From: "H. Peter Anvin" <hpa@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
Chuck Ebbert <cebbert.lkml@gmail.com>
Cc: linux-kernel@vger.kernel.org,
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>,
Jiri Kosina <jkosina@suse.cz>,
Steven Rostedt <rostedt@goodmis.org>,
Jason Baron <jbaron@akamai.com>,
yrl.pp-manager.tt@hitachi.com, Borislav Petkov <bpetkov@suse.de>,
Ingo Molnar <mingo@kernel.org>,
Daniel Vetter <daniel.vetter@ffwll.ch>,
x86@kernel.org, Thomas Gleixner <tglx@linutronix.de>
Subject: Re: i915.ko WC writes are slow after ea8596bb2d8d379
Date: Wed, 08 Oct 2014 14:36:49 -0700 [thread overview]
Message-ID: <5435AE71.60304@linux.intel.com> (raw)
In-Reply-To: <20141008194955.GF12897@nuc-i3427.alporthouse.com>
On 10/08/2014 12:49 PM, Chris Wilson wrote:
>
> Indeed, this appears to be the explanation. (And here I thought PAT
> superseded mtrrs - i915.ko stopped trying to use assign an mtrr for its
> GTT quite a while ago.)
>
> Replacing the stop_machine there with on_each_cpu does the trick:
>
It should, but there seem to be quite a few drivers which still muck
with MTRRs. However, i915 is not one of them, it calls
io_mapping_create_wc() followed by arch_phys_wc_add(), so I'm wondering
what the heck is going on here.
> Naively I would say that we lost the wc on our ioremap.
> /sys/kernel/debug/x86/pat_memtype_list remained the same across repeated
> runs.
Could you tell me what the above looks like?
-hpa
next prev parent reply other threads:[~2014-10-08 21:37 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-08 9:03 i915.ko WC writes are slow after ea8596bb2d8d379 Chris Wilson
2014-10-08 10:10 ` Chuck Ebbert
2014-10-08 19:49 ` Chris Wilson
2014-10-08 21:36 ` H. Peter Anvin [this message]
2014-10-09 6:53 ` Chris Wilson
2014-10-09 12:44 ` Chuck Ebbert
2014-10-09 13:00 ` Chris Wilson
2014-10-09 14:46 ` Chuck Ebbert
2014-10-09 15:14 ` Chris Wilson
2015-11-18 14:48 ` Chris Wilson
2015-11-18 15:57 ` Andy Lutomirski
2015-11-19 8:14 ` Ingo Molnar
2015-11-19 10:03 ` [PATCH] kernel: Remove stop_machine() Kconfig dependency Chris Wilson
2015-11-19 8:16 ` i915.ko WC writes are slow after ea8596bb2d8d379 Ingo Molnar
2014-10-08 17:47 ` Chuck Ebbert
2014-10-09 1:45 ` Masami Hiramatsu
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=5435AE71.60304@linux.intel.com \
--to=hpa@linux.intel.com \
--cc=bpetkov@suse.de \
--cc=cebbert.lkml@gmail.com \
--cc=chris@chris-wilson.co.uk \
--cc=daniel.vetter@ffwll.ch \
--cc=jbaron@akamai.com \
--cc=jkosina@suse.cz \
--cc=linux-kernel@vger.kernel.org \
--cc=masami.hiramatsu.pt@hitachi.com \
--cc=mingo@kernel.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
--cc=yrl.pp-manager.tt@hitachi.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;
as well as URLs for NNTP newsgroup(s).