From: Paul Mackerras <paulus@samba.org>
To: torvalds@osdl.org
Cc: linux-kernel@vger.kernel.org, benh@kernel.crashing.org,
akpm@osdl.org, segher@kernel.crashing.org, davem@davemloft.net
Subject: Opinion on ordering of writel vs. stores to RAM
Date: Sat, 9 Sep 2006 12:03:29 +1000 [thread overview]
Message-ID: <17666.8433.533221.866510@cargo.ozlabs.ibm.com> (raw)
Linus,
An issue has come up in the tg3 ethernet driver, where we are seeing
data corruption on ppc64 machines that is attributable to a lack of
ordering between writes to normal RAM and writes to an MMIO register.
Basically the driver does writes to RAM and then a writel to an MMIO
register to trigger DMA, and occasionally the device then reads old
values from memory.
Do you have an opinion about whether the MMIO write in writel() should
be ordered with respect to preceding writes to normal memory?
Currently we have a sync instruction after the store in writel() but
not one before. The sync after is to keep the writel inside
spinlocked regions and to ensure that the store is ordered with
respect to the load in readl() and friends.
Paul.
next reply other threads:[~2006-09-09 2:04 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-09 2:03 Paul Mackerras [this message]
2006-09-09 2:42 ` Linus Torvalds
2006-09-09 3:02 ` Paul Mackerras
2006-09-09 3:54 ` Linus Torvalds
2006-09-09 7:24 ` Benjamin Herrenschmidt
2006-09-09 9:34 ` David Miller
2006-09-09 9:55 ` Jeff Garzik
2006-09-09 10:08 ` David Miller
2006-09-10 17:18 ` Jesse Barnes
2006-09-10 19:35 ` Alan Cox
2006-09-10 21:25 ` Benjamin Herrenschmidt
2006-09-10 22:23 ` Alan Cox
2006-09-10 22:18 ` Benjamin Herrenschmidt
2006-09-11 13:19 ` Jes Sorensen
2006-09-10 23:35 ` Segher Boessenkool
2006-09-11 0:12 ` Benjamin Herrenschmidt
2006-09-11 0:34 ` Jesse Barnes
2006-09-11 1:04 ` Benjamin Herrenschmidt
2006-09-11 1:13 ` Segher Boessenkool
2006-09-11 1:35 ` Benjamin Herrenschmidt
2006-09-11 9:02 ` Alan Cox
2006-09-11 9:23 ` Benjamin Herrenschmidt
2006-09-11 0:25 ` Jesse Barnes
2006-09-11 0:54 ` Segher Boessenkool
2006-09-11 1:10 ` Benjamin Herrenschmidt
2006-09-11 1:48 ` Segher Boessenkool
2006-09-11 3:53 ` Benjamin Herrenschmidt
2006-09-11 18:12 ` Jesse Barnes
2006-09-11 1:00 ` Benjamin Herrenschmidt
2006-09-11 18:08 ` Jesse Barnes
2006-09-11 21:32 ` Benjamin Herrenschmidt
2006-09-10 20:01 ` Segher Boessenkool
2006-09-11 13:21 ` David Miller
2006-09-11 14:17 ` Segher Boessenkool
2006-09-12 0:32 ` David Miller
2006-09-12 0:49 ` Benjamin Herrenschmidt
2006-09-12 16:47 ` Segher Boessenkool
2006-09-12 0:54 ` Roland Dreier
2006-09-09 11:16 ` Paul Mackerras
2006-09-09 7:23 ` Benjamin Herrenschmidt
2006-09-09 9:38 ` David Miller
2006-09-09 15:09 ` Alan Cox
2006-09-10 17:19 ` Jesse Barnes
2006-09-10 17:35 ` Michael Buesch
2006-09-10 17:49 ` Linus Torvalds
2006-09-10 18:02 ` Michael Buesch
2006-09-09 15:08 ` Alan Cox
2006-09-09 18:34 ` Auke Kok
2006-09-09 19:10 ` Patrick McFarland
2006-09-09 15:06 ` Alan Cox
2006-09-11 5:03 Michael Chan
2006-09-11 5:21 ` Benjamin Herrenschmidt
2006-09-12 4:30 Albert Cahalan
2006-09-12 5:30 ` Benjamin Herrenschmidt
2006-09-12 6:04 ` Albert Cahalan
2006-09-12 6:12 ` Benjamin Herrenschmidt
2006-09-12 7:09 ` Albert Cahalan
2006-09-12 7:17 ` Benjamin Herrenschmidt
2006-09-12 7:21 ` Albert Cahalan
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=17666.8433.533221.866510@cargo.ozlabs.ibm.com \
--to=paulus@samba.org \
--cc=akpm@osdl.org \
--cc=benh@kernel.crashing.org \
--cc=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=segher@kernel.crashing.org \
--cc=torvalds@osdl.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®