[Lldb-commits] [lldb] SBDebugger: define callback_token_t for win32 (PR #92870)
https://github.com/antmox created https://github.com/llvm/llvm-project/pull/92870 fix lldb-aarch64-windows bot failure introduced by #89868 https://lab.llvm.org/buildbot/#/builders/219/builds/11583 >From 83ac7ff11a694795dccf3fa09c82ddab5d0096f3 Mon Sep 17 00:00:00 2001 From: Antoine Moynault Date: Tue, 21 May 2024 07:21:10 + Subject: [PATCH] SBDebugger: define callback_token_t for win32 fix lldb-aarch64-windows bot failure introduced by #89868 https://lab.llvm.org/buildbot/#/builders/219/builds/11583 --- lldb/include/lldb/lldb-types.h | 1 + 1 file changed, 1 insertion(+) diff --git a/lldb/include/lldb/lldb-types.h b/lldb/include/lldb/lldb-types.h index 8e717c62d3259..6b9289f0a0743 100644 --- a/lldb/include/lldb/lldb-types.h +++ b/lldb/include/lldb/lldb-types.h @@ -47,6 +47,7 @@ typedef void *thread_arg_t; // Host thread argument type typedef unsigned thread_result_t; // Host thread result type typedef thread_result_t (*thread_func_t)(void *); // Host thread function type typedef void *pipe_t; // Host pipe type is HANDLE +typedef int callback_token_t; #else ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: define callback_token_t for win32 (PR #92870)
llvmbot wrote: @llvm/pr-subscribers-lldb Author: antoine moynault (antmox) Changes fix lldb-aarch64-windows bot failure introduced by #89868 https://lab.llvm.org/buildbot/#/builders/219/builds/11583 --- Full diff: https://github.com/llvm/llvm-project/pull/92870.diff 1 Files Affected: - (modified) lldb/include/lldb/lldb-types.h (+1) ``diff diff --git a/lldb/include/lldb/lldb-types.h b/lldb/include/lldb/lldb-types.h index 8e717c62d3259..6b9289f0a0743 100644 --- a/lldb/include/lldb/lldb-types.h +++ b/lldb/include/lldb/lldb-types.h @@ -47,6 +47,7 @@ typedef void *thread_arg_t; // Host thread argument type typedef unsigned thread_result_t; // Host thread result type typedef thread_result_t (*thread_func_t)(void *); // Host thread function type typedef void *pipe_t; // Host pipe type is HANDLE +typedef int callback_token_t; #else `` https://github.com/llvm/llvm-project/pull/92870 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: Add new APIs `AddDestroyCallback` and `RemoveDestroyCallback` (PR #89868)
antmox wrote: @royitaqi @clayborg I proposed a fix here : https://github.com/llvm/llvm-project/pull/92870 https://github.com/llvm/llvm-project/pull/89868 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: define callback_token_t for win32 (PR #92870)
mstorsjo wrote: +1, this fixes my build errors for Windows targets as well: https://github.com/mstorsjo/llvm-mingw/actions/runs/9167001362/job/25206642195 https://github.com/llvm/llvm-project/pull/92870 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] af8f155 - [lldb-dap] Don't send expanded descriptions for "hover" expressions (#92726)
Author: Pavel Labath
Date: 2024-05-21T10:37:41+02:00
New Revision: af8f1554b8e7844304ce61b49b06e5b7946e9c47
URL:
https://github.com/llvm/llvm-project/commit/af8f1554b8e7844304ce61b49b06e5b7946e9c47
DIFF:
https://github.com/llvm/llvm-project/commit/af8f1554b8e7844304ce61b49b06e5b7946e9c47.diff
LOG: [lldb-dap] Don't send expanded descriptions for "hover" expressions
(#92726)
VSCode will automatically ask for the children (in structured form) so
there's no point in sending the textual representation. This can make
displaying hover popups for complex variables with complicated data
formatters much faster. See discussion on #77026 for context.
Added:
Modified:
lldb/test/API/tools/lldb-dap/evaluate/TestDAP_evaluate.py
lldb/test/API/tools/lldb-dap/variables/TestDAP_variables.py
lldb/tools/lldb-dap/JSONUtils.cpp
Removed:
diff --git a/lldb/test/API/tools/lldb-dap/evaluate/TestDAP_evaluate.py
b/lldb/test/API/tools/lldb-dap/evaluate/TestDAP_evaluate.py
index 57cabf5b7f411..68c57ad775544 100644
--- a/lldb/test/API/tools/lldb-dap/evaluate/TestDAP_evaluate.py
+++ b/lldb/test/API/tools/lldb-dap/evaluate/TestDAP_evaluate.py
@@ -27,7 +27,7 @@ def assertEvaluateFailure(self, expression):
)
def isResultExpandedDescription(self):
-return self.context == "repl" or self.context == "hover"
+return self.context == "repl"
def isExpressionParsedExpected(self):
return self.context != "hover"
diff --git a/lldb/test/API/tools/lldb-dap/variables/TestDAP_variables.py
b/lldb/test/API/tools/lldb-dap/variables/TestDAP_variables.py
index 07ab6d5a63eb6..57c17e5ea9d3d 100644
--- a/lldb/test/API/tools/lldb-dap/variables/TestDAP_variables.py
+++ b/lldb/test/API/tools/lldb-dap/variables/TestDAP_variables.py
@@ -502,29 +502,12 @@ def do_test_scopes_and_evaluate_expansion(self,
enableAutoVariableSummaries: boo
},
"hover": {
"equals": {"type": "PointType"},
-"equals": {
-"result": """(PointType) pt = {
- x = 11
- y = 22
- buffer = {
-[0] = 0
-[1] = 1
-[2] = 2
-[3] = 3
-[4] = 4
-[5] = 5
-[6] = 6
-[7] = 7
-[8] = 8
-[9] = 9
-[10] = 10
-[11] = 11
-[12] = 12
-[13] = 13
-[14] = 14
-[15] = 15
- }
-}"""
+"startswith": {
+"result": (
+"{x:11, y:22, buffer:{...}}"
+if enableAutoVariableSummaries
+else "PointType @ 0x"
+)
},
"missing": ["indexedVariables"],
"hasVariablesReference": True,
diff --git a/lldb/tools/lldb-dap/JSONUtils.cpp
b/lldb/tools/lldb-dap/JSONUtils.cpp
index bec277332bcf0..069877dbab339 100644
--- a/lldb/tools/lldb-dap/JSONUtils.cpp
+++ b/lldb/tools/lldb-dap/JSONUtils.cpp
@@ -1065,9 +1065,9 @@ llvm::json::Object
VariableDescription::GetVariableExtensionsJSON() {
}
std::string VariableDescription::GetResult(llvm::StringRef context) {
- // In repl and hover context, the results can be displayed as multiple lines
- // so more detailed descriptions can be returned.
- if (context != "repl" && context != "hover")
+ // In repl context, the results can be displayed as multiple lines so more
+ // detailed descriptions can be returned.
+ if (context != "repl")
return display_value;
if (!v.IsValid())
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb-dap] Don't send expanded descriptions for "hover" expressions (PR #92726)
https://github.com/labath closed https://github.com/llvm/llvm-project/pull/92726 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (PR #92503)
labath wrote: > Yes, originally lldb's unwinder was recursive for any register propagation > and it was easy to hit the problem of lldb blowing out its own stack on a > recursive inferior that had crashed. I changed most of the propagation to a > loop to solve this (years and years ago) but it looks like we still have a > case where it is recursing. > > We still need to skip the test case on macOS until I can come up with some > idea to get proper unwind instruction for sigtramp on arm64. Most of the CI > bots are x86_64 so it may pass on them, but that's The Past and I would > prefer to skip this on Darwin until I can figure something out, I wrote a > little TODO on myself in rdar://128031075 Sounds good (I meant to xfail this on darwin, but it looks like I forgot). https://github.com/llvm/llvm-project/pull/92503 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (PR #92503)
https://github.com/labath updated
https://github.com/llvm/llvm-project/pull/92503
>From 47858edddfbccf989213d61f3761fcdb04814839 Mon Sep 17 00:00:00 2001
From: Pavel Labath
Date: Fri, 17 May 2024 06:34:08 +
Subject: [PATCH 1/2] Reapply "[lldb/aarch64] Fix unwinding when signal
interrupts a leaf function (#91321)"
This reapplies fd1bd53ba5a06f344698a55578f6a5d79c457e30, which was
reverted due to a test failure on aarch64/windows. The failure was
caused by a combination of several factors:
- clang targeting aarch64-windows (unlike msvc, and unlike clang
targeting other aarch64 platforms) defaults to -fomit-frame-pointers
- lldb's code for looking up register values for `` unwind rules
is recursive
- the test binary creates a very long chain of fp-less function frames
(it manages to fit about 22k frames before it blows its stack)
Together, these things have caused lldb to recreate the same deep
recursion when unwinding through this, and blow its own stack as well.
Since lldb frames are larger, about 4k frames like this was sufficient
to trigger the stack overflow.
This version of the patch works around this problem by increasing the
frame size of the test binary, thereby causing it to blow its stack
sooner. This doesn't fix the issue -- the same problem can occur with a
real binary -- but it's not very likely, as it requires an infinite
recursion in a simple (so it doesn't use the frame pointer) function
with a very small frame (so you can fit a lot of them on the stack).
A more principled fix would be to make lldb's lookup code non-recursive,
but I believe that's out of scope for this patch.
The original patch description follows:
A leaf function may not store the link register to stack, but we it can
still end up being a non-zero frame if it gets interrupted by a signal.
Currently, we were unable to unwind past this function because we could
not read the link register value.
To make this work, this patch:
- changes the function-entry unwind plan to include the `fp|lr = `
rules. This in turn necessitated an adjustment in the generic
instruction emulation logic to ensure that `lr=[sp-X]` can override the
`` rule.
- allows the `` rule for pc and lr in all
`m_all_registers_available` frames (and not just frame zero).
The test verifies that we can unwind in a situation like this, and that
the backtrace matches the one we computed before getting a signal.
---
.../ARM64/EmulateInstructionARM64.cpp | 2 ++
.../UnwindAssemblyInstEmulation.cpp | 4 +--
lldb/source/Target/RegisterContextUnwind.cpp | 6 ++---
.../API/functionalities/bt-interrupt/main.c | 1 +
.../Inputs/signal-in-leaf-function-aarch64.c | 15 +++
.../signal-in-leaf-function-aarch64.test | 27 +++
.../ARM64/TestArm64InstEmulation.cpp | 24 ++---
7 files changed, 69 insertions(+), 10 deletions(-)
create mode 100644
lldb/test/Shell/Unwind/Inputs/signal-in-leaf-function-aarch64.c
create mode 100644 lldb/test/Shell/Unwind/signal-in-leaf-function-aarch64.test
diff --git a/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
b/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
index 6ca4fb052457e..62ecac3e0831d 100644
--- a/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
+++ b/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
@@ -444,6 +444,8 @@ bool EmulateInstructionARM64::CreateFunctionEntryUnwind(
// Our previous Call Frame Address is the stack pointer
row->GetCFAValue().SetIsRegisterPlusOffset(gpr_sp_arm64, 0);
+ row->SetRegisterLocationToSame(gpr_lr_arm64, /*must_replace=*/false);
+ row->SetRegisterLocationToSame(gpr_fp_arm64, /*must_replace=*/false);
unwind_plan.AppendRow(row);
unwind_plan.SetSourceName("EmulateInstructionARM64");
diff --git
a/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
b/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
index c4a171ec7d01b..49edd40544e32 100644
---
a/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
+++
b/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
@@ -424,8 +424,6 @@ size_t UnwindAssemblyInstEmulation::WriteMemory(
log->PutString(strm.GetString());
}
- const bool cant_replace = false;
-
switch (context.type) {
default:
case EmulateInstruction::eContextInvalid:
@@ -467,7 +465,7 @@ size_t UnwindAssemblyInstEmulation::WriteMemory(
m_pushed_regs[reg_num] = addr;
const int32_t offset = addr - m_initial_sp;
m_curr_row->SetRegisterLocationToAtCFAPlusOffset(reg_num, offset,
- cant_replace);
+ /*can_replace=*/true);
m_curr_row_modified = true;
}
}
diff --git a/lldb/source/Target/RegisterContextUnwind.cpp
b/lldb/source/Target/Reg
[Lldb-commits] [lldb] bbd54e0 - Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (#92503)
Author: Pavel Labath
Date: 2024-05-21T10:56:26+02:00
New Revision: bbd54e08b08f5ccd38c4665178e65c58f7b14459
URL:
https://github.com/llvm/llvm-project/commit/bbd54e08b08f5ccd38c4665178e65c58f7b14459
DIFF:
https://github.com/llvm/llvm-project/commit/bbd54e08b08f5ccd38c4665178e65c58f7b14459.diff
LOG: Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f…
(#92503)
…unction (#91321)"
This reapplies fd1bd53ba5a06f344698a55578f6a5d79c457e30, which was
reverted due to a test failure on aarch64/windows. The failure was
caused by a combination of several factors:
- clang targeting aarch64-windows (unlike msvc, and unlike clang
targeting other aarch64 platforms) defaults to -fomit-frame-pointers
- lldb's code for looking up register values for `` unwind rules
is recursive
- the test binary creates a very long chain of fp-less function frames
(it manages to fit about 22k frames before it blows its stack)
Together, these things have caused lldb to recreate the same deep
recursion when unwinding through this, and blow its own stack as well.
Since lldb frames are larger, about 4k frames like this was sufficient
to trigger the stack overflow.
This version of the patch works around this problem by increasing the
frame size of the test binary, thereby causing it to blow its stack
sooner. This doesn't fix the issue -- the same problem can occur with a
real binary -- but it's not very likely, as it requires an infinite
recursion in a simple (so it doesn't use the frame pointer) function
with a very small frame (so you can fit a lot of them on the stack).
A more principled fix would be to make lldb's lookup code non-recursive,
but I believe that's out of scope for this patch.
The original patch description follows:
A leaf function may not store the link register to stack, but we it can
still end up being a non-zero frame if it gets interrupted by a signal.
Currently, we were unable to unwind past this function because we could
not read the link register value.
To make this work, this patch:
- changes the function-entry unwind plan to include the `fp|lr = `
rules. This in turn necessitated an adjustment in the generic
instruction emulation logic to ensure that `lr=[sp-X]` can override the
`` rule.
- allows the `` rule for pc and lr in all
`m_all_registers_available` frames (and not just frame zero).
The test verifies that we can unwind in a situation like this, and that
the backtrace matches the one we computed before getting a signal.
Added:
lldb/test/Shell/Unwind/Inputs/signal-in-leaf-function-aarch64.c
lldb/test/Shell/Unwind/signal-in-leaf-function-aarch64.test
Modified:
lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
lldb/source/Target/RegisterContextUnwind.cpp
lldb/test/API/functionalities/bt-interrupt/main.c
lldb/unittests/UnwindAssembly/ARM64/TestArm64InstEmulation.cpp
Removed:
diff --git a/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
b/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
index 6ca4fb052457e..62ecac3e0831d 100644
--- a/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
+++ b/lldb/source/Plugins/Instruction/ARM64/EmulateInstructionARM64.cpp
@@ -444,6 +444,8 @@ bool EmulateInstructionARM64::CreateFunctionEntryUnwind(
// Our previous Call Frame Address is the stack pointer
row->GetCFAValue().SetIsRegisterPlusOffset(gpr_sp_arm64, 0);
+ row->SetRegisterLocationToSame(gpr_lr_arm64, /*must_replace=*/false);
+ row->SetRegisterLocationToSame(gpr_fp_arm64, /*must_replace=*/false);
unwind_plan.AppendRow(row);
unwind_plan.SetSourceName("EmulateInstructionARM64");
diff --git
a/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
b/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
index c4a171ec7d01b..49edd40544e32 100644
---
a/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
+++
b/lldb/source/Plugins/UnwindAssembly/InstEmulation/UnwindAssemblyInstEmulation.cpp
@@ -424,8 +424,6 @@ size_t UnwindAssemblyInstEmulation::WriteMemory(
log->PutString(strm.GetString());
}
- const bool cant_replace = false;
-
switch (context.type) {
default:
case EmulateInstruction::eContextInvalid:
@@ -467,7 +465,7 @@ size_t UnwindAssemblyInstEmulation::WriteMemory(
m_pushed_regs[reg_num] = addr;
const int32_t offset = addr - m_initial_sp;
m_curr_row->SetRegisterLocationToAtCFAPlusOffset(reg_num, offset,
- cant_replace);
+ /*can_replace=*/true);
m_curr_row_modified = true;
}
}
diff --git a/lldb/source/Target/RegisterContextUnwind.cpp
b/lldb/source/Tar
[Lldb-commits] [lldb] Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (PR #92503)
https://github.com/labath closed https://github.com/llvm/llvm-project/pull/92503 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] a098bea - [lldb] Move callback_token_t out of #ifdef windows
Author: Pavel Labath Date: 2024-05-21T09:37:38Z New Revision: a098beaff7670e5986b790b6ba93999fa21c6b48 URL: https://github.com/llvm/llvm-project/commit/a098beaff7670e5986b790b6ba93999fa21c6b48 DIFF: https://github.com/llvm/llvm-project/commit/a098beaff7670e5986b790b6ba93999fa21c6b48.diff LOG: [lldb] Move callback_token_t out of #ifdef windows Also make it a int32_t to match the surrounding code. Added: Modified: lldb/include/lldb/lldb-types.h Removed: diff --git a/lldb/include/lldb/lldb-types.h b/lldb/include/lldb/lldb-types.h index 8e717c62d3259..d88b8232ee6b4 100644 --- a/lldb/include/lldb/lldb-types.h +++ b/lldb/include/lldb/lldb-types.h @@ -62,7 +62,6 @@ typedef void *thread_arg_t; // Host thread argument type typedef void *thread_result_t; // Host thread result type typedef void *(*thread_func_t)(void *); // Host thread function type typedef int pipe_t; // Host pipe type -typedef int callback_token_t; #endif // _WIN32 @@ -79,6 +78,7 @@ typedef bool (*ExpressionCancelCallback)(ExpressionEvaluationPhase phase, typedef void *ScriptObjectPtr; typedef uint64_t addr_t; +typedef int32_t callback_token_t; typedef uint64_t user_id_t; typedef uint64_t pid_t; typedef uint64_t tid_t; ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: define callback_token_t for win32 (PR #92870)
labath wrote: Moving the typedef out of platform-specific code would be a better, as this is not in any way platform-specific. I just pushed something along those lines. https://github.com/llvm/llvm-project/pull/92870 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: define callback_token_t for win32 (PR #92870)
https://github.com/antmox closed https://github.com/llvm/llvm-project/pull/92870 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: define callback_token_t for win32 (PR #92870)
antmox wrote: fixed elsewhere https://github.com/llvm/llvm-project/pull/92870 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Improve performance of .debug_names lookups when DW_IDX_parent attributes are used (PR #91808)
https://github.com/labath edited https://github.com/llvm/llvm-project/pull/91808 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Improve performance of .debug_names lookups when DW_IDX_parent attributes are used (PR #91808)
https://github.com/labath commented: I think #77696 is justification enough to add this check, but I don't think we should be implicating DW_IDX_parent_entries until we know how those came about. Both me and David have tried to produce entries like you describe and failed (in fact I was not able to produce a DW_IDX_parent_entries referring to any kind of class type whatsoever (only to namespace entries)). Even when this check goes in, I think it would be useful to get to the bottom of the DW_IDX_parent_entries issue, if nothing else, then because it may affect the decision of when/if to remove this workaround. https://github.com/llvm/llvm-project/pull/91808 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Improve performance of .debug_names lookups when DW_IDX_parent attributes are used (PR #91808)
@@ -83,6 +83,10 @@ bool DebugNamesDWARFIndex::ProcessEntry( DWARFDIE die = dwarf.GetDIE(*ref); if (!die) return true; + // Watch out for forward declarations that appear in the .debug_names tables + // only due to being there for a DW_IDX_parent. labath wrote: ```suggestion // Clang erroneously emits index entries for declaration DIEs in case when the // definition is in a type unit (llvm.org/pr77696). Weed those out. ``` https://github.com/llvm/llvm-project/pull/91808 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Improve performance of .debug_names lookups when DW_IDX_parent attributes are used (PR #91808)
labath wrote: > What is "nameless index entry"? I don't quite understand from the spec. ``` It is possible that an indexed debugging information entry has a parent that is not indexed (for example, if its parent does not have a name attribute). In such a case, a parent attribute may point to a nameless index entry (that is, one that cannot be reached from any entry in the name table), or it may point to the nearest ancestor that does have an index entry. ``` My interpretation of this paragraph is that we're allowed to insert extra "floating" entries into the debug_names table. Such entries must not be reachable from the hash or name tables, but they can be referred to via DW_IDX_parent_entry attributes of other (reachable) entries. Using Greg's example from https://github.com/llvm/llvm-project/pull/91808#issue-2290404282, I believe we could add a debug_names entry for `std::ios_base` (and refer to it from the entry for `std::ios_base::seekdir`), as long as the ios_base entry was not present in any of the name tables. https://github.com/llvm/llvm-project/pull/91808 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb/dwarf] Fix DW_IDX_parent processing for split dwarf (PR #92745)
https://github.com/labath updated
https://github.com/llvm/llvm-project/pull/92745
>From ae72d530c723a8d0fb2eac38224e8cf74cc43e70 Mon Sep 17 00:00:00 2001
From: Pavel Labath
Date: Mon, 20 May 2024 12:30:46 +
Subject: [PATCH 1/2] [lldb/dwarf] Fix DW_IDX_parent processing for split dwarf
DWARFDebugInfo only knows how to resolve references in its own file, but
in split dwarf, the index entries will refer to DIEs in the separate
(DWO) file. To resolve the DIERef correctly we'd either need to go
through the SymbolFileDWARF to get the full logic for resolving a
DIERef, or use the fact that ToDIERef already looks up the correct unit
while computing its result.
This patch does the latter.
This bug manifested itself in not being able to find type definitions
for types in namespaces, so I've modified one of our type resolving test
cases to run with debug_names, and added a namespaced class into it (it
originally contained only a top-level class).
---
.../SymbolFile/DWARF/DWARFDebugInfo.cpp | 6 ---
.../Plugins/SymbolFile/DWARF/DWARFDebugInfo.h | 5 ---
.../SymbolFile/DWARF/DebugNamesDWARFIndex.cpp | 28 +++--
.../SymbolFile/DWARF/DebugNamesDWARFIndex.h | 2 +-
.../API/lang/cpp/limit-debug-info/Makefile| 2 -
.../TestWithLimitDebugInfo.py | 40 +++
.../API/lang/cpp/limit-debug-info/base.cpp| 9 ++---
.../test/API/lang/cpp/limit-debug-info/base.h | 22 +++---
.../API/lang/cpp/limit-debug-info/derived.cpp | 11 +++--
.../API/lang/cpp/limit-debug-info/derived.h | 37 -
.../API/lang/cpp/limit-debug-info/main.cpp| 6 +--
.../SymbolFile/DWARF/DWARFDIETest.cpp | 24 +--
12 files changed, 106 insertions(+), 86 deletions(-)
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
index 44febcfac3b00..d28da728728e5 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
@@ -259,9 +259,3 @@ DWARFDebugInfo::GetDIE(const DIERef &die_ref) {
return cu->GetNonSkeletonUnit().GetDIE(die_ref.die_offset());
return DWARFDIE(); // Not found
}
-
-llvm::StringRef DWARFDebugInfo::PeekDIEName(const DIERef &die_ref) {
- if (DWARFUnit *cu = GetUnit(die_ref))
-return cu->GetNonSkeletonUnit().PeekDIEName(die_ref.die_offset());
- return llvm::StringRef();
-}
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
index c1f0cb0203fb7..456ebd908ccb2 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
@@ -44,11 +44,6 @@ class DWARFDebugInfo {
bool ContainsTypeUnits();
DWARFDIE GetDIE(const DIERef &die_ref);
- /// Returns the AT_Name of this DIE, if it exists, without parsing the entire
- /// compile unit. An empty is string is returned upon error or if the
- /// attribute is not present.
- llvm::StringRef PeekDIEName(const DIERef &die_ref);
-
enum {
eDumpFlag_Verbose = (1 << 0), // Verbose dumping
eDumpFlag_ShowForm = (1 << 1), // Show the DW_form type
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
index 4da0d56fdcacb..79400e36e04f3 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
@@ -9,7 +9,7 @@
#include "Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h"
#include "Plugins/SymbolFile/DWARF/DWARFDebugInfo.h"
#include "Plugins/SymbolFile/DWARF/DWARFDeclContext.h"
-#include "Plugins/SymbolFile/DWARF/SymbolFileDWARFDwo.h"
+#include "Plugins/SymbolFile/DWARF/LogChannelDWARF.h"
#include "lldb/Core/Module.h"
#include "lldb/Utility/RegularExpression.h"
#include "lldb/Utility/Stream.h"
@@ -48,26 +48,30 @@ DebugNamesDWARFIndex::GetUnits(const DebugNames
&debug_names) {
return result;
}
-std::optional
-DebugNamesDWARFIndex::ToDIERef(const DebugNames::Entry &entry) const {
+DWARFUnit *
+DebugNamesDWARFIndex::GetNonSkeletonUnit(const DebugNames::Entry &entry) const
{
// Look for a DWARF unit offset (CU offset or local TU offset) as they are
// both offsets into the .debug_info section.
std::optional unit_offset = entry.getCUOffset();
if (!unit_offset) {
unit_offset = entry.getLocalTUOffset();
if (!unit_offset)
- return std::nullopt;
+ return nullptr;
}
DWARFUnit *cu =
m_debug_info.GetUnitAtOffset(DIERef::Section::DebugInfo, *unit_offset);
- if (!cu)
-return std::nullopt;
+ return cu ? &cu->GetNonSkeletonUnit() : nullptr;
+}
- cu = &cu->GetNonSkeletonUnit();
+std::optional
+DebugNamesDWARFIndex::ToDIERef(const DebugNames::Entry &entry) const {
+ DWARFUnit *unit = GetNonSkeletonUnit(entry);
+ if (!unit)
+return std::nullopt;
if (std::optional die_offset = entry.getDIE
[Lldb-commits] [lldb] [lldb/dwarf] Fix DW_IDX_parent processing for split dwarf (PR #92745)
@@ -9,7 +9,7 @@ #include "Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h" #include "Plugins/SymbolFile/DWARF/DWARFDebugInfo.h" #include "Plugins/SymbolFile/DWARF/DWARFDeclContext.h" -#include "Plugins/SymbolFile/DWARF/SymbolFileDWARFDwo.h" +#include "Plugins/SymbolFile/DWARF/LogChannelDWARF.h" labath wrote: Not exactly. I wanted to remove (unused) SymbolFileDWARFDwo since it was potentially signalling breaking of abstractions, and then I got carried away include gardening (my editor now highlights unused headers :/). Let me put those in separately. https://github.com/llvm/llvm-project/pull/92745 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] 9677e81 - [lldb/dwarf] Fix DW_IDX_parent processing for split dwarf (#92745)
Author: Pavel Labath
Date: 2024-05-21T12:45:59+02:00
New Revision: 9677e8171c19f1a15c0234724f083bc9473c545a
URL:
https://github.com/llvm/llvm-project/commit/9677e8171c19f1a15c0234724f083bc9473c545a
DIFF:
https://github.com/llvm/llvm-project/commit/9677e8171c19f1a15c0234724f083bc9473c545a.diff
LOG: [lldb/dwarf] Fix DW_IDX_parent processing for split dwarf (#92745)
DWARFDebugInfo only knows how to resolve references in its own file, but
in split dwarf, the index entries will refer to DIEs in the separate
(DWO) file. To resolve the DIERef correctly we'd either need to go
through the SymbolFileDWARF to get the full logic for resolving a
DIERef, or use the fact that ToDIERef already looks up the correct unit
while computing its result.
This patch does the latter.
This bug manifested itself in not being able to find type definitions
for types in namespaces, so I've modified one of our type resolving test
cases to run with debug_names, and added a namespaced class into it (it
originally contained only a top-level class).
Added:
Modified:
lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h
lldb/test/API/lang/cpp/limit-debug-info/Makefile
lldb/test/API/lang/cpp/limit-debug-info/TestWithLimitDebugInfo.py
lldb/test/API/lang/cpp/limit-debug-info/base.cpp
lldb/test/API/lang/cpp/limit-debug-info/base.h
lldb/test/API/lang/cpp/limit-debug-info/derived.cpp
lldb/test/API/lang/cpp/limit-debug-info/derived.h
lldb/test/API/lang/cpp/limit-debug-info/main.cpp
lldb/unittests/SymbolFile/DWARF/DWARFDIETest.cpp
Removed:
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
index 44febcfac3b00..d28da728728e5 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
@@ -259,9 +259,3 @@ DWARFDebugInfo::GetDIE(const DIERef &die_ref) {
return cu->GetNonSkeletonUnit().GetDIE(die_ref.die_offset());
return DWARFDIE(); // Not found
}
-
-llvm::StringRef DWARFDebugInfo::PeekDIEName(const DIERef &die_ref) {
- if (DWARFUnit *cu = GetUnit(die_ref))
-return cu->GetNonSkeletonUnit().PeekDIEName(die_ref.die_offset());
- return llvm::StringRef();
-}
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
index c1f0cb0203fb7..456ebd908ccb2 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
@@ -44,11 +44,6 @@ class DWARFDebugInfo {
bool ContainsTypeUnits();
DWARFDIE GetDIE(const DIERef &die_ref);
- /// Returns the AT_Name of this DIE, if it exists, without parsing the entire
- /// compile unit. An empty is string is returned upon error or if the
- /// attribute is not present.
- llvm::StringRef PeekDIEName(const DIERef &die_ref);
-
enum {
eDumpFlag_Verbose = (1 << 0), // Verbose dumping
eDumpFlag_ShowForm = (1 << 1), // Show the DW_form type
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
index 4da0d56fdcacb..fcf29470b9485 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
@@ -48,26 +48,30 @@ DebugNamesDWARFIndex::GetUnits(const DebugNames
&debug_names) {
return result;
}
-std::optional
-DebugNamesDWARFIndex::ToDIERef(const DebugNames::Entry &entry) const {
+DWARFUnit *
+DebugNamesDWARFIndex::GetNonSkeletonUnit(const DebugNames::Entry &entry) const
{
// Look for a DWARF unit offset (CU offset or local TU offset) as they are
// both offsets into the .debug_info section.
std::optional unit_offset = entry.getCUOffset();
if (!unit_offset) {
unit_offset = entry.getLocalTUOffset();
if (!unit_offset)
- return std::nullopt;
+ return nullptr;
}
DWARFUnit *cu =
m_debug_info.GetUnitAtOffset(DIERef::Section::DebugInfo, *unit_offset);
- if (!cu)
-return std::nullopt;
+ return cu ? &cu->GetNonSkeletonUnit() : nullptr;
+}
- cu = &cu->GetNonSkeletonUnit();
+std::optional
+DebugNamesDWARFIndex::ToDIERef(const DebugNames::Entry &entry) const {
+ DWARFUnit *unit = GetNonSkeletonUnit(entry);
+ if (!unit)
+return std::nullopt;
if (std::optional die_offset = entry.getDIEUnitOffset())
-return DIERef(cu->GetSymbolFileDWARF().GetFileIndex(),
- DIERef::Section::DebugInfo, cu->GetOffset() + *die_offset);
+return DIERef(unit->GetSymbolFileDWARF().GetFileIndex(),
+ DIERef::Section::DebugInfo, unit->GetOffset() + *die_offset);
[Lldb-commits] [lldb] [lldb/dwarf] Fix DW_IDX_parent processing for split dwarf (PR #92745)
https://github.com/labath closed https://github.com/llvm/llvm-project/pull/92745 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] 45a5d8d - [lldb] Clean up unused includes in DebugNamesDWARFIndex
Author: Pavel Labath Date: 2024-05-21T10:48:39Z New Revision: 45a5d8d6dba6156c24825b3a04378e999f8af82d URL: https://github.com/llvm/llvm-project/commit/45a5d8d6dba6156c24825b3a04378e999f8af82d DIFF: https://github.com/llvm/llvm-project/commit/45a5d8d6dba6156c24825b3a04378e999f8af82d.diff LOG: [lldb] Clean up unused includes in DebugNamesDWARFIndex Added: Modified: lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h Removed: diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp index fcf29470b9485..79400e36e04f3 100644 --- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp +++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp @@ -9,7 +9,7 @@ #include "Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h" #include "Plugins/SymbolFile/DWARF/DWARFDebugInfo.h" #include "Plugins/SymbolFile/DWARF/DWARFDeclContext.h" -#include "Plugins/SymbolFile/DWARF/SymbolFileDWARFDwo.h" +#include "Plugins/SymbolFile/DWARF/LogChannelDWARF.h" #include "lldb/Core/Module.h" #include "lldb/Utility/RegularExpression.h" #include "lldb/Utility/Stream.h" diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h index 392ec559dfb12..81fb8f88b805a 100644 --- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h +++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.h @@ -10,7 +10,6 @@ #define LLDB_SOURCE_PLUGINS_SYMBOLFILE_DWARF_DEBUGNAMESDWARFINDEX_H #include "Plugins/SymbolFile/DWARF/DWARFIndex.h" -#include "Plugins/SymbolFile/DWARF/LogChannelDWARF.h" #include "Plugins/SymbolFile/DWARF/ManualDWARFIndex.h" #include "Plugins/SymbolFile/DWARF/SymbolFileDWARF.h" #include "lldb/Utility/ConstString.h" ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Remove DWARFDebugInfo DIERef footguns (PR #92894)
https://github.com/labath created
https://github.com/llvm/llvm-project/pull/92894
DWARFDebugInfo doesn't know how to resolve the "file_index" component of a
DIERef. This patch removes GetUnit (in favor of existing
GetUnitContainingDIEOffset) and changes GetDIE to take only the components it
actually uses.
In the process it fixes GetCompleteObjCClass, although that bug was unlikely to
be noticed, since ObjC is only really used on darwin, and split DWARF isn't a
thing there.
>From e416051b09147b1083765795f711bb972e42a893 Mon Sep 17 00:00:00 2001
From: Pavel Labath
Date: Mon, 20 May 2024 14:51:52 +
Subject: [PATCH] [lldb] Remove DWARFDebugInfo DIERef footguns
DWARFDebugInfo doesn't know how to resolve the "file_index" component of
a DIERef. This patch removes GetUnit (in favor of existing
GetUnitContainingDIEOffset) and changes GetDIE to take only the
components it actually uses.
In the process it fixes GetCompleteObjCClass, although that bug was
unlikely to be noticed, since ObjC is only really used on darwin, and
split DWARF isn't a thing there.
---
.../Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp| 11 +++
.../Plugins/SymbolFile/DWARF/DWARFDebugInfo.h | 3 +--
.../SymbolFile/DWARF/DebugNamesDWARFIndex.cpp | 14 +++---
.../Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp | 7 ---
.../SymbolFile/DWARF/SymbolFileDWARFDwo.cpp| 2 +-
5 files changed, 16 insertions(+), 21 deletions(-)
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
index d28da728728e5..c37cc91e08ed1 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
@@ -222,10 +222,6 @@ DWARFUnit *DWARFDebugInfo::GetUnitAtOffset(DIERef::Section
section,
return result;
}
-DWARFUnit *DWARFDebugInfo::GetUnit(const DIERef &die_ref) {
- return GetUnitContainingDIEOffset(die_ref.section(), die_ref.die_offset());
-}
-
DWARFUnit *
DWARFDebugInfo::GetUnitContainingDIEOffset(DIERef::Section section,
dw_offset_t die_offset) {
@@ -253,9 +249,8 @@ bool DWARFDebugInfo::ContainsTypeUnits() {
//
// Get the DIE (Debug Information Entry) with the specified offset.
DWARFDIE
-DWARFDebugInfo::GetDIE(const DIERef &die_ref) {
- DWARFUnit *cu = GetUnit(die_ref);
- if (cu)
-return cu->GetNonSkeletonUnit().GetDIE(die_ref.die_offset());
+DWARFDebugInfo::GetDIE(DIERef::Section section, dw_offset_t die_offset) {
+ if (DWARFUnit *cu = GetUnitContainingDIEOffset(section, die_offset))
+return cu->GetNonSkeletonUnit().GetDIE(die_offset);
return DWARFDIE(); // Not found
}
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
index 456ebd908ccb2..4706b55d38ea9 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
@@ -38,11 +38,10 @@ class DWARFDebugInfo {
uint32_t *idx_ptr = nullptr);
DWARFUnit *GetUnitContainingDIEOffset(DIERef::Section section,
dw_offset_t die_offset);
- DWARFUnit *GetUnit(const DIERef &die_ref);
DWARFUnit *GetSkeletonUnit(DWARFUnit *dwo_unit);
DWARFTypeUnit *GetTypeUnitForHash(uint64_t hash);
bool ContainsTypeUnits();
- DWARFDIE GetDIE(const DIERef &die_ref);
+ DWARFDIE GetDIE(DIERef::Section section, dw_offset_t die_offset);
enum {
eDumpFlag_Verbose = (1 << 0), // Verbose dumping
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
index 79400e36e04f3..31bbad3e56269 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
@@ -195,17 +195,17 @@ void DebugNamesDWARFIndex::GetCompleteObjCClass(
if (!ref)
continue;
-DWARFUnit *cu = m_debug_info.GetUnit(*ref);
-if (!cu || !cu->Supports_DW_AT_APPLE_objc_complete_type()) {
- incomplete_types.push_back(*ref);
- continue;
-}
-
-DWARFDIE die = m_debug_info.GetDIE(*ref);
+SymbolFileDWARF &dwarf = *llvm::cast(
+m_module.GetSymbolFile()->GetBackingSymbolFile());
+DWARFDIE die = dwarf.GetDIE(*ref);
if (!die) {
ReportInvalidDIERef(*ref, class_name.GetStringRef());
continue;
}
+if (!die.GetCU()->Supports_DW_AT_APPLE_objc_complete_type()) {
+ incomplete_types.push_back(*ref);
+ continue;
+}
if (die.GetAttributeValueAsUnsigned(DW_AT_APPLE_objc_complete_type, 0)) {
// If we find the complete version we're done.
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
index f6f152726bf74..2bca27936c76b 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
+
[Lldb-commits] [lldb] [lldb] Remove DWARFDebugInfo DIERef footguns (PR #92894)
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Pavel Labath (labath)
Changes
DWARFDebugInfo doesn't know how to resolve the "file_index" component of a
DIERef. This patch removes GetUnit (in favor of existing
GetUnitContainingDIEOffset) and changes GetDIE to take only the components it
actually uses.
In the process it fixes GetCompleteObjCClass, although that bug was unlikely to
be noticed, since ObjC is only really used on darwin, and split DWARF isn't a
thing there.
---
Full diff: https://github.com/llvm/llvm-project/pull/92894.diff
5 Files Affected:
- (modified) lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp (+3-8)
- (modified) lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h (+1-2)
- (modified) lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
(+7-7)
- (modified) lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp (+4-3)
- (modified) lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARFDwo.cpp (+1-1)
``diff
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
index d28da728728e5..c37cc91e08ed1 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.cpp
@@ -222,10 +222,6 @@ DWARFUnit *DWARFDebugInfo::GetUnitAtOffset(DIERef::Section
section,
return result;
}
-DWARFUnit *DWARFDebugInfo::GetUnit(const DIERef &die_ref) {
- return GetUnitContainingDIEOffset(die_ref.section(), die_ref.die_offset());
-}
-
DWARFUnit *
DWARFDebugInfo::GetUnitContainingDIEOffset(DIERef::Section section,
dw_offset_t die_offset) {
@@ -253,9 +249,8 @@ bool DWARFDebugInfo::ContainsTypeUnits() {
//
// Get the DIE (Debug Information Entry) with the specified offset.
DWARFDIE
-DWARFDebugInfo::GetDIE(const DIERef &die_ref) {
- DWARFUnit *cu = GetUnit(die_ref);
- if (cu)
-return cu->GetNonSkeletonUnit().GetDIE(die_ref.die_offset());
+DWARFDebugInfo::GetDIE(DIERef::Section section, dw_offset_t die_offset) {
+ if (DWARFUnit *cu = GetUnitContainingDIEOffset(section, die_offset))
+return cu->GetNonSkeletonUnit().GetDIE(die_offset);
return DWARFDIE(); // Not found
}
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
index 456ebd908ccb2..4706b55d38ea9 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfo.h
@@ -38,11 +38,10 @@ class DWARFDebugInfo {
uint32_t *idx_ptr = nullptr);
DWARFUnit *GetUnitContainingDIEOffset(DIERef::Section section,
dw_offset_t die_offset);
- DWARFUnit *GetUnit(const DIERef &die_ref);
DWARFUnit *GetSkeletonUnit(DWARFUnit *dwo_unit);
DWARFTypeUnit *GetTypeUnitForHash(uint64_t hash);
bool ContainsTypeUnits();
- DWARFDIE GetDIE(const DIERef &die_ref);
+ DWARFDIE GetDIE(DIERef::Section section, dw_offset_t die_offset);
enum {
eDumpFlag_Verbose = (1 << 0), // Verbose dumping
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
index 79400e36e04f3..31bbad3e56269 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DebugNamesDWARFIndex.cpp
@@ -195,17 +195,17 @@ void DebugNamesDWARFIndex::GetCompleteObjCClass(
if (!ref)
continue;
-DWARFUnit *cu = m_debug_info.GetUnit(*ref);
-if (!cu || !cu->Supports_DW_AT_APPLE_objc_complete_type()) {
- incomplete_types.push_back(*ref);
- continue;
-}
-
-DWARFDIE die = m_debug_info.GetDIE(*ref);
+SymbolFileDWARF &dwarf = *llvm::cast(
+m_module.GetSymbolFile()->GetBackingSymbolFile());
+DWARFDIE die = dwarf.GetDIE(*ref);
if (!die) {
ReportInvalidDIERef(*ref, class_name.GetStringRef());
continue;
}
+if (!die.GetCU()->Supports_DW_AT_APPLE_objc_complete_type()) {
+ incomplete_types.push_back(*ref);
+ continue;
+}
if (die.GetAttributeValueAsUnsigned(DW_AT_APPLE_objc_complete_type, 0)) {
// If we find the complete version we're done.
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
b/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
index f6f152726bf74..2bca27936c76b 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp
@@ -1748,7 +1748,8 @@ SymbolFileDWARF::GetDIE(const DIERef &die_ref) {
if (SymbolFileDWARFDebugMap *debug_map = GetDebugMapSymfile()) {
symbol_file = debug_map->GetSymbolFileByOSOIndex(*file_index); // OSO
case
if (symbol_file)
-return symbol_file->DebugInfo().GetDIE(die_ref);
+return symbol_file->DebugInfo().GetDIE(die_ref.section(),
+
[Lldb-commits] [lldb] New ThreadPlanSingleThreadTimeout to resolve potential deadlock in single thread stepping (PR #90930)
jeffreytan81 wrote: @jimingham, I have revised the PR, let me know when you can review again. Thanks. https://github.com/llvm/llvm-project/pull/90930 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] A few updates around "transcript" (PR #92843)
royitaqi wrote: cc @clayborg and @jeffreytan81 https://github.com/llvm/llvm-project/pull/92843 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply [lldb][DWARF] Delay struct/class/union definition DIE searching when parsing declaration DIEs. (PR #92328)
@@ -2306,6 +2345,11 @@ bool DWARFASTParserClang::CompleteTypeFromDWARF(const DWARFDIE &die, if (!die) return false; + ParsedDWARFTypeAttributes attrs(die); ZequanWu wrote: > Why don't those cases lead to a crash? For example, what would happen if the > name was simply not in the index? That is specific to debug_names because the index gives us a declaration DIE when we are searching for definition DIE in 'FindDefinitionTypeForDWARFDeclContext'. Before, we didn't have the extra check, so we tries to complete the type from declaration DIE, which triggers an assertion in clang. However, it doesn't happen in manual index because we already explicitly checked `DW_AT_declaration` attributes when creating the manual index. It's guaranteed to find a definition DIE from `FindDefinitionTypeForDWARFDeclContext` or nothing (early bailout, won't try to complete it). > So it seems perfectly reasonable to have this check somewhere. I just want to > check whether this is the right place. I assume Greg's change at https://github.com/llvm/llvm-project/pull/91808 will also fix this problem by skipping forward declaration DIE when processing it, which is earlier check than this extra check added here. https://github.com/llvm/llvm-project/pull/92328 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Improve performance of .debug_names lookups when DW_IDX_parent attributes are used (PR #91808)
felipepiovezan wrote:
> I think #77696 is justification enough to add this check, but I don't think
> we should be implicating DW_IDX_parent_entries until we know how those came
> about.
This is pretty much my point, the workaround is fine, but I don't think the
source-code comment in this PR is. The suggestion provided LGTM.
My guess is that the examples Greg is encountering are just hitting the
slowness of DWARF parsing through other code paths (anything that causes too
many calls to "ProcessEntries", and IDX_parent addressed _one_ of those,
@Michael137 was just asking me the other day if we can address some of the
other queries as well).
> I believe we could add a debug_names entry for std::ios_base (and refer to it
> from the entry for std::ios_base::seekdir), as long as the ios_base entry was
> not present in any of the name tables.
Yup, this is very doable!
> I was not able to produce a DW_IDX_parent_entries referring to any kind of
> class type
When I compile the following with `clang++ test.cpp -c -gdwarf-5 -O0 -o - |
dwarfdump --debug-names -`, I get the right parent chain. Is this not true when
you try it?
```
namespace A { namespace B { struct State { class InnerState{}; }; } }
A::B::State::InnerState get_state() { return A::B::State::InnerState(); }
```
```
String: 0x00c4 "InnerState"
Entry @ 0xbe {
Abbrev: 0x138
Tag: DW_TAG_class_type
DW_IDX_die_offset: 0x003f
DW_IDX_parent: Entry @ 0xde
}
String: 0x00be "State"
Entry @ 0xde {
Abbrev: 0x9b8
Tag: DW_TAG_structure_type
DW_IDX_die_offset: 0x0039
DW_IDX_parent: Entry @ 0xe9
}
String: 0x00bc "B"
Entry @ 0xe9 {
Abbrev: 0x1cb8
Tag: DW_TAG_namespace
DW_IDX_die_offset: 0x0037
DW_IDX_parent: Entry @ 0xd7
}
String: 0x00ba "A"
Entry @ 0xd7 {
Abbrev: 0x1c98
Tag: DW_TAG_namespace
DW_IDX_die_offset: 0x0035
DW_IDX_parent:
}
```
https://github.com/llvm/llvm-project/pull/91808
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Improve performance of .debug_names lookups when DW_IDX_parent attributes are used (PR #91808)
labath wrote: > > I was not able to produce a DW_IDX_parent_entries referring to any kind of > > class type > > When I compile the following with `clang++ test.cpp -c -gdwarf-5 -O0 -o - | > dwarfdump --debug-names -`, I get the right parent chain. Is this not true > when you try it? I do get it. It was my bad, I had a leftover -fdebug-types-section flag in my test cmd from when I was experimenting with the debug types section. I guess that's another thing to chalk up to https://github.com/llvm/llvm-project/issues/77696 https://github.com/llvm/llvm-project/pull/91808 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [compiler-rt] [lldb] [llvm] [Support] Remove terminfo dependency (PR #92865)
keith wrote: looks like the .bazelrc still has mentions of this https://github.com/llvm/llvm-project/pull/92865 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] 98f105a - [lldb] Add the word "backtrace" to bt help string (#92618)
Author: Dave Lee
Date: 2024-05-21T10:16:51-07:00
New Revision: 98f105a1304e8d3fd0a0234d1d488ed513833df8
URL:
https://github.com/llvm/llvm-project/commit/98f105a1304e8d3fd0a0234d1d488ed513833df8
DIFF:
https://github.com/llvm/llvm-project/commit/98f105a1304e8d3fd0a0234d1d488ed513833df8.diff
LOG: [lldb] Add the word "backtrace" to bt help string (#92618)
We noticed that `apropos backtrace` did not return the `bt` alias. This change
adds the
word "backtrace" to the help for `bt`. It also updates `thread backtrace` to
keep the
language used roughly in sync.
Added:
Modified:
lldb/source/Commands/CommandObjectThread.cpp
lldb/source/Interpreter/CommandInterpreter.cpp
Removed:
diff --git a/lldb/source/Commands/CommandObjectThread.cpp
b/lldb/source/Commands/CommandObjectThread.cpp
index 4397ee14ea074..db96ee2cec383 100644
--- a/lldb/source/Commands/CommandObjectThread.cpp
+++ b/lldb/source/Commands/CommandObjectThread.cpp
@@ -114,8 +114,8 @@ class CommandObjectThreadBacktrace : public
CommandObjectIterateOverThreads {
CommandObjectThreadBacktrace(CommandInterpreter &interpreter)
: CommandObjectIterateOverThreads(
interpreter, "thread backtrace",
-"Show thread call stacks. Defaults to the current thread, thread "
-"indexes can be specified as arguments.\n"
+"Show backtraces of thread call stacks. Defaults to the current "
+"thread, thread indexes can be specified as arguments.\n"
"Use the thread-index \"all\" to see all threads.\n"
"Use the thread-index \"unique\" to see threads grouped by unique "
"call stacks.\n"
diff --git a/lldb/source/Interpreter/CommandInterpreter.cpp
b/lldb/source/Interpreter/CommandInterpreter.cpp
index 811726e30af4d..7f21f382adb83 100644
--- a/lldb/source/Interpreter/CommandInterpreter.cpp
+++ b/lldb/source/Interpreter/CommandInterpreter.cpp
@@ -828,11 +828,11 @@ void CommandInterpreter::LoadCommandDictionary() {
std::unique_ptr bt_regex_cmd_up(
new CommandObjectRegexCommand(
*this, "_regexp-bt",
- "Show the current thread's call stack. Any numeric argument "
- "displays at most that many "
- "frames. The argument 'all' displays all threads. Use 'settings"
- " set frame-format' to customize the printing of individual frames "
- "and 'settings set thread-format' to customize the thread header.",
+ "Show backtrace of the current thread's call stack. Any numeric "
+ "argument displays at most that many frames. The argument 'all' "
+ "displays all threads. Use 'settings set frame-format' to customize
"
+ "the printing of individual frames and 'settings set thread-format' "
+ "to customize the thread header.",
"bt [ | all]", 0, false));
if (bt_regex_cmd_up) {
// accept but don't document "bt -c " -- before bt was a regex
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Add the word "backtrace" to bt help string (PR #92618)
https://github.com/kastiglione closed https://github.com/llvm/llvm-project/pull/92618 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [compiler-rt] [lldb] [llvm] [Support] Remove terminfo dependency (PR #92865)
JDevlieghere wrote: No objections in the context of LLDB. We don't use terminfo directly (although I think editline does, but that isn't affected by this) and if we want the TUI we depend on curses anyway. https://github.com/llvm/llvm-project/pull/92865 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply [lldb][DWARF] Delay struct/class/union definition DIE searching when parsing declaration DIEs. (PR #92328)
@@ -2306,6 +2345,11 @@ bool DWARFASTParserClang::CompleteTypeFromDWARF(const DWARFDIE &die, if (!die) return false; + ParsedDWARFTypeAttributes attrs(die); labath wrote: > > Why don't those cases lead to a crash? For example, what would happen if > > the name was simply not in the index? > > That is specific to debug_names because the index gives us a declaration DIE > when we are searching for definition DIE in > 'FindDefinitionTypeForDWARFDeclContext'. Before, we didn't have the extra > check, so we tries to complete the type from declaration DIE, which triggers > an assertion in clang. However, it doesn't happen in manual index because we > already explicitly checked `DW_AT_declaration` attributes when creating the > manual index. It's guaranteed to find a definition DIE from > `FindDefinitionTypeForDWARFDeclContext` or nothing (early bailout, won't try > to complete it). Okay, so when I said "specific to debug_names", I was mainly thinking of the case where the index doesn't find anything (i.e., we attempt to complete a type and the index (any kind of index) says we can't, because it doesn't have the definition). Based on what you've said (and some more pondering), I think can now answer that question for myself (please let me know if this is wrong): The reason why this crash does not happen when attempting to complete a type with no definition is because this function is the actual implementation of "completing a type from a defintion". I.e., it expects to be called with a definition DIE, and will never be called if there is no definition (we will bail out, and possibly "forcefully" complete the type somewhere higher up the stack). If that is true, then I think Greg's patch is a better fix for this problem, and it also sidesteps all of the performance/memory tradeoff discussions. https://github.com/llvm/llvm-project/pull/92328 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Summarize std::string's when created from data. (PR #89110)
https://github.com/Jlalond updated
https://github.com/llvm/llvm-project/pull/89110
>From e0316188d22605c670079e37855d3d8b5c944cee Mon Sep 17 00:00:00 2001
From: Jacob John Lalonde
Date: Wed, 10 Apr 2024 14:33:40 -0700
Subject: [PATCH 1/5] Fix bug where an sbvalue containing a std::string created
from data would not use the built in summary provider, but default to the
valueobjectprinter
std::string children are normally formatted as their summary [my_string_prop] =
"hello world".
Sbvalues created from data do not follow the same summary formatter path and
instead hit the value object printer.
This change fixes that and adds tests for the future.
Added tests in TestDataFormatterStdString.py and a formatter for a custom
struct.
---
.../Plugins/Language/CPlusPlus/LibStdcpp.cpp | 31 ++--
.../string/ConvertToDataFormatter.py | 50 +++
.../string/TestDataFormatterStdString.py | 28 +++
.../libstdcpp/string/main.cpp | 13 +
4 files changed, 118 insertions(+), 4 deletions(-)
create mode 100644
lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/ConvertToDataFormatter.py
diff --git a/lldb/source/Plugins/Language/CPlusPlus/LibStdcpp.cpp
b/lldb/source/Plugins/Language/CPlusPlus/LibStdcpp.cpp
index 86bb575af5ca3..0da01e9ca07fb 100644
--- a/lldb/source/Plugins/Language/CPlusPlus/LibStdcpp.cpp
+++ b/lldb/source/Plugins/Language/CPlusPlus/LibStdcpp.cpp
@@ -254,13 +254,13 @@ bool
lldb_private::formatters::LibStdcppStringSummaryProvider(
} else
addr_of_string =
valobj.GetAddressOf(scalar_is_load_addr, &addr_type);
- if (addr_of_string != LLDB_INVALID_ADDRESS) {
+ if (addr_of_string != LLDB_INVALID_ADDRESS ||
+addr_type == eAddressTypeHost) {
switch (addr_type) {
case eAddressTypeLoad: {
ProcessSP process_sp(valobj.GetProcessSP());
if (!process_sp)
return false;
-
StringPrinter::ReadStringAndDumpToStreamOptions options(valobj);
Status error;
lldb::addr_t addr_of_data =
@@ -287,8 +287,31 @@ bool
lldb_private::formatters::LibStdcppStringSummaryProvider(
} else
return true;
} break;
-case eAddressTypeHost:
- break;
+case eAddressTypeHost: {
+ // We have the host address of our std::string
+ // But we need to read the pointee data from the debugged process.
+ ProcessSP process_sp(valobj.GetProcessSP());
+ DataExtractor data;
+ Status error;
+ valobj.GetData(data, error);
+ if (error.Fail())
+return false;
+ // We want to read the address from std::string, which is the first 8
bytes.
+ lldb::offset_t offset = 0;
+ lldb::addr_t addr = data.GetAddress(&offset);
+ if (!addr)
+ {
+stream.Printf("nullptr");
+return true;
+ }
+
+ std::string contents;
+ process_sp->ReadCStringFromMemory(addr, contents, error);
+ if (error.Fail())
+return false;
+ stream.Printf("%s", contents.c_str());
+ return true;
+} break;
case eAddressTypeInvalid:
case eAddressTypeFile:
break;
diff --git
a/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/ConvertToDataFormatter.py
b/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/ConvertToDataFormatter.py
new file mode 100644
index 0..57e42c920f829
--- /dev/null
+++
b/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/ConvertToDataFormatter.py
@@ -0,0 +1,50 @@
+# coding=utf8
+"""
+Helper formmater to verify Std::String by created via SBData
+"""
+
+import lldb
+
+class SyntheticFormatter:
+def __init__(self, valobj, dict):
+self.valobj = valobj
+
+def num_children(self):
+return 6
+
+def has_children(self):
+return True
+
+def get_child_at_index(self, index):
+name = None
+match index:
+case 0:
+name = "short_string"
+case 1:
+name = "long_string"
+case 2:
+name = "short_string_ptr"
+case 3:
+name = "long_string_ptr"
+case 4:
+name = "short_string_ref"
+case 5:
+name = "long_string_ref"
+case _:
+return None
+
+child = self.valobj.GetChildMemberWithName(name)
+valType = child.GetType()
+return self.valobj.CreateValueFromData(name,
+child.GetData(),
+valType)
+
+
+def __lldb_init_module(debugger, dict):
+typeName = "string_container"
+debugger.HandleCommand(
+'type synthetic add -x "'
++ typeName
++ '" --python-class '
++ f"{__name__}.SyntheticFormatter"
+)
diff --git
a/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/TestDataFormatterStdString.py
b/lldb/test/A
[Lldb-commits] [lldb] [lldb] Refactor string manipulation in Debugger.cpp (#91209) (PR #92565)
https://github.com/augusto2112 approved this pull request. https://github.com/llvm/llvm-project/pull/92565 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Summarize std::string's when created from data. (PR #89110)
github-actions[bot] wrote:
:warning: Python code formatter, darker found issues in your code. :warning:
You can test this locally with the following command:
``bash
darker --check --diff -r
3074060d6a1d7d2e74cb767876bd9e5192d12007...29b55fcba7a7f7b7ab0581f7804b9ad70863b0ff
lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/ConvertToDataFormatter.py
lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/TestDataFormatterStdString.py
``
View the diff from darker here.
``diff
--- ConvertToDataFormatter.py 2024-04-16 22:18:59.00 +
+++ ConvertToDataFormatter.py 2024-05-21 17:57:50.207877 +
@@ -2,10 +2,11 @@
"""
Helper formmater to verify Std::String by created via SBData
"""
import lldb
+
class SyntheticFormatter:
def __init__(self, valobj, dict):
self.valobj = valobj
@@ -33,13 +34,11 @@
case _:
return None
child = self.valobj.GetChildMemberWithName(name)
valType = child.GetType()
-return self.valobj.CreateValueFromData(name,
-child.GetData(),
-valType)
+return self.valobj.CreateValueFromData(name, child.GetData(), valType)
def __lldb_init_module(debugger, dict):
typeName = "string_container"
debugger.HandleCommand(
--- TestDataFormatterStdString.py 2024-04-24 20:52:10.00 +
+++ TestDataFormatterStdString.py 2024-05-21 17:57:50.260723 +
@@ -92,11 +92,10 @@
self.runCmd("next")
self.assertEqual(var_S.GetSummary(), 'L"!"', "new S summary wrong")
-
@add_test_categories(["libstdcxx"])
@expectedFailureAll(bugnumber="llvm.org/pr50861", compiler="gcc")
def test_std_string_as_data(self):
"""Test that strings created via SBValue::CreateFromData display
correctly."""
self.build()
@@ -122,26 +121,32 @@
var_str_container = self.frame().FindVariable("sc")
self.assertEqual(
var_str_container.GetChildAtIndex(0).GetSummary(),
'"u22"',
-"string container child wrong")
+"string container child wrong",
+)
self.assertEqual(
var_str_container.GetChildAtIndex(1).GetSummary(),
'"quite a long std::string with lots of info inside it inside a
struct"',
-"string container child wrong")
+"string container child wrong",
+)
self.assertEqual(
var_str_container.GetChildAtIndex(2).GetSummary(),
'"u22"',
-"string container child wrong")
+"string container child wrong",
+)
self.assertEqual(
var_str_container.GetChildAtIndex(3).GetSummary(),
'"quite a long std::string with lots of info inside it inside a
struct"',
-"string container child wrong")
+"string container child wrong",
+)
self.assertEqual(
var_str_container.GetChildAtIndex(4).GetSummary(),
'"u22"',
-"string container child wrong")
+"string container child wrong",
+)
self.assertEqual(
var_str_container.GetChildAtIndex(5).GetSummary(),
'"quite a long std::string with lots of info inside it inside a
struct"',
-"string container child wrong")
+"string container child wrong",
+)
``
https://github.com/llvm/llvm-project/pull/89110
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Summarize std::string's when created from data. (PR #89110)
github-actions[bot] wrote:
:warning: C/C++ code formatter, clang-format found issues in your code.
:warning:
You can test this locally with the following command:
``bash
git-clang-format --diff 3074060d6a1d7d2e74cb767876bd9e5192d12007
29b55fcba7a7f7b7ab0581f7804b9ad70863b0ff --
lldb/source/Plugins/Language/CPlusPlus/LibStdcpp.cpp
lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/main.cpp
``
View the diff from clang-format here.
``diff
diff --git
a/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/main.cpp
b/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/main.cpp
index b169a2e694..838781935f 100644
---
a/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/main.cpp
+++
b/lldb/test/API/functionalities/data-formatter/data-formatter-stl/libstdcpp/string/main.cpp
@@ -1,12 +1,12 @@
#include
struct string_container {
-std::string short_string;
-std::string long_string;
-std::string *short_string_ptr = &short_string;
-std::string *long_string_ptr = &long_string;
-std::string &short_string_ref = short_string;
-std::string &long_string_ref = long_string;
+ std::string short_string;
+ std::string long_string;
+ std::string *short_string_ptr = &short_string;
+ std::string *long_string_ptr = &long_string;
+ std::string &short_string_ref = short_string;
+ std::string &long_string_ref = long_string;
};
int main()
@@ -22,7 +22,8 @@ int main()
std::string *pq = &q, *pQ = &Q;
string_container sc;
sc.short_string = "u22";
-sc.long_string = "quite a long std::string with lots of info inside it
inside a struct";
+sc.long_string =
+"quite a long std::string with lots of info inside it inside a struct";
S.assign(L"!"); // Set break point at this line.
return 0;
``
https://github.com/llvm/llvm-project/pull/89110
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply [lldb][DWARF] Delay struct/class/union definition DIE searching when parsing declaration DIEs. (PR #92328)
@@ -2306,6 +2345,11 @@ bool DWARFASTParserClang::CompleteTypeFromDWARF(const DWARFDIE &die, if (!die) return false; + ParsedDWARFTypeAttributes attrs(die); ZequanWu wrote: > The reason why this crash does not happen when attempting to complete a type > with no definition is because this function is the actual implementation of > "completing a type from a defintion". I.e., it expects to be called with a > definition DIE, and will never be called if there is no definition (we will > bail out, and possibly "forcefully" complete the type somewhere higher up the > stack). Yes, it should only be called with definition DIE. This extra check just a double-check for it with a bit performance cost. If there's no definition DIE, `SymbolFileDWARF::CompleteType` does an early return. This behaviour is unchanged. Forceful completion behaviour is also remained unchanged, happens when failed to find definition DIE of a containing record type. > If that is true, then I think Greg's patch is a better fix for this problem, > and it also sidesteps all of the performance/memory tradeoff discussions. I agree. If we can ensure the DIE is always a definition DIE for both manual index and debug_names index at a earlier time, there's no need for this check. I'll remove this check when Greg's change is in. https://github.com/llvm/llvm-project/pull/92328 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,28 @@
+%extend lldb::SBAddressRangeList {
+#ifdef SWIGPYTHON
+%pythoncode%{
+def __len__(self):
+ '''Return the number of address ranges in a lldb.SBAddressRangeList
object.'''
+ return self.GetSize()
+
+def __iter__(self):
+ '''Iterate over all the address ranges in a lldb.SBAddressRangeList
object.'''
+ return lldb_iter(self, 'GetSize', 'GetAddressRangeAtIndex')
+
+def __getitem__(self, idx):
+ '''Get the address range at a given index in an lldb.SBAddressRangeList
object.'''
+ if type(idx) == int:
+if idx >= self.GetSize():
+ raise IndexError("list index out of range")
+return self.GetAddressRangeAtIndex(idx)
+ else:
+print("error: unsupported idx type: %s" % type(key))
+return None
bulbazord wrote:
If the type of index is not an int, why not raise an exception? We raise an
exception if the index is an int and is out of range.
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,28 @@
+%extend lldb::SBAddressRangeList {
+#ifdef SWIGPYTHON
+%pythoncode%{
+def __len__(self):
+ '''Return the number of address ranges in a lldb.SBAddressRangeList
object.'''
+ return self.GetSize()
+
+def __iter__(self):
+ '''Iterate over all the address ranges in a lldb.SBAddressRangeList
object.'''
+ return lldb_iter(self, 'GetSize', 'GetAddressRangeAtIndex')
+
+def __getitem__(self, idx):
+ '''Get the address range at a given index in an lldb.SBAddressRangeList
object.'''
+ if type(idx) == int:
+if idx >= self.GetSize():
+ raise IndexError("list index out of range")
+return self.GetAddressRangeAtIndex(idx)
+ else:
+print("error: unsupported idx type: %s" % type(key))
+return None
+def __repr__(self):
+ import lldb
bulbazord wrote:
Same here, why does this import lldb?
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,11 @@
+%extend lldb::SBAddressRange {
+#ifdef SWIGPYTHON
+%pythoncode%{
+ def __repr__(self):
+import lldb
bulbazord wrote:
Why does this import lldb? Shouldn't it already be imported?
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,91 @@
+//===-- SBAddressRangeList.cpp
===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/API/SBAddressRangeList.h"
+#include "Utils.h"
+#include "lldb/API/SBAddressRange.h"
+#include "lldb/API/SBStream.h"
+#include "lldb/Core/AddressRangeListImpl.h"
+#include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Stream.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+SBAddressRangeList::SBAddressRangeList()
+: m_opaque_up(new AddressRangeListImpl()) {
+ LLDB_INSTRUMENT_VA(this);
+}
+
+SBAddressRangeList::SBAddressRangeList(const SBAddressRangeList &rhs)
+: m_opaque_up(new AddressRangeListImpl(*rhs.m_opaque_up)) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+}
+
+SBAddressRangeList::~SBAddressRangeList() = default;
+
+const SBAddressRangeList &
+SBAddressRangeList::operator=(const SBAddressRangeList &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ if (this != &rhs)
+*m_opaque_up = *rhs.m_opaque_up;
+ return *this;
+}
+
+uint32_t SBAddressRangeList::GetSize() const {
+ LLDB_INSTRUMENT_VA(this);
+
+ return m_opaque_up->GetSize();
+}
+
+SBAddressRange SBAddressRangeList::GetAddressRangeAtIndex(uint64_t idx) {
+ LLDB_INSTRUMENT_VA(this, idx);
+
+ SBAddressRange sb_addr_range;
+ (*sb_addr_range.m_opaque_up) = m_opaque_up->GetAddressRangeAtIndex(idx);
+ return sb_addr_range;
+}
+
+void SBAddressRangeList::Clear() {
+ LLDB_INSTRUMENT_VA(this);
+
+ m_opaque_up->Clear();
+}
+
+void SBAddressRangeList::Append(const SBAddressRange &sb_addr_range) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range);
+
+ m_opaque_up->Append(*sb_addr_range.m_opaque_up);
+}
+
+void SBAddressRangeList::Append(const SBAddressRangeList &sb_addr_range_list) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range_list);
+
+ m_opaque_up->Append(*sb_addr_range_list.m_opaque_up);
+}
+
+bool SBAddressRangeList::GetDescription(SBStream &description,
+SBTarget *target) {
+ LLDB_INSTRUMENT_VA(this, description);
bulbazord wrote:
Missing target from the instrumentation macro.
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,91 @@
+//===-- SBAddressRangeList.cpp
===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/API/SBAddressRangeList.h"
+#include "Utils.h"
+#include "lldb/API/SBAddressRange.h"
+#include "lldb/API/SBStream.h"
+#include "lldb/Core/AddressRangeListImpl.h"
+#include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Stream.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+SBAddressRangeList::SBAddressRangeList()
+: m_opaque_up(new AddressRangeListImpl()) {
+ LLDB_INSTRUMENT_VA(this);
+}
+
+SBAddressRangeList::SBAddressRangeList(const SBAddressRangeList &rhs)
+: m_opaque_up(new AddressRangeListImpl(*rhs.m_opaque_up)) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+}
+
+SBAddressRangeList::~SBAddressRangeList() = default;
+
+const SBAddressRangeList &
+SBAddressRangeList::operator=(const SBAddressRangeList &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ if (this != &rhs)
+*m_opaque_up = *rhs.m_opaque_up;
+ return *this;
+}
+
+uint32_t SBAddressRangeList::GetSize() const {
+ LLDB_INSTRUMENT_VA(this);
+
+ return m_opaque_up->GetSize();
+}
+
+SBAddressRange SBAddressRangeList::GetAddressRangeAtIndex(uint64_t idx) {
+ LLDB_INSTRUMENT_VA(this, idx);
+
+ SBAddressRange sb_addr_range;
+ (*sb_addr_range.m_opaque_up) = m_opaque_up->GetAddressRangeAtIndex(idx);
+ return sb_addr_range;
+}
+
+void SBAddressRangeList::Clear() {
+ LLDB_INSTRUMENT_VA(this);
+
+ m_opaque_up->Clear();
+}
+
+void SBAddressRangeList::Append(const SBAddressRange &sb_addr_range) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range);
+
+ m_opaque_up->Append(*sb_addr_range.m_opaque_up);
+}
+
+void SBAddressRangeList::Append(const SBAddressRangeList &sb_addr_range_list) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range_list);
+
+ m_opaque_up->Append(*sb_addr_range_list.m_opaque_up);
+}
+
+bool SBAddressRangeList::GetDescription(SBStream &description,
+SBTarget *target) {
+ LLDB_INSTRUMENT_VA(this, description);
+
+ const uint32_t num_ranges = GetSize();
+ bool is_first = true;
+ Stream &stream = description.ref();
+ stream << "[";
+ for (uint32_t i = 0; i < num_ranges; ++i) {
+if (is_first) {
+ is_first = false;
+} else {
+ stream.Printf(", ");
+}
bulbazord wrote:
I think this doesn't do what you intended it to do. For example, a list with 3
elements will look like this:
```
[ , , ]
```
When you probably wanted something like this:
```
[, , ]
```
The first element isn't the one you care about, it's the last one that doesn't
need the `", "` appended to the end.
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,65 @@
+//===-- SBAddressRange.h *- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#ifndef LLDB_API_SBADDRESSRANGE_H
+#define LLDB_API_SBADDRESSRANGE_H
+
+#include "lldb/API/SBDefines.h"
+
+namespace lldb {
+
+class LLDB_API SBAddressRange {
+public:
+ SBAddressRange();
+
+ SBAddressRange(const lldb::SBAddressRange &rhs);
+
+ SBAddressRange(lldb::SBAddress addr, lldb::addr_t byte_size);
+
+ ~SBAddressRange();
+
+ const lldb::SBAddressRange &operator=(const lldb::SBAddressRange &rhs);
+
+ void Clear();
+
+ /// Check the address range refers to a valid base address and has a byte
+ /// size greater than zero.
+ ///
+ /// \return
+ /// True if the address range is valid, false otherwise.
+ bool IsValid() const;
+
+ /// Get the base address of the range.
+ ///
+ /// \return
+ /// Base address object.
+ lldb::SBAddress GetBaseAddress() const;
+
+ /// Get the byte size of this range.
+ ///
+ /// \return
+ /// The size in bytes of this address range.
+ lldb::addr_t GetByteSize() const;
+
+ bool operator==(const SBAddressRange &rhs);
+
+ bool operator!=(const SBAddressRange &rhs);
+
+ bool GetDescription(lldb::SBStream &description, SBTarget *target);
bulbazord wrote:
In addition, if this doesn't modify the `SBTarget`, please mark it as const --
`const SBTarget &target`. We have some tooling downstream that works better if
parameters that aren't mutated are marked as `const`.
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,91 @@
+//===-- SBAddressRangeList.cpp
===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/API/SBAddressRangeList.h"
+#include "Utils.h"
+#include "lldb/API/SBAddressRange.h"
+#include "lldb/API/SBStream.h"
+#include "lldb/Core/AddressRangeListImpl.h"
+#include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Stream.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+SBAddressRangeList::SBAddressRangeList()
+: m_opaque_up(new AddressRangeListImpl()) {
bulbazord wrote:
Can we use `make_unique` here instead of a raw `new`?
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,54 @@
+//===-- SBAddressRangeList.h *- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#ifndef LLDB_API_SBADDRESSRANGELIST_H
+#define LLDB_API_SBADDRESSRANGELIST_H
+
+#include
+
+#include "lldb/API/SBDefines.h"
+
+namespace lldb_private {
+class AddressRangeListImpl;
+}
+
+namespace lldb {
+
+class LLDB_API SBAddressRangeList {
+public:
+ SBAddressRangeList();
+
+ SBAddressRangeList(const lldb::SBAddressRangeList &rhs);
+
+ ~SBAddressRangeList();
+
+ const lldb::SBAddressRangeList &
+ operator=(const lldb::SBAddressRangeList &rhs);
+
+ uint32_t GetSize() const;
+
+ void Clear();
+
+ SBAddressRange GetAddressRangeAtIndex(uint64_t idx);
bulbazord wrote:
There's still a difference here. Why is the index uint64_t while `GetSize`
returns a `uint32_t`?
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,103 @@
+//===-- SBAddressRange.cpp
===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/API/SBAddressRange.h"
+#include "Utils.h"
+#include "lldb/API/SBAddress.h"
+#include "lldb/API/SBStream.h"
+#include "lldb/API/SBTarget.h"
+#include "lldb/Core/AddressRange.h"
+#include "lldb/Core/Section.h"
+#include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Stream.h"
+#include
+#include
+
+using namespace lldb;
+using namespace lldb_private;
+
+SBAddressRange::SBAddressRange()
+: m_opaque_up(std::make_unique()) {
+ LLDB_INSTRUMENT_VA(this);
+}
+
+SBAddressRange::SBAddressRange(const SBAddressRange &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ m_opaque_up = clone(rhs.m_opaque_up);
+}
+
+SBAddressRange::SBAddressRange(lldb::SBAddress addr, lldb::addr_t byte_size)
+: m_opaque_up(std::make_unique(addr.ref(), byte_size)) {
+ LLDB_INSTRUMENT_VA(this, addr, byte_size);
+}
+
+SBAddressRange::~SBAddressRange() = default;
+
+const SBAddressRange &SBAddressRange::operator=(const SBAddressRange &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ if (this != &rhs)
+m_opaque_up = clone(rhs.m_opaque_up);
+ return *this;
+}
+
+bool SBAddressRange::operator==(const SBAddressRange &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ if (!IsValid() || !rhs.IsValid())
+return false;
+ return m_opaque_up->operator==(*(rhs.m_opaque_up));
+}
+
+bool SBAddressRange::operator!=(const SBAddressRange &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ return !(*this == rhs);
+}
+
+void SBAddressRange::Clear() {
+ LLDB_INSTRUMENT_VA(this);
+
+ m_opaque_up.reset();
+}
+
+bool SBAddressRange::IsValid() const {
+ LLDB_INSTRUMENT_VA(this);
+
+ return m_opaque_up && m_opaque_up->IsValid();
+}
+
+lldb::SBAddress SBAddressRange::GetBaseAddress() const {
+ LLDB_INSTRUMENT_VA(this);
+
+ if (!IsValid())
+return lldb::SBAddress();
+ return lldb::SBAddress(m_opaque_up->GetBaseAddress());
+}
+
+lldb::addr_t SBAddressRange::GetByteSize() const {
+ LLDB_INSTRUMENT_VA(this);
+
+ if (!IsValid())
+return 0;
+ return m_opaque_up->GetByteSize();
+}
+
+bool SBAddressRange::GetDescription(SBStream &description, SBTarget *target) {
+ LLDB_INSTRUMENT_VA(this, description);
bulbazord wrote:
instrumentation macro is missing target
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,91 @@
+//===-- SBAddressRangeList.cpp
===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/API/SBAddressRangeList.h"
+#include "Utils.h"
+#include "lldb/API/SBAddressRange.h"
+#include "lldb/API/SBStream.h"
+#include "lldb/Core/AddressRangeListImpl.h"
+#include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Stream.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+SBAddressRangeList::SBAddressRangeList()
+: m_opaque_up(new AddressRangeListImpl()) {
+ LLDB_INSTRUMENT_VA(this);
+}
+
+SBAddressRangeList::SBAddressRangeList(const SBAddressRangeList &rhs)
+: m_opaque_up(new AddressRangeListImpl(*rhs.m_opaque_up)) {
bulbazord wrote:
Same here
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,11 @@
+%extend lldb::SBAddressRange {
+#ifdef SWIGPYTHON
+%pythoncode%{
+ def __repr__(self):
+import lldb
mbucko wrote:
it's not, produces the following error:
`NameError: name 'lldb' is not defined`
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,28 @@
+%extend lldb::SBAddressRangeList {
+#ifdef SWIGPYTHON
+%pythoncode%{
+def __len__(self):
+ '''Return the number of address ranges in a lldb.SBAddressRangeList
object.'''
+ return self.GetSize()
+
+def __iter__(self):
+ '''Iterate over all the address ranges in a lldb.SBAddressRangeList
object.'''
+ return lldb_iter(self, 'GetSize', 'GetAddressRangeAtIndex')
+
+def __getitem__(self, idx):
+ '''Get the address range at a given index in an lldb.SBAddressRangeList
object.'''
+ if type(idx) == int:
+if idx >= self.GetSize():
+ raise IndexError("list index out of range")
+return self.GetAddressRangeAtIndex(idx)
+ else:
+print("error: unsupported idx type: %s" % type(key))
+return None
+def __repr__(self):
+ import lldb
mbucko wrote:
same as above
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,28 @@
+%extend lldb::SBAddressRangeList {
+#ifdef SWIGPYTHON
+%pythoncode%{
+def __len__(self):
+ '''Return the number of address ranges in a lldb.SBAddressRangeList
object.'''
+ return self.GetSize()
+
+def __iter__(self):
+ '''Iterate over all the address ranges in a lldb.SBAddressRangeList
object.'''
+ return lldb_iter(self, 'GetSize', 'GetAddressRangeAtIndex')
+
+def __getitem__(self, idx):
+ '''Get the address range at a given index in an lldb.SBAddressRangeList
object.'''
+ if type(idx) == int:
+if idx >= self.GetSize():
+ raise IndexError("list index out of range")
+return self.GetAddressRangeAtIndex(idx)
+ else:
+print("error: unsupported idx type: %s" % type(key))
+return None
mbucko wrote:
I'm following the convention from other Extensions files
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
https://github.com/Endilll created
https://github.com/llvm/llvm-project/pull/92953
This patch remove 36 checks for compiler flags that are done via invoking the
compiler across LLVM, Clang, and LLDB. It's was made possible by raising the
bar for supported compilers that has been happening over the years since the
checks were added.
This is going to improve CMake configuration times. This topic was highlighted
in
https://discourse.llvm.org/t/cmake-compiler-flag-checks-are-really-slow-ideas-to-speed-them-up/78882.
>From 66e05ac24613435dbe774d49684d8ff9d119c4c3 Mon Sep 17 00:00:00 2001
From: Vlad Serebrennikov
Date: Tue, 21 May 2024 21:41:24 +0300
Subject: [PATCH] Remove some `try_compile` CMake checks for compiler flags
This patch remove 36 checks for compiler flags that are done via invoking the
compiler across LLVM, Clang, and LLDB. It's was made possible by raising the
bar for supported compilers that has been happening over the years since the
checks were added.
This is going to improve CMake configuration times. This topic was highlighted
in
https://discourse.llvm.org/t/cmake-compiler-flag-checks-are-really-slow-ideas-to-speed-them-up/78882.
---
clang/CMakeLists.txt | 5 +-
.../tests/functional/exec/CMakeLists.txt | 6 +-
lldb/cmake/modules/LLDBConfig.cmake | 20 +--
llvm/cmake/config-ix.cmake| 19 +--
llvm/cmake/modules/AddLLVM.cmake | 4 +-
llvm/cmake/modules/HandleLLVMOptions.cmake| 142 --
third-party/unittest/CMakeLists.txt | 4 +-
7 files changed, 80 insertions(+), 120 deletions(-)
diff --git a/clang/CMakeLists.txt b/clang/CMakeLists.txt
index c20ce47a12abb..a6bcb853a464c 100644
--- a/clang/CMakeLists.txt
+++ b/clang/CMakeLists.txt
@@ -349,10 +349,7 @@ if (LLVM_COMPILER_IS_GCC_COMPATIBLE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -pedantic -Wno-long-long")
endif ()
- check_cxx_compiler_flag("-Werror -Wnested-anon-types"
CXX_SUPPORTS_NO_NESTED_ANON_TYPES_FLAG)
- if( CXX_SUPPORTS_NO_NESTED_ANON_TYPES_FLAG )
-set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-nested-anon-types" )
- endif()
+ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-nested-anon-types" )
endif ()
# Determine HOST_LINK_VERSION on Darwin.
diff --git a/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
b/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
index 95c6fdb610e0f..cb6ebda183725 100644
--- a/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
+++ b/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
@@ -2,11 +2,7 @@ project(exec C)
cmake_minimum_required(VERSION 3.20.0)
-include(CheckCCompilerFlag)
-check_c_compiler_flag("-std=c99" C99_SUPPORTED)
-if (C99_SUPPORTED)
-set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -std=c99")
-endif()
+set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -std=c99")
include(CheckFunctionExists)
include(CheckSymbolExists)
diff --git a/lldb/cmake/modules/LLDBConfig.cmake
b/lldb/cmake/modules/LLDBConfig.cmake
index 3c6223b015bb1..6458f2e174643 100644
--- a/lldb/cmake/modules/LLDBConfig.cmake
+++ b/lldb/cmake/modules/LLDBConfig.cmake
@@ -187,24 +187,18 @@
include_directories("${CMAKE_CURRENT_BINARY_DIR}/../clang/include")
# form -W, and if supported, add the corresponding -Wno- option.
# Disable GCC warnings
-check_cxx_compiler_flag("-Wdeprecated-declarations"
CXX_SUPPORTS_DEPRECATED_DECLARATIONS)
-append_if(CXX_SUPPORTS_DEPRECATED_DECLARATIONS "-Wno-deprecated-declarations"
CMAKE_CXX_FLAGS)
-
-check_cxx_compiler_flag("-Wunknown-pragmas" CXX_SUPPORTS_UNKNOWN_PRAGMAS)
-append_if(CXX_SUPPORTS_UNKNOWN_PRAGMAS "-Wno-unknown-pragmas" CMAKE_CXX_FLAGS)
-
-check_cxx_compiler_flag("-Wstrict-aliasing" CXX_SUPPORTS_STRICT_ALIASING)
-append_if(CXX_SUPPORTS_STRICT_ALIASING "-Wno-strict-aliasing" CMAKE_CXX_FLAGS)
+append("-Wno-deprecated-declarations" CMAKE_CXX_FLAGS)
+append("-Wno-unknown-pragmas" CMAKE_CXX_FLAGS)
+append("-Wno-strict-aliasing" CMAKE_CXX_FLAGS)
check_cxx_compiler_flag("-Wstringop-truncation"
CXX_SUPPORTS_STRINGOP_TRUNCATION)
append_if(CXX_SUPPORTS_STRINGOP_TRUNCATION "-Wno-stringop-truncation"
CMAKE_CXX_FLAGS)
# Disable Clang warnings
-check_cxx_compiler_flag("-Wdeprecated-register"
CXX_SUPPORTS_DEPRECATED_REGISTER)
-append_if(CXX_SUPPORTS_DEPRECATED_REGISTER "-Wno-deprecated-register"
CMAKE_CXX_FLAGS)
-
-check_cxx_compiler_flag("-Wvla-extension" CXX_SUPPORTS_VLA_EXTENSION)
-append_if(CXX_SUPPORTS_VLA_EXTENSION "-Wno-vla-extension" CMAKE_CXX_FLAGS)
+if (CMAKE_CXX_COMPILER_ID MATCHES "Clang")
+ append("-Wno-deprecated-register" CMAKE_CXX_FLAGS)
+ append("-Wno-vla-extension" CMAKE_CXX_FLAGS)
+endif()
# Disable MSVC warnings
if( MSVC )
diff --git a/llvm/cmake/config-ix.cmake b/llvm/cmake/config-ix.cmake
index bf1b110245bb2..0900e1107076e 100644
--- a/llvm/cmake/config-ix.cmake
+++ b/llvm/cmake/config-ix.cmake
@@ -415,15 +415,14 @@ if( LLVM_ENABLE_PIC )
set(ENABLE_PIC 1)
else()
set(ENABLE_P
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,54 @@
+//===-- SBAddressRangeList.h *- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#ifndef LLDB_API_SBADDRESSRANGELIST_H
+#define LLDB_API_SBADDRESSRANGELIST_H
+
+#include
+
+#include "lldb/API/SBDefines.h"
+
+namespace lldb_private {
+class AddressRangeListImpl;
+}
+
+namespace lldb {
+
+class LLDB_API SBAddressRangeList {
+public:
+ SBAddressRangeList();
+
+ SBAddressRangeList(const lldb::SBAddressRangeList &rhs);
+
+ ~SBAddressRangeList();
+
+ const lldb::SBAddressRangeList &
+ operator=(const lldb::SBAddressRangeList &rhs);
+
+ uint32_t GetSize() const;
+
+ void Clear();
+
+ SBAddressRange GetAddressRangeAtIndex(uint64_t idx);
mbucko wrote:
https://github.com/llvm/llvm-project/pull/92014#discussion_r1599157212
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
llvmbot wrote:
@llvm/pr-subscribers-lldb
@llvm/pr-subscribers-clang
Author: Vlad Serebrennikov (Endilll)
Changes
This patch remove 36 checks for compiler flags that are done via invoking the
compiler across LLVM, Clang, and LLDB. It's was made possible by raising the
bar for supported compilers that has been happening over the years since the
checks were added.
This is going to improve CMake configuration times. This topic was highlighted
in
https://discourse.llvm.org/t/cmake-compiler-flag-checks-are-really-slow-ideas-to-speed-them-up/78882.
---
Full diff: https://github.com/llvm/llvm-project/pull/92953.diff
7 Files Affected:
- (modified) clang/CMakeLists.txt (+1-4)
- (modified) clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
(+1-5)
- (modified) lldb/cmake/modules/LLDBConfig.cmake (+7-13)
- (modified) llvm/cmake/config-ix.cmake (+8-11)
- (modified) llvm/cmake/modules/AddLLVM.cmake (+1-3)
- (modified) llvm/cmake/modules/HandleLLVMOptions.cmake (+61-81)
- (modified) third-party/unittest/CMakeLists.txt (+1-3)
``diff
diff --git a/clang/CMakeLists.txt b/clang/CMakeLists.txt
index c20ce47a12abb..a6bcb853a464c 100644
--- a/clang/CMakeLists.txt
+++ b/clang/CMakeLists.txt
@@ -349,10 +349,7 @@ if (LLVM_COMPILER_IS_GCC_COMPATIBLE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -pedantic -Wno-long-long")
endif ()
- check_cxx_compiler_flag("-Werror -Wnested-anon-types"
CXX_SUPPORTS_NO_NESTED_ANON_TYPES_FLAG)
- if( CXX_SUPPORTS_NO_NESTED_ANON_TYPES_FLAG )
-set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-nested-anon-types" )
- endif()
+ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-nested-anon-types" )
endif ()
# Determine HOST_LINK_VERSION on Darwin.
diff --git a/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
b/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
index 95c6fdb610e0f..cb6ebda183725 100644
--- a/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
+++ b/clang/tools/scan-build-py/tests/functional/exec/CMakeLists.txt
@@ -2,11 +2,7 @@ project(exec C)
cmake_minimum_required(VERSION 3.20.0)
-include(CheckCCompilerFlag)
-check_c_compiler_flag("-std=c99" C99_SUPPORTED)
-if (C99_SUPPORTED)
-set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -std=c99")
-endif()
+set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -std=c99")
include(CheckFunctionExists)
include(CheckSymbolExists)
diff --git a/lldb/cmake/modules/LLDBConfig.cmake
b/lldb/cmake/modules/LLDBConfig.cmake
index 3c6223b015bb1..6458f2e174643 100644
--- a/lldb/cmake/modules/LLDBConfig.cmake
+++ b/lldb/cmake/modules/LLDBConfig.cmake
@@ -187,24 +187,18 @@
include_directories("${CMAKE_CURRENT_BINARY_DIR}/../clang/include")
# form -W, and if supported, add the corresponding -Wno- option.
# Disable GCC warnings
-check_cxx_compiler_flag("-Wdeprecated-declarations"
CXX_SUPPORTS_DEPRECATED_DECLARATIONS)
-append_if(CXX_SUPPORTS_DEPRECATED_DECLARATIONS "-Wno-deprecated-declarations"
CMAKE_CXX_FLAGS)
-
-check_cxx_compiler_flag("-Wunknown-pragmas" CXX_SUPPORTS_UNKNOWN_PRAGMAS)
-append_if(CXX_SUPPORTS_UNKNOWN_PRAGMAS "-Wno-unknown-pragmas" CMAKE_CXX_FLAGS)
-
-check_cxx_compiler_flag("-Wstrict-aliasing" CXX_SUPPORTS_STRICT_ALIASING)
-append_if(CXX_SUPPORTS_STRICT_ALIASING "-Wno-strict-aliasing" CMAKE_CXX_FLAGS)
+append("-Wno-deprecated-declarations" CMAKE_CXX_FLAGS)
+append("-Wno-unknown-pragmas" CMAKE_CXX_FLAGS)
+append("-Wno-strict-aliasing" CMAKE_CXX_FLAGS)
check_cxx_compiler_flag("-Wstringop-truncation"
CXX_SUPPORTS_STRINGOP_TRUNCATION)
append_if(CXX_SUPPORTS_STRINGOP_TRUNCATION "-Wno-stringop-truncation"
CMAKE_CXX_FLAGS)
# Disable Clang warnings
-check_cxx_compiler_flag("-Wdeprecated-register"
CXX_SUPPORTS_DEPRECATED_REGISTER)
-append_if(CXX_SUPPORTS_DEPRECATED_REGISTER "-Wno-deprecated-register"
CMAKE_CXX_FLAGS)
-
-check_cxx_compiler_flag("-Wvla-extension" CXX_SUPPORTS_VLA_EXTENSION)
-append_if(CXX_SUPPORTS_VLA_EXTENSION "-Wno-vla-extension" CMAKE_CXX_FLAGS)
+if (CMAKE_CXX_COMPILER_ID MATCHES "Clang")
+ append("-Wno-deprecated-register" CMAKE_CXX_FLAGS)
+ append("-Wno-vla-extension" CMAKE_CXX_FLAGS)
+endif()
# Disable MSVC warnings
if( MSVC )
diff --git a/llvm/cmake/config-ix.cmake b/llvm/cmake/config-ix.cmake
index bf1b110245bb2..0900e1107076e 100644
--- a/llvm/cmake/config-ix.cmake
+++ b/llvm/cmake/config-ix.cmake
@@ -415,15 +415,14 @@ if( LLVM_ENABLE_PIC )
set(ENABLE_PIC 1)
else()
set(ENABLE_PIC 0)
- check_cxx_compiler_flag("-fno-pie" SUPPORTS_NO_PIE_FLAG)
- if(SUPPORTS_NO_PIE_FLAG)
-set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -fno-pie")
- endif()
+ set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -fno-pie")
endif()
-check_cxx_compiler_flag("-Wvariadic-macros" SUPPORTS_VARIADIC_MACROS_FLAG)
-check_cxx_compiler_flag("-Wgnu-zero-variadic-macro-arguments"
-SUPPORTS_GNU_ZERO_VARIADIC_MACRO_ARGUMENTS_FLAG)
+if (CMAKE_CXX_COMPILER_ID MATCHES "Clang")
+ set(SUPPORTS_GNU_ZERO_VARIADIC_MACRO_ARGUMENTS_FLAG 1)
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
Endilll wrote: On my setup, this patch improves CMake configuration times (from a clean state) from 51 seconds down to 46 seconds (average of 3 measurements). My setup: ancient x86 hardware, Debian Sid, nightly Clang, build directory on a RAM disk. CMake invocation: `cmake -DLLVM_ENABLE_PROJECTS="clang;lldb" -DLLVM_ENABLE_RUNTIMES="libunwind;libcxx;libcxxabi" -DCMAKE_BUILD_TYPE=Debug -DLLVM_OPTIMIZED_TABLEGEN=ON -DLLVM_TARGETS_TO_BUILD="X86" -DLLVM_ENABLE_DOXYGEN=ON -DLLVM_ENABLE_LIBCXX=ON -DBUILD_SHARED_LIBS=ON -DLLDB_ENABLE_PYTHON=ON ~/endill/llvm-project/llvm`. https://github.com/llvm/llvm-project/pull/92953 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,91 @@
+//===-- SBAddressRangeList.cpp
===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/API/SBAddressRangeList.h"
+#include "Utils.h"
+#include "lldb/API/SBAddressRange.h"
+#include "lldb/API/SBStream.h"
+#include "lldb/Core/AddressRangeListImpl.h"
+#include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Stream.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+SBAddressRangeList::SBAddressRangeList()
+: m_opaque_up(new AddressRangeListImpl()) {
+ LLDB_INSTRUMENT_VA(this);
+}
+
+SBAddressRangeList::SBAddressRangeList(const SBAddressRangeList &rhs)
+: m_opaque_up(new AddressRangeListImpl(*rhs.m_opaque_up)) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+}
+
+SBAddressRangeList::~SBAddressRangeList() = default;
+
+const SBAddressRangeList &
+SBAddressRangeList::operator=(const SBAddressRangeList &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ if (this != &rhs)
+*m_opaque_up = *rhs.m_opaque_up;
+ return *this;
+}
+
+uint32_t SBAddressRangeList::GetSize() const {
+ LLDB_INSTRUMENT_VA(this);
+
+ return m_opaque_up->GetSize();
+}
+
+SBAddressRange SBAddressRangeList::GetAddressRangeAtIndex(uint64_t idx) {
+ LLDB_INSTRUMENT_VA(this, idx);
+
+ SBAddressRange sb_addr_range;
+ (*sb_addr_range.m_opaque_up) = m_opaque_up->GetAddressRangeAtIndex(idx);
+ return sb_addr_range;
+}
+
+void SBAddressRangeList::Clear() {
+ LLDB_INSTRUMENT_VA(this);
+
+ m_opaque_up->Clear();
+}
+
+void SBAddressRangeList::Append(const SBAddressRange &sb_addr_range) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range);
+
+ m_opaque_up->Append(*sb_addr_range.m_opaque_up);
+}
+
+void SBAddressRangeList::Append(const SBAddressRangeList &sb_addr_range_list) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range_list);
+
+ m_opaque_up->Append(*sb_addr_range_list.m_opaque_up);
+}
+
+bool SBAddressRangeList::GetDescription(SBStream &description,
+SBTarget *target) {
+ LLDB_INSTRUMENT_VA(this, description);
+
+ const uint32_t num_ranges = GetSize();
+ bool is_first = true;
+ Stream &stream = description.ref();
+ stream << "[";
+ for (uint32_t i = 0; i < num_ranges; ++i) {
+if (is_first) {
+ is_first = false;
+} else {
+ stream.Printf(", ");
+}
mbucko wrote:
This is actually correct, it prints out:
`[a.out[0x115f-0x1167), a.out[0x1134-0x1144), a.out[0x1144-0x1164)]`
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
Endilll wrote: Here are CE links with the set of minimum required compilers that should cover almost all the changes I made: https://godbolt.org/z/j98xzhrGa https://godbolt.org/z/errv4WhfP https://godbolt.org/z/vnoh8YqEP Windows-targeted flags were tested both on MSVC on CE, and on clang-cl 5.0 locally. https://github.com/llvm/llvm-project/pull/92953 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
https://github.com/MaskRay approved this pull request. Great 👍 https://github.com/llvm/llvm-project/pull/92953 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
https://github.com/mbucko updated
https://github.com/llvm/llvm-project/pull/92014
>From bcca8420736283bf564cf86e1442f023ba63b9e9 Mon Sep 17 00:00:00 2001
From: Miro Bucko
Date: Fri, 10 May 2024 12:42:03 -0700
Subject: [PATCH] Add AddressRange to SB API
Summary:
This adds new SB API calls and classes to allow a user of the SB API to
obtain an address range from SBFunction and SBBlock.
Test Plan:
llvm-lit -sv
llvm-project/lldb/test/API/python_api/address_range/TestAddressRange.py
Reviewers: clayborg
Subscribers: lldb-commits
Tasks:
Tags:
---
lldb/bindings/headers.swig| 2 +
.../interface/SBAddressRangeDocstrings.i | 3 +
.../interface/SBAddressRangeExtensions.i | 11 +
.../interface/SBAddressRangeListDocstrings.i | 3 +
.../interface/SBAddressRangeListExtensions.i | 28 ++
lldb/bindings/interfaces.swig | 6 +
lldb/include/lldb/API/LLDB.h | 2 +
lldb/include/lldb/API/SBAddress.h | 1 +
lldb/include/lldb/API/SBAddressRange.h| 65 +
lldb/include/lldb/API/SBAddressRangeList.h| 54
lldb/include/lldb/API/SBBlock.h | 4 +
lldb/include/lldb/API/SBDefines.h | 2 +
lldb/include/lldb/API/SBFunction.h| 3 +
lldb/include/lldb/API/SBStream.h | 2 +
lldb/include/lldb/API/SBTarget.h | 1 +
lldb/include/lldb/Core/AddressRange.h | 14 +
lldb/include/lldb/Core/AddressRangeListImpl.h | 51
lldb/include/lldb/Symbol/Block.h | 2 +
lldb/include/lldb/lldb-forward.h | 3 +
lldb/source/API/CMakeLists.txt| 2 +
lldb/source/API/SBAddressRange.cpp| 103
lldb/source/API/SBAddressRangeList.cpp| 93 +++
lldb/source/API/SBBlock.cpp | 10 +
lldb/source/API/SBFunction.cpp| 14 +
lldb/source/Core/AddressRange.cpp | 43
lldb/source/Core/AddressRangeListImpl.cpp | 50
lldb/source/Core/CMakeLists.txt | 1 +
lldb/source/Symbol/Block.cpp | 16 ++
.../API/python_api/address_range/Makefile | 3 +
.../address_range/TestAddressRange.py | 240 ++
.../API/python_api/address_range/main.cpp | 8 +
31 files changed, 840 insertions(+)
create mode 100644 lldb/bindings/interface/SBAddressRangeDocstrings.i
create mode 100644 lldb/bindings/interface/SBAddressRangeExtensions.i
create mode 100644 lldb/bindings/interface/SBAddressRangeListDocstrings.i
create mode 100644 lldb/bindings/interface/SBAddressRangeListExtensions.i
create mode 100644 lldb/include/lldb/API/SBAddressRange.h
create mode 100644 lldb/include/lldb/API/SBAddressRangeList.h
create mode 100644 lldb/include/lldb/Core/AddressRangeListImpl.h
create mode 100644 lldb/source/API/SBAddressRange.cpp
create mode 100644 lldb/source/API/SBAddressRangeList.cpp
create mode 100644 lldb/source/Core/AddressRangeListImpl.cpp
create mode 100644 lldb/test/API/python_api/address_range/Makefile
create mode 100644 lldb/test/API/python_api/address_range/TestAddressRange.py
create mode 100644 lldb/test/API/python_api/address_range/main.cpp
diff --git a/lldb/bindings/headers.swig b/lldb/bindings/headers.swig
index ffdc3c31ec883..c91504604b6ac 100644
--- a/lldb/bindings/headers.swig
+++ b/lldb/bindings/headers.swig
@@ -8,6 +8,8 @@
%{
#include "lldb/lldb-public.h"
#include "lldb/API/SBAddress.h"
+#include "lldb/API/SBAddressRange.h"
+#include "lldb/API/SBAddressRangeList.h"
#include "lldb/API/SBAttachInfo.h"
#include "lldb/API/SBBlock.h"
#include "lldb/API/SBBreakpoint.h"
diff --git a/lldb/bindings/interface/SBAddressRangeDocstrings.i
b/lldb/bindings/interface/SBAddressRangeDocstrings.i
new file mode 100644
index 0..650195704d73e
--- /dev/null
+++ b/lldb/bindings/interface/SBAddressRangeDocstrings.i
@@ -0,0 +1,3 @@
+%feature("docstring",
+"API clients can get address range information."
+) lldb::SBAddressRange;
diff --git a/lldb/bindings/interface/SBAddressRangeExtensions.i
b/lldb/bindings/interface/SBAddressRangeExtensions.i
new file mode 100644
index 0..9ff8e78080eb2
--- /dev/null
+++ b/lldb/bindings/interface/SBAddressRangeExtensions.i
@@ -0,0 +1,11 @@
+%extend lldb::SBAddressRange {
+#ifdef SWIGPYTHON
+%pythoncode%{
+ def __repr__(self):
+import lldb
+stream = lldb.SBStream()
+self.GetDescription(stream, lldb.target)
+return stream.GetData()
+%}
+#endif
+}
diff --git a/lldb/bindings/interface/SBAddressRangeListDocstrings.i
b/lldb/bindings/interface/SBAddressRangeListDocstrings.i
new file mode 100644
index 0..e4b96b9ca5931
--- /dev/null
+++ b/lldb/bindings/interface/SBAddressRangeListDocstrings.i
@@ -0,0 +1,3 @@
+%feature("docstring",
+"Represents a list of :py:class:`SBAddressRange`."
+) lldb::SBAddressRangeList;
diff --git a/lldb/bindings/interface/SBAddressRangeListExtensions.i
b/lldb/bindings
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,91 @@
+//===-- SBAddressRangeList.cpp
===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#include "lldb/API/SBAddressRangeList.h"
+#include "Utils.h"
+#include "lldb/API/SBAddressRange.h"
+#include "lldb/API/SBStream.h"
+#include "lldb/Core/AddressRangeListImpl.h"
+#include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Stream.h"
+
+using namespace lldb;
+using namespace lldb_private;
+
+SBAddressRangeList::SBAddressRangeList()
+: m_opaque_up(new AddressRangeListImpl()) {
+ LLDB_INSTRUMENT_VA(this);
+}
+
+SBAddressRangeList::SBAddressRangeList(const SBAddressRangeList &rhs)
+: m_opaque_up(new AddressRangeListImpl(*rhs.m_opaque_up)) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+}
+
+SBAddressRangeList::~SBAddressRangeList() = default;
+
+const SBAddressRangeList &
+SBAddressRangeList::operator=(const SBAddressRangeList &rhs) {
+ LLDB_INSTRUMENT_VA(this, rhs);
+
+ if (this != &rhs)
+*m_opaque_up = *rhs.m_opaque_up;
+ return *this;
+}
+
+uint32_t SBAddressRangeList::GetSize() const {
+ LLDB_INSTRUMENT_VA(this);
+
+ return m_opaque_up->GetSize();
+}
+
+SBAddressRange SBAddressRangeList::GetAddressRangeAtIndex(uint64_t idx) {
+ LLDB_INSTRUMENT_VA(this, idx);
+
+ SBAddressRange sb_addr_range;
+ (*sb_addr_range.m_opaque_up) = m_opaque_up->GetAddressRangeAtIndex(idx);
+ return sb_addr_range;
+}
+
+void SBAddressRangeList::Clear() {
+ LLDB_INSTRUMENT_VA(this);
+
+ m_opaque_up->Clear();
+}
+
+void SBAddressRangeList::Append(const SBAddressRange &sb_addr_range) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range);
+
+ m_opaque_up->Append(*sb_addr_range.m_opaque_up);
+}
+
+void SBAddressRangeList::Append(const SBAddressRangeList &sb_addr_range_list) {
+ LLDB_INSTRUMENT_VA(this, sb_addr_range_list);
+
+ m_opaque_up->Append(*sb_addr_range_list.m_opaque_up);
+}
+
+bool SBAddressRangeList::GetDescription(SBStream &description,
+SBTarget *target) {
+ LLDB_INSTRUMENT_VA(this, description);
+
+ const uint32_t num_ranges = GetSize();
+ bool is_first = true;
+ Stream &stream = description.ref();
+ stream << "[";
+ for (uint32_t i = 0; i < num_ranges; ++i) {
+if (is_first) {
+ is_first = false;
+} else {
+ stream.Printf(", ");
+}
bulbazord wrote:
Oh, yes, I see. That was definitely my mistake, sorry for the noise.
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Add AddressRange to SB API (PR #92014)
@@ -0,0 +1,65 @@
+//===-- SBAddressRange.h *- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+
+#ifndef LLDB_API_SBADDRESSRANGE_H
+#define LLDB_API_SBADDRESSRANGE_H
+
+#include "lldb/API/SBDefines.h"
+
+namespace lldb {
+
+class LLDB_API SBAddressRange {
+public:
+ SBAddressRange();
+
+ SBAddressRange(const lldb::SBAddressRange &rhs);
+
+ SBAddressRange(lldb::SBAddress addr, lldb::addr_t byte_size);
+
+ ~SBAddressRange();
+
+ const lldb::SBAddressRange &operator=(const lldb::SBAddressRange &rhs);
+
+ void Clear();
+
+ /// Check the address range refers to a valid base address and has a byte
+ /// size greater than zero.
+ ///
+ /// \return
+ /// True if the address range is valid, false otherwise.
+ bool IsValid() const;
+
+ /// Get the base address of the range.
+ ///
+ /// \return
+ /// Base address object.
+ lldb::SBAddress GetBaseAddress() const;
+
+ /// Get the byte size of this range.
+ ///
+ /// \return
+ /// The size in bytes of this address range.
+ lldb::addr_t GetByteSize() const;
+
+ bool operator==(const SBAddressRange &rhs);
+
+ bool operator!=(const SBAddressRange &rhs);
+
+ bool GetDescription(lldb::SBStream &description, SBTarget *target);
mbucko wrote:
Target might be null though
https://github.com/llvm/llvm-project/pull/92014
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] A few updates around "transcript" (PR #92843)
@@ -362,6 +363,36 @@ llvm::json::Value DebuggerStats::ReportStatistics(
global_stats.try_emplace("modules", std::move(json_modules));
global_stats.try_emplace("memory", std::move(json_memory));
global_stats.try_emplace("commands", std::move(cmd_stats));
+
+// When transcript is available, add it to the to-be-returned statistics.
+//
+// NOTE:
+// When the statistics is polled by an LLDB command:
+// - The transcript in the returned statistics *will NOT* contain the
+// returned statistics itself.
+// - The returned statistics *will* be written to the internal transcript
+// buffer as the output of the said LLDB command. It *will* appear in
+// the next statistcs or transcript poll.
+//
+// For example, let's say the following commands are run in order:
+// - "version"
+// - "statistics dump" <- call it "A"
+// - "statistics dump" <- call it "B"
+// The output of "A" will contain the transcript of "version" and
+// "statistics dump" (A), with the latter having empty output. The output
+// of B will contain the trascnript of "version", "statistics dump" (A),
+// "statistics dump" (B), with A's output populated and B's output empty.
royitaqi wrote:
I feel the quoted behavior is a bit difficult to understand/reason.
--
A (cleaner) alternative behavior might to completely remove the output of
"statistics dump" commands in the "transcript" in the "statistics dump":
```
(lldb) statistics dump
{
...,
"transcript" : [
{
"command": "version",
"output": "Meta LLDB version ...",
...
},
{
"command": "statistics dump",
"output": "(wiped empty)",
},
]
}
```
--
@clayborg Which behavior do you think is better?
https://github.com/llvm/llvm-project/pull/92843
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] A few updates around "transcript" (PR #92843)
https://github.com/royitaqi edited https://github.com/llvm/llvm-project/pull/92843 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] A few updates around "transcript" (PR #92843)
https://github.com/royitaqi edited https://github.com/llvm/llvm-project/pull/92843 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] A few updates around "transcript" (PR #92843)
https://github.com/royitaqi edited https://github.com/llvm/llvm-project/pull/92843 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
https://github.com/cor3ntin approved this pull request. https://github.com/llvm/llvm-project/pull/92953 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
https://github.com/aeubanks approved this pull request. awesome, thanks! https://github.com/llvm/llvm-project/pull/92953 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (PR #92503)
jasonmolenda wrote: btw @labath I was thinking about "sigtramp routines" and fault / trap / interrupt handlers in general, and how lldb has this list of function names that it treats as fault handlers (`target.trap-handler-names`). And in the unwinder we have the same idea of "the frame above a designed sigtramp/fault/trap/interrupt function can retrieve all registers". But what if the unwinder had a method saying "Do you have an unwind rule to give me a value for register n, without iterating down the stack?" e.g. frame 5 is interrupted and frame 4 is sigtramp, with full eh_frame. From frame 5, I can say "does frame 4 have an unwind rule to provide x0, without iterating down to frame 3, etc." This also means if we have a sigtramp which is missing its eh_frame, we won't apply our "all registers available" rules to the frame above. An interesting case for a return-address target like aarch64 where normally when we ask for a caller frame's pc, we fetch the link register. But above a fault/trap/interrupt frame, we can retrieve both the pc and the link register and they are different values. Just something I started kicking around in my head, I don't have concrete plans to implement an overhaul like this but the more I think about it, the more I like it. https://github.com/llvm/llvm-project/pull/92503 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (PR #92503)
jasonmolenda wrote: We should be able to work correctly with a trap handler that has full eh_frame without knowing the function name. And we shouldn't treat a sigtramp missing eh_frame as having all registers. https://github.com/llvm/llvm-project/pull/92503 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [lldb] Fixed the TestDebuggerAPI test on x86_64 Windows host (PR #90580)
slydiman wrote: We are almost done with configuring the buildbot for cross lldb (Windows x86_64 host and Linux Aarch64 target). We are trying to get it green with minimal local patches. https://github.com/llvm/llvm-project/pull/90580 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [lldb] [llvm] Remove some `try_compile` CMake checks for compiler flags (PR #92953)
https://github.com/JDevlieghere approved this pull request. https://github.com/llvm/llvm-project/pull/92953 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (PR #92503)
jasonmolenda wrote: (with the caveat that a register location of IsSame for a volatile aka non-callee-spilled register would be treated as "did not have a location") https://github.com/llvm/llvm-project/pull/92503 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Change GetChildCompilerTypeAtIndex to return Expected (NFC) (PR #92979)
https://github.com/adrian-prantl created
https://github.com/llvm/llvm-project/pull/92979
This change is a general improvement of the internal API. My motivation is to
use this in the Swift typesystem plugin.
>From 8a6202f8ae92a62ce577271ec580379f1424f9e3 Mon Sep 17 00:00:00 2001
From: Adrian Prantl
Date: Tue, 21 May 2024 17:41:31 -0700
Subject: [PATCH] Change GetChildCompilerTypeAtIndex to return Expected (NFC)
---
lldb/include/lldb/Symbol/CompilerType.h | 2 +-
lldb/include/lldb/Symbol/TypeSystem.h | 2 +-
lldb/source/Core/ValueObject.cpp | 31 ++-
.../Core/ValueObjectConstResultImpl.cpp | 12 +--
.../Plugins/ABI/PowerPC/ABISysV_ppc64.cpp | 3 +-
.../Language/CPlusPlus/BlockPointer.cpp | 18 ++-
.../Plugins/Language/CPlusPlus/LibCxxMap.cpp | 6 ++--
.../TypeSystem/Clang/TypeSystemClang.cpp | 18 ++-
.../TypeSystem/Clang/TypeSystemClang.h| 2 +-
lldb/source/Symbol/CompilerType.cpp | 2 +-
10 files changed, 63 insertions(+), 33 deletions(-)
diff --git a/lldb/include/lldb/Symbol/CompilerType.h
b/lldb/include/lldb/Symbol/CompilerType.h
index 28c723abf2794..70dacdcb7986f 100644
--- a/lldb/include/lldb/Symbol/CompilerType.h
+++ b/lldb/include/lldb/Symbol/CompilerType.h
@@ -436,7 +436,7 @@ class CompilerType {
uint32_t *bitfield_bit_size_ptr = nullptr,
bool *is_bitfield_ptr = nullptr) const;
- CompilerType GetChildCompilerTypeAtIndex(
+ llvm::Expected GetChildCompilerTypeAtIndex(
ExecutionContext *exe_ctx, size_t idx, bool transparent_pointers,
bool omit_empty_base_classes, bool ignore_array_bounds,
std::string &child_name, uint32_t &child_byte_size,
diff --git a/lldb/include/lldb/Symbol/TypeSystem.h
b/lldb/include/lldb/Symbol/TypeSystem.h
index 7bcb8d69387a0..b4025c173a186 100644
--- a/lldb/include/lldb/Symbol/TypeSystem.h
+++ b/lldb/include/lldb/Symbol/TypeSystem.h
@@ -359,7 +359,7 @@ class TypeSystem : public PluginInterface,
return CompilerDecl();
}
- virtual CompilerType GetChildCompilerTypeAtIndex(
+ virtual llvm::Expected GetChildCompilerTypeAtIndex(
lldb::opaque_compiler_type_t type, ExecutionContext *exe_ctx, size_t idx,
bool transparent_pointers, bool omit_empty_base_classes,
bool ignore_array_bounds, std::string &child_name,
diff --git a/lldb/source/Core/ValueObject.cpp b/lldb/source/Core/ValueObject.cpp
index f39bd07a25536..1443d9dfc3280 100644
--- a/lldb/source/Core/ValueObject.cpp
+++ b/lldb/source/Core/ValueObject.cpp
@@ -505,15 +505,23 @@ ValueObject *ValueObject::CreateChildAtIndex(size_t idx,
uint64_t language_flags = 0;
const bool transparent_pointers = !synthetic_array_member;
- CompilerType child_compiler_type;
ExecutionContext exe_ctx(GetExecutionContextRef());
- child_compiler_type = GetCompilerType().GetChildCompilerTypeAtIndex(
- &exe_ctx, idx, transparent_pointers, omit_empty_base_classes,
- ignore_array_bounds, child_name_str, child_byte_size, child_byte_offset,
- child_bitfield_bit_size, child_bitfield_bit_offset, child_is_base_class,
- child_is_deref_of_parent, this, language_flags);
+ auto child_compiler_type_or_err =
+ GetCompilerType().GetChildCompilerTypeAtIndex(
+ &exe_ctx, idx, transparent_pointers, omit_empty_base_classes,
+ ignore_array_bounds, child_name_str, child_byte_size,
+ child_byte_offset, child_bitfield_bit_size,
child_bitfield_bit_offset,
+ child_is_base_class, child_is_deref_of_parent, this, language_flags);
+ CompilerType child_compiler_type;
+ if (!child_compiler_type_or_err)
+LLDB_LOG_ERROR(GetLog(LLDBLog::Types),
+ child_compiler_type_or_err.takeError(),
+ "could not find child: {0}");
+ else
+child_compiler_type = *child_compiler_type_or_err;
+
if (child_compiler_type) {
if (synthetic_index)
child_byte_offset += child_byte_size * synthetic_index;
@@ -2624,16 +2632,23 @@ ValueObjectSP ValueObject::Dereference(Status &error) {
bool child_is_deref_of_parent = false;
const bool transparent_pointers = false;
CompilerType compiler_type = GetCompilerType();
-CompilerType child_compiler_type;
uint64_t language_flags = 0;
ExecutionContext exe_ctx(GetExecutionContextRef());
-child_compiler_type = compiler_type.GetChildCompilerTypeAtIndex(
+CompilerType child_compiler_type;
+auto child_compiler_type_or_err =
compiler_type.GetChildCompilerTypeAtIndex(
&exe_ctx, 0, transparent_pointers, omit_empty_base_classes,
ignore_array_bounds, child_name_str, child_byte_size,
child_byte_offset,
child_bitfield_bit_size, child_bitfield_bit_offset,
child_is_base_class,
child_is_deref_of_parent, this, language_flags);
+if (!child_compiler_type_or_err)
+ LLDB_LOG_ERROR(GetLog(LLDBLog::Types),
+
[Lldb-commits] [lldb] Change GetChildCompilerTypeAtIndex to return Expected (NFC) (PR #92979)
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Adrian Prantl (adrian-prantl)
Changes
This change is a general improvement of the internal API. My motivation is to
use this in the Swift typesystem plugin.
---
Full diff: https://github.com/llvm/llvm-project/pull/92979.diff
10 Files Affected:
- (modified) lldb/include/lldb/Symbol/CompilerType.h (+1-1)
- (modified) lldb/include/lldb/Symbol/TypeSystem.h (+1-1)
- (modified) lldb/source/Core/ValueObject.cpp (+23-8)
- (modified) lldb/source/Core/ValueObjectConstResultImpl.cpp (+10-2)
- (modified) lldb/source/Plugins/ABI/PowerPC/ABISysV_ppc64.cpp (+2-1)
- (modified) lldb/source/Plugins/Language/CPlusPlus/BlockPointer.cpp (+11-7)
- (modified) lldb/source/Plugins/Language/CPlusPlus/LibCxxMap.cpp (+3-3)
- (modified) lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp (+10-8)
- (modified) lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.h (+1-1)
- (modified) lldb/source/Symbol/CompilerType.cpp (+1-1)
``diff
diff --git a/lldb/include/lldb/Symbol/CompilerType.h
b/lldb/include/lldb/Symbol/CompilerType.h
index 28c723abf2794..70dacdcb7986f 100644
--- a/lldb/include/lldb/Symbol/CompilerType.h
+++ b/lldb/include/lldb/Symbol/CompilerType.h
@@ -436,7 +436,7 @@ class CompilerType {
uint32_t *bitfield_bit_size_ptr = nullptr,
bool *is_bitfield_ptr = nullptr) const;
- CompilerType GetChildCompilerTypeAtIndex(
+ llvm::Expected GetChildCompilerTypeAtIndex(
ExecutionContext *exe_ctx, size_t idx, bool transparent_pointers,
bool omit_empty_base_classes, bool ignore_array_bounds,
std::string &child_name, uint32_t &child_byte_size,
diff --git a/lldb/include/lldb/Symbol/TypeSystem.h
b/lldb/include/lldb/Symbol/TypeSystem.h
index 7bcb8d69387a0..b4025c173a186 100644
--- a/lldb/include/lldb/Symbol/TypeSystem.h
+++ b/lldb/include/lldb/Symbol/TypeSystem.h
@@ -359,7 +359,7 @@ class TypeSystem : public PluginInterface,
return CompilerDecl();
}
- virtual CompilerType GetChildCompilerTypeAtIndex(
+ virtual llvm::Expected GetChildCompilerTypeAtIndex(
lldb::opaque_compiler_type_t type, ExecutionContext *exe_ctx, size_t idx,
bool transparent_pointers, bool omit_empty_base_classes,
bool ignore_array_bounds, std::string &child_name,
diff --git a/lldb/source/Core/ValueObject.cpp b/lldb/source/Core/ValueObject.cpp
index f39bd07a25536..1443d9dfc3280 100644
--- a/lldb/source/Core/ValueObject.cpp
+++ b/lldb/source/Core/ValueObject.cpp
@@ -505,15 +505,23 @@ ValueObject *ValueObject::CreateChildAtIndex(size_t idx,
uint64_t language_flags = 0;
const bool transparent_pointers = !synthetic_array_member;
- CompilerType child_compiler_type;
ExecutionContext exe_ctx(GetExecutionContextRef());
- child_compiler_type = GetCompilerType().GetChildCompilerTypeAtIndex(
- &exe_ctx, idx, transparent_pointers, omit_empty_base_classes,
- ignore_array_bounds, child_name_str, child_byte_size, child_byte_offset,
- child_bitfield_bit_size, child_bitfield_bit_offset, child_is_base_class,
- child_is_deref_of_parent, this, language_flags);
+ auto child_compiler_type_or_err =
+ GetCompilerType().GetChildCompilerTypeAtIndex(
+ &exe_ctx, idx, transparent_pointers, omit_empty_base_classes,
+ ignore_array_bounds, child_name_str, child_byte_size,
+ child_byte_offset, child_bitfield_bit_size,
child_bitfield_bit_offset,
+ child_is_base_class, child_is_deref_of_parent, this, language_flags);
+ CompilerType child_compiler_type;
+ if (!child_compiler_type_or_err)
+LLDB_LOG_ERROR(GetLog(LLDBLog::Types),
+ child_compiler_type_or_err.takeError(),
+ "could not find child: {0}");
+ else
+child_compiler_type = *child_compiler_type_or_err;
+
if (child_compiler_type) {
if (synthetic_index)
child_byte_offset += child_byte_size * synthetic_index;
@@ -2624,16 +2632,23 @@ ValueObjectSP ValueObject::Dereference(Status &error) {
bool child_is_deref_of_parent = false;
const bool transparent_pointers = false;
CompilerType compiler_type = GetCompilerType();
-CompilerType child_compiler_type;
uint64_t language_flags = 0;
ExecutionContext exe_ctx(GetExecutionContextRef());
-child_compiler_type = compiler_type.GetChildCompilerTypeAtIndex(
+CompilerType child_compiler_type;
+auto child_compiler_type_or_err =
compiler_type.GetChildCompilerTypeAtIndex(
&exe_ctx, 0, transparent_pointers, omit_empty_base_classes,
ignore_array_bounds, child_name_str, child_byte_size,
child_byte_offset,
child_bitfield_bit_size, child_bitfield_bit_offset,
child_is_base_class,
child_is_deref_of_parent, this, language_flags);
+if (!child_compiler_type_or_err)
+ LLDB_LOG_ERROR(GetLog(LLDBLog::Types),
+ child_compiler_type_or_err.takeError(),
+ "coul
[Lldb-commits] [lldb] Change GetChildCompilerTypeAtIndex to return Expected (NFC) (PR #92979)
github-actions[bot] wrote: :warning: C/C++ code formatter, clang-format found issues in your code. :warning: You can test this locally with the following command: ``bash git-clang-format --diff 0bc710f7c19910817ccff254c43496602635bbc9 8a6202f8ae92a62ce577271ec580379f1424f9e3 -- lldb/include/lldb/Symbol/CompilerType.h lldb/include/lldb/Symbol/TypeSystem.h lldb/source/Core/ValueObject.cpp lldb/source/Core/ValueObjectConstResultImpl.cpp lldb/source/Plugins/ABI/PowerPC/ABISysV_ppc64.cpp lldb/source/Plugins/Language/CPlusPlus/BlockPointer.cpp lldb/source/Plugins/Language/CPlusPlus/LibCxxMap.cpp lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.h lldb/source/Symbol/CompilerType.cpp `` View the diff from clang-format here. ``diff diff --git a/lldb/source/Plugins/Language/CPlusPlus/LibCxxMap.cpp b/lldb/source/Plugins/Language/CPlusPlus/LibCxxMap.cpp index d4cc47b454..0929d49e55 100644 --- a/lldb/source/Plugins/Language/CPlusPlus/LibCxxMap.cpp +++ b/lldb/source/Plugins/Language/CPlusPlus/LibCxxMap.cpp @@ -295,12 +295,12 @@ void lldb_private::formatters::LibcxxStdMapSyntheticFrontEnd::GetValueOffset( bool child_is_base_class; bool child_is_deref_of_parent; uint64_t language_flags; -auto child_type = llvm::expectedToStdOptional(tree_node_type -.GetChildCompilerTypeAtIndex( -nullptr, 4, true, true, true, child_name, child_byte_size, -child_byte_offset, child_bitfield_bit_size, -child_bitfield_bit_offset, child_is_base_class, -child_is_deref_of_parent, nullptr, language_flags)); +auto child_type = +llvm::expectedToStdOptional(tree_node_type.GetChildCompilerTypeAtIndex( +nullptr, 4, true, true, true, child_name, child_byte_size, +child_byte_offset, child_bitfield_bit_size, +child_bitfield_bit_offset, child_is_base_class, +child_is_deref_of_parent, nullptr, language_flags)); if (child_type && child_type->IsValid()) m_skip_size = (uint32_t)child_byte_offset; } `` https://github.com/llvm/llvm-project/pull/92979 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: Add new APIs `AddDestroyCallback` and `RemoveDestroyCallback` (PR #89868)
royitaqi wrote: Hi @antmox, I'm sorry that it broke the lldb-aarch64-windows build. My bad. Your fix makes sense, though I do have the same comment as @labath had, i.e. to move that typedef outside platform-specific region. From the latest comments in that PR it seems it's already got moved. If that's the case then I think we are all good. My apologies that I didn't respond earlier - I was watching my email and the build bot (https://lab.llvm.org/buildbot/#/changes/134099), saw all green and I thought it's went well. **Questions for my learning** (I'm noob to all these): I realize that the build bot I quoted didn't run the windows build you mentioned (lldb-aarch64-windows). This tells me that there are multiple build bots that I need to watch in the future. **Is there an exhaustive list somewhere?** Email obviously failed to inform me about the build break (i.e. I searched for "lldb-aarch64-windows" in my email and nothing except your message). Thanks, Roy https://github.com/llvm/llvm-project/pull/89868 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] SBDebugger: define callback_token_t for win32 (PR #92870)
royitaqi wrote: Thanks to @antmox and @labath for fixing this. My apologies for breaking the build. https://github.com/llvm/llvm-project/pull/92870 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [clang] [libcxx] [lldb] [llvm] [mlir] [BOLT][BAT] Add entries for deleted basic blocks (PR #91906)
https://github.com/rafaelauler approved this pull request. Could you elaborate a bit better on why do we need a deleted block to be present in the table? My memory fails me, aren't we using the translation table just to map samples collected on the bolted binary? Where do the deleted blocks become a problem? Other than the motivation, the implementation itself looks good to me. https://github.com/llvm/llvm-project/pull/91906 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] [LLDB] Add more helper functions to ValueObject class. (PR #87197)
@@ -719,6 +776,10 @@ class ValueObject {
ClearUserVisibleData(eClearUserVisibleDataItemsSummary);
}
+ void SetDerefValobj(ValueObject *deref) { m_deref_valobj = deref; }
cmtice wrote:
I use the synthetic child providers way of getting the deref ValueObject to
get the value I use for my call to SetDerefValobj (when I'm converting a
ValueObject containing a SmartPointer into a ValueObject containing the pointer
the SmartPointer was referencing).
https://github.com/llvm/llvm-project/pull/87197
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [lldb] Reapply "[lldb/aarch64] Fix unwinding when signal interrupts a leaf f… (PR #92503)
jasonmolenda wrote: as cool as this idea is, I do worry that it will make the code less readable, where instead of saying "BehavesLikeFrameZero / HasAllRegisters", we now need to ask "can the frame below supply register x", I don't know. it's just something I have running around my head today. https://github.com/llvm/llvm-project/pull/92503 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
