[llvm] Modify dwarfdump verification to allow sub-category counts (PR #125062)
via llvm-commits
llvm-commits at lists.llvm.org
Thu Jan 30 06:23:23 PST 2025
https://github.com/youngdfb created https://github.com/llvm/llvm-project/pull/125062
It was discovered that BOLT had several distinct issues of missing debug information by various tags for debug names (119493 & 119023 as examples), but the verification of a DWARF with llvm-dwarfdump prior to those fixes only gave one 'missing name' category.
{"error-categories":{"Name Index DIE entry missing name":{"count":36355210}},"error-count":36355210}
To more easily leverage dwarf verification for debug health, the JSON output will be improved to allow having detailed counts by a sub-category when it makes sense.
For now, this is only implemented on the missing tags, but can be extended to more.
{"error-categories":{"Name Index DIE entry missing name":{"count":10,"details":{"DW_TAG_inlined_subroutine":1,"DW_TAG_label":1,"DW_TAG_namespace":2,"DW_TAG_subprogram":2,"DW_TAG_variable":4}}},"error-count":10}
This diff also modifies the tests created in pull request 124936 (not yet landed) to ensure the JSON switches. Ideally this lands after that but it did not correctly create a stack of pull requests.
>From 5d326487c7a2371e3e1ceca2072106e9c03ec26f Mon Sep 17 00:00:00 2001
From: David Young <davidayoung at meta.com>
Date: Wed, 29 Jan 2025 06:32:33 -0800
Subject: [PATCH 1/3] Add test for dwarf verification JSON output
Summary:
6244dfef5cd45f1395c66abbe061c6a7eb002676 LLVM commit added the ability
for llvm-dwarfdump to specify --verify-json and get a JSON aggregation of t
he DWARF errors. This diff improves the testing by ensuring we validate th
e expected JSON shape.
A follow up diff will modify the JSON and this ensures we can verify.
he DWARF errors. This diff improves the testing by ensuring we validate th
e expected JSON shape.
A follow up diff will modify the JSON and this ensures we can verify.
Test Plan:
ninja check-llvm-tools-llvm-dwarfdump
---
...g-names-verify--completeness-json-output.s | 172 ++++++++++++++++++
.../X86/debug-names-verify-completeness.s | 1 -
2 files changed, 172 insertions(+), 1 deletion(-)
create mode 100644 llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify--completeness-json-output.s
diff --git a/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify--completeness-json-output.s b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify--completeness-json-output.s
new file mode 100644
index 00000000000000..c4b7ffe1d2d8e7
--- /dev/null
+++ b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify--completeness-json-output.s
@@ -0,0 +1,172 @@
+# RUN: llvm-mc -triple x86_64-pc-linux %s -filetype=obj -o - | not llvm-dwarfdump -verify --verify-json=%t.json -
+# RUN: FileCheck %s --input-file %t.json
+
+# CHECK: {"error-categories":{"Name Index DIE entry missing name":{"count":10}},"error-count":10}
+# CHECK-NOT: error: Name Index @ 0x0: Entry for DIE @ {{.*}} (DW_TAG_variable) with name var_block_addr missing.
+
+ .section .debug_loc,"", at progbits
+.Ldebug_loc0:
+ .quad 0
+ .quad 1
+ .short .Lloc0_end-.Lloc0_start # Loc expr size
+.Lloc0_start:
+ .byte 3 # DW_OP_addr
+ .quad 0x47
+.Lloc0_end:
+ .quad 0
+ .quad 0
+
+ .section .debug_abbrev,"", at progbits
+ .byte 1 # Abbreviation Code
+ .byte 17 # DW_TAG_compile_unit
+ .byte 1 # DW_CHILDREN_yes
+ .byte 37 # DW_AT_producer
+ .byte 8 # DW_FORM_string
+ .byte 17 # DW_AT_low_pc
+ .byte 1 # DW_FORM_addr
+ .byte 18 # DW_AT_high_pc
+ .byte 6 # DW_FORM_data4
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 2 # Abbreviation Code
+ .byte 52 # DW_TAG_variable
+ .byte 0 # DW_CHILDREN_no
+ .byte 3 # DW_AT_name
+ .byte 8 # DW_FORM_string
+ .byte 2 # DW_AT_location
+ .byte 24 # DW_FORM_exprloc
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 3 # Abbreviation Code
+ .byte 46 # DW_TAG_subprogram
+ .byte 1 # DW_CHILDREN_yes
+ .byte 3 # DW_AT_name
+ .byte 8 # DW_FORM_string
+ .byte 110 # DW_AT_linkage_name
+ .byte 8 # DW_FORM_string
+ .byte 82 # DW_AT_entry_pc
+ .byte 1 # DW_FORM_addr
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 4 # Abbreviation Code
+ .byte 57 # DW_TAG_namespace
+ .byte 1 # DW_CHILDREN_yes
+ .byte 3 # DW_AT_name
+ .byte 8 # DW_FORM_string
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 5 # Abbreviation Code
+ .byte 52 # DW_TAG_variable
+ .byte 0 # DW_CHILDREN_no
+ .byte 3 # DW_AT_name
+ .byte 8 # DW_FORM_string
+ .byte 2 # DW_AT_location
+ .byte 23 # DW_FORM_sec_offset
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 6 # Abbreviation Code
+ .byte 57 # DW_TAG_namespace
+ .byte 1 # DW_CHILDREN_yes
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 7 # Abbreviation Code
+ .byte 29 # DW_TAG_inlined_subroutine
+ .byte 0 # DW_CHILDREN_no
+ .byte 3 # DW_AT_name
+ .byte 8 # DW_FORM_string
+ .byte 17 # DW_AT_low_pc
+ .byte 1 # DW_FORM_addr
+ .byte 18 # DW_AT_high_pc
+ .byte 1 # DW_FORM_addr
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 8 # Abbreviation Code
+ .byte 10 # DW_TAG_label
+ .byte 0 # DW_CHILDREN_no
+ .byte 3 # DW_AT_name
+ .byte 8 # DW_FORM_string
+ .byte 82 # DW_AT_entry_pc
+ .byte 1 # DW_FORM_addr
+ .byte 0 # EOM(1)
+ .byte 0 # EOM(2)
+
+ .byte 0 # EOM(3)
+ .section .debug_info,"", at progbits
+
+.Lcu_begin0:
+ .long .Lcu_end0-.Lcu_start0 # Length of Unit
+.Lcu_start0:
+ .short 4 # DWARF version number
+ .long .debug_abbrev # Offset Into Abbrev. Section
+ .byte 8 # Address Size (in bytes)
+ .byte 1 # Abbrev [1] DW_TAG_compile_unit
+ .asciz "hand-written DWARF" # DW_AT_producer
+ .quad 0x0 # DW_AT_low_pc
+ .long 0x100 # DW_AT_high_pc
+
+ .byte 4 # Abbrev [4] DW_TAG_namespace
+ .asciz "namesp" # DW_AT_name
+ .byte 2 # Abbrev [2] DW_TAG_variable
+ .asciz "var_block_addr" # DW_AT_name
+ .byte 9 # DW_AT_location
+ .byte 3 # DW_OP_addr
+ .quad 0x47
+ .byte 0 # End Of Children Mark
+
+ .byte 6 # Abbrev [6] DW_TAG_namespace
+ .byte 5 # Abbrev [5] DW_TAG_variable
+ .asciz "var_loc_addr" # DW_AT_name
+ .long .Ldebug_loc0 # DW_AT_location
+ .byte 0 # End Of Children Mark
+
+ .byte 2 # Abbrev [2] DW_TAG_variable
+ .asciz "var_loc_tls" # DW_AT_name
+ .byte 1 # DW_AT_location
+ .byte 0x9b # DW_OP_form_tls_address
+
+ .byte 2 # Abbrev [2] DW_TAG_variable
+ .asciz "var_loc_gnu_tls" # DW_AT_name
+ .byte 1 # DW_AT_location
+ .byte 0xe0 # DW_OP_GNU_push_tls_address
+
+ .byte 3 # Abbrev [3] DW_TAG_subprogram
+ .asciz "fun_name" # DW_AT_name
+ .asciz "_Z8fun_name" # DW_AT_linkage_name
+ .quad 0x47 # DW_AT_entry_pc
+ .byte 7 # Abbrev [7] DW_TAG_inlined_subroutine
+ .asciz "fun_inline" # DW_AT_name
+ .quad 0x48 # DW_AT_low_pc
+ .quad 0x49 # DW_AT_high_pc
+ .byte 8 # Abbrev [8] DW_TAG_label
+ .asciz "label" # DW_AT_name
+ .quad 0x4a # DW_AT_entry_pc
+ .byte 0 # End Of Children Mark
+
+ .byte 0 # End Of Children Mark
+.Lcu_end0:
+
+ .section .debug_names,"", at progbits
+ .long .Lnames_end0-.Lnames_start0 # Header: contribution length
+.Lnames_start0:
+ .short 5 # Header: version
+ .short 0 # Header: padding
+ .long 1 # Header: compilation unit count
+ .long 0 # Header: local type unit count
+ .long 0 # Header: foreign type unit count
+ .long 0 # Header: bucket count
+ .long 0 # Header: name count
+ .long .Lnames_abbrev_end0-.Lnames_abbrev_start0 # Header: abbreviation table size
+ .long 0 # Header: augmentation length
+ .long .Lcu_begin0 # Compilation unit 0
+.Lnames_abbrev_start0:
+ .byte 0 # End of abbrev list
+.Lnames_abbrev_end0:
+.Lnames_entries0:
+.Lnames_end0:
diff --git a/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s
index b16f8658f87eca..9886968fdab996 100644
--- a/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s
+++ b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s
@@ -177,4 +177,3 @@
.Lnames_abbrev_end0:
.Lnames_entries0:
.Lnames_end0:
-
>From 63113cf699762747ee56a064fdf75d26dde7b5cb Mon Sep 17 00:00:00 2001
From: David Young <davidayoung at meta.com>
Date: Wed, 29 Jan 2025 12:26:55 -0800
Subject: [PATCH 2/3] Add test for dwarf verification JSON output (2/2)
---
.../X86/debug-names-verify-completeness.s | 1 +
.../X86/dwarf-verify-good-json-output.s | 32 +++++++++++++++++++
2 files changed, 33 insertions(+)
create mode 100644 llvm/test/tools/llvm-dwarfdump/X86/dwarf-verify-good-json-output.s
diff --git a/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s
index 9886968fdab996..b16f8658f87eca 100644
--- a/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s
+++ b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness.s
@@ -177,3 +177,4 @@
.Lnames_abbrev_end0:
.Lnames_entries0:
.Lnames_end0:
+
diff --git a/llvm/test/tools/llvm-dwarfdump/X86/dwarf-verify-good-json-output.s b/llvm/test/tools/llvm-dwarfdump/X86/dwarf-verify-good-json-output.s
new file mode 100644
index 00000000000000..6f4b1cdedf83ab
--- /dev/null
+++ b/llvm/test/tools/llvm-dwarfdump/X86/dwarf-verify-good-json-output.s
@@ -0,0 +1,32 @@
+# RUN: llvm-mc %s -filetype obj -triple x86_64-apple-darwin -o - | llvm-dwarfdump -verify --verify-json=%t.json -
+# RUN: FileCheck %s --input-file %t.json
+
+# CHECK: {"error-categories":{},"error-count":0}
+
+# This test is meant to verify that the -verify option
+# in llvm-dwarfdump doesn't produce any .apple_names related
+# output when there's no such section in the object.
+# The test was manually modified to exclude the
+# .apple_names section from the apple_names_verify_num_atoms.s
+# test file in the same directory.
+
+ .section __TEXT,__text,regular,pure_instructions
+ .file 1 "basic.c"
+ .comm _i,4,2 ## @i
+ .comm _j,4,2 ## @j
+ .section __DWARF,__debug_str,regular,debug
+Linfo_string:
+ .asciz "Apple LLVM version 8.1.0 (clang-802.0.35)" ## string offset=0
+ .asciz "basic.c" ## string offset=42
+ .asciz "/Users/sgravani/Development/tests" ## string offset=50
+ .asciz "i" ## string offset=84
+ .asciz "int" ## string offset=86
+ .asciz "j" ## string offset=90
+
+ .section __DWARF,__debug_info,regular,debug
+Lsection_info:
+
+.subsections_via_symbols
+ .section __DWARF,__debug_line,regular,debug
+Lsection_line:
+Lline_table_start0:
>From ebf1cab2db2d8e26b014372147b3644f52a1ab4d Mon Sep 17 00:00:00 2001
From: David Young <davidayoung at meta.com>
Date: Thu, 30 Jan 2025 05:54:39 -0800
Subject: [PATCH 3/3] Modify dwarf verification JSON to include detailed counts
by sub-category
---
.../llvm/DebugInfo/DWARF/DWARFVerifier.h | 15 +++++-
llvm/lib/DebugInfo/DWARF/DWARFVerifier.cpp | 46 +++++++++++++++----
...g-names-verify-completeness-json-output.s} | 2 +-
3 files changed, 51 insertions(+), 12 deletions(-)
rename llvm/test/tools/llvm-dwarfdump/X86/{debug-names-verify--completeness-json-output.s => debug-names-verify-completeness-json-output.s} (98%)
diff --git a/llvm/include/llvm/DebugInfo/DWARF/DWARFVerifier.h b/llvm/include/llvm/DebugInfo/DWARF/DWARFVerifier.h
index 7b51bb63cd15ba..6e79619e156cb6 100644
--- a/llvm/include/llvm/DebugInfo/DWARF/DWARFVerifier.h
+++ b/llvm/include/llvm/DebugInfo/DWARF/DWARFVerifier.h
@@ -30,9 +30,15 @@ class DWARFDebugAbbrev;
class DataExtractor;
struct DWARFSection;
+struct AggregationData {
+ unsigned OverallCount;
+ std::map<std::string, unsigned> DetailedCounts;
+ AggregationData() = default;
+};
+
class OutputCategoryAggregator {
private:
- std::map<std::string, unsigned> Aggregation;
+ std::map<std::string, AggregationData> Aggregation;
bool IncludeDetail;
public:
@@ -40,8 +46,13 @@ class OutputCategoryAggregator {
: IncludeDetail(includeDetail) {}
void ShowDetail(bool showDetail) { IncludeDetail = showDetail; }
size_t GetNumCategories() const { return Aggregation.size(); }
- void Report(StringRef s, std::function<void()> detailCallback);
+ void Report(StringRef category, std::function<void()> detailCallback);
+ void Report(StringRef category, StringRef sub_category,
+ std::function<void()> detailCallback);
void EnumerateResults(std::function<void(StringRef, unsigned)> handleCounts);
+ void EnumerateDetailedResultsFor(
+ StringRef category,
+ std::function<void(StringRef, unsigned)> handleCounts);
};
/// A class that verifies DWARF debug information given a DWARF Context.
diff --git a/llvm/lib/DebugInfo/DWARF/DWARFVerifier.cpp b/llvm/lib/DebugInfo/DWARF/DWARFVerifier.cpp
index 8bf513538de7c7..362461114b95e6 100644
--- a/llvm/lib/DebugInfo/DWARF/DWARFVerifier.cpp
+++ b/llvm/lib/DebugInfo/DWARF/DWARFVerifier.cpp
@@ -1942,12 +1942,14 @@ unsigned DWARFVerifier::verifyNameIndexCompleteness(
if (none_of(NI.equal_range(Name), [&](const DWARFDebugNames::Entry &E) {
return E.getDIEUnitOffset() == DieUnitOffset;
})) {
- ErrorCategory.Report("Name Index DIE entry missing name", [&]() {
- error() << formatv(
- "Name Index @ {0:x}: Entry for DIE @ {1:x} ({2}) with "
- "name {3} missing.\n",
- NI.getUnitOffset(), Die.getOffset(), Die.getTag(), Name);
- });
+ ErrorCategory.Report(
+ "Name Index DIE entry missing name",
+ llvm::dwarf::TagString(Die.getTag()), [&]() {
+ error() << formatv(
+ "Name Index @ {0:x}: Entry for DIE @ {1:x} ({2}) with "
+ "name {3} missing.\n",
+ NI.getUnitOffset(), Die.getOffset(), Die.getTag(), Name);
+ });
++NumErrors;
}
}
@@ -2169,15 +2171,35 @@ bool DWARFVerifier::verifyDebugStrOffsets(
void OutputCategoryAggregator::Report(
StringRef s, std::function<void(void)> detailCallback) {
- Aggregation[std::string(s)]++;
+ this->Report(s, "", detailCallback);
+}
+
+void OutputCategoryAggregator::Report(
+ StringRef category, StringRef sub_category,
+ std::function<void(void)> detailCallback) {
+ std::string category_str = std::string(category);
+ AggregationData *Agg = &Aggregation[category_str];
+ Agg->OverallCount++;
+ if (!sub_category.empty()) {
+ Agg->DetailedCounts[std::string(sub_category)]++;
+ }
if (IncludeDetail)
detailCallback();
}
void OutputCategoryAggregator::EnumerateResults(
std::function<void(StringRef, unsigned)> handleCounts) {
- for (auto &&[name, count] : Aggregation) {
- handleCounts(name, count);
+ for (auto &&[name, aggData] : Aggregation) {
+ handleCounts(name, aggData.OverallCount);
+ }
+}
+void OutputCategoryAggregator::EnumerateDetailedResultsFor(
+ StringRef category, std::function<void(StringRef, unsigned)> handleCounts) {
+ auto Agg = Aggregation.find(std::string(category));
+ if (Agg != Aggregation.end()) {
+ for (auto &&[name, count] : Agg->second.DetailedCounts) {
+ handleCounts(name, count);
+ }
}
}
@@ -2204,6 +2226,12 @@ void DWARFVerifier::summarize() {
ErrorCategory.EnumerateResults([&](StringRef Category, unsigned Count) {
llvm::json::Object Val;
Val.try_emplace("count", Count);
+ llvm::json::Object Details;
+ ErrorCategory.EnumerateDetailedResultsFor(
+ Category, [&](StringRef SubCategory, unsigned SubCount) {
+ Details.try_emplace(SubCategory, SubCount);
+ });
+ Val.try_emplace("details", std::move(Details));
Categories.try_emplace(Category, std::move(Val));
ErrorCount += Count;
});
diff --git a/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify--completeness-json-output.s b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness-json-output.s
similarity index 98%
rename from llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify--completeness-json-output.s
rename to llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness-json-output.s
index c4b7ffe1d2d8e7..9c9658308c2f90 100644
--- a/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify--completeness-json-output.s
+++ b/llvm/test/tools/llvm-dwarfdump/X86/debug-names-verify-completeness-json-output.s
@@ -1,7 +1,7 @@
# RUN: llvm-mc -triple x86_64-pc-linux %s -filetype=obj -o - | not llvm-dwarfdump -verify --verify-json=%t.json -
# RUN: FileCheck %s --input-file %t.json
-# CHECK: {"error-categories":{"Name Index DIE entry missing name":{"count":10}},"error-count":10}
+# CHECK: {"error-categories":{"Name Index DIE entry missing name":{"count":10,"details":{"DW_TAG_inlined_subroutine":1,"DW_TAG_label":1,"DW_TAG_namespace":2,"DW_TAG_subprogram":2,"DW_TAG_variable":4}}},"error-count":10}
# CHECK-NOT: error: Name Index @ 0x0: Entry for DIE @ {{.*}} (DW_TAG_variable) with name var_block_addr missing.
.section .debug_loc,"", at progbits
More information about the llvm-commits
mailing list