Date: Tue, 10 Feb 1998 23:58:21 -0800 From: David Greenman <dg@root.com> To: Terry Lambert <tlambert@primenet.com> Cc: jb@cimlogic.com.au, jkh@time.cdrom.com, jbryant@unix.tfs.net, freebsd-current@FreeBSD.ORG Subject: Re: merging win95 and nt filesystem changes into msdosfs Message-ID: <199802110758.XAA15322@implode.root.com> In-Reply-To: Your message of "Wed, 11 Feb 1998 03:15:15 GMT." <199802110315.UAA00548@usr04.primenet.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>> In many cases, Terry's solutions to the problems that he is trying to >> solve aren't the ones we want to adopt. >> There are far more instances where Kirk McKusick and other people >> in-the-know have objected to the direction that Terry wants to take >> us than there are in favor. > >I've talked to Kirk about many of the changes since I've been in the >Bay Area, and you are misquoting him here. I have neither quoted nor paraphrased Kirk or anyone else in my last message. > He has agreed that >a number of the layering issues I have been harping on since day one >are, in fact, legitimate problems. Yes, and I agree as well about some issues. For instance, there is agreement that VOP_UNLOCK should be handled entirely by the VOP_LOCK'er, and not haphazardly by the lower levels depending on the phase of the moon and other factors. Those of us who have looked at making this change realize that this is riddled with potential problems, however, especially when code such as ufs_rename has to be rewritten. I don't even trust Kirk (the author) to do this, so why would I trust you? On the other hand, there is general disagreement on changing the symantics of advisory locking. >> Terry's unwillingness to carefully explain/justify the changes that >> he proposes in a manor that reasonable kernel developers can >> understand has further resulted in a serious lack of trust. For ... >I'll be happy to explain anything you have questions about. This is inconsistent with what you have said in the past which was escentially that you had neither the time nor the willingness to explain the changes you were proposing and that we should just take it on faith that what you want to do is good. >occur. The only rationale against these has been "we fear change, >specifically in the form of divergence". Now that that's blown >out, so long as it's possible to revert such changes if they are seen >to have a detrimental effect, I don't see why there's a problem. Backing out changes after some amount of time has gone by (and thus the code has been modified by other people for other reasons) is not a trivial undertaking. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199802110758.XAA15322>