This is the mail archive of the mailing list for the GDB project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[RFA/commit] gdb/dwarf2read: Minimal handling of non-constant struct sizes.


Using the gdb.ada/var_rec_arr.exp test, where the program declares
an array of variant records...

   type Record_Type (I : Small_Type := 0) is record
      S : String (1 .. I);
   end record;
   type Array_Type is array (Integer range <>) of Record_Type;

... and then a variable A1 of type Array_Type, the following command
ocassionally trigger an internal error trying to allocate more memory
than we have left:

    (gdb) ptype a1(1)
    [...]/utils.c:1089: internal-error: virtual memory exhausted.
    A problem internal to GDB has been detected,

What happens is that recent versions of GNAT are able to generate
DWARF expressions for type Record_Type, and therefore the record's
DW_AT_byte_size is not a constant, which unfortunately breaks
an assumption made by dwarf2read.c:read_structure_type when it does:

   attr = dwarf2_attr (die, DW_AT_byte_size, cu);
   if (attr)
       TYPE_LENGTH (type) = DW_UNSND (attr);

As a result of this, when ada_evaluate_subexp tries to create
a value_zero for a1(1) while processing the OP_FUNCALL operator
as part of evaluating the subscripting operation in no-side-effect
mode, we try to allocate a value with a bogus size, potentially
triggering the out-of-memory internal error.

This patch avoids this issue by setting the length to zero in
this case.  Until we decide to start supporting dynamic type
lengths in GDB's type struct, and it's not clear yet that
this is worth the effort (see added comment), that's probably
the best we can do.


        * dwarf2read.c (read_structure_type): Set the type's length
        to zero if it has a DW_AT_byte_size attribute which is not
        a constant.


        * testsuite/gdb.ada/var_rec_arr.exp: Add "ptype a1(1)" test.

Tested on x86_64-linux.


 gdb/dwarf2read.c                      | 15 ++++++++++++++-
 gdb/testsuite/gdb.ada/var_rec_arr.exp |  6 ++++++
 2 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/gdb/dwarf2read.c b/gdb/dwarf2read.c
index 87dc8b4..ceb915f 100644
--- a/gdb/dwarf2read.c
+++ b/gdb/dwarf2read.c
@@ -13218,7 +13218,20 @@ read_structure_type (struct die_info *die, struct dwarf2_cu *cu)
   attr = dwarf2_attr (die, DW_AT_byte_size, cu);
   if (attr)
-      TYPE_LENGTH (type) = DW_UNSND (attr);
+      if (attr_form_is_constant (attr))
+        TYPE_LENGTH (type) = DW_UNSND (attr);
+      else
+	{
+	  /* For the moment, dynamic type sizes are not supported
+	     by GDB's struct type.  The actual size is determined
+	     on-demand when resolving the type of a given object,
+	     so set the type's length to zero for now.  Otherwise,
+	     we record an expression as the length, and that expression
+	     could lead to a very large value, which could eventually
+	     lead to us trying to allocate that much memory when creating
+	     a value of that type.  */
+          TYPE_LENGTH (type) = 0;
+	}
diff --git a/gdb/testsuite/gdb.ada/var_rec_arr.exp b/gdb/testsuite/gdb.ada/var_rec_arr.exp
index 82ca857..66525e7 100644
--- a/gdb/testsuite/gdb.ada/var_rec_arr.exp
+++ b/gdb/testsuite/gdb.ada/var_rec_arr.exp
@@ -49,3 +49,9 @@ gdb_test "print a2(2)" \
 gdb_test "print a2(3)" \
          " = \\(i => 0, s => \"\"\\)"
+gdb_test "ptype a1(1)" \
+         [multi_line "type = record" \
+                     "    i: pck\\.small_type;" \
+                     "    s: access array \\((<>|1 \\.\\. i)\\) of character;" \
+                     "end record"]

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]