diff mbox series

DWARF reader: use size_t for more DWARF expression length

Message ID 20220118130354.8649-1-mark@klomp.org
State New
Headers
Series DWARF reader: use size_t for more DWARF expression length |

Commit Message

Mark Wielaard Jan. 18, 2022, 1:03 p.m. UTC 
  From: Mark Wielaard <mark@klomp.org>

commit 6e0d3f7 "DWARF reader: use size_t for DWARF expression
length" is a partial fix for the assumption that size_t == uint64_t.

This commit updates a couple more functions where this (wrong)
assumption was made.

	* src/abg-dwarf-reader.cc (eval_last_constant_dwarf_sub_expr):
	Change expr_len argument type to size_t*.
	(op_pushes_constant_value): Change ops_len and index argument
	types to size_t; change next_index argument type to size_t&.
	(op_pushes_non_constant_value): Likewise.
	(op_is_arith_logic): Change expr_len and index argument
	types to size_t; change next_index argument type to size_t&.
	(op_is_control_flow): Likewise.

Fixes: 16207c4af7bc ("Bug 28191 - Interpret DWARF 5 addrx locations")
Signed-off-by: Mark Wielaard <mark@klomp.org>
---
 src/abg-dwarf-reader.cc | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

Comments

Mark Wielaard Jan. 18, 2022, 1:14 p.m. UTC | #1 
Hi,

On Tue, 2022-01-18 at 14:03 +0100, Mark J. Wielaard wrote:
> From: Mark Wielaard <mark@klomp.org>
> 
> commit 6e0d3f7 "DWARF reader: use size_t for DWARF expression
> length" is a partial fix for the assumption that size_t == uint64_t.
> 
> This commit updates a couple more functions where this (wrong)
> assumption was made.
> 
> 	* src/abg-dwarf-reader.cc (eval_last_constant_dwarf_sub_expr):
> 	Change expr_len argument type to size_t*.
> 	(op_pushes_constant_value): Change ops_len and index argument
> 	types to size_t; change next_index argument type to size_t&.
> 	(op_pushes_non_constant_value): Likewise.
> 	(op_is_arith_logic): Change expr_len and index argument
> 	types to size_t; change next_index argument type to size_t&.
> 	(op_is_control_flow): Likewise.
> 
> Fixes: 16207c4af7bc ("Bug 28191 - Interpret DWARF 5 addrx locations")
> Signed-off-by: Mark Wielaard <mark@klomp.org>

Sorry, this is equivalent to Giuliano's patch in 
https://sourceware.org/pipermail/libabigail/2022q1/004114.html
I was confused about was was/wasn't committed.

Cheers,

Mark
diff mbox series

Patch

diff --git a/src/abg-dwarf-reader.cc b/src/abg-dwarf-reader.cc
index 7ee8c04a..d8545b4c 100644
--- a/src/abg-dwarf-reader.cc
+++ b/src/abg-dwarf-reader.cc
@@ -363,7 +363,7 @@  die_is_at_class_scope(const read_context& ctxt,
 		      Dwarf_Die& class_scope_die);
 static bool
 eval_last_constant_dwarf_sub_expr(Dwarf_Op*	expr,
-				  uint64_t	expr_len,
+				  size_t	expr_len,
 				  int64_t&	value,
 				  bool&	is_tls_address);
 
@@ -7732,9 +7732,9 @@  die_location_expr(const Dwarf_Die* die,
 /// value onto the DEVM stack, false otherwise.
 static bool
 op_pushes_constant_value(Dwarf_Op*			ops,
-			 uint64_t			ops_len,
-			 uint64_t			index,
-			 uint64_t&			next_index,
+			 size_t				ops_len,
+			 size_t				index,
+			 size_t&			next_index,
 			 dwarf_expr_eval_context&	ctxt)
 {
   ABG_ASSERT(index < ops_len);
@@ -7896,9 +7896,9 @@  op_pushes_constant_value(Dwarf_Op*			ops,
 /// non-constant value onto the DEVM stack, false otherwise.
 static bool
 op_pushes_non_constant_value(Dwarf_Op* ops,
-			     uint64_t ops_len,
-			     uint64_t index,
-			     uint64_t& next_index,
+			     size_t ops_len,
+			     size_t index,
+			     size_t& next_index,
 			     dwarf_expr_eval_context& ctxt)
 {
   ABG_ASSERT(index < ops_len);
@@ -8146,9 +8146,9 @@  op_manipulates_stack(Dwarf_Op* expr,
 /// arithmetic or logic operation.
 static bool
 op_is_arith_logic(Dwarf_Op* expr,
-		  uint64_t expr_len,
-		  uint64_t index,
-		  uint64_t& next_index,
+		  size_t expr_len,
+		  size_t index,
+		  size_t& next_index,
 		  dwarf_expr_eval_context& ctxt)
 {
   ABG_ASSERT(index < expr_len);
@@ -8279,9 +8279,9 @@  op_is_arith_logic(Dwarf_Op* expr,
 /// control flow operation, false otherwise.
 static bool
 op_is_control_flow(Dwarf_Op* expr,
-		   uint64_t expr_len,
-		   uint64_t index,
-		   uint64_t& next_index,
+		   size_t expr_len,
+		   size_t index,
+		   size_t& next_index,
 		   dwarf_expr_eval_context& ctxt)
 {
   ABG_ASSERT(index < expr_len);