize() {
+ assert(!InstanceImpl() && "Already initialized.");
+ InstanceImpl().emplace();
+}
+
+void ProgressManager::Terminate() {
+ assert(InstanceImpl() && "Already terminated.");
+ InstanceImpl().reset();
+}
+
+bool ProgressManager::Enabled() { return InstanceImpl().operator bool(); }
+
ProgressManager &ProgressManager::Instance() {
- static std::once_flag g_once_flag;
- static ProgressManager *g_progress_manager = nullptr;
- std::call_once(g_once_flag, []() {
-// NOTE: known leak to avoid global destructor chain issues.
-g_progress_manager = new ProgressManager();
- });
- return *g_progress_manager;
+ assert(InstanceImpl() && "ProgressManager must be initialized");
+ return *InstanceImpl();
+}
+
+std::optional &ProgressManager::InstanceImpl() {
+ static std::optional g_progress_manager;
+ return g_progress_manager;
}
void ProgressManager::Increment(const Progress::ProgressData &progress_data) {
- std::lock_guard lock(m_progress_map_mutex);
- // If the current category exists in the map then it is not an initial
report,
- // therefore don't broadcast to the category bit. Also, store the current
- // progress data in the map so that we have a note of the ID used for the
- // initial progress report.
- if (!m_progress_category_map.contains(progress_data.title)) {
-m_progress_category_map[progress_data.title].second = progress_data;
+ std::lock_guard lock(m_entries_mutex);
+ llvm::StringRef key = progress_data.title;
+
+ // This is a new progress event.
+ if (!m_entries.contains(key)) {
ReportProgress(progress_data, EventType::Begin);
+Entry &entry = m_entries[key];
+entry.data = progress_data;
+entry.refcount = 1;
+return;
+ }
+
+ // This is an existing progress event.
+ Entry &entry = m_entries[key];
clayborg wrote:
Since we are always calling `m_entries[key]` anyway, this code could be a bit
simpler:
```
Entry &entry = m_entries[key];
if (entry.refcount == 0) {
// This is a new progress event.
entry.data = progress_data;
entry.refcount = 1;
return;
}
// This is an existing progress event.
```
I assume that accessing a non existent key in the string map will return an
default constructed object right?
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -75,45 +81,86 @@ void Progress::ReportProgress() {
}
}
-ProgressManager::ProgressManager() : m_progress_category_map() {}
+ProgressManager::ProgressManager()
+: m_alarm(std::chrono::milliseconds(100)), m_entries() {}
clayborg wrote:
Make this a lldb setting instead of hard coding to 100ms?
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -75,45 +81,86 @@ void Progress::ReportProgress() {
}
}
-ProgressManager::ProgressManager() : m_progress_category_map() {}
+ProgressManager::ProgressManager()
+: m_alarm(std::chrono::milliseconds(100)), m_entries() {}
ProgressManager::~ProgressManager() {}
+void ProgressManager::Initialize() {
+ assert(!InstanceImpl() && "Already initialized.");
+ InstanceImpl().emplace();
+}
+
+void ProgressManager::Terminate() {
+ assert(InstanceImpl() && "Already terminated.");
+ InstanceImpl().reset();
+}
+
+bool ProgressManager::Enabled() { return InstanceImpl().operator bool(); }
+
ProgressManager &ProgressManager::Instance() {
- static std::once_flag g_once_flag;
- static ProgressManager *g_progress_manager = nullptr;
- std::call_once(g_once_flag, []() {
-// NOTE: known leak to avoid global destructor chain issues.
-g_progress_manager = new ProgressManager();
- });
- return *g_progress_manager;
+ assert(InstanceImpl() && "ProgressManager must be initialized");
+ return *InstanceImpl();
+}
+
+std::optional &ProgressManager::InstanceImpl() {
+ static std::optional g_progress_manager;
+ return g_progress_manager;
}
void ProgressManager::Increment(const Progress::ProgressData &progress_data) {
- std::lock_guard lock(m_progress_map_mutex);
- // If the current category exists in the map then it is not an initial
report,
- // therefore don't broadcast to the category bit. Also, store the current
- // progress data in the map so that we have a note of the ID used for the
- // initial progress report.
- if (!m_progress_category_map.contains(progress_data.title)) {
-m_progress_category_map[progress_data.title].second = progress_data;
+ std::lock_guard lock(m_entries_mutex);
+ llvm::StringRef key = progress_data.title;
+
+ // This is a new progress event.
+ if (!m_entries.contains(key)) {
ReportProgress(progress_data, EventType::Begin);
+Entry &entry = m_entries[key];
+entry.data = progress_data;
+entry.refcount = 1;
+return;
+ }
+
+ // This is an existing progress event.
+ Entry &entry = m_entries[key];
+
+ // The progress event was scheduled to be deleted but a new one came in
before
+ // the timer expired.
+ if (entry.refcount == 0) {
+assert(entry.handle != Alarm::INVALID_HANDLE);
+
+if (!m_alarm.Cancel(entry.handle)) {
+ // The timer expired before we had a chance to cancel it. We have to
treat
+ // this as an entirely new progress event.
+ ReportProgress(progress_data, EventType::Begin);
+}
+entry.refcount = 1;
+entry.handle = Alarm::INVALID_HANDLE;
+ } else {
+entry.refcount++;
}
clayborg wrote:
One idea for this code could be:
```
if (entry.refcount++ == 1) {
// This entry was scheduled to be expired, but no longer is.
assert(entry.handle != Alarm::INVALID_HANDLE);
if (!m_alarm.Cancel(entry.handle)) {
// The timer expired before we had a chance to cancel it. We have to treat
// this as an entirely new progress event.
ReportProgress(progress_data, EventType::Begin);
}
entry.handle = Alarm::INVALID_HANDLE;
}
```
This just removes the need to manually set `entry.refcount = 1;` or to
separately increment refcount in the else body
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -75,45 +81,86 @@ void Progress::ReportProgress() {
}
}
-ProgressManager::ProgressManager() : m_progress_category_map() {}
+ProgressManager::ProgressManager()
+: m_alarm(std::chrono::milliseconds(100)), m_entries() {}
ProgressManager::~ProgressManager() {}
+void ProgressManager::Initialize() {
+ assert(!InstanceImpl() && "Already initialized.");
+ InstanceImpl().emplace();
+}
+
+void ProgressManager::Terminate() {
+ assert(InstanceImpl() && "Already terminated.");
+ InstanceImpl().reset();
+}
+
+bool ProgressManager::Enabled() { return InstanceImpl().operator bool(); }
+
ProgressManager &ProgressManager::Instance() {
- static std::once_flag g_once_flag;
- static ProgressManager *g_progress_manager = nullptr;
- std::call_once(g_once_flag, []() {
-// NOTE: known leak to avoid global destructor chain issues.
-g_progress_manager = new ProgressManager();
- });
- return *g_progress_manager;
+ assert(InstanceImpl() && "ProgressManager must be initialized");
+ return *InstanceImpl();
+}
+
+std::optional &ProgressManager::InstanceImpl() {
+ static std::optional g_progress_manager;
+ return g_progress_manager;
}
void ProgressManager::Increment(const Progress::ProgressData &progress_data) {
- std::lock_guard lock(m_progress_map_mutex);
- // If the current category exists in the map then it is not an initial
report,
- // therefore don't broadcast to the category bit. Also, store the current
- // progress data in the map so that we have a note of the ID used for the
- // initial progress report.
- if (!m_progress_category_map.contains(progress_data.title)) {
-m_progress_category_map[progress_data.title].second = progress_data;
+ std::lock_guard lock(m_entries_mutex);
+ llvm::StringRef key = progress_data.title;
+
+ // This is a new progress event.
+ if (!m_entries.contains(key)) {
ReportProgress(progress_data, EventType::Begin);
+Entry &entry = m_entries[key];
+entry.data = progress_data;
+entry.refcount = 1;
+return;
+ }
+
+ // This is an existing progress event.
+ Entry &entry = m_entries[key];
+
+ // The progress event was scheduled to be deleted but a new one came in
before
+ // the timer expired.
+ if (entry.refcount == 0) {
+assert(entry.handle != Alarm::INVALID_HANDLE);
+
+if (!m_alarm.Cancel(entry.handle)) {
+ // The timer expired before we had a chance to cancel it. We have to
treat
+ // this as an entirely new progress event.
+ ReportProgress(progress_data, EventType::Begin);
+}
+entry.refcount = 1;
+entry.handle = Alarm::INVALID_HANDLE;
+ } else {
+entry.refcount++;
}
- m_progress_category_map[progress_data.title].first++;
}
void ProgressManager::Decrement(const Progress::ProgressData &progress_data) {
- std::lock_guard lock(m_progress_map_mutex);
- auto pos = m_progress_category_map.find(progress_data.title);
+ std::lock_guard lock(m_entries_mutex);
+ llvm::StringRef key = progress_data.title;
- if (pos == m_progress_category_map.end())
+ if (!m_entries.contains(key))
return;
- if (pos->second.first <= 1) {
-ReportProgress(pos->second.second, EventType::End);
-m_progress_category_map.erase(progress_data.title);
+ Entry &entry = m_entries[key];
+ if (entry.refcount <= 1) {
+assert(entry.handle == Alarm::INVALID_HANDLE);
+// Start a timer. If it expires before we see another progress event, it
+// will be reported.
+entry.refcount = 0;
+// Copy the key to a std::string so we can pass it by value to the lambda.
+// The underlying StringRef will not exist by the time the callback is
+// called.
+std::string key_str = std::string(key);
+entry.handle = m_alarm.Create([=]() { Expire(key_str); });
} else {
---pos->second.first;
+entry.refcount--;
}
clayborg wrote:
This can similarly always decrement, similar to previous suggestion:
```
Entry &entry = m_entries[key];
if (entry.refcount > 0 && --entry.refcount == 0) {
assert(entry.handle == Alarm::INVALID_HANDLE);
// Start a timer. If it expires before we see another progress event, it
// will be reported.
// Copy the key to a std::string so we can pass it by value to the lambda.
// The underlying StringRef will not exist by the time the callback is
// called.
std::string key_str = std::string(key);
entry.handle = m_alarm.Create([=]() { Expire(key_str); });
}
```
This stops the manual editing of `entry.refcount = 0;` and the separate
decrement in the else
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Allow languages to filter breakpoints set by line (PR #83908)
https://github.com/felipepiovezan updated
https://github.com/llvm/llvm-project/pull/83908
>From 51307b548d09c34ee06037ccf459110e451970a9 Mon Sep 17 00:00:00 2001
From: Felipe de Azevedo Piovezan
Date: Mon, 4 Mar 2024 09:56:18 -0800
Subject: [PATCH 1/5] [lldb] Allow languages to filter breakpoints set by line
Some languages may create artificial functions that have no real user code, even
though there is line table information for them. One such case is with coroutine
code that receives the CoroSplitter transformation in LLVM IR. That code
transformation creates many different Functions, cloning one Instruction into
many Instructions in many different Functions and copying the associated debug
locations.
It would be difficult to make that pass delete debug locations of cloned
instructions in a language agnostic way (is it even possible?), but LLDB can
ignore certain locations by querying its Language APIs and having it decide
based on, for example, mangling information.
---
lldb/include/lldb/Target/Language.h | 4
lldb/source/Breakpoint/BreakpointResolver.cpp | 12
2 files changed, 16 insertions(+)
diff --git a/lldb/include/lldb/Target/Language.h
b/lldb/include/lldb/Target/Language.h
index 0cbd8a32dccd54..957c40eb7c0772 100644
--- a/lldb/include/lldb/Target/Language.h
+++ b/lldb/include/lldb/Target/Language.h
@@ -339,6 +339,10 @@ class Language : public PluginInterface {
virtual llvm::StringRef GetInstanceVariableName() { return {}; }
+ virtual bool IsInterestingCtxForLineBreakpoint(const SymbolContext &) const {
+return true;
+ }
+
protected:
// Classes that inherit from Language can see and modify these
diff --git a/lldb/source/Breakpoint/BreakpointResolver.cpp
b/lldb/source/Breakpoint/BreakpointResolver.cpp
index bc6348716ef418..876b30c6d76d55 100644
--- a/lldb/source/Breakpoint/BreakpointResolver.cpp
+++ b/lldb/source/Breakpoint/BreakpointResolver.cpp
@@ -23,6 +23,7 @@
#include "lldb/Symbol/CompileUnit.h"
#include "lldb/Symbol/Function.h"
#include "lldb/Symbol/SymbolContext.h"
+#include "lldb/Target/Language.h"
#include "lldb/Target/Target.h"
#include "lldb/Utility/LLDBLog.h"
#include "lldb/Utility/Log.h"
@@ -199,6 +200,15 @@ bool operator<(const SourceLoc lhs, const SourceLoc rhs) {
}
} // namespace
+static void
+ApplyLanguageFilters(llvm::SmallVectorImpl &sc_list) {
+ llvm::erase_if(sc_list, [](SymbolContext &sc) {
+if (Language *lang = Language::FindPlugin(sc.GetLanguage()))
+ return !lang->IsInterestingCtxForLineBreakpoint(sc);
+return false;
+ });
+}
+
void BreakpointResolver::SetSCMatchesByLine(
SearchFilter &filter, SymbolContextList &sc_list, bool skip_prologue,
llvm::StringRef log_ident, uint32_t line, std::optional column) {
@@ -206,6 +216,8 @@ void BreakpointResolver::SetSCMatchesByLine(
for (uint32_t i = 0; i < sc_list.GetSize(); ++i)
all_scs.push_back(sc_list[i]);
+ ApplyLanguageFilters(all_scs);
+
while (all_scs.size()) {
uint32_t closest_line = UINT32_MAX;
>From 7ea76d6c04d063d50da52756179639f2037aa793 Mon Sep 17 00:00:00 2001
From: Felipe de Azevedo Piovezan
Date: Tue, 5 Mar 2024 18:47:51 -0800
Subject: [PATCH 2/5] fixup! add target option
---
lldb/include/lldb/Target/Language.h | 9 ++--
lldb/include/lldb/Target/Target.h | 2 ++
lldb/source/Breakpoint/BreakpointResolver.cpp | 23 +--
lldb/source/Target/Target.cpp | 8 +++
lldb/source/Target/TargetProperties.td| 3 +++
5 files changed, 31 insertions(+), 14 deletions(-)
diff --git a/lldb/include/lldb/Target/Language.h
b/lldb/include/lldb/Target/Language.h
index 957c40eb7c0772..9db9f4e297e114 100644
--- a/lldb/include/lldb/Target/Language.h
+++ b/lldb/include/lldb/Target/Language.h
@@ -339,8 +339,13 @@ class Language : public PluginInterface {
virtual llvm::StringRef GetInstanceVariableName() { return {}; }
- virtual bool IsInterestingCtxForLineBreakpoint(const SymbolContext &) const {
-return true;
+ // Returns true if this SymbolContext should be used when setting breakpoints
+ // by line (number or regex). This is useful for languages that create
+ // artificial functions without any meaningful user code associated with them
+ // (e.g. code that gets expanded in late compilation stages, like by
+ // CoroSplitter).
+ virtual bool IsArtificialCtxForLineBreakpoint(const SymbolContext &) const {
+return false;
}
protected:
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index 8f57358981d4d2..ab04aa57f17300 100644
--- a/lldb/include/lldb/Target/Target.h
+++ b/lldb/include/lldb/Target/Target.h
@@ -258,6 +258,8 @@ class TargetProperties : public Properties {
bool GetDebugUtilityExpression() const;
+ bool GetIgnoreBreakpointsFromLanguageArtificialLocations() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
diff --git a/lldb/source/Breakpoint/
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
https://github.com/JDevlieghere updated
https://github.com/llvm/llvm-project/pull/84854
>From aae699eb956d1e235682b34e6407f6a9990028b3 Mon Sep 17 00:00:00 2001
From: Jonas Devlieghere
Date: Tue, 5 Mar 2024 22:57:43 -0800
Subject: [PATCH 1/2] [lldb] Add an Alarm class
Add an Alarm class which allows scheduling callbacks after a specific
timeout has elapsed.
---
lldb/include/lldb/Host/Alarm.h | 112 +
lldb/source/Host/CMakeLists.txt| 1 +
lldb/source/Host/common/Alarm.cpp | 193 +
lldb/unittests/Host/AlarmTest.cpp | 164
lldb/unittests/Host/CMakeLists.txt | 1 +
5 files changed, 471 insertions(+)
create mode 100644 lldb/include/lldb/Host/Alarm.h
create mode 100644 lldb/source/Host/common/Alarm.cpp
create mode 100644 lldb/unittests/Host/AlarmTest.cpp
diff --git a/lldb/include/lldb/Host/Alarm.h b/lldb/include/lldb/Host/Alarm.h
new file mode 100644
index 00..7bdbc8f2b0ed74
--- /dev/null
+++ b/lldb/include/lldb/Host/Alarm.h
@@ -0,0 +1,112 @@
+//===-- Alarm.h -*- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#ifndef LLDB_HOST_ALARM_H
+#define LLDB_HOST_ALARM_H
+
+#include "lldb/Host/HostThread.h"
+#include "lldb/lldb-types.h"
+#include "llvm/Support/Chrono.h"
+
+namespace lldb_private {
+
+/// \class Alarm abstraction that enables scheduling a callback function after
a
+/// specified timeout. Creating an alarm for a callback returns a Handle that
+/// can be used to restart or cancel the alarm.
+class Alarm {
+public:
+ using Handle = uint64_t;
+ using Callback = std::function;
+ using TimePoint = llvm::sys::TimePoint<>;
+ using Duration = std::chrono::milliseconds;
+
+ Alarm(Duration timeout, bool run_callback_on_exit = false);
+ ~Alarm();
+
+ /// Create an alarm for the given callback. The alarm will expire and the
+ /// callback will be called after the timeout.
+ ///
+ /// \returns
+ /// Handle which can be used to restart or cancel the alarm.
+ Handle Create(Callback callback);
+
+ /// Restart the alarm for the given Handle. The alarm will expire and the
+ /// callback will be called after the timeout.
+ ///
+ /// \returns
+ /// True if the alarm was successfully restarted. False if there is no
alarm
+ /// for the given Handle or the alarm already expired.
+ bool Restart(Handle handle);
+
+ /// Cancel the alarm for the given Handle. The alarm and its handle will be
+ /// removed.
+ ///
+ /// \returns
+ /// True if the alarm was successfully canceled and the Handle removed.
+ /// False if there is no alarm for the given Handle or the alarm already
+ /// expired.
+ bool Cancel(Handle handle);
+
+ static constexpr Handle INVALID_HANDLE = 0;
+
+private:
+ /// Helper functions to start, stop and check the status of the alarm thread.
+ /// @{
+ void StartAlarmThread();
+ void StopAlarmThread();
+ bool AlarmThreadRunning();
+ /// @}
+
+ /// Return an unique, monotonically increasing handle.
+ static Handle GetNextUniqueHandle();
+
+ /// Helper to compute the next time the alarm thread needs to wake up.
+ TimePoint GetNextExpiration() const;
+
+ /// Alarm entry.
+ struct Entry {
+Handle handle;
+Callback callback;
+TimePoint expiration;
+
+Entry(Callback callback, TimePoint expiration);
+bool operator==(const Entry &rhs) { return handle == rhs.handle; }
+ };
+
+ /// List of alarm entries.
+ std::vector m_entries;
+
+ /// Timeout between when an alarm is created and when it fires.
+ Duration m_timeout;
+
+ /// The alarm thread.
+ /// @{
+ HostThread m_alarm_thread;
+ lldb::thread_result_t AlarmThread();
+ /// @}
+
+ /// Synchronize access between the alarm thread and the main thread.
+ std::mutex m_alarm_mutex;
+
+ /// Condition variable used to wake up the alarm thread.
+ std::condition_variable m_alarm_cv;
+
+ /// Flag to signal the alarm thread that something changed and we need to
+ /// recompute the next alarm.
+ bool m_recompute_next_alarm = false;
+
+ /// Flag to signal the alarm thread to exit.
+ bool m_exit = false;
+
+ /// Flag to signal we should run all callbacks on exit.
+ bool m_run_callbacks_on_exit = false;
+};
+
+} // namespace lldb_private
+
+#endif // LLDB_HOST_ALARM_H
diff --git a/lldb/source/Host/CMakeLists.txt b/lldb/source/Host/CMakeLists.txt
index fe6e539f758fda..c2e091ee8555b7 100644
--- a/lldb/source/Host/CMakeLists.txt
+++ b/lldb/source/Host/CMakeLists.txt
@@ -13,6 +13,7 @@ macro(add_host_subdirectory group)
endmacro()
add_host_subdirectory(common
+ common/Alarm.cpp
common/FileAction.cpp
common/FileCache.cpp
common/File.cpp
diff --git a/lldb/source/Host/common/Alarm.cpp
b
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -0,0 +1,164 @@
+//===-- AlarmTest.cpp
-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/Host/Alarm.h"
+#include "gtest/gtest.h"
+
+#include
+#include
+
+using namespace lldb_private;
+using namespace std::chrono_literals;
+
+static constexpr auto ALARM_TIMEOUT = 500ms;
+static constexpr auto TEST_TIMEOUT = 1000ms;
+static constexpr bool RUN_CALLBACKS_ON_EXIT = true;
+
+// Enable strict checking of the ALARM_TIMEOUTs. This should only be enabled
for
+// development and never during automated testing where scheduling and
+// ALARM_TIMEOUTs are unpredictable.
+#define STRICT_ALARM_TIMEOUT 1
+
+#if STRICT_ALARM_TIMEOUT
+static constexpr auto EPSILON = 10ms;
+#endif
+
+TEST(AlarmTest, Create) {
+ std::mutex m;
+
+ std::vector callbacks_actual;
+ std::vector callbacks_expected;
+
+ Alarm alarm(ALARM_TIMEOUT, RUN_CALLBACKS_ON_EXIT);
+
+ // Create 5 alarms some time apart.
+ for (size_t i = 0; i < 5; ++i) {
+callbacks_actual.emplace_back();
+callbacks_expected.emplace_back(std::chrono::system_clock::now() +
+ALARM_TIMEOUT);
+
+alarm.Create([&callbacks_actual, &m, i]() {
+ std::lock_guard guard(m);
+ callbacks_actual[i] = std::chrono::system_clock::now();
+});
+
+std::this_thread::sleep_for(ALARM_TIMEOUT / 5);
+ }
+
+ // Leave plenty of time for all the alarms to fire.
+ std::this_thread::sleep_for(TEST_TIMEOUT);
+
+ // Make sure all the alarms around the expected time.
+ for (size_t i = 0; i < 5; ++i) {
+EXPECT_GE(callbacks_actual[i], callbacks_expected[i]);
+#if STRICT_ALARM_TIMEOUT
+EXPECT_LE(callbacks_actual[i], callbacks_expected[i] + EPSILON);
+#endif
+ }
+}
+
+TEST(AlarmTest, Exit) {
+ std::mutex m;
+
+ std::vector handles;
+ std::vector callbacks;
+
+ {
+Alarm alarm(ALARM_TIMEOUT, RUN_CALLBACKS_ON_EXIT);
+
+// Create 5 alarms.
+for (size_t i = 0; i < 5; ++i) {
+ callbacks.emplace_back(false);
+
+ handles.push_back(alarm.Create([&callbacks, &m, i]() {
+std::lock_guard guard(m);
+callbacks[i] = true;
+ }));
+}
+
+// Let the alarm go out of scope before any alarm had a chance to fire.
+ }
+
+ // Make sure none of the first 4 alarms fired.
+ for (bool callback : callbacks)
+EXPECT_TRUE(callback);
JDevlieghere wrote:
Yup, that's a typo.
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] 88bf640 - [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (#84890)
Author: Michael Buch
Date: 2024-03-12T22:19:27Z
New Revision: 88bf64097e453deca73c91ec7de1af7eebe296a9
URL:
https://github.com/llvm/llvm-project/commit/88bf64097e453deca73c91ec7de1af7eebe296a9
DIFF:
https://github.com/llvm/llvm-project/commit/88bf64097e453deca73c91ec7de1af7eebe296a9.diff
LOG: [lldb][test] TestExprCompletion.py: add tests for completion of reserved
identifiers (#84890)
Added:
lldb/test/API/commands/expression/completion/sys/reserved.h
Modified:
lldb/test/API/commands/expression/completion/Makefile
lldb/test/API/commands/expression/completion/TestExprCompletion.py
lldb/test/API/commands/expression/completion/main.cpp
Removed:
diff --git a/lldb/test/API/commands/expression/completion/Makefile
b/lldb/test/API/commands/expression/completion/Makefile
index 020dce7c31d11d..9882622b2189ea 100644
--- a/lldb/test/API/commands/expression/completion/Makefile
+++ b/lldb/test/API/commands/expression/completion/Makefile
@@ -1,3 +1,4 @@
CXX_SOURCES := main.cpp other.cpp
+CXXFLAGS += -isystem $(SRCDIR)/sys
include Makefile.rules
diff --git
a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
index c6a1e3c0f42275..022b9436ee8ea6 100644
--- a/lldb/test/API/commands/expression/completion/TestExprCompletion.py
+++ b/lldb/test/API/commands/expression/completion/TestExprCompletion.py
@@ -246,6 +246,11 @@ def test_expr_completion(self):
"expr some_expr.Self(). FooNoArgs", "expr some_expr.Self().
FooNoArgsBar()"
)
+self.complete_from_to("expr myVec.__f", "expr myVec.__func()")
+self.complete_from_to("expr myVec._F", "expr myVec._Func()")
+self.complete_from_to("expr myVec.__m", "expr myVec.__mem")
+self.complete_from_to("expr myVec._M", "expr myVec._Mem")
+
def test_expr_completion_with_descriptions(self):
self.build()
self.main_source = "main.cpp"
diff --git a/lldb/test/API/commands/expression/completion/main.cpp
b/lldb/test/API/commands/expression/completion/main.cpp
index 908bebbebff568..5e03805a7a4d5c 100644
--- a/lldb/test/API/commands/expression/completion/main.cpp
+++ b/lldb/test/API/commands/expression/completion/main.cpp
@@ -1,3 +1,5 @@
+#include
+
namespace LongNamespaceName { class NestedClass { long m; }; }
// Defined in other.cpp, we only have a forward declaration here.
@@ -31,5 +33,8 @@ int main()
some_expr.FooNumbersBar1();
Expr::StaticMemberMethodBar();
ForwardDecl *fwd_decl_ptr = &fwd_decl;
+MyVec myVec;
+myVec.__func();
+myVec._Func();
return 0; // Break here
}
diff --git a/lldb/test/API/commands/expression/completion/sys/reserved.h
b/lldb/test/API/commands/expression/completion/sys/reserved.h
new file mode 100644
index 00..0ce10ebec62bfd
--- /dev/null
+++ b/lldb/test/API/commands/expression/completion/sys/reserved.h
@@ -0,0 +1,8 @@
+class MyVec {
+ int __mem;
+ int _Mem;
+
+public:
+ void __func() {}
+ void _Func() {}
+};
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][test] TestExprCompletion.py: add tests for completion of reserved identifiers (PR #84890)
https://github.com/Michael137 closed https://github.com/llvm/llvm-project/pull/84890 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Implement coalescing of disjoint progress events (PR #84854)
@@ -75,45 +81,86 @@ void Progress::ReportProgress() {
}
}
-ProgressManager::ProgressManager() : m_progress_category_map() {}
+ProgressManager::ProgressManager()
+: m_alarm(std::chrono::milliseconds(100)), m_entries() {}
JDevlieghere wrote:
I initially wanted to make this a setting, but you need this when initializing
the ProgressManager, which happens way before we even have a debugger or a
command interpreter, so there wouldn't be a way to set the setting even. I
could move the timeout out of the constructor and allow you to set the timeout
after the fact, but that would still leave the question of where to get the
setting from: the ProgressManager is shared across debuggers.
https://github.com/llvm/llvm-project/pull/84854
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Avoid a potential exit(1) in LLVMContext::diagnose() (PR #84992)
https://github.com/adrian-prantl created
https://github.com/llvm/llvm-project/pull/84992
by handling *all* errors in IRExecDiagnosticHandler. The function that call
this handles all unhandled errors with an `exit(1)`.
rdar://124459751
I don't really have a testcase for this, since the crash report I got for this
involved the Swift language plugin.
>From 0c805bfc91d92a364b02bd2e555192bc996fc83f Mon Sep 17 00:00:00 2001
From: Adrian Prantl
Date: Tue, 12 Mar 2024 16:22:33 -0700
Subject: [PATCH] Avoid a potential exit(1) in LLVMContext::diagnose()
by handling *all* errors in IRExecDiagnosticHandler.
rdar://124459751
---
lldb/source/Expression/IRExecutionUnit.cpp | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/lldb/source/Expression/IRExecutionUnit.cpp
b/lldb/source/Expression/IRExecutionUnit.cpp
index e4e131d70d4319..cb9bee8733e15d 100644
--- a/lldb/source/Expression/IRExecutionUnit.cpp
+++ b/lldb/source/Expression/IRExecutionUnit.cpp
@@ -212,18 +212,17 @@ struct IRExecDiagnosticHandler : public
llvm::DiagnosticHandler {
Status *err;
IRExecDiagnosticHandler(Status *err) : err(err) {}
bool handleDiagnostics(const llvm::DiagnosticInfo &DI) override {
-if (DI.getKind() == llvm::DK_SrcMgr) {
+if (DI.getSeverity() == llvm::DS_Error) {
const auto &DISM = llvm::cast(DI);
if (err && err->Success()) {
err->SetErrorToGenericError();
err->SetErrorStringWithFormat(
-"Inline assembly error: %s",
+"IRExecution error: %s",
DISM.getSMDiag().getMessage().str().c_str());
}
- return true;
}
-return false;
+return true;
}
};
} // namespace
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Avoid a potential exit(1) in LLVMContext::diagnose() (PR #84992)
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Adrian Prantl (adrian-prantl)
Changes
by handling *all* errors in IRExecDiagnosticHandler. The function that call
this handles all unhandled errors with an `exit(1)`.
rdar://124459751
I don't really have a testcase for this, since the crash report I got for this
involved the Swift language plugin.
---
Full diff: https://github.com/llvm/llvm-project/pull/84992.diff
1 Files Affected:
- (modified) lldb/source/Expression/IRExecutionUnit.cpp (+3-4)
``diff
diff --git a/lldb/source/Expression/IRExecutionUnit.cpp
b/lldb/source/Expression/IRExecutionUnit.cpp
index e4e131d70d4319..cb9bee8733e15d 100644
--- a/lldb/source/Expression/IRExecutionUnit.cpp
+++ b/lldb/source/Expression/IRExecutionUnit.cpp
@@ -212,18 +212,17 @@ struct IRExecDiagnosticHandler : public
llvm::DiagnosticHandler {
Status *err;
IRExecDiagnosticHandler(Status *err) : err(err) {}
bool handleDiagnostics(const llvm::DiagnosticInfo &DI) override {
-if (DI.getKind() == llvm::DK_SrcMgr) {
+if (DI.getSeverity() == llvm::DS_Error) {
const auto &DISM = llvm::cast(DI);
if (err && err->Success()) {
err->SetErrorToGenericError();
err->SetErrorStringWithFormat(
-"Inline assembly error: %s",
+"IRExecution error: %s",
DISM.getSMDiag().getMessage().str().c_str());
}
- return true;
}
-return false;
+return true;
}
};
} // namespace
``
https://github.com/llvm/llvm-project/pull/84992
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Allow languages to filter breakpoints set by line (PR #83908)
https://github.com/felipepiovezan updated
https://github.com/llvm/llvm-project/pull/83908
>From 51307b548d09c34ee06037ccf459110e451970a9 Mon Sep 17 00:00:00 2001
From: Felipe de Azevedo Piovezan
Date: Mon, 4 Mar 2024 09:56:18 -0800
Subject: [PATCH 1/5] [lldb] Allow languages to filter breakpoints set by line
Some languages may create artificial functions that have no real user code, even
though there is line table information for them. One such case is with coroutine
code that receives the CoroSplitter transformation in LLVM IR. That code
transformation creates many different Functions, cloning one Instruction into
many Instructions in many different Functions and copying the associated debug
locations.
It would be difficult to make that pass delete debug locations of cloned
instructions in a language agnostic way (is it even possible?), but LLDB can
ignore certain locations by querying its Language APIs and having it decide
based on, for example, mangling information.
---
lldb/include/lldb/Target/Language.h | 4
lldb/source/Breakpoint/BreakpointResolver.cpp | 12
2 files changed, 16 insertions(+)
diff --git a/lldb/include/lldb/Target/Language.h
b/lldb/include/lldb/Target/Language.h
index 0cbd8a32dccd54..957c40eb7c0772 100644
--- a/lldb/include/lldb/Target/Language.h
+++ b/lldb/include/lldb/Target/Language.h
@@ -339,6 +339,10 @@ class Language : public PluginInterface {
virtual llvm::StringRef GetInstanceVariableName() { return {}; }
+ virtual bool IsInterestingCtxForLineBreakpoint(const SymbolContext &) const {
+return true;
+ }
+
protected:
// Classes that inherit from Language can see and modify these
diff --git a/lldb/source/Breakpoint/BreakpointResolver.cpp
b/lldb/source/Breakpoint/BreakpointResolver.cpp
index bc6348716ef418..876b30c6d76d55 100644
--- a/lldb/source/Breakpoint/BreakpointResolver.cpp
+++ b/lldb/source/Breakpoint/BreakpointResolver.cpp
@@ -23,6 +23,7 @@
#include "lldb/Symbol/CompileUnit.h"
#include "lldb/Symbol/Function.h"
#include "lldb/Symbol/SymbolContext.h"
+#include "lldb/Target/Language.h"
#include "lldb/Target/Target.h"
#include "lldb/Utility/LLDBLog.h"
#include "lldb/Utility/Log.h"
@@ -199,6 +200,15 @@ bool operator<(const SourceLoc lhs, const SourceLoc rhs) {
}
} // namespace
+static void
+ApplyLanguageFilters(llvm::SmallVectorImpl &sc_list) {
+ llvm::erase_if(sc_list, [](SymbolContext &sc) {
+if (Language *lang = Language::FindPlugin(sc.GetLanguage()))
+ return !lang->IsInterestingCtxForLineBreakpoint(sc);
+return false;
+ });
+}
+
void BreakpointResolver::SetSCMatchesByLine(
SearchFilter &filter, SymbolContextList &sc_list, bool skip_prologue,
llvm::StringRef log_ident, uint32_t line, std::optional column) {
@@ -206,6 +216,8 @@ void BreakpointResolver::SetSCMatchesByLine(
for (uint32_t i = 0; i < sc_list.GetSize(); ++i)
all_scs.push_back(sc_list[i]);
+ ApplyLanguageFilters(all_scs);
+
while (all_scs.size()) {
uint32_t closest_line = UINT32_MAX;
>From 7ea76d6c04d063d50da52756179639f2037aa793 Mon Sep 17 00:00:00 2001
From: Felipe de Azevedo Piovezan
Date: Tue, 5 Mar 2024 18:47:51 -0800
Subject: [PATCH 2/5] fixup! add target option
---
lldb/include/lldb/Target/Language.h | 9 ++--
lldb/include/lldb/Target/Target.h | 2 ++
lldb/source/Breakpoint/BreakpointResolver.cpp | 23 +--
lldb/source/Target/Target.cpp | 8 +++
lldb/source/Target/TargetProperties.td| 3 +++
5 files changed, 31 insertions(+), 14 deletions(-)
diff --git a/lldb/include/lldb/Target/Language.h
b/lldb/include/lldb/Target/Language.h
index 957c40eb7c0772..9db9f4e297e114 100644
--- a/lldb/include/lldb/Target/Language.h
+++ b/lldb/include/lldb/Target/Language.h
@@ -339,8 +339,13 @@ class Language : public PluginInterface {
virtual llvm::StringRef GetInstanceVariableName() { return {}; }
- virtual bool IsInterestingCtxForLineBreakpoint(const SymbolContext &) const {
-return true;
+ // Returns true if this SymbolContext should be used when setting breakpoints
+ // by line (number or regex). This is useful for languages that create
+ // artificial functions without any meaningful user code associated with them
+ // (e.g. code that gets expanded in late compilation stages, like by
+ // CoroSplitter).
+ virtual bool IsArtificialCtxForLineBreakpoint(const SymbolContext &) const {
+return false;
}
protected:
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index 8f57358981d4d2..ab04aa57f17300 100644
--- a/lldb/include/lldb/Target/Target.h
+++ b/lldb/include/lldb/Target/Target.h
@@ -258,6 +258,8 @@ class TargetProperties : public Properties {
bool GetDebugUtilityExpression() const;
+ bool GetIgnoreBreakpointsFromLanguageArtificialLocations() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
diff --git a/lldb/source/Breakpoint/
[Lldb-commits] [lldb] [lldb] [Mach-O] ProcessMachCore needs to strip TBI data from addrs (PR #84998)
https://github.com/jasonmolenda created
https://github.com/llvm/llvm-project/pull/84998
Darwin AArch64 application processors are run with Top Byte Ignore mode enabled
so metadata may be stored in the top byte, it needs to be ignored when
reading/writing memory. David Spickett handled this already in the base class
Process::ReadMemory but ProcessMachCore overrides that method (to avoid the
memory cache) and did not pick up the same change. I add a test case that
creates a pointer with metadata in the top byte and dereferences it with a live
process and with a corefile.
rdar://123784501
>From 4278537c262b01b1d6432391bd9d8017eb96c60a Mon Sep 17 00:00:00 2001
From: Jason Molenda
Date: Tue, 12 Mar 2024 17:09:30 -0700
Subject: [PATCH] [lldb] [Mach-O] ProcessMachCore needs to strip TBI data from
addrs
Darwin AArch64 application processors are run with Top Byte Ignore
mode enabled so metadata may be stored in the top byte, it needs
to be ignored when reading/writing memory. David Spickett handled
this already in the base class Process::ReadMemory but ProcessMachCore
overrides that method (to avoid the memory cache) and did not pick
up the same change. I add a test case that creates a pointer with
metadata in the top byte and dereferences it with a live process and
with a corefile.
rdar://123784501
---
.../Process/mach-core/ProcessMachCore.cpp | 2 +-
lldb/test/API/macosx/tbi-honored/Makefile | 3 ++
.../API/macosx/tbi-honored/TestTBIHonored.py | 49 +++
lldb/test/API/macosx/tbi-honored/main.c | 13 +
4 files changed, 66 insertions(+), 1 deletion(-)
create mode 100644 lldb/test/API/macosx/tbi-honored/Makefile
create mode 100644 lldb/test/API/macosx/tbi-honored/TestTBIHonored.py
create mode 100644 lldb/test/API/macosx/tbi-honored/main.c
diff --git a/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
b/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
index 3961dcf0fbcc0e..7b9938d4f02020 100644
--- a/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
+++ b/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
@@ -652,7 +652,7 @@ size_t ProcessMachCore::ReadMemory(addr_t addr, void *buf,
size_t size,
Status &error) {
// Don't allow the caching that lldb_private::Process::ReadMemory does since
// in core files we have it all cached our our core file anyway.
- return DoReadMemory(addr, buf, size, error);
+ return DoReadMemory(FixAnyAddress(addr), buf, size, error);
}
size_t ProcessMachCore::DoReadMemory(addr_t addr, void *buf, size_t size,
diff --git a/lldb/test/API/macosx/tbi-honored/Makefile
b/lldb/test/API/macosx/tbi-honored/Makefile
new file mode 100644
index 00..10495940055b63
--- /dev/null
+++ b/lldb/test/API/macosx/tbi-honored/Makefile
@@ -0,0 +1,3 @@
+C_SOURCES := main.c
+
+include Makefile.rules
diff --git a/lldb/test/API/macosx/tbi-honored/TestTBIHonored.py
b/lldb/test/API/macosx/tbi-honored/TestTBIHonored.py
new file mode 100644
index 00..d38685359af6d1
--- /dev/null
+++ b/lldb/test/API/macosx/tbi-honored/TestTBIHonored.py
@@ -0,0 +1,49 @@
+"""Test that lldb on Darwin ignores metadata in the top byte of addresses."""
+
+import os
+import re
+import subprocess
+
+import lldb
+from lldbsuite.test.decorators import *
+from lldbsuite.test.lldbtest import *
+from lldbsuite.test import lldbutil
+
+
+class TestTBIHonored(TestBase):
+@no_debug_info_test
+@skipUnlessDarwin
+@skipIf(archs=no_match(["arm64", "arm64e"]))
+@skipIfRemote
+def do_variable_access_tests(self, frame):
+self.assertEqual(
+frame.variables["pb"][0]
+.GetChildMemberWithName("p")
+.Dereference()
+.GetValueAsUnsigned(),
+15,
+)
+addr =
frame.variables["pb"][0].GetChildMemberWithName("p").GetValueAsUnsigned()
+self.expect("expr -- *pb.p", substrs=["15"])
+self.expect("frame variable *pb.p", substrs=["15"])
+self.expect("expr -- *(int*)0x%x" % addr, substrs=["15"])
+
+def test(self):
+corefile = self.getBuildArtifact("process.core")
+self.build()
+(target, process, thread, bkpt) = lldbutil.run_to_source_breakpoint(
+self, "// break here", lldb.SBFileSpec("main.c")
+)
+
+self.do_variable_access_tests(thread.GetFrameAtIndex(0))
+
+self.runCmd("process save-core -s stack " + corefile)
+self.dbg.DeleteTarget(target)
+
+# Now load the corefile
+target = self.dbg.CreateTarget("")
+process = target.LoadCore(corefile)
+thread = process.GetSelectedThread()
+self.assertTrue(process.GetSelectedThread().IsValid())
+
+self.do_variable_access_tests(thread.GetFrameAtIndex(0))
diff --git a/lldb/test/API/macosx/tbi-honored/main.c
b/lldb/test/API/macosx/tbi-honored/main.c
new file mode 100644
index 00..3d7ad0b04cd664
--- /dev/null
+++ b/lldb/test/API/
[Lldb-commits] [lldb] [lldb] [Mach-O] ProcessMachCore needs to strip TBI data from addrs (PR #84998)
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Jason Molenda (jasonmolenda)
Changes
Darwin AArch64 application processors are run with Top Byte Ignore mode enabled
so metadata may be stored in the top byte, it needs to be ignored when
reading/writing memory. David Spickett handled this already in the base class
Process::ReadMemory but ProcessMachCore overrides that method (to avoid the
memory cache) and did not pick up the same change. I add a test case that
creates a pointer with metadata in the top byte and dereferences it with a live
process and with a corefile.
rdar://123784501
---
Full diff: https://github.com/llvm/llvm-project/pull/84998.diff
4 Files Affected:
- (modified) lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp (+1-1)
- (added) lldb/test/API/macosx/tbi-honored/Makefile (+3)
- (added) lldb/test/API/macosx/tbi-honored/TestTBIHonored.py (+49)
- (added) lldb/test/API/macosx/tbi-honored/main.c (+13)
``diff
diff --git a/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
b/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
index 3961dcf0fbcc0e..7b9938d4f02020 100644
--- a/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
+++ b/lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp
@@ -652,7 +652,7 @@ size_t ProcessMachCore::ReadMemory(addr_t addr, void *buf,
size_t size,
Status &error) {
// Don't allow the caching that lldb_private::Process::ReadMemory does since
// in core files we have it all cached our our core file anyway.
- return DoReadMemory(addr, buf, size, error);
+ return DoReadMemory(FixAnyAddress(addr), buf, size, error);
}
size_t ProcessMachCore::DoReadMemory(addr_t addr, void *buf, size_t size,
diff --git a/lldb/test/API/macosx/tbi-honored/Makefile
b/lldb/test/API/macosx/tbi-honored/Makefile
new file mode 100644
index 00..10495940055b63
--- /dev/null
+++ b/lldb/test/API/macosx/tbi-honored/Makefile
@@ -0,0 +1,3 @@
+C_SOURCES := main.c
+
+include Makefile.rules
diff --git a/lldb/test/API/macosx/tbi-honored/TestTBIHonored.py
b/lldb/test/API/macosx/tbi-honored/TestTBIHonored.py
new file mode 100644
index 00..d38685359af6d1
--- /dev/null
+++ b/lldb/test/API/macosx/tbi-honored/TestTBIHonored.py
@@ -0,0 +1,49 @@
+"""Test that lldb on Darwin ignores metadata in the top byte of addresses."""
+
+import os
+import re
+import subprocess
+
+import lldb
+from lldbsuite.test.decorators import *
+from lldbsuite.test.lldbtest import *
+from lldbsuite.test import lldbutil
+
+
+class TestTBIHonored(TestBase):
+@no_debug_info_test
+@skipUnlessDarwin
+@skipIf(archs=no_match(["arm64", "arm64e"]))
+@skipIfRemote
+def do_variable_access_tests(self, frame):
+self.assertEqual(
+frame.variables["pb"][0]
+.GetChildMemberWithName("p")
+.Dereference()
+.GetValueAsUnsigned(),
+15,
+)
+addr =
frame.variables["pb"][0].GetChildMemberWithName("p").GetValueAsUnsigned()
+self.expect("expr -- *pb.p", substrs=["15"])
+self.expect("frame variable *pb.p", substrs=["15"])
+self.expect("expr -- *(int*)0x%x" % addr, substrs=["15"])
+
+def test(self):
+corefile = self.getBuildArtifact("process.core")
+self.build()
+(target, process, thread, bkpt) = lldbutil.run_to_source_breakpoint(
+self, "// break here", lldb.SBFileSpec("main.c")
+)
+
+self.do_variable_access_tests(thread.GetFrameAtIndex(0))
+
+self.runCmd("process save-core -s stack " + corefile)
+self.dbg.DeleteTarget(target)
+
+# Now load the corefile
+target = self.dbg.CreateTarget("")
+process = target.LoadCore(corefile)
+thread = process.GetSelectedThread()
+self.assertTrue(process.GetSelectedThread().IsValid())
+
+self.do_variable_access_tests(thread.GetFrameAtIndex(0))
diff --git a/lldb/test/API/macosx/tbi-honored/main.c
b/lldb/test/API/macosx/tbi-honored/main.c
new file mode 100644
index 00..3d7ad0b04cd664
--- /dev/null
+++ b/lldb/test/API/macosx/tbi-honored/main.c
@@ -0,0 +1,13 @@
+#include
+#include
+union ptrbytes {
+ int *p;
+ uint8_t bytes[8];
+};
+int main() {
+ int c = 15;
+ union ptrbytes pb;
+ pb.p = &c;
+ pb.bytes[7] = 0xfe;
+ printf("%d\n", *pb.p); // break here
+}
``
https://github.com/llvm/llvm-project/pull/84998
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] e2468bf - [lldb][debugserver] Update flags past to app launch request
Author: Jason Molenda Date: 2024-03-12T17:19:46-07:00 New Revision: e2468bf16a0c1f63a39aa417c15c03ebd77fab9e URL: https://github.com/llvm/llvm-project/commit/e2468bf16a0c1f63a39aa417c15c03ebd77fab9e DIFF: https://github.com/llvm/llvm-project/commit/e2468bf16a0c1f63a39aa417c15c03ebd77fab9e.diff LOG: [lldb][debugserver] Update flags past to app launch request rdar://117421999 Added: Modified: lldb/tools/debugserver/source/MacOSX/MachProcess.mm Removed: diff --git a/lldb/tools/debugserver/source/MacOSX/MachProcess.mm b/lldb/tools/debugserver/source/MacOSX/MachProcess.mm index 87bdbf835bfd10..70b4564a027b1b 100644 --- a/lldb/tools/debugserver/source/MacOSX/MachProcess.mm +++ b/lldb/tools/debugserver/source/MacOSX/MachProcess.mm @@ -472,6 +472,8 @@ static bool FBSAddEventDataToOptions(NSMutableDictionary *options, // And there are some other options at the top level in this dictionary: [options setObject:[NSNumber numberWithBool:YES] forKey:FBSOpenApplicationOptionKeyUnlockDevice]; + [options setObject:[NSNumber numberWithBool:YES] + forKey:FBSOpenApplicationOptionKeyPromptUnlockDevice]; // We have to get the "sequence ID & UUID" for this app bundle path and send // them to FBS: ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Avoid a potential exit(1) in LLVMContext::diagnose() (PR #84992)
jimingham wrote: This makes sense to me. It still gives me the willies that there's exit's lurking in the llvm libraries, but so long as there are, we really do need to side-step them. https://github.com/llvm/llvm-project/pull/84992 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Allow languages to filter breakpoints set by line (PR #83908)
https://github.com/jimingham approved this pull request. LGTM https://github.com/llvm/llvm-project/pull/83908 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Allow languages to filter breakpoints set by line (PR #83908)
https://github.com/felipepiovezan updated
https://github.com/llvm/llvm-project/pull/83908
>From 51307b548d09c34ee06037ccf459110e451970a9 Mon Sep 17 00:00:00 2001
From: Felipe de Azevedo Piovezan
Date: Mon, 4 Mar 2024 09:56:18 -0800
Subject: [PATCH 1/6] [lldb] Allow languages to filter breakpoints set by line
Some languages may create artificial functions that have no real user code, even
though there is line table information for them. One such case is with coroutine
code that receives the CoroSplitter transformation in LLVM IR. That code
transformation creates many different Functions, cloning one Instruction into
many Instructions in many different Functions and copying the associated debug
locations.
It would be difficult to make that pass delete debug locations of cloned
instructions in a language agnostic way (is it even possible?), but LLDB can
ignore certain locations by querying its Language APIs and having it decide
based on, for example, mangling information.
---
lldb/include/lldb/Target/Language.h | 4
lldb/source/Breakpoint/BreakpointResolver.cpp | 12
2 files changed, 16 insertions(+)
diff --git a/lldb/include/lldb/Target/Language.h
b/lldb/include/lldb/Target/Language.h
index 0cbd8a32dccd54..957c40eb7c0772 100644
--- a/lldb/include/lldb/Target/Language.h
+++ b/lldb/include/lldb/Target/Language.h
@@ -339,6 +339,10 @@ class Language : public PluginInterface {
virtual llvm::StringRef GetInstanceVariableName() { return {}; }
+ virtual bool IsInterestingCtxForLineBreakpoint(const SymbolContext &) const {
+return true;
+ }
+
protected:
// Classes that inherit from Language can see and modify these
diff --git a/lldb/source/Breakpoint/BreakpointResolver.cpp
b/lldb/source/Breakpoint/BreakpointResolver.cpp
index bc6348716ef418..876b30c6d76d55 100644
--- a/lldb/source/Breakpoint/BreakpointResolver.cpp
+++ b/lldb/source/Breakpoint/BreakpointResolver.cpp
@@ -23,6 +23,7 @@
#include "lldb/Symbol/CompileUnit.h"
#include "lldb/Symbol/Function.h"
#include "lldb/Symbol/SymbolContext.h"
+#include "lldb/Target/Language.h"
#include "lldb/Target/Target.h"
#include "lldb/Utility/LLDBLog.h"
#include "lldb/Utility/Log.h"
@@ -199,6 +200,15 @@ bool operator<(const SourceLoc lhs, const SourceLoc rhs) {
}
} // namespace
+static void
+ApplyLanguageFilters(llvm::SmallVectorImpl &sc_list) {
+ llvm::erase_if(sc_list, [](SymbolContext &sc) {
+if (Language *lang = Language::FindPlugin(sc.GetLanguage()))
+ return !lang->IsInterestingCtxForLineBreakpoint(sc);
+return false;
+ });
+}
+
void BreakpointResolver::SetSCMatchesByLine(
SearchFilter &filter, SymbolContextList &sc_list, bool skip_prologue,
llvm::StringRef log_ident, uint32_t line, std::optional column) {
@@ -206,6 +216,8 @@ void BreakpointResolver::SetSCMatchesByLine(
for (uint32_t i = 0; i < sc_list.GetSize(); ++i)
all_scs.push_back(sc_list[i]);
+ ApplyLanguageFilters(all_scs);
+
while (all_scs.size()) {
uint32_t closest_line = UINT32_MAX;
>From 7ea76d6c04d063d50da52756179639f2037aa793 Mon Sep 17 00:00:00 2001
From: Felipe de Azevedo Piovezan
Date: Tue, 5 Mar 2024 18:47:51 -0800
Subject: [PATCH 2/6] fixup! add target option
---
lldb/include/lldb/Target/Language.h | 9 ++--
lldb/include/lldb/Target/Target.h | 2 ++
lldb/source/Breakpoint/BreakpointResolver.cpp | 23 +--
lldb/source/Target/Target.cpp | 8 +++
lldb/source/Target/TargetProperties.td| 3 +++
5 files changed, 31 insertions(+), 14 deletions(-)
diff --git a/lldb/include/lldb/Target/Language.h
b/lldb/include/lldb/Target/Language.h
index 957c40eb7c0772..9db9f4e297e114 100644
--- a/lldb/include/lldb/Target/Language.h
+++ b/lldb/include/lldb/Target/Language.h
@@ -339,8 +339,13 @@ class Language : public PluginInterface {
virtual llvm::StringRef GetInstanceVariableName() { return {}; }
- virtual bool IsInterestingCtxForLineBreakpoint(const SymbolContext &) const {
-return true;
+ // Returns true if this SymbolContext should be used when setting breakpoints
+ // by line (number or regex). This is useful for languages that create
+ // artificial functions without any meaningful user code associated with them
+ // (e.g. code that gets expanded in late compilation stages, like by
+ // CoroSplitter).
+ virtual bool IsArtificialCtxForLineBreakpoint(const SymbolContext &) const {
+return false;
}
protected:
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index 8f57358981d4d2..ab04aa57f17300 100644
--- a/lldb/include/lldb/Target/Target.h
+++ b/lldb/include/lldb/Target/Target.h
@@ -258,6 +258,8 @@ class TargetProperties : public Properties {
bool GetDebugUtilityExpression() const;
+ bool GetIgnoreBreakpointsFromLanguageArtificialLocations() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
diff --git a/lldb/source/Breakpoint/
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
https://github.com/rupprecht updated
https://github.com/llvm/llvm-project/pull/84860
>From b5e04eb49c89c8c654305c6ba5db5e2f237bccec Mon Sep 17 00:00:00 2001
From: Jordan Rupprecht
Date: Mon, 11 Mar 2024 11:23:59 -0700
Subject: [PATCH 1/4] [lldb][test] Add `pexpect` category for tests that
`import pexpect`
- Add pexpect category
- Replace skipIfWindows with this
- Make categories apply to test classes too
---
.../Python/lldbsuite/test/decorators.py | 4
lldb/packages/Python/lldbsuite/test/dotest.py | 20 +++
.../Python/lldbsuite/test/lldbpexpect.py | 2 +-
.../Python/lldbsuite/test/test_categories.py | 1 +
.../Python/lldbsuite/test/test_result.py | 10 +-
.../expression/TestExpressionCmd.py | 5 +
.../expression/TestRepeatedExprs.py | 5 +
.../TestFrameVariableResponse.py | 5 +
.../benchmarks/startup/TestStartupDelays.py | 5 +
.../benchmarks/stepping/TestSteppingSpeed.py | 5 +
.../TestCompileRunToBreakpointTurnaround.py | 5 +
.../API/terminal/TestSTTYBeforeAndAfter.py| 2 +-
12 files changed, 34 insertions(+), 35 deletions(-)
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index b691f82b90652c..8e13aa6a13882f 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -409,10 +409,6 @@ def add_test_categories(cat):
cat = test_categories.validate(cat, True)
def impl(func):
-if isinstance(func, type) and issubclass(func, unittest.TestCase):
-raise Exception(
-"@add_test_categories can only be used to decorate a test
method"
-)
try:
if hasattr(func, "categories"):
cat.extend(func.categories)
diff --git a/lldb/packages/Python/lldbsuite/test/dotest.py
b/lldb/packages/Python/lldbsuite/test/dotest.py
index 291d7bad5c0897..268c02e6b49dde 100644
--- a/lldb/packages/Python/lldbsuite/test/dotest.py
+++ b/lldb/packages/Python/lldbsuite/test/dotest.py
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
+)
+
+
def run_suite():
# On MacOS X, check to make sure that domain for com.apple.DebugSymbols
defaults
# does not exist before proceeding to running the test suite.
@@ -1013,6 +1032,7 @@ def run_suite():
checkDebugServerSupport()
checkObjcSupport()
checkForkVForkSupport()
+checkPexpectSupport()
skipped_categories_list = ", ".join(configuration.skip_categories)
print(
diff --git a/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
b/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
index 9d216d90307473..998a080565b6b3 100644
--- a/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
+++ b/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
@@ -10,7 +10,7 @@
@skipIfRemote
-@skipIfWindows # llvm.org/pr22274: need a pexpect replacement for windows
+@add_test_categories(["pexpect"])
class PExpectTest(TestBase):
NO_DEBUG_INFO_TESTCASE = True
PROMPT = "(lldb) "
diff --git a/lldb/packages/Python/lldbsuite/test/test_categories.py
b/lldb/packages/Python/lldbsuite/test/test_categories.py
index 3f8de175e29df3..036bda9c957d11 100644
--- a/lldb/packages/Python/lldbsuite/test/test_categories.py
+++ b/lldb/packages/Python/lldbsuite/test/test_categories.py
@@ -33,6 +33,7 @@
"lldb-server": "Tests related to lldb-server",
"lldb-dap": "Tests for the Debug Adaptor Protocol with lldb-dap",
"llgs": "Tests for the gdb-server functionality of lldb-server",
+"pexpect": "Tests requiring the pexpect library to be available",
"objc": "Tests related to the Objective-C programming language support",
"pyapi": "Tests related to the Python API",
"std-module": "Tests related to importing the std module",
diff --git a/lldb/packages/Python/lldbsuite/test/test_result.py
b/lldb/packages/Python/lldbsuite/test/test_result.py
index 20365f53a67541..cda67ae27d4674 100644
--- a/lldb/packages/Python/lldbsuite/test/test_result.py
+++ b/lldb/packages/Python/lldbsuite/test/test_result.py
@@ -148,9 +148,11 @@ def getCategoriesForTest(self, test):
Gets all the categories for the currently running test meth
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
https://github.com/rupprecht updated
https://github.com/llvm/llvm-project/pull/84860
>From b5e04eb49c89c8c654305c6ba5db5e2f237bccec Mon Sep 17 00:00:00 2001
From: Jordan Rupprecht
Date: Mon, 11 Mar 2024 11:23:59 -0700
Subject: [PATCH 1/5] [lldb][test] Add `pexpect` category for tests that
`import pexpect`
- Add pexpect category
- Replace skipIfWindows with this
- Make categories apply to test classes too
---
.../Python/lldbsuite/test/decorators.py | 4
lldb/packages/Python/lldbsuite/test/dotest.py | 20 +++
.../Python/lldbsuite/test/lldbpexpect.py | 2 +-
.../Python/lldbsuite/test/test_categories.py | 1 +
.../Python/lldbsuite/test/test_result.py | 10 +-
.../expression/TestExpressionCmd.py | 5 +
.../expression/TestRepeatedExprs.py | 5 +
.../TestFrameVariableResponse.py | 5 +
.../benchmarks/startup/TestStartupDelays.py | 5 +
.../benchmarks/stepping/TestSteppingSpeed.py | 5 +
.../TestCompileRunToBreakpointTurnaround.py | 5 +
.../API/terminal/TestSTTYBeforeAndAfter.py| 2 +-
12 files changed, 34 insertions(+), 35 deletions(-)
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index b691f82b90652c..8e13aa6a13882f 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -409,10 +409,6 @@ def add_test_categories(cat):
cat = test_categories.validate(cat, True)
def impl(func):
-if isinstance(func, type) and issubclass(func, unittest.TestCase):
-raise Exception(
-"@add_test_categories can only be used to decorate a test
method"
-)
try:
if hasattr(func, "categories"):
cat.extend(func.categories)
diff --git a/lldb/packages/Python/lldbsuite/test/dotest.py
b/lldb/packages/Python/lldbsuite/test/dotest.py
index 291d7bad5c0897..268c02e6b49dde 100644
--- a/lldb/packages/Python/lldbsuite/test/dotest.py
+++ b/lldb/packages/Python/lldbsuite/test/dotest.py
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
+)
+
+
def run_suite():
# On MacOS X, check to make sure that domain for com.apple.DebugSymbols
defaults
# does not exist before proceeding to running the test suite.
@@ -1013,6 +1032,7 @@ def run_suite():
checkDebugServerSupport()
checkObjcSupport()
checkForkVForkSupport()
+checkPexpectSupport()
skipped_categories_list = ", ".join(configuration.skip_categories)
print(
diff --git a/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
b/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
index 9d216d90307473..998a080565b6b3 100644
--- a/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
+++ b/lldb/packages/Python/lldbsuite/test/lldbpexpect.py
@@ -10,7 +10,7 @@
@skipIfRemote
-@skipIfWindows # llvm.org/pr22274: need a pexpect replacement for windows
+@add_test_categories(["pexpect"])
class PExpectTest(TestBase):
NO_DEBUG_INFO_TESTCASE = True
PROMPT = "(lldb) "
diff --git a/lldb/packages/Python/lldbsuite/test/test_categories.py
b/lldb/packages/Python/lldbsuite/test/test_categories.py
index 3f8de175e29df3..036bda9c957d11 100644
--- a/lldb/packages/Python/lldbsuite/test/test_categories.py
+++ b/lldb/packages/Python/lldbsuite/test/test_categories.py
@@ -33,6 +33,7 @@
"lldb-server": "Tests related to lldb-server",
"lldb-dap": "Tests for the Debug Adaptor Protocol with lldb-dap",
"llgs": "Tests for the gdb-server functionality of lldb-server",
+"pexpect": "Tests requiring the pexpect library to be available",
"objc": "Tests related to the Objective-C programming language support",
"pyapi": "Tests related to the Python API",
"std-module": "Tests related to importing the std module",
diff --git a/lldb/packages/Python/lldbsuite/test/test_result.py
b/lldb/packages/Python/lldbsuite/test/test_result.py
index 20365f53a67541..cda67ae27d4674 100644
--- a/lldb/packages/Python/lldbsuite/test/test_result.py
+++ b/lldb/packages/Python/lldbsuite/test/test_result.py
@@ -148,9 +148,11 @@ def getCategoriesForTest(self, test):
Gets all the categories for the currently running test meth
[Lldb-commits] [lldb] [lldb][test] Add `pexpect` category for tests that `import pexpect` (PR #84860)
@@ -914,6 +914,25 @@ def checkForkVForkSupport():
configuration.skip_categories.append("fork")
+def checkPexpectSupport():
+from lldbsuite.test import lldbplatformutil
+
+platform = lldbplatformutil.getPlatform()
+
+# llvm.org/pr22274: need a pexpect replacement for windows
+if platform in ["windows"]:
+if configuration.verbose:
+print("pexpect tests will be skipped because of unsupported
platform")
+configuration.skip_categories.append("pexpect")
+elif not configuration.shouldSkipBecauseOfCategories(["pexpect"]):
+try:
+import pexpect
+except:
+print(
+"Warning: pexpect is not installed, but pexpect tests are not
being skipped."
rupprecht wrote:
Proposed some more specific wording in
https://github.com/llvm/llvm-project/pull/84860/commits/bcfc5e8f2f18e8eb8f5a96e18c649f3cb2e248c0
But this `elif` block is the least thing that I care about actually landing :)
I think it's helpful, but if this is contentious, it's not a big deal to drop
it and just have a potentially worse error message. So I dropped it altogether
in
https://github.com/llvm/llvm-project/pull/84860/commits/03e13152eac8416c31414b4f469e281c40b80deb.
I can always add it back if needed, either in this PR or as a followup commit.
> So what would happens without this code? Wouldn't we try to import pexpect
> anyway in the relevant test? In other words, do we even need this?
Yep, see the snippet above ([link](#discussion_r1521684820)) for what it'd look
like if we just remove it. Having a specific check+error text here is not
essential, but might help some developer if they aren't aware of how to skip
this category.
> It seems that printing a warning for non-pexpect either adds noise to the
> already verbose test output or will be invisible if the test passes.
Agreed about spamminess; that was changed in
https://github.com/llvm/llvm-project/pull/84860/commits/7f2ec70a35d1f5426afcf354f9b16b0052e81df2
to make this a hard error.
> If the error message is poor enough that we would want to print something
> more actionable, could we do it from the PExpectTest base class?
That would not work for tests that use pexpect outside of `PExpectTest`, which
are:
- lldb/test/API/terminal/TestSTTYBeforeAndAfter.py
- lldb/test/API/macosx/nslog/TestDarwinNSLogOutput.py
- lldb/test/API/benchmarks/expression/TestRepeatedExprs.py (and many other
benchmark tests)
So if the user just runs `ninja check-lldb`, they'd get an actionable message
from PExpectTest-based tests, but a potentially confusing one from
TestSTTYBeforeAndAfter.
https://github.com/llvm/llvm-project/pull/84860
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
