Implement the ldr-pseudo opcode for ARM assembly

The ldr-pseudo opcode is a convenience for loading 32-bit constants.
It is converted into a pc-relative load from a constant pool. For
example,

  ldr r0, =0x10001
  ldr r1, =bar

will generate this output in the final assembly

  ldr r0, .Ltmp0
  ldr r1, .Ltmp1
  ...
  .Ltmp0: .long 0x10001
  .Ltmp1: .long bar

Sketch of the LDR pseudo implementation:
  Keep a map from Section => ConstantPool

  When parsing ldr r0, =val
    parse val as an MCExpr
    get ConstantPool for current Section
    Label = CreateTempSymbol()
    remember val in ConstantPool at next free slot
    add operand to ldr that is MCSymbolRef of Label

  On finishParse() callback
    Write out all non-empty constant pools
    for each Entry in ConstantPool
      Emit Entry.Label
      Emit Entry.Value

Possible improvements to be added in a later patch:
  1. Does not convert load of small constants to mov
     (e.g. ldr r0, =0x1 => mov r0, 0x1)
  2. Does reuse constant pool entries for same constant

The implementation was tested for ARM, Thumb1, and Thumb2 targets on
linux and darwin.

llvm-svn: 197708

4 files changed, 489 insertions, 0 deletions

diff --git a/llvm/test/MC/ARM/ldr-pseudo-darwin.s b/llvm/test/MC/ARM/ldr-pseudo-darwin.s
new file mode 100644
index 00000000000..a77f6d5cb0e
--- /dev/null
+++ b/llvm/test/MC/ARM/ldr-pseudo-darwin.s
@@ -0,0 +1,241 @@
+@ This test has a partner (ldr-pseudo.s) that contains matching
+@ tests for the ldr-pseudo on linux targets. We need separate files
+@ because the syntax for switching sections and temporary labels differs
+@ between darwin and linux. Any tests added here should have a matching
+@ test added there.
+
+@RUN: llvm-mc -triple   armv7-apple-darwin %s | FileCheck %s
+@RUN: llvm-mc -triple thumbv5-apple-darwin %s | FileCheck %s
+@RUN: llvm-mc -triple thumbv7-apple-darwin %s | FileCheck %s
+
+@
+@ Check that large constants are converted to ldr from constant pool
+@
+@ simple test
+.section __TEXT,b,regular,pure_instructions
+@ CHECK-LABEL: f3:
+f3:
+  ldr r0, =0x10001
+@ CHECK: ldr r0, Ltmp0
+
+@ loading multiple constants
+.section __TEXT,c,regular,pure_instructions
+@ CHECK-LABEL: f4:
+f4:
+  ldr r0, =0x10002
+@ CHECK: ldr r0, Ltmp1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =0x10003
+@ CHECK: ldr r0, Ltmp2
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+@ TODO: the same constants should have the same constant pool location
+.section __TEXT,d,regular,pure_instructions
+@ CHECK-LABEL: f5:
+f5:
+  ldr r0, =0x10004
+@ CHECK: ldr r0, Ltmp3
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =0x10004
+@ CHECK: ldr r0, Ltmp4
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+@ a section defined in multiple pieces should be merged and use a single constant pool
+.section __TEXT,e,regular,pure_instructions
+@ CHECK-LABEL: f6:
+f6:
+  ldr r0, =0x10006
+@ CHECK: ldr r0, Ltmp5
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+.section __TEXT,f,regular,pure_instructions
+@ CHECK-LABEL: f7:
+f7:
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+.section __TEXT,e,regular,pure_instructions
+@ CHECK-LABEL: f8:
+f8:
+  adds r0, r0, #1
+  ldr r0, =0x10007
+@ CHECK: ldr r0, Ltmp6
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+@
+@ Check that symbols can be loaded using ldr pseudo
+@
+
+@ load an undefined symbol
+.section __TEXT,g,regular,pure_instructions
+@ CHECK-LABEL: f9:
+f9:
+  ldr r0, =foo
+@ CHECK: ldr r0, Ltmp7
+
+@ load a symbol from another section
+.section __TEXT,h,regular,pure_instructions
+@ CHECK-LABEL: f10:
+f10:
+  ldr r0, =f5
+@ CHECK: ldr r0, Ltmp8
+
+@ load a symbol from the same section
+.section __TEXT,i,regular,pure_instructions
+@ CHECK-LABEL: f11:
+f11:
+  ldr r0, =f12
+@ CHECK: ldr r0, Ltmp9
+
+@ CHECK-LABEL: f12:
+f12:
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+.section __TEXT,j,regular,pure_instructions
+@ mix of symbols and constants
+@ CHECK-LABEL: f13:
+f13:
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =0x101
+@ CHECK: ldr r0, Ltmp10
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =bar
+@ CHECK: ldr r0, Ltmp11
+  adds r0, r0, #1
+  adds r0, r0, #1
+@
+@ Check for correct usage in other contexts
+@
+
+@ usage in macro
+.macro useit_in_a_macro
+  ldr r0, =0x10008
+  ldr r0, =baz
+.endm
+.section __TEXT,k,regular,pure_instructions
+@ CHECK-LABEL: f14:
+f14:
+  useit_in_a_macro
+@ CHECK: ldr r0, Ltmp12
+@ CHECK: ldr r0, Ltmp13
+
+@ usage with expressions
+.section __TEXT,l,regular,pure_instructions
+@ CHECK-LABEL: f15:
+f15:
+  ldr r0, =0x10001+8
+@ CHECK: ldr r0, Ltmp14
+  adds r0, r0, #1
+  ldr r0, =bar+4
+@ CHECK: ldr r0, Ltmp15
+  adds r0, r0, #1
+
+@
+@ Constant Pools
+@
+@ CHECK: .section __TEXT,b,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp0:
+@ CHECK: .long 65537
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,c,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp1:
+@ CHECK: .long 65538
+@ CHECK-LABEL: Ltmp2:
+@ CHECK: .long 65539
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,d,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp3:
+@ CHECK: .long 65540
+@ CHECK-LABEL: Ltmp4:
+@ CHECK: .long 65540
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,e,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp5:
+@ CHECK: .long 65542
+@ CHECK-LABEL: Ltmp6:
+@ CHECK: .long 65543
+@ CHECK: .end_data_region
+
+@ Should not switch to section because it has no constant pool
+@ CHECK-NOT: .section __TEXT,f,regular,pure_instructions
+
+@ CHECK: .section __TEXT,g,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp7:
+@ CHECK: .long foo
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,h,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp8:
+@ CHECK: .long f5
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,i,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp9:
+@ CHECK: .long f12
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,j,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp10:
+@ CHECK: .long 257
+@ CHECK-LABEL: Ltmp11:
+@ CHECK: .long bar
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,k,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp12:
+@ CHECK: .long 65544
+@ CHECK-LABEL: Ltmp13:
+@ CHECK: .long baz
+@ CHECK: .end_data_region
+
+@ CHECK: .section __TEXT,l,regular,pure_instructions
+@ CHECK: .align 2
+@ CHECK: .data_region
+@ CHECK-LABEL: Ltmp14:
+@ CHECK: .long 65545
+@ CHECK-LABEL: Ltmp15:
+@ CHECK: .long bar+4
+@ CHECK: .end_data_region
diff --git a/llvm/test/MC/ARM/ldr-pseudo-obj-errors.s b/llvm/test/MC/ARM/ldr-pseudo-obj-errors.s
new file mode 100644
index 00000000000..dce88f058f8
--- /dev/null
+++ b/llvm/test/MC/ARM/ldr-pseudo-obj-errors.s
@@ -0,0 +1,17 @@
+@RUN: not llvm-mc -triple=armv7-unknown-linux-gnueabi -filetype=obj %s -o %t1 2> %t2
+@RUN: cat %t2 | FileCheck %s
+@RUN: not llvm-mc -triple=armv7-darwin-apple -filetype=obj %s -o %t1_darwin 2> %t2_darwin
+@RUN: cat %t2_darwin | FileCheck %s
+
+@These tests look for errors that should be reported for invalid object layout
+@with the ldr pseudo. They are tested separately from parse errors because they
+@only trigger when the file has successfully parsed and the object file is about
+@to be written out.
+
+.text
+foo:
+  ldr r0, =0x101
+  .space 8000
+@ CHECK: error: out of range pc-relative fixup value
+@ CHECK: ldr r0, =0x101
+@ CHECK: ^
diff --git a/llvm/test/MC/ARM/ldr-pseudo-parse-errors.s b/llvm/test/MC/ARM/ldr-pseudo-parse-errors.s
new file mode 100644
index 00000000000..2e6114d6fe0
--- /dev/null
+++ b/llvm/test/MC/ARM/ldr-pseudo-parse-errors.s
@@ -0,0 +1,10 @@
+@RUN: not llvm-mc -triple=armv7-unknown-linux-gnueabi < %s  2>&1 | FileCheck %s
+@RUN: not llvm-mc -triple=armv7-apple-darwin < %s  2>&1 | FileCheck %s
+
+.text
+bar:
+  mov r0, =0x101
+@ CHECK: error: unexpected token in operand
+@ CHECK: mov r0, =0x101
+@ CHECK:         ^
+
diff --git a/llvm/test/MC/ARM/ldr-pseudo.s b/llvm/test/MC/ARM/ldr-pseudo.s
new file mode 100644
index 00000000000..889cb8e7bcb
--- /dev/null
+++ b/llvm/test/MC/ARM/ldr-pseudo.s
@@ -0,0 +1,221 @@
+@ This test has a partner (ldr-pseudo-darwin.s) that contains matching
+@ tests for the ldr-pseudo on darwin targets. We need separate files
+@ because the syntax for switching sections and temporary labels differs
+@ between darwin and linux. Any tests added here should have a matching
+@ test added there.
+
+@RUN: llvm-mc -triple   armv7-unknown-linux-gnueabi %s | FileCheck %s
+@RUN: llvm-mc -triple thumbv5-unknown-linux-gnueabi %s | FileCheck %s
+@RUN: llvm-mc -triple thumbv7-unknown-linux-gnueabi %s | FileCheck %s
+
+@
+@ Check that large constants are converted to ldr from constant pool
+@
+@ simple test
+.section b,"ax",%progbits
+@ CHECK-LABEL: f3:
+f3:
+  ldr r0, =0x10001
+@ CHECK: ldr r0, .Ltmp0
+
+@ loading multiple constants
+.section c,"ax",%progbits
+@ CHECK-LABEL: f4:
+f4:
+  ldr r0, =0x10002
+@ CHECK: ldr r0, .Ltmp1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =0x10003
+@ CHECK: ldr r0, .Ltmp2
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+@ TODO: the same constants should have the same constant pool location
+.section d,"ax",%progbits
+@ CHECK-LABEL: f5:
+f5:
+  ldr r0, =0x10004
+@ CHECK: ldr r0, .Ltmp3
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =0x10004
+@ CHECK: ldr r0, .Ltmp4
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+@ a section defined in multiple pieces should be merged and use a single constant pool
+.section e,"ax",%progbits
+@ CHECK-LABEL: f6:
+f6:
+  ldr r0, =0x10006
+@ CHECK: ldr r0, .Ltmp5
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+.section f, "ax", %progbits
+@ CHECK-LABEL: f7:
+f7:
+  adds r0, r0, #1
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+.section e, "ax", %progbits
+@ CHECK-LABEL: f8:
+f8:
+  adds r0, r0, #1
+  ldr r0, =0x10007
+@ CHECK: ldr r0, .Ltmp6
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+@
+@ Check that symbols can be loaded using ldr pseudo
+@
+
+@ load an undefined symbol
+.section g,"ax",%progbits
+@ CHECK-LABEL: f9:
+f9:
+  ldr r0, =foo
+@ CHECK: ldr r0, .Ltmp7
+
+@ load a symbol from another section
+.section h,"ax",%progbits
+@ CHECK-LABEL: f10:
+f10:
+  ldr r0, =f5
+@ CHECK: ldr r0, .Ltmp8
+
+@ load a symbol from the same section
+.section i,"ax",%progbits
+@ CHECK-LABEL: f11:
+f11:
+  ldr r0, =f12
+@ CHECK: ldr r0, .Ltmp9
+
+@ CHECK-LABEL: f12:
+f12:
+  adds r0, r0, #1
+  adds r0, r0, #1
+
+.section j,"ax",%progbits
+@ mix of symbols and constants
+@ CHECK-LABEL: f13:
+f13:
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =0x101
+@ CHECK: ldr r0, .Ltmp10
+  adds r0, r0, #1
+  adds r0, r0, #1
+  ldr r0, =bar
+@ CHECK: ldr r0, .Ltmp11
+  adds r0, r0, #1
+  adds r0, r0, #1
+@
+@ Check for correct usage in other contexts
+@
+
+@ usage in macro
+.macro useit_in_a_macro
+  ldr r0, =0x10008
+  ldr r0, =baz
+.endm
+.section k,"ax",%progbits
+@ CHECK-LABEL: f14:
+f14:
+  useit_in_a_macro
+@ CHECK: ldr r0, .Ltmp12
+@ CHECK: ldr r0, .Ltmp13
+
+@ usage with expressions
+.section l, "ax", %progbits
+@ CHECK-LABEL: f15:
+f15:
+  ldr r0, =0x10001+8
+@ CHECK: ldr r0, .Ltmp14
+  adds r0, r0, #1
+  ldr r0, =bar+4
+@ CHECK: ldr r0, .Ltmp15
+  adds r0, r0, #1
+
+@
+@ Constant Pools
+@
+@ CHECK: .section b,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp0:
+@ CHECK: .long 65537
+
+@ CHECK: .section c,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp1:
+@ CHECK: .long 65538
+@ CHECK-LABEL: .Ltmp2:
+@ CHECK: .long 65539
+
+@ CHECK: .section d,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp3:
+@ CHECK: .long 65540
+@ CHECK-LABEL: .Ltmp4:
+@ CHECK: .long 65540
+
+@ CHECK: .section e,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp5:
+@ CHECK: .long 65542
+@ CHECK-LABEL: .Ltmp6:
+@ CHECK: .long 65543
+
+@ Should not switch to section because it has no constant pool
+@ CHECK-NOT: .section f,"ax",%progbits
+
+@ CHECK: .section g,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp7:
+@ CHECK: .long foo
+
+@ CHECK: .section h,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp8:
+@ CHECK: .long f5
+
+@ CHECK: .section i,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp9:
+@ CHECK: .long f12
+
+@ CHECK: .section j,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp10:
+@ CHECK: .long 257
+@ CHECK-LABEL: .Ltmp11:
+@ CHECK: .long bar
+
+@ CHECK: .section k,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp12:
+@ CHECK: .long 65544
+@ CHECK-LABEL: .Ltmp13:
+@ CHECK: .long baz
+
+@ CHECK: .section l,"ax",%progbits
+@ CHECK: .align 2
+@ CHECK-LABEL: .Ltmp14:
+@ CHECK: .long 65545
+@ CHECK-LABEL: .Ltmp15:
+@ CHECK: .long bar+4