Re: New function get-internal-run-time?
* Steve Youngs <steve(a)sxemacs.org> writes:
* Lars Brinkhoff <lars(a)nocrew.org> writes:
>> ** The new primitive `get-internal-run-time' returns the
processor
>> run time used by Emacs since start-up.
> I'll take that as a no, then.
Not so fast! Personally, I think this would be a cool addition. So
much so, that I'm planning on adding it to SXEmacs.
Please, Lars, cons up a patch against XEmacs 21.5 CVS HEAD and send
it
in to xemacs-patches(a)xemacs.org.
Lars, (S)XEmacs has a function, `current-process-time'[1], which
looks like it does a similar thing to your function...
,----[ C-h f current-process-time RET ]
| `current-process-time' is a built-in function
| (current-process-time)
|
| Documentation:
| Return the amount of time used by this SXEmacs process so far.
| The return value is a list of three floating-point numbers, expressing
| the user, system, and real times used by the process. The user time
| measures the time actually spent by the CPU executing the code in this
| process. The system time measures time spent by the CPU executing kernel
| code on behalf of this process (e.g. I/O requests made by the process).
|
| Note that the user and system times measure processor time, as opposed
| to real time, and only accrue when the processor is actually doing
| something: Time spent in an idle wait (waiting for user events to come
| in or for I/O on a disk drive or other device to complete) does not
| count. Thus, the user and system times will often be considerably
| less than the real time.
|
| Some systems do not allow the user and system times to be distinguished.
| In this case, the user time will be the total processor time used by
| the process, and the system time will be 0.
|
| Some systems do not allow the real and processor times to be distinguished.
| In this case, the user and real times will be the same and the system
| time will be 0.
`----
(current-process-time)
=> (15.09 0.38 4742.375269)
(get-internal-run-time)
=> (0 15 484645)
So I wonder if (S)XEmacs needs `get-internal-run-time' at all? Lars,
any comments?
Footnotes:
[1] I don't think GNU/Emacs has it... a quick grep through the source
didn't find it anyway.
--
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
| I am Dyslexic of Borg. |
| Fusistance is retile. Your arse will be laminated. |
|------------------------------------<steve(a)sxemacs.org>---|
Attachments:
- attachment.sig (application/pgp-signature — 256 bytes)