2015-09-27 09:47:46 +08:00
|
|
|
; RUN: opt -S -simplifycfg < %s | FileCheck %s
|
|
|
|
|
|
|
|
|
|
declare void @Personality()
|
|
|
|
|
declare void @f()
|
|
|
|
|
|
|
|
|
|
; CHECK-LABEL: define void @test1()
|
|
|
|
|
define void @test1() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
; CHECK: call void @f()
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %unreachable.unwind
|
|
|
|
|
exit:
|
|
|
|
|
ret void
|
|
|
|
|
unreachable.unwind:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
cleanuppad within none []
|
2015-09-27 09:47:46 +08:00
|
|
|
unreachable
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; CHECK-LABEL: define void @test2()
|
|
|
|
|
define void @test2() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %catch.pad
|
|
|
|
|
catch.pad:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
%cs1 = catchswitch within none [label %catch.body] unwind label %unreachable.unwind
|
|
|
|
|
; CHECK: catch.pad:
|
|
|
|
|
; CHECK-NEXT: catchswitch within none [label %catch.body] unwind to caller
|
2015-09-27 09:47:46 +08:00
|
|
|
catch.body:
|
|
|
|
|
; CHECK: catch.body:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
; CHECK-NEXT: catchpad within %cs1
|
2015-09-27 09:47:46 +08:00
|
|
|
; CHECK-NEXT: call void @f()
|
|
|
|
|
; CHECK-NEXT: unreachable
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
%catch = catchpad within %cs1 []
|
2015-09-27 09:47:46 +08:00
|
|
|
call void @f()
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
catchret from %catch to label %unreachable
|
2015-09-27 09:47:46 +08:00
|
|
|
exit:
|
|
|
|
|
ret void
|
|
|
|
|
unreachable.unwind:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
cleanuppad within none []
|
2015-09-27 09:47:46 +08:00
|
|
|
unreachable
|
|
|
|
|
unreachable:
|
|
|
|
|
unreachable
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; CHECK-LABEL: define void @test3()
|
|
|
|
|
define void @test3() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %cleanup.pad
|
|
|
|
|
cleanup.pad:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
; CHECK: %cleanup = cleanuppad within none []
|
2015-09-27 09:47:46 +08:00
|
|
|
; CHECK-NEXT: call void @f()
|
|
|
|
|
; CHECK-NEXT: unreachable
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
%cleanup = cleanuppad within none []
|
2015-09-27 09:47:46 +08:00
|
|
|
invoke void @f()
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
to label %cleanup.ret unwind label %unreachable.unwind
|
2015-09-27 09:47:46 +08:00
|
|
|
cleanup.ret:
|
|
|
|
|
; This cleanupret should be rewritten to unreachable,
|
|
|
|
|
; and merged into the pred block.
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
cleanupret from %cleanup unwind label %unreachable.unwind
|
2015-09-27 09:47:46 +08:00
|
|
|
exit:
|
|
|
|
|
ret void
|
|
|
|
|
unreachable.unwind:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
cleanuppad within none []
|
2015-09-27 09:47:46 +08:00
|
|
|
unreachable
|
|
|
|
|
}
|
|
|
|
|
|
2015-10-28 06:43:56 +08:00
|
|
|
; CHECK-LABEL: define void @test5()
|
|
|
|
|
define void @test5() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %catch.pad
|
|
|
|
|
|
|
|
|
|
catch.pad:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
%cs1 = catchswitch within none [label %catch.body] unwind to caller
|
2015-10-28 06:43:56 +08:00
|
|
|
|
|
|
|
|
catch.body:
|
[IR] Reformulate LLVM's EH funclet IR
While we have successfully implemented a funclet-oriented EH scheme on
top of LLVM IR, our scheme has some notable deficiencies:
- catchendpad and cleanupendpad are necessary in the current design
but they are difficult to explain to others, even to seasoned LLVM
experts.
- catchendpad and cleanupendpad are optimization barriers. They cannot
be split and force all potentially throwing call-sites to be invokes.
This has a noticable effect on the quality of our code generation.
- catchpad, while similar in some aspects to invoke, is fairly awkward.
It is unsplittable, starts a funclet, and has control flow to other
funclets.
- The nesting relationship between funclets is currently a property of
control flow edges. Because of this, we are forced to carefully
analyze the flow graph to see if there might potentially exist illegal
nesting among funclets. While we have logic to clone funclets when
they are illegally nested, it would be nicer if we had a
representation which forbade them upfront.
Let's clean this up a bit by doing the following:
- Instead, make catchpad more like cleanuppad and landingpad: no control
flow, just a bunch of simple operands; catchpad would be splittable.
- Introduce catchswitch, a control flow instruction designed to model
the constraints of funclet oriented EH.
- Make funclet scoping explicit by having funclet instructions consume
the token produced by the funclet which contains them.
- Remove catchendpad and cleanupendpad. Their presence can be inferred
implicitly using coloring information.
N.B. The state numbering code for the CLR has been updated but the
veracity of it's output cannot be spoken for. An expert should take a
look to make sure the results are reasonable.
Reviewers: rnk, JosephTremoulet, andrew.w.kaylor
Differential Revision: http://reviews.llvm.org/D15139
llvm-svn: 255422
2015-12-12 13:38:55 +08:00
|
|
|
%catch = catchpad within %cs1 []
|
|
|
|
|
catchret from %catch to label %exit
|
2015-10-28 06:43:56 +08:00
|
|
|
|
|
|
|
|
exit:
|
|
|
|
|
unreachable
|
|
|
|
|
}
|
2016-01-05 14:27:50 +08:00
|
|
|
|
|
|
|
|
; CHECK-LABEL: define void @test6()
|
|
|
|
|
define void @test6() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %catch.pad
|
|
|
|
|
|
|
|
|
|
catch.pad:
|
|
|
|
|
%cs1 = catchswitch within none [label %catch.body, label %catch.body] unwind to caller
|
|
|
|
|
; CHECK: catchswitch within none [label %catch.body] unwind to caller
|
|
|
|
|
|
|
|
|
|
catch.body:
|
|
|
|
|
%catch = catchpad within %cs1 [i8* null, i32 0, i8* null]
|
|
|
|
|
catchret from %catch to label %exit
|
|
|
|
|
|
|
|
|
|
exit:
|
|
|
|
|
ret void
|
|
|
|
|
}
|
2016-01-05 15:42:17 +08:00
|
|
|
|
|
|
|
|
; CHECK-LABEL: define void @test7()
|
|
|
|
|
define void @test7() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %catch.pad
|
|
|
|
|
|
|
|
|
|
catch.pad:
|
|
|
|
|
%cs1 = catchswitch within none [label %catch.body, label %catch.body2] unwind to caller
|
|
|
|
|
; CHECK: catchswitch within none [label %catch.body] unwind to caller
|
|
|
|
|
|
|
|
|
|
catch.body:
|
|
|
|
|
%catch = catchpad within %cs1 [i8* null, i32 0, i8* null]
|
|
|
|
|
catchret from %catch to label %exit
|
|
|
|
|
|
|
|
|
|
catch.body2:
|
|
|
|
|
%catch2 = catchpad within %cs1 [i8* null, i32 0, i8* null]
|
|
|
|
|
catchret from %catch2 to label %exit
|
|
|
|
|
|
|
|
|
|
exit:
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; CHECK-LABEL: define void @test8()
|
|
|
|
|
define void @test8() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %catch.pad
|
|
|
|
|
|
|
|
|
|
catch.pad:
|
|
|
|
|
%cs1 = catchswitch within none [label %catch.body, label %catch.body2] unwind to caller
|
|
|
|
|
; CHECK: catchswitch within none [label %catch.body] unwind to caller
|
|
|
|
|
|
|
|
|
|
catch.body2:
|
|
|
|
|
%catch2 = catchpad within %cs1 [i8* null, i32 0, i8* null]
|
|
|
|
|
catchret from %catch2 to label %exit
|
|
|
|
|
|
|
|
|
|
catch.body:
|
|
|
|
|
%catch = catchpad within %cs1 [i8* null, i32 0, i8* null]
|
|
|
|
|
catchret from %catch to label %exit
|
|
|
|
|
|
|
|
|
|
exit:
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; CHECK-LABEL: define void @test9()
|
|
|
|
|
define void @test9() personality i8* bitcast (void ()* @Personality to i8*) {
|
|
|
|
|
entry:
|
|
|
|
|
invoke void @f()
|
|
|
|
|
to label %exit unwind label %catch.pad
|
|
|
|
|
|
|
|
|
|
catch.pad:
|
|
|
|
|
%cs1 = catchswitch within none [label %catch.body, label %catch.body2] unwind to caller
|
|
|
|
|
; CHECK: catchswitch within none [label %catch.body, label %catch.body2] unwind to caller
|
|
|
|
|
|
|
|
|
|
catch.body:
|
|
|
|
|
%catch = catchpad within %cs1 [i8* null, i32 0, i8* null]
|
|
|
|
|
catchret from %catch to label %exit
|
|
|
|
|
|
|
|
|
|
catch.body2:
|
|
|
|
|
%catch2 = catchpad within %cs1 [i8* null, i32 64, i8* null]
|
|
|
|
|
catchret from %catch2 to label %exit
|
|
|
|
|
|
|
|
|
|
exit:
|
|
|
|
|
ret void
|
|
|
|
|
}
|