avr-libc-dev
[Top][All Lists]
Advanced

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

RE: [avr-libc-dev] Re: [RFC] Unified ELF file


From: Michael Hennebry
Subject: RE: [avr-libc-dev] Re: [RFC] Unified ELF file
Date: Thu, 27 Sep 2007 17:09:53 -0500 (CDT)

 On Thu, 27 Sep 2007, Eric Weddington wrote:

> > -----Original Message-----
> > From: Michael Hennebry [mailto:address@hidden

> > How would a value specified by the user produce surprise?
> > In case I wasn't clear, I was suggesting that in the absence
> > of a user-specified value that failure, preferably with a
> > useful error message, is better than a default.
>
> If a user specifes lfuse and efuse, but not hfuse, how are they going to be
> ordered by the linker script in memory? What gets put in the hole? The ELF

It can fail.
The next time the user can specify hfuse.

If one wants to allow for lfuse and efuse without hfuse,
one can use separate sections.

> file is just a group of data. It requires a program to decide if there is an
> error and to report one. The ELF file itself can't decide if there was an
> error. And I'm not about to patch the compiler toolchain with custom code
> that can't be committed upstream.

I wasn't suggesting that you do so.
I don't recall suggesting anything that couldn't
be done with header files and linker scripts.

A linker script can be made to fail based on an
undefined reference or the size of a section.

-- 
Mike   address@hidden
"Horse guts never lie."  -- Cherek Bear-Shoulders






reply via email to

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