On Thu, Aug 24, 2000 at 01:32:10PM -0400, Joshua N Pritikin <[EMAIL PROTECTED]>
wrote:
> > mechanism we could use" that I got on the mailinglist, which becomes
> > increasingly untrue (sigio & friends (like sigtimedwait) are optional
> > posix extensions that get really popular), and the more pro
On Thu, Aug 24, 2000 at 07:19:38PM +0200, [EMAIL PROTECTED] wrote:
> However, my concern was the impression of "there is no portable event
> mechanism we could use" that I got on the mailinglist, which becomes
> increasingly untrue (sigio & friends (like sigtimedwait) are optional
> posix extensio
On Thu, Aug 24, 2000 at 01:09:57PM -0400, Uri Guttman <[EMAIL PROTECTED]> wrote:
> please quote the docs that show this. according to my solaris 7 docs,
> you have to call aio_wait to get an aio_result back. this can be one of
I am talking about SIGIO and friends, not the posix aio_*-functions. a
On Thu, Aug 24, 2000 at 12:29:24PM -0400, Joshua N Pritikin <[EMAIL PROTECTED]>
wrote:
> > i/o, for example you could just transfer the data inside the signal
> > handler ;)
>
> THAT's portable. :-)
Well, at least so much as using SIGIO. It only works this way on newer
systems, and I *do* care
> "ML" == Marc Lehmann <[EMAIL PROTECTED]> writes:
ML> Solaris can tell you which event caused the signal, so there is no
ML> need to scan your list of operations. After all, that's the point
ML> of sigio.
please quote the docs that show this. according to my solaris 7 docs,
you have t
On Thu, Aug 24, 2000 at 06:13:03PM +0200, [EMAIL PROTECTED] wrote:
> I forgot to mention that there are a lot of other ways to implement asynch
> i/o, for example you could just transfer the data inside the signal
> handler ;)
THAT's portable. :-)
> However, as for Event I think one could imple
On Thu, Aug 24, 2000 at 11:45:35AM -0400, Joshua N Pritikin <[EMAIL PROTECTED]>
wrote:
> > Just as linux: you clone/block/signal (i am not talking about pthreads). Its
I forgot to mention that there are a lot of other ways to implement asynch
i/o, for example you could just transfer the data ins
On Thu, Aug 24, 2000 at 05:34:58PM +0200, [EMAIL PROTECTED] wrote:
> On Wed, Aug 23, 2000 at 02:50:51PM -0400, Uri Guttman <[EMAIL PROTECTED]> wrote:
> > solaris has true aynch file i/o.
>
> Just as linux: you clone/block/signal (i am not talking about pthreads). Its
> just implemented differentl
ith Event, sinc eEvent does not do any I/O,
rather it does event processing. And sigio+friends is exactly that (and
very similar to kqueue).
> but it uses sigio to tell you when it something happens. then you have
> to scan you pending operations to see which one it was. the idjits
> should
y nice if you want to do aio. Why do you
consider it a hack? It seems to me exactly equivalent to kqueue: you tell
the kernel "please deliver me signal number xyz" and the kernel delivers
you the signal + the event information you need. Vastly better than select
in most cases Event is being
On Wed, Aug 23, 2000 at 11:12:01AM -0400, Uri Guttman <[EMAIL PROTECTED]> wrote:
> > "ML" == Marc Lehmann <[EMAIL PROTECTED]> writes:
>
> ML> Well, most unices support asynchroneous I/O, which gives most of the same
> ML> and scales very well.
>
> please show me where in linux is the sup
On Wed, Aug 23, 2000 at 02:50:51PM -0400, [EMAIL PROTECTED] wrote:
> ... i am proposing to perl6 that async file
> i/o be supported in a more integrated way.
Sounds good to me. No objections here. :-)
--
May the best description of competition prevail.
(via, but not speaking for Deutsch
> "JNP" == Joshua N Pritikin <[EMAIL PROTECTED]> writes:
JNP> Agreed. SIGIO != asynchronous I/O; SIGIO is a hack.
solaris has true aynch file i/o. but it uses sigio to tell you when it
something happens. then you have to scan you pending operations to see
which one it was. the idjits shou
On Wed, Aug 23, 2000 at 11:12:01AM -0400, [EMAIL PROTECTED] wrote:
> > "ML" == Marc Lehmann <[EMAIL PROTECTED]> writes:
>
> ML> Well, most unices support asynchroneous I/O, which gives most of
> ML> the same and scales very well.
>
> please show me where in linux is the support for async
> "ML" == Marc Lehmann <[EMAIL PROTECTED]> writes:
ML> Well, most unices support asynchroneous I/O, which gives most of the same
ML> and scales very well.
please show me where in linux is the support for asynch file I/O. i have
not found it.
uri
--
Uri Guttman - [EMAIL PROTE
Sorry for the late reply, just read through the archives ;)
On Thu, Jul 27, 2000 at 04:45:17PM -0400, Uri Guttman <[EMAIL PROTECTED]> wrote:
> JNP> ftp://ftp.freebsd.org/pub/FreeBSD/releases/i386/4.1-RELEASE/RELNOTES.TXT
>
> and whether this propogates to other unixes is a big issue. i don't
>>>>> "JNP" == Joshua N Pritikin <[EMAIL PROTECTED]> writes:
JNP> ftp://ftp.freebsd.org/pub/FreeBSD/releases/i386/4.1-RELEASE/RELNOTES.TXT
JNP> "A new event notification facility called kqueue was added to the
JNP> FreeBSD kernel. This is a
ftp://ftp.freebsd.org/pub/FreeBSD/releases/i386/4.1-RELEASE/RELNOTES.TXT
"A new event notification facility called kqueue was added to the
FreeBSD kernel. This is a new interface which is able to replace
poll/select, offering improved performance, as well as the ability
to report many diff
18 matches
Mail list logo
