What does CCB stand for?!!

NTFSD
1

sorry all,
but what does this CCB stand for in the context of shadow file object implementation where one place the CCB in fscontext2?

couldn’t find anything googling :frowning:

2

On Tue, 14 Mar 2017, xxxxx@yahoo.com wrote:

but what does this CCB stand for in the context of shadow file object implementation where one place the CCB in fscontext2?

CCB stands for context control block and it is typicaly a structure that
contains information that is uniq to every open handle to a file, for
example the current position for read/write.

Bo Branten

3

thanks

4

Just to add…This terminology is borrowed from the Windows file systems
which have historically called this structure the CCB (search FASTFAT for
CCB). I don’t know if the name originated any further back than that.

I would prefer to call it the Open Control Block (OCB), but in the end I
usually give in and call it the CCB. I guess deep down I’m really a
conformist.

-scott
OSR
@OSRDrivers

wrote in message news:xxxxx@ntfsd…

thanks

5

> I don’t know if the name originated any further back than that.

I’m pretty sure that VMS had a Channel Control Block. The handle/file
object equivalent to a negative index into a fixed array of them. I have a
vague memory that, inter alia, it pointed to the structures where the VBN to
LBN mappings happened.

But it’s been too long and I’m not even sure where by micro-fiche reader is,
let alone the V3 micro-fiches

R

6

> I’m pretty sure that VMS had a Channel Control Block

Of course! How foolish, should have assumed it came from VMS.

Quick scan of technical documentation from 1994 tells me that to send I/O on
VMS I call $ASSIGN to allocate a Channel Control Block (CCB) and then call
$QIO/$QIOW to generate an I/O Request Packet (I/O) using that channel. As
someone who never touched VMS it always feels like bizarro Windows, which is
pretty fun.

Thanks!

-scott

“Rod Widdowson” wrote in message news:xxxxx@ntfsd…

I don’t know if the name originated any further back than that.

I’m pretty sure that VMS had a Channel Control Block. The handle/file
object equivalent to a negative index into a fixed array of them. I have a
vague memory that, inter alia, it pointed to the structures where the VBN to
LBN mappings happened.

But it’s been too long and I’m not even sure where by micro-fiche reader is,
let alone the V3 micro-fiches

R

7

> Quick scan of technical documentation from 1994

quite modern then :slight_smile: Grey wall (Magic/5.5) or Orange wall (4.?) ?

tells me that to send I/O on VMS I call $ASSIGN to allocate a Channel
Control Block (CCB) and then call $QIO/$QIOW to generate an I/O Request
Packet (I/O) using that channel.

Oh gosh it all comes back.

I remembered while walking the dogs yesterday what

pointed to the structures where the VBN to
LBN mappings happened.

was. It was the Window Control Block (WCB). If you needed to fully map a
fragmented file it generated a “Cathedral Window”, otherwise you (the file
system) would get called back to “turn the windows” if the LBN/VBN mapping
failed. However you were not called for the usermode read or write so there
was no re-entrancy there.

As someone who never touched VMS it always feels like bizarro Windows,
which is pretty fun.

It has to be said that the windows FS interface is way cleaner that the VMS
one (which looked a lot like the RSX one - and in that one you could only
have a directory hierarchy if you built it by hand).

Why do I feel old?