maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/X86/frameaddr.ll

84 lines
2.9 KiB
LLVM
Raw Normal View History

[FastISel][X86] Add support for the frameaddress intrinsic. llvm-svn: 210709
2014-06-12 05:44:44 +08:00
; RUN: llc < %s -march=x86 | FileCheck %s --check-prefix=CHECK-32
Change the fast-isel-abort option from bool to int to enable "levels" Summary: Currently fast-isel-abort will only abort for regular instructions, and just warn for function calls, terminators, function arguments. There is already fast-isel-abort-args but nothing for calls and terminators. This change turns the fast-isel-abort options into an integer option, so that multiple levels of strictness can be defined. This will help no being surprised when the "abort" option indeed does not abort, and enables the possibility to write test that verifies that no intrinsics are forgotten by fast-isel. Reviewers: resistor, echristo Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D7941 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 230775
2015-02-28 02:32:11 +08:00
; RUN: llc < %s -march=x86 -fast-isel -fast-isel-abort=1 | FileCheck %s --check-prefix=CHECK-32
X86: @llvm.frameaddress should defer to SelectionDAG for Win CFI llvm-svn: 228754
2015-02-11 06:00:34 +08:00
; RUN: llc < %s -mtriple=x86_64-pc-win32 -fast-isel | FileCheck %s --check-prefix=CHECK-W64
X86: Emit an ABI compliant prologue and epilogue for Win64 Win64 has specific contraints on what valid prologues and epilogues look like. This constraint is born from the flexibility and descriptiveness of Win64's unwind opcodes. Prologues previously emitted by LLVM could not be represented by the unwind opcodes, preventing operations powered by stack unwinding to successfully work. Differential Revision: http://reviews.llvm.org/D7520 llvm-svn: 228641
2015-02-10 08:57:42 +08:00
; RUN: llc < %s -mtriple=x86_64-unknown | FileCheck %s --check-prefix=CHECK-64
Change the fast-isel-abort option from bool to int to enable "levels" Summary: Currently fast-isel-abort will only abort for regular instructions, and just warn for function calls, terminators, function arguments. There is already fast-isel-abort-args but nothing for calls and terminators. This change turns the fast-isel-abort options into an integer option, so that multiple levels of strictness can be defined. This will help no being surprised when the "abort" option indeed does not abort, and enables the possibility to write test that verifies that no intrinsics are forgotten by fast-isel. Reviewers: resistor, echristo Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D7941 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 230775
2015-02-28 02:32:11 +08:00
; RUN: llc < %s -mtriple=x86_64-unknown -fast-isel -fast-isel-abort=1 | FileCheck %s --check-prefix=CHECK-64
[x32] Use ebp/esp as frame and stack pointer Summary: Since pointers are 32-bit on x32 we can use ebp and esp as frame and stack pointer. Some operations like PUSH/POP and CFI_INSTRUCTION still require 64-bit register, so using 64-bit MachineFramePtr where required. X86_64 NaCl uses 64-bit frame/stack pointers, however it's been found that both isTarget64BitLP64 and isTarget64BitILP32 are true for NaCl. Addressing this issue here as well by making isTarget64BitLP64 false. Also mark hasReservedSpillSlot unreachable on X86. See inlined comments. Test Plan: Add one new simple test and upgrade 2 existing with x32 target case. Reviewers: nadav, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D4617 llvm-svn: 215091
2014-08-07 17:41:19 +08:00
; RUN: llc < %s -mtriple=x86_64-gnux32 | FileCheck %s --check-prefix=CHECK-X32ABI
Change the fast-isel-abort option from bool to int to enable "levels" Summary: Currently fast-isel-abort will only abort for regular instructions, and just warn for function calls, terminators, function arguments. There is already fast-isel-abort-args but nothing for calls and terminators. This change turns the fast-isel-abort options into an integer option, so that multiple levels of strictness can be defined. This will help no being surprised when the "abort" option indeed does not abort, and enables the possibility to write test that verifies that no intrinsics are forgotten by fast-isel. Reviewers: resistor, echristo Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D7941 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 230775
2015-02-28 02:32:11 +08:00
; RUN: llc < %s -mtriple=x86_64-gnux32 -fast-isel -fast-isel-abort=1 | FileCheck %s --check-prefix=CHECK-X32ABI
Use 32-bit ebp for NaCl64 in a limited case: llvm.frameaddress. Summary: Follow up to [x32] "Use ebp/esp as frame and stack pointer": http://reviews.llvm.org/D4617 In that earlier patch, NaCl64 was made to always use rbp. That's needed for most cases because rbp should hold a full 64-bit address within the NaCl sandbox so that load/stores off of rbp don't require sandbox adjustment (zeroing the top 32-bits, then filling those by adding r15). However, llvm.frameaddress returns a pointer and pointers are 32-bit for NaCl64. In this case, use ebp instead, which will make the register copy type check. A similar mechanism may be needed for llvm.eh.return, but is not added in this change. Test Plan: test/CodeGen/X86/frameaddr.ll Reviewers: dschuff, nadav Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D6514 llvm-svn: 223510
2014-12-06 04:55:53 +08:00
; RUN: llc < %s -mtriple=x86_64-nacl | FileCheck %s --check-prefix=CHECK-NACL64
Change the fast-isel-abort option from bool to int to enable "levels" Summary: Currently fast-isel-abort will only abort for regular instructions, and just warn for function calls, terminators, function arguments. There is already fast-isel-abort-args but nothing for calls and terminators. This change turns the fast-isel-abort options into an integer option, so that multiple levels of strictness can be defined. This will help no being surprised when the "abort" option indeed does not abort, and enables the possibility to write test that verifies that no intrinsics are forgotten by fast-isel. Reviewers: resistor, echristo Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D7941 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 230775
2015-02-28 02:32:11 +08:00
; RUN: llc < %s -mtriple=x86_64-nacl -fast-isel -fast-isel-abort=1 | FileCheck %s --check-prefix=CHECK-NACL64
[FastISel][X86] Add support for the frameaddress intrinsic. llvm-svn: 210709
2014-06-12 05:44:44 +08:00
define i8* @test1() nounwind {
entry:
[x86] Improve frameaddress test from r210709. llvm-svn: 210743
2014-06-12 11:29:29 +08:00
; CHECK-32-LABEL: test1
; CHECK-32: push
; CHECK-32-NEXT: movl %esp, %ebp
; CHECK-32-NEXT: movl %ebp, %eax
; CHECK-32-NEXT: pop
; CHECK-32-NEXT: ret
X86: @llvm.frameaddress should defer to SelectionDAG for Win CFI llvm-svn: 228754
2015-02-11 06:00:34 +08:00
; CHECK-W64-LABEL: test1
; CHECK-W64: push
X86: Use 'mov' instead of 'lea' in Win64 SEH prologues when possible 'mov' and 'lea' are equivalent when the displacement applied with 'lea' is zero. However, 'mov' should encode smaller. llvm-svn: 230269
2015-02-24 05:50:27 +08:00
; CHECK-W64-NEXT: movq %rsp, %rbp
[X86] Optimization for replacing LEA with MOV at frame index elimination time Summary: Replace a LEA instruction of the form 'lea (%esp), %ebx' --> 'mov %esp, %ebx' MOV is preferable over LEA because usually there are more issue-slots available to execute MOVs than LEAs. Latest processors also support zero-latency MOVs. Fixes pr29022. Reviewers: hfinkel, delena, igorb, myatsina, mkuper Differential Revision: https://reviews.llvm.org/D24705 llvm-svn: 282385
2016-09-26 14:42:07 +08:00
; CHECK-W64-NEXT: movq %rbp, %rax
X86: @llvm.frameaddress should defer to SelectionDAG for Win CFI llvm-svn: 228754
2015-02-11 06:00:34 +08:00
; CHECK-W64-NEXT: pop
; CHECK-W64-NEXT: ret
[x86] Improve frameaddress test from r210709. llvm-svn: 210743
2014-06-12 11:29:29 +08:00
; CHECK-64-LABEL: test1
; CHECK-64: push
; CHECK-64-NEXT: movq %rsp, %rbp
; CHECK-64-NEXT: movq %rbp, %rax
; CHECK-64-NEXT: pop
; CHECK-64-NEXT: ret
[x32] Use ebp/esp as frame and stack pointer Summary: Since pointers are 32-bit on x32 we can use ebp and esp as frame and stack pointer. Some operations like PUSH/POP and CFI_INSTRUCTION still require 64-bit register, so using 64-bit MachineFramePtr where required. X86_64 NaCl uses 64-bit frame/stack pointers, however it's been found that both isTarget64BitLP64 and isTarget64BitILP32 are true for NaCl. Addressing this issue here as well by making isTarget64BitLP64 false. Also mark hasReservedSpillSlot unreachable on X86. See inlined comments. Test Plan: Add one new simple test and upgrade 2 existing with x32 target case. Reviewers: nadav, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D4617 llvm-svn: 215091
2014-08-07 17:41:19 +08:00
; CHECK-X32ABI-LABEL: test1
; CHECK-X32ABI: pushq %rbp
; CHECK-X32ABI-NEXT: movl %esp, %ebp
; CHECK-X32ABI-NEXT: movl %ebp, %eax
; CHECK-X32ABI-NEXT: popq %rbp
; CHECK-X32ABI-NEXT: ret
Use 32-bit ebp for NaCl64 in a limited case: llvm.frameaddress. Summary: Follow up to [x32] "Use ebp/esp as frame and stack pointer": http://reviews.llvm.org/D4617 In that earlier patch, NaCl64 was made to always use rbp. That's needed for most cases because rbp should hold a full 64-bit address within the NaCl sandbox so that load/stores off of rbp don't require sandbox adjustment (zeroing the top 32-bits, then filling those by adding r15). However, llvm.frameaddress returns a pointer and pointers are 32-bit for NaCl64. In this case, use ebp instead, which will make the register copy type check. A similar mechanism may be needed for llvm.eh.return, but is not added in this change. Test Plan: test/CodeGen/X86/frameaddr.ll Reviewers: dschuff, nadav Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D6514 llvm-svn: 223510
2014-12-06 04:55:53 +08:00
; CHECK-NACL64-LABEL: test1
; CHECK-NACL64: pushq %rbp
; CHECK-NACL64-NEXT: movq %rsp, %rbp
; CHECK-NACL64-NEXT: movl %ebp, %eax
[FastISel][X86] Add support for the frameaddress intrinsic. llvm-svn: 210709
2014-06-12 05:44:44 +08:00
%0 = tail call i8* @llvm.frameaddress(i32 0)
ret i8* %0
}
define i8* @test2() nounwind {
entry:
[x86] Improve frameaddress test from r210709. llvm-svn: 210743
2014-06-12 11:29:29 +08:00
; CHECK-32-LABEL: test2
; CHECK-32: push
; CHECK-32-NEXT: movl %esp, %ebp
; CHECK-32-NEXT: movl (%ebp), %eax
; CHECK-32-NEXT: movl (%eax), %eax
; CHECK-32-NEXT: pop
; CHECK-32-NEXT: ret
X86: @llvm.frameaddress should defer to SelectionDAG for Win CFI llvm-svn: 228754
2015-02-11 06:00:34 +08:00
; CHECK-W64-LABEL: test2
; CHECK-W64: push
X86: Use 'mov' instead of 'lea' in Win64 SEH prologues when possible 'mov' and 'lea' are equivalent when the displacement applied with 'lea' is zero. However, 'mov' should encode smaller. llvm-svn: 230269
2015-02-24 05:50:27 +08:00
; CHECK-W64-NEXT: movq %rsp, %rbp
[X86] Optimization for replacing LEA with MOV at frame index elimination time Summary: Replace a LEA instruction of the form 'lea (%esp), %ebx' --> 'mov %esp, %ebx' MOV is preferable over LEA because usually there are more issue-slots available to execute MOVs than LEAs. Latest processors also support zero-latency MOVs. Fixes pr29022. Reviewers: hfinkel, delena, igorb, myatsina, mkuper Differential Revision: https://reviews.llvm.org/D24705 llvm-svn: 282385
2016-09-26 14:42:07 +08:00
; CHECK-W64-NEXT: movq %rbp, %rax
X86: @llvm.frameaddress should defer to SelectionDAG for Win CFI llvm-svn: 228754
2015-02-11 06:00:34 +08:00
; CHECK-W64-NEXT: pop
; CHECK-W64-NEXT: ret
[x86] Improve frameaddress test from r210709. llvm-svn: 210743
2014-06-12 11:29:29 +08:00
; CHECK-64-LABEL: test2
; CHECK-64: push
; CHECK-64-NEXT: movq %rsp, %rbp
; CHECK-64-NEXT: movq (%rbp), %rax
; CHECK-64-NEXT: movq (%rax), %rax
; CHECK-64-NEXT: pop
; CHECK-64-NEXT: ret
[x32] Use ebp/esp as frame and stack pointer Summary: Since pointers are 32-bit on x32 we can use ebp and esp as frame and stack pointer. Some operations like PUSH/POP and CFI_INSTRUCTION still require 64-bit register, so using 64-bit MachineFramePtr where required. X86_64 NaCl uses 64-bit frame/stack pointers, however it's been found that both isTarget64BitLP64 and isTarget64BitILP32 are true for NaCl. Addressing this issue here as well by making isTarget64BitLP64 false. Also mark hasReservedSpillSlot unreachable on X86. See inlined comments. Test Plan: Add one new simple test and upgrade 2 existing with x32 target case. Reviewers: nadav, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D4617 llvm-svn: 215091
2014-08-07 17:41:19 +08:00
; CHECK-X32ABI-LABEL: test2
; CHECK-X32ABI: pushq %rbp
; CHECK-X32ABI-NEXT: movl %esp, %ebp
; CHECK-X32ABI-NEXT: movl (%ebp), %eax
; CHECK-X32ABI-NEXT: movl (%eax), %eax
; CHECK-X32ABI-NEXT: popq %rbp
; CHECK-X32ABI-NEXT: ret
Use 32-bit ebp for NaCl64 in a limited case: llvm.frameaddress. Summary: Follow up to [x32] "Use ebp/esp as frame and stack pointer": http://reviews.llvm.org/D4617 In that earlier patch, NaCl64 was made to always use rbp. That's needed for most cases because rbp should hold a full 64-bit address within the NaCl sandbox so that load/stores off of rbp don't require sandbox adjustment (zeroing the top 32-bits, then filling those by adding r15). However, llvm.frameaddress returns a pointer and pointers are 32-bit for NaCl64. In this case, use ebp instead, which will make the register copy type check. A similar mechanism may be needed for llvm.eh.return, but is not added in this change. Test Plan: test/CodeGen/X86/frameaddr.ll Reviewers: dschuff, nadav Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D6514 llvm-svn: 223510
2014-12-06 04:55:53 +08:00
; CHECK-NACL64-LABEL: test2
; CHECK-NACL64: pushq %rbp
; CHECK-NACL64-NEXT: movq %rsp, %rbp
; CHECK-NACL64-NEXT: movl (%ebp), %eax
; CHECK-NACL64-NEXT: movl (%eax), %eax
[FastISel][X86] Add support for the frameaddress intrinsic. llvm-svn: 210709
2014-06-12 05:44:44 +08:00
%0 = tail call i8* @llvm.frameaddress(i32 2)
ret i8* %0
}
[x86] Improve frameaddress test from r210709. llvm-svn: 210743
2014-06-12 11:29:29 +08:00
[FastISel][X86] Add support for the frameaddress intrinsic. llvm-svn: 210709
2014-06-12 05:44:44 +08:00
declare i8* @llvm.frameaddress(i32) nounwind readnone