RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-30 23:42:52 +01:00
Code Issues Projects Releases Wiki Activity
84c6cd222d
llvm-mirror/test/CodeGen/PowerPC/sjlj.ll

161 lines
4.9 KiB
LLVM
Raw Normal View History

Implement builtin_{setjmp/longjmp} on PPC This implements SJLJ lowering on PPC, making the Clang functions __builtin_{setjmp/longjmp} functional on PPC platforms. The implementation strategy is similar to that on X86, with the exception that a branch-and-link variant is used to get the right jump address. Credit goes to Bill Schmidt for suggesting the use of the unconditional bcl form (instead of the regular bl instruction) to limit return-address-cache pollution. Benchmarking the speed at -O3 of: static jmp_buf env_sigill; void foo() { __builtin_longjmp(env_sigill,1); } main() { ... for (int i = 0; i < c; ++i) { if (__builtin_setjmp(env_sigill)) { goto done; } else { foo(); } done:; } ... } vs. the same code using the libc setjmp/longjmp functions on a P7 shows that this builtin implementation is ~4x faster with Altivec enabled and ~7.25x faster with Altivec disabled. This comparison is somewhat unfair because the libc version must also save/restore the VSX registers which we don't yet support. llvm-svn: 177666
2013-03-21 22:37:52 +01:00
; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck %s
; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=a2 | FileCheck -check-prefix=CHECK-NOAV %s
target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64"
target triple = "powerpc64-unknown-linux-gnu"
%struct.__jmp_buf_tag = type { [64 x i64], i32, %struct.__sigset_t, [8 x i8] }
%struct.__sigset_t = type { [16 x i64] }
@env_sigill = internal global [1 x %struct.__jmp_buf_tag] zeroinitializer, align 16
define void @foo() #0 {
entry:
call void @llvm.eh.sjlj.longjmp(i8* bitcast ([1 x %struct.__jmp_buf_tag]* @env_sigill to i8*))
unreachable
; CHECK: @foo
; CHECK: addis [[REG:[0-9]+]], 2, env_sigill@toc@ha
; CHECK: addi [[REG]], [[REG]], env_sigill@toc@l
; CHECK: ld 31, 0([[REG]])
; CHECK: ld [[REG2:[0-9]+]], 8([[REG]])
; CHECK: ld 1, 16([[REG]])
; CHECK: mtctr [[REG2]]
PPC: Add base-pointer support to builtin setjmp/longjmp First, this changes the base-pointer implementation to remove an unnecessary complication (and one that is incompatible with how builtin SjLj is implemented): instead of using r31 as the base pointer when it is not needed as a frame pointer, now the base pointer will always be r30 when needed. Second, we introduce another pseudo register, BP, which is used just like the FP pseudo register to refer to the base register before we know for certain what register it will be. Third, we now save BP into the jmp_buf, and restore r30 from that slot in longjmp. If the function that called setjmp did not use a base pointer, then r30 will be overwritten by the setjmp-calling-function's restore code. FP restoration (which is restored into r31) works the same way. llvm-svn: 186545
2013-07-18 01:50:51 +02:00
; CHECK: ld 30, 32([[REG]])
Implement builtin_{setjmp/longjmp} on PPC This implements SJLJ lowering on PPC, making the Clang functions __builtin_{setjmp/longjmp} functional on PPC platforms. The implementation strategy is similar to that on X86, with the exception that a branch-and-link variant is used to get the right jump address. Credit goes to Bill Schmidt for suggesting the use of the unconditional bcl form (instead of the regular bl instruction) to limit return-address-cache pollution. Benchmarking the speed at -O3 of: static jmp_buf env_sigill; void foo() { __builtin_longjmp(env_sigill,1); } main() { ... for (int i = 0; i < c; ++i) { if (__builtin_setjmp(env_sigill)) { goto done; } else { foo(); } done:; } ... } vs. the same code using the libc setjmp/longjmp functions on a P7 shows that this builtin implementation is ~4x faster with Altivec enabled and ~7.25x faster with Altivec disabled. This comparison is somewhat unfair because the libc version must also save/restore the VSX registers which we don't yet support. llvm-svn: 177666
2013-03-21 22:37:52 +01:00
; CHECK: ld 2, 24([[REG]])
; CHECK: bctr
return: ; No predecessors!
ret void
}
declare void @llvm.eh.sjlj.longjmp(i8*) #1
define signext i32 @main() #0 {
entry:
%retval = alloca i32, align 4
store i32 0, i32* %retval
%0 = call i8* @llvm.frameaddress(i32 0)
store i8* %0, i8** bitcast ([1 x %struct.__jmp_buf_tag]* @env_sigill to i8**)
%1 = call i8* @llvm.stacksave()
store i8* %1, i8** getelementptr (i8** bitcast ([1 x %struct.__jmp_buf_tag]* @env_sigill to i8**), i32 2)
%2 = call i32 @llvm.eh.sjlj.setjmp(i8* bitcast ([1 x %struct.__jmp_buf_tag]* @env_sigill to i8*))
%tobool = icmp ne i32 %2, 0
br i1 %tobool, label %if.then, label %if.else
if.then: ; preds = %entry
store i32 1, i32* %retval
br label %return
if.else: ; preds = %entry
call void @foo()
br label %if.end
if.end: ; preds = %if.else
store i32 0, i32* %retval
br label %return
return: ; preds = %if.end, %if.then
%3 = load i32* %retval
ret i32 %3
Don't spill PPC VRSAVE on non-Darwin (even in SjLj) As Bill Schmidt pointed out to me, only on Darwin do we need to spill/restore VRSAVE in the SjLj code. For non-Darwin, don't spill/restore VRSAVE (and I've added some asserts to make sure that we're not). As it turns out, we're not currently handling the Darwin case correctly (I've added a FIXME in the test case). I've tried adding various implied register definitions/uses to force the spill without success, so I'll need to address this later. llvm-svn: 178096
2013-03-27 01:02:20 +01:00
; FIXME: We should be saving VRSAVE on Darwin, but we're not!
Implement builtin_{setjmp/longjmp} on PPC This implements SJLJ lowering on PPC, making the Clang functions __builtin_{setjmp/longjmp} functional on PPC platforms. The implementation strategy is similar to that on X86, with the exception that a branch-and-link variant is used to get the right jump address. Credit goes to Bill Schmidt for suggesting the use of the unconditional bcl form (instead of the regular bl instruction) to limit return-address-cache pollution. Benchmarking the speed at -O3 of: static jmp_buf env_sigill; void foo() { __builtin_longjmp(env_sigill,1); } main() { ... for (int i = 0; i < c; ++i) { if (__builtin_setjmp(env_sigill)) { goto done; } else { foo(); } done:; } ... } vs. the same code using the libc setjmp/longjmp functions on a P7 shows that this builtin implementation is ~4x faster with Altivec enabled and ~7.25x faster with Altivec disabled. This comparison is somewhat unfair because the libc version must also save/restore the VSX registers which we don't yet support. llvm-svn: 177666
2013-03-21 22:37:52 +01:00
; CHECK: @main
; CHECK: std
Don't spill PPC VRSAVE on non-Darwin (even in SjLj) As Bill Schmidt pointed out to me, only on Darwin do we need to spill/restore VRSAVE in the SjLj code. For non-Darwin, don't spill/restore VRSAVE (and I've added some asserts to make sure that we're not). As it turns out, we're not currently handling the Darwin case correctly (I've added a FIXME in the test case). I've tried adding various implied register definitions/uses to force the spill without success, so I'll need to address this later. llvm-svn: 178096
2013-03-27 01:02:20 +01:00
; Make sure that we're not saving VRSAVE on non-Darwin:
; CHECK-NOT: mfspr
Implement builtin_{setjmp/longjmp} on PPC This implements SJLJ lowering on PPC, making the Clang functions __builtin_{setjmp/longjmp} functional on PPC platforms. The implementation strategy is similar to that on X86, with the exception that a branch-and-link variant is used to get the right jump address. Credit goes to Bill Schmidt for suggesting the use of the unconditional bcl form (instead of the regular bl instruction) to limit return-address-cache pollution. Benchmarking the speed at -O3 of: static jmp_buf env_sigill; void foo() { __builtin_longjmp(env_sigill,1); } main() { ... for (int i = 0; i < c; ++i) { if (__builtin_setjmp(env_sigill)) { goto done; } else { foo(); } done:; } ... } vs. the same code using the libc setjmp/longjmp functions on a P7 shows that this builtin implementation is ~4x faster with Altivec enabled and ~7.25x faster with Altivec disabled. This comparison is somewhat unfair because the libc version must also save/restore the VSX registers which we don't yet support. llvm-svn: 177666
2013-03-21 22:37:52 +01:00
PPC: Enable aggressive anti-dependency breaking Aggressive anti-dependency breaking is enabled by default for all PPC cores. This provides a general speedup on the P7 and other platforms (among other factors, the instruction group formation for the non-embedded PPC cores is done during post-RA scheduling). In order to do this safely, the incompatibility between uses of the MFOCRF instruction and anti-dependency breaking are resolved by marking MFOCRF with hasExtraSrcRegAllocReq. As noted in the removed FIXME, the problem was that MFOCRF's output is sensitive to the identify of the source register, and always paired with a shift to undo this effect. Because anti-dependency breaking is unaware of this hidden dependency of the shift amount on the source register of the MFOCRF instruction, changing that register must be inhibited. Two test cases were adjusted: The SjLj test was made more insensitive to register choices and scheduling; the saveCR test disabled anti-dependency breaking because part of what it is testing is proper register reuse. llvm-svn: 190587
2013-09-12 07:24:49 +02:00
; CHECK-DAG: stfd
; CHECK-DAG: stvx
; CHECK-DAG: addis [[REG:[0-9]+]], 2, env_sigill@toc@ha
; CHECK-DAG: std 31, env_sigill@toc@l([[REG]])
; CHECK-DAG: addi [[REGA:[0-9]+]], [[REG]], env_sigill@toc@l
; CHECK-DAG: std [[REGA]], [[OFF:[0-9]+]](31) # 8-byte Folded Spill
; CHECK-DAG: std 1, 16([[REGA]])
; CHECK-DAG: std 2, 24([[REGA]])
Implement builtin_{setjmp/longjmp} on PPC This implements SJLJ lowering on PPC, making the Clang functions __builtin_{setjmp/longjmp} functional on PPC platforms. The implementation strategy is similar to that on X86, with the exception that a branch-and-link variant is used to get the right jump address. Credit goes to Bill Schmidt for suggesting the use of the unconditional bcl form (instead of the regular bl instruction) to limit return-address-cache pollution. Benchmarking the speed at -O3 of: static jmp_buf env_sigill; void foo() { __builtin_longjmp(env_sigill,1); } main() { ... for (int i = 0; i < c; ++i) { if (__builtin_setjmp(env_sigill)) { goto done; } else { foo(); } done:; } ... } vs. the same code using the libc setjmp/longjmp functions on a P7 shows that this builtin implementation is ~4x faster with Altivec enabled and ~7.25x faster with Altivec disabled. This comparison is somewhat unfair because the libc version must also save/restore the VSX registers which we don't yet support. llvm-svn: 177666
2013-03-21 22:37:52 +01:00
; CHECK: bcl 20, 31, .LBB1_1
; CHECK: li 3, 1
; CHECK: #EH_SjLj_Setup .LBB1_1
; CHECK: b .LBB1_2
; CHECK: .LBB1_1:
; CHECK: mflr [[REGL:[0-9]+]]
; CHECK: ld [[REG2:[0-9]+]], [[OFF]](31) # 8-byte Folded Reload
; CHECK: std [[REGL]], 8([[REG2]])
; CHECK: li 3, 0
; CHECK: .LBB1_2:
; CHECK: lfd
; CHECK: lvx
; CHECK: ld
; CHECK: blr
; CHECK-NOAV: @main
; CHECK-NOAV-NOT: stvx
; CHECK-NOAV: bcl
; CHECK-NOAV: mflr
; CHECK-NOAV: bl foo
; CHECK-NOAV-NOT: lvx
; CHECK-NOAV: blr
}
PPC: Add base-pointer support to builtin setjmp/longjmp First, this changes the base-pointer implementation to remove an unnecessary complication (and one that is incompatible with how builtin SjLj is implemented): instead of using r31 as the base pointer when it is not needed as a frame pointer, now the base pointer will always be r30 when needed. Second, we introduce another pseudo register, BP, which is used just like the FP pseudo register to refer to the base register before we know for certain what register it will be. Third, we now save BP into the jmp_buf, and restore r30 from that slot in longjmp. If the function that called setjmp did not use a base pointer, then r30 will be overwritten by the setjmp-calling-function's restore code. FP restoration (which is restored into r31) works the same way. llvm-svn: 186545
2013-07-18 01:50:51 +02:00
define signext i32 @main2() #0 {
entry:
%a = alloca i8, align 64
call void @bar(i8* %a)
%retval = alloca i32, align 4
store i32 0, i32* %retval
%0 = call i8* @llvm.frameaddress(i32 0)
store i8* %0, i8** bitcast ([1 x %struct.__jmp_buf_tag]* @env_sigill to i8**)
%1 = call i8* @llvm.stacksave()
store i8* %1, i8** getelementptr (i8** bitcast ([1 x %struct.__jmp_buf_tag]* @env_sigill to i8**), i32 2)
%2 = call i32 @llvm.eh.sjlj.setjmp(i8* bitcast ([1 x %struct.__jmp_buf_tag]* @env_sigill to i8*))
%tobool = icmp ne i32 %2, 0
br i1 %tobool, label %if.then, label %if.else
if.then: ; preds = %entry
store i32 1, i32* %retval
br label %return
if.else: ; preds = %entry
call void @foo()
br label %if.end
if.end: ; preds = %if.else
store i32 0, i32* %retval
br label %return
return: ; preds = %if.end, %if.then
%3 = load i32* %retval
ret i32 %3
; CHECK: @main2
; CHECK: addis [[REG:[0-9]+]], 2, env_sigill@toc@ha
; CHECK: std 31, env_sigill@toc@l([[REG]])
PPC: Enable aggressive anti-dependency breaking Aggressive anti-dependency breaking is enabled by default for all PPC cores. This provides a general speedup on the P7 and other platforms (among other factors, the instruction group formation for the non-embedded PPC cores is done during post-RA scheduling). In order to do this safely, the incompatibility between uses of the MFOCRF instruction and anti-dependency breaking are resolved by marking MFOCRF with hasExtraSrcRegAllocReq. As noted in the removed FIXME, the problem was that MFOCRF's output is sensitive to the identify of the source register, and always paired with a shift to undo this effect. Because anti-dependency breaking is unaware of this hidden dependency of the shift amount on the source register of the MFOCRF instruction, changing that register must be inhibited. Two test cases were adjusted: The SjLj test was made more insensitive to register choices and scheduling; the saveCR test disabled anti-dependency breaking because part of what it is testing is proper register reuse. llvm-svn: 190587
2013-09-12 07:24:49 +02:00
; CHECK: addi [[REGB:[0-9]+]], [[REG]], env_sigill@toc@l
; CHECK-DAG: std [[REGB]], [[OFF:[0-9]+]](31) # 8-byte Folded Spill
; CHECK-DAG: std 1, 16([[REGB]])
; CHECK-DAG: std 2, 24([[REGB]])
; CHECK-DAG: std 30, 32([[REGB]])
PPC: Add base-pointer support to builtin setjmp/longjmp First, this changes the base-pointer implementation to remove an unnecessary complication (and one that is incompatible with how builtin SjLj is implemented): instead of using r31 as the base pointer when it is not needed as a frame pointer, now the base pointer will always be r30 when needed. Second, we introduce another pseudo register, BP, which is used just like the FP pseudo register to refer to the base register before we know for certain what register it will be. Third, we now save BP into the jmp_buf, and restore r30 from that slot in longjmp. If the function that called setjmp did not use a base pointer, then r30 will be overwritten by the setjmp-calling-function's restore code. FP restoration (which is restored into r31) works the same way. llvm-svn: 186545
2013-07-18 01:50:51 +02:00
; CHECK: bcl 20, 31,
; CHECK: blr
}
declare void @bar(i8*) #3
Implement builtin_{setjmp/longjmp} on PPC This implements SJLJ lowering on PPC, making the Clang functions __builtin_{setjmp/longjmp} functional on PPC platforms. The implementation strategy is similar to that on X86, with the exception that a branch-and-link variant is used to get the right jump address. Credit goes to Bill Schmidt for suggesting the use of the unconditional bcl form (instead of the regular bl instruction) to limit return-address-cache pollution. Benchmarking the speed at -O3 of: static jmp_buf env_sigill; void foo() { __builtin_longjmp(env_sigill,1); } main() { ... for (int i = 0; i < c; ++i) { if (__builtin_setjmp(env_sigill)) { goto done; } else { foo(); } done:; } ... } vs. the same code using the libc setjmp/longjmp functions on a P7 shows that this builtin implementation is ~4x faster with Altivec enabled and ~7.25x faster with Altivec disabled. This comparison is somewhat unfair because the libc version must also save/restore the VSX registers which we don't yet support. llvm-svn: 177666
2013-03-21 22:37:52 +01:00
declare i8* @llvm.frameaddress(i32) #2
declare i8* @llvm.stacksave() #3
declare i32 @llvm.eh.sjlj.setjmp(i8*) #3
Update to remove the no-frame-pointer-elim-non-leaf flag if it was set to 'false'. llvm-svn: 189068
2013-08-22 23:28:54 +02:00
attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "unsafe-fp-math"="false" "use-soft-float"="false" }
Implement builtin_{setjmp/longjmp} on PPC This implements SJLJ lowering on PPC, making the Clang functions __builtin_{setjmp/longjmp} functional on PPC platforms. The implementation strategy is similar to that on X86, with the exception that a branch-and-link variant is used to get the right jump address. Credit goes to Bill Schmidt for suggesting the use of the unconditional bcl form (instead of the regular bl instruction) to limit return-address-cache pollution. Benchmarking the speed at -O3 of: static jmp_buf env_sigill; void foo() { __builtin_longjmp(env_sigill,1); } main() { ... for (int i = 0; i < c; ++i) { if (__builtin_setjmp(env_sigill)) { goto done; } else { foo(); } done:; } ... } vs. the same code using the libc setjmp/longjmp functions on a P7 shows that this builtin implementation is ~4x faster with Altivec enabled and ~7.25x faster with Altivec disabled. This comparison is somewhat unfair because the libc version must also save/restore the VSX registers which we don't yet support. llvm-svn: 177666
2013-03-21 22:37:52 +01:00
attributes #1 = { noreturn nounwind }
attributes #2 = { nounwind readnone }
attributes #3 = { nounwind }
Reference in New Issue Copy Permalink