This is the mail archive of the
binutils@sourceware.cygnus.com
mailing list for the binutils project.
Re: Remove alignment value check in gasp
- To: nickc at cygnus dot com
- Subject: Re: Remove alignment value check in gasp
- From: Ian Lance Taylor <ian at zembu dot com>
- Date: 29 Feb 2000 20:19:42 -0500
- CC: binutils at sourceware dot cygnus dot com
- References: <200003010116.RAA28450@elmo.cygnus.com>
Date: Tue, 29 Feb 2000 17:16:06 -0800
From: Nick Clifton <nickc@cygnus.com>
Unless anyone objects I plan to check in the following patch
tomorrow. All it does is to remove the code in GASP that checks to
see if the .align directive has been given an argument of 1, 2 or 4,
This test is, I believe, bogus, since the .align directive takes a
power-of-two as its argument and, for example ".align 3" is a valid
alignment directive.
Sometimes the .align directive takes a power of 2, and sometimes it
takes a byte count. gasp basically just passes it through to the
assembler. Perhaps removing the check is right, because gasp can't
know how the assembler will interpret it. Alternatively, perhaps gasp
should present a unified model and pass it through as .p2align.
Actually, I don't care at all about gasp, and I think we should remove
it from a future binutils release. These days gasp has no feature
which is not in gas itself.
Ian