Re: Memory corruption for host double format different from target double format

[Thomas Schwinge <thomas@codesourcery.com>]

[-- Attachment #1: Type: text/plain, Size: 5424 bytes --]

Hi!

On Wed, 29 Aug 2012 17:36:54 +0200, I wrote:
> On Fri, 10 Aug 2012 16:31:47 +0200, Mark Kettenis <mark.kettenis@xs4all.nl> wrote:
> > > Date: Fri, 10 Aug 2012 14:56:46 +0200 (CEST)
> > > From: "Ulrich Weigand" <uweigand@de.ibm.com>
> > > 
> > > Yao Qi wrote:
> > > > On Friday, August 10, 2012 11:32:53 AM Thomas Schwinge wrote:
> > > > > That is, if set_gdbarch_double_format has not been called, it will
> > > > > default to floatformats_ieee_double -- even though set_gdbarch_double_bit
> > > > > may have been called setting it unequal to the 64-bit double format.
> > > > > Hmm, and gdbarch.c:verify_gdbarch has the following comment on top of it:
> > > > > Ensure that all values in a GDBARCH are reasonable.  ;-)
> > > > 
> > > > Looks like some checking like this is missing?
> > > > 
> > > >   gdbarch->float_format->totalsize <= gdbarch->float_bit
> > > >   gdbarch->double_format->totalsize <= gdbarch->double_bit
> > > 
> > > In fact, I'd prefer to make gdbarch_double_format etc. *mandatory*
> > > and gdbarch_double_bit etc. optional, defaulting to the format size.
> > > (Currently, _bit is mandatory and _format is optional.)
> > > 
> > > This would mean that nearly all calls to set_gdbarch_double_bit
> > > could go away, with the exception of special cases like "long double"
> > > on i386 ...
> > 
> > Initializing _bit based on _format by default makes sense, but I don't
> > think this is easy to implement given the way how the gdbarch.c code
> > is generated.
> > 
> > Making _format mandatory doesn't make sense to me though.  I'd say
> > that ieee_single and ieee_double are perfectly reasonable defaults for
> > float_format and double_format.
> 
> Is there a reasonable way for at least detecting the mismatch that I
> happened to observe for SH?
> 
> 
> Other than that, OK to check in the following?  I have only tested the SH
> bits; no maintainer listed for h8300, Stephane CCed for m68hc11.

Stephane Carrez' email address <stcarrez@nerim.fr> (as listed in
gdb/MAINTAINERS) bounces saying »unknown user«, but I found another one
in the GCC context -- Stephane, is this you?  If yes, please update the
three occurences of your old email address in gdb/MAINTAINERS (and
possibly other files, too).

Kevin, I'm also adding you to the CC list, as you've been helpful with SH
issues before -- should you be listed as a maintainer for SH?

And what about the h8300 bits?

> gdb/
> 	* h8300-tdep.c (h8300_gdbarch_init): Invoke
> 	set_gdbarch_double_format and set_gdbarch_long_double_format.
> 	* m68hc11-tdep.c (m68hc11_gdbarch_init): Invoke
> 	set_gdbarch_double_format.
> 	* sh-tdep.c (sh_gdbarch_init): Likewise.
> 
> diff --git a/gdb/h8300-tdep.c b/gdb/h8300-tdep.c
> index 7fc4daa..bcb769e 100644
> --- a/gdb/h8300-tdep.c
> +++ b/gdb/h8300-tdep.c
> @@ -1351,7 +1351,9 @@ h8300_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
>    set_gdbarch_long_bit (gdbarch, 4 * TARGET_CHAR_BIT);
>    set_gdbarch_long_long_bit (gdbarch, 8 * TARGET_CHAR_BIT);
>    set_gdbarch_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
> +  set_gdbarch_double_format (gdbarch, floatformats_ieee_single);
>    set_gdbarch_long_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
> +  set_gdbarch_long_double_format (gdbarch, floatformats_ieee_single);
>  
>    set_gdbarch_believe_pcc_promotion (gdbarch, 1);
>  
> diff --git a/gdb/m68hc11-tdep.c b/gdb/m68hc11-tdep.c
> index 79629ef..cd32459 100644
> --- a/gdb/m68hc11-tdep.c
> +++ b/gdb/m68hc11-tdep.c
> @@ -1498,7 +1498,16 @@ m68hc11_gdbarch_init (struct gdbarch_info info,
>    set_gdbarch_short_bit (gdbarch, 16);
>    set_gdbarch_int_bit (gdbarch, elf_flags & E_M68HC11_I32 ? 32 : 16);
>    set_gdbarch_float_bit (gdbarch, 32);
> -  set_gdbarch_double_bit (gdbarch, elf_flags & E_M68HC11_F64 ? 64 : 32);
> +  if (elf_flags & E_M68HC11_F64)
> +    {
> +      set_gdbarch_double_bit (gdbarch, 64);
> +      set_gdbarch_double_format (gdbarch, floatformats_ieee_double);
> +    }
> +  else
> +    {
> +      set_gdbarch_double_bit (gdbarch, 32);
> +      set_gdbarch_double_format (gdbarch, floatformats_ieee_single);
> +    }
>    set_gdbarch_long_double_bit (gdbarch, 64);
>    set_gdbarch_long_bit (gdbarch, 32);
>    set_gdbarch_ptr_bit (gdbarch, 16);
> diff --git a/gdb/sh-tdep.c b/gdb/sh-tdep.c
> index 1ede13a..caf940d 100644
> --- a/gdb/sh-tdep.c
> +++ b/gdb/sh-tdep.c
> @@ -2299,6 +2299,7 @@ sh_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
>      case bfd_mach_sh2e:
>        /* doubles on sh2e and sh3e are actually 4 byte.  */
>        set_gdbarch_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
> +      set_gdbarch_double_format (gdbarch, floatformats_ieee_single);
>  
>        set_gdbarch_register_name (gdbarch, sh_sh2e_register_name);
>        set_gdbarch_register_type (gdbarch, sh_sh3e_register_type);
> @@ -2344,6 +2345,7 @@ sh_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
>      case bfd_mach_sh2a_or_sh3e:
>        /* doubles on sh2e and sh3e are actually 4 byte.  */
>        set_gdbarch_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
> +      set_gdbarch_double_format (gdbarch, floatformats_ieee_single);
>  
>        set_gdbarch_register_name (gdbarch, sh_sh3e_register_name);
>        set_gdbarch_register_type (gdbarch, sh_sh3e_register_type);


Grüße,
 Thomas

[-- Attachment #2: Type: application/pgp-signature, Size: 489 bytes --]