[PATCH] gas: Disable --gcodeview on PE targets with no O_secrel

Alan Modra amodra@gmail.com
Wed Nov 23 22:20:26 GMT 2022


On Wed, Nov 23, 2022 at 07:43:30PM +0000, Mark Harmstone wrote:
> As per https://sourceware.org/pipermail/binutils/2022-November/124611.html
> 
> ---
>  gas/as.c       | 6 +++---
>  gas/codeview.c | 4 ++--
>  2 files changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/gas/as.c b/gas/as.c
> index 74ffa71925f..e49b1f0a796 100644
> --- a/gas/as.c
> +++ b/gas/as.c
> @@ -334,7 +334,7 @@ Options:\n\
>    --gdwarf-cie-version=<N> generate version 1, 3 or 4 DWARF CIEs\n"));
>    fprintf (stream, _("\
>    --gdwarf-sections       generate per-function section names for DWARF line information\n"));
> -#ifdef TE_PE
> +#if defined (TE_PE) && defined (O_secrel)
>    fprintf (stream, _("\
>    --gcodeview             generate CodeView debugging information\n"));
>  #endif
> @@ -551,7 +551,7 @@ parse_args (int * pargc, char *** pargv)
>      ,{"gdwarf2", no_argument, NULL, OPTION_GDWARF_2}
>      ,{"gdwarf-sections", no_argument, NULL, OPTION_GDWARF_SECTIONS}
>      ,{"gdwarf-cie-version", required_argument, NULL, OPTION_GDWARF_CIE_VERSION}
> -#ifdef TE_PE
> +#if defined (TE_PE) && defined (O_secrel)
>      ,{"gcodeview", no_argument, NULL, OPTION_GCODEVIEW}
>  #endif
>      ,{"gen-debug", no_argument, NULL, 'g'}
> @@ -879,7 +879,7 @@ This program has absolutely no warranty.\n"));
>  	  flag_dwarf_sections = true;
>  	  break;
>  
> -#ifdef TE_PE
> +#if defined (TE_PE) && defined (O_secrel)
>  	case OPTION_GCODEVIEW:
>  	  debug_type = DEBUG_CODEVIEW;
>  	  break;
> diff --git a/gas/codeview.c b/gas/codeview.c
> index da7145659f5..13f1a3a3223 100644
> --- a/gas/codeview.c
> +++ b/gas/codeview.c
> @@ -24,7 +24,7 @@
>  #include "filenames.h"
>  #include "md5.h"
>  
> -#ifdef TE_PE
> +#if defined (TE_PE) && defined (O_secrel)
>  
>  #define NUM_MD5_BYTES       	16
>  
> @@ -538,4 +538,4 @@ codeview_generate_asm_lineno (void)
>  {
>  }
>  
> -#endif /* TE_PE */
> +#endif /* TE_PE && O_secrel */
> -- 
> 2.37.4

This is OK, thanks.

-- 
Alan Modra
Australia Development Lab, IBM


More information about the Binutils mailing list