From: Richard Henderson <rth@twiddle.net>
To: David Mosberger <davidm@hpl.hp.com>
Cc: Paul Mackerras <paulus@samba.org>,
torvalds@transmeta.com, linux-kernel@vger.kernel.org,
davem@redhat.com
Subject: Re: [PATCH] avoid unnecessary cache flushes
Date: Mon, 3 Sep 2001 13:41:25 -0700 [thread overview]
Message-ID: <20010903134125.B16069@twiddle.net> (raw)
In-Reply-To: <15247.29338.3671.548678@cargo.ozlabs.ibm.com> <20010903131436.A16069@twiddle.net> <15251.59286.154267.431231@napali.hpl.hp.com>
In-Reply-To: <15251.59286.154267.431231@napali.hpl.hp.com>; from davidm@hpl.hp.com on Mon, Sep 03, 2001 at 01:27:02PM -0700
On Mon, Sep 03, 2001 at 01:27:02PM -0700, David Mosberger wrote:
> >> + if (!test_bit(PG_arch_1, &page->flags)) {
> >> + __flush_dcache_icache((unsigned long)kmap(page));
> >> + kunmap(page);
> >> + set_bit(PG_arch_1, &page->flags);
>
> Richard> Race. Use test_and_set_bit.
>
> Nope, the old code is correct: you must turn on PG_arch_1 *after*
> flushing the cache. Yes, you might sometimes double flush, but that's
> both safe and rare.
You can get a missed flush from
bit == 0
flush cache
modify page
bit = 0
bit = 1
unless this is protected from some outer lock of which
I am not aware.
I do see your point about the early set though.
r~
next prev parent reply other threads:[~2001-09-03 20:41 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-08-31 11:18 [PATCH] avoid unnecessary cache flushes Paul Mackerras
2001-08-31 22:45 ` David S. Miller
2001-09-01 8:35 ` Paul Mackerras
2001-09-01 8:55 ` David S. Miller
2001-09-01 11:47 ` Paul Mackerras
2001-09-01 12:24 ` David S. Miller
2001-09-03 20:14 ` Richard Henderson
2001-09-03 20:27 ` David Mosberger
2001-09-03 20:41 ` Richard Henderson [this message]
2001-09-03 21:00 ` David Mosberger
2001-09-04 16:37 ` Richard Henderson
2001-09-04 17:06 ` David Mosberger
2001-09-04 1:53 ` Paul Mackerras
2001-09-04 2:31 ` Andrea Arcangeli
2001-09-04 4:08 ` Paul Mackerras
2001-09-04 12:58 ` Andrea Arcangeli
2001-09-04 16:44 ` Richard Henderson
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=20010903134125.B16069@twiddle.net \
--to=rth@twiddle.net \
--cc=davem@redhat.com \
--cc=davidm@hpl.hp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=paulus@samba.org \
--cc=torvalds@transmeta.com \
/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