Date: Tue, 9 Feb 1999 22:52:56 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: jdp@polstra.com (John Polstra) Cc: tlambert@primenet.com, hackers@FreeBSD.ORG, wes@softweyr.com Subject: Re: ldconfig and libraries Message-ID: <199902092252.PAA11134@usr02.primenet.com> In-Reply-To: <XFMail.990209142317.jdp@polstra.com> from "John Polstra" at Feb 9, 99 02:23:17 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > OK, next question: Why can't it be at the end of the data? > > Because then it wouldn't be in a read-only segment at execution > time. OK, next next question: who cares that it's not in a read-only segment? > And it wouldn't help anyway, because it would still come before BSS, > causing BSS to move if the RPATHs size were changed. :-) Next next next question: What's wrong with putting it at the end of the data, and putting Wes's 1K offset into the BSS start offset, instead of into the image file on disk? E.g., it takes what it takes up to MAX_PATH, but only the space used is accounted against the image size. > Anticipating another likely question: > Q. Why not make it read-write, then? What could it hurt? > A. It would violate the ELF spec. Hmmm. I thought Doug Rabson just did this on a mapping location change on the Alpha? Well... how about supporting more sections in a process image? Doesn't WINE require this, so it's already somewhat supported? Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199902092252.PAA11134>