linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thierry Reding <thierry.reding@avionic-design.de>
To: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Cc: Stephen Warren <swarren@wwwdotorg.org>,
	Arnd Bergmann <arnd@arndb.de>,
	linux-tegra@vger.kernel.org,
	Grant Likely <grant.likely@secretlab.ca>,
	Rob Herring <rob.herring@calxeda.com>,
	Russell King <linux@arm.linux.org.uk>,
	Bjorn Helgaas <bhelgaas@google.com>,
	Andrew Murray <andrew.murray@arm.com>,
	Thomas Petazzoni <thomas.petazzoni@free-electrons.com>,
	devicetree-discuss@lists.ozlabs.org,
	linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, linux-pci@vger.kernel.org
Subject: Re: [PATCH 05/14] lib: Add I/O map cache implementation
Date: Thu, 10 Jan 2013 08:19:37 +0100	[thread overview]
Message-ID: <20130110071937.GG15212@avionic-0098.adnet.avionic-design.de> (raw)
In-Reply-To: <20130109231758.GA27065@obsidianresearch.com>

[-- Attachment #1: Type: text/plain, Size: 1930 bytes --]

On Wed, Jan 09, 2013 at 04:17:58PM -0700, Jason Gunthorpe wrote:
> On Wed, Jan 09, 2013 at 04:12:31PM -0700, Stephen Warren wrote:
> > On 01/09/2013 03:10 PM, Arnd Bergmann wrote:
> > > On Wednesday 09 January 2013, Thierry Reding wrote:
> > >> What happens on Tegra is that we need to map 256 MiB of physical memory
> > >> to access all the PCIe extended configuration space. However, ioremap()
> > >> on such a large region fails if not enough vmalloc() space is available.
> > >>
> > >> This was observed when somebody tested this on CardHu which has a 1 GiB
> > >> of RAM and therefore remapping the full 256 MiB fails.
> > ...
> > > Have you checked if the hardware supports an alternative config
> > > space access mechanism that does not depend on a huge address range?
> > > A lot of them provide an index/data register pair somewhere, as the
> > > original PC implementation did.
> > 
> > That would be nice, but I've talked to the HW engineers, and there's no
> > indication that any alternative mechanism exists.
> 
> It seems to be convention that extended config space is often only
> accessible through mmio space, that was true on x86 last I checked
> too..
> 
> You could decrease the size of the mapping to only span the bus
> numbers that are configured for use via DT.

That won't work, unfortunately. The mapping is such that the bus number
is not encoded in the uppermost bits, the extended register number is.
So the only thing that we could do is decrease the size of the extended
register space for *all* devices.

> Are there any concerns about these config registers being accessed
> from a context where a new mapping can't be made? Interrupt? Machine
> Check? PCI-E Advanced Error Reporting?

I haven't checked but I would expect configuration space accesses to not
happen in interrupt context. Usually they are limited to enumeration and
driver probe.

Thierry

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

  reply	other threads:[~2013-01-10  7:19 UTC|newest]

Thread overview: 93+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-01-09 20:43 [PATCH 00/14] Rewrite Tegra PCIe driver Thierry Reding
2013-01-09 20:43 ` [PATCH 01/14] of/pci: Provide support for parsing PCI DT ranges property Thierry Reding
2013-01-11  0:06   ` Stephen Warren
2013-01-11  4:02     ` Thierry Reding
2013-01-09 20:43 ` [PATCH 02/14] of/pci: Add of_pci_get_devfn() function Thierry Reding
2013-01-11  0:09   ` Stephen Warren
2013-01-11  4:06     ` Thierry Reding
2013-01-09 20:43 ` [PATCH 03/14] of/pci: Add of_pci_get_bus() function Thierry Reding
2013-01-09 20:43 ` [PATCH 04/14] of/pci: Add of_pci_parse_bus_range() function Thierry Reding
2013-01-09 20:43 ` [PATCH 05/14] lib: Add I/O map cache implementation Thierry Reding
2013-01-09 21:19   ` Arnd Bergmann
2013-01-09 21:54     ` Thierry Reding
2013-01-09 22:10       ` Arnd Bergmann
2013-01-09 23:12         ` Stephen Warren
2013-01-09 23:17           ` Jason Gunthorpe
2013-01-10  7:19             ` Thierry Reding [this message]
2013-01-10  9:17               ` Arnd Bergmann
2013-01-10 10:25                 ` Thierry Reding
2013-01-10 18:20                   ` Jason Gunthorpe
2013-01-10 18:55                     ` Thierry Reding
2013-01-10 19:03                       ` Thierry Reding
2013-01-10 19:24                         ` Jason Gunthorpe
2013-01-10 20:20                           ` Thierry Reding
2013-01-10 21:06                             ` Jason Gunthorpe
2013-01-16 10:18                           ` Thierry Reding
2013-01-16 11:25                             ` Russell King - ARM Linux
2013-01-16 11:52                               ` Thierry Reding
2013-01-10 18:26                   ` Arnd Bergmann
2013-01-10 18:57                     ` Thierry Reding
2013-01-10  7:10         ` Thierry Reding
2013-01-09 21:28   ` Russell King - ARM Linux
2013-01-09 21:57     ` Thierry Reding
2013-01-09 20:43 ` [PATCH 06/14] ARM: pci: Keep pci_common_init() around after init Thierry Reding
2013-02-05 20:41   ` Thierry Reding
2013-02-06 16:30     ` Russell King - ARM Linux
2013-02-06 19:35       ` Thierry Reding
2013-02-06  8:36   ` Thomas Petazzoni
2013-02-06 16:38   ` Linus Walleij
2013-02-07  0:54     ` Arnd Bergmann
2013-02-06 17:07       ` Linus Walleij
2013-02-07  1:20         ` Arnd Bergmann
2013-01-09 20:43 ` [PATCH 07/14] ARM: pci: Allow passing per-controller private data Thierry Reding
2013-01-09 20:43 ` [PATCH 08/14] ARM: tegra: Move tegra_pcie_xclk_clamp() to PMC Thierry Reding
2013-01-09 20:43 ` [PATCH 09/14] ARM: tegra: Move pmc.h to include/mach Thierry Reding
2013-01-11  0:15   ` Stephen Warren
2013-01-11  4:08     ` Thierry Reding
2013-01-09 20:43 ` [PATCH 10/14] PCI: tegra: Move PCIe driver to drivers/pci/host Thierry Reding
2013-01-09 21:22   ` Arnd Bergmann
2013-01-09 21:58     ` Thierry Reding
2013-01-09 22:03       ` Arnd Bergmann
2013-01-10 23:54   ` Stephen Warren
2013-01-11  3:40     ` Thierry Reding
2013-01-11 15:36       ` Arnd Bergmann
2013-01-11 15:45         ` Thierry Reding
2013-01-12 12:36           ` Thierry Reding
2013-01-12 21:12             ` Arnd Bergmann
2013-01-13  9:58               ` Thierry Reding
2013-01-14  9:57                 ` Andrew Murray
2013-01-15 12:08                   ` Thierry Reding
2013-01-15 12:44                     ` Arnd Bergmann
2013-01-15 15:40                       ` Andrew Murray
2013-01-15 21:14                         ` Thierry Reding
2013-01-16 14:00                           ` Arnd Bergmann
2013-01-16 16:17                             ` Andrew Murray
2013-01-16 18:31                               ` Thierry Reding
2013-01-17 15:42                                 ` Andrew Murray
2013-01-17 16:05                                   ` Thierry Reding
2013-01-17 16:22                                     ` Andrew Murray
2013-01-17 20:30                                       ` Thierry Reding
2013-01-18  9:18                                         ` Andrew Murray
2013-01-22 19:29                                       ` Jason Gunthorpe
2013-01-29 13:31                                         ` Andrew Murray
2013-01-11  0:48   ` Stephen Warren
2013-01-11  3:52     ` Thierry Reding
2013-01-11 20:34       ` Stephen Warren
2013-01-18  9:56   ` Andrew Murray
2013-01-18 10:09     ` Thierry Reding
2013-02-13 23:11   ` Thomas Petazzoni
2013-01-09 20:43 ` [PATCH 11/14] ARM: tegra: tamonten: Add PCIe support Thierry Reding
2013-01-09 21:23   ` Arnd Bergmann
2013-01-10 20:21     ` Thierry Reding
2013-01-09 20:43 ` [PATCH 12/14] ARM: tegra: tec: " Thierry Reding
2013-01-11  0:22   ` Stephen Warren
2013-01-11  4:34     ` Thierry Reding
2013-01-09 20:43 ` [PATCH 13/14] ARM: tegra: harmony: Initialize PCIe from DT Thierry Reding
2013-01-10 23:58   ` Stephen Warren
2013-01-09 20:43 ` [PATCH 14/14] ARM: tegra: trimslice: " Thierry Reding
2013-01-10 23:56   ` Stephen Warren
2013-01-11 18:48     ` Thierry Reding
2013-01-09 21:25 ` [PATCH 00/14] Rewrite Tegra PCIe driver Thomas Petazzoni
2013-01-10  6:55   ` Thierry Reding
2013-01-10  8:34     ` Thomas Petazzoni
2013-01-28 18:15 ` Bjorn Helgaas

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=20130110071937.GG15212@avionic-0098.adnet.avionic-design.de \
    --to=thierry.reding@avionic-design.de \
    --cc=andrew.murray@arm.com \
    --cc=arnd@arndb.de \
    --cc=bhelgaas@google.com \
    --cc=devicetree-discuss@lists.ozlabs.org \
    --cc=grant.likely@secretlab.ca \
    --cc=jgunthorpe@obsidianresearch.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-tegra@vger.kernel.org \
    --cc=linux@arm.linux.org.uk \
    --cc=rob.herring@calxeda.com \
    --cc=swarren@wwwdotorg.org \
    --cc=thomas.petazzoni@free-electrons.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).