Posted on 06/29/2011 6:16:30 AM PDT by ShadowAce
Thanks man, good article.
Two responses:
1. You can argue the theory now until forever but these questions are always complex enough that it all comes down to what works. When linux stared a micro-kernel was the “right” theoretical answer but the monolith worked. And it won.
2. So what? If Linux doesn’t like userland FS - so what - since it’s in user space, doesn’t that mean it can succeed with or without love from Linus? Or, alternatively, does it need certain kernel hooks to survive and indeed thrive?
Ouch, “If Linux doesn’t like userland FS” should have been “If Linus doesn’t like userland FS” - my fingers are too used to typing the “x” there LOL!
I am not a big fan of the whole concept of “the cloud”.
I simply do not see the point. I want the computing power and storage on my machine, independent of the web.
Storage is cheap and ram is relatively. Big file transfer takes time.
I am an old Linux user, since some time in 1997. Have been a Ham Op since 1976 and hold commercial radio license. Am very comfortable with all electronics, but not a programmer.
I question the motives for “the cloud” and do not think it is in the interest of the user.
OTOH, if the particular FS (XFS) is only user-space, then I don't see a problem.
Those who depend significantly on “the cloud” will find themselves sitting around wondering “WTF?” when hackers begin targeting “the cloud.”
In my opinion, “the cloud” is an idea to extend the control of the platform to a few big players in computing.
It is about compensating for their inability to significantly improve software to justify the high price. It is especially noticeable due to the Open Source apps first for Linux and now ported for Windows. For the average user there is no longer a need to purchase an upgraded office suite each time they change computers.
Some of the Open Source apps have functionality that the best close source does not. Example: regex find/replace in a spreadsheet. Gnumeric does that and it is ported for Windows and is free. (there are some things that Excel will do that Gnumeric will not)
BINGO!
don’t pay your monthly MS/Apple bill then no data for you.
ANYONE with confidential client data has to be insane to use the cloud system. It is just virtual servers with a HUGE liability stick.
Cloud is all about mobility and multiple devices - take those two factors away and it’s no longer a compelling argument. Add those factors back in and I figure it’s here to stay.
In my opinion the security issue outweighs the portability issue. At least for me.
The web compromises us all, but I try to minimize it.
If your data does not reside in your physical control (on your machine), it’s not your data.
Is “cloud” necessarily the same thing as a user space file system?
No and they’ve actually got nothing (that I can see) to do with each other - it’s just that the “cloud” is a sexier topic and more people are aware of the issues. Good question!
No, “the cloud” also includes access to apps as I understand it.
Quote:
Or, alternatively, does it need certain kernel hooks to survive and indeed thrive?
I would think it is like the vfs and vfs_ops.h family of header files and accompanying calls.
You produce alternate functions to be called when the user
mounts/dismounts a file system as well as alternate functions for when any code uses typical file manipulation library calls (open(), stat(), read(), seek() or a dozen other typical calls).
You create kernel modules to load these into the kernel, and if any utility (anything from “ls” to “write()”, etc)performs an action within your file system, then your function gets the callout, and you can have your way with the data.
Well of course this makes sense - everything eventually relates back to the kernel, doesn’t it?
So is the suggestion that Linus can effectively block these hooks from being integrated in, or does he basically say I don’t like it but do it if you must and we’ll see who comes out the winner in the end?
Quote:
“So is the suggestion that Linus can effectively block these hooks from being integrated in, or does he basically say I don’t like it but do it if you must and we’ll see who comes out the winner in the end?”
Well, they aren’t ‘Linux’ hooks alone.
It goes back to good-old raw down-to-earth UNIX days,
and VFS (virtual file systems) grew out of that.
AIX has forms of the calls, Solaris, HPUX, Apple’s OSX (since it is a BSD Unix), and all sorts of versions of Linux, etc.
I *think* Windows has VFS calls too.
I could be mistaken, though.
I only did a bit of kernel module stuff, and it has been awhile.
It sounded like good sexy work when I first started coding, but 20 years later when I actually had to do it I found that dealing with the kernel can be nasty.
Timing issues, callouts between system and user space, etc, can be a nightmare to write - much less debug.
I imagine all kernel hooks could be blocked - but then that wouldn’t be the UNIX way.
I think what Linus is saying is a tool like fuser (though I’ve never used it) is slow and not applicable to a direct hook into the kernel to be able to perform a lot of file manipulation.
That’s my guess.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.