The Linux Kernel Mailing List
 help / color / mirror / Atom feed
From: Suresh Siddha <suresh.b.siddha@intel.com>
To: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Eric Anholt <eric@anholt.net>,
	Thomas Hellstrom <thellstrom@vmware.com>,
	Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Arjan van de Ven <arjan@linux.intel.com>,
	"x86@kernel.org" <x86@kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Jan Beulich <JBeulich@novell.com>, Ingo Molnar <mingo@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Henrique de Moraes Holschuh <hmh@hmh.eng.br>,
	"Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>,
	Tejun Heo <tj@kernel.org>,
	Thomas Schlichter <thomas.schlichter@web.de>,
	"dri-devel@lists.sourceforge.net"
	<dri-devel@lists.sourceforge.net>,
	Yinghai Lu <yinghai@kernel.org>, Ingo Molnar <mingo@elte.hu>,
	Robert Hancock <hancockrwd@gmail.com>
Subject: Re: [RFC Patch] use MTRR for write combining if PAT is not available
Date: Thu, 22 Oct 2009 21:58:15 -0700	[thread overview]
Message-ID: <1256273896.2569.10.camel@sbs-t61> (raw)
In-Reply-To: <20091023133152.1a15ba69@jbarnes-x200>

On Thu, 2009-10-22 at 21:31 -0700, Jesse Barnes wrote:
> Here's a quick & dirty version, totally untested.  A cleaner approach
> would be to separate the WC mapping routines and hide the return
> -EINVAL in arch specific code...

Jesse How about this patch? Doing this in x86 is cleaner.

I would like Acks/sign-offs-by Thomas, Eric and Jesse, if it is ok with
this patch and works.

thanks,
suresh
---

From: Suresh Siddha <suresh.b.siddha@intel.com>
Subject: x86, pat: return EINVAL for pci mmap WC request for !pat_enabled

Thomas Schlichter reported:
> X.org uses libpciaccess which tries to mmap with write combining enabled via
> /sys/bus/pci/devices/*/resource0_wc. Currently, when PAT is not enabled, the
> kernel does fall back to uncached mmap. Then libpciaccess thinks it succeeded
> mapping with write combining enabled and does not set up suited MTRR entries.
> ;-(

Instead of silently mapping pci mmap region as UC minus in the case
of !pat_enabled and wc request, we can return error. Eric Anholt mentioned
that caller (like X) typically follows up with UC minus pci mmap request and
if there is a free mtrr slot, caller will manage adding WC mtrr.

Jesse Barnes says:
> Older versions of libpciaccess will behave better if we do it that way
> (iirc it only allocates an MTRR if the resource_wc file doesn't exist or
> fails to get mapped).

Reported-by: Thomas Schlichter <thomas.schlichter@web.de>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---

diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c
index b22d13b..a672f12 100644
--- a/arch/x86/pci/i386.c
+++ b/arch/x86/pci/i386.c
@@ -282,6 +282,15 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
 		return -EINVAL;
 
 	prot = pgprot_val(vma->vm_page_prot);
+
+	/*
+ 	 * Return error if pat is not enabled and write_combine is requested.
+ 	 * Caller can followup with UC MINUS request and add a WC mtrr if there
+ 	 * is a free mtrr slot.
+ 	 */
+	if (!pat_enabled && write_combine)
+		return -EINVAL;
+
 	if (pat_enabled && write_combine)
 		prot |= _PAGE_CACHE_WC;
 	else if (pat_enabled || boot_cpu_data.x86 > 3)



  reply	other threads:[~2009-10-23  4:58 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-10-21 13:45 [RFC Patch] use MTRR for write combining if PAT is not available Thomas Schlichter
2009-10-21 14:11 ` Jan Beulich
2009-10-21 17:35   ` Ingo Molnar
2009-10-21 20:01     ` Thomas Schlichter
2009-10-22  9:53       ` Suresh Siddha
2009-10-22 15:34         ` Eric Anholt
2009-10-22 21:47           ` Suresh Siddha
2009-10-22 23:10             ` Jesse Barnes
2009-10-23  0:11               ` Suresh Siddha
2009-10-23  1:53                 ` Eric Anholt
2009-10-23  4:31                   ` Jesse Barnes
2009-10-23  4:58                     ` Suresh Siddha [this message]
2009-10-23  7:24                       ` Thomas Schlichter
2009-10-23 14:24                         ` Suresh Siddha
2009-10-23 14:37                           ` Ingo Molnar
2009-10-23  4:33                   ` Suresh Siddha
  -- strict thread matches above, loose matches on Subject: below --
2009-10-22 14:41 Thomas Schlichter
2009-10-22 14:27 Thomas Schlichter
2009-10-22 12:08 Thomas Schlichter
2009-10-22 12:14 ` H. Peter Anvin
2009-10-22 13:26   ` Suresh Siddha
2009-10-22 13:35 ` Suresh Siddha
2009-10-21 14:38 Thomas Schlichter
2009-10-21 15:14 ` Jan Beulich
2009-10-19 15:10 Thomas Schlichter
2009-10-19 15:28 ` Ingo Molnar
2009-10-19 15:07 Thomas Schlichter
2009-10-19 14:59 Thomas Schlichter
2009-10-19 15:31 ` Ingo Molnar
2009-10-19 21:49   ` Suresh Siddha
2009-10-20 20:35   ` Thomas Schlichter
2009-10-20 21:59     ` Suresh Siddha
2009-10-21 11:52       ` Ingo Molnar
2009-10-19 14:47 Thomas Schlichter
2009-10-20 19:54 ` Thomas Schlichter
2009-10-21 11:57   ` Ingo Molnar
2009-10-13  7:34 Jan Beulich
2009-10-13 21:29 ` Thomas Schlichter
2009-10-14  8:13   ` Jan Beulich
2009-10-14 19:14     ` Thomas Schlichter
2009-10-15  7:48       ` Jan Beulich
2009-10-17 19:48         ` Thomas Schlichter
2009-10-19  9:16           ` Jan Beulich
2009-10-19 13:44             ` Suresh Siddha
2009-10-19 13:54               ` Ingo Molnar
2009-10-19 13:36           ` Konrad Rzeszutek Wilk
     [not found] <200910122032.52168.thomas.schlichter@web.de>
2009-10-12 19:16 ` Thomas Hellstrom
2009-10-12 19:45   ` Thomas Schlichter
     [not found]     ` <1255378684.2063.5.camel@gaiman.anholt.net>
2009-10-13 21:05       ` Thomas Schlichter
2009-10-10  1:22 Thomas Schlichter
2009-10-10  4:24 ` Arjan van de Ven
2009-10-10  8:31   ` Thomas Schlichter
2009-10-10 15:45     ` Arjan van de Ven
2009-10-10 17:50       ` Roland Dreier
2009-10-11  7:40       ` Ingo Molnar
2009-10-11  9:56       ` Thomas Schlichter
2009-10-11 18:51   ` Henrique de Moraes Holschuh
2009-10-11 18:54     ` Arjan van de Ven
2009-10-11 20:19       ` Henrique de Moraes Holschuh
2009-10-12 18:09         ` Robert Hancock

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=1256273896.2569.10.camel@sbs-t61 \
    --to=suresh.b.siddha@intel.com \
    --cc=JBeulich@novell.com \
    --cc=arjan@linux.intel.com \
    --cc=dri-devel@lists.sourceforge.net \
    --cc=eric@anholt.net \
    --cc=hancockrwd@gmail.com \
    --cc=hmh@hmh.eng.br \
    --cc=hpa@zytor.com \
    --cc=jbarnes@virtuousgeek.org \
    --cc=jeremy.fitzhardinge@citrix.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=mingo@redhat.com \
    --cc=tglx@linutronix.de \
    --cc=thellstrom@vmware.com \
    --cc=thomas.schlichter@web.de \
    --cc=tj@kernel.org \
    --cc=venkatesh.pallipadi@intel.com \
    --cc=x86@kernel.org \
    --cc=yinghai@kernel.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