public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
From: Richard Biener <rguenth@gcc.gnu.org>
To: gcc-cvs@gcc.gnu.org
Subject: [gcc r13-132] Embed real_value into REAL_CST
Date: Thu,  5 May 2022 11:19:55 +0000 (GMT)	[thread overview]
Message-ID: <20220505111955.4261A3856256@sourceware.org> (raw)

https://gcc.gnu.org/g:f1d8a2d9bc3a817309c299147ff85e5091b6693e

commit r13-132-gf1d8a2d9bc3a817309c299147ff85e5091b6693e
Author: Richard Biener <rguenther@suse.de>
Date:   Wed Apr 27 13:48:49 2022 +0200

    Embed real_value into REAL_CST
    
    The following removes the indirection to real_value from REAL_CST
    which doesn't seem to serve any useful purpose.  Any sharing can
    be achieved by sharing the actual REAL_CST (which is what usually
    happens when copying trees) and sharing of real_value amongst
    different REAL_CST doesn't happen as far as I can see and would
    only lead to further issues like mismatching type and real_value.
    
    2022-04-27  Richard Biener  <rguenther@suse.de>
    
            * tree-core.h (tree_real_cst::real_cst_ptr): Remove pointer
            to real_value field.
            (tree_real_cst::value): Add real_value field.
            * tree.h (TREE_REAL_CST_PTR): Adjust.
            * tree.cc (build_real): Remove separate allocation.
            * tree-streamer-in.cc (unpack_ts_real_cst_value_fields):
            Likewise.
    
    gcc/cp/
            * module.cc (trees_in::core_vals): Remove separate allocation
            for REAL_CST.

Diff:
---
 gcc/cp/module.cc        | 4 +---
 gcc/tree-core.h         | 2 +-
 gcc/tree-streamer-in.cc | 5 +----
 gcc/tree.cc             | 6 +-----
 gcc/tree.h              | 2 +-
 5 files changed, 5 insertions(+), 14 deletions(-)

diff --git a/gcc/cp/module.cc b/gcc/cp/module.cc
index cebf9c35c1d..18dabfcc9ac 100644
--- a/gcc/cp/module.cc
+++ b/gcc/cp/module.cc
@@ -6468,9 +6468,7 @@ trees_in::core_vals (tree t)
 
     case REAL_CST:
       if (const void *bytes = buf (sizeof (real_value)))
-	TREE_REAL_CST_PTR (t)
-	  = reinterpret_cast<real_value *> (memcpy (ggc_alloc<real_value> (),
-						    bytes, sizeof (real_value)));
+	memcpy (TREE_REAL_CST_PTR (t), bytes, sizeof (real_value));
       break;
 
     case STRING_CST:
diff --git a/gcc/tree-core.h b/gcc/tree-core.h
index 07e76e659e4..c79b8b28da3 100644
--- a/gcc/tree-core.h
+++ b/gcc/tree-core.h
@@ -1461,7 +1461,7 @@ struct GTY(()) tree_int_cst {
 
 struct GTY(()) tree_real_cst {
   struct tree_typed typed;
-  struct real_value * real_cst_ptr;
+  struct real_value value;
 };
 
 struct GTY(()) tree_fixed_cst {
diff --git a/gcc/tree-streamer-in.cc b/gcc/tree-streamer-in.cc
index a35a810f4d1..196f19c759f 100644
--- a/gcc/tree-streamer-in.cc
+++ b/gcc/tree-streamer-in.cc
@@ -190,7 +190,6 @@ unpack_ts_real_cst_value_fields (struct bitpack_d *bp, tree expr)
 {
   unsigned i;
   REAL_VALUE_TYPE r;
-  REAL_VALUE_TYPE *rp;
 
   /* Clear all bits of the real value type so that we can later do
      bitwise comparisons to see if two values are the same.  */
@@ -204,9 +203,7 @@ unpack_ts_real_cst_value_fields (struct bitpack_d *bp, tree expr)
   for (i = 0; i < SIGSZ; i++)
     r.sig[i] = (unsigned long) bp_unpack_value (bp, HOST_BITS_PER_LONG);
 
-  rp = ggc_alloc<real_value> ();
-  memcpy (rp, &r, sizeof (REAL_VALUE_TYPE));
-  TREE_REAL_CST_PTR (expr) = rp;
+  memcpy (TREE_REAL_CST_PTR (expr), &r, sizeof (REAL_VALUE_TYPE));
 }
 
 
diff --git a/gcc/tree.cc b/gcc/tree.cc
index 4cf3785270b..5e8876d2b38 100644
--- a/gcc/tree.cc
+++ b/gcc/tree.cc
@@ -2381,7 +2381,6 @@ tree
 build_real (tree type, REAL_VALUE_TYPE d)
 {
   tree v;
-  REAL_VALUE_TYPE *dp;
   int overflow = 0;
 
   /* dconst{1,2,m1,half} are used in various places in
@@ -2408,11 +2407,8 @@ build_real (tree type, REAL_VALUE_TYPE d)
      Consider doing it via real_convert now.  */
 
   v = make_node (REAL_CST);
-  dp = ggc_alloc<real_value> ();
-  memcpy (dp, &d, sizeof (REAL_VALUE_TYPE));
-
   TREE_TYPE (v) = type;
-  TREE_REAL_CST_PTR (v) = dp;
+  memcpy (TREE_REAL_CST_PTR (v), &d, sizeof (REAL_VALUE_TYPE));
   TREE_OVERFLOW (v) = overflow;
   return v;
 }
diff --git a/gcc/tree.h b/gcc/tree.h
index 8844471e9a5..82eb8ba39d2 100644
--- a/gcc/tree.h
+++ b/gcc/tree.h
@@ -1048,7 +1048,7 @@ extern void omp_clause_range_check_failed (const_tree, const char *, int,
 #define POLY_INT_CST_COEFF(NODE, I) \
   (POLY_INT_CST_CHECK (NODE)->poly_int_cst.coeffs[I])
 
-#define TREE_REAL_CST_PTR(NODE) (REAL_CST_CHECK (NODE)->real_cst.real_cst_ptr)
+#define TREE_REAL_CST_PTR(NODE) (&REAL_CST_CHECK (NODE)->real_cst.value)
 #define TREE_REAL_CST(NODE) (*TREE_REAL_CST_PTR (NODE))
 
 #define TREE_FIXED_CST_PTR(NODE) \


                 reply	other threads:[~2022-05-05 11:19 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220505111955.4261A3856256@sourceware.org \
    --to=rguenth@gcc.gnu.org \
    --cc=gcc-cvs@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).