[Top] [All Lists]

Re: Icache coherency problems for R3400, DS5000/240

To: Dominic Sweetman <>
Subject: Re: Icache coherency problems for R3400, DS5000/240
From: "Maciej W. Rozycki" <>
Date: Tue, 20 Jun 2000 14:50:05 +0200 (MET DST)
Cc: Ralf Baechle <>, Ralf Baechle <>, Harald Koerfgen <>,,
In-reply-to: <>
Organization: Technical University of Gdansk
On Tue, 20 Jun 2000, Dominic Sweetman wrote:

> Yes, the original R3000 chip could be wired to produce the appearance
> of multi-word lines in its I-cache, and some derivative CPUs were built
> that way.  Four was popular - I don't think anyone did 8.

 DEC docs claim the cache subsystem of KN03 is configured with single-word
lines for the icache, but upon an icache fill the MB ASIC fills four

> And while I'm here, I'll continue my lonely campaign.  I suggest you
> don't say "flush" because nobody knows whether it means invalidate,
> write-back, or both[1].  Instead, say "invalidate", "writeback", or
> "both".  Even if it means changing St Linus' function names...

 OK, by "flush" I always meant "invalidate" (see also ia32's invd and
wbinvd instructions -- the first one causes an external flush cycle and
the other one an external writeback cycle and then a flush cycle -- that's
Intel's original naming).  On the other hand, "invalidate" is surely not
more ambiguous than "flush", so feel free to continue your campaign... 

+  Maciej W. Rozycki, Technical University of Gdansk, Poland   +
+        e-mail:, PGP key available        +

<Prev in Thread] Current Thread [Next in Thread>