[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [incomplete patch] Get objdump to report exception tables for ARM/SH

From: Danny Backx
Subject: Re: [incomplete patch] Get objdump to report exception tables for ARM/SH4 correctly
Date: Sat, 05 Apr 2008 10:43:31 +0200

On Fri, 2008-04-04 at 08:10 +0100, Nick Clifton wrote:
> > Didn't know about that. I've used it, and added one or more lines to 30
> > other .c files. In most cases this is just
> > 
> > #define bfd_pe_print_pdata NULL
> Shouldn't that be:
>    #define bfd_pe_print_pdata pe_print_data
> ie use the default print_pdata routine if compressed pdata is not 
> supported by the target.

One of the earlier suggestions was for the default print_pdata to be
used when the new field is NULL. This is how I implemented it.

> > One more question though. I moved my pe_print_ce_compressed_pdata
> > function into pe-arm-wince.c , it was in peXXigen.c in my initial patch.
> > The peXXigen.c doesn't seem right for this because this function is not
> > to be treated with the XX replacement stuff. However, pe-arm-wince.c
> > doesn't seem right either because the pe_print_ce_compressed_pdata
> > function is to be used both for ARM and for SH.
> > 
> > So where should I put it (and its helper functions) ?
> I would actually suggest leaving it in peXXigen.c.  The current 
> pe_print_pdata function lives there and it does not make use of the XX 
> replacement stuff.  Make sure that it is static though, since peXXigen.c 
> can be included multiple times in a single build.

I don't think I understand. If it's a static function in peXXigen.c then
I can't use it in e.g. pe-arm-wince.c to initialise the structure.

Also right now there are two functions. The original one, and one to
handle the ARM and SH4 compressed pdata cases. Putting them together in
one file is fine by me, but putting them in a source file that can be
included more than once seems wrong. This would, I think, apply to both
the function I'm adding and to the default one. Should they not be in
another source file (one that doesn't get the XX treatment) so they get
included just once ?

So I still think I need a source file that doesn't get the XX treatment
to put both functions in as non-static functions. But I am probably
overlooking something, I'm new to the binutils source.


Attachment: signature.asc
Description: This is a digitally signed message part

reply via email to

[Prev in Thread] Current Thread [Next in Thread]