From: Rene Herman <rene.herman@keyaccess.nl>
To: Robert Hancock <hancockr@shaw.ca>
Cc: "David P. Reed" <dpreed@reed.com>,
linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>
Subject: Re: RFC: outb 0x80 in inb_p, outb_p harmful on some modern AMD64 with MCP51 laptops
Date: Fri, 07 Dec 2007 06:09:24 +0100 [thread overview]
Message-ID: <4758D584.90105@keyaccess.nl> (raw)
In-Reply-To: <4758927F.50500@shaw.ca>
On 07-12-07 01:23, Robert Hancock wrote:
> David P. Reed wrote:
>> After much, much testing (months, off and on, pursuing hypotheses),
>> I've discovered that the use of "outb al,0x80" instructions to "delay"
>> after inb and outb instructions causes solid freezes on my HP dv9000z
>> laptop, when ACPI is enabled.
>>
>> It takes a fair number of out's to 0x80, but the hard freeze is
>> reliably reproducible by writing a driver that solely does a loop of
>> 50 outb's to 0x80 and calling it in a loop 1000 times from user
>> space. !!!
>>
>> The serious impact is that the /dev/rtc and /dev/nvram devices are
>> very unreliable - thus "hwclock" freezes very reliably while looping
>> waiting for a new second value and calling "cat /dev/nvram" in a loop
>> freezes the machine if done a few times in a row.
>>
>> This is reproducible, but requires a fair number of outb's to the 0x80
>> diagnostic port, and seems to require ACPI to be on.
>>
>> io_64.h is the source of these particular instructions, via the
>> CMOS_READ and CMOS_WRITE macros, which are defined in mc146818_64.h.
>> (I wonder if the same problem occurs in 32-bit mode).
>>
>> I'm happy to complete and test a patch, but I'm curious what the right
>> approach ought to be. I have to say I have no clue as to what ACPI is
>> doing on this chipset (nvidia MCP51) that would make port 80 do
>> this. A raw random guess is that something is logging POST codes, but
>> if so, not clear what is problematic in ACPI mode.
>>
>> ANy help/suggestions?
>>
>> Changing the delay instruction sequence from the outb to short jumps
>> might be the safe thing. But Linus, et al. may have experience with
>> that on other architectures like older Pentiums etc.
>
> The fact that these "pausing" calls are needed in the first place seems
> rather cheesy. If there's hardware that's unable to respond to IO port
> writes as fast as possible, then surely there's a better solution than
> trying to stall the IOs by an arbitrary and hardware-dependent amount of
> time, like udelay calls, etc. Does any remotely recent hardware even
> need this?
The idea is that the delay is not in fact hardware dependent. With in the
the absense of a POST board port 0x80 being sort of guaranteeed to not be
decoded on PCI but forwarded to and left to die on ISA/LPC one should get
the effect that the _next_ write will have survived an ISA/LPC bus address
cycle acknowledgement timeout.
I believe.
And no, I don't believe any remotely recent hardware needs it and have in
fact wondered about it since actual 386 days, having since that time never
found a device that wouldn't in fact take back to back I/O even. Even back
then (ie, legacy only systems, no forwarding from PCI or anything) BIOSes
provided ISA bus wait-state settings which should be involved in getting
insanely stupid and old hardware to behave...
Port 0xed has been suggested as an alternate port. Probably not a great
"fix" but if replacing the out with a simple udelay() isn't that simple
(during early boot I gather) then it might at least be something for you to
try. I'd hope that the 0x80 in include/asm/io.h:native_io_delay() would be
the only one you are running into, so you could change that to 0xed and see
what catches fire.
If there are no sensible fixes, an 0x80/0xed choice could I assume be hung
of DMI or something (if that _is_ parsed soon enough).
Rene.
next prev parent reply other threads:[~2007-12-07 5:10 UTC|newest]
Thread overview: 109+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <fa./27SNSh+L5T3iqFNPdHClEu+yT0@ifi.uio.no>
2007-12-07 0:23 ` RFC: outb 0x80 in inb_p, outb_p harmful on some modern AMD64 with MCP51 laptops Robert Hancock
2007-12-07 5:09 ` Rene Herman [this message]
2007-12-07 5:54 ` David P. Reed
2007-12-07 7:17 ` Rene Herman
2007-12-07 7:34 ` Rene Herman
2007-12-07 10:49 ` Andi Kleen
2007-12-06 22:38 David P. Reed
2007-12-07 0:15 ` Alan Cox
2007-12-07 10:44 ` Andi Kleen
2007-12-07 14:50 ` David P. Reed
2007-12-05 11:10 ` Pavel Machek
2007-12-08 0:21 ` Andi Kleen
2007-12-07 14:54 ` Andi Kleen
2007-12-07 15:43 ` Rene Herman
2007-12-07 16:28 ` Rene Herman
2007-12-11 0:31 ` H. Peter Anvin
2007-12-11 5:53 ` H. Peter Anvin
2007-12-07 16:04 ` Alan Cox
2007-12-07 16:31 ` Andi Kleen
2007-12-07 17:19 ` Alan Cox
2007-12-07 18:45 ` Rene Herman
2007-12-07 18:42 ` Alan Cox
2007-12-07 19:25 ` Rene Herman
2007-12-07 21:45 ` Alan Cox
2007-12-08 19:25 ` David P. Reed
2007-12-08 19:50 ` Andi Kleen
2007-12-08 20:47 ` David P. Reed
2007-12-08 21:04 ` Alan Cox
2007-12-08 20:26 ` Alan Cox
2007-12-11 5:58 ` H. Peter Anvin
2007-12-09 5:04 ` Rene Herman
2007-12-09 13:22 ` Pavel Machek
2007-12-11 15:14 ` Lennart Sorensen
2007-12-09 12:54 ` Pavel Machek
2007-12-09 13:41 ` Dr. David Alan Gilbert
2007-12-09 15:54 ` Ondrej Zary
2007-12-09 16:59 ` Andi Kleen
2007-12-09 21:25 ` Pavel Machek
2007-12-09 22:29 ` Alan Cox
2007-12-09 23:22 ` Pavel Machek
2007-12-10 12:02 ` Alan Cox
2007-12-10 4:17 ` Rene Herman
2007-12-10 11:30 ` Krzysztof Halasa
2007-12-10 12:08 ` Rene Herman
2007-12-10 18:02 ` Ondrej Zary
2007-12-11 1:10 ` David Newall
2007-12-11 1:25 ` H. Peter Anvin
2007-12-11 1:42 ` David Newall
2007-12-11 1:46 ` H. Peter Anvin
2007-12-11 1:51 ` H. Peter Anvin
2007-12-11 7:40 ` Paul Rolland
2007-12-11 9:50 ` Rene Herman
2007-12-11 12:08 ` David Newall
2007-12-11 13:16 ` Rene Herman
2007-12-11 13:32 ` Paul Rolland
2007-12-11 14:15 ` Rene Herman
2007-12-11 15:28 ` Rene Herman
2007-12-11 15:37 ` Paul Rolland
2007-12-11 15:53 ` Rene Herman
2007-12-11 16:58 ` David P. Reed
2007-12-11 17:01 ` Rene Herman
2007-12-11 17:05 ` H. Peter Anvin
2007-12-11 17:32 ` Alan Cox
2007-12-11 19:19 ` David P. Reed
2007-12-11 19:36 ` Pavel Machek
2007-12-11 20:16 ` Alan Cox
2007-12-11 20:27 ` linux-os (Dick Johnson)
2007-12-11 20:34 ` Rene Herman
2007-12-11 21:03 ` David P. Reed
2007-12-11 23:56 ` David P. Reed
2007-12-12 13:11 ` linux-os (Dick Johnson)
2007-12-12 16:12 ` Alan Cox
2007-12-14 14:33 ` Ingo Molnar
2007-12-16 21:26 ` Pavel Machek
2007-12-17 0:02 ` Alan Cox
2007-12-17 0:03 ` Alan Cox
2007-12-17 0:28 ` Pavel Machek
2007-12-17 14:42 ` Ingo Molnar
2007-12-27 10:39 ` Pavel Machek
2007-12-11 16:32 ` John Stoffel
2007-12-11 16:40 ` Rene Herman
2007-12-11 13:50 ` David Newall
2007-12-11 14:14 ` Rene Herman
2007-12-11 14:25 ` Alan Cox
2007-12-12 22:18 ` David Newall
2007-12-12 23:00 ` Alan Cox
2007-12-13 13:13 ` David P. Reed
2007-12-13 13:21 ` Alan Cox
2007-12-14 1:50 ` David P. Reed
2007-12-14 15:16 ` Alan Cox
2007-12-11 15:41 ` linux-os (Dick Johnson)
2007-12-11 16:30 ` Andi Kleen
2007-12-11 16:50 ` Rene Herman
2007-12-11 17:00 ` David P. Reed
2007-12-11 17:04 ` Rene Herman
2007-12-11 17:27 ` Rene Herman
2007-12-11 19:18 ` Pavel Machek
2007-12-11 19:16 ` Pavel Machek
2007-12-11 19:59 ` Rene Herman
2007-12-11 19:59 ` Rene Herman
2007-12-11 20:00 ` Rene Herman
2007-12-11 20:00 ` Rene Herman
2007-12-11 13:14 ` Alan Cox
2007-12-11 13:32 ` Andi Kleen
2007-12-11 13:47 ` Pavel Machek
2007-12-11 13:50 ` Andi Kleen
2007-12-14 13:33 ` Ingo Molnar
2007-12-11 6:54 ` Rene Herman
2007-12-11 17:01 ` H. Peter Anvin
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=4758D584.90105@keyaccess.nl \
--to=rene.herman@keyaccess.nl \
--cc=dpreed@reed.com \
--cc=hancockr@shaw.ca \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
/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;
as well as URLs for NNTP newsgroup(s).