Path: eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: anton@mips.complang.tuwien.ac.at (Anton Ertl)
Newsgroups: comp.arch
Subject: Interrupts in OoO (was: Microarchitectural support for counting)
Date: Sat, 05 Oct 2024 17:57:12 GMT
Organization: Institut fuer Computersprachen, Technische Universitaet Wien
Lines: 23
Message-ID: <2024Oct5.195712@mips.complang.tuwien.ac.at>
References: <2024Oct3.160055@mips.complang.tuwien.ac.at> <vdmrk6$3rksr$1@dont-email.me> <LyELO.69485$2nv5.62232@fx39.iad> <TdWLO.282116$FzW1.158190@fx14.iad> <963a276fd8d43e4212477cefae7f6e46@www.novabbs.org> <8IcMO.249144$v8v2.147178@fx18.iad>
Injection-Date: Sat, 05 Oct 2024 20:05:19 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="4b486aea911af6f0cec786c98c1ac92e";
	logging-data="885056"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+ScjO0trmN1YxHHuk/0R1S"
Cancel-Lock: sha1:MY5L64OAhdQhdn/bEac+k3bxqa8=
X-newsreader: xrn 10.11

EricP <ThatWouldBeTelling@thevillage.com> writes:
>That's difficult with a circular buffer for the instruction queue/rob
>as you can't edit the order.

What's wrong with performing an asynchronous interrupt at the ROB
level rather than inserting it at the decoder?  Just stop commiting at
some point, record this at the interrupt return address and start
decoding the interrupt code.

Ok, it's more efficient to insert an interrupt call into the
instruction stream in the decoder: all the in-flight instructions will
be completed instead of going to waste.  However, the interrupt will
usually be serviced later, and, as you point out, if the instruction
stream is redirected for some other reason, you may have to replay the
interrupt.

As for counting, it seems to me that Brett has understood nothing of
what he cited from my posting.

- anton
-- 
'Anyone trying for "industrial quality" ISA should avoid undefined behavior.'
  Mitch Alsup, <c17fcd89-f024-40e7-a594-88a85ac10d20o@googlegroups.com>