Date: Mon, 21 Mar 2005 11:21:04 -0500 (EST) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Krzysztof Drewicz <hunter24@pancake.sggw.waw.pl> Cc: Andrew Gallatin <gallatin@cs.duke.edu> Subject: Re: user-mode-linux in fbsd Message-ID: <16958.62576.271134.811502@grasshopper.cs.duke.edu> In-Reply-To: <20050319224648.GA30079@pancake.sggw.waw.pl> References: <20050316151639.GA6424@pancake.sggw.waw.pl> <20050316182800.GC91771@hub.freebsd.org> <16955.20467.474306.849814@grasshopper.cs.duke.edu> <20050319224648.GA30079@pancake.sggw.waw.pl>
next in thread | previous in thread | raw e-mail | index | archive | help
Krzysztof Drewicz writes: > On Fri, Mar 18, 2005 at 05:02:27PM -0500, Andrew Gallatin wrote: > > Only s/=4096/+=4096/ or cast 4096 to a pointer type in your patch. Yes, it was intended to be mostly a hack to prove that the problem was with the algorithm that selected a bad location for mmap if no hint was given. > I've figured out: > linux: ptrace(24, ...) not implemented > > And any of 'ptrace' in sys/compat/ is not implemented, uml says "we depend > on ptrace" so it's not possilble right now. FreeBSD also has a ptrace, and a quick glance makes it look like there may be a 1:1 mapping for most things. But the PTRACE_PEEKUSER / PTRACE_POKEUSER that uml is trying to do to change sycall numbers sounds rather scary. I'm not sure if there would be a 1:1 mapping for that, and I don't think I understand what they are doing well enough to write it from scratch. Drew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?16958.62576.271134.811502>