maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/CodeGen/exceptions-seh.c

284 lines
10 KiB
C
Raw Normal View History

Remove -fnew-ms-eh This flag no longer controls any behavior inside of clang. llvm-svn: 261423
2016-02-20 17:23:41 +08:00
// RUN: %clang_cc1 %s -triple x86_64-pc-win32 -fms-extensions -emit-llvm -o - \
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// RUN: | FileCheck %s --check-prefix=CHECK --check-prefix=X64
Remove -fnew-ms-eh This flag no longer controls any behavior inside of clang. llvm-svn: 261423
2016-02-20 17:23:41 +08:00
// RUN: %clang_cc1 %s -triple i686-pc-win32 -fms-extensions -emit-llvm -o - \
Re-enable 32-bit SEH after the alignment fix llvm-svn: 241878
2015-07-10 08:16:25 +08:00
// RUN: | FileCheck %s --check-prefix=CHECK --check-prefix=X86
Remove -fnew-ms-eh This flag no longer controls any behavior inside of clang. llvm-svn: 261423
2016-02-20 17:23:41 +08:00
// RUN: %clang_cc1 %s -triple i686-pc-windows-gnu -fms-extensions -emit-llvm -o - \
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// RUN: | FileCheck %s --check-prefix=X86-GNU
Remove -fnew-ms-eh This flag no longer controls any behavior inside of clang. llvm-svn: 261423
2016-02-20 17:23:41 +08:00
// RUN: %clang_cc1 %s -triple x86_64-pc-windows-gnu -fms-extensions -emit-llvm -o - \
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// RUN: | FileCheck %s --check-prefix=X64-GNU
Emit an error when attempting to generate IR for SEH __try Currently we silently omit the code in the try and finally bodies, which is pretty bad. This way we fail loudly. llvm-svn: 190809
2013-09-17 05:46:30 +08:00
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
void try_body(int numerator, int denominator, int *myres) {
*myres = numerator / denominator;
}
// CHECK-LABEL: define void @try_body(i32 %numerator, i32 %denominator, i32* %myres)
// CHECK: sdiv i32
// CHECK: store i32 %{{.*}}, i32*
// CHECK: ret void
Emit an error when attempting to generate IR for SEH __try Currently we silently omit the code in the try and finally bodies, which is pretty bad. This way we fail loudly. llvm-svn: 190809
2013-09-17 05:46:30 +08:00
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
int safe_div(int numerator, int denominator, int *res) {
Emit an error when attempting to generate IR for SEH __try Currently we silently omit the code in the try and finally bodies, which is pretty bad. This way we fail loudly. llvm-svn: 190809
2013-09-17 05:46:30 +08:00
int myres = 0;
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
int success = 1;
Emit an error when attempting to generate IR for SEH __try Currently we silently omit the code in the try and finally bodies, which is pretty bad. This way we fail loudly. llvm-svn: 190809
2013-09-17 05:46:30 +08:00
__try {
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
try_body(numerator, denominator, &myres);
Emit an error when attempting to generate IR for SEH __try Currently we silently omit the code in the try and finally bodies, which is pretty bad. This way we fail loudly. llvm-svn: 190809
2013-09-17 05:46:30 +08:00
} __except (1) {
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
success = -42;
Emit an error when attempting to generate IR for SEH __try Currently we silently omit the code in the try and finally bodies, which is pretty bad. This way we fail loudly. llvm-svn: 190809
2013-09-17 05:46:30 +08:00
}
*res = myres;
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
return success;
}
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK-LABEL: define i32 @safe_div(i32 %numerator, i32 %denominator, i32* %res)
// X64-SAME: personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
// X86-SAME: personality i8* bitcast (i32 (...)* @_except_handler3 to i8*)
// CHECK: invoke void @try_body(i32 %{{.*}}, i32 %{{.*}}, i32* %{{.*}}) #[[NOINLINE:[0-9]+]]
// CHECK: to label %{{.*}} unwind label %[[catchpad:[^ ]*]]
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[catchpad]]
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// X64: %[[padtoken:[^ ]*]] = catchpad within %{{[^ ]*}} [i8* null]
// X86: %[[padtoken:[^ ]*]] = catchpad within %{{[^ ]*}} [i8* bitcast (i32 ()* @"\01?filt$0@0@safe_div@@" to i8*)]
// CHECK-NEXT: catchret from %[[padtoken]] to label %[[except:[^ ]*]]
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[except]]
// CHECK: store i32 -42, i32* %[[success:[^ ]*]]
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: %[[res:[^ ]*]] = load i32, i32* %[[success]]
// CHECK: ret i32 %[[res]]
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// 32-bit SEH needs this filter to save the exception code.
//
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// X86-LABEL: define internal i32 @"\01?filt$0@0@safe_div@@"()
// X86: %[[ebp:[^ ]*]] = call i8* @llvm.frameaddress(i32 1)
// X86: %[[fp:[^ ]*]] = call i8* @llvm.x86.seh.recoverfp(i8* bitcast (i32 (i32, i32, i32*)* @safe_div to i8*), i8* %[[ebp]])
Update clang for intrinsic rename of framerecover to localrecover llvm-svn: 241634
2015-07-08 06:26:07 +08:00
// X86: call i8* @llvm.localrecover(i8* bitcast (i32 (i32, i32, i32*)* @safe_div to i8*), i8* %[[fp]], i32 0)
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// X86: load i8*, i8**
// X86: load i32*, i32**
// X86: load i32, i32*
// X86: store i32 %{{.*}}, i32*
// X86: ret i32 1
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// Mingw uses msvcrt, so it can also use _except_handler3.
// X86-GNU-LABEL: define i32 @safe_div(i32 %numerator, i32 %denominator, i32* %res)
// X86-GNU-SAME: personality i8* bitcast (i32 (...)* @_except_handler3 to i8*)
// X64-GNU-LABEL: define i32 @safe_div(i32 %numerator, i32 %denominator, i32* %res)
// X64-GNU-SAME: personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
void j(void);
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
int filter_expr_capture(void) {
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
int r = 42;
__try {
j();
Remove stale FIXMEs from test case llvm-svn: 235937
2015-04-28 07:14:24 +08:00
} __except(r = -1) {
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
r = 13;
}
return r;
}
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK-LABEL: define i32 @filter_expr_capture()
// X64-SAME: personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
// X86-SAME: personality i8* bitcast (i32 (...)* @_except_handler3 to i8*)
Update clang for intrinsic rename of framerecover to localrecover llvm-svn: 241634
2015-07-08 06:26:07 +08:00
// X64: call void (...) @llvm.localescape(i32* %[[r:[^ ,]*]])
// X86: call void (...) @llvm.localescape(i32* %[[r:[^ ,]*]], i32* %[[code:[^ ,]*]])
Remove stale FIXMEs from test case llvm-svn: 235937
2015-04-28 07:14:24 +08:00
// CHECK: store i32 42, i32* %[[r]]
Add the 'noinline' attribute to call sites within __try bodies LLVM doesn't support non-call exceptions, so inlining makes it harder to catch such asynchronous exceptions. llvm-svn: 228876
2015-02-12 05:40:48 +08:00
// CHECK: invoke void @j() #[[NOINLINE]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
//
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: catchpad within %{{[^ ]*}} [i8* bitcast (i32 ({{.*}})* @"\01?filt$0@0@filter_expr_capture@@" to i8*)]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
// CHECK: store i32 13, i32* %[[r]]
//
Update Clang tests to handle explicitly typed load changes in LLVM. llvm-svn: 230795
2015-02-28 05:19:58 +08:00
// CHECK: %[[rv:[^ ]*]] = load i32, i32* %[[r]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
// CHECK: ret i32 %[[rv]]
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// X64-LABEL: define internal i32 @"\01?filt$0@0@filter_expr_capture@@"(i8* %exception_pointers, i8* %frame_pointer)
Reland "[SEH] Use llvm.x86.seh.recoverfp for SEH filters on x64 as well as 32bit" I forgot to initialize RecoverFP. llvm-svn: 255731
2015-12-16 08:26:37 +08:00
// X64: %[[fp:[^ ]*]] = call i8* @llvm.x86.seh.recoverfp(i8* bitcast (i32 ()* @filter_expr_capture to i8*), i8* %frame_pointer)
// X64: call i8* @llvm.localrecover(i8* bitcast (i32 ()* @filter_expr_capture to i8*), i8* %[[fp]], i32 0)
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
//
// X86-LABEL: define internal i32 @"\01?filt$0@0@filter_expr_capture@@"()
// X86: %[[ebp:[^ ]*]] = call i8* @llvm.frameaddress(i32 1)
// X86: %[[fp:[^ ]*]] = call i8* @llvm.x86.seh.recoverfp(i8* bitcast (i32 ()* @filter_expr_capture to i8*), i8* %[[ebp]])
Update clang for intrinsic rename of framerecover to localrecover llvm-svn: 241634
2015-07-08 06:26:07 +08:00
// X86: call i8* @llvm.localrecover(i8* bitcast (i32 ()* @filter_expr_capture to i8*), i8* %[[fp]], i32 0)
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
//
Remove stale FIXMEs from test case llvm-svn: 235937
2015-04-28 07:14:24 +08:00
// CHECK: store i32 -1, i32* %{{.*}}
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
// CHECK: ret i32 -1
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
int nested_try(void) {
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
int r = 42;
__try {
__try {
j();
r = 0;
} __except(_exception_code() == 123) {
r = 123;
}
} __except(_exception_code() == 456) {
r = 456;
}
return r;
}
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK-LABEL: define i32 @nested_try()
// X64-SAME: personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
// X86-SAME: personality i8* bitcast (i32 (...)* @_except_handler3 to i8*)
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
// CHECK: store i32 42, i32* %[[r:[^ ,]*]]
Add the 'noinline' attribute to call sites within __try bodies LLVM doesn't support non-call exceptions, so inlining makes it harder to catch such asynchronous exceptions. llvm-svn: 228876
2015-02-12 05:40:48 +08:00
// CHECK: invoke void @j() #[[NOINLINE]]
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: to label %[[cont:[^ ]*]] unwind label %[[cswitch_inner:[^ ]*]]
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
//
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: [[cswitch_inner]]
// CHECK: %[[cs_inner:[^ ]*]] = catchswitch within none [label %[[cpad_inner:[^ ]*]]] unwind label %[[cswitch_outer:[^ ]*]]
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
//
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: [[cswitch_outer]]
// CHECK: %[[cs_outer:[^ ]*]] = catchswitch within none [label %[[cpad_outer:[^ ]*]]] unwind to caller
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[cpad_outer]]
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: catchpad within %{{[^ ]*}} [i8* bitcast (i32 ({{.*}})* @"\01?filt$0@0@nested_try@@" to i8*)]
// CHECK-NEXT: catchret {{.*}} to label %[[except_outer:[^ ]*]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[except_outer]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
// CHECK: store i32 456, i32* %[[r]]
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: br label %[[outer_try_cont:[^ ]*]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
//
// CHECK: [[outer_try_cont]]
Update Clang tests to handle explicitly typed load changes in LLVM. llvm-svn: 230795
2015-02-28 05:19:58 +08:00
// CHECK: %[[r_load:[^ ]*]] = load i32, i32* %[[r]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
// CHECK: ret i32 %[[r_load]]
//
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: [[cpad_inner]]
// CHECK: catchpad within %[[cs_inner]] [i8* bitcast (i32 ({{.*}})* @"\01?filt$1@0@nested_try@@" to i8*)]
// CHECK-NEXT: catchret {{.*}} to label %[[except_inner:[^ ]*]]
//
// CHECK: [[except_inner]]
// CHECK: store i32 123, i32* %[[r]]
// CHECK: br label %[[inner_try_cont:[^ ]*]]
//
// CHECK: [[inner_try_cont]]
// CHECK: br label %[[outer_try_cont]]
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[cont]]
// CHECK: store i32 0, i32* %[[r]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
// CHECK: br label %[[inner_try_cont]]
//
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK-LABEL: define internal i32 @"\01?filt$0@0@nested_try@@"({{.*}})
// X86: call i8* @llvm.x86.seh.recoverfp({{.*}})
// CHECK: load i32*, i32**
// CHECK: load i32, i32*
// CHECK: icmp eq i32 %{{.*}}, 456
//
// CHECK-LABEL: define internal i32 @"\01?filt$1@0@nested_try@@"({{.*}})
// X86: call i8* @llvm.x86.seh.recoverfp({{.*}})
// CHECK: load i32*, i32**
// CHECK: load i32, i32*
// CHECK: icmp eq i32 %{{.*}}, 123
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
int basic_finally(int g) {
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
__try {
j();
} __finally {
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
++g;
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
}
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
return g;
Emit an error when attempting to generate IR for SEH __try Currently we silently omit the code in the try and finally bodies, which is pretty bad. This way we fail loudly. llvm-svn: 190809
2013-09-17 05:46:30 +08:00
}
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK-LABEL: define i32 @basic_finally(i32 %g)
// X64-SAME: personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
// X86-SAME: personality i8* bitcast (i32 (...)* @_except_handler3 to i8*)
// CHECK: %[[g_addr:[^ ]*]] = alloca i32, align 4
Update clang for intrinsic rename of framerecover to localrecover llvm-svn: 241634
2015-07-08 06:26:07 +08:00
// CHECK: call void (...) @llvm.localescape(i32* %[[g_addr]])
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK: store i32 %g, i32* %[[g_addr]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
//
// CHECK: invoke void @j()
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: to label %[[cont:[^ ]*]] unwind label %[[cleanuppad:[^ ]*]]
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
//
// CHECK: [[cont]]
[SEH] Switch from frameaddress(0) to localaddress This should do the right thing for stack realignment prologues. llvm-svn: 241644
2015-07-08 07:23:31 +08:00
// CHECK: %[[fp:[^ ]*]] = call i8* @llvm.localaddress()
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK: call void @"\01?fin$0@0@basic_finally@@"({{i8( zeroext)?}} 0, i8* %[[fp]])
// CHECK: load i32, i32* %[[g_addr]], align 4
// CHECK: ret i32
Initial support for Win64 SEH IR emission The lowering looks a lot like normal EH lowering, with the exception that the exceptions are caught by executing filter expression code instead of matching typeinfo globals. The filter expressions are outlined into functions which are used in landingpad clauses where typeinfo would normally go. Major aspects that still need work: - Non-call exceptions in __try bodies won't work yet. The plan is to outline the __try block in the frontend to keep things simple. - Filter expressions cannot use local variables until capturing is implemented. - __finally blocks will not run after exceptions. Fixing this requires work in the LLVM SEH preparation pass. The IR lowering looks like this: // C code: bool safe_div(int n, int d, int *r) { __try { *r = normal_div(n, d); } __except(_exception_code() == EXCEPTION_INT_DIVIDE_BY_ZERO) { return false; } return true; } ; LLVM IR: define i32 @filter(i8* %e, i8* %fp) { %ehptrs = bitcast i8* %e to i32** %ehrec = load i32** %ehptrs %code = load i32* %ehrec %matches = icmp eq i32 %code, i32 u0xC0000094 %matches.i32 = zext i1 %matches to i32 ret i32 %matches.i32 } define i1 zeroext @safe_div(i32 %n, i32 %d, i32* %r) { %rr = invoke i32 @normal_div(i32 %n, i32 %d) to label %normal unwind to label %lpad normal: store i32 %rr, i32* %r ret i1 1 lpad: %ehvals = landingpad {i8*, i32} personality i32 (...)* @__C_specific_handler catch i8* bitcast (i32 (i8*, i8*)* @filter to i8*) %ehptr = extractvalue {i8*, i32} %ehvals, i32 0 %sel = extractvalue {i8*, i32} %ehvals, i32 1 %filter_sel = call i32 @llvm.eh.seh.typeid.for(i8* bitcast (i32 (i8*, i8*)* @filter to i8*)) %matches = icmp eq i32 %sel, %filter_sel br i1 %matches, label %eh.except, label %eh.resume eh.except: ret i1 false eh.resume: resume } Reviewers: rjmccall, rsmith, majnemer Differential Revision: http://reviews.llvm.org/D5607 llvm-svn: 226760
2015-01-22 09:36:17 +08:00
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[cleanuppad]]
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: %[[padtoken:[^ ]*]] = cleanuppad within none []
[SEH] Switch from frameaddress(0) to localaddress This should do the right thing for stack realignment prologues. llvm-svn: 241644
2015-07-08 07:23:31 +08:00
// CHECK: %[[fp:[^ ]*]] = call i8* @llvm.localaddress()
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: call void @"\01?fin$0@0@basic_finally@@"({{i8( zeroext)?}} 1, i8* %[[fp]])
// CHECK: cleanupret from %[[padtoken]] unwind to caller
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK: define internal void @"\01?fin$0@0@basic_finally@@"({{i8( zeroext)?}} %abnormal_termination, i8* %frame_pointer)
Update clang for intrinsic rename of framerecover to localrecover llvm-svn: 241634
2015-07-08 06:26:07 +08:00
// CHECK: call i8* @llvm.localrecover(i8* bitcast (i32 (i32)* @basic_finally to i8*), i8* %frame_pointer, i32 0)
Revert "Revert 241171, 241187, 241199 (32-bit SEH)." This reverts commit r241244, but restricts SEH support to Win64. This way, Chromium builds will still fall back on TUs with SEH, and Clang developers can work on this incrementally upstream while patching this small predicate locally. It'll also make it easier to review small fixes. llvm-svn: 241533
2015-07-07 08:36:30 +08:00
// CHECK: load i32, i32* %{{.*}}, align 4
// CHECK: add nsw i32 %{{.*}}, 1
// CHECK: store i32 %{{.*}}, i32* %{{.*}}, align 4
Reland r234613 (and follow-ups 234614, 234616, 234618) The frameescape intrinsic cannot be inlined, so I fixed the inliner in r234937. This should address PR23216. llvm-svn: 234942
2015-04-15 04:59:00 +08:00
// CHECK: ret void
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
int returns_int(void);
int except_return(void) {
__try {
return returns_int();
} __except(1) {
return 42;
}
}
// CHECK-LABEL: define i32 @except_return()
// CHECK: %[[tmp:[^ ]*]] = invoke i32 @returns_int()
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: to label %[[cont:[^ ]*]] unwind label %[[catchpad:[^ ]*]]
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[catchpad]]
// CHECK: catchpad
// CHECK: catchret
// CHECK: store i32 42, i32* %[[rv:[^ ]*]]
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
// CHECK: br label %[[retbb:[^ ]*]]
//
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// CHECK: [[cont]]
// CHECK: store i32 %[[tmp]], i32* %[[rv]]
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
// CHECK: br label %[[retbb]]
//
// CHECK: [[retbb]]
Update Clang tests to handle explicitly typed load changes in LLVM. llvm-svn: 230795
2015-02-28 05:19:58 +08:00
// CHECK: %[[r:[^ ]*]] = load i32, i32* %[[rv]]
SEH: Don't jump to an unreachable continuation block If both the __try and __except blocks do not return, we want to delete the continuation block as unreachable instead. llvm-svn: 227627
2015-01-31 06:16:45 +08:00
// CHECK: ret i32 %[[r]]
Add the 'noinline' attribute to call sites within __try bodies LLVM doesn't support non-call exceptions, so inlining makes it harder to catch such asynchronous exceptions. llvm-svn: 228876
2015-02-12 05:40:48 +08:00
Collect SEH captures in a set instead of a vector to avoid doing redundant work if a variable is used multiple times. Fixes PR24751. llvm-svn: 247075
2015-09-09 05:15:22 +08:00
// PR 24751: don't assert if a variable is used twice in a __finally block.
// Also, make sure we don't do redundant work to capture/project it.
void finally_capture_twice(int x) {
__try {
} __finally {
int y = x;
int z = x;
}
}
[SEH] Use catchret in the new EH IR like we do for C++ Also add tests for SEH with the new IRGen. llvm-svn: 247318
2015-09-11 02:39:41 +08:00
//
Collect SEH captures in a set instead of a vector to avoid doing redundant work if a variable is used multiple times. Fixes PR24751. llvm-svn: 247075
2015-09-09 05:15:22 +08:00
// CHECK-LABEL: define void @finally_capture_twice(
// CHECK: [[X:%.*]] = alloca i32, align 4
// CHECK: call void (...) @llvm.localescape(i32* [[X]])
// CHECK-NEXT: store i32 {{.*}}, i32* [[X]], align 4
// CHECK-NEXT: [[LOCAL:%.*]] = call i8* @llvm.localaddress()
// CHECK-NEXT: call void [[FINALLY:@.*]](i8{{ zeroext | }}0, i8* [[LOCAL]])
// CHECK: define internal void [[FINALLY]](
// CHECK: [[LOCAL:%.*]] = call i8* @llvm.localrecover(
// CHECK: [[X:%.*]] = bitcast i8* [[LOCAL]] to i32*
// CHECK-NEXT: [[Y:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[Z:%.*]] = alloca i32, align 4
// CHECK-NEXT: store i8*
// CHECK-NEXT: store i8
// CHECK-NEXT: [[T0:%.*]] = load i32, i32* [[X]], align 4
// CHECK-NEXT: store i32 [[T0]], i32* [[Y]], align 4
// CHECK-NEXT: [[T0:%.*]] = load i32, i32* [[X]], align 4
// CHECK-NEXT: store i32 [[T0]], i32* [[Z]], align 4
// CHECK-NEXT: ret void
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
int exception_code_in_except(void) {
__try {
try_body(0, 0, 0);
} __except(1) {
return _exception_code();
}
}
// CHECK-LABEL: define i32 @exception_code_in_except()
// CHECK: %[[ret_slot:[^ ]*]] = alloca i32
// CHECK: %[[code_slot:[^ ]*]] = alloca i32
// CHECK: invoke void @try_body(i32 0, i32 0, i32* null)
// CHECK: %[[pad:[^ ]*]] = catchpad
Update clang to use the updated LLVM EH instructions Depends on D15139. Reviewers: rnk Differential Revision: http://reviews.llvm.org/D15140 llvm-svn: 255423
2015-12-12 13:39:21 +08:00
// CHECK: catchret from %[[pad]]
[WinEH] Remove NewMSEH and enable its behavior by default Testing has shown that it is at least as reliable as the old landingpad pattern matching code. llvm-svn: 249647
2015-10-08 09:13:52 +08:00
// X64: %[[code:[^ ]*]] = call i32 @llvm.eh.exceptioncode(token %[[pad]])
// X64: store i32 %[[code]], i32* %[[code_slot]]
// CHECK: %[[ret1:[^ ]*]] = load i32, i32* %[[code_slot]]
// CHECK: store i32 %[[ret1]], i32* %[[ret_slot]]
// CHECK: %[[ret2:[^ ]*]] = load i32, i32* %[[ret_slot]]
// CHECK: ret i32 %[[ret2]]
Add the 'noinline' attribute to call sites within __try bodies LLVM doesn't support non-call exceptions, so inlining makes it harder to catch such asynchronous exceptions. llvm-svn: 228876
2015-02-12 05:40:48 +08:00
// CHECK: attributes #[[NOINLINE]] = { {{.*noinline.*}} }