bug-binutils
[Top][All Lists]
Advanced

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

[Bug binutils/24942] objcopy: Add option for setting section alignment


From: jh7370 at my dot bristol.ac.uk
Subject: [Bug binutils/24942] objcopy: Add option for setting section alignment
Date: Tue, 24 Sep 2019 09:00:39 +0000

https://sourceware.org/bugzilla/show_bug.cgi?id=24942

James Henderson <jh7370 at my dot bristol.ac.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jh7370 at my dot bristol.ac.uk

--- Comment #10 from James Henderson <jh7370 at my dot bristol.ac.uk> ---
I'm with Fangrui on this one. Although I don't know of any instances of
non-powers-of-2 alignment making sense, I still talk about an alignment of,
say, 4, as "4" not "2". Similarly, tools print alignment as the actual value,
not as a power of 2, in my experience. As such, I expect switches to conform to
that. Indeed, as pointed out, that's exactly what .align does on many
platforms, and also every tool I've ever used that has anything to do with
alignment (in fact, I didn't even know .align had a different behaviour on some
platforms).

Additionally, I think it's much more likely that a user who hasn't looked in
depth at the documentation will try to specify the raw value, and then will end
up with an incorrect modification, than the other way around. Furthermore, it
is simpler for a user to not have to do the conversion of, say, 128 to 2 to the
whatever.

If bfd can't support non-powers-of-2 alignment values, it should simply reject
them at command-line parsing time, in my opinion.

-- 
You are receiving this mail because:
You are on the CC list for the bug.


reply via email to

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