Date: Wed, 3 Jun 1998 11:13:37 -0500 (CDT) From: Tony Kimball <alk@pobox.com> To: chat@FreeBSD.ORG Subject: Re: kernfs/procfs questions... Message-ID: <199806031613.LAA29012@compound.east.sun.com> References: <199806031145.EAA26532@hub.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
I'm pretty strongly on the side of simplicity and orthogonality of use, and universal applicability of tools myself. That's what the FS interface provides, and sysctl, or IP services for that matter, fail to provide. (Who goofed with sockets?) The FS interface needs to be abstracted, generalized in order to accomodate these broader applications, admittedly. Ioctls are eggregious. I can understand the performance consideration which leads to SHM, but in general communication between processes should be accomplished by read/write, not ioctl and setsockopt and listen and accept and sysctl and traps, and... One could argue for substantial changes in many historical interfaces, but codebase and portability considerations generally prohibit such changes in any general-purpose system. FreeBSD, however, seems to draw the line by historical reference to 4.4 Lite 2, constraining the system more strongly than does the application codebase (since the former includes design decisions which are not honored by the latter). This is unfortunate when it causes divergence from the mainstream application codebase, which is more Linux/SVR4- oriented with each passing day. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199806031613.LAA29012>