Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Sanimir Agovic <sanimir.agovic@intel.com>
To: gdb-patches@sourceware.org
Subject: [PATCH 03/10] vla: enable sizeof operator to work with variable length arrays
Date: Mon, 21 Oct 2013 14:40:00 -0000	[thread overview]
Message-ID: <1382366424-21010-4-git-send-email-sanimir.agovic@intel.com> (raw)
In-Reply-To: <1382366424-21010-1-git-send-email-sanimir.agovic@intel.com>

In C99 the sizeof operator computes the size of a variable length array
at runtime (6.5.3.4 The sizeof operator). This patch reflects the semantic
change in the debugger.

We now are able to get the size of a vla:

1| void foo (size_t n) {
2|   int vla[n];
3| }

(gdb) p sizeof(vla)

yields N.

2013-10-18  Sanimir Agovic  <sanimir.agovic@intel.com>
            Keven Boell  <keven.boell@intel.com>

	* eval.c (evaluate_subexp_for_sizeof): If the type passed to sizeof is
	dynamic	evaluate the argument to compute the length.
	* gdbtypes.c (is_dynamic_type): Make extern.
	* gdbtypes.c (is_dynamic_type): Make extern.

Change-Id: I490f695f7a164be5f1f824fe3ba33f805d1ab689
---
 gdb/eval.c     |    8 +++++++-
 gdb/gdbtypes.c |    2 +-
 gdb/gdbtypes.h |    2 ++
 3 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/gdb/eval.c b/gdb/eval.c
index e83bfdf..9e4afa3 100644
--- a/gdb/eval.c
+++ b/gdb/eval.c
@@ -3041,8 +3041,14 @@ evaluate_subexp_for_sizeof (struct expression *exp, int *pos)
       return value_from_longest (size_type, (LONGEST) TYPE_LENGTH (type));
 
     case OP_VAR_VALUE:
-      (*pos) += 4;
       type = check_typedef (SYMBOL_TYPE (exp->elts[pc + 2].symbol));
+      if (is_dynamic_type (type))
+	{
+	  val = evaluate_subexp (NULL_TYPE, exp, pos, EVAL_NORMAL);
+	  type = value_type (val);
+	}
+      else
+	(*pos) += 4;
       return
 	value_from_longest (size_type, (LONGEST) TYPE_LENGTH (type));
 
diff --git a/gdb/gdbtypes.c b/gdb/gdbtypes.c
index c0f375d..9774a4b 100644
--- a/gdb/gdbtypes.c
+++ b/gdb/gdbtypes.c
@@ -1624,7 +1624,7 @@ resolve_dynamic_prop (const struct dwarf2_prop *prop, CORE_ADDR address,
 
 /* Predicates if the type has dynamic values, which are not resolved yet.  */
 
-static int
+int
 is_dynamic_type (const struct type *type)
 {
   if (type == NULL)
diff --git a/gdb/gdbtypes.h b/gdb/gdbtypes.h
index 96e0ed7..1252fd1 100644
--- a/gdb/gdbtypes.h
+++ b/gdb/gdbtypes.h
@@ -1580,6 +1580,8 @@ extern struct type *lookup_unsigned_typename (const struct language_defn *,
 extern struct type *lookup_signed_typename (const struct language_defn *,
 					    struct gdbarch *, const char *);
 
+extern int is_dynamic_type (const struct type *type);
+
 extern struct type *resolve_dynamic_type (struct type *, CORE_ADDR);
 
 extern struct type *check_typedef (struct type *);
-- 
1.7.0.7


  parent reply	other threads:[~2013-10-21 14:40 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-21 14:40 C99 variable length array support Sanimir Agovic
2013-10-21 14:40 ` [PATCH 09/10] test: basic c99 vla tests Sanimir Agovic
2013-11-07 21:23   ` Tom Tromey
2013-10-21 14:40 ` [PATCH 07/10] test: evaluate pointers to C99 vla correctly Sanimir Agovic
2013-11-07 20:57   ` Tom Tromey
2013-11-08  6:37     ` Tom Tromey
2013-11-07 21:14   ` Tom Tromey
2013-10-21 14:40 ` [PATCH 02/10] type: add c99 variable length array support Sanimir Agovic
2013-11-07 19:02   ` Tom Tromey
2013-11-19 15:31     ` Agovic, Sanimir
2013-11-22 20:00       ` Tom Tromey
2013-11-27 17:08         ` Agovic, Sanimir
     [not found]         ` <0377C58828D86C4588AEEC42FC3B85A7176BC3DE@IRSMSX105.ger.corp.intel.com>
2013-11-27 17:15           ` Agovic, Sanimir
2013-11-07 19:10   ` Tom Tromey
2013-10-21 14:40 ` [PATCH 05/10] vla: allow side effects for sizeof argument Sanimir Agovic
2013-10-24 19:55   ` Tom Tromey
2013-10-25  8:13     ` Agovic, Sanimir
2013-10-28 21:00       ` Tom Tromey
2013-11-18  9:37         ` Agovic, Sanimir
2013-11-18 15:56           ` Tom Tromey
2013-11-20 12:18             ` Agovic, Sanimir
2013-11-19 17:08           ` Pedro Alves
2013-11-20 12:47             ` Agovic, Sanimir
2013-11-20 13:24               ` Pedro Alves
2013-10-21 14:40 ` [PATCH 06/10] vla: update type from newly created value Sanimir Agovic
2013-11-07 20:56   ` Tom Tromey
2013-11-20  7:56     ` Agovic, Sanimir
2013-11-20 11:02   ` Pedro Alves
2013-11-20 13:08     ` Agovic, Sanimir
2013-11-21 18:50       ` Pedro Alves
2013-11-23 19:27     ` Doug Evans
2013-10-21 14:40 ` Sanimir Agovic [this message]
2013-11-07 19:10   ` [PATCH 03/10] vla: enable sizeof operator to work with variable length arrays Tom Tromey
2013-10-21 14:40 ` [PATCH 04/10] vla: enable sizeof operator for indirection Sanimir Agovic
2013-11-07 19:57   ` Tom Tromey
2013-10-21 14:40 ` [PATCH 08/10] test: multi-dimensional c99 vla Sanimir Agovic
2013-11-07 21:19   ` Tom Tromey
2013-10-21 14:40 ` [PATCH 01/10] vla: introduce new bound type abstraction adapt uses Sanimir Agovic
2013-11-07 19:00   ` Tom Tromey
2013-11-18 11:15     ` Agovic, Sanimir
2013-10-21 14:40 ` [PATCH 10/10] test: add mi vla test Sanimir Agovic
2013-11-07 21:31   ` Tom Tromey
2013-11-21 18:52 ` C99 variable length array support Pedro Alves
2013-11-21 19:01   ` Pedro Alves
2013-11-21 19:10     ` Pedro Alves
2013-11-22 10:53       ` Agovic, Sanimir
2013-11-22 12:35         ` Pedro Alves
2013-11-22 17:06           ` Agovic, Sanimir

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=1382366424-21010-4-git-send-email-sanimir.agovic@intel.com \
    --to=sanimir.agovic@intel.com \
    --cc=gdb-patches@sourceware.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