Jiufu Guo <guoji...@linux.ibm.com> writes:
> Hi,
>
> As mentioned in the previous version patch:
> https://gcc.gnu.org/pipermail/gcc-patches/2022-October/604646.html
> The suboptimal code is generated for "assigning from parameter" or
> "assigning to return value".
> This patch enhances the assignment from parameters like the below
> cases:
> /////case1.c
> typedef struct SA {double a[3];long l; } A;
> A ret_arg (A a) {return a;}
> void st_arg (A a, A *p) {*p = a;}
>
> ////case2.c
> typedef struct SA {double a[3];} A;
> A ret_arg (A a) {return a;}
> void st_arg (A a, A *p) {*p = a;}
>
> For this patch, bootstrap and regtest pass on ppc64{,le}
> and x86_64.
> * Besides asking for help reviewing this patch, I would like to
> consult comments about enhancing for "assigning to returns".
I updated the patch to fix the issue for returns. This patch
adds a flag DECL_USEDBY_RETURN_P to indicate if a var is used
by a return stmt. This patch fix the issue in expand pass only,
so, we would try to update the patch to avoid this flag.
diff --git a/gcc/cfgexpand.cc b/gcc/cfgexpand.cc
index dd29ffffc03..09b8ec64cea 100644
--- a/gcc/cfgexpand.cc
+++ b/gcc/cfgexpand.cc
@@ -2158,6 +2158,20 @@ expand_used_vars (bitmap forced_stack_vars)
frame_phase = off ? align - off : 0;
}
+ /* Collect VARs on returns. */
+ if (DECL_RESULT (current_function_decl))
+ {
+ edge_iterator ei;
+ edge e;
+ FOR_EACH_EDGE (e, ei, EXIT_BLOCK_PTR_FOR_FN (cfun)->preds)
+ if (greturn *ret = safe_dyn_cast<greturn *> (last_stmt (e->src)))
+ {
+ tree val = gimple_return_retval (ret);
+ if (val && VAR_P (val))
+ DECL_USEDBY_RETURN_P (val) = 1;
+ }
+ }
+
/* Set TREE_USED on all variables in the local_decls. */
FOR_EACH_LOCAL_DECL (cfun, i, var)
TREE_USED (var) = 1;
diff --git a/gcc/expr.cc b/gcc/expr.cc
index d9407432ea5..20973649963 100644
--- a/gcc/expr.cc
+++ b/gcc/expr.cc
@@ -6045,6 +6045,52 @@ expand_assignment (tree to, tree from, bool nontemporal)
return;
}
+ if ((TREE_CODE (from) == PARM_DECL && DECL_INCOMING_RTL (from)
+ && TYPE_MODE (TREE_TYPE (from)) == BLKmode
+ && (GET_CODE (DECL_INCOMING_RTL (from)) == PARALLEL
+ || REG_P (DECL_INCOMING_RTL (from))))
+ || (VAR_P (to) && DECL_USEDBY_RETURN_P (to)
+ && TYPE_MODE (TREE_TYPE (to)) == BLKmode
+ && GET_CODE (DECL_RTL (DECL_RESULT (current_function_decl)))
+ == PARALLEL))
+ {
+ push_temp_slots ();
+ rtx par_ret;
+ machine_mode mode;
+ par_ret = TREE_CODE (from) == PARM_DECL
+ ? DECL_INCOMING_RTL (from)
+ : DECL_RTL (DECL_RESULT (current_function_decl));
+ mode = GET_CODE (par_ret) == PARALLEL
+ ? GET_MODE (XEXP (XVECEXP (par_ret, 0, 0), 0))
+ : word_mode;
+ int mode_size = GET_MODE_SIZE (mode).to_constant ();
+ int size = INTVAL (expr_size (from));
+
+ /* If/How the parameter using submode, it dependes on the size and
+ position of the parameter. Here using heurisitic number. */
+ int hurstc_num = 8;
+ if (size < mode_size || (size % mode_size) != 0
+ || size > (mode_size * hurstc_num))
+ result = store_expr (from, to_rtx, 0, nontemporal, false);
+ else
+ {
+ rtx from_rtx
+ = expand_expr (from, NULL_RTX, GET_MODE (to_rtx), EXPAND_NORMAL);
+ for (int i = 0; i < size / mode_size; i++)
+ {
+ rtx temp = gen_reg_rtx (mode);
+ rtx src = adjust_address (from_rtx, mode, mode_size * i);
+ rtx dest = adjust_address (to_rtx, mode, mode_size * i);
+ emit_move_insn (temp, src);
+ emit_move_insn (dest, temp);
+ }
+ result = to_rtx;
+ }
+ preserve_temp_slots (result);
+ pop_temp_slots ();
+ return;
+ }
+
/* Compute FROM and store the value in the rtx we got. */
push_temp_slots ();
diff --git a/gcc/tree-core.h b/gcc/tree-core.h
index af75522504f..be42e1464de 100644
--- a/gcc/tree-core.h
+++ b/gcc/tree-core.h
@@ -1808,7 +1808,8 @@ struct GTY(()) tree_decl_common {
In VAR_DECL, PARM_DECL and RESULT_DECL, this is
DECL_HAS_VALUE_EXPR_P. */
unsigned decl_flag_2 : 1;
- /* In FIELD_DECL, this is DECL_PADDING_P. */
+ /* In FIELD_DECL, this is DECL_PADDING_P
+ In VAR_DECL, this is DECL_USEDBY_RETURN_P. */
unsigned decl_flag_3 : 1;
/* Logically, these two would go in a theoretical base shared by var and
parm decl. */
diff --git a/gcc/tree.h b/gcc/tree.h
index a863d2e50e5..73c0314dac1 100644
--- a/gcc/tree.h
+++ b/gcc/tree.h
@@ -3011,6 +3011,10 @@ extern void decl_value_expr_insert (tree, tree);
#define DECL_PADDING_P(NODE) \
(FIELD_DECL_CHECK (NODE)->decl_common.decl_flag_3)
+/* Used in a VAR_DECL to indicate that it is used by a return stmt. */
+#define DECL_USEDBY_RETURN_P(NODE) \
+ (VAR_DECL_CHECK (NODE)->decl_common.decl_flag_3)
+
/* Used in a FIELD_DECL to indicate whether this field is not a flexible
array member. This is only valid for the last array type field of a
structure. */
>
> On some targets(ppc64), for below case:
> ////case3.c
> typedef struct SA {double a[3]; long l; } A;
> A ret_arg_pt (A *a) {return *a;}
>
> The optimized GIMPLE code looks like:
> <retval> = *a_2(D);
> return <retval>;
> Here, <retval>(aka. RESULT_DECL) is MEM, and "aggregate_value_p"
> returns true for <retval>.
>
> * While for below case, the generated code is still suboptimal.
> ////case4.c
> typedef struct SA {double a[3];} A;
> A ret_arg_pt (A *a) {return *a;}
>
> The optimized GIMPLE code looks like:
> D.3951 = *a_2(D);
> return D.3951;
> The "return/assign" stmts are using D.3951(VAR_DECL) instead
> "<retval>(RESULT_DECL)". The mode of D.3951/<retval> is BLK.
> The RTL of D.3951 is MEM, and RTL of <retval> is PARALLEL. For
> PARALLEL, aggregate_value_p returns false.
>
> In function expand_assignment, there is code:
> if (TREE_CODE (to) == RESULT_DECL
> && (REG_P (to_rtx) || GET_CODE (to_rtx) == PARALLEL))
> This code can handle "<retval>", but can not handle "D.3951".
>
> I'm thinking of one way to handle this issue is to update the
> GIMPLE sequence as: "<retval> = *a_2(D); return <retval>;"
> Or, collecting VARs which are used by return stmts; and for
> assignments to those VARs, using sub scalar mode for the block
> move.
>
> Thanks for any comments and suggestions!
>
>
> BR,
> Jeff (Jiufu)
>
> ---
> gcc/expr.cc | 40 ++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 40 insertions(+)
>
> diff --git a/gcc/expr.cc b/gcc/expr.cc
> index d9407432ea5..420f9cf3662 100644
> --- a/gcc/expr.cc
> +++ b/gcc/expr.cc
> @@ -6045,6 +6045,46 @@ expand_assignment (tree to, tree from, bool
> nontemporal)
> return;
> }
>
> + if (TREE_CODE (from) == PARM_DECL && DECL_INCOMING_RTL (from)
> + && TYPE_MODE (TREE_TYPE (from)) == BLKmode
> + && (GET_CODE (DECL_INCOMING_RTL (from)) == PARALLEL
> + || REG_P (DECL_INCOMING_RTL (from))))
> + {
> + rtx parm = DECL_INCOMING_RTL (from);
> +
> + push_temp_slots ();
> + machine_mode mode;
> + mode = GET_CODE (parm) == PARALLEL
> + ? GET_MODE (XEXP (XVECEXP (parm, 0, 0), 0))
> + : word_mode;
> + int mode_size = GET_MODE_SIZE (mode).to_constant ();
> + int size = INTVAL (expr_size (from));
> +
> + /* If/How the parameter using submode, it dependes on the size and
> + position of the parameter. Here using heurisitic number. */
> + int hurstc_num = 8;
> + if (size < mode_size || (size % mode_size) != 0
> + || size > (mode_size * hurstc_num))
> + result = store_expr (from, to_rtx, 0, nontemporal, false);
> + else
> + {
> + rtx from_rtx
> + = expand_expr (from, NULL_RTX, GET_MODE (to_rtx), EXPAND_NORMAL);
> + for (int i = 0; i < size / mode_size; i++)
> + {
> + rtx temp = gen_reg_rtx (mode);
> + rtx src = adjust_address (from_rtx, mode, mode_size * i);
> + rtx dest = adjust_address (to_rtx, mode, mode_size * i);
> + emit_move_insn (temp, src);
> + emit_move_insn (dest, temp);
> + }
> + result = to_rtx;
> + }
> + preserve_temp_slots (result);
> + pop_temp_slots ();
> + return;
> + }
> +
> /* Compute FROM and store the value in the rtx we got. */
>
> push_temp_slots ();
