From: Segher Boessenkool <segher@kernel.crashing.org>
To: Chris Friesen <cfriesen@nortelnetworks.com>
Cc: Tom Rini <trini@kernel.crashing.org>,
Linux Kernel list <linux-kernel@vger.kernel.org>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>
Subject: Re: problem with cache flush routine for G5?
Date: Sat, 6 Mar 2004 02:34:03 +0100 [thread overview]
Message-ID: <59D75402-6F0E-11D8-9B46-000A95A4DC02@kernel.crashing.org> (raw)
In-Reply-To: <4048B720.4010403@nortelnetworks.com>
> After doing some digging in the 970fx specs, it seems that we may not
> need to explicitly force a store of the L1 dcache at all. According
> to the docs, the L1 dcache is unconditionally store-through. Thus, for
> a brute-force implementation we should be able to just invalidate the
> whole icache, do the appropriate sync/isync, and it should pick up the
> changed instructions from the L2 cache. Do you see any problems with
> this? Do I actually still need the store?
You need a sync instruction before the instruction cache invalidate,
to make sure all stores to L2 have completed.
I don't know which "970fx specs" you mean, but if it's the programming
manual, it should tell you how to invalidate the entire instruction
cache (and how to flush the L2 cache, if you really must. But just
don't).
Segher
next prev parent reply other threads:[~2004-03-08 7:51 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-03-04 21:06 Chris Friesen
2004-03-04 23:51 ` Benjamin Herrenschmidt
2004-03-04 23:57 ` Tom Rini
2004-03-05 0:04 ` Benjamin Herrenschmidt
2004-03-05 0:44 ` Chris Friesen
2004-03-05 15:15 ` Tom Rini
2004-03-05 23:51 ` Benjamin Herrenschmidt
2004-03-05 0:37 ` Chris Friesen
2004-03-05 2:10 ` Benjamin Herrenschmidt
2004-03-05 5:39 ` Chris Friesen
2004-03-05 5:46 ` Benjamin Herrenschmidt
2004-03-05 17:21 ` Chris Friesen
2004-03-05 23:53 ` Benjamin Herrenschmidt
2004-03-06 1:34 ` Segher Boessenkool [this message]
2004-03-06 2:31 ` Anton Blanchard
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=59D75402-6F0E-11D8-9B46-000A95A4DC02@kernel.crashing.org \
--to=segher@kernel.crashing.org \
--cc=benh@kernel.crashing.org \
--cc=cfriesen@nortelnetworks.com \
--cc=linux-kernel@vger.kernel.org \
--cc=trini@kernel.crashing.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®