amdgcn: Support basic DWARF

This is enough DWARF support for "-O0 -g" to work OK, within a single frame,
using the new rocgdb debugger from AMD.  Debugging with optimization enabled
also works when the values are located in SGPRs or memory.

Scalars in VGPRs are problematic, EXEC_HI and VCC_HI are unmappable, and
CFI remains unimplemented.

gcc/ChangeLog:

	Backport from mainline (eff23b7961f):

	* config/gcn/gcn-hsa.h (DBX_REGISTER_NUMBER): New macro.
	* config/gcn/gcn-protos.h (gcn_dwarf_register_number): New prototype.
	* config/gcn/gcn.c (gcn_expand_prologue): Add RTX_FRAME_RELATED_P
	and REG_FRAME_RELATED_EXPR to stack and frame pointer adjustments.
	(gcn_dwarf_register_number): New function.
	(gcn_dwarf_register_span): New function.
	(TARGET_DWARF_REGISTER_SPAN): New hook macro.

diff --git a/gcc/ChangeLog.omp b/gcc/ChangeLog.omp
index e93424072e697368cb5791e50f5f2052fd626296..799aaab3468670dcb044da4d0f0216a92a12c58a 100644 (file)
--- a/gcc/ChangeLog.omp
+++ b/gcc/ChangeLog.omp
@@ -1,3 +1,15 @@
+2020-07-16  Andrew Stubbs  <ams@codesourcery.com>
+
+       Backport from mainline (eff23b7961f):
+
+       * config/gcn/gcn-hsa.h (DBX_REGISTER_NUMBER): New macro.
+       * config/gcn/gcn-protos.h (gcn_dwarf_register_number): New prototype.
+       * config/gcn/gcn.c (gcn_expand_prologue): Add RTX_FRAME_RELATED_P
+       and REG_FRAME_RELATED_EXPR to stack and frame pointer adjustments.
+       (gcn_dwarf_register_number): New function.
+       (gcn_dwarf_register_span): New function.
+       (TARGET_DWARF_REGISTER_SPAN): New hook macro.
+
 2020-07-15  Kwok Cheung Yeung  <kcy@codesourcery.com>
            Andrew Stubbs   <ams@codesourcery.com>
 

diff --git a/gcc/config/gcn/gcn-hsa.h b/gcc/config/gcn/gcn-hsa.h
index 5797c59941d2349aacd576777b637fc5ccd17ba3..3c4b07fd8d0cc7ee647c7cc2c2bbbfc5bcc0b305 100644 (file)
--- a/gcc/config/gcn/gcn-hsa.h
+++ b/gcc/config/gcn/gcn-hsa.h
@@ -113,3 +113,4 @@ extern const char *last_arg_spec_function (int argc, const char **argv);
 #define DWARF2_DEBUGGING_INFO      1
 #define DWARF2_ASM_LINE_DEBUG_INFO 1
 #define EH_FRAME_THROUGH_COLLECT2  1
+#define DBX_REGISTER_NUMBER(REGNO) gcn_dwarf_register_number (REGNO)

diff --git a/gcc/config/gcn/gcn-protos.h b/gcc/config/gcn/gcn-protos.h
index 5bd9f44572f82abb5766d7006fba14ec62105bb2..575c64e66cd141d4cd02b77630039dfcd8361ec0 100644 (file)
--- a/gcc/config/gcn/gcn-protos.h
+++ b/gcc/config/gcn/gcn-protos.h
@@ -23,6 +23,7 @@ extern bool gcn_can_split_p (machine_mode, rtx);
 extern bool gcn_constant64_p (rtx);
 extern bool gcn_constant_p (rtx);
 extern rtx gcn_convert_mask_mode (rtx reg);
+extern unsigned int gcn_dwarf_register_number (unsigned int regno);
 extern char * gcn_expand_dpp_shr_insn (machine_mode, const char *, int, int);
 extern void gcn_expand_epilogue ();
 extern rtx gcn_expand_scaled_offsets (addr_space_t as, rtx base, rtx offsets,

diff --git a/gcc/config/gcn/gcn.c b/gcc/config/gcn/gcn.c
index 38dda166160bde1debc6c70aa9f31f8864b266db..4b2df3c7df061229f29aa7df0bbc9d41d849531e 100644 (file)
--- a/gcc/config/gcn/gcn.c
+++ b/gcc/config/gcn/gcn.c
@@ -2809,7 +2809,11 @@ gcn_expand_prologue ()
   if (!cfun || !cfun->machine || cfun->machine->normal_function)
     {
       rtx sp = gen_rtx_REG (Pmode, STACK_POINTER_REGNUM);
+      rtx sp_hi = gcn_operand_part (Pmode, sp, 1);
+      rtx sp_lo = gcn_operand_part (Pmode, sp, 0);
       rtx fp = gen_rtx_REG (Pmode, HARD_FRAME_POINTER_REGNUM);
+      rtx fp_hi = gcn_operand_part (Pmode, fp, 1);
+      rtx fp_lo = gcn_operand_part (Pmode, fp, 0);
 
       start_sequence ();
 
@@ -2826,14 +2830,40 @@ gcn_expand_prologue ()
        + offsets->callee_saves
        + offsets->local_vars + offsets->outgoing_args_size;
       if (sp_adjust > 0)
-       emit_insn (gen_adddi3_scc (sp, sp, gen_int_mode (sp_adjust, DImode)));
+       {
+         /* Adding RTX_FRAME_RELATED_P effectively disables spliting, so
+            we use split add explictly, and specify the DImode add in
+            the note.  */
+         rtx scc = gen_rtx_REG (BImode, SCC_REG);
+         rtx adjustment = gen_int_mode (sp_adjust, SImode);
+         rtx insn = emit_insn (gen_addsi3_scalar_carry (sp_lo, sp_lo,
+                                                        adjustment, scc));
+         RTX_FRAME_RELATED_P (insn) = 1;
+         add_reg_note (insn, REG_FRAME_RELATED_EXPR,
+                       gen_rtx_SET (sp,
+                                    gen_rtx_PLUS (DImode, sp, adjustment)));
+         emit_insn (gen_addcsi3_scalar_zero (sp_hi, sp_hi, scc));
+       }
 
       if (offsets->need_frame_pointer)
-       emit_insn (gen_adddi3_scc (fp, sp,
-                                  gen_int_mode
-                                  (-(offsets->local_vars +
-                                     offsets->outgoing_args_size),
-                                   DImode)));
+       {
+         /* Adding RTX_FRAME_RELATED_P effectively disables spliting, so
+            we use split add explictly, and specify the DImode add in
+            the note.  */
+         rtx scc = gen_rtx_REG (BImode, SCC_REG);
+         int fp_adjust = -(offsets->local_vars + offsets->outgoing_args_size);
+         rtx adjustment = gen_int_mode (fp_adjust, SImode);
+         rtx insn = emit_insn (gen_addsi3_scalar_carry(fp_lo, sp_lo,
+                                                       adjustment, scc));
+         RTX_FRAME_RELATED_P (insn) = 1;
+         add_reg_note (insn, REG_FRAME_RELATED_EXPR,
+                       gen_rtx_SET (fp,
+                                    gen_rtx_PLUS (DImode, sp, adjustment)));
+         emit_insn (gen_addcsi3_scalar (fp_hi, sp_hi,
+                                        (fp_adjust < 0 ? GEN_INT (-1)
+                                         : const0_rtx),
+                                        scc, scc));
+       }
 
       rtx_insn *seq = get_insns ();
       end_sequence ();
@@ -2879,6 +2909,8 @@ gcn_expand_prologue ()
 
       /* Set up frame pointer and stack pointer.  */
       rtx sp = gen_rtx_REG (DImode, STACK_POINTER_REGNUM);
+      rtx sp_hi = simplify_gen_subreg (SImode, sp, DImode, 4);
+      rtx sp_lo = simplify_gen_subreg (SImode, sp, DImode, 0);
       rtx fp = gen_rtx_REG (DImode, HARD_FRAME_POINTER_REGNUM);
       rtx fp_hi = simplify_gen_subreg (SImode, fp, DImode, 4);
       rtx fp_lo = simplify_gen_subreg (SImode, fp, DImode, 0);
@@ -2894,10 +2926,28 @@ gcn_expand_prologue ()
       emit_insn (gen_addsi3_scalar_carry (fp_lo, fp_lo, wave_offset, scc));
       emit_insn (gen_addcsi3_scalar_zero (fp_hi, fp_hi, scc));
 
+      /* Adding RTX_FRAME_RELATED_P effectively disables spliting, so we use
+        split add explictly, and specify the DImode add in the note.
+         The DWARF info expects that the callee-save data is in the frame,
+         even though it isn't (because this is the entry point), so we
+         make a notional adjustment to the DWARF frame offset here.  */
+      rtx dbg_adjustment = gen_int_mode (sp_adjust + offsets->callee_saves,
+                                        DImode);
+      rtx insn;
       if (sp_adjust > 0)
-       emit_insn (gen_adddi3_scc (sp, fp, gen_int_mode (sp_adjust, DImode)));
+       {
+         rtx scc = gen_rtx_REG (BImode, SCC_REG);
+         rtx adjustment = gen_int_mode (sp_adjust, DImode);
+         insn = emit_insn (gen_addsi3_scalar_carry(sp_lo, fp_lo, adjustment,
+                                                   scc));
+         emit_insn (gen_addcsi3_scalar_zero (sp_hi, fp_hi, scc));
+       }
       else
-       emit_move_insn (sp, fp);
+       insn = emit_move_insn (sp, fp);
+      RTX_FRAME_RELATED_P (insn) = 1;
+      add_reg_note (insn, REG_FRAME_RELATED_EXPR,
+                   gen_rtx_SET (sp, gen_rtx_PLUS (DImode, sp,
+                                                  dbg_adjustment)));
 
       /* Make sure the flat scratch reg doesn't get optimised away.  */
       emit_insn (gen_prologue_use (gen_rtx_REG (DImode, FLAT_SCRATCH_REG)));
@@ -6146,6 +6196,64 @@ print_operand (FILE *file, rtx x, int code)
   gcc_unreachable ();
 }
 
+/* Implement DBX_REGISTER_NUMBER macro.
+ 
+   Return the DWARF register number that corresponds to the GCC internal
+   REGNO.  */
+
+unsigned int
+gcn_dwarf_register_number (unsigned int regno)
+{
+  /* Registers defined in DWARF.  */
+  if (regno == EXEC_LO_REG)
+    return 17;
+  /* We need to use a more complex DWARF expression for this
+  else if (regno == EXEC_HI_REG)
+    return 17; */
+  else if (regno == VCC_LO_REG)
+    return 768;
+  /* We need to use a more complex DWARF expression for this
+  else if (regno == VCC_HI_REG)
+    return 768;  */
+  else if (regno == SCC_REG)
+    return 128;
+  else if (SGPR_REGNO_P (regno))
+    {
+      if (regno - FIRST_SGPR_REG < 64)
+       return (regno - FIRST_SGPR_REG + 32);
+      else
+       return (regno - FIRST_SGPR_REG + 1024);
+    }
+  else if (VGPR_REGNO_P (regno))
+    return (regno - FIRST_VGPR_REG + 2560);
+
+  /* Otherwise, there's nothing sensible to do.  */
+  return regno + 100000;
+}
+
+/* Implement TARGET_DWARF_REGISTER_SPAN.
+ 
+   DImode and Vector DImode require additional registers.  */
+
+static rtx
+gcn_dwarf_register_span (rtx rtl)
+{
+  machine_mode mode = GET_MODE (rtl);
+
+  if (VECTOR_MODE_P (mode))
+    mode = GET_MODE_INNER (mode);
+
+  if (GET_MODE_SIZE (mode) != 8)
+    return NULL_RTX;
+
+  rtx p = gen_rtx_PARALLEL (VOIDmode, rtvec_alloc (2));
+  unsigned regno = REGNO (rtl);
+  XVECEXP (p, 0, 0) = gen_rtx_REG (SImode, regno);
+  XVECEXP (p, 0, 1) = gen_rtx_REG (SImode, regno + 1);
+
+  return p;
+}
+
 /* }}}  */
 /* {{{ TARGET hook overrides.  */
 
@@ -6194,6 +6302,8 @@ print_operand (FILE *file, rtx x, int code)
 #define TARGET_CONSTANT_ALIGNMENT gcn_constant_alignment
 #undef  TARGET_DEBUG_UNWIND_INFO
 #define TARGET_DEBUG_UNWIND_INFO gcn_debug_unwind_info
+#undef  TARGET_DWARF_REGISTER_SPAN
+#define TARGET_DWARF_REGISTER_SPAN gcn_dwarf_register_span
 #undef  TARGET_EMUTLS_VAR_INIT
 #define TARGET_EMUTLS_VAR_INIT gcn_emutls_var_init
 #undef  TARGET_EXPAND_BUILTIN