[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei created
https://github.com/llvm/llvm-project/pull/96802
This is the same diff I've put up at many times before. I've been trying to add
some brand new functionality to the LLDB test infrastucture, and we all know
that no good deed goes unpunished. The last attempt was reverted because it
didn't work on the *Fuchsia* build.
My understanding of Fuchsia is based on the only thing I know about it: one of
my friends was moved from the project when it was basically killed during the
Google layoffs in January 2023. I really don't understand the value of the
LLVM/LLDB infra requiring functional Fuchsia builds when it's such a weird,
tiny, _possibly dead_ niche. But I've put this off for long enough, so I've
added a single bit of logic to the test the dwp Makefile.rules assignment to
try using llvm-dwp if it's sitting in one particular place. I would give this
particular change a chance of addressing the fuchsia problem at about 10%, but
I've got to start somewhere.
If anyone can fix the Fuchsia build, or help me figure out how to disable some
tests for Fuchsia (because it actively lies to the infra and claims it's linux
:( ) I'd really appreciate the help.
>From 5db458e34f9db981d78bdac0ba18a97d506fb404 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/12] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
kevinfrei wrote: > @petrhosek Is there a way you can help @kevinfrei validate this on Fuchsia Specifically, I'm adding tests that require dwp/llvm-dwp laying around somewhere, and it looks like Fuchsia doesn't put it in the 'normal' location, and I have no idea how to simply disable an LLDB test for Fuchsia, because it lies to the infra and says it's linux (which is probably the right lie to tell, but I can't find any mechanism to see that it's a flower-flavored variant of Linux...) https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/96802
>From 5db458e34f9db981d78bdac0ba18a97d506fb404 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/13] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
kevinfrei wrote: > What do you mean by "normal" location? "normal" as defined by whoever wrote Makefile.rules originally and thought that was the correct place to file tools like ar and objcopy, I guess? So, on your build machine, do you know where the llvm-dwp file might be installed? Because honestly, I'd rather just use that thing instead of assuming the GNU dwp tool is around. https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/96802
>From 4f0477c3ab3bbd6dcd83e755ad484031301b6a82 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/13] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei ready_for_review https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/96802
>From 95832768ffb3b115e95df19ae5ef14231cad32cc Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/13] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/96802
>From 95832768ffb3b115e95df19ae5ef14231cad32cc Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/13] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/96802
>From 95832768ffb3b115e95df19ae5ef14231cad32cc Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/13] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/96802
>From a7894b87df9afa9f22263e4d8ddec955978dea45 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/13] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/96802
>From 790db0c3dece1699a6cc6e670ddd0f7099840386 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/13] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
kevinfrei wrote: @JDevlieghere could you stamp this so I can see if I've managed to make the tests work on the Fuchsia build (which I can't reproduce locally) https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] DebugInfoD tests: attempt to fix Fuchsia build (PR #96802)
kevinfrei wrote: Friendly ping @JDevlieghere https://github.com/llvm/llvm-project/pull/96802 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Revert "[LLDB] DebugInfoD tests: attempt to fix Fuchsia build" (PR #98101)
https://github.com/kevinfrei created
https://github.com/llvm/llvm-project/pull/98101
Reverts llvm/llvm-project#96802
Attempt #5 fails. It's been 6 months. I despise Makefile.rules and have no
ability to even *detect* these failures without _landing_ a diff. In the mean
time, we have no testing for DWP files at all (and a regression that was
introduced, that I fix with this diff) so I'm going to just remove some of the
tests and try to land it again, but with less testing I guess.
>From 76f102112c8a8a1292929b65ebb7db8b033ee344 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 8 Jul 2024 17:07:16 -0700
Subject: [PATCH] Revert "[LLDB] DebugInfoD tests: attempt to fix Fuchsia build
(#96802)"
This reverts commit 2a7abb04e258542679476fa6527418c34412283c.
---
lldb/include/lldb/Host/Config.h.cmake | 2 -
.../Python/lldbsuite/test/decorators.py | 4 -
.../Python/lldbsuite/test/make/Makefile.rules | 36 +---
lldb/source/API/SBDebugger.cpp| 13 +-
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 38 ++--
.../Plugins/SymbolLocator/CMakeLists.txt | 7 +-
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 29 +--
lldb/test/API/debuginfod/Normal/Makefile | 19 --
.../API/debuginfod/Normal/TestDebuginfod.py | 186 -
lldb/test/API/debuginfod/Normal/main.c| 7 -
lldb/test/API/debuginfod/SplitDWARF/Makefile | 23 --
.../SplitDWARF/TestDebuginfodDWP.py | 196 --
lldb/test/API/debuginfod/SplitDWARF/main.c| 7 -
13 files changed, 22 insertions(+), 545 deletions(-)
delete mode 100644 lldb/test/API/debuginfod/Normal/Makefile
delete mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
delete mode 100644 lldb/test/API/debuginfod/Normal/main.c
delete mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile
delete mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
delete mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c
diff --git a/lldb/include/lldb/Host/Config.h.cmake
b/lldb/include/lldb/Host/Config.h.cmake
index 9e538534086a2..3defa454f6d42 100644
--- a/lldb/include/lldb/Host/Config.h.cmake
+++ b/lldb/include/lldb/Host/Config.h.cmake
@@ -33,8 +33,6 @@
#cmakedefine01 LLDB_ENABLE_LZMA
-#cmakedefine01 LLVM_ENABLE_CURL
-
#cmakedefine01 LLDB_ENABLE_CURSES
#cmakedefine01 CURSES_HAVE_NCURSES_CURSES_H
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index 0e8ca159efd55..ecc7b81035f11 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -1053,10 +1053,6 @@ def _get_bool_config_skip_if_decorator(key):
return unittest.skipIf(not have, "requires " + key)
-def skipIfCurlSupportMissing(func):
-return _get_bool_config_skip_if_decorator("curl")(func)
-
-
def skipIfCursesSupportMissing(func):
return _get_bool_config_skip_if_decorator("curses")(func)
diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
index a7d9d34509f5d..bd8eea3d6f5a0 100644
--- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
+++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
@@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../
#
# GNUWin32 uname gives "windows32" or "server version windows32" while
# some versions of MSYS uname return "MSYS_NT*", but most environments
-# standardize on "Windows_NT", so we'll make it consistent here.
+# standardize on "Windows_NT", so we'll make it consistent here.
# When running tests from Visual Studio, the environment variable isn't
# inherited all the way down to the process spawned for make.
#--
@@ -210,12 +210,6 @@ else
ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES"
DSYM = $(EXE).debug
endif
-
- ifeq "$(MAKE_DWP)" "YES"
- MAKE_DWO := YES
- DWP_NAME = $(EXE).dwp
- DYLIB_DWP_NAME = $(DYLIB_NAME).dwp
- endif
endif
LIMIT_DEBUG_INFO_FLAGS =
@@ -364,17 +358,6 @@ ifneq "$(OS)" "Darwin"
OBJCOPY ?= $(call replace_cc_with,objcopy)
ARCHIVER ?= $(call replace_cc_with,ar)
- # Look for llvm-dwp or gnu dwp
- DWP ?= $(call replace_cc_with,llvm-dwp)
- ifeq ($(wildcard $(DWP)),)
- DWP = $(call replace_cc_with,dwp)
- ifeq ($(wildcard $(DWP)),)
- DWP = $(shell command -v llvm-dwp 2> /dev/null)
- ifeq ($(wildcard $(DWP)),)
- DWP = $(shell command -v dwp 2> /dev/null)
- endif
- endif
- endif
override AR = $(ARCHIVER)
endif
@@ -545,10 +528,6 @@ ifneq "$(CXX)" ""
endif
endif
-ifeq "$(GEN_GNU_BUILD_ID)" "YES"
- LDFLAGS += -Wl,--build-id
-endif
-
#-
[Lldb-commits] [lldb] Revert "[LLDB] DebugInfoD tests: attempt to fix Fuchsia build" (PR #98101)
https://github.com/kevinfrei ready_for_review https://github.com/llvm/llvm-project/pull/98101 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] DebugInfoD tests & fixes (but with dwp testing disabled) (PR #98344)
https://github.com/kevinfrei created
https://github.com/llvm/llvm-project/pull/98344
This is all the tests and fixes I've had percolating since my first attempt at
this in January. After 6 months of trying, I've given up on adding the ability
to test DWP files in LLDB API tests. I've left both the tests (disabled) and
the changes to Makefile.rules in place, in the hopes that someone who can
configure the build bots will be able to enable the tests once a non-borked dwp
tool is widely available.
Other than disabling the DWP tests, this continues to be the same diff that
I've tried to land and [not](https://github.com/llvm/llvm-project/pull/90622)
[revert](https://github.com/llvm/llvm-project/pull/87676)
[five](https://github.com/llvm/llvm-project/pull/86812)
[times](https://github.com/llvm/llvm-project/pull/85693)
[before](https://github.com/llvm/llvm-project/pull/96802). There are a couple
of fixes that the testing exposed, and I've abandoned the DWP tests because I
want to get those fixes finally upstreamed, as without them DebugInfoD is less
useful.
>From 6f885b0c6b255ec7066bbe2a2493608a12c531fb Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/14] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
[Lldb-commits] [lldb] [lldb] DebugInfoD tests & fixes (but with dwp testing disabled) (PR #98344)
kevinfrei wrote: @JDevlieghere or @clayborg if one of you could please approve & land this, I *believe* this disables the tests that caused the previous diff to be reverted. https://github.com/llvm/llvm-project/pull/98344 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Missed disabling the baseline test for DWP stuff (PR #98351)
https://github.com/kevinfrei created https://github.com/llvm/llvm-project/pull/98351 This should disable the failing test on the ubuntu build bots @JDevlieghere (I forgot to disable the 'baseline' test, as it tests the debugger's basic handling of DWP files, but again, the API infrastructure doesn't quite support DWP generation) https://github.com/llvm/llvm-project/pull/98344#issuecomment-2221000566 >From 9840146580768d72eed4ec43941751d5a3859f9c Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Wed, 10 Jul 2024 10:02:47 -0700 Subject: [PATCH] Missed disabling the baseline test for DWP stuff --- lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py b/lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py index 129f402c6a124..437c83a820fb7 100644 --- a/lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py +++ b/lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py @@ -25,7 +25,7 @@ class DebugInfodDWPTests(TestBase): # No need to try every flavor of debug inf. NO_DEBUG_INFO_TESTCASE = True -@skipUnlessPlatform(["linux", "freebsd"]) + @skipUnlessPlatform(["linux_freebsd_but_old_dwp_tools_on_build_bots_are_broken"]) def test_normal_stripped(self): """ Validate behavior with a stripped binary, no symbols or symbol locator. ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] DebugInfoD tests & fixes (but with dwp testing disabled) (PR #98344)
kevinfrei wrote: "fixed forward" with [PR98351](https://github.com/llvm/llvm-project/pull/98351/commits) if someone can approve & land that (it just disabled the baseline test for DWP stuff, as well: I missed it because the dwp tests run on all the machines I have access to :/ ) https://github.com/llvm/llvm-project/pull/98344 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] DebugInfoD tests & fixes (but with dwp testing disabled) (PR #98344)
kevinfrei wrote: > @kevinfrei , I'm so sorry to tell you that I have reverted this patch. The > revert commit is > [27b2f4f](https://github.com/llvm/llvm-project/commit/27b2f4f861b8aeeabc4eb1a97649062de8fa3992) > and I left some notes there, which I also copy here: > > The patch #98344 causes a crash in LLDB when parsing some files like > `numpy.libs/libgfortran-daac5196.so.5.0.0` on graviton (you can download it > in > https://drive.google.com/file/d/12ygLjJwWpzdYsrzBPp1JGiFHxcgM0-XY/view?usp=drive_link > if you want to troubleshoot yourself). > > ``` > The assert that is hit is the following: > > ``` > llvm-project/lldb/source/Plugins/ObjectFile/ELF/ObjectFileELF.cpp:2452: > std::pair lldb_private::AddressClass> > > ObjectFileELF::ParseSymbolTable(lldb_private::Symtab*, lldb::user_id_t, > lldb_private::Section*): Assertion `strtab->GetObjectFile() == this' failed. > ``` Reverted with a repro instead of some inscrutable "This didn't work on a build machine configured by someone 4 years ago who promptly forgot what they did and there's no documentation, but here's this log: good luck!" is glorious! I'll get on it- thanks 😃 https://github.com/llvm/llvm-project/pull/98344 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
https://github.com/kevinfrei created
https://github.com/llvm/llvm-project/pull/99362
@walter-erquinigo found the the [PR with testing and a fix for
DebugInfoD](https://github.com/llvm/llvm-project/pull/98344) caused an issue
when working with stripped binaries.
The issue is that when you're working with split-dwarf, there are *3* possible
files: The stripped binary the user is debugging, the "only-keep-debug" *or*
unstripped binary, plus the `.dwp` file. The debuginfod plugin should provide
the unstripped/OKD binary. However, if the debuginfod plugin fails, the default
symbol locator plugin will just return the stripped binary, which doesn't help.
So, to address that, the SymbolVendorELF code checks to see if the
SymbolLocator's ExecutableObjectFile request returned the same file, and bails
if that's the case. You can see the specific diff as the second commit in the
PR.
I'm investigating adding a test: I can't quite get a simple repro, and I'm
unwilling to make any additional changes to Makefile.rules to this diff, for
Pavlovian reasons.
>From 1494617c82065e44617b013c96ba51e578706d75 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 1/2] Trying to deal with Linux AArch64 test failures :/
Reapply "DebugInfoD tests + fixing issues exposed by tests (#85693)"
This reverts commit 7fc2fbb3f1961e0ad0722c2d749ddd6264195a1c.
Switched to using LLDB to get UUID at @clayborg's suggestion
Disable tests on non-x86 Linux platforms, as they appear to fail on AArch64/Arm
buildbots
Moved the @skipIf to each test, off of the class itself
Added CURL dependency to lit configuration 'stuff'
Fixed comments in the tests
Fix stupid formatter issue
Updated to respond to (very late) code review feedback
Fixed the script to acquire the UUID without creating a target
Updated tests to not run on Mac/Windows
Added LLVM_ENABLE_CURL to the config.h.cmake to work in test's @skipIf thing
Attempting to prefer llvm-dwp over gnu's dwp
Disabled the DWP tests
Missed disabling the baseline test for DWP stuff
---
lldb/include/lldb/Host/Config.h.cmake | 2 +
.../Python/lldbsuite/test/decorators.py | 4 +
.../Python/lldbsuite/test/make/Makefile.rules | 36 +++-
lldb/source/API/SBDebugger.cpp| 13 +-
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 38 ++--
.../Plugins/SymbolLocator/CMakeLists.txt | 7 +-
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 29 ++-
lldb/test/API/debuginfod/Normal/Makefile | 19 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 186 +
lldb/test/API/debuginfod/Normal/main.c| 7 +
lldb/test/API/debuginfod/SplitDWARF/Makefile | 23 ++
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
lldb/test/API/debuginfod/SplitDWARF/main.c| 7 +
13 files changed, 545 insertions(+), 22 deletions(-)
create mode 100644 lldb/test/API/debuginfod/Normal/Makefile
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/Normal/main.c
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c
diff --git a/lldb/include/lldb/Host/Config.h.cmake
b/lldb/include/lldb/Host/Config.h.cmake
index 3defa454f6d42..9e538534086a2 100644
--- a/lldb/include/lldb/Host/Config.h.cmake
+++ b/lldb/include/lldb/Host/Config.h.cmake
@@ -33,6 +33,8 @@
#cmakedefine01 LLDB_ENABLE_LZMA
+#cmakedefine01 LLVM_ENABLE_CURL
+
#cmakedefine01 LLDB_ENABLE_CURSES
#cmakedefine01 CURSES_HAVE_NCURSES_CURSES_H
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index ecc7b81035f11..0e8ca159efd55 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -1053,6 +1053,10 @@ def _get_bool_config_skip_if_decorator(key):
return unittest.skipIf(not have, "requires " + key)
+def skipIfCurlSupportMissing(func):
+return _get_bool_config_skip_if_decorator("curl")(func)
+
+
def skipIfCursesSupportMissing(func):
return _get_bool_config_skip_if_decorator("curses")(func)
diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
index 3d562285ce9cc..d1a2de8b2478a 100644
--- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
+++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
@@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../
#
# GNUWin32 uname gives "windows32" or "server version windows32" while
# some versions of MSYS uname return "MSYS_NT*", but most environments
-# standardize on "Windows_NT", so we'll make it consistent here.
+# standardize on "Windows_NT", so we'll make it consistent here.
# When running tests from Visual Studio, the environment v
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/99362
>From 1494617c82065e44617b013c96ba51e578706d75 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 1/2] Trying to deal with Linux AArch64 test failures :/
Reapply "DebugInfoD tests + fixing issues exposed by tests (#85693)"
This reverts commit 7fc2fbb3f1961e0ad0722c2d749ddd6264195a1c.
Switched to using LLDB to get UUID at @clayborg's suggestion
Disable tests on non-x86 Linux platforms, as they appear to fail on AArch64/Arm
buildbots
Moved the @skipIf to each test, off of the class itself
Added CURL dependency to lit configuration 'stuff'
Fixed comments in the tests
Fix stupid formatter issue
Updated to respond to (very late) code review feedback
Fixed the script to acquire the UUID without creating a target
Updated tests to not run on Mac/Windows
Added LLVM_ENABLE_CURL to the config.h.cmake to work in test's @skipIf thing
Attempting to prefer llvm-dwp over gnu's dwp
Disabled the DWP tests
Missed disabling the baseline test for DWP stuff
---
lldb/include/lldb/Host/Config.h.cmake | 2 +
.../Python/lldbsuite/test/decorators.py | 4 +
.../Python/lldbsuite/test/make/Makefile.rules | 36 +++-
lldb/source/API/SBDebugger.cpp| 13 +-
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 38 ++--
.../Plugins/SymbolLocator/CMakeLists.txt | 7 +-
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 29 ++-
lldb/test/API/debuginfod/Normal/Makefile | 19 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 186 +
lldb/test/API/debuginfod/Normal/main.c| 7 +
lldb/test/API/debuginfod/SplitDWARF/Makefile | 23 ++
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
lldb/test/API/debuginfod/SplitDWARF/main.c| 7 +
13 files changed, 545 insertions(+), 22 deletions(-)
create mode 100644 lldb/test/API/debuginfod/Normal/Makefile
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/Normal/main.c
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c
diff --git a/lldb/include/lldb/Host/Config.h.cmake
b/lldb/include/lldb/Host/Config.h.cmake
index 3defa454f6d42..9e538534086a2 100644
--- a/lldb/include/lldb/Host/Config.h.cmake
+++ b/lldb/include/lldb/Host/Config.h.cmake
@@ -33,6 +33,8 @@
#cmakedefine01 LLDB_ENABLE_LZMA
+#cmakedefine01 LLVM_ENABLE_CURL
+
#cmakedefine01 LLDB_ENABLE_CURSES
#cmakedefine01 CURSES_HAVE_NCURSES_CURSES_H
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index ecc7b81035f11..0e8ca159efd55 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -1053,6 +1053,10 @@ def _get_bool_config_skip_if_decorator(key):
return unittest.skipIf(not have, "requires " + key)
+def skipIfCurlSupportMissing(func):
+return _get_bool_config_skip_if_decorator("curl")(func)
+
+
def skipIfCursesSupportMissing(func):
return _get_bool_config_skip_if_decorator("curses")(func)
diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
index 3d562285ce9cc..d1a2de8b2478a 100644
--- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
+++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
@@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../
#
# GNUWin32 uname gives "windows32" or "server version windows32" while
# some versions of MSYS uname return "MSYS_NT*", but most environments
-# standardize on "Windows_NT", so we'll make it consistent here.
+# standardize on "Windows_NT", so we'll make it consistent here.
# When running tests from Visual Studio, the environment variable isn't
# inherited all the way down to the process spawned for make.
#--
@@ -213,6 +213,12 @@ else
ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES"
DSYM = $(EXE).debug
endif
+
+ ifeq "$(MAKE_DWP)" "YES"
+ MAKE_DWO := YES
+ DWP_NAME = $(EXE).dwp
+ DYLIB_DWP_NAME = $(DYLIB_NAME).dwp
+ endif
endif
LIMIT_DEBUG_INFO_FLAGS =
@@ -361,6 +367,17 @@ ifneq "$(OS)" "Darwin"
OBJCOPY ?= $(call replace_cc_with,objcopy)
ARCHIVER ?= $(call replace_cc_with,ar)
+ # Look for llvm-dwp or gnu dwp
+ DWP ?= $(call replace_cc_with,llvm-dwp)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(call replace_cc_with,dwp)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(shell command -v llvm-dwp 2> /dev/null)
+ ifeq ($(wildcard $(DWP)),)
+
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
@@ -106,12 +106,19 @@ SymbolVendorELF::CreateInstance(const lldb::ModuleSP
&module_sp,
FileSpec dsym_fspec =
PluginManager::LocateExecutableSymbolFile(module_spec, search_paths);
if (!dsym_fspec || IsDwpSymbolFile(module_sp, dsym_fspec)) {
-// If we have a stripped binary or if we got a DWP file, we should prefer
-// symbols in the executable acquired through a plugin.
+// If we have a stripped binary or if we have a DWP file, symbol locator
+// plugins may be able to give us an unstripped binary or an
+// 'only-keep-debug' stripped file.
ModuleSpec unstripped_spec =
PluginManager::LocateExecutableObjectFile(module_spec);
if (!unstripped_spec)
return nullptr;
+// If we got the original binary back from the plugin manager, the plugins
+// didn't find anything more useful that the stripped binary. (The default
+// symbol locator plugin returns the original binary if it can't find
+// anything else.)
+if (unstripped_spec.GetFileSpec() == module_spec.GetFileSpec())
kevinfrei wrote:
Cleaned the comment up a bit.
https://github.com/llvm/llvm-project/pull/99362
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
kevinfrei wrote: > This LGTM, although I'd rather have someone like @clayborg taking a look Yeah, I try not to bug him too much, since he's a coworker and I already get his time in a few recurring meetings. I might actually try to grab a VC with him to see if there's some other scenario I'm missing. https://github.com/llvm/llvm-project/pull/99362 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/99362
>From b3fe99591b7bc6dad2e9cc46bd4fc2c2f87d1eff Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 1/2] Trying to deal with Linux AArch64 test failures :/
Reapply "DebugInfoD tests + fixing issues exposed by tests (#85693)"
This reverts commit 7fc2fbb3f1961e0ad0722c2d749ddd6264195a1c.
Switched to using LLDB to get UUID at @clayborg's suggestion
Disable tests on non-x86 Linux platforms, as they appear to fail on AArch64/Arm
buildbots
Moved the @skipIf to each test, off of the class itself
Added CURL dependency to lit configuration 'stuff'
Fixed comments in the tests
Fix stupid formatter issue
Updated to respond to (very late) code review feedback
Fixed the script to acquire the UUID without creating a target
Updated tests to not run on Mac/Windows
Added LLVM_ENABLE_CURL to the config.h.cmake to work in test's @skipIf thing
Attempting to prefer llvm-dwp over gnu's dwp
Disabled the DWP tests
Missed disabling the baseline test for DWP stuff
---
lldb/include/lldb/Host/Config.h.cmake | 2 +
.../Python/lldbsuite/test/decorators.py | 4 +
.../Python/lldbsuite/test/make/Makefile.rules | 34 +++
lldb/source/API/SBDebugger.cpp| 13 +-
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 38 ++--
.../Plugins/SymbolLocator/CMakeLists.txt | 7 +-
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 29 ++-
lldb/test/API/debuginfod/Normal/Makefile | 19 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 186 +
lldb/test/API/debuginfod/Normal/main.c| 7 +
lldb/test/API/debuginfod/SplitDWARF/Makefile | 23 ++
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
lldb/test/API/debuginfod/SplitDWARF/main.c| 7 +
13 files changed, 544 insertions(+), 21 deletions(-)
create mode 100644 lldb/test/API/debuginfod/Normal/Makefile
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/Normal/main.c
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c
diff --git a/lldb/include/lldb/Host/Config.h.cmake
b/lldb/include/lldb/Host/Config.h.cmake
index 3defa454f6d42..9e538534086a2 100644
--- a/lldb/include/lldb/Host/Config.h.cmake
+++ b/lldb/include/lldb/Host/Config.h.cmake
@@ -33,6 +33,8 @@
#cmakedefine01 LLDB_ENABLE_LZMA
+#cmakedefine01 LLVM_ENABLE_CURL
+
#cmakedefine01 LLDB_ENABLE_CURSES
#cmakedefine01 CURSES_HAVE_NCURSES_CURSES_H
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index ecc7b81035f11..0e8ca159efd55 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -1053,6 +1053,10 @@ def _get_bool_config_skip_if_decorator(key):
return unittest.skipIf(not have, "requires " + key)
+def skipIfCurlSupportMissing(func):
+return _get_bool_config_skip_if_decorator("curl")(func)
+
+
def skipIfCursesSupportMissing(func):
return _get_bool_config_skip_if_decorator("curses")(func)
diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
index be3ad684dd736..7ecfa8aac520b 100644
--- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
+++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
@@ -190,6 +190,12 @@ else
ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES"
DSYM = $(EXE).debug
endif
+
+ ifeq "$(MAKE_DWP)" "YES"
+ MAKE_DWO := YES
+ DWP_NAME = $(EXE).dwp
+ DYLIB_DWP_NAME = $(DYLIB_NAME).dwp
+ endif
endif
LIMIT_DEBUG_INFO_FLAGS =
@@ -338,6 +344,17 @@ ifneq "$(OS)" "Darwin"
OBJCOPY ?= $(call replace_cc_with,objcopy)
ARCHIVER ?= $(call replace_cc_with,ar)
+ # Look for llvm-dwp or gnu dwp
+ DWP ?= $(call replace_cc_with,llvm-dwp)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(call replace_cc_with,dwp)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(shell command -v llvm-dwp 2> /dev/null)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(shell command -v dwp 2> /dev/null)
+ endif
+ endif
+ endif
override AR = $(ARCHIVER)
endif
@@ -508,6 +525,10 @@ ifneq "$(CXX)" ""
endif
endif
+ifeq "$(GEN_GNU_BUILD_ID)" "YES"
+ LDFLAGS += -Wl,--build-id
+endif
+
#--
# DYLIB_ONLY variable can be used to skip the building of a.out.
# See the sections below regarding dSYM file as well as the building of
@@ -546,11 +567,18 @@ else
endif
else
ifeq "$(SP
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/99362
>From f498534a60636a6b8ee06f27dcd04a5074537180 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 1/2] Trying to deal with Linux AArch64 test failures :/
Reapply "DebugInfoD tests + fixing issues exposed by tests (#85693)"
This reverts commit 7fc2fbb3f1961e0ad0722c2d749ddd6264195a1c.
Switched to using LLDB to get UUID at @clayborg's suggestion
Disable tests on non-x86 Linux platforms, as they appear to fail on AArch64/Arm
buildbots
Moved the @skipIf to each test, off of the class itself
Added CURL dependency to lit configuration 'stuff'
Fixed comments in the tests
Fix stupid formatter issue
Updated to respond to (very late) code review feedback
Fixed the script to acquire the UUID without creating a target
Updated tests to not run on Mac/Windows
Added LLVM_ENABLE_CURL to the config.h.cmake to work in test's @skipIf thing
Attempting to prefer llvm-dwp over gnu's dwp
Disabled the DWP tests
Missed disabling the baseline test for DWP stuff
---
lldb/include/lldb/Host/Config.h.cmake | 2 +
.../Python/lldbsuite/test/decorators.py | 4 +
.../Python/lldbsuite/test/make/Makefile.rules | 34 +++
lldb/source/API/SBDebugger.cpp| 13 +-
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 38 ++--
.../Plugins/SymbolLocator/CMakeLists.txt | 7 +-
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 29 ++-
lldb/test/API/debuginfod/Normal/Makefile | 19 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 186 +
lldb/test/API/debuginfod/Normal/main.c| 7 +
lldb/test/API/debuginfod/SplitDWARF/Makefile | 23 ++
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
lldb/test/API/debuginfod/SplitDWARF/main.c| 7 +
13 files changed, 544 insertions(+), 21 deletions(-)
create mode 100644 lldb/test/API/debuginfod/Normal/Makefile
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/Normal/main.c
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c
diff --git a/lldb/include/lldb/Host/Config.h.cmake
b/lldb/include/lldb/Host/Config.h.cmake
index 3defa454f6d42..9e538534086a2 100644
--- a/lldb/include/lldb/Host/Config.h.cmake
+++ b/lldb/include/lldb/Host/Config.h.cmake
@@ -33,6 +33,8 @@
#cmakedefine01 LLDB_ENABLE_LZMA
+#cmakedefine01 LLVM_ENABLE_CURL
+
#cmakedefine01 LLDB_ENABLE_CURSES
#cmakedefine01 CURSES_HAVE_NCURSES_CURSES_H
diff --git a/lldb/packages/Python/lldbsuite/test/decorators.py
b/lldb/packages/Python/lldbsuite/test/decorators.py
index ecc7b81035f11..0e8ca159efd55 100644
--- a/lldb/packages/Python/lldbsuite/test/decorators.py
+++ b/lldb/packages/Python/lldbsuite/test/decorators.py
@@ -1053,6 +1053,10 @@ def _get_bool_config_skip_if_decorator(key):
return unittest.skipIf(not have, "requires " + key)
+def skipIfCurlSupportMissing(func):
+return _get_bool_config_skip_if_decorator("curl")(func)
+
+
def skipIfCursesSupportMissing(func):
return _get_bool_config_skip_if_decorator("curses")(func)
diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
index be3ad684dd736..7ecfa8aac520b 100644
--- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
+++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules
@@ -190,6 +190,12 @@ else
ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES"
DSYM = $(EXE).debug
endif
+
+ ifeq "$(MAKE_DWP)" "YES"
+ MAKE_DWO := YES
+ DWP_NAME = $(EXE).dwp
+ DYLIB_DWP_NAME = $(DYLIB_NAME).dwp
+ endif
endif
LIMIT_DEBUG_INFO_FLAGS =
@@ -338,6 +344,17 @@ ifneq "$(OS)" "Darwin"
OBJCOPY ?= $(call replace_cc_with,objcopy)
ARCHIVER ?= $(call replace_cc_with,ar)
+ # Look for llvm-dwp or gnu dwp
+ DWP ?= $(call replace_cc_with,llvm-dwp)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(call replace_cc_with,dwp)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(shell command -v llvm-dwp 2> /dev/null)
+ ifeq ($(wildcard $(DWP)),)
+ DWP = $(shell command -v dwp 2> /dev/null)
+ endif
+ endif
+ endif
override AR = $(ARCHIVER)
endif
@@ -508,6 +525,10 @@ ifneq "$(CXX)" ""
endif
endif
+ifeq "$(GEN_GNU_BUILD_ID)" "YES"
+ LDFLAGS += -Wl,--build-id
+endif
+
#--
# DYLIB_ONLY variable can be used to skip the building of a.out.
# See the sections below regarding dSYM file as well as the building of
@@ -546,11 +567,18 @@ else
endif
else
ifeq "$(SP
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
kevinfrei wrote: @JDevlieghere I think this still wants your approval. I squashed all the original commits, but kept the single final change to address the issue Walter found as a separate commit. https://github.com/llvm/llvm-project/pull/99362 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb][debuginfod] Fix the DebugInfoD PR that caused issues when working with stripped binaries (PR #99362)
kevinfrei wrote: Ping @JDevlieghere https://github.com/llvm/llvm-project/pull/99362 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Read and store gnu build id from loaded core file (PR #92492)
@@ -983,6 +1005,73 @@ llvm::Error
ProcessElfCore::ParseThreadContextsFromNoteSegment(
}
}
+bool ProcessElfCore::IsElf(const lldb::addr_t address) {
+ uint8_t buf[4];
+ Status error;
+ size_t byte_read = ReadMemory(address, buf, 4, error);
+ if (byte_read != 4)
+return false;
+ return elf::ELFHeader::MagicBytesMatch(buf);
+}
+
+std::optional ProcessElfCore::FindNote(const lldb::addr_t address,
kevinfrei wrote:
I'd suggest instead, implement support for the Note section in the same fashion
as the other ELF components, with a "parse" method and the like (check out
ThreadElfCore.h) and integrated it into the top level loop, same as the other
Program Header structures.
https://github.com/llvm/llvm-project/pull/92492
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Read and store gnu build id from loaded core file (PR #92492)
@@ -271,6 +282,17 @@ Status ProcessElfCore::DoLoadCore() {
return error;
}
+void ProcessElfCore::UpdateBuildIdForNTFileEntries() {
+ if (!m_nt_file_entries.empty()) {
kevinfrei wrote:
I'd suggest adding it to that pass as well, though it will be a little bit
messier, as NT_PRPSINFO and NT_GNU_BUILD_ID have the same values, so they
depend on the Name field (CORE and GNU, respectively...). Do a check for the
GNU name at the very top (line 902ish?) and handle NT_GNU_BUILD_ID's before the
rest of that loop body, or dealing with PRPSINFO logic will get much messier...
https://github.com/llvm/llvm-project/pull/92492
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Read and store gnu build id from loaded core file (PR #92492)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/92492 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Read and store gnu build id from loaded core file (PR #92492)
https://github.com/kevinfrei requested changes to this pull request. https://github.com/llvm/llvm-project/pull/92492 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Read and store gnu build id from loaded core file (PR #92492)
@@ -117,6 +117,13 @@ class ProcessElfCore : public
lldb_private::PostMortemProcess {
lldb::addr_t end;
lldb::addr_t file_ofs;
std::string path;
+lldb_private::UUID uuid; //.note.gnu.build-id
+ };
+
+ struct Section_Note {
+uint32_t namesz;
+uint32_t descsz;
+uint32_t type;
kevinfrei wrote:
Move this to the ThreadElfCore.h header and add a 'parse' method to it, so the
UUID is populated much like the PRPSINFO data and other useful info from the
coredump.
https://github.com/llvm/llvm-project/pull/92492
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Added Debuginfod tests and fixed a couple issues (PR #92572)
https://github.com/kevinfrei created
https://github.com/llvm/llvm-project/pull/92572
Here we go with attempt #5.
Again, no changes to the LLDB code diffs that have been looked at several times.
For the tests, I added a `@skipIfCurlSupportMissing` annotation so that the
Debuginfod mocked server stuff won't run, and I also disabled non-Linux/FreeBSD
hosts altogether, as they fail for platform reasons on macOS and Windows. In
addition, I updated the process for extracting the GNU BuildID to no create a
target, per some feedback on the previous diff.
>From 8a544973fcdb0391f3342b6e6e44b4cf422d0429 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 01/12] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a..a881218a56cef 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 0..a662fb9fc6e68
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.
[Lldb-commits] [lldb] [lldb] Added Debuginfod tests and fixed a couple issues (PR #92572)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/92572 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Added Debuginfod tests and fixed a couple issues (PR #92572)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/92572 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Added Debuginfod tests and fixed a couple issues (PR #92572)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/92572 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Added Debuginfod tests and fixed a couple issues (PR #92572)
https://github.com/kevinfrei ready_for_review https://github.com/llvm/llvm-project/pull/92572 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Added Debuginfod tests and fixed a couple issues (PR #92572)
kevinfrei wrote: FYI: The previous diff exposed a bug in some ARM code that @DavidSpicket fixed in #91585, but (for understandable reasons) he didn't want to be left holding the back on restoring this diff, so I integrated some post-land feedback from @labath and put the diff back up for review. https://github.com/llvm/llvm-project/pull/92572 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Add 'FindInMemory()' overload for PostMortemProcess. (PR #102536)
@@ -2835,6 +2835,34 @@ void PruneThreadPlans();
AddressRanges &matches, size_t alignment,
size_t max_matches);
+ template
+ lldb::addr_t FindInMemoryGeneric(IT &&iterator, lldb::addr_t low,
+ lldb::addr_t high, const uint8_t *buf,
+ size_t size) {
+const size_t region_size = high - low;
+
+if (region_size < size)
+ return LLDB_INVALID_ADDRESS;
+
+std::vector bad_char_heuristic(256, size);
+
+for (size_t idx = 0; idx < size - 1; idx++) {
+ decltype(bad_char_heuristic)::size_type bcu_idx = buf[idx];
+ bad_char_heuristic[bcu_idx] = size - idx - 1;
+}
+for (size_t s = 0; s <= (region_size - size);) {
+ int64_t j = size - 1;
+ while (j >= 0 && buf[j] == iterator[s + j])
+j--;
+ if (j < 0)
+return low + s;
+ else
+s += bad_char_heuristic[iterator[s + size - 1]];
+}
+
+return LLDB_INVALID_ADDRESS;
+ }
+
kevinfrei wrote:
This algorithm is probably a performance win, but obscures the simplicity of
the function. Obscuring simplicity for good reasons requires an explanation. At
least name drop BMH in the comments or link to the wikipedia page
(https://en.wikipedia.org/wiki/Boyer%E2%80%93Moore%E2%80%93Horspool_algorithm)
https://github.com/llvm/llvm-project/pull/102536
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
@@ -0,0 +1,142 @@
+//===-- SymbolLocatorDebuginfod.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 "SymbolLocatorDebuginfod.h"
+
+#include "lldb/Core/PluginManager.h"
+#include "lldb/Utility/Args.h"
+
+#include "llvm/Debuginfod/Debuginfod.h"
+#include "llvm/Debuginfod/HTTPClient.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+LLDB_PLUGIN_DEFINE(SymbolLocatorDebuginfod)
+
+namespace {
+
+#define LLDB_PROPERTIES_symbollocatordebuginfod
+#include "SymbolLocatorDebuginfodProperties.inc"
+
+enum {
+#define LLDB_PROPERTIES_symbollocatordebuginfod
+#include "SymbolLocatorDebuginfodPropertiesEnum.inc"
+};
+
+class PluginProperties : public Properties {
+public:
+ static llvm::StringRef GetSettingName() {
+return SymbolLocatorDebuginfod::GetPluginNameStatic();
+ }
+
+ PluginProperties() {
+m_collection_sp =
std::make_shared(GetSettingName());
+m_collection_sp->Initialize(g_symbollocatordebuginfod_properties);
+
+// We need to read the default value first to read the environment
variable.
+llvm::SmallVector urls = llvm::getDefaultDebuginfodUrls();
+Args arg_urls{urls};
+m_collection_sp->SetPropertyAtIndexFromArgs(ePropertyServerURLs, arg_urls);
+
+m_collection_sp->SetValueChangedCallback(
+ePropertyServerURLs, [this] { ServerURLsChangedCallback(); });
+ }
+
+ Args GetDebugInfoDURLs() const {
+Args urls;
+m_collection_sp->GetPropertyAtIndexAsArgs(ePropertyServerURLs, urls);
+return urls;
+ }
+
+private:
+ void ServerURLsChangedCallback() {
+Args urls = GetDebugInfoDURLs();
+llvm::SmallVector dbginfod_urls;
+llvm::transform(urls, dbginfod_urls.end(),
+[](const auto &obj) { return obj.ref(); });
+llvm::setDefaultDebuginfodUrls(dbginfod_urls);
+ }
+};
+
+} // namespace
+
+static PluginProperties &GetGlobalPluginProperties() {
+ static PluginProperties g_settings;
+ return g_settings;
+}
+
+SymbolLocatorDebuginfod::SymbolLocatorDebuginfod() : SymbolLocator() {}
+
+void SymbolLocatorDebuginfod::Initialize() {
+ static llvm::once_flag g_once_flag;
+
+ llvm::call_once(g_once_flag, []() {
+PluginManager::RegisterPlugin(
+GetPluginNameStatic(), GetPluginDescriptionStatic(), CreateInstance,
+LocateExecutableObjectFile, LocateExecutableSymbolFile, nullptr,
+nullptr, SymbolLocatorDebuginfod::DebuggerInitialize);
+llvm::HTTPClient::initialize();
+ });
+}
+
+void SymbolLocatorDebuginfod::DebuggerInitialize(Debugger &debugger) {
+ if (!PluginManager::GetSettingForSymbolLocatorPlugin(
+ debugger, PluginProperties::GetSettingName())) {
+const bool is_global_setting = true;
+PluginManager::CreateSettingForSymbolLocatorPlugin(
+debugger, GetGlobalPluginProperties().GetValueProperties(),
+"Properties for the Debuginfod Symbol Locator plug-in.",
+is_global_setting);
+ }
+}
+
+void SymbolLocatorDebuginfod::Terminate() {
+ PluginManager::UnregisterPlugin(CreateInstance);
+ llvm::HTTPClient::cleanup();
+}
+
+llvm::StringRef SymbolLocatorDebuginfod::GetPluginDescriptionStatic() {
+ return "Debuginfod symbol locator.";
+}
+
+SymbolLocator *SymbolLocatorDebuginfod::CreateInstance() {
+ return new SymbolLocatorDebuginfod();
+}
+
+std::optional SymbolLocatorDebuginfod::LocateExecutableObjectFile(
+const ModuleSpec &module_spec) {
+ const UUID &module_uuid = module_spec.GetUUID();
+ if (module_uuid.IsValid() && llvm::canUseDebuginfod()) {
+llvm::object::BuildID build_id(module_uuid.GetBytes());
+llvm::Expected result =
+llvm::getCachedOrDownloadExecutable(build_id);
+if (result)
+ return FileSpec(*result);
+// An error here should be logged as a failure in the Debuginfod library,
+// so just consume it here
+consumeError(result.takeError());
+ }
+ return {};
+}
+
+std::optional SymbolLocatorDebuginfod::LocateExecutableSymbolFile(
+const ModuleSpec &module_spec, const FileSpecList &default_search_paths) {
+ const UUID &module_uuid = module_spec.GetUUID();
+ if (module_uuid.IsValid() && llvm::canUseDebuginfod()) {
+llvm::object::BuildID build_id(module_uuid.GetBytes());
+llvm::Expected result =
+llvm::getCachedOrDownloadDebuginfo(build_id);
+if (result)
+ return FileSpec(*result);
+// An error here should be logged as a failure in the Debuginfod library,
+// so just consume it here
+consumeError(result.takeError());
+ }
kevinfrei wrote:
Not quite: They're calling different functions
(llvm::getCachedOrDownloadDebugInfo vs. llvm::getCachedOrDownloadExecutable)
but I could factor that out and pass the function pointers instead.
ht
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/70996
>From b04c85dbed0b369e747aa2a3823789203156736b Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Wed, 18 Oct 2023 14:37:34 -0700
Subject: [PATCH 1/7] DEBUGINFOD based DWP acquisition for LLDB
Summary:
I've plumbed the LLVM DebugInfoD client into LLDB, and added automatic
downloading of
DWP files to the SymbolFileDWARF.cpp plugin. If you have `DEBUGINFOD_URLS` set
to a
space delimited set of web servers, LLDB will try to use them as a last resort
when
searching for DWP files. If you do *not* have that environment variable set,
nothing
should be changed. There's also a setting, per Greg Clayton's request, that will
override the env variable, or can be used instead of the env var. This setting
is the
reason for the additional API added to the llvm's Debuginfod library.
Test Plan:
Suggestions are welcome here. I should probably have some positive and negative
tests,
but I wanted to get the diff up for people who have a clue what they're doing
to rip it
to pieces before spending too much time validating my implementation.
---
lldb/include/lldb/Target/Target.h | 3 +++
lldb/source/Core/CoreProperties.td| 2 +-
lldb/source/Core/Debugger.cpp | 5
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 1 +
lldb/source/Symbol/CMakeLists.txt | 1 +
lldb/source/Target/Target.cpp | 19 +-
lldb/source/Target/TargetProperties.td| 4 +++
llvm/include/llvm/Debuginfod/Debuginfod.h | 4 +++
llvm/lib/Debuginfod/Debuginfod.cpp| 26 ++-
9 files changed, 57 insertions(+), 8 deletions(-)
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index c37682e2a03859f..7f10f0409fb1315 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;
+ Args GetDebugInfoDURLs() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
@@ -270,6 +272,7 @@ class TargetProperties : public Properties {
void DisableASLRValueChangedCallback();
void InheritTCCValueChangedCallback();
void DisableSTDIOValueChangedCallback();
+ void DebugInfoDURLsChangedCallback();
// Settings checker for target.jit-save-objects-dir:
void CheckJITObjectsDir();
diff --git a/lldb/source/Core/CoreProperties.td
b/lldb/source/Core/CoreProperties.td
index 92884258347e9be..865030b0133bbb2 100644
--- a/lldb/source/Core/CoreProperties.td
+++ b/lldb/source/Core/CoreProperties.td
@@ -4,7 +4,7 @@ let Definition = "modulelist" in {
def EnableExternalLookup: Property<"enable-external-lookup", "Boolean">,
Global,
DefaultTrue,
-Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched.">;
+Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched. If all other methods fail, and
the DEBUGINFOD_URLS environment variable is specified, the Debuginfod protocol
is used to acquire symbols from a compatible Debuginfod service.">;
def EnableBackgroundLookup: Property<"enable-background-lookup", "Boolean">,
Global,
DefaultFalse,
diff --git a/lldb/source/Core/Debugger.cpp b/lldb/source/Core/Debugger.cpp
index 21f71e449ca5ed0..9a3e82f3e6a2adf 100644
--- a/lldb/source/Core/Debugger.cpp
+++ b/lldb/source/Core/Debugger.cpp
@@ -61,6 +61,8 @@
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/iterator.h"
+#include "llvm/Debuginfod/Debuginfod.h"
+#include "llvm/Debuginfod/HTTPClient.h"
#include "llvm/Support/DynamicLibrary.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Process.h"
@@ -594,6 +596,9 @@ lldb::DWIMPrintVerbosity Debugger::GetDWIMPrintVerbosity()
const {
void Debugger::Initialize(LoadPluginCallbackType load_plugin_callback) {
assert(g_debugger_list_ptr == nullptr &&
"Debugger::Initialize called more than once!");
+ // We might be using the Debuginfod service, s
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
@@ -0,0 +1,142 @@
+//===-- SymbolLocatorDebuginfod.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 "SymbolLocatorDebuginfod.h"
+
+#include "lldb/Core/PluginManager.h"
+#include "lldb/Utility/Args.h"
+
+#include "llvm/Debuginfod/Debuginfod.h"
+#include "llvm/Debuginfod/HTTPClient.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+LLDB_PLUGIN_DEFINE(SymbolLocatorDebuginfod)
+
+namespace {
+
+#define LLDB_PROPERTIES_symbollocatordebuginfod
+#include "SymbolLocatorDebuginfodProperties.inc"
+
+enum {
+#define LLDB_PROPERTIES_symbollocatordebuginfod
+#include "SymbolLocatorDebuginfodPropertiesEnum.inc"
+};
+
+class PluginProperties : public Properties {
+public:
+ static llvm::StringRef GetSettingName() {
+return SymbolLocatorDebuginfod::GetPluginNameStatic();
+ }
+
+ PluginProperties() {
+m_collection_sp =
std::make_shared(GetSettingName());
+m_collection_sp->Initialize(g_symbollocatordebuginfod_properties);
+
+// We need to read the default value first to read the environment
variable.
+llvm::SmallVector urls = llvm::getDefaultDebuginfodUrls();
+Args arg_urls{urls};
+m_collection_sp->SetPropertyAtIndexFromArgs(ePropertyServerURLs, arg_urls);
+
+m_collection_sp->SetValueChangedCallback(
+ePropertyServerURLs, [this] { ServerURLsChangedCallback(); });
+ }
+
+ Args GetDebugInfoDURLs() const {
+Args urls;
+m_collection_sp->GetPropertyAtIndexAsArgs(ePropertyServerURLs, urls);
+return urls;
+ }
+
+private:
+ void ServerURLsChangedCallback() {
+Args urls = GetDebugInfoDURLs();
+llvm::SmallVector dbginfod_urls;
+llvm::transform(urls, dbginfod_urls.end(),
+[](const auto &obj) { return obj.ref(); });
+llvm::setDefaultDebuginfodUrls(dbginfod_urls);
+ }
+};
+
+} // namespace
+
+static PluginProperties &GetGlobalPluginProperties() {
+ static PluginProperties g_settings;
+ return g_settings;
+}
+
+SymbolLocatorDebuginfod::SymbolLocatorDebuginfod() : SymbolLocator() {}
+
+void SymbolLocatorDebuginfod::Initialize() {
+ static llvm::once_flag g_once_flag;
+
+ llvm::call_once(g_once_flag, []() {
+PluginManager::RegisterPlugin(
+GetPluginNameStatic(), GetPluginDescriptionStatic(), CreateInstance,
+LocateExecutableObjectFile, LocateExecutableSymbolFile, nullptr,
+nullptr, SymbolLocatorDebuginfod::DebuggerInitialize);
+llvm::HTTPClient::initialize();
+ });
+}
+
+void SymbolLocatorDebuginfod::DebuggerInitialize(Debugger &debugger) {
+ if (!PluginManager::GetSettingForSymbolLocatorPlugin(
+ debugger, PluginProperties::GetSettingName())) {
+const bool is_global_setting = true;
+PluginManager::CreateSettingForSymbolLocatorPlugin(
+debugger, GetGlobalPluginProperties().GetValueProperties(),
+"Properties for the Debuginfod Symbol Locator plug-in.",
+is_global_setting);
+ }
+}
+
+void SymbolLocatorDebuginfod::Terminate() {
+ PluginManager::UnregisterPlugin(CreateInstance);
+ llvm::HTTPClient::cleanup();
+}
+
+llvm::StringRef SymbolLocatorDebuginfod::GetPluginDescriptionStatic() {
+ return "Debuginfod symbol locator.";
+}
+
+SymbolLocator *SymbolLocatorDebuginfod::CreateInstance() {
+ return new SymbolLocatorDebuginfod();
+}
+
+std::optional SymbolLocatorDebuginfod::LocateExecutableObjectFile(
+const ModuleSpec &module_spec) {
+ const UUID &module_uuid = module_spec.GetUUID();
+ if (module_uuid.IsValid() && llvm::canUseDebuginfod()) {
+llvm::object::BuildID build_id(module_uuid.GetBytes());
+llvm::Expected result =
+llvm::getCachedOrDownloadExecutable(build_id);
+if (result)
+ return FileSpec(*result);
+// An error here should be logged as a failure in the Debuginfod library,
+// so just consume it here
+consumeError(result.takeError());
+ }
+ return {};
+}
+
+std::optional SymbolLocatorDebuginfod::LocateExecutableSymbolFile(
+const ModuleSpec &module_spec, const FileSpecList &default_search_paths) {
+ const UUID &module_uuid = module_spec.GetUUID();
+ if (module_uuid.IsValid() && llvm::canUseDebuginfod()) {
+llvm::object::BuildID build_id(module_uuid.GetBytes());
+llvm::Expected result =
+llvm::getCachedOrDownloadDebuginfo(build_id);
+if (result)
+ return FileSpec(*result);
+// An error here should be logged as a failure in the Debuginfod library,
+// so just consume it here
+consumeError(result.takeError());
+ }
kevinfrei wrote:
I factored it into a helper to reduce copy-pasta
https://github.com/llvm/llvm-project/pull/70996
___
lldb-commits mailing list
lldb-commits@lists.llvm.o
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
@@ -46,6 +46,10 @@ bool canUseDebuginfod(); /// environment variable. SmallVector getDefaultDebuginfodUrls(); +/// Sets the list of debuginfod server URLs to query. This overrides the +/// environment variable DEBUGINFOD_URLS. +void setDefaultDebuginfodUrls(SmallVector URLs); kevinfrei wrote: Done. https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
@@ -62,15 +66,23 @@ bool canUseDebuginfod() {
}
SmallVector getDefaultDebuginfodUrls() {
- const char *DebuginfodUrlsEnv = std::getenv("DEBUGINFOD_URLS");
- if (DebuginfodUrlsEnv == nullptr)
-return SmallVector();
-
- SmallVector DebuginfodUrls;
- StringRef(DebuginfodUrlsEnv).split(DebuginfodUrls, " ");
+ if (!DebuginfodUrlsSet) {
+// Only read from the environment variable if the user hasn't already
+// set the value
+if (const char *DebuginfodUrlsEnv = std::getenv("DEBUGINFOD_URLS"))
+ StringRef(DebuginfodUrlsEnv).split(DebuginfodUrls, " ", -1, false);
+DebuginfodUrlsSet = true;
+ }
return DebuginfodUrls;
}
+// Set the default debuginfod URL list, override the environment variable
+void setDefaultDebuginfodUrls(SmallVector URLs) {
+ DebuginfodUrls.clear();
+ DebuginfodUrls.insert(DebuginfodUrls.begin(), URLs.begin(), URLs.end());
+ DebuginfodUrlsSet = true;
kevinfrei wrote:
So much cleaner!
https://github.com/llvm/llvm-project/pull/70996
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
@@ -62,15 +66,23 @@ bool canUseDebuginfod() {
}
SmallVector getDefaultDebuginfodUrls() {
- const char *DebuginfodUrlsEnv = std::getenv("DEBUGINFOD_URLS");
- if (DebuginfodUrlsEnv == nullptr)
-return SmallVector();
-
- SmallVector DebuginfodUrls;
- StringRef(DebuginfodUrlsEnv).split(DebuginfodUrls, " ");
+ if (!DebuginfodUrlsSet) {
+// Only read from the environment variable if the user hasn't already
+// set the value
+if (const char *DebuginfodUrlsEnv = std::getenv("DEBUGINFOD_URLS"))
+ StringRef(DebuginfodUrlsEnv).split(DebuginfodUrls, " ", -1, false);
+DebuginfodUrlsSet = true;
kevinfrei wrote:
Much better
https://github.com/llvm/llvm-project/pull/70996
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
@@ -47,6 +47,10 @@ namespace llvm {
using llvm::object::BuildIDRef;
+SmallVector DebuginfodUrls;
kevinfrei wrote:
Done
https://github.com/llvm/llvm-project/pull/70996
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
@@ -62,15 +66,23 @@ bool canUseDebuginfod() {
}
SmallVector getDefaultDebuginfodUrls() {
- const char *DebuginfodUrlsEnv = std::getenv("DEBUGINFOD_URLS");
- if (DebuginfodUrlsEnv == nullptr)
-return SmallVector();
-
- SmallVector DebuginfodUrls;
- StringRef(DebuginfodUrlsEnv).split(DebuginfodUrls, " ");
+ if (!DebuginfodUrlsSet) {
kevinfrei wrote:
std::optional is very nice
https://github.com/llvm/llvm-project/pull/70996
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/70996
>From b04c85dbed0b369e747aa2a3823789203156736b Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Wed, 18 Oct 2023 14:37:34 -0700
Subject: [PATCH 1/8] DEBUGINFOD based DWP acquisition for LLDB
Summary:
I've plumbed the LLVM DebugInfoD client into LLDB, and added automatic
downloading of
DWP files to the SymbolFileDWARF.cpp plugin. If you have `DEBUGINFOD_URLS` set
to a
space delimited set of web servers, LLDB will try to use them as a last resort
when
searching for DWP files. If you do *not* have that environment variable set,
nothing
should be changed. There's also a setting, per Greg Clayton's request, that will
override the env variable, or can be used instead of the env var. This setting
is the
reason for the additional API added to the llvm's Debuginfod library.
Test Plan:
Suggestions are welcome here. I should probably have some positive and negative
tests,
but I wanted to get the diff up for people who have a clue what they're doing
to rip it
to pieces before spending too much time validating my implementation.
---
lldb/include/lldb/Target/Target.h | 3 +++
lldb/source/Core/CoreProperties.td| 2 +-
lldb/source/Core/Debugger.cpp | 5
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 1 +
lldb/source/Symbol/CMakeLists.txt | 1 +
lldb/source/Target/Target.cpp | 19 +-
lldb/source/Target/TargetProperties.td| 4 +++
llvm/include/llvm/Debuginfod/Debuginfod.h | 4 +++
llvm/lib/Debuginfod/Debuginfod.cpp| 26 ++-
9 files changed, 57 insertions(+), 8 deletions(-)
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index c37682e2a03859f..7f10f0409fb1315 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;
+ Args GetDebugInfoDURLs() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
@@ -270,6 +272,7 @@ class TargetProperties : public Properties {
void DisableASLRValueChangedCallback();
void InheritTCCValueChangedCallback();
void DisableSTDIOValueChangedCallback();
+ void DebugInfoDURLsChangedCallback();
// Settings checker for target.jit-save-objects-dir:
void CheckJITObjectsDir();
diff --git a/lldb/source/Core/CoreProperties.td
b/lldb/source/Core/CoreProperties.td
index 92884258347e9be..865030b0133bbb2 100644
--- a/lldb/source/Core/CoreProperties.td
+++ b/lldb/source/Core/CoreProperties.td
@@ -4,7 +4,7 @@ let Definition = "modulelist" in {
def EnableExternalLookup: Property<"enable-external-lookup", "Boolean">,
Global,
DefaultTrue,
-Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched.">;
+Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched. If all other methods fail, and
the DEBUGINFOD_URLS environment variable is specified, the Debuginfod protocol
is used to acquire symbols from a compatible Debuginfod service.">;
def EnableBackgroundLookup: Property<"enable-background-lookup", "Boolean">,
Global,
DefaultFalse,
diff --git a/lldb/source/Core/Debugger.cpp b/lldb/source/Core/Debugger.cpp
index 21f71e449ca5ed0..9a3e82f3e6a2adf 100644
--- a/lldb/source/Core/Debugger.cpp
+++ b/lldb/source/Core/Debugger.cpp
@@ -61,6 +61,8 @@
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/iterator.h"
+#include "llvm/Debuginfod/Debuginfod.h"
+#include "llvm/Debuginfod/HTTPClient.h"
#include "llvm/Support/DynamicLibrary.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Process.h"
@@ -594,6 +596,9 @@ lldb::DWIMPrintVerbosity Debugger::GetDWIMPrintVerbosity()
const {
void Debugger::Initialize(LoadPluginCallbackType load_plugin_callback) {
assert(g_debugger_list_ptr == nullptr &&
"Debugger::Initialize called more than once!");
+ // We might be using the Debuginfod service, s
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/70996
>From b04c85dbed0b369e747aa2a3823789203156736b Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Wed, 18 Oct 2023 14:37:34 -0700
Subject: [PATCH 1/9] DEBUGINFOD based DWP acquisition for LLDB
Summary:
I've plumbed the LLVM DebugInfoD client into LLDB, and added automatic
downloading of
DWP files to the SymbolFileDWARF.cpp plugin. If you have `DEBUGINFOD_URLS` set
to a
space delimited set of web servers, LLDB will try to use them as a last resort
when
searching for DWP files. If you do *not* have that environment variable set,
nothing
should be changed. There's also a setting, per Greg Clayton's request, that will
override the env variable, or can be used instead of the env var. This setting
is the
reason for the additional API added to the llvm's Debuginfod library.
Test Plan:
Suggestions are welcome here. I should probably have some positive and negative
tests,
but I wanted to get the diff up for people who have a clue what they're doing
to rip it
to pieces before spending too much time validating my implementation.
---
lldb/include/lldb/Target/Target.h | 3 +++
lldb/source/Core/CoreProperties.td| 2 +-
lldb/source/Core/Debugger.cpp | 5
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 1 +
lldb/source/Symbol/CMakeLists.txt | 1 +
lldb/source/Target/Target.cpp | 19 +-
lldb/source/Target/TargetProperties.td| 4 +++
llvm/include/llvm/Debuginfod/Debuginfod.h | 4 +++
llvm/lib/Debuginfod/Debuginfod.cpp| 26 ++-
9 files changed, 57 insertions(+), 8 deletions(-)
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index c37682e2a03859f..7f10f0409fb1315 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;
+ Args GetDebugInfoDURLs() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
@@ -270,6 +272,7 @@ class TargetProperties : public Properties {
void DisableASLRValueChangedCallback();
void InheritTCCValueChangedCallback();
void DisableSTDIOValueChangedCallback();
+ void DebugInfoDURLsChangedCallback();
// Settings checker for target.jit-save-objects-dir:
void CheckJITObjectsDir();
diff --git a/lldb/source/Core/CoreProperties.td
b/lldb/source/Core/CoreProperties.td
index 92884258347e9be..865030b0133bbb2 100644
--- a/lldb/source/Core/CoreProperties.td
+++ b/lldb/source/Core/CoreProperties.td
@@ -4,7 +4,7 @@ let Definition = "modulelist" in {
def EnableExternalLookup: Property<"enable-external-lookup", "Boolean">,
Global,
DefaultTrue,
-Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched.">;
+Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched. If all other methods fail, and
the DEBUGINFOD_URLS environment variable is specified, the Debuginfod protocol
is used to acquire symbols from a compatible Debuginfod service.">;
def EnableBackgroundLookup: Property<"enable-background-lookup", "Boolean">,
Global,
DefaultFalse,
diff --git a/lldb/source/Core/Debugger.cpp b/lldb/source/Core/Debugger.cpp
index 21f71e449ca5ed0..9a3e82f3e6a2adf 100644
--- a/lldb/source/Core/Debugger.cpp
+++ b/lldb/source/Core/Debugger.cpp
@@ -61,6 +61,8 @@
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/iterator.h"
+#include "llvm/Debuginfod/Debuginfod.h"
+#include "llvm/Debuginfod/HTTPClient.h"
#include "llvm/Support/DynamicLibrary.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Process.h"
@@ -594,6 +596,9 @@ lldb::DWIMPrintVerbosity Debugger::GetDWIMPrintVerbosity()
const {
void Debugger::Initialize(LoadPluginCallbackType load_plugin_callback) {
assert(g_debugger_list_ptr == nullptr &&
"Debugger::Initialize called more than once!");
+ // We might be using the Debuginfod service, s
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/70996
>From fc24f4c0ba2328009796a09ae2c2c925c773af44 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Wed, 18 Oct 2023 14:37:34 -0700
Subject: [PATCH 1/9] DEBUGINFOD based DWP acquisition for LLDB
Summary:
I've plumbed the LLVM DebugInfoD client into LLDB, and added automatic
downloading of
DWP files to the SymbolFileDWARF.cpp plugin. If you have `DEBUGINFOD_URLS` set
to a
space delimited set of web servers, LLDB will try to use them as a last resort
when
searching for DWP files. If you do *not* have that environment variable set,
nothing
should be changed. There's also a setting, per Greg Clayton's request, that will
override the env variable, or can be used instead of the env var. This setting
is the
reason for the additional API added to the llvm's Debuginfod library.
Test Plan:
Suggestions are welcome here. I should probably have some positive and negative
tests,
but I wanted to get the diff up for people who have a clue what they're doing
to rip it
to pieces before spending too much time validating my implementation.
---
lldb/include/lldb/Target/Target.h | 3 +++
lldb/source/Core/CoreProperties.td| 2 +-
lldb/source/Core/Debugger.cpp | 5
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 1 +
lldb/source/Symbol/CMakeLists.txt | 1 +
lldb/source/Target/Target.cpp | 19 +-
lldb/source/Target/TargetProperties.td| 4 +++
llvm/include/llvm/Debuginfod/Debuginfod.h | 4 +++
llvm/lib/Debuginfod/Debuginfod.cpp| 26 ++-
9 files changed, 57 insertions(+), 8 deletions(-)
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index c37682e2a03859f..7f10f0409fb1315 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;
+ Args GetDebugInfoDURLs() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
@@ -270,6 +272,7 @@ class TargetProperties : public Properties {
void DisableASLRValueChangedCallback();
void InheritTCCValueChangedCallback();
void DisableSTDIOValueChangedCallback();
+ void DebugInfoDURLsChangedCallback();
// Settings checker for target.jit-save-objects-dir:
void CheckJITObjectsDir();
diff --git a/lldb/source/Core/CoreProperties.td
b/lldb/source/Core/CoreProperties.td
index 92884258347e9be..865030b0133bbb2 100644
--- a/lldb/source/Core/CoreProperties.td
+++ b/lldb/source/Core/CoreProperties.td
@@ -4,7 +4,7 @@ let Definition = "modulelist" in {
def EnableExternalLookup: Property<"enable-external-lookup", "Boolean">,
Global,
DefaultTrue,
-Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched.">;
+Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched. If all other methods fail, and
the DEBUGINFOD_URLS environment variable is specified, the Debuginfod protocol
is used to acquire symbols from a compatible Debuginfod service.">;
def EnableBackgroundLookup: Property<"enable-background-lookup", "Boolean">,
Global,
DefaultFalse,
diff --git a/lldb/source/Core/Debugger.cpp b/lldb/source/Core/Debugger.cpp
index 21f71e449ca5ed0..9a3e82f3e6a2adf 100644
--- a/lldb/source/Core/Debugger.cpp
+++ b/lldb/source/Core/Debugger.cpp
@@ -61,6 +61,8 @@
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/iterator.h"
+#include "llvm/Debuginfod/Debuginfod.h"
+#include "llvm/Debuginfod/HTTPClient.h"
#include "llvm/Support/DynamicLibrary.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Process.h"
@@ -594,6 +596,9 @@ lldb::DWIMPrintVerbosity Debugger::GetDWIMPrintVerbosity()
const {
void Debugger::Initialize(LoadPluginCallbackType load_plugin_callback) {
assert(g_debugger_list_ptr == nullptr &&
"Debugger::Initialize called more than once!");
+ // We might be using the Debuginfod service, s
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/70996
>From 190fb1529bfe0ecdb5e5a30d335ccdca45463c32 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Wed, 18 Oct 2023 14:37:34 -0700
Subject: [PATCH 01/10] DEBUGINFOD based DWP acquisition for LLDB
Summary:
I've plumbed the LLVM DebugInfoD client into LLDB, and added automatic
downloading of
DWP files to the SymbolFileDWARF.cpp plugin. If you have `DEBUGINFOD_URLS` set
to a
space delimited set of web servers, LLDB will try to use them as a last resort
when
searching for DWP files. If you do *not* have that environment variable set,
nothing
should be changed. There's also a setting, per Greg Clayton's request, that will
override the env variable, or can be used instead of the env var. This setting
is the
reason for the additional API added to the llvm's Debuginfod library.
Test Plan:
Suggestions are welcome here. I should probably have some positive and negative
tests,
but I wanted to get the diff up for people who have a clue what they're doing
to rip it
to pieces before spending too much time validating my implementation.
---
lldb/include/lldb/Target/Target.h | 3 +++
lldb/source/Core/CoreProperties.td| 2 +-
lldb/source/Core/Debugger.cpp | 5
.../SymbolFile/DWARF/SymbolFileDWARF.cpp | 1 +
lldb/source/Symbol/CMakeLists.txt | 1 +
lldb/source/Target/Target.cpp | 19 +-
lldb/source/Target/TargetProperties.td| 4 +++
llvm/include/llvm/Debuginfod/Debuginfod.h | 4 +++
llvm/lib/Debuginfod/Debuginfod.cpp| 26 ++-
9 files changed, 57 insertions(+), 8 deletions(-)
diff --git a/lldb/include/lldb/Target/Target.h
b/lldb/include/lldb/Target/Target.h
index c37682e2a0385..7f10f0409fb13 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;
+ Args GetDebugInfoDURLs() const;
+
private:
// Callbacks for m_launch_info.
void Arg0ValueChangedCallback();
@@ -270,6 +272,7 @@ class TargetProperties : public Properties {
void DisableASLRValueChangedCallback();
void InheritTCCValueChangedCallback();
void DisableSTDIOValueChangedCallback();
+ void DebugInfoDURLsChangedCallback();
// Settings checker for target.jit-save-objects-dir:
void CheckJITObjectsDir();
diff --git a/lldb/source/Core/CoreProperties.td
b/lldb/source/Core/CoreProperties.td
index 92884258347e9..865030b0133bb 100644
--- a/lldb/source/Core/CoreProperties.td
+++ b/lldb/source/Core/CoreProperties.td
@@ -4,7 +4,7 @@ let Definition = "modulelist" in {
def EnableExternalLookup: Property<"enable-external-lookup", "Boolean">,
Global,
DefaultTrue,
-Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched.">;
+Desc<"Control the use of external tools and repositories to locate symbol
files. Directories listed in target.debug-file-search-paths and directory of
the executable are always checked first for separate debug info files. Then
depending on this setting: On macOS, Spotlight would be also used to locate a
matching .dSYM bundle based on the UUID of the executable. On NetBSD, directory
/usr/libdata/debug would be also searched. On platforms other than NetBSD
directory /usr/lib/debug would be also searched. If all other methods fail, and
the DEBUGINFOD_URLS environment variable is specified, the Debuginfod protocol
is used to acquire symbols from a compatible Debuginfod service.">;
def EnableBackgroundLookup: Property<"enable-background-lookup", "Boolean">,
Global,
DefaultFalse,
diff --git a/lldb/source/Core/Debugger.cpp b/lldb/source/Core/Debugger.cpp
index 21f71e449ca5e..9a3e82f3e6a2a 100644
--- a/lldb/source/Core/Debugger.cpp
+++ b/lldb/source/Core/Debugger.cpp
@@ -61,6 +61,8 @@
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/iterator.h"
+#include "llvm/Debuginfod/Debuginfod.h"
+#include "llvm/Debuginfod/HTTPClient.h"
#include "llvm/Support/DynamicLibrary.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Process.h"
@@ -594,6 +596,9 @@ lldb::DWIMPrintVerbosity Debugger::GetDWIMPrintVerbosity()
const {
void Debugger::Initialize(LoadPluginCallbackType load_plugin_callback) {
assert(g_debugger_list_ptr == nullptr &&
"Debugger::Initialize called more than once!");
+ // We might be using the Debuginfod service, so we have
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: > We need to respect the setting `symbols.enable-external-lookup` and not do > anything with debuginfod if this is set to false. > > You can get this value by calling: > > ``` > ModuleList::GetGlobalModuleListProperties().GetEnableExternalLookup() > ``` > > Otherwise I fear that the linux buildbots will start failing if we get extra > system library symbols if debuginfod is enabled via the environment variable. Fixed that up, rebased and push -f 'ed https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: > This breaks targeting macOS versions older than 10.12 for unnecessary reasons: > > ``` > ../../llvm/lib/Debuginfod/Debuginfod.cpp:54:6: error: 'shared_mutex' is > unavailable: introduced in macOS 10.12 > std::shared_mutex UrlsMutex; > ^ > ``` > > You should be able to use llvm::sys::RWMutex so please switch to that. See > https://reviews.llvm.org/D138423 for an example. N! I'll get this fixed up ASAP. Should I open an issue, or just put up a new PR? (Thanks for the example!) https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: > This change broke building against LLVM dylib: > > ``` > /usr/lib/gcc/x86_64-pc-linux-gnu/13/../../../../x86_64-pc-linux-gnu/bin/ld: > cannot find -lLLVMDebuginfod: No such file or directory > collect2: error: ld returned 1 exit status > samu: subcommand failed > ``` > > Full build log: > [dev-util:lldb-18.0.0_pre20231206:20231206-163149.log](https://github.com/llvm/llvm-project/files/13588891/dev-util.lldb-18.0.0_pre20231206.20231206-163149.log) Sounds like I get to dig in deeper to LLVM's build, now. Hooray 🤣 https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: > This change broke building against LLVM dylib: > > ``` > /usr/lib/gcc/x86_64-pc-linux-gnu/13/../../../../x86_64-pc-linux-gnu/bin/ld: > cannot find -lLLVMDebuginfod: No such file or directory > collect2: error: ld returned 1 exit status > samu: subcommand failed > ``` > > Full build log: > [dev-util:lldb-18.0.0_pre20231206:20231206-163149.log](https://github.com/llvm/llvm-project/files/13588891/dev-util.lldb-18.0.0_pre20231206.20231206-163149.log) @mgorny Can you please point me at where to find all the configuration for those builds? I'm stuck on a weirdo CentOS build (or MacOS) and can't manage to get my CMake configuration stuff to let me build a dylib/so. I'll get on this first thing tomorrow. https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: > > This change broke building against LLVM dylib: > > ``` > > /usr/lib/gcc/x86_64-pc-linux-gnu/13/../../../../x86_64-pc-linux-gnu/bin/ld: > > cannot find -lLLVMDebuginfod: No such file or directory > > collect2: error: ld returned 1 exit status > > samu: subcommand failed > > ``` > > > > > > > > > > > > > > > > > > > > > > > > Full build log: > > [dev-util:lldb-18.0.0_pre20231206:20231206-163149.log](https://github.com/llvm/llvm-project/files/13588891/dev-util.lldb-18.0.0_pre20231206.20231206-163149.log) > > @mgorny Can you please point me at where to find all the configuration for > those builds? I'm stuck on a weirdo CentOS build (or MacOS) and can't manage > to get my CMake configuration stuff to let me build a dylib/so. I'll get on > this first thing tomorrow. > This change broke building against LLVM dylib: > > ``` > /usr/lib/gcc/x86_64-pc-linux-gnu/13/../../../../x86_64-pc-linux-gnu/bin/ld: > cannot find -lLLVMDebuginfod: No such file or directory > collect2: error: ld returned 1 exit status > samu: subcommand failed > ``` > > Full build log: > [dev-util:lldb-18.0.0_pre20231206:20231206-163149.log](https://github.com/llvm/llvm-project/files/13588891/dev-util.lldb-18.0.0_pre20231206.20231206-163149.log) @mgorny I've burned 3 hours this morning trying to get the visible configuration flags from that log file to function _before_ my patch on (my weirdo CentOS 9) Linux box, or cause problems _after_ my patch on macOS, and failed on both fronts. I'm stuck until you can get me the contents of `gentoo_common_config.cmake` (or tell me where to find it...) https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: > Sounds like some configs don't create `LLVMDebuginfod` target possibly > because it isn't enabled. Is there a cmake option to disable `LLVMDebuginfod` > or some magic in the Cmake config file for this target? Nothing that I could find. The library is written such that if the configuration doesn't include either libcurl or httplib then all the functions just fail to do anything, so from what I can tell there'd never be a reason to disable it. https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: Is there some documentation that both me & google are missing somewhere? I'm trying to read between the lines of the scenario that's broken, with nothing but a log (that refers to a bunch of pretty important configuration files from somewhere I can't find) with a repo configuration that, again, I can't figure out. You say that we can't "build against LLMV as a dylib". Is this trying to build LLDB stand-alone, without a full LLVM repo around it? I'm also assuming you're actually talking about an SO and not a dylib, since the build is on Linux, not macOS. I really want to help, but I'm recently new to this space, so I'm missing a bunch of tribal knowledge. Alternatively, if you have a repro, and can fix it by adding LLVMDebuginfod to libLLVM, that seems reasonable to me, given the design of LLVMDebuginfod (small, falls back to no functionality without libcurl, etc...) I've burned all day today trying to get a build & repro going, and have failed (my build is quite complicated). I'll try this on a personal machine tomorrow which should be simpler (though a couple orders of magnitude slower) than my work behemoth... https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: I've spent 5 or 6 more hours fighting this and I'm fully stuck on "No Repro". I have a libLLVM-git18.so built, independent of building LLDB. Then I configured LLDB to build against that as a standalone entity and validated that it worked (all on my personal machine, which is running Kali Linux under WSL2: basically just Ubuntu). Below, you'll find shell scripts that I've been using to configure the build (both for LLVM and then for the standalone LLDB build). Please tell me what I need to change in these scripts to repro the problem. ``` #!/bin/sh # Configure LLVM for subsequent use by an LLDB build cmake -G Ninja \ -B rel-llvm \ -S llvm/llvm \ -DCMAKE_BUILD_TYPE=Release \ -DLLVM_ENABLE_PROJECTS=clang \ -DCMAKE_INSTALL_PREFIX=/usr \ -DLLVM_APPEND_VC_REV=OFF \ -DCMAKE_INSTALL_PREFIX=/usr/lib/llvm/18 \ -DBUILD_SHARED_LIBS=OFF \ -DLLVM_BUILD_LLVM_DYLIB=ON \ -DLLVM_LINK_LLVM_DYLIB=ON \ -DLLVM_TARGETS_TO_BUILD="AArch64;X86" \ -DLLVM_INCLUDE_BENCHMARKS=OFF \ -DLLVM_ENABLE_ZSTD=yes \ -DLLVM_ENABLE_CURL=yes \ -DLLVM_HOST_TRIPLE=i686-pc-linux-gnu \ -DLLVM_DISTRIBUTION_COMPONENTS="LLVM;LTO;Remarks;llvm-config;cmake-exports;llvm-headers;LLVMDemangle;LLVMSupport;LLVMTableGen;llvm-tblgen;dsymutil;llc;lli;llvm-addr2line;llvm-ar;llvm-as;llvm-bcanalyzer;llvm-bitcode-strip;llvm-c-test;llvm-cat;llvm-cfi-verify;llvm-config;llvm-cov;llvm-cvtres;llvm-cxxdump;llvm-cxxfilt;llvm-cxxmap;llvm-debuginfo-analyzer;llvm-debuginfod-find;llvm-diff;llvm-dis;llvm-dlltool;llvm-dwarfdump;llvm-dwarfutil;llvm-dwp;llvm-exegesis;llvm-extract;llvm-gsymutil;llvm-ifs;llvm-install-name-tool;llvm-jitlink;llvm-lib;llvm-libtool-darwin;llvm-link;llvm-lipo;llvm-lto;llvm-lto2;llvm-mc;llvm-mca;llvm-ml;llvm-modextract;llvm-mt;llvm-nm;llvm-objcopy;llvm-objdump;llvm-opt-report;llvm-otool;llvm-pdbutil;llvm-profdata;llvm-profgen;llvm-ranlib;llvm-rc;llvm-readelf;llvm-readobj;llvm-readtapi;llvm-reduce;llvm-remarkutil;llvm-rtdyld;llvm-sim;llvm-size;llvm-split;llvm-stress;llvm-strings;llvm-strip;llvm-symbolizer;llvm-tli-checker;llvm-undname;llvm-windres;llvm-xray;opt;sancov;sanstats;verify-uselistorder;opt-viewer;llvm-debuginfod" ``` and this one for the lldb build: ``` #!/bin/sh cmake -G Ninja \ -B rel-lldb \ -S llvm/lldb \ -DCMAKE_BUILD_TYPE=Release \ -DLLVM_TARGETS_TO_BUILD="AArch64;X86" \ -DCMAKE_INSTALL_PREFIX=/usr \ -DLLVM_ENABLE_LIBEDIT=yes \ -DLLVM_ENABLE_TERMINFO=yes \ -DLLVM_ENABLE_LIBXML2=yes \ -DLLVM_ENABLE_EH=ON \ -DLLVM_ENABLE_RTTI=ON \ -DLLVM_ENABLE_CURL=yes \ -DLLVM_HOST_TRIPLE=i686-pc-linux-gnu \ -DPython3_EXECUTABLE=/usr/bin/python3 \ -DLLVM_DIR=/home/freik/src/rel-llvm/lib/cmake/llvm ``` https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: Ran the first script, sudo nina install-distribution, then ran the second script. I don't see the build depending on what's been installed, as it looks like setting LLVM_DIR makes the .so dependency on the binary sitting in the rel-llvm/lib directory instead. [build.log.gz](https://github.com/llvm/llvm-project/files/13653581/build.log.gz) https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Fix a crash from character type confusion interaction with libedit (PR #75388)
https://github.com/kevinfrei created
https://github.com/llvm/llvm-project/pull/75388
If you type `settings show ` LLDB might crash, depending on the version of
libedit you're compiled with, and whether you're compiled with
`-DLLDB_EDITLINE_USE_WCHAR=0` (and depending on how the optimizer lays out the
stack...)
The issue has to do with trying to figure out whether the libedit `getchar`
callback is supposed to read a wide or 8 bit character. In order to maintain
backward compatibility, there's really no 'clean' way to do it. We just have to
make sure that we're invoking el_[w]getc with a buffer that is as wide as the
getchar callback (registered by the `SetGetCharacterFunction` function further
down in `Editline.cpp`.
So, it's 'fixed' with a comment, and a wider version of the 'reply' variable.
>From 16a58d40ad0ff4a7f0fd73c3033cb758aed25d5d Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Wed, 13 Dec 2023 13:36:20 -0800
Subject: [PATCH] Fix a crash from character type confusion interaction with
libedit
---
lldb/include/lldb/Host/Editline.h| 2 ++
lldb/source/Host/common/Editline.cpp | 10 --
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/lldb/include/lldb/Host/Editline.h
b/lldb/include/lldb/Host/Editline.h
index c598244150788d..9049b106f02a34 100644
--- a/lldb/include/lldb/Host/Editline.h
+++ b/lldb/include/lldb/Host/Editline.h
@@ -75,6 +75,8 @@ using EditLineCharType = char;
// to wchar_t. It is not possible to detect differentiate between the two
// versions exactly, but this is a pretty good approximation and allows us to
// build against almost any editline version out there.
+// It does, however, require extra care when invoking el_getc, as the type
+// of the input is a single char buffer, but the callback will write a wchar_t.
#if LLDB_EDITLINE_USE_WCHAR || defined(EL_CLIENTDATA) || LLDB_HAVE_EL_RFUNC_T
using EditLineGetCharType = wchar_t;
#else
diff --git a/lldb/source/Host/common/Editline.cpp
b/lldb/source/Host/common/Editline.cpp
index 82e17ec753ab23..ce707e530d008b 100644
--- a/lldb/source/Host/common/Editline.cpp
+++ b/lldb/source/Host/common/Editline.cpp
@@ -978,8 +978,14 @@ void Editline::DisplayCompletions(
break;
fprintf(editline.m_output_file, "More (Y/n/a): ");
-char reply = 'n';
-int got_char = el_getc(editline.m_editline, &reply);
+// The type for the output and the type for the parameter are different,
+// to allow interoperability with older versions of libedit. The container
+// for the reply must be as wide as what our implementation is using,
+// but libedit may use a narrower type depending on the build
+// configuration.
+EditLineGetCharType reply = L'n';
+int got_char = el_wgetc(editline.m_editline,
+reinterpret_cast(&reply));
// Check for a ^C or other interruption.
if (editline.m_editor_status == EditorStatus::Interrupted) {
editline.m_editor_status = EditorStatus::Editing;
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Fix a crash from character type confusion interaction with libedit (PR #75388)
@@ -978,8 +978,14 @@ void Editline::DisplayCompletions( break; fprintf(editline.m_output_file, "More (Y/n/a): "); -char reply = 'n'; -int got_char = el_getc(editline.m_editline, &reply); +// The type for the output and the type for the parameter are different, +// to allow interoperability with older versions of libedit. The container +// for the reply must be as wide as what our implementation is using, +// but libedit may use a narrower type depending on the build +// configuration. +EditLineGetCharType reply = L'n'; +int got_char = el_wgetc(editline.m_editline, kevinfrei wrote: el_wgetc is #defined to el_getc if there's no wchar_t support (it's not at all consistent in this file, but I wanted to targeted fix for the crash). The core problem is actually with the #if code below the comment in the header: ``` #if LLDB_EDITLINE_USE_WCHAR || defined(EL_CLIENTDATA) || LLDB_HAVE_EL_RFUNC_T using EditLineGetCharType = wchar_t; #else using EditLineGetCharType = char; #endif ``` We set EditLineGetCharType to wchar_t, *even if you are building with `LLDB_EDITLINE_USE_WCHAR=0`* with newer versions of libedit (the ones that define EL_CLIENTDATA) so the function that reads a character (called from within libedit) is always reading a wchar_t *and writing it to the buffer that's backed by `&reply`*. https://github.com/llvm/llvm-project/pull/75388 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Fix a crash from character type confusion interaction with libedit (PR #75388)
@@ -978,8 +978,14 @@ void Editline::DisplayCompletions( break; fprintf(editline.m_output_file, "More (Y/n/a): "); -char reply = 'n'; -int got_char = el_getc(editline.m_editline, &reply); +// The type for the output and the type for the parameter are different, +// to allow interoperability with older versions of libedit. The container +// for the reply must be as wide as what our implementation is using, +// but libedit may use a narrower type depending on the build +// configuration. +EditLineGetCharType reply = L'n'; +int got_char = el_wgetc(editline.m_editline, kevinfrei wrote: Just realized I didn't point you at the code to answer your simple question: At the top of EditLine.cpp, all the el_w* functions are #defined to el_* functions if you're build without LLDB_EDITLINE_USE_WCHAR https://github.com/llvm/llvm-project/pull/75388 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: I'll try to get this together tonight or tomorrow morning. https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [llvm] [lldb] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: doBuild.sh: ``` #!/bin/sh ./llvm-config.sh cd rel-llvm ninja -v sudo ninja -v install-distribution cd .. ./lldb-config.sh cd rel-lldb ninja -v ``` llvm-config.sh: ``` #!/bin/sh cmake -G Ninja \ -B rel-llvm \ -S llvm/llvm \ -DCMAKE_BUILD_TYPE=Release \ -DLLVM_ENABLE_PROJECTS=clang \ -DCMAKE_INSTALL_PREFIX=/usr \ -DLLVM_APPEND_VC_REV=OFF \ -DCMAKE_INSTALL_PREFIX=/usr/lib/llvm/18 \ -DBUILD_SHARED_LIBS=OFF \ -DLLVM_BUILD_LLVM_DYLIB=ON \ -DLLVM_LINK_LLVM_DYLIB=ON \ -DLLVM_TARGETS_TO_BUILD="AArch64;X86" \ -DLLVM_INCLUDE_BENCHMARKS=OFF \ -DLLVM_ENABLE_ZSTD=yes \ -DLLVM_ENABLE_CURL=yes \ -DLLVM_HOST_TRIPLE=i686-pc-linux-gnu \ -DLLVM_DISTRIBUTION_COMPONENTS="LLVM;LTO;Remarks;llvm-config;cmake-exports;llvm-headers;LLVMDemangle;LLVMSupport;LLVMTableGen;llvm-tblgen;dsymutil;llc;lli;llvm-addr2line;llvm-ar;llvm-as;llvm-bcanalyzer;llvm-bitcode-strip;llvm-c-test;llvm-cat;llvm-cfi-verify;llvm-config;llvm-cov;llvm-cvtres;llvm-cxxdump;llvm-cxxfilt;llvm-cxxmap;llvm-debuginfo-analyzer;llvm-debuginfod-find;llvm-diff;llvm-dis;llvm-dlltool;llvm-dwarfdump;llvm-dwarfutil;llvm-dwp;llvm-exegesis;llvm-extract;llvm-gsymutil;llvm-ifs;llvm-install-name-tool;llvm-jitlink;llvm-lib;llvm-libtool-darwin;llvm-link;llvm-lipo;llvm-lto;llvm-lto2;llvm-mc;llvm-mca;llvm-ml;llvm-modextract;llvm-mt;llvm-nm;llvm-objcopy;llvm-objdump;llvm-opt-report;llvm-otool;llvm-pdbutil;llvm-profdata;llvm-profgen;llvm-ranlib;llvm-rc;llvm-readelf;llvm-readobj;llvm-readtapi;llvm-reduce;llvm-remarkutil;llvm-rtdyld;llvm-sim;llvm-size;llvm-split;llvm-stress;llvm-strings;llvm-strip;llvm-symbolizer;llvm-tli-checker;llvm-undname;llvm-windres;llvm-xray;opt;sancov;sanstats;verify-uselistorder;opt-viewer;llvm-debuginfod" ``` lldb-config.sh: ``` #!/bin/sh cmake -G Ninja \ -B rel-lldb \ -S llvm/lldb \ -DCMAKE_BUILD_TYPE=Release \ -DLLVM_TARGETS_TO_BUILD="AArch64;X86" \ -DCMAKE_INSTALL_PREFIX=/usr \ -DLLVM_ENABLE_LIBEDIT=yes \ -DLLVM_ENABLE_TERMINFO=yes \ -DLLVM_ENABLE_LIBXML2=yes \ -DLLVM_ENABLE_EH=ON \ -DLLVM_ENABLE_RTTI=ON \ -DLLVM_ENABLE_CURL=yes \ -DLLVM_HOST_TRIPLE=i686-pc-linux-gnu \ -DPython3_EXECUTABLE=/usr/bin/python3 \ -DLLVM_DIR=/home/freik/src/rel-llvm/lib/cmake/llvm ``` And, here's the output of that whole mess: [hopefully-enough.log.gz](https://github.com/llvm/llvm-project/files/13681262/hopefully-enough.log.gz) https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] DEBUGINFOD based DWP acquisition for LLDB (PR #70996)
kevinfrei wrote: Sorry: nice long holiday break. So calling that library a "left over from a previous build" confuses me. What's the point of the `-DLLVM_DIR=/home/freik/src/rel-llvm/lib/cmake/llvm` argument to the second cmake config if not to instruct the build where to find stuff from the full LLVM build? If I delete that library, I can, indeed, reproduce the issue [now working on it] but that also leads me to believe that my repro still isn't actually doing what your configuration is. Because of that, I have no confidence that what I do to fix the issue is going to *actually* fix it 😞 https://github.com/llvm/llvm-project/pull/70996 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei created https://github.com/llvm/llvm-project/pull/85693 Finally getting back to Debuginfod tests, I've migrated my tests from shell to API (at @JDevlieghere's suggestion) and addressed a couple issues that came about during testing. >From 9713607cd4839ad355c7fd2e786ae7eb5a96f637 Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/2] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLI
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei updated https://github.com/llvm/llvm-project/pull/85693 >From 9713607cd4839ad355c7fd2e786ae7eb5a96f637 Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/3] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o $(DYLIB_DWP_FILE) $(DYLIB_DWOS) +endif endif #
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei updated https://github.com/llvm/llvm-project/pull/85693 >From 9713607cd4839ad355c7fd2e786ae7eb5a96f637 Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/3] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o $(DYLIB_DWP_FILE) $(DYLIB_DWOS) +endif endif #
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei ready_for_review https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei converted_to_draft https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei updated https://github.com/llvm/llvm-project/pull/85693 >From 9713607cd4839ad355c7fd2e786ae7eb5a96f637 Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/4] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o $(DYLIB_DWP_FILE) $(DYLIB_DWOS) +endif endif #
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei updated https://github.com/llvm/llvm-project/pull/85693 >From 9713607cd4839ad355c7fd2e786ae7eb5a96f637 Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/4] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o $(DYLIB_DWP_FILE) $(DYLIB_DWOS) +endif endif #
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei ready_for_review https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [llvm] LLDB Debuginfod usage tests (with fixes) (PR #79181)
https://github.com/kevinfrei closed https://github.com/llvm/llvm-project/pull/79181 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
@@ -4377,26 +4377,40 @@ const std::shared_ptr
&SymbolFileDWARF::GetDwpSymbolFile() {
FileSpecList search_paths = Target::GetDefaultDebugFileSearchPaths();
ModuleSpec module_spec;
module_spec.GetFileSpec() = m_objfile_sp->GetFileSpec();
+FileSpec dwp_filespec;
for (const auto &symfile : symfiles.files()) {
module_spec.GetSymbolFileSpec() =
FileSpec(symfile.GetPath() + ".dwp", symfile.GetPathStyle());
LLDB_LOG(log, "Searching for DWP using: \"{0}\"",
module_spec.GetSymbolFileSpec());
- FileSpec dwp_filespec =
+ dwp_filespec =
PluginManager::LocateExecutableSymbolFile(module_spec, search_paths);
if (FileSystem::Instance().Exists(dwp_filespec)) {
-LLDB_LOG(log, "Found DWP file: \"{0}\"", dwp_filespec);
-DataBufferSP dwp_file_data_sp;
-lldb::offset_t dwp_file_data_offset = 0;
-ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
-GetObjectFile()->GetModule(), &dwp_filespec, 0,
-FileSystem::Instance().GetByteSize(dwp_filespec), dwp_file_data_sp,
-dwp_file_data_offset);
-if (dwp_obj_file) {
- m_dwp_symfile = std::make_shared(
- *this, dwp_obj_file, DIERef::k_file_index_mask);
- break;
-}
+break;
+ }
+}
+if (!FileSystem::Instance().Exists(dwp_filespec)) {
+ LLDB_LOG(log, "No DWP file found locally");
+ // Fill in the UUID for the module we're trying to match for, so we can
+ // find the correct DWP file, as the Debuginfod plugin uses *only* this
+ // data to correctly match the DWP file with the binary.
+ module_spec.GetUUID() = m_objfile_sp->GetUUID();
+ dwp_filespec =
+ PluginManager::LocateExecutableSymbolFile(module_spec, search_paths);
+ // Set it back so it's not outliving the m_objfile_sp shared pointer.
+ module_spec.GetUUID() = {};
kevinfrei wrote:
Actually, that line of code is left-over from a previous iteration (before I
moved it to after the original loop). @clayborg had removed the assignment for
other reasons, but it needs to be there for the Debuginfod plugin. It's not
used again, and doesn't hurt anything (the original value is actually {},
because that module_spec is a file-loc only skeleton). I've removed the line.
https://github.com/llvm/llvm-project/pull/85693
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei updated https://github.com/llvm/llvm-project/pull/85693 >From 9713607cd4839ad355c7fd2e786ae7eb5a96f637 Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/5] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o $(DYLIB_DWP_FILE) $(DYLIB_DWOS) +endif endif #
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
kevinfrei wrote: @JDevlieghere ping :) https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei updated https://github.com/llvm/llvm-project/pull/85693 >From fc0eda99c7cceeaefd33477c9b08e7221a5a6e2a Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/5] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o $(DYLIB_DWP_FILE) $(DYLIB_DWOS) +endif endif #
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
@@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" kevinfrei wrote: Updated. It also turns out that MAKE_DWP implies MAKE_DWO, so I removed it from the test Makefile while I was at it. https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
https://github.com/kevinfrei updated https://github.com/llvm/llvm-project/pull/85693 >From fc0eda99c7cceeaefd33477c9b08e7221a5a6e2a Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 15 Mar 2024 08:54:04 -0700 Subject: [PATCH 1/6] Tests (w/fixes) for initial DebugInfoD LLDB integration Summary: While adding tests for DebugInfoD integration, there were a couple issues that came up: DWP from /debuginfo for a stripped binary needs to return symbols from /executable Test Plan: Added some API tests Reviewers: gclayton, hyubo Subscribers: Tasks: T179375937 Tags: debuginfod Differential Revision: https://phabricator.intern.facebook.com/D54953389 --- .../Python/lldbsuite/test/make/Makefile.rules | 33 +++- .../SymbolFile/DWARF/SymbolFileDWARF.cpp | 40 +++-- .../Plugins/SymbolLocator/CMakeLists.txt | 7 +- .../SymbolVendor/ELF/SymbolVendorELF.cpp | 32 +++- lldb/test/API/debuginfod/Normal/Makefile | 25 +++ .../API/debuginfod/Normal/TestDebuginfod.py | 159 + lldb/test/API/debuginfod/Normal/main.c| 7 + lldb/test/API/debuginfod/SplitDWARF/Makefile | 29 +++ .../SplitDWARF/TestDebuginfodDWP.py | 167 ++ lldb/test/API/debuginfod/SplitDWARF/main.c| 7 + 10 files changed, 489 insertions(+), 17 deletions(-) create mode 100644 lldb/test/API/debuginfod/Normal/Makefile create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py create mode 100644 lldb/test/API/debuginfod/Normal/main.c create mode 100644 lldb/test/API/debuginfod/SplitDWARF/Makefile create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py create mode 100644 lldb/test/API/debuginfod/SplitDWARF/main.c diff --git a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules index bfd249ccd43f2e..b33c087357a79b 100644 --- a/lldb/packages/Python/lldbsuite/test/make/Makefile.rules +++ b/lldb/packages/Python/lldbsuite/test/make/Makefile.rules @@ -51,7 +51,7 @@ LLDB_BASE_DIR := $(THIS_FILE_DIR)/../../../../../ # # GNUWin32 uname gives "windows32" or "server version windows32" while # some versions of MSYS uname return "MSYS_NT*", but most environments -# standardize on "Windows_NT", so we'll make it consistent here. +# standardize on "Windows_NT", so we'll make it consistent here. # When running tests from Visual Studio, the environment variable isn't # inherited all the way down to the process spawned for make. #-- @@ -210,6 +210,12 @@ else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" DSYM = $(EXE).debug endif + + ifeq "$(MERGE_DWOS)" "YES" + MAKE_DWO := YES + DWP_NAME = $(EXE).dwp + DYLIB_DWP_NAME = $(DYLIB_NAME).dwp + endif endif LIMIT_DEBUG_INFO_FLAGS = @@ -357,6 +363,7 @@ ifneq "$(OS)" "Darwin" OBJCOPY ?= $(call replace_cc_with,objcopy) ARCHIVER ?= $(call replace_cc_with,ar) + DWP ?= $(call replace_cc_with,dwp) override AR = $(ARCHIVER) endif @@ -527,6 +534,10 @@ ifneq "$(CXX)" "" endif endif +ifeq "$(GEN_GNU_BUILD_ID)" "YES" + LDFLAGS += -Wl,--build-id +endif + #-- # DYLIB_ONLY variable can be used to skip the building of a.out. # See the sections below regarding dSYM file as well as the building of @@ -565,11 +576,25 @@ else endif else ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" +ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(EXE)" "$(EXE).full" +endif $(OBJCOPY) --only-keep-debug "$(EXE)" "$(DSYM)" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DSYM)" "$(EXE)" "$(EXE)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o "$(DWP_NAME)" $(DWOS) +endif endif + +#-- +# Support emitting the context of the GNU build-id into a file +# This needs to be used in conjunction with GEN_GNU_BUILD_ID := YES +#-- +$(EXE).uuid : $(EXE) + $(OBJCOPY) --dump-section=.note.gnu.build-id=$@ $< + #-- # Make the dylib #-- @@ -610,9 +635,15 @@ endif else $(LD) $(DYLIB_OBJECTS) $(LDFLAGS) -shared -o "$(DYLIB_FILENAME)" ifeq "$(SPLIT_DEBUG_SYMBOLS)" "YES" + ifeq "$(SAVE_FULL_DEBUG_BINARY)" "YES" + cp "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).full" + endif $(OBJCOPY) --only-keep-debug "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME).debug" $(OBJCOPY) --strip-debug --add-gnu-debuglink="$(DYLIB_FILENAME).debug" "$(DYLIB_FILENAME)" "$(DYLIB_FILENAME)" endif +ifeq "$(MERGE_DWOS)" "YES" + $(DWP) -o $(DYLIB_DWP_FILE) $(DYLIB_DWOS) +endif endif #
[Lldb-commits] [lldb] Missed a null-ptr check in previous PR for Debuginfod testing (PR #86292)
https://github.com/kevinfrei created https://github.com/llvm/llvm-project/pull/86292 @GeorgeHuyubo noticed an unchecked shared pointer result in https://github.com/llvm/llvm-project/pull/85693/. This is the fix for that issue. >From 6bc82c0820963050752e6faabef7ec10e8c81f2f Mon Sep 17 00:00:00 2001 From: Kevin Frei Date: Fri, 22 Mar 2024 08:04:37 -0700 Subject: [PATCH] Missed a null-ptr check in previous PR --- lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp index 91b8b4a979e0c7..a9956aa9075fe2 100644 --- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp +++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp @@ -53,7 +53,7 @@ static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp, ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin( module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec), dwp_file_data_sp, dwp_file_data_offset); - if (!ObjectFileELF::classof(dwp_obj_file.get())) + if (!dwp_obj_file || !ObjectFileELF::classof(dwp_obj_file.get())) return false; // The presence of a debug_cu_index section is the key identifying feature of // a DWP file. Make sure we don't fill in the section list on dwp_obj_file ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
@@ -44,6 +44,25 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ if (!ObjectFileELF::classof(dwp_obj_file.get()))
kevinfrei wrote:
Fixed in PR https://github.com/llvm/llvm-project/pull/86292 Good catch,
@GeorgeHuyubo !
https://github.com/llvm/llvm-project/pull/85693
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Missed a null-ptr check in previous PR for Debuginfod testing (PR #86292)
https://github.com/kevinfrei ready_for_review https://github.com/llvm/llvm-project/pull/86292 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD tests + fixing issues exposed by tests (PR #85693)
kevinfrei wrote: > New tests apparently are failing on Arm/AArch64 Linux. The failure indicates that the failing tests' build don't include a UUID, which *should* be generated by the changes to Makefile.rules. Any idea how I ought to proceed with reproducing a build on a system that I have no access to? I could probably dust off an RPi 4 at home tonight, if that would be a reasonable repro platform. Alternatively, I can make the test resolve (but not succeed...) by checking to make sure that the UUID is actually found. https://github.com/llvm/llvm-project/pull/85693 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD issues, take 2 (PR #86812)
https://github.com/kevinfrei created
https://github.com/llvm/llvm-project/pull/86812
> The previous diff (and it's subsequent fix) were reverted as the tests didn't
> work properly on the AArch64 & ARM LLDB buildbots. I made a couple more minor
> changes to tests (from @clayborg's feedback) and disabled them for non
> Linux-x86(_64) builds, as I don't have the ability do anything about an ARM64
> Linux failure. If I had to guess, I'd say the toolchain on the buildbots
> isn't respecting the `-Wl,--build-id` flag. Maybe, one day, when I have a
> Linux AArch64 system I'll dig in to it.
Finally getting back to DebugInfoD tests:
I've migrated the tests in my https://github.com/llvm/llvm-project/pull/79181
from shell to API (at @JDevlieghere's suggestion) and addressed a couple issues
that came about during testing.
The tests first test the "normal" situation (no DebugInfoD involvement, just
normal debug files sitting around), then the "no debug info" situation (to make
sure the test is seeing failure properly), then it tests to validate that when
DebugInfoD returns the symbols, things work properly. This is duplicated for
DWP/split-dwarf scenarios.
>From ff90236364eed7fe71df2411ba694ea1603a9897 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 1/4] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a8..a881218a56cef3 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 00..a662fb9fc6e686
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+
[Lldb-commits] [lldb] DebugInfoD issues, take 2 (PR #86812)
https://github.com/kevinfrei edited https://github.com/llvm/llvm-project/pull/86812 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD issues, take 2 (PR #86812)
https://github.com/kevinfrei updated
https://github.com/llvm/llvm-project/pull/86812
>From 5e3a35bb69b0bd6c3950deaba35a78c085bc5728 Mon Sep 17 00:00:00 2001
From: Kevin Frei
Date: Mon, 25 Mar 2024 08:23:47 -0700
Subject: [PATCH 1/4] Trying to deal with Linux AArch64 test failures :/
---
.../SymbolVendor/ELF/SymbolVendorELF.cpp | 18 ++
.../API/debuginfod/Normal/TestDebuginfod.py | 187 +
.../SplitDWARF/TestDebuginfodDWP.py | 196 ++
3 files changed, 401 insertions(+)
create mode 100644 lldb/test/API/debuginfod/Normal/TestDebuginfod.py
create mode 100644 lldb/test/API/debuginfod/SplitDWARF/TestDebuginfodDWP.py
diff --git a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
index b5fe35d71032a8..a881218a56cef3 100644
--- a/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
+++ b/lldb/source/Plugins/SymbolVendor/ELF/SymbolVendorELF.cpp
@@ -44,6 +44,24 @@ llvm::StringRef
SymbolVendorELF::GetPluginDescriptionStatic() {
"executables.";
}
+// If this is needed elsewhere, it can be exported/moved.
+static bool IsDwpSymbolFile(const lldb::ModuleSP &module_sp,
+const FileSpec &file_spec) {
+ DataBufferSP dwp_file_data_sp;
+ lldb::offset_t dwp_file_data_offset = 0;
+ // Try to create an ObjectFile from the file_spec.
+ ObjectFileSP dwp_obj_file = ObjectFile::FindPlugin(
+ module_sp, &file_spec, 0, FileSystem::Instance().GetByteSize(file_spec),
+ dwp_file_data_sp, dwp_file_data_offset);
+ // The presence of a debug_cu_index section is the key identifying feature of
+ // a DWP file. Make sure we don't fill in the section list on dwp_obj_file
+ // (by calling GetSectionList(false)) as this is invoked before we may have
+ // all the symbol files collected and available.
+ return dwp_obj_file && ObjectFileELF::classof(dwp_obj_file.get()) &&
+ dwp_obj_file->GetSectionList(false)->FindSectionByType(
+ eSectionTypeDWARFDebugCuIndex, false);
+}
+
// CreateInstance
//
// Platforms can register a callback to use when creating symbol vendors to
diff --git a/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
new file mode 100644
index 00..a662fb9fc6e686
--- /dev/null
+++ b/lldb/test/API/debuginfod/Normal/TestDebuginfod.py
@@ -0,0 +1,187 @@
+import os
+import shutil
+import tempfile
+import struct
+
+import lldb
+from lldbsuite.test.decorators import *
+import lldbsuite.test.lldbutil as lldbutil
+from lldbsuite.test.lldbtest import *
+
+
+def getUUID(aoutuuid):
+"""
+Pull the 20 byte UUID out of the .note.gnu.build-id section that was dumped
+to a file already, as part of the build.
+"""
+with open(aoutuuid, "rb") as f:
+data = f.read(36)
+if len(data) != 36:
+return None
+header = struct.unpack_from("<4I", data)
+if len(header) != 4:
+return None
+# 4 element 'prefix', 20 bytes of uuid, 3 byte long string: 'GNU':
+if header[0] != 4 or header[1] != 20 or header[2] != 3 or header[3] !=
0x554E47:
+return None
+return data[16:].hex()
+
+
+"""
+Test support for the DebugInfoD network symbol acquisition protocol.
+This one is for simple / no split-dwarf scenarios.
+
+For no-split-dwarf scenarios, there are 2 variations:
+1 - A stripped binary with it's corresponding unstripped binary:
+2 - A stripped binary with a corresponding --only-keep-debug symbols file
+"""
+
+
+@skipUnlessPlatform(["linux", "freebsd"])
+class DebugInfodTests(TestBase):
+# No need to try every flavor of debug inf.
+NO_DEBUG_INFO_TESTCASE = True
+
+def test_normal_no_symbols(self):
+"""
+Validate behavior with no symbols or symbol locator.
+('baseline negative' behavior)
+"""
+test_root = self.config_test(["a.out"])
+self.try_breakpoint(False)
+
+def test_normal_default(self):
+"""
+Validate behavior with symbols, but no symbol locator.
+('baseline positive' behavior)
+"""
+test_root = self.config_test(["a.out", "a.out.debug"])
+self.try_breakpoint(True)
+
+def test_debuginfod_symbols(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'debuginfo' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_executable(self):
+"""
+Test behavior with the full binary available from Debuginfod as
+'executable' from the plug-in.
+"""
+test_root = self.config_test(["a.out"], None, "a.out.full")
+self.try_breakpoint(True)
+
+def test_debuginfod_okd_symbols(self):
+"""
+Test behavior with the 'only-keep-debug' symbols available from
Debuginfod.
+"""
+
[Lldb-commits] [lldb] DebugInfoD issues, take 2 (PR #86812)
https://github.com/kevinfrei ready_for_review https://github.com/llvm/llvm-project/pull/86812 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] DebugInfoD issues, take 2 (PR #86812)
kevinfrei wrote: > I have such a system and have dug into the DebugInfoD "space" recently and > would be more than willing (eager, even) to help! If you are on the Discord > server, we can message and find a way to collaborate (I am hawkinsw). Just joined. I can sort of use Discord (I mentor high school robotics students, and definitely can't keep up with them). Name is 'frei4all'. I'm in the Seattle area, so 8-5, PDT. If you're free Thursday, I'm happy to see if you can help me repro (and address) the issue... https://github.com/llvm/llvm-project/pull/86812 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
