[4/4][llvm-objdump][RISCV] Support --symbolize-operands for RISC-V

[jhenderson]

llvm/docs/CommandGuide/llvm-objdump.rst

229

I'd probably add a comma after "objects".

[asb]

Address review comment.

[MaskRay]

LGTM.

[asb]

Rebase.

llvm/docs/CommandGuide/llvm-objdump.rst

.. option:: --stop-address=<address>
   When printing symbols, only print symbols with a value up to ``address``.
 
 .. option:: --symbolize-operands
 
   When disassembling, symbolize a branch target operand to print a label instead of a real address.
 
   When printing a PC-relative global symbol reference, print it as an offset from the leading symbol.
 
-  Only works with PowerPC objects or X86 linked images.
+  Only works with PowerPC or RISC-V objects or X86 linked images.
 
   Example:
     A non-symbolized branch instruction with a local target and pc-relative memory access like
 
   .. code-block:: none
 
       cmp eax, dword ptr [rip + 4112]
       jge 0x20117e <_start+0x25>

llvm/test/tools/llvm-objdump/ELF/RISCV/branches.s

-# RUN: llvm-mc -filetype=obj -triple riscv32 -mattr=+c < %s \
-# RUN:     | llvm-objdump -d -M no-aliases --no-show-raw-insn - \
-# RUN:     | FileCheck --match-full-lines %s
+# RUN: llvm-mc -filetype=obj -triple riscv32 -mattr=+c < %s -o %t
+# RUN: llvm-objdump %t -d -M no-aliases --no-show-raw-insn \
+# RUN:     | FileCheck -check-prefixes=CHECK-NOSYMBOLIZE,CHECK-COMMON --match-full-lines %s
+# RUN: llvm-objdump %t -d -M no-aliases --no-show-raw-insn --symbolize-operands \
+# RUN:     | FileCheck -check-prefixes=CHECK-SYMBOLIZE,CHECK-COMMON --match-full-lines %s
 
-# CHECK: 00000000 <foo>:
+# CHECK-COMMON: 00000000 <foo>:
 foo:
-# CHECK: 0: beq t0, t1, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 0: beq t0, t1, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 0: beq t0, t1, 0x58 <L0>
 beq t0, t1, .Llocal
-# CHECK: 4: bne t0, t1, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 4: bne t0, t1, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 4: bne t0, t1, 0x58 <L0>
 bne t0, t1, .Llocal
-# CHECK: 8: blt t0, t1, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 8: blt t0, t1, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 8: blt t0, t1, 0x58 <L0>
 blt t0, t1, .Llocal
-# CHECK: c: bge t0, t1, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: c: bge t0, t1, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: c: bge t0, t1, 0x58 <L0>
 bge t0, t1, .Llocal
-# CHECK: 10: bltu t0, t1, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 10: bltu t0, t1, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 10: bltu t0, t1, 0x58 <L0>
 bltu t0, t1, .Llocal
-# CHECK: 14: bgeu t0, t1, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 14: bgeu t0, t1, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 14: bgeu t0, t1, 0x58 <L0>
 bgeu t0, t1, .Llocal
 
-# CHECK: 18: c.beqz a0, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 18: c.beqz a0, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 18: c.beqz a0, 0x58 <L0>
 beq a0, zero, .Llocal
-# CHECK: 1a: c.bnez a0, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 1a: c.bnez a0, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 1a: c.bnez a0, 0x58 <L0>
 bne a0, zero, .Llocal
 
-# CHECK: 1c: beq t0, t1, 0x60 <bar>
+# CHECK-COMMON: 1c: beq t0, t1, 0x60 <bar>
 beq t0, t1, bar
-# CHECK: 20: bne t0, t1, 0x60 <bar>
+# CHECK-COMMON: 20: bne t0, t1, 0x60 <bar>
 bne t0, t1, bar
-# CHECK: 24: blt t0, t1, 0x60 <bar>
+# CHECK-COMMON: 24: blt t0, t1, 0x60 <bar>
 blt t0, t1, bar
-# CHECK: 28: bge t0, t1, 0x60 <bar>
+# CHECK-COMMON: 28: bge t0, t1, 0x60 <bar>
 bge t0, t1, bar
-# CHECK: 2c: bltu t0, t1, 0x60 <bar>
+# CHECK-COMMON: 2c: bltu t0, t1, 0x60 <bar>
 bltu t0, t1, bar
-# CHECK: 30: bgeu t0, t1, 0x60 <bar>
+# CHECK-COMMON: 30: bgeu t0, t1, 0x60 <bar>
 bgeu t0, t1, bar
 
-# CHECK: 34: c.beqz a0, 0x60 <bar>
+# CHECK-COMMON: 34: c.beqz a0, 0x60 <bar>
 beq a0, zero, bar
-# CHECK: 36: c.bnez a0, 0x60 <bar>
+# CHECK-COMMON: 36: c.bnez a0, 0x60 <bar>
 bne a0, zero, bar
 
-# CHECK: 38: jal t0, 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 38: jal t0, 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 38: jal t0, 0x58 <L0>
 jal t0, .Llocal
-# CHECK: 3c: c.jal 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 3c: c.jal 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 3c: c.jal 0x58 <L0>
 c.jal .Llocal
 
-# CHECK: 3e: c.j 0x58 <foo+0x58>
+# CHECK-NOSYMBOLIZE: 3e: c.j 0x58 <foo+0x58>
+# CHECK-SYMBOLIZE: 3e: c.j 0x58 <L0>
 c.j .Llocal
 
-# CHECK: 40: jal t0, 0x60 <bar>
+# CHECK-COMMON: 40: jal t0, 0x60 <bar>
 jal t0, bar
-# CHECK: 44: c.jal 0x60 <bar>
+# CHECK-COMMON: 44: c.jal 0x60 <bar>
 c.jal bar
 
-# CHECK: 46: c.j 0x60 <bar>
+# CHECK-COMMON: 46: c.j 0x60 <bar>
 c.j bar
 
-# CHECK: 48: auipc  ra, 0
-# CHECK: 4c: jalr ra, 16(ra) <foo+0x58>
+# CHECK-COMMON: 48: auipc ra, 0
+# CHECK-NOSYMBOLIZE: 4c: jalr ra, 16(ra) <foo+0x58>
+# CHECK-SYMBOLIZE: 4c: jalr ra, 16(ra) <L0>
 call .Llocal
 
-# CHECK: 50: auipc  ra, 0
-# CHECK: 54: jalr ra, 16(ra) <bar>
+# CHECK-COMMON: 50: auipc ra, 0
+# CHECK-COMMON: 54: jalr ra, 16(ra) <bar>
 call bar
 
 .Llocal:
-# CHECK: 58: c.nop
-# CHECK: 5a: c.nop
-# CHECK: 5c: c.nop
-# CHECK: 5e: c.nop
+# CHECK-COMMON: 58: c.nop
+# CHECK-COMMON: 5a: c.nop
+# CHECK-COMMON: 5c: c.nop
+# CHECK-COMMON: 5e: c.nop
 nop
 nop
 nop
 nop
 
-# CHECK: 00000060 <bar>:
+# CHECK-COMMON: 00000060 <bar>:
 bar:
-# CHECK: 60: c.nop
+# CHECK-COMMON: 60: c.nop
 nop

llvm/tools/llvm-objdump/llvm-objdump.cpp

 }
 
 static void
 collectLocalBranchTargets(ArrayRef<uint8_t> Bytes, const MCInstrAnalysis *MIA,
                           MCDisassembler *DisAsm, MCInstPrinter *IP,
                           const MCSubtargetInfo *STI, uint64_t SectionAddr,
                           uint64_t Start, uint64_t End,
                           std::unordered_map<uint64_t, std::string> &Labels) {
-  // So far only supports PowerPC and X86.
-  if (!STI->getTargetTriple().isPPC() && !STI->getTargetTriple().isX86())
+  // So far only supports PowerPC, RISC-V, and X86.
+  if (!STI->getTargetTriple().isPPC() && !STI->getTargetTriple().isRISCV() &&
+      !STI->getTargetTriple().isX86())
     return;
 
   Labels.clear();
   unsigned LabelCount = 0;
   Start += SectionAddr;
   End += SectionAddr;
   uint64_t Index = Start;
   Optional<MCInst> PreviousInst;