[Lldb-commits] [PATCH] D57907: lldb: Fix compilation on OpenBSD
mortimer created this revision.
mortimer added a reviewer: krytarowski.
Herald added a project: LLDB.
Herald added a subscriber: lldb-commits.
Update the OpenBSD Host.cpp for the new SetFile() function signature. Fixes
compiling lldb on OpenBSD.
Repository:
rLLDB LLDB
https://reviews.llvm.org/D57907
Files:
lldb/source/Host/openbsd/Host.cpp
Index: lldb/source/Host/openbsd/Host.cpp
===
--- lldb/source/Host/openbsd/Host.cpp
+++ lldb/source/Host/openbsd/Host.cpp
@@ -67,8 +67,7 @@
cstr = data.GetCStr(&offset);
if (cstr) {
-process_info.GetExecutableFile().SetFile(cstr, false,
- FileSpec::Style::native);
+process_info.GetExecutableFile().SetFile(cstr,
FileSpec::Style::native);
if (!(match_info_ptr == NULL ||
NameMatches(
Index: lldb/source/Host/openbsd/Host.cpp
===
--- lldb/source/Host/openbsd/Host.cpp
+++ lldb/source/Host/openbsd/Host.cpp
@@ -67,8 +67,7 @@
cstr = data.GetCStr(&offset);
if (cstr) {
-process_info.GetExecutableFile().SetFile(cstr, false,
- FileSpec::Style::native);
+process_info.GetExecutableFile().SetFile(cstr, FileSpec::Style::native);
if (!(match_info_ptr == NULL ||
NameMatches(
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [PATCH] D57928: Fix x86 return pattern detection
mortimer created this revision.
mortimer added a reviewer: jasonmolenda.
Herald added a project: LLDB.
Herald added a subscriber: lldb-commits.
Replace 0xc9 (LEAVE) with 0xcb (RETF) in ret_pattern_p(). Also put 0xc3 first,
since it is the most common form and will match first.
Repository:
rLLDB LLDB
https://reviews.llvm.org/D57928
Files:
lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
Index: lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
===
--- lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
+++ lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
@@ -665,10 +665,10 @@
return false;
}
-// ret [0xc9] or [0xc2 imm8] or [0xca imm8]
+// ret [0xc3] or [0xcb] or [0xc2 imm16] or [0xca imm16]
bool x86AssemblyInspectionEngine::ret_pattern_p() {
uint8_t *p = m_cur_insn;
- return *p == 0xc9 || *p == 0xc2 || *p == 0xca || *p == 0xc3;
+ return *p == 0xc3 || *p == 0xc2 || *p == 0xca || *p == 0xcb;
}
uint32_t x86AssemblyInspectionEngine::extract_4(uint8_t *b) {
Index: lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
===
--- lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
+++ lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
@@ -665,10 +665,10 @@
return false;
}
-// ret [0xc9] or [0xc2 imm8] or [0xca imm8]
+// ret [0xc3] or [0xcb] or [0xc2 imm16] or [0xca imm16]
bool x86AssemblyInspectionEngine::ret_pattern_p() {
uint8_t *p = m_cur_insn;
- return *p == 0xc9 || *p == 0xc2 || *p == 0xca || *p == 0xc3;
+ return *p == 0xc3 || *p == 0xc2 || *p == 0xca || *p == 0xcb;
}
uint32_t x86AssemblyInspectionEngine::extract_4(uint8_t *b) {
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [PATCH] D57928: Fix x86 return pattern detection
mortimer added a comment. For sure I can add a test. I found this only because I was looking to fix something else (prologue detection on OpenBSD with -fret-protector) and noticed that the return pattern detect function was wrong. So nothing was broken that I found, though the inclusion of 0xc9 as a return pattern could conceivably cause problems under some circumstances. In any event, I figured it would be easy enough to submit a fix, since it’s a tiny change. Repository: rLLDB LLDB CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57928/new/ https://reviews.llvm.org/D57928 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [PATCH] D57928: Fix x86 return pattern detection
mortimer updated this revision to Diff 186144.
mortimer added a comment.
Added a test to verify that the unwind state is reset after each kind of return.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D57928/new/
https://reviews.llvm.org/D57928
Files:
lldb/source/Plugins/UnwindAssembly/x86/x86AssemblyInspectionEngine.cpp
lldb/unittests/UnwindAssembly/x86/Testx86AssemblyInspectionEngine.cpp
Index: lldb/unittests/UnwindAssembly/x86/Testx86AssemblyInspectionEngine.cpp
===
--- lldb/unittests/UnwindAssembly/x86/Testx86AssemblyInspectionEngine.cpp
+++ lldb/unittests/UnwindAssembly/x86/Testx86AssemblyInspectionEngine.cpp
@@ -2552,3 +2552,173 @@
}
+TEST_F(Testx86AssemblyInspectionEngine, TestReturnDetect) {
+ std::unique_ptr engine = Getx86_64Inspector();
+
+ // Single fragment with all four return forms.
+ // We want to verify that the unwind state is reset after each ret.
+ uint8_t data[] = {
+ 0x55, // offset 0 -- pushq %rbp
+ 0x48, 0x89, 0xe5, // offset 1 -- movq %rsp, %rbp
+ 0x31, 0xc0, // offset 4 -- xorl %eax, %eax
+ 0x5d, // offset 6 -- popq %rbp
+ 0xc3, // offset 7 -- retq
+ 0x31, 0xc0, // offset 8 -- xorl %eax, %eax
+ 0x5d, // offset 10 -- popq %rbp
+ 0xcb, // offset 11 -- retf
+ 0x31, 0xc0, // offset 12 -- xorl %eax, %eax
+ 0x5d, // offset 14 -- popq %rbp
+ 0xc2, 0x22, 0x11, // offset 15 -- retq 0x1122
+ 0x31, 0xc0, // offset 18 -- xorl %eax, %eax
+ 0x5d, // offset 20 -- popq %rbp
+ 0xca, 0x44, 0x33, // offset 21 -- retf 0x3344
+ 0x31, 0xc0, // offset 24 -- xorl %eax, %eax
+ };
+
+ AddressRange sample_range(0x1000, sizeof(data));
+
+ UnwindPlan unwind_plan(eRegisterKindLLDB);
+ EXPECT_TRUE(engine->GetNonCallSiteUnwindPlanFromAssembly(
+ data, sizeof(data), sample_range, unwind_plan));
+
+ // Expect following unwind rows:
+ // 0: CFA=rsp +8 => rsp=CFA+0 rip=[CFA-8]
+ // 1: CFA=rsp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ // 4: CFA=rbp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ // 7: CFA=rsp +8 => rsp=CFA+0 rip=[CFA-8]
+ // 8: CFA=rbp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ // 11: CFA=rsp +8 => rsp=CFA+0 rip=[CFA-8]
+ // 12: CFA=rbp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ // 15: CFA=rsp +8 => rsp=CFA+0 rip=[CFA-8]
+ // 18: CFA=rbp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ // 21: CFA=rsp +8 => rsp=CFA+0 rip=[CFA-8]
+ // 24: CFA=rbp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+
+ EXPECT_TRUE(unwind_plan.GetInitialCFARegister() == k_rsp);
+ EXPECT_TRUE(unwind_plan.GetUnwindPlanValidAtAllInstructions() ==
+ eLazyBoolYes);
+ EXPECT_TRUE(unwind_plan.GetSourcedFromCompiler() == eLazyBoolNo);
+
+ UnwindPlan::Row::RegisterLocation regloc;
+
+ // 0: CFA=rsp +8 => rsp=CFA+0 rip=[CFA-8]
+ UnwindPlan::RowSP row_sp = unwind_plan.GetRowForFunctionOffset(0);
+ EXPECT_EQ(0ull, row_sp->GetOffset());
+ EXPECT_TRUE(row_sp->GetCFAValue().GetRegisterNumber() == k_rsp);
+ EXPECT_TRUE(row_sp->GetCFAValue().IsRegisterPlusOffset() == true);
+ EXPECT_EQ(8, row_sp->GetCFAValue().GetOffset());
+
+ EXPECT_TRUE(row_sp->GetRegisterInfo(k_rip, regloc));
+ EXPECT_TRUE(regloc.IsAtCFAPlusOffset());
+ EXPECT_EQ(-8, regloc.GetOffset());
+
+ // 1: CFA=rsp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ row_sp = unwind_plan.GetRowForFunctionOffset(1);
+ EXPECT_EQ(1ull, row_sp->GetOffset());
+ EXPECT_TRUE(row_sp->GetCFAValue().GetRegisterNumber() == k_rsp);
+ EXPECT_TRUE(row_sp->GetCFAValue().IsRegisterPlusOffset() == true);
+ EXPECT_EQ(16, row_sp->GetCFAValue().GetOffset());
+
+ EXPECT_TRUE(row_sp->GetRegisterInfo(k_rip, regloc));
+ EXPECT_TRUE(regloc.IsAtCFAPlusOffset());
+ EXPECT_EQ(-8, regloc.GetOffset());
+
+ // 4: CFA=rbp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ row_sp = unwind_plan.GetRowForFunctionOffset(4);
+ EXPECT_EQ(4ull, row_sp->GetOffset());
+ EXPECT_TRUE(row_sp->GetCFAValue().GetRegisterNumber() == k_rbp);
+ EXPECT_TRUE(row_sp->GetCFAValue().IsRegisterPlusOffset() == true);
+ EXPECT_EQ(16, row_sp->GetCFAValue().GetOffset());
+
+ EXPECT_TRUE(row_sp->GetRegisterInfo(k_rip, regloc));
+ EXPECT_TRUE(regloc.IsAtCFAPlusOffset());
+ EXPECT_EQ(-8, regloc.GetOffset());
+
+ // 7: CFA=rsp +8 => rsp=CFA+0 rip=[CFA-8]
+ row_sp = unwind_plan.GetRowForFunctionOffset(7);
+ EXPECT_EQ(7ull, row_sp->GetOffset());
+ EXPECT_TRUE(row_sp->GetCFAValue().GetRegisterNumber() == k_rsp);
+ EXPECT_TRUE(row_sp->GetCFAValue().IsRegisterPlusOffset() == true);
+ EXPECT_EQ(8, row_sp->GetCFAValue().GetOffset());
+
+ EXPECT_TRUE(row_sp->GetRegisterInfo(k_rip, regloc));
+ EXPECT_TRUE(regloc.IsAtCFAPlusOffset());
+ EXPECT_EQ(-8, regloc.GetOffset());
+
+ // 8: CFA=rbp+16 => rbp=[CFA-16] rsp=CFA+0 rip=[CFA-8]
+ row_sp = unwind_plan.GetRowForFunctionOffset(8);
+ EXPECT_EQ(8ull, row_sp->GetOffset());
+ EX
[Lldb-commits] [PATCH] D57928: Fix x86 return pattern detection
mortimer added a comment. Happy to help out. :-) I do not have commit access. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57928/new/ https://reviews.llvm.org/D57928 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
[Lldb-commits] [PATCH] D57907: lldb: Fix compilation on OpenBSD
mortimer added a comment. I don't have commit access, so if there is ready to land, could someone commit it? Thanks! Repository: rLLDB LLDB CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57907/new/ https://reviews.llvm.org/D57907 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
