public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Nathan Sidwell <nathan@acm.org>
To: GCC Patches <gcc-patches@gcc.gnu.org>
Cc: Jason Merrill <jason@redhat.com>
Subject: c++: Adjust synthetic template parm creation
Date: Tue, 25 Oct 2022 16:16:11 -0400	[thread overview]
Message-ID: <f068e565-e2a0-2b51-cb63-952e16b7c024@acm.org> (raw)

[-- Attachment #1: Type: text/plain, Size: 866 bytes --]


We intend to mark synthetic template parameters (coming from use of auto
parms), as DECL_VIRTUAL_P.  The API of process_template_parm is
awkwardly confusing, and we were marking the previous template parm
(unless this was the first parm).  process_template_parm returns the list
of parms, when most (all?) users really want the newly-added final node.
That's a bigger change, so let's not do it right now.  With this, we
correctly mark such synthetic parms DECL_VIRTUAL_P.

I fell over this trying to update template lambda mangling.  The API of 
process_template_parm is somewhat awkaward, as it returns the whole list, and we 
end up doing a lot of tree_last calls, to get to the newly added node.  I think 
all (or nearly all?) uses could be adapted to it returning the newly added list 
node, but I didn't want to go there right now.

nathan

-- 
Nathan Sidwell

[-- Attachment #2: 0001-c-Adjust-synthetic-template-parm-creation.patch --]
[-- Type: text/x-patch, Size: 3740 bytes --]

From 43e654afeba484d75fbee080262a038c1da00ad5 Mon Sep 17 00:00:00 2001
From: Nathan Sidwell <nathan@acm.org>
Date: Tue, 25 Oct 2022 09:39:00 -0400
Subject: [PATCH] c++: Adjust synthetic template parm creation

We intend to mark synthetic template parameters (coming from use of auto
parms), as DECL_VIRTUAL_P.  The API of process_template_parm is
awkwardly confusing, and we were marking the previous template parm
(unless this was the first parm).  process_template_parm returns the list
of parms, when most (all?) users really want the newly-added final node.
That's a bigger change, so let's not do it right now.  With this, we
correctly mark such synthetic parms DECL_VIRTUAL_P.

	gcc/cp/
	* parser.cc (synthesize_implicit_template_parm): Fix thinko about
	mark the new parm DECL_VIRTUAL_P.  Avoid unneccessary tree_last call.
---
 gcc/cp/parser.cc | 26 +++++++++++++++-----------
 1 file changed, 15 insertions(+), 11 deletions(-)

diff --git a/gcc/cp/parser.cc b/gcc/cp/parser.cc
index a39c5f0d24b..e685f190b3d 100644
--- a/gcc/cp/parser.cc
+++ b/gcc/cp/parser.cc
@@ -48996,12 +48996,11 @@ synthesize_implicit_template_parm  (cp_parser *parser, tree constr)
 
   tree proto = constr ? DECL_INITIAL (constr) : NULL_TREE;
   tree synth_id = make_generic_type_name ();
-  tree synth_tmpl_parm;
   bool non_type = false;
 
   /* Synthesize the type template parameter.  */
   gcc_assert(!proto || TREE_CODE (proto) == TYPE_DECL);
-  synth_tmpl_parm = finish_template_type_parm (class_type_node, synth_id);
+  tree synth_tmpl_parm = finish_template_type_parm (class_type_node, synth_id);
 
   if (become_template)
     current_template_parms = tree_cons (size_int (current_template_depth + 1),
@@ -49016,22 +49015,27 @@ synthesize_implicit_template_parm  (cp_parser *parser, tree constr)
 			     node,
 			     /*non_type=*/non_type,
 			     /*param_pack=*/false);
+  // Process_template_parm returns the list of parms, and
+  // parser->implicit_template_parms holds the final node of the parm
+  // list.  We really want to manipulate the newly appended element.
+  gcc_checking_assert (!parser->implicit_template_parms
+		       || parser->implicit_template_parms == new_parm);
+  if (parser->implicit_template_parms)
+    new_parm = TREE_CHAIN (new_parm);
+  gcc_checking_assert (!TREE_CHAIN (new_parm));
+
+  // Record the last implicit parm node
+  parser->implicit_template_parms = new_parm;
 
   /* Mark the synthetic declaration "virtual". This is used when
      comparing template-heads to determine if whether an abbreviated
      function template is equivalent to an explicit template.
 
-     Note that DECL_ARTIFICIAL is used elsewhere for template parameters.  */
+     Note that DECL_ARTIFICIAL is used elsewhere for template
+     parameters.  */
   if (TREE_VALUE (new_parm) != error_mark_node)
     DECL_VIRTUAL_P (TREE_VALUE (new_parm)) = true;
 
-  // Chain the new parameter to the list of implicit parameters.
-  if (parser->implicit_template_parms)
-    parser->implicit_template_parms
-      = TREE_CHAIN (parser->implicit_template_parms);
-  else
-    parser->implicit_template_parms = new_parm;
-
   tree new_decl = get_local_decls ();
   if (non_type)
     /* Return the TEMPLATE_PARM_INDEX, not the PARM_DECL.  */
@@ -49059,7 +49063,7 @@ synthesize_implicit_template_parm  (cp_parser *parser, tree constr)
 
   /* If the new parameter was constrained, we need to add that to the
      constraints in the template parameter list.  */
-  if (tree req = TEMPLATE_PARM_CONSTRAINTS (tree_last (new_parm)))
+  if (tree req = TEMPLATE_PARM_CONSTRAINTS (new_parm))
     {
       tree reqs = TEMPLATE_PARMS_CONSTRAINTS (current_template_parms);
       reqs = combine_constraint_expressions (reqs, req);
-- 
2.37.3


                 reply	other threads:[~2022-10-25 20:16 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=f068e565-e2a0-2b51-cb63-952e16b7c024@acm.org \
    --to=nathan@acm.org \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jason@redhat.com \
    /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).