On Tue, Aug 10, 2010 at 05:15:35AM +0000, Neel Natu wrote:
> Author: neel
> Date: Tue Aug 10 05:15:35 2010
> New Revision: 211130
> URL: http://svn.freebsd.org/changeset/base/211130
> 
> Log:
>   Fix compilation error for 64-bit little endian build:
>   libexec/rtld-elf/mips/reloc.c:196: warning: right shift count >= width of 
> type
>   
>   When the expression '(r_info) >> 32' was passed to bswap32() it was promptly
>   changed to '(uint32_t)(r_info) >> 32' which is not what we intended.

Wouldn't it be better to fix the bswap32 macro instead?

> Modified:
>   head/libexec/rtld-elf/mips/reloc.c
> 
> Modified: head/libexec/rtld-elf/mips/reloc.c
> ==============================================================================
> --- head/libexec/rtld-elf/mips/reloc.c        Mon Aug  9 23:32:37 2010        
> (r211129)
> +++ head/libexec/rtld-elf/mips/reloc.c        Tue Aug 10 05:15:35 2010        
> (r211130)
> @@ -75,7 +75,7 @@ void _rtld_relocate_nonplt_self(Elf_Dyn 
>   * ELF64 MIPS encodes the relocs uniquely.  The first 32-bits of info contain
>   * the symbol index.  The top 32-bits contain three relocation types encoded
>   * in big-endian integer with first relocation in LSB.  This means for little
> - * endian we have to byte swap that interger (r_type).
> + * endian we have to byte swap that integer (r_type).
>   */
>  #define      Elf_Sxword                      Elf64_Sxword
>  #define      ELF_R_NXTTYPE_64_P(r_type)      ((((r_type) >> 8) & 0xff) == 
> R_TYPE(64))
> @@ -83,7 +83,7 @@ void _rtld_relocate_nonplt_self(Elf_Dyn 
>  #undef ELF_R_SYM
>  #undef ELF_R_TYPE
>  #define ELF_R_SYM(r_info)            ((r_info) & 0xffffffff)
> -#define ELF_R_TYPE(r_info)           bswap32((r_info) >> 32)
> +#define ELF_R_TYPE(r_info)           bswap32(((r_info) >> 32))
>  #endif
>  #else
>  #define      ELF_R_NXTTYPE_64_P(r_type)      (0)
> 
_______________________________________________
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Reply via email to