[PATCH 1/2] gdb: fix ia64-tdep.c build with g++ 4.8

[PATCH 1/2] gdb: fix ia64-tdep.c build with g++ 4.8

[Simon Marchi]

When building with g++ 4.8, I get:

      CXX    ia64-tdep.o
    /home/smarchi/src/binutils-gdb/gdb/ia64-tdep.c:3862:1: error: could not convert '{ia64_allocate_new_rse_frame, ia64_store_argument_in_slot, ia64_set_function_addr}' from '<brace
-enclosed initializer list>' to 'const ia64_infcall_ops'
     };
     ^

This happens since commit 345bd07cce3 ("gdb: fix gdbarch_tdep ODR
violation"), which added default values for ia64_infcall_ops fields.  It
looks like g++ 4.8 doesn't like initializing the ia64_infcall_ops object
using the brace-enclosed initializer list when the ia64_infcall_ops
fields are assigned default values.

Later compilers don't have a problem with that, so I suppose that the
code is correct, but still, change it to make gcc 4.8 happy.  Don't
initialize the fields of ia64_infcall_ops directly, instead
default-initialize ia64_gdbarch_tdep::infcall_ops.

Change-Id: I35e3a61abd7b7bbcafe6cb207078c738c5266d76
---
 gdb/ia64-tdep.h | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/gdb/ia64-tdep.h b/gdb/ia64-tdep.h
index 82319d50894..d831484ee57 100644
--- a/gdb/ia64-tdep.h
+++ b/gdb/ia64-tdep.h
@@ -216,18 +216,17 @@ struct ia64_infcall_ops
 
      Should do nothing if this operation is not permitted by the OS.  */
   void (*allocate_new_rse_frame) (struct regcache *regcache, ULONGEST bsp,
-				  int sof) = nullptr;
+				  int sof);
 
   /* Store the argument stored in BUF into the appropriate location
      given the BSP and the SLOTNUM.  */
   void (*store_argument_in_slot) (struct regcache *regcache, CORE_ADDR bsp,
-				  int slotnum, gdb_byte *buf) = nullptr;
+				  int slotnum, gdb_byte *buf);
 
   /* For targets where we cannot call the function directly, store
      the address of the function we want to call at the location
      expected by the calling sequence.  */
-  void (*set_function_addr) (struct regcache *regcache, CORE_ADDR func_addr)
-    = nullptr;
+  void (*set_function_addr) (struct regcache *regcache, CORE_ADDR func_addr);
 };
 
 struct ia64_gdbarch_tdep : gdbarch_tdep
@@ -255,7 +254,7 @@ struct ia64_gdbarch_tdep : gdbarch_tdep
   /* ISA-specific data types.  */
   struct type *ia64_ext_type = nullptr;
 
-  struct ia64_infcall_ops infcall_ops;
+  struct ia64_infcall_ops infcall_ops {};
 };
 
 extern void ia64_write_pc (struct regcache *, CORE_ADDR);
-- 
2.33.1

[PATCH 2/2] gdb: fix array-view-selftests.c build with g++ 4.8

[Simon Marchi]

When building with g++ 4.8, I get:

    CXX    unittests/array-view-selftests.o
  /home/smarchi/src/binutils-gdb/gdb/unittests/array-view-selftests.c:123:42: error: expected 'class' before 'Container'
   template<template<typename ...> typename Container>
					    ^

I am no C++ template expert, but it looks like if I change "typename" for
"class", as the compiler kind of suggests, the code compiles.

Change-Id: I9c3edd29fb2b190069f0ce0dbf3bc3604d175f48
---
 gdb/unittests/array-view-selftests.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gdb/unittests/array-view-selftests.c b/gdb/unittests/array-view-selftests.c
index fe211a647b5..9df48db3912 100644
--- a/gdb/unittests/array-view-selftests.c
+++ b/gdb/unittests/array-view-selftests.c
@@ -120,7 +120,7 @@ check ()
 /* Check that there's no container->view conversion for containers of derived
    types or subclasses.  */
 
-template<template<typename ...> typename Container>
+template<template<typename ...> class Container>
 static constexpr bool
 check_ctor_from_container ()
 {
-- 
2.33.1

Re: [PATCH 2/2] gdb: fix array-view-selftests.c build with g++ 4.8

[Lancelot SIX]

On Thu, Nov 18, 2021 at 03:50:53PM -0500, Simon Marchi via Gdb-patches wrote:
> When building with g++ 4.8, I get:
> 
>     CXX    unittests/array-view-selftests.o
>   /home/smarchi/src/binutils-gdb/gdb/unittests/array-view-selftests.c:123:42: error: expected 'class' before 'Container'
>    template<template<typename ...> typename Container>
> 					    ^
> 
> I am no C++ template expert, but it looks like if I change "typename" for
> "class", as the compiler kind of suggests, the code compiles.

Indeed. Apparently 'typename' can be used instead of 'class' in a
template template parameter only starting c++17[1].

Thanks for taking care of this.

Lancelot.

[1] https://en.cppreference.com/w/cpp/language/template_parameters#Template_template_parameter

q> 
> Change-Id: I9c3edd29fb2b190069f0ce0dbf3bc3604d175f48
> ---
>  gdb/unittests/array-view-selftests.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/gdb/unittests/array-view-selftests.c b/gdb/unittests/array-view-selftests.c
> index fe211a647b5..9df48db3912 100644
> --- a/gdb/unittests/array-view-selftests.c
> +++ b/gdb/unittests/array-view-selftests.c
> @@ -120,7 +120,7 @@ check ()
>  /* Check that there's no container->view conversion for containers of derived
>     types or subclasses.  */
>  
> -template<template<typename ...> typename Container>
> +template<template<typename ...> class Container>
>  static constexpr bool
>  check_ctor_from_container ()
>  {
> -- 
> 2.33.1
> 

Re: [PATCH 2/2] gdb: fix array-view-selftests.c build with g++ 4.8

[Simon Marchi]

On 2021-11-18 4:11 p.m., Lancelot SIX via Gdb-patches wrote:
> On Thu, Nov 18, 2021 at 03:50:53PM -0500, Simon Marchi via Gdb-patches wrote:
>> When building with g++ 4.8, I get:
>>
>>     CXX    unittests/array-view-selftests.o
>>   /home/smarchi/src/binutils-gdb/gdb/unittests/array-view-selftests.c:123:42: error: expected 'class' before 'Container'
>>    template<template<typename ...> typename Container>
>> 					    ^
>>
>> I am no C++ template expert, but it looks like if I change "typename" for
>> "class", as the compiler kind of suggests, the code compiles.
> 
> Indeed. Apparently 'typename' can be used instead of 'class' in a
> template template parameter only starting c++17[1].
> 
> Thanks for taking care of this.
> 
> Lancelot.
> 
> [1] https://en.cppreference.com/w/cpp/language/template_parameters#Template_template_parameter

Ok, thanks for the info.  Since you acked this patch and patch 1/2
fixes a regression I introduced, I pushed the two patches.

Simon