RE: [PATCH c++ 1/5] lm32: Add (const gdb_byte *) cast

["Jon Beniston" <jon@beniston.com>]

Hi Pedro,

Yes, I believe people are still using this port. (Probably no one on a host
with different byte ordering though!)

Regards,
Jon

-----Original Message-----
From: Pedro Alves [mailto:palves@redhat.com] 
Sent: 12 October 2015 12:11
To: Simon Marchi; gdb-patches@sourceware.org
Cc: Jon Beniston
Subject: Re: [PATCH c++ 1/5] lm32: Add (const gdb_byte *) cast

On 10/11/2015 05:37 AM, Simon Marchi wrote:
> There might be a cleaner sequence of function calls to do what is done 
> here (I don't know), but since I don't want to risk breaking anything, 
> it's safer to just add the required cast.
> 
> gdb/ChangeLog:
> 
> 	* lm32-tdep.c (lm32_push_dummy_call): Add cast.
> ---
>  gdb/lm32-tdep.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/gdb/lm32-tdep.c b/gdb/lm32-tdep.c index 25a7e1e..a0defad 
> 100644
> --- a/gdb/lm32-tdep.c
> +++ b/gdb/lm32-tdep.c
> @@ -289,7 +289,7 @@ lm32_push_dummy_call (struct gdbarch *gdbarch, struct
value *function,
>  	regcache_cooked_write_unsigned (regcache, first_arg_reg + i, val);
>        else
>  	{
> -	  write_memory (sp, (void *) &val, TYPE_LENGTH (arg_type));
> +	  write_memory (sp, (const gdb_byte *) &val, TYPE_LENGTH 
> +(arg_type));

This reveals that the code has a host-dependency.  It is assuming the byte
order of the target is the same as host's.

Please replace this with a call to write_memory_unsigned_integer.

Looks like this port hasn't been touched ever since it was originally
contributed.  Jon, is there still interest in this port?

Thanks,
Pedro Alves