[PATCH, i386] Move common definitions of gnu-user.h and gnu-user64.h to separate file

[Richard Guenther]

On Fri, Apr 13, 2012 at 12:37 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote:
>> Hello!
>>
>>> Here is a patch which creates new gnu-user-common.h file and moves all
>>> common gnu-user.h and gnu-user64.h definitions to this new file. New
>>> file is required to avoid duplication of Android specific changes in
>>> gnu-user.h and gnu-user64.h. This patch is actually a non Android
>>> specific part of previously submitted patch to support Android in x86
>>> target (http://gcc.gnu.org/ml/gcc-patches/2012-04/msg00127.html).
>>> Bootstrapped and tested on linux-x86_64. Is it OK for mainline and
>>> 4.7?
>>
>>> 2012-04-13  Enkovich Ilya  <ilya.enkovich@intel.com>
>>>
>>>       * config/i386/gnu-user.h: Include gnu-user-common.h.
>>>       (CPP_SPEC): Removed.
>>>       (CC1_SPEC): Removed.
>>>       (ENDFILE_SPEC): Removed.
>>>       (DEFAULT_PCC_STRUCT_RETURN): Removed.
>>>       (TARGET_TLS_DIRECT_SEG_REFS_DEFAULT): Removed.
>>>       (TARGET_OS_CPP_BUILTINS): Removed.
>>>       (LIBGCC2_HAS_TF_MODE): Removed.
>>>       (LIBGCC2_TF_CEXT): Removed.
>>>       (TF_SIZE): Removed.
>>>       (TARGET_ASM_FILE_END): Removed.
>>>       (STACK_CHECK_MOVING_SP): Removed.
>>>       (STACK_CHECK_STATIC_BUILTIN): Removed.
>>>       (TARGET_THREAD_SSP_OFFSET): Removed.
>>>       (TARGET_CAN_SPLIT_STACK): Removed.
>>>       (TARGET_THREAD_SPLIT_STACK_OFFSET): Removed.
>>>
>>>       * config/i386/gnu-user64.h: Likewise.
>>>
>>>       * config/i386/gnu-user-common.h: New.
>>
>> This ChangeLog is wrong, you didn't remove these files but move them
>> to new file.
>>
>> +#ifdef TARGET_LIBC_PROVIDES_SSP
>> +/* i386 glibc provides __stack_chk_guard in %gs:0x14,
>> +   x32 glibc provides it in %fs:0x18.
>> +   x86_64 glibc provides it in %fs:0x28.  */
>> +#define TARGET_THREAD_SSP_OFFSET \
>> +  (TARGET_64BIT ? (TARGET_X32 ? 0x18 : 0x28) : 0x14)
>> +
>> +/* We steal the last transactional memory word.  */
>> +#define TARGET_CAN_SPLIT_STACK
>> +#define TARGET_THREAD_SPLIT_STACK_OFFSET \
>> +  (TARGET_64BIT ? (TARGET_X32 ? 0x40 : 0x70) : 0x30)
>> +#endif
>>
>> Shouldn't TARGET_64BIT part remain in gnu_user64.h and !TARGET_64BIT
>> in gnu-user.h? I don't see the reason to put these conditinal defines
>> in shared file. However, this may be due to biarch handling that I'm
>> not familiar with in all details.
>>
>> +#include "config/i386/gnu-user-common.h"
>>
>> You shouldn't include new file here, list it in config.gcc before
>> i386/gnu-user.h or i386/gnu-user64.c
>>
>> Uros.
>
> Hello,
>
> Thanks for review! Here is a new version with all issues fixed.
>
> Bootstrapped on linux-x86_64 and check is in progress. Will it be OK
> for trunk and 4.7 after successfull check?

For sure such changes are not appropriate for a release branch.

Richard.

> Thanks,
> Ilya
> ---
> 2012-04-13  Enkovich Ilya  <ilya.enkovich@intel.com>
>
>        * config/i386/gnu-user-common.h: New.
>
>        * config/i386/gnu-user.h (CPP_SPEC): Moved to
>        gnu-user-common.h.
>        (CC1_SPEC): Likewise.
>        (ENDFILE_SPEC): Likewise.
>        (DEFAULT_PCC_STRUCT_RETURN): Likewise.
>        (TARGET_TLS_DIRECT_SEG_REFS_DEFAULT): Likewise.
>        (TARGET_OS_CPP_BUILTINS): Likewise.
>        (LIBGCC2_HAS_TF_MODE): Likewise.
>        (LIBGCC2_TF_CEXT): Likewise.
>        (TF_SIZE): Likewise.
>        (TARGET_ASM_FILE_END): Likewise.
>        (STACK_CHECK_MOVING_SP): Likewise.
>        (STACK_CHECK_STATIC_BUILTIN): Likewise.
>
>        * config/i386/gnu-user64.h: Likewise.