remove workaround for GCC 4.1-4.3

Message ID ZROy7OxMTzQrhvRn@tucnak
State Unresolved
Headers
Series remove workaround for GCC 4.1-4.3 |

Checks

Context Check Description
snail/gcc-patch-check warning Git am fail log

Commit Message

Jakub Jelinek Sept. 27, 2023, 4:43 a.m. UTC
  Hi!

While looking into vec.h, I've noticed we still have a workaround for
GCC 4.1-4.3 bugs.
As we now use C++11 and thus need to be built by GCC 4.8 or later,
I think this is now never used.

Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

2023-09-27  Jakub Jelinek  <jakub@redhat.com>

	* system.h (BROKEN_VALUE_INITIALIZATION): Don't define.
	* vec.h (vec_default_construct): Remove BROKEN_VALUE_INITIALIZATION
	workaround.
	* function.cc (assign_parm_find_data_types): Likewise.


	Jakub
  

Comments

Bernhard Reutner-Fischer Sept. 27, 2023, 4:46 a.m. UTC | #1
On 27 September 2023 06:43:24 CEST, Jakub Jelinek <jakub@redhat.com> wrote:
>Hi!
>
>While looking into vec.h, I've noticed we still have a workaround for
>GCC 4.1-4.3 bugs.


This is https://gcc.gnu.org/PR105656
thanks,

>As we now use C++11 and thus need to be built by GCC 4.8 or later,
>I think this is now never used.
>
>Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
>
>2023-09-27  Jakub Jelinek  <jakub@redhat.com>
>
>	* system.h (BROKEN_VALUE_INITIALIZATION): Don't define.
>	* vec.h (vec_default_construct): Remove BROKEN_VALUE_INITIALIZATION
>	workaround.
>	* function.cc (assign_parm_find_data_types): Likewise.
>
>--- gcc/system.h.jj	2023-04-22 20:14:03.502203388 +0200
>+++ gcc/system.h	2023-09-26 16:41:44.384204843 +0200
>@@ -905,12 +905,6 @@ extern void fancy_abort (const char *, i
> /* Some compilers do not allow the use of unsigned char in bitfields.  */
> #define BOOL_BITFIELD unsigned int
> 
>-/* GCC older than 4.4 have broken C++ value initialization handling, see
>-   PR11309, PR30111, PR33916, PR82939 and PR84405 for more details.  */
>-#if GCC_VERSION > 0 && GCC_VERSION < 4004 && !defined(__clang__)
>-# define BROKEN_VALUE_INITIALIZATION
>-#endif
>-
> /* As the last action in this file, we poison the identifiers that
>    shouldn't be used.  Note, luckily gcc-3.0's token-based integrated
>    preprocessor won't trip on poisoned identifiers that arrive from
>--- gcc/vec.h.jj	2023-07-11 13:40:40.392430080 +0200
>+++ gcc/vec.h	2023-09-26 16:44:30.637902359 +0200
>@@ -512,21 +512,6 @@ template <typename T>
> inline void
> vec_default_construct (T *dst, unsigned n)
> {
>-#ifdef BROKEN_VALUE_INITIALIZATION
>-  /* Versions of GCC before 4.4 sometimes leave certain objects
>-     uninitialized when value initialized, though if the type has
>-     user defined default ctor, that ctor is invoked.  As a workaround
>-     perform clearing first and then the value initialization, which
>-     fixes the case when value initialization doesn't initialize due to
>-     the bugs and should initialize to all zeros, but still allows
>-     vectors for types with user defined default ctor that initializes
>-     some or all elements to non-zero.  If T has no user defined
>-     default ctor and some non-static data members have user defined
>-     default ctors that initialize to non-zero the workaround will
>-     still not work properly; in that case we just need to provide
>-     user defined default ctor.  */
>-  memset (dst, '\0', sizeof (T) * n);
>-#endif
>   for ( ; n; ++dst, --n)
>     ::new (static_cast<void*>(dst)) T ();
> }
>--- gcc/function.cc.jj	2023-07-11 13:40:38.992448821 +0200
>+++ gcc/function.cc	2023-09-26 16:44:54.865567722 +0200
>@@ -2429,15 +2429,7 @@ assign_parm_find_data_types (struct assi
> {
>   int unsignedp;
> 
>-#ifndef BROKEN_VALUE_INITIALIZATION
>   *data = assign_parm_data_one ();
>-#else
>-  /* Old versions of GCC used to miscompile the above by only initializing
>-     the members with explicit constructors and copying garbage
>-     to the other members.  */
>-  assign_parm_data_one zero_data = {};
>-  *data = zero_data;
>-#endif
> 
>   /* NAMED_ARG is a misnomer.  We really mean 'non-variadic'. */
>   if (!cfun->stdarg)
>
>	Jakub
>
  
Richard Biener Sept. 27, 2023, 5:14 a.m. UTC | #2
> Am 27.09.2023 um 06:43 schrieb Jakub Jelinek <jakub@redhat.com>:
> 
> Hi!
> 
> While looking into vec.h, I've noticed we still have a workaround for
> GCC 4.1-4.3 bugs.
> As we now use C++11 and thus need to be built by GCC 4.8 or later,
> I think this is now never used.
> 
> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

Ok

Richard 

> 2023-09-27  Jakub Jelinek  <jakub@redhat.com>
> 
>    * system.h (BROKEN_VALUE_INITIALIZATION): Don't define.
>    * vec.h (vec_default_construct): Remove BROKEN_VALUE_INITIALIZATION
>    workaround.
>    * function.cc (assign_parm_find_data_types): Likewise.
> 
> --- gcc/system.h.jj    2023-04-22 20:14:03.502203388 +0200
> +++ gcc/system.h    2023-09-26 16:41:44.384204843 +0200
> @@ -905,12 +905,6 @@ extern void fancy_abort (const char *, i
> /* Some compilers do not allow the use of unsigned char in bitfields.  */
> #define BOOL_BITFIELD unsigned int
> 
> -/* GCC older than 4.4 have broken C++ value initialization handling, see
> -   PR11309, PR30111, PR33916, PR82939 and PR84405 for more details.  */
> -#if GCC_VERSION > 0 && GCC_VERSION < 4004 && !defined(__clang__)
> -# define BROKEN_VALUE_INITIALIZATION
> -#endif
> -
> /* As the last action in this file, we poison the identifiers that
>    shouldn't be used.  Note, luckily gcc-3.0's token-based integrated
>    preprocessor won't trip on poisoned identifiers that arrive from
> --- gcc/vec.h.jj    2023-07-11 13:40:40.392430080 +0200
> +++ gcc/vec.h    2023-09-26 16:44:30.637902359 +0200
> @@ -512,21 +512,6 @@ template <typename T>
> inline void
> vec_default_construct (T *dst, unsigned n)
> {
> -#ifdef BROKEN_VALUE_INITIALIZATION
> -  /* Versions of GCC before 4.4 sometimes leave certain objects
> -     uninitialized when value initialized, though if the type has
> -     user defined default ctor, that ctor is invoked.  As a workaround
> -     perform clearing first and then the value initialization, which
> -     fixes the case when value initialization doesn't initialize due to
> -     the bugs and should initialize to all zeros, but still allows
> -     vectors for types with user defined default ctor that initializes
> -     some or all elements to non-zero.  If T has no user defined
> -     default ctor and some non-static data members have user defined
> -     default ctors that initialize to non-zero the workaround will
> -     still not work properly; in that case we just need to provide
> -     user defined default ctor.  */
> -  memset (dst, '\0', sizeof (T) * n);
> -#endif
>   for ( ; n; ++dst, --n)
>     ::new (static_cast<void*>(dst)) T ();
> }
> --- gcc/function.cc.jj    2023-07-11 13:40:38.992448821 +0200
> +++ gcc/function.cc    2023-09-26 16:44:54.865567722 +0200
> @@ -2429,15 +2429,7 @@ assign_parm_find_data_types (struct assi
> {
>   int unsignedp;
> 
> -#ifndef BROKEN_VALUE_INITIALIZATION
>   *data = assign_parm_data_one ();
> -#else
> -  /* Old versions of GCC used to miscompile the above by only initializing
> -     the members with explicit constructors and copying garbage
> -     to the other members.  */
> -  assign_parm_data_one zero_data = {};
> -  *data = zero_data;
> -#endif
> 
>   /* NAMED_ARG is a misnomer.  We really mean 'non-variadic'. */
>   if (!cfun->stdarg)
> 
>    Jakub
>
  
Bernhard Reutner-Fischer Sept. 29, 2023, 5:57 p.m. UTC | #3
On 27 September 2023 06:46:29 CEST, Bernhard Reutner-Fischer <rep.dot.nop@gmail.com> wrote:
>On 27 September 2023 06:43:24 CEST, Jakub Jelinek <jakub@redhat.com> wrote:
>>Hi!
>>
>>While looking into vec.h, I've noticed we still have a workaround for
>>GCC 4.1-4.3 bugs.
>
>
>This is https://gcc.gnu.org/PR105656
>thanks,

Mere cosmetics, but just for consistency:
I think you fat-fingered the number.
Since I would not suggest to backport this, should I close this reminder PR with a manually crafted commit link, or is there a preferred, other way to adjust the commited PR reference (also for the typoed PR which got the unrelated commit associated) and close the PR?

thanks,

>
>>As we now use C++11 and thus need to be built by GCC 4.8 or later,
>>I think this is now never used.
>>
>>Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
>>
>>2023-09-27  Jakub Jelinek  <jakub@redhat.com>
>>
>>	* system.h (BROKEN_VALUE_INITIALIZATION): Don't define.
>>	* vec.h (vec_default_construct): Remove BROKEN_VALUE_INITIALIZATION
>>	workaround.
>>	* function.cc (assign_parm_find_data_types): Likewise.
>>
>>--- gcc/system.h.jj	2023-04-22 20:14:03.502203388 +0200
>>+++ gcc/system.h	2023-09-26 16:41:44.384204843 +0200
>>@@ -905,12 +905,6 @@ extern void fancy_abort (const char *, i
>> /* Some compilers do not allow the use of unsigned char in bitfields.  */
>> #define BOOL_BITFIELD unsigned int
>> 
>>-/* GCC older than 4.4 have broken C++ value initialization handling, see
>>-   PR11309, PR30111, PR33916, PR82939 and PR84405 for more details.  */
>>-#if GCC_VERSION > 0 && GCC_VERSION < 4004 && !defined(__clang__)
>>-# define BROKEN_VALUE_INITIALIZATION
>>-#endif
>>-
>> /* As the last action in this file, we poison the identifiers that
>>    shouldn't be used.  Note, luckily gcc-3.0's token-based integrated
>>    preprocessor won't trip on poisoned identifiers that arrive from
>>--- gcc/vec.h.jj	2023-07-11 13:40:40.392430080 +0200
>>+++ gcc/vec.h	2023-09-26 16:44:30.637902359 +0200
>>@@ -512,21 +512,6 @@ template <typename T>
>> inline void
>> vec_default_construct (T *dst, unsigned n)
>> {
>>-#ifdef BROKEN_VALUE_INITIALIZATION
>>-  /* Versions of GCC before 4.4 sometimes leave certain objects
>>-     uninitialized when value initialized, though if the type has
>>-     user defined default ctor, that ctor is invoked.  As a workaround
>>-     perform clearing first and then the value initialization, which
>>-     fixes the case when value initialization doesn't initialize due to
>>-     the bugs and should initialize to all zeros, but still allows
>>-     vectors for types with user defined default ctor that initializes
>>-     some or all elements to non-zero.  If T has no user defined
>>-     default ctor and some non-static data members have user defined
>>-     default ctors that initialize to non-zero the workaround will
>>-     still not work properly; in that case we just need to provide
>>-     user defined default ctor.  */
>>-  memset (dst, '\0', sizeof (T) * n);
>>-#endif
>>   for ( ; n; ++dst, --n)
>>     ::new (static_cast<void*>(dst)) T ();
>> }
>>--- gcc/function.cc.jj	2023-07-11 13:40:38.992448821 +0200
>>+++ gcc/function.cc	2023-09-26 16:44:54.865567722 +0200
>>@@ -2429,15 +2429,7 @@ assign_parm_find_data_types (struct assi
>> {
>>   int unsignedp;
>> 
>>-#ifndef BROKEN_VALUE_INITIALIZATION
>>   *data = assign_parm_data_one ();
>>-#else
>>-  /* Old versions of GCC used to miscompile the above by only initializing
>>-     the members with explicit constructors and copying garbage
>>-     to the other members.  */
>>-  assign_parm_data_one zero_data = {};
>>-  *data = zero_data;
>>-#endif
>> 
>>   /* NAMED_ARG is a misnomer.  We really mean 'non-variadic'. */
>>   if (!cfun->stdarg)
>>
>>	Jakub
>>
>
  

Patch

--- gcc/system.h.jj	2023-04-22 20:14:03.502203388 +0200
+++ gcc/system.h	2023-09-26 16:41:44.384204843 +0200
@@ -905,12 +905,6 @@  extern void fancy_abort (const char *, i
 /* Some compilers do not allow the use of unsigned char in bitfields.  */
 #define BOOL_BITFIELD unsigned int
 
-/* GCC older than 4.4 have broken C++ value initialization handling, see
-   PR11309, PR30111, PR33916, PR82939 and PR84405 for more details.  */
-#if GCC_VERSION > 0 && GCC_VERSION < 4004 && !defined(__clang__)
-# define BROKEN_VALUE_INITIALIZATION
-#endif
-
 /* As the last action in this file, we poison the identifiers that
    shouldn't be used.  Note, luckily gcc-3.0's token-based integrated
    preprocessor won't trip on poisoned identifiers that arrive from
--- gcc/vec.h.jj	2023-07-11 13:40:40.392430080 +0200
+++ gcc/vec.h	2023-09-26 16:44:30.637902359 +0200
@@ -512,21 +512,6 @@  template <typename T>
 inline void
 vec_default_construct (T *dst, unsigned n)
 {
-#ifdef BROKEN_VALUE_INITIALIZATION
-  /* Versions of GCC before 4.4 sometimes leave certain objects
-     uninitialized when value initialized, though if the type has
-     user defined default ctor, that ctor is invoked.  As a workaround
-     perform clearing first and then the value initialization, which
-     fixes the case when value initialization doesn't initialize due to
-     the bugs and should initialize to all zeros, but still allows
-     vectors for types with user defined default ctor that initializes
-     some or all elements to non-zero.  If T has no user defined
-     default ctor and some non-static data members have user defined
-     default ctors that initialize to non-zero the workaround will
-     still not work properly; in that case we just need to provide
-     user defined default ctor.  */
-  memset (dst, '\0', sizeof (T) * n);
-#endif
   for ( ; n; ++dst, --n)
     ::new (static_cast<void*>(dst)) T ();
 }
--- gcc/function.cc.jj	2023-07-11 13:40:38.992448821 +0200
+++ gcc/function.cc	2023-09-26 16:44:54.865567722 +0200
@@ -2429,15 +2429,7 @@  assign_parm_find_data_types (struct assi
 {
   int unsignedp;
 
-#ifndef BROKEN_VALUE_INITIALIZATION
   *data = assign_parm_data_one ();
-#else
-  /* Old versions of GCC used to miscompile the above by only initializing
-     the members with explicit constructors and copying garbage
-     to the other members.  */
-  assign_parm_data_one zero_data = {};
-  *data = zero_data;
-#endif
 
   /* NAMED_ARG is a misnomer.  We really mean 'non-variadic'. */
   if (!cfun->stdarg)