Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Wed, 11 Mar 2015 18:19:47 -0400
From: Rich Felker <dalias@...c.org>
To: musl@...ts.openwall.com
Subject: Re: musl-gcc wrapper patch

On Wed, Mar 11, 2015 at 10:32:08PM +0100, Szabolcs Nagy wrote:
> i assume the most common expectation is that the installed
> musl-gcc tool uses the same compiler as the musl build was
> using
> 
> at least this is what i would want for cross compilation
> 
> i guess this breaks if relative path or multiple words were
> used (cc with args) in $(CC)
> 
> comments?
> [...]
>  tools/musl-gcc: config.mak
> -	printf '#!/bin/sh\nexec "$${REALGCC:-gcc}" "$$@" -specs "%s/musl-gcc.specs"\n' "$(libdir)" > $@
> +	printf '#!/bin/sh\nexec "$${REALGCC:-%s}" "$$@" -specs "%s/musl-gcc.specs"\n' "$(CC)" "$(libdir)" > $@
>  	chmod +x $@

Since $CC can contain multiple shell words (e.g. a -m32 or similar)
perhaps we should remove the double-quotes around the expansion of
$REALGCC? Unfortunately this makes for a trade-off between supporting
'standard' usage of $CC and supporting pathnames containing
whitespace. Any opinions on the matter? Any ways to avoid the
tradeoff? Overall I'm in favor of this change anyway but I'd like to
avoid having any negative impact.

Rich

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.