97 changes: 54 additions & 43 deletions llvm/test/CodeGen/PowerPC/mcm-obj.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,12 @@
; RUN: llc -O0 -mcpu=pwr7 -code-model=large -filetype=obj -fast-isel=false %s -o - | \
; RUN: llvm-readobj -r | FileCheck -check-prefix=LARGE %s

; Run jump table test separately since jump tables aren't generated at -O0.
; RUN: llc -mcpu=pwr7 -code-model=medium -filetype=obj -fast-isel=false %s -o - | \
; RUN: llvm-readobj -r | FileCheck -check-prefix=MEDIUM-JT %s
; RUN: llc -mcpu=pwr7 -code-model=large -filetype=obj -fast-isel=false %s -o - | \
; RUN: llvm-readobj -r | FileCheck -check-prefix=LARGE-JT %s

; FIXME: When asm-parse is available, could make this an assembly test.

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"
Expand Down Expand Up @@ -92,6 +98,46 @@ entry:
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM4:[^ ]+]]
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM4]]

@ti = common global i32 0, align 4

define signext i32 @test_tentative() nounwind {
entry:
%0 = load i32, i32* @ti, align 4
%inc = add nsw i32 %0, 1
store i32 %inc, i32* @ti, align 4
ret i32 %0
}

; Verify generation of R_PPC64_TOC16_HA and R_PPC64_TOC16_LO_DS for
; accessing tentatively declared variable ti.
;
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM6:[^ ]+]]
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM6]]
;
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM6:[^ ]+]]
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM6]]

define i8* @test_fnaddr() nounwind {
entry:
%func = alloca i32 (i32)*, align 8
store i32 (i32)* @foo, i32 (i32)** %func, align 8
%0 = load i32 (i32)*, i32 (i32)** %func, align 8
%1 = bitcast i32 (i32)* %0 to i8*
ret i8* %1
}

declare signext i32 @foo(i32 signext)

; Verify generation of R_PPC64_TOC16_HA and R_PPC64_TOC16_LO_DS for
; accessing function address foo.
;
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM7:[^ ]+]]
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM7]]
;
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM7:[^ ]+]]
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM7]]


define signext i32 @test_jump_table(i32 signext %i) nounwind {
entry:
%i.addr = alloca i32, align 4
Expand Down Expand Up @@ -139,47 +185,12 @@ sw.epilog: ; preds = %sw.bb3, %sw.default
; Verify generation of R_PPC64_TOC16_HA and R_PPC64_TOC16_LO_DS for
; accessing a jump table address.
;
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM5:[^ ]+]]
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM5]]
;
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM5:[^ ]+]]
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM5]]

@ti = common global i32 0, align 4

define signext i32 @test_tentative() nounwind {
entry:
%0 = load i32, i32* @ti, align 4
%inc = add nsw i32 %0, 1
store i32 %inc, i32* @ti, align 4
ret i32 %0
}

; Verify generation of R_PPC64_TOC16_HA and R_PPC64_TOC16_LO_DS for
; accessing tentatively declared variable ti.
;
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM6:[^ ]+]]
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM6]]
;
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM6:[^ ]+]]
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM6]]

define i8* @test_fnaddr() nounwind {
entry:
%func = alloca i32 (i32)*, align 8
store i32 (i32)* @foo, i32 (i32)** %func, align 8
%0 = load i32 (i32)*, i32 (i32)** %func, align 8
%1 = bitcast i32 (i32)* %0 to i8*
ret i8* %1
}

declare signext i32 @foo(i32 signext)

; Verify generation of R_PPC64_TOC16_HA and R_PPC64_TOC16_LO_DS for
; accessing function address foo.
;
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM7:[^ ]+]]
; MEDIUM-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM7]]
; MEDIUM-JT: Relocations [
; MEDIUM-JT: Section (2) .rela.text {
; MEDIUM-JT-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM:[^ ]+]]
; MEDIUM-JT-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM]]
;
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM7:[^ ]+]]
; LARGE-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM7]]
; LARGE-JT: Relocations [
; LARGE-JT: Section (2) .rela.text {
; LARGE-JT-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_HA [[SYM:[^ ]+]]
; LARGE-JT-NEXT: 0x{{[0-9,A-F]+}} R_PPC64_TOC16_LO_DS [[SYM]]
4 changes: 3 additions & 1 deletion llvm/test/CodeGen/X86/pic_jumptable.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -55,13 +55,15 @@ entry:
]

bb: ; preds = %entry, %entry, %entry, %entry, %entry, %entry, %entry, %entry, %entry, %entry
call void @_Z3bari( i32 0 )
br label %bb1

bb1: ; preds = %bb, %entry
call void @_Z3bari( i32 1 )
br label %bb2

bb2: ; preds = %bb1, %entry
call void @_Z3bari( i32 1 )
call void @_Z3bari( i32 2 )
br label %bb11

bb3: ; preds = %entry
Expand Down
12 changes: 5 additions & 7 deletions llvm/test/CodeGen/X86/switch-bt.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -140,19 +140,17 @@ sw.epilog:

; The balanced binary switch here would start with a comparison against 39, but
; it is currently starting with 29 because of the density-sum heuristic.
; CHECK: cmpl $29
; CHECK: cmpl $39
; CHECK: jg
; CHECK: cmpl $10
; CHECK: jne
; CHECK: cmpl $49
; CHECK: jg
; CHECK: cmpl $30
; CHECK: jne
; CHECK: je
; CHECK: cmpl $20
; CHECK: jne
; CHECK: cmpl $40
; CHECK: je
; CHECK: cmpl $50
; CHECK: jne
; CHECK: cmpl $40
; CHECK: cmpl $30
; CHECK: jne
; CHECK: cmpl $60
; CHECK: jne
Expand Down
288 changes: 288 additions & 0 deletions llvm/test/CodeGen/X86/switch.ll
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,288 @@
; RUN: llc -march=x86-64 %s -o - | FileCheck %s
; RUN: llc -march=x86-64 %s -o - -O0 | FileCheck --check-prefix=NOOPT %s

declare void @g(i32)

define void @basic(i32 %x) {
entry:
switch i32 %x, label %return [
i32 3, label %bb0
i32 1, label %bb1
i32 4, label %bb1
i32 5, label %bb0
]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
return: ret void

; Should be lowered as straight compares in -O0 mode.
; NOOPT-LABEL: basic
; NOOPT: subl $3, %eax
; NOOPT: je
; NOOPT: subl $1, %eax
; NOOPT: je
; NOOPT: subl $4, %eax
; NOOPT: je
; NOOPT: subl $5, %eax
; NOOPT: je

; Jump table otherwise.
; CHECK-LABEL: basic
; CHECK: decl
; CHECK: cmpl $4
; CHECK: ja
; CHECK: jmpq *.LJTI
}


define void @simple_ranges(i32 %x) {
entry:
switch i32 %x, label %return [
i32 0, label %bb0
i32 1, label %bb0
i32 2, label %bb0
i32 3, label %bb0
i32 100, label %bb1
i32 101, label %bb1
i32 102, label %bb1
i32 103, label %bb1
]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
return: ret void

; Should be lowered to two range checks.
; CHECK-LABEL: simple_ranges
; CHECK: leal -100
; CHECK: cmpl $4
; CHECK: jae
; CHECK: cmpl $3
; CHECK: ja
}


define void @jt_is_better(i32 %x) {
entry:
switch i32 %x, label %return [
i32 0, label %bb0
i32 2, label %bb0
i32 4, label %bb0
i32 1, label %bb1
i32 3, label %bb1
i32 5, label %bb1

i32 6, label %bb2
i32 7, label %bb3
i32 8, label %bb4
]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
bb2: tail call void @g(i32 2) br label %return
bb3: tail call void @g(i32 3) br label %return
bb4: tail call void @g(i32 4) br label %return
return: ret void

; Cases 0-5 could be lowered with two bit tests,
; but with 6-8, the whole switch is suitable for a jump table.
; CHECK-LABEL: jt_is_better
; CHECK: cmpl $8
; CHECK: jbe
; CHECK: jmpq *.LJTI
}


define void @bt_is_better(i32 %x) {
entry:
switch i32 %x, label %return [
i32 0, label %bb0
i32 3, label %bb0
i32 6, label %bb0
i32 1, label %bb1
i32 4, label %bb1
i32 7, label %bb1
i32 2, label %bb2
i32 5, label %bb2
i32 8, label %bb2

]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
bb2: tail call void @g(i32 2) br label %return
return: ret void

; This could be lowered as a jump table, but bit tests is more efficient.
; CHECK-LABEL: bt_is_better
; 73 = 2^0 + 2^3 + 2^6
; CHECK: movl $73
; CHECK: btl
; 146 = 2^1 + 2^4 + 2^7
; CHECK: movl $146
; CHECK: btl
; 292 = 2^2 + 2^5 + 2^8
; CHECK: movl $292
; CHECK: btl
}


define void @optimal_pivot1(i32 %x) {
entry:
switch i32 %x, label %return [
i32 100, label %bb0
i32 200, label %bb1
i32 300, label %bb0
i32 400, label %bb1
i32 500, label %bb0
i32 600, label %bb1

]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
return: ret void

; Should pivot around 400 for two subtrees of equal size.
; CHECK-LABEL: optimal_pivot1
; CHECK-NOT: cmpl
; CHECK: cmpl $399
}


define void @optimal_pivot2(i32 %x) {
entry:
switch i32 %x, label %return [
i32 100, label %bb0 i32 101, label %bb1 i32 102, label %bb2 i32 103, label %bb3
i32 200, label %bb0 i32 201, label %bb1 i32 202, label %bb2 i32 203, label %bb3
i32 300, label %bb0 i32 301, label %bb1 i32 302, label %bb2 i32 303, label %bb3
i32 400, label %bb0 i32 401, label %bb1 i32 402, label %bb2 i32 403, label %bb3

]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
bb2: tail call void @g(i32 2) br label %return
bb3: tail call void @g(i32 3) br label %return
return: ret void

; Should pivot around 300 for two subtrees with two jump tables each.
; CHECK-LABEL: optimal_pivot2
; CHECK-NOT: cmpl
; CHECK: cmpl $299
; CHECK: jmpq *.LJTI
; CHECK: jmpq *.LJTI
; CHECK: jmpq *.LJTI
; CHECK: jmpq *.LJTI
}


define void @optimal_jump_table1(i32 %x) {
entry:
switch i32 %x, label %return [
i32 0, label %bb0
i32 5, label %bb1
i32 6, label %bb2
i32 12, label %bb3
i32 13, label %bb4
i32 15, label %bb5
]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
bb2: tail call void @g(i32 2) br label %return
bb3: tail call void @g(i32 3) br label %return
bb4: tail call void @g(i32 4) br label %return
bb5: tail call void @g(i32 5) br label %return
return: ret void

; Splitting in the largest gap (between 6 and 12) would yield suboptimal result.
; Expecting a jump table from 5 to 15.
; CHECK-LABEL: optimal_jump_table1
; CHECK: leal -5
; CHECK: cmpl $10
; CHECK: jmpq *.LJTI
}


define void @optimal_jump_table2(i32 %x) {
entry:
switch i32 %x, label %return [
i32 0, label %bb0
i32 1, label %bb1
i32 2, label %bb2
i32 9, label %bb3
i32 14, label %bb4
i32 15, label %bb5
]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
bb2: tail call void @g(i32 2) br label %return
bb3: tail call void @g(i32 3) br label %return
bb4: tail call void @g(i32 4) br label %return
bb5: tail call void @g(i32 5) br label %return
return: ret void

; Partitioning the cases to the minimum number of dense sets is not good enough.
; This can be partitioned as {0,1,2,9},{14,15} or {0,1,2},{9,14,15}. The former
; should be preferred. Expecting a table from 0-9.
; CHECK-LABEL: optimal_jump_table2
; CHECK: cmpl $9
; CHECK: jmpq *.LJTI
}


define void @optimal_jump_table3(i32 %x) {
entry:
switch i32 %x, label %return [
i32 1, label %bb0
i32 2, label %bb1
i32 3, label %bb2
i32 10, label %bb3
i32 13, label %bb0
i32 14, label %bb1
i32 15, label %bb2
i32 20, label %bb3
i32 25, label %bb4
]
bb0: tail call void @g(i32 0) br label %return
bb1: tail call void @g(i32 1) br label %return
bb2: tail call void @g(i32 2) br label %return
bb3: tail call void @g(i32 3) br label %return
bb4: tail call void @g(i32 4) br label %return
return: ret void

; Splitting to maximize left-right density sum and gap size would split this
; between 3 and 10, and then between 20 and 25. It's better to build a table
; from 1-20.
; CHECK-LABEL: optimal_jump_table3
; CHECK: leal -1
; CHECK: cmpl $19
; CHECK: jmpq *.LJTI
}

%struct.S = type { %struct.S*, i32 }
define void @phi_node_trouble(%struct.S* %s) {
entry:
br label %header
header:
%ptr = phi %struct.S* [ %s, %entry ], [ %next, %loop ]
%bool = icmp eq %struct.S* %ptr, null
br i1 %bool, label %exit, label %loop
loop:
%nextptr = getelementptr inbounds %struct.S, %struct.S* %ptr, i64 0, i32 0
%next = load %struct.S*, %struct.S** %nextptr
%xptr = getelementptr inbounds %struct.S, %struct.S* %next, i64 0, i32 1
%x = load i32, i32* %xptr
switch i32 %x, label %exit [
i32 4, label %header
i32 36, label %exit2
i32 69, label %exit2
i32 25, label %exit2
]
exit:
ret void
exit2:
ret void

; This will be lowered to a comparison with 4 and then bit tests. Make sure
; that the phi node in %header gets a value from the comparison block.
; CHECK-LABEL: phi_node_trouble
; CHECK: movq (%[[REG1:[a-z]+]]), %[[REG1]]
; CHECK: movl 8(%[[REG1]]), %[[REG2:[a-z]+]]
; CHECK: cmpl $4, %[[REG2]]
}
99 changes: 11 additions & 88 deletions llvm/test/MC/ARM/data-in-code.ll
View file
Original file line number Diff line number Diff line change
@@ -1,8 +1,8 @@
;; RUN: llc -O0 -verify-machineinstrs -fast-isel-abort=1 \
;; RUN: llc -verify-machineinstrs \
;; RUN: -mtriple=armv7-linux-gnueabi -filetype=obj %s -o - | \
;; RUN: llvm-readobj -t | FileCheck -check-prefix=ARM %s

;; RUN: llc -O0 -verify-machineinstrs -fast-isel-abort=1 \
;; RUN: llc -verify-machineinstrs \
;; RUN: -mtriple=thumbv7-linux-gnueabi -filetype=obj %s -o - | \
;; RUN: llvm-readobj -t | FileCheck -check-prefix=TMB %s

Expand All @@ -11,102 +11,25 @@

define void @foo(i32* %ptr) nounwind ssp {
%tmp = load i32, i32* %ptr, align 4
switch i32 %tmp, label %default [
i32 11, label %bb0
i32 10, label %bb1
i32 8, label %bb2
i32 4, label %bb3
i32 2, label %bb4
i32 6, label %bb5
i32 9, label %bb6
i32 15, label %bb7
i32 1, label %bb8
i32 3, label %bb9
i32 5, label %bb10
i32 30, label %bb11
i32 31, label %bb12
i32 13, label %bb13
i32 14, label %bb14
i32 20, label %bb15
i32 19, label %bb16
i32 17, label %bb17
i32 18, label %bb18
i32 21, label %bb19
i32 22, label %bb20
i32 16, label %bb21
i32 24, label %bb22
i32 25, label %bb23
i32 26, label %bb24
i32 27, label %bb25
i32 28, label %bb26
i32 23, label %bb27
i32 12, label %bb28
switch i32 %tmp, label %exit [
i32 0, label %bb0
i32 1, label %bb1
i32 2, label %bb2
i32 3, label %bb3
]

default:
br label %exit
bb0:
store i32 0, i32* %ptr, align 4
br label %exit
bb1:
store i32 1, i32* %ptr, align 4
br label %exit
bb2:
store i32 2, i32* %ptr, align 4
br label %exit
bb3:
store i32 3, i32* %ptr, align 4
br label %exit
bb4:
br label %exit
bb5:
br label %exit
bb6:
br label %exit
bb7:
br label %exit
bb8:
br label %exit
bb9:
br label %exit
bb10:
br label %exit
bb11:
br label %exit
bb12:
br label %exit
bb13:
br label %exit
bb14:
br label %exit
bb15:
br label %exit
bb16:
br label %exit
bb17:
br label %exit
bb18:
br label %exit
bb19:
br label %exit
bb20:
br label %exit
bb21:
br label %exit
bb22:
br label %exit
bb23:
br label %exit
bb24:
br label %exit
bb25:
br label %exit
bb26:
br label %exit
bb27:
br label %exit
bb28:
br label %exit


exit:

ret void
}

Expand Down