Date: 02 Dec 1999 12:04:31 -0500 From: Randell Jesup <rjesup@wgate.com> To: Matthew Dillon <dillon@apollo.backplane.com>, freebsd-hackers@freebsd.org Subject: Re: writing to an mmap()'ed region requires read access? Message-ID: <ybuogc9i89s.fsf@jesup.eng.tvol.net.jesup.eng.tvol.net> In-Reply-To: Matthew Dillon's message of "Tue, 30 Nov 1999 23:28:44 -0800 (PST)"
next in thread | raw e-mail | index | archive | help
Matthew Dillon <dillon@apollo.backplane.com> writes:
>:I was just playing with mmap() really for the first time, and I noticed
>:something a bit odd. When you mmap() a file for writing, it also appears
>:to require that you give it read permissions, else it dies on a signal 10.
>
> I've got a few hundred terrabytes of WOM to sell ya!
I'll sell you a few petabytes!
> This is just a byproduct of the MMU implementation. Most MMU's do
> not support write-only maps because most do not implement separate
> read and write bits. You get a valid and a write-enable bit instead,
> or you just get a valid bit and have to use the dirty bit to make the
> page writable (no write-enable bit at all).
Makes sense.
Shouldn't the mmap() return MAP_FAILED, probably with EACCES,
instead of causing a signal 10?
--
Randell Jesup, Worldgate Communications, ex-Scala, ex-Amiga OS team ('88-94)
rjesup@wgate.com
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?ybuogc9i89s.fsf>