2017-06-03 00:36:37 +08:00
|
|
|
; RUN: llc -verify-machineinstrs -stack-symbol-ordering=0 < %s | FileCheck %s
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
|
|
|
|
|
target datalayout = "e-i64:64-f80:128-n8:16:32:64-S128"
|
|
|
|
|
target triple = "x86_64-pc-linux-gnu"
|
|
|
|
|
|
|
|
|
|
; This test is checking to make sure that we reuse the same stack slots
|
|
|
|
|
; for GC values spilled over two different call sites. Since the order
|
|
|
|
|
; of GC arguments differ, niave lowering code would insert loads and
|
|
|
|
|
; stores to rearrange items on the stack. We need to make sure (for
|
|
|
|
|
; performance) that this doesn't happen.
|
2015-03-27 13:09:33 +08:00
|
|
|
define i32 @back_to_back_calls(i32 addrspace(1)* %a, i32 addrspace(1)* %b, i32 addrspace(1)* %c) #1 gc "statepoint-example" {
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
; CHECK-LABEL: back_to_back_calls
|
|
|
|
|
; The exact stores don't matter, but there need to be three stack slots created
|
[DAG] Improve Aliasing of operations to static alloca
Re-recommiting after landing DAG extension-crash fix.
Recommiting after adding check to avoid miscomputing alias information
on addresses of the same base but different subindices.
Memory accesses offset from frame indices may alias, e.g., we
may merge write from function arguments passed on the stack when they
are contiguous. As a result, when checking aliasing, we consider the
underlying frame index's offset from the stack pointer.
Static allocs are realized as stack objects in SelectionDAG, but its
offset is not set until post-DAG causing DAGCombiner's alias check to
consider access to static allocas to frequently alias. Modify isAlias
to consider access between static allocas and access from other frame
objects to be considered aliasing.
Many test changes are included here. Most are fixes for tests which
indirectly relied on our aliasing ability and needed to be modified to
preserve their original intent.
The remaining tests have minor improvements due to relaxed
ordering. The exception is CodeGen/X86/2011-10-19-widen_vselect.ll
which has a minor degradation dispite though the pre-legalized DAG is
improved.
Reviewers: rnk, mkuper, jonpa, hfinkel, uweigand
Reviewed By: rnk
Subscribers: sdardis, nemanjai, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33345
llvm-svn: 308350
2017-07-19 04:06:24 +08:00
|
|
|
; CHECK-DAG: movq %rdi, 16(%rsp)
|
|
|
|
|
; CHECK-DAG: movq %rdx, 8(%rsp)
|
|
|
|
|
; CHECK-DAG: movq %rsi, (%rsp)
|
2015-06-17 03:07:05 +08:00
|
|
|
; There should be no more than three moves
|
|
|
|
|
; CHECK-NOT: movq
|
2015-12-26 15:54:32 +08:00
|
|
|
%safepoint_token = tail call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0, i32 addrspace(1)* %a, i32 addrspace(1)* %b, i32 addrspace(1)* %c)
|
|
|
|
|
%a1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 12)
|
|
|
|
|
%b1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 13)
|
|
|
|
|
%c1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 14)
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
; CHECK: callq
|
|
|
|
|
; This is the key check. There should NOT be any memory moves here
|
|
|
|
|
; CHECK-NOT: movq
|
2015-12-26 15:54:32 +08:00
|
|
|
%safepoint_token2 = tail call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0, i32 addrspace(1)* %c1, i32 addrspace(1)* %b1, i32 addrspace(1)* %a1)
|
|
|
|
|
%a2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 14)
|
|
|
|
|
%b2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 13)
|
|
|
|
|
%c2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 12)
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
; CHECK: callq
|
|
|
|
|
ret i32 1
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; This test simply checks that minor changes in vm state don't prevent slots
|
|
|
|
|
; being reused for gc values.
|
2015-03-27 13:09:33 +08:00
|
|
|
define i32 @reserve_first(i32 addrspace(1)* %a, i32 addrspace(1)* %b, i32 addrspace(1)* %c) #1 gc "statepoint-example" {
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
; CHECK-LABEL: reserve_first
|
|
|
|
|
; The exact stores don't matter, but there need to be three stack slots created
|
[DAG] Improve Aliasing of operations to static alloca
Re-recommiting after landing DAG extension-crash fix.
Recommiting after adding check to avoid miscomputing alias information
on addresses of the same base but different subindices.
Memory accesses offset from frame indices may alias, e.g., we
may merge write from function arguments passed on the stack when they
are contiguous. As a result, when checking aliasing, we consider the
underlying frame index's offset from the stack pointer.
Static allocs are realized as stack objects in SelectionDAG, but its
offset is not set until post-DAG causing DAGCombiner's alias check to
consider access to static allocas to frequently alias. Modify isAlias
to consider access between static allocas and access from other frame
objects to be considered aliasing.
Many test changes are included here. Most are fixes for tests which
indirectly relied on our aliasing ability and needed to be modified to
preserve their original intent.
The remaining tests have minor improvements due to relaxed
ordering. The exception is CodeGen/X86/2011-10-19-widen_vselect.ll
which has a minor degradation dispite though the pre-legalized DAG is
improved.
Reviewers: rnk, mkuper, jonpa, hfinkel, uweigand
Reviewed By: rnk
Subscribers: sdardis, nemanjai, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33345
llvm-svn: 308350
2017-07-19 04:06:24 +08:00
|
|
|
; CHECK-DAG: movq %rdi, 16(%rsp)
|
|
|
|
|
; CHECK-DAG: movq %rdx, 8(%rsp)
|
|
|
|
|
; CHECK-DAG: movq %rsi, (%rsp)
|
2015-12-26 15:54:32 +08:00
|
|
|
%safepoint_token = tail call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0, i32 addrspace(1)* %a, i32 addrspace(1)* %b, i32 addrspace(1)* %c)
|
|
|
|
|
%a1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 12)
|
|
|
|
|
%b1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 13)
|
|
|
|
|
%c1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 14)
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
; CHECK: callq
|
|
|
|
|
; This is the key check. There should NOT be any memory moves here
|
|
|
|
|
; CHECK-NOT: movq
|
2015-12-26 15:54:32 +08:00
|
|
|
%safepoint_token2 = tail call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 5, i32 addrspace(1)* %a1, i32 0, i32 addrspace(1)* %c1, i32 0, i32 0, i32 addrspace(1)* %c1, i32 addrspace(1)* %b1, i32 addrspace(1)* %a1)
|
|
|
|
|
%a2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 14)
|
|
|
|
|
%b2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 13)
|
|
|
|
|
%c2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 12)
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
; CHECK: callq
|
|
|
|
|
ret i32 1
|
|
|
|
|
}
|
|
|
|
|
|
2016-12-13 09:21:15 +08:00
|
|
|
; Check that we reuse the same stack slot across multiple calls. The use of
|
|
|
|
|
; more than two calls here is critical. We've had a bug which allowed reuse
|
|
|
|
|
; exactly once which went undetected for a long time.
|
|
|
|
|
define i32 @back_to_back_deopt(i32 %a, i32 %b, i32 %c) #1
|
|
|
|
|
gc "statepoint-example" {
|
|
|
|
|
; CHECK-LABEL: back_to_back_deopt
|
|
|
|
|
; The exact stores don't matter, but there need to be three stack slots created
|
[DAG] Improve Aliasing of operations to static alloca
Re-recommiting after landing DAG extension-crash fix.
Recommiting after adding check to avoid miscomputing alias information
on addresses of the same base but different subindices.
Memory accesses offset from frame indices may alias, e.g., we
may merge write from function arguments passed on the stack when they
are contiguous. As a result, when checking aliasing, we consider the
underlying frame index's offset from the stack pointer.
Static allocs are realized as stack objects in SelectionDAG, but its
offset is not set until post-DAG causing DAGCombiner's alias check to
consider access to static allocas to frequently alias. Modify isAlias
to consider access between static allocas and access from other frame
objects to be considered aliasing.
Many test changes are included here. Most are fixes for tests which
indirectly relied on our aliasing ability and needed to be modified to
preserve their original intent.
The remaining tests have minor improvements due to relaxed
ordering. The exception is CodeGen/X86/2011-10-19-widen_vselect.ll
which has a minor degradation dispite though the pre-legalized DAG is
improved.
Reviewers: rnk, mkuper, jonpa, hfinkel, uweigand
Reviewed By: rnk
Subscribers: sdardis, nemanjai, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33345
llvm-svn: 308350
2017-07-19 04:06:24 +08:00
|
|
|
; CHECK-DAG: movl %ebx, 12(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %ebp, 8(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %r14d, 4(%rsp)
|
2016-12-13 09:21:15 +08:00
|
|
|
; CHECK: callq
|
[DAG] Improve Aliasing of operations to static alloca
Re-recommiting after landing DAG extension-crash fix.
Recommiting after adding check to avoid miscomputing alias information
on addresses of the same base but different subindices.
Memory accesses offset from frame indices may alias, e.g., we
may merge write from function arguments passed on the stack when they
are contiguous. As a result, when checking aliasing, we consider the
underlying frame index's offset from the stack pointer.
Static allocs are realized as stack objects in SelectionDAG, but its
offset is not set until post-DAG causing DAGCombiner's alias check to
consider access to static allocas to frequently alias. Modify isAlias
to consider access between static allocas and access from other frame
objects to be considered aliasing.
Many test changes are included here. Most are fixes for tests which
indirectly relied on our aliasing ability and needed to be modified to
preserve their original intent.
The remaining tests have minor improvements due to relaxed
ordering. The exception is CodeGen/X86/2011-10-19-widen_vselect.ll
which has a minor degradation dispite though the pre-legalized DAG is
improved.
Reviewers: rnk, mkuper, jonpa, hfinkel, uweigand
Reviewed By: rnk
Subscribers: sdardis, nemanjai, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33345
llvm-svn: 308350
2017-07-19 04:06:24 +08:00
|
|
|
; CHECK-DAG: movl %ebx, 12(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %ebp, 8(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %r14d, 4(%rsp)
|
2016-12-13 09:21:15 +08:00
|
|
|
; CHECK: callq
|
[DAG] Improve Aliasing of operations to static alloca
Re-recommiting after landing DAG extension-crash fix.
Recommiting after adding check to avoid miscomputing alias information
on addresses of the same base but different subindices.
Memory accesses offset from frame indices may alias, e.g., we
may merge write from function arguments passed on the stack when they
are contiguous. As a result, when checking aliasing, we consider the
underlying frame index's offset from the stack pointer.
Static allocs are realized as stack objects in SelectionDAG, but its
offset is not set until post-DAG causing DAGCombiner's alias check to
consider access to static allocas to frequently alias. Modify isAlias
to consider access between static allocas and access from other frame
objects to be considered aliasing.
Many test changes are included here. Most are fixes for tests which
indirectly relied on our aliasing ability and needed to be modified to
preserve their original intent.
The remaining tests have minor improvements due to relaxed
ordering. The exception is CodeGen/X86/2011-10-19-widen_vselect.ll
which has a minor degradation dispite though the pre-legalized DAG is
improved.
Reviewers: rnk, mkuper, jonpa, hfinkel, uweigand
Reviewed By: rnk
Subscribers: sdardis, nemanjai, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33345
llvm-svn: 308350
2017-07-19 04:06:24 +08:00
|
|
|
; CHECK-DAG: movl %ebx, 12(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %ebp, 8(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %r14d, 4(%rsp)
|
2016-12-13 09:21:15 +08:00
|
|
|
; CHECK: callq
|
[DAG] Improve Aliasing of operations to static alloca
Re-recommiting after landing DAG extension-crash fix.
Recommiting after adding check to avoid miscomputing alias information
on addresses of the same base but different subindices.
Memory accesses offset from frame indices may alias, e.g., we
may merge write from function arguments passed on the stack when they
are contiguous. As a result, when checking aliasing, we consider the
underlying frame index's offset from the stack pointer.
Static allocs are realized as stack objects in SelectionDAG, but its
offset is not set until post-DAG causing DAGCombiner's alias check to
consider access to static allocas to frequently alias. Modify isAlias
to consider access between static allocas and access from other frame
objects to be considered aliasing.
Many test changes are included here. Most are fixes for tests which
indirectly relied on our aliasing ability and needed to be modified to
preserve their original intent.
The remaining tests have minor improvements due to relaxed
ordering. The exception is CodeGen/X86/2011-10-19-widen_vselect.ll
which has a minor degradation dispite though the pre-legalized DAG is
improved.
Reviewers: rnk, mkuper, jonpa, hfinkel, uweigand
Reviewed By: rnk
Subscribers: sdardis, nemanjai, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33345
llvm-svn: 308350
2017-07-19 04:06:24 +08:00
|
|
|
; CHECK-DAG: movl %ebx, 12(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %ebp, 8(%rsp)
|
|
|
|
|
; CHECK-DAG: movl %r14d, 4(%rsp)
|
2016-12-13 09:21:15 +08:00
|
|
|
; CHECK: callq
|
|
|
|
|
call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 3, i32 %a, i32 %b, i32 %c)
|
|
|
|
|
call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 3, i32 %a, i32 %b, i32 %c)
|
|
|
|
|
call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 3, i32 %a, i32 %b, i32 %c)
|
|
|
|
|
call token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 3, i32 %a, i32 %b, i32 %c)
|
|
|
|
|
ret i32 1
|
|
|
|
|
}
|
|
|
|
|
|
2015-06-10 20:31:53 +08:00
|
|
|
; Test that stack slots are reused for invokes
|
2015-06-18 04:52:32 +08:00
|
|
|
define i32 @back_to_back_invokes(i32 addrspace(1)* %a, i32 addrspace(1)* %b, i32 addrspace(1)* %c) #1 gc "statepoint-example" personality i32 ()* @"personality_function" {
|
2015-06-10 20:31:53 +08:00
|
|
|
; CHECK-LABEL: back_to_back_invokes
|
|
|
|
|
entry:
|
|
|
|
|
; The exact stores don't matter, but there need to be three stack slots created
|
[DAG] Improve Aliasing of operations to static alloca
Re-recommiting after landing DAG extension-crash fix.
Recommiting after adding check to avoid miscomputing alias information
on addresses of the same base but different subindices.
Memory accesses offset from frame indices may alias, e.g., we
may merge write from function arguments passed on the stack when they
are contiguous. As a result, when checking aliasing, we consider the
underlying frame index's offset from the stack pointer.
Static allocs are realized as stack objects in SelectionDAG, but its
offset is not set until post-DAG causing DAGCombiner's alias check to
consider access to static allocas to frequently alias. Modify isAlias
to consider access between static allocas and access from other frame
objects to be considered aliasing.
Many test changes are included here. Most are fixes for tests which
indirectly relied on our aliasing ability and needed to be modified to
preserve their original intent.
The remaining tests have minor improvements due to relaxed
ordering. The exception is CodeGen/X86/2011-10-19-widen_vselect.ll
which has a minor degradation dispite though the pre-legalized DAG is
improved.
Reviewers: rnk, mkuper, jonpa, hfinkel, uweigand
Reviewed By: rnk
Subscribers: sdardis, nemanjai, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33345
llvm-svn: 308350
2017-07-19 04:06:24 +08:00
|
|
|
; CHECK-DAG: movq %rdi, 16(%rsp)
|
|
|
|
|
; CHECK-DAG: movq %rdx, 8(%rsp)
|
|
|
|
|
; CHECK-DAG: movq %rsi, (%rsp)
|
2015-06-10 20:31:53 +08:00
|
|
|
; CHECK: callq
|
2015-12-26 15:54:32 +08:00
|
|
|
%safepoint_token = invoke token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0, i32 addrspace(1)* %a, i32 addrspace(1)* %b, i32 addrspace(1)* %c)
|
2015-06-10 20:31:53 +08:00
|
|
|
to label %normal_return unwind label %exceptional_return
|
|
|
|
|
|
|
|
|
|
normal_return:
|
2015-12-26 15:54:32 +08:00
|
|
|
%a1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 12)
|
|
|
|
|
%b1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 13)
|
|
|
|
|
%c1 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token, i32 12, i32 14)
|
2015-06-10 20:31:53 +08:00
|
|
|
; Should work even through bitcasts
|
|
|
|
|
%c1.casted = bitcast i32 addrspace(1)* %c1 to i8 addrspace(1)*
|
|
|
|
|
; This is the key check. There should NOT be any memory moves here
|
|
|
|
|
; CHECK-NOT: movq
|
|
|
|
|
; CHECK: callq
|
2015-12-26 15:54:32 +08:00
|
|
|
%safepoint_token2 = invoke token (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* undef, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0, i8 addrspace(1)* %c1.casted, i32 addrspace(1)* %b1, i32 addrspace(1)* %a1)
|
2015-06-10 20:31:53 +08:00
|
|
|
to label %normal_return2 unwind label %exceptional_return2
|
|
|
|
|
|
|
|
|
|
normal_return2:
|
2015-12-26 15:54:32 +08:00
|
|
|
%a2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 14)
|
|
|
|
|
%b2 = tail call coldcc i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token %safepoint_token2, i32 12, i32 13)
|
|
|
|
|
%c2 = tail call coldcc i8 addrspace(1)* @llvm.experimental.gc.relocate.p1i8(token %safepoint_token2, i32 12, i32 12)
|
2015-06-10 20:31:53 +08:00
|
|
|
ret i32 1
|
|
|
|
|
|
|
|
|
|
exceptional_return:
|
2015-06-18 04:52:32 +08:00
|
|
|
%landing_pad = landingpad { i8*, i32 }
|
2015-06-10 20:31:53 +08:00
|
|
|
cleanup
|
|
|
|
|
ret i32 0
|
|
|
|
|
|
|
|
|
|
exceptional_return2:
|
2015-06-18 04:52:32 +08:00
|
|
|
%landing_pad2 = landingpad { i8*, i32 }
|
2015-06-10 20:31:53 +08:00
|
|
|
cleanup
|
|
|
|
|
ret i32 0
|
|
|
|
|
}
|
|
|
|
|
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
; Function Attrs: nounwind
|
2015-12-26 15:54:32 +08:00
|
|
|
declare i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(token, i32, i32) #3
|
|
|
|
|
declare i8 addrspace(1)* @llvm.experimental.gc.relocate.p1i8(token, i32, i32) #3
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
|
2015-12-26 15:54:32 +08:00
|
|
|
declare token @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
|
[Statepoints 3/4] Statepoint infrastructure for garbage collection: SelectionDAGBuilder
This is the third patch in a small series. It contains the CodeGen support for lowering the gc.statepoint intrinsic sequences (223078) to the STATEPOINT pseudo machine instruction (223085). The change also includes the set of helper routines and classes for working with gc.statepoints, gc.relocates, and gc.results since the lowering code uses them.
With this change, gc.statepoints should be functionally complete. The documentation will follow in the fourth change, and there will likely be some cleanup changes, but interested parties can start experimenting now.
I'm not particularly happy with the amount of code or complexity involved with the lowering step, but at least it's fairly well isolated. The statepoint lowering code is split into it's own files and anyone not working on the statepoint support itself should be able to ignore it.
During the lowering process, we currently spill aggressively to stack. This is not entirely ideal (and we have plans to do better), but it's functional, relatively straight forward, and matches closely the implementations of the patchpoint intrinsics. Most of the complexity comes from trying to keep relocated copies of values in the same stack slots across statepoints. Doing so avoids the insertion of pointless load and store instructions to reshuffle the stack. The current implementation isn't as effective as I'd like, but it is functional and 'good enough' for many common use cases.
In the long term, I'd like to figure out how to integrate the statepoint lowering with the register allocator. In principal, we shouldn't need to eagerly spill at all. The register allocator should do any spilling required and the statepoint should simply record that fact. Depending on how challenging that turns out to be, we may invest in a smarter global stack slot assignment mechanism as a stop gap measure.
Reviewed by: atrick, ributzka
llvm-svn: 223137
2014-12-03 02:50:36 +08:00
|
|
|
|
2015-06-10 20:31:53 +08:00
|
|
|
declare i32 @"personality_function"()
|
|
|
|
|
|
|
|
|
|
attributes #1 = { uwtable }
|