Re: Are We Back to the "Wars" Now ?

John Ames <commodorejohn@gmail.com> wrote:

On Wed, 18 Dec 2024 14:02:52 -0000 (UTC)
Rich <rich@example.invalid> wrote:
 

The amount of data you can transfer over a pipe is not in any way
limited by system memory size or any other system imposed limits. 

 
Quite. I’m not sure why this discussion has restarted but it was
clear from last time round that some of the participants don’t know
what a pipe is, and aren’t particularly interested in finding out. 

 
Yes, our local nymshift troll seems to clearly not know what a pipe
is, nor care to learn either.

 
I *think* what he's meaning to say is this: while you can transfer any
arbitrary amount of data *through* a pipe, there is an upper limit to
how much you can have *in* a pipe at any one time; eventually, you hit
either *A.* an OS-imposed limit on buffer size, at which point things
start blocking as already discussed, or *B.* the upper bounds of system
memory, at which point the system will either start swapping (in which
case you lose any speed advantage) or blocking (as with limited buffer
size.)

Even if that is what the nymshift troll means, the second half is still
wrong when discussing pipes.  There is a pipe buffer size, when you hit
that then the OS begins blocking the pipe writer to stop the flow until
the pipe reader removes some of the data.

But you never exhaust system memory via data in a pipe, the pipe API
simply does not work that way.

Where the nymshift troll's idea of pipes seems to be stuck is MSDOS 2.0
and command.com's fake pipes that were really temporary disk files. 
This seems to be his concept of a pipe and he's never learned anything
nor updated his belief since MSDOS 2.0.

That said, what probably shouldn't need saying here is that if you're
filling up all available space in a pipe such that you're regularly
hitting these limits, you're probably doing pipes wrong.

You can't fill system memory via data buffered inside a pipe.  The pipe
will only hold whatever the system buffer space is (recent Linux's seem
to be 64k buffer size within the pipe).  You'd need to create a
significant number of pipes (and never close them) to exhaust system
memory, in which case yes, you are "doing pipes wrong".