On Wed, May 17, 2000 at 02:49:09PM -0400, Theodore Y. Ts'o wrote:
[SNIP]
> Is the linux-serial archived anywhere? It seems just last week I
> answered the question about using TIOCMIWAIT to allow you a process to
> sleep until a specified set of control lines change. Unfortunately, I
> can't fin
On Wed, 17 May 2000, Theodore Y. Ts'o wrote:
> Is the linux-serial archived anywhere? It seems just last week I
> answered the question about using TIOCMIWAIT to allow you a process
> to sleep until a specified set of control lines change.
> Unfortunately, I can't find my posting on the subject,
Date:Wed, 17 May 2000 12:13:01 +0200
From: Craig Schlenter <[EMAIL PROTECTED]>
Entirely correct AFAIK. I don't have seriously hard deadlines to
meet timing-wise in my software so I don't need RTLinux. My objection
to polling is that it's not an elegant way to solve the pro
On Wed, May 17, 2000 at 07:28:32AM -0700, Tom Glass wrote:
> What kind of time resolution are you needing? I have an application
> that monitors one serial port for incoming messages in an infinite
> loop. For some of the messages I initiate a child process that
> communicates with a robot, volt
What kind of time resolution are you needing? I have an application
that monitors one serial port for incoming messages in an infinite
loop. For some of the messages I initiate a child process that
communicates with a robot, voltmeter, and other hardware on two other
serial ports. The parent pr
On Wed, May 17, 2000 at 01:15:07PM -0700, Michael Harig wrote:
[snip]
> real-time or not, RTLinux may give you the capability to avoid this. my
> objection is, that linux kernel really should be able to schedule & notify
> a process in about 1 ms if a IRQ occured and the process demanded the
> not
-- Craig Schlenter schrieb über modem control line checking - is polling the only way?
> I don't have seriously hard deadlines to
> meet timing-wise in my software so I don't need RTLinux. My objection
> to polling is that it's not an elegant way to solve the problem of
> detecting a change in the
On Wed, May 17, 2000 at 04:16:24AM -0400, [EMAIL PROTECTED] wrote:
> Well, at a guess, you can wait for me to do a courtesy-call capable
> kernel, which may be a while, and a good while more for it to be
> accepted, or you can do it yourself, or you can solve the problem in a
> totally unexpected
On Wed, May 17, 2000 at 10:52:04AM -0700, Michael Harig wrote:
> Craig,
>
> remember that Linux is a multitasking kernel with no (generic) real time
> capabilities. Your process frequently gets interrupted for an unspecified
> amount of time. If a (serial) IRQ occurs then, the signal has to be me
Craig,
remember that Linux is a multitasking kernel with no (generic) real time
capabilities. Your process frequently gets interrupted for an unspecified
amount of time. If a (serial) IRQ occurs then, the signal has to be memorized
and your process gets aware of it when scheduled the next time. N
On Tue, May 16, 2000 at 04:06:50PM -0400, [EMAIL PROTECTED] wrote:
> It is a pity you couldn't be bothered to read linux-serial. It is not a
> terribly busy list, but this very question has been asked and answered
> in the last week. Maybe you can find some list archives somewhere.
It's not a q
11 matches
Mail list logo
