Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.arch > #5946

Re: BI option

Show all headers | View raw

Robert Wessel wrote:
> On Tue, 14 Feb 2012 21:20:55 +0000 (UTC), jgk@panix.com (Joe keane)
> wrote:
>
>> How about some 'branch-on-interrupt' instruction?
>
>
> Like the S/370 Test Pending Interruption instruction?  It lets a CPU
> check to see if an I/O interrupt is pending.  Usually used near the
> end of the I/O interupt handler to see if another is pending (and thus
> can be processed without returning control to the interruptee and
> taking another interrupt).

Isn't the x86 INTO opcode somewhere in the same allpark?

I.e. "we've done some integer code that has a small but not negligible 
chance of having overflowed, and in that case we'd like to know, so 
branch to the Overflow interrupt handler if this is the case".

Terje

-- 
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"

Back to comp.arch | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

BI option jgk@panix.com (Joe keane) - 2012-02-14 21:20 +0000
  Re: BI option Robert Wessel <robertwessel2@yahoo.com> - 2012-02-14 15:26 -0600
    Re: BI option Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-02-15 08:50 +0100
      Re: BI option Robert Wessel <robertwessel2@yahoo.com> - 2012-02-15 02:58 -0600
  Re: BI option Nomen Nescio <nobody@dizum.com> - 2012-02-15 07:47 +0100
  Re: BI option jacko <jackokring@gmail.com> - 2012-02-15 01:43 -0800
  Re: BI option MitchAlsup <MitchAlsup@aol.com> - 2012-02-15 10:08 -0800

csiph-web