LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Sergio Paracuellos <sergio.paracuellos@gmail.com>
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
	Bjorn Helgaas <bhelgaas@google.com>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	Greg KH <gregkh@linuxfoundation.org>,
	"open list:MIPS" <linux-mips@vger.kernel.org>,
	linux-pci <linux-pci@vger.kernel.org>,
	linux-staging@lists.linux.dev, NeilBrown <neil@brown.name>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Liviu Dudau <Liviu.Dudau@arm.com>, Rob Herring <robh@kernel.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Arnd Bergmann <arnd@arndb.de>
Subject: Re: [PATCH v2 2/3] PCI: fix 'pci_remap_iospace' for architectures with PCI_IOBASE not defined
Date: Tue, 21 Sep 2021 20:36:07 +0200	[thread overview]
Message-ID: <CAMhs-H90yvbnPN26C7P_431cYqD4=8yrCUrxUY3L5k1GdsCeiQ@mail.gmail.com> (raw)
In-Reply-To: <20210921174924.GA104830@bhelgaas>

Hi Bjorn,

On Tue, Sep 21, 2021 at 7:49 PM Bjorn Helgaas <helgaas@kernel.org> wrote:
>
> [+cc Liviu, Rob, Catalin, Arnd, since this warning was added by
> 8b921acfeffd ("PCI: Add pci_remap_iospace() to map bus I/O resources"),
> https://git.kernel.org/linus/8b921acfeffd]
>
> On Sun, Aug 22, 2021 at 06:10:04PM +0200, Sergio Paracuellos wrote:
> > Request for I/O resources from device tree call 'pci_remap_iospace' from
> > 'devm_pci_remap_iospace' which is also called from device tree function
> > 'pci_parse_request_of_pci_ranges'. if PCI_IOBASE is not defined and I/O
> > resources are requested the following warning appears:
> >
> > ------------[ cut here ]------------
> > WARNING: CPU: 2 PID: 1 at ../drivers/pci/pci.c:4066 pci_remap_iospace+0x3c/0x54
> > This architecture does not support memory mapped I/O
> > Modules linked in:
> > CPU: 2 PID: 1 Comm: swapper/0 Not tainted 5.10.1+ #1228
> > Stack : 00000000 00000000 807fa974 00000000 827ffa80 80066b48 80710000 0000000b
> >         00000000 00000000 81c59aac 7d06ddec 80850000 00000001 81c59a40 7d06ddec
> >         00000000 00000000 807c909c 81c598f0 00000001 81c59904 00000000 0000000a
> >         203a6d6d 80708880 0000000f 70617773 80850000 00000000 00000000 807d0000
> >         807ffecc 1e160000 00000001 00000200 00000000 8054e920 00000008 815e0008
> >         ...
> > Call Trace:
> > [<80008efc>] show_stack+0x8c/0x130
> > [<806e1674>] dump_stack+0x9c/0xc8
> > [<80024a3c>] __warn+0xc0/0xe8
> > [<80024ad0>] warn_slowpath_fmt+0x6c/0xbc
> > [<80410ca8>] pci_remap_iospace+0x3c/0x54
> > [<80410d20>] devm_pci_remap_iospace+0x58/0xa4
> > [<8042019c>] devm_of_pci_bridge_init+0x4dc/0x55c
> > [<80408de8>] devm_pci_alloc_host_bridge+0x78/0x88
> > [<80424e44>] mt7621_pci_probe+0x68/0x9a4
> > [<80464804>] platform_drv_probe+0x40/0x7c
> > [<804628bc>] really_probe+0x2fc/0x4e4
> > [<80463214>] device_driver_attach+0x4c/0x74
> > [<80463384>] __driver_attach+0x148/0x150
> > [<8046047c>] bus_for_each_dev+0x6c/0xb0
> > [<804614dc>] bus_add_driver+0x1b4/0x1fc
> > [<80463aa0>] driver_register+0xd0/0x110
> > [<80001714>] do_one_initcall+0x84/0x1c0
> > [<808e7fd0>] kernel_init_freeable+0x214/0x24c
> > [<806e4164>] kernel_init+0x14/0x118
> > [<80003358>] ret_from_kernel_thread+0x14/0x1c
> >
> > ---[ end trace 1c9d4412bd51b53c ]---
> > mt7621-pci 1e140000.pcie: error -19: failed to map resource [io  0x1e160000-0x1e16ffff]
> >
> > Since there are architectures (like MIPS ralink) that can request I/O
> > resources from device tree but have not mapeable I/O space and also PCI_IOBASE
> > not defined, avoid this warning and just return zero to make the I/O ranges
> > assignment work.
> >
> > Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
> > ---
> >  drivers/pci/pci.c | 12 ++++++------
> >  1 file changed, 6 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> > index aacf575c15cf..10bb2191f376 100644
> > --- a/drivers/pci/pci.c
> > +++ b/drivers/pci/pci.c
> > @@ -4102,9 +4102,9 @@ unsigned long __weak pci_address_to_pio(phys_addr_t address)
> >   * @phys_addr: physical address of range to be mapped
> >   *
> >   * Remap the memory mapped I/O space described by the @res and the CPU
> > - * physical address @phys_addr into virtual address space.  Only
> > - * architectures that have memory mapped IO functions defined (and the
> > - * PCI_IOBASE value defined) should call this function.
> > + * physical address @phys_addr into virtual address space. There
> > + * are architectures that don't define PCI_IOBASE but can have not
> > + * mapeable IO space. Return zero for those cases.
> >   */
> >  int pci_remap_iospace(const struct resource *res, phys_addr_t phys_addr)
> >  {
> > @@ -4122,10 +4122,10 @@ int pci_remap_iospace(const struct resource *res, phys_addr_t phys_addr)
> >  #else
> >       /*
> >        * This architecture does not have memory mapped I/O space,
> > -      * so this function should never be called
> > +      * but can have not mapeable I/O space, so just return ok
> > +      * here.
> >        */
> > -     WARN_ONCE(1, "This architecture does not support memory mapped I/O\n");
> > -     return -ENODEV;
> > +     return 0;
>
> This doesn't seem right to me.  pci_remap_iospace() remaps
> memory-mapped I/O space into virtual address space.
>
> If the architecture doesn't support that remapping, we shouldn't claim
> that it succeeded.
>
> The analogous path for ACPI is in acpi_pci_root_remap_iospace(), where
> we only call pci_remap_iospace() if PCI_IOBASE is defined.  Maybe we
> should use the same approach here, i.e., add the corresonding #ifdef
> in pci_parse_request_of_pci_ranges()?

This approach you are talking here is what I did in v1 of this series
[0] and you tell me to properly fix that in the 'pci_remap_iospace'.
Both of these v1 and v2 obviously end up in a working pci system for
mt7621 ralink.
Another option I guess would be to mark 'pci_remap_iospace' as weak
and redefine the symbol for ralink but since it is only for one simple
return zero maybe it makes no sense at all... Please, tell me the
correct approach to properly fix this and make things work for this
platform.

Thanks in advance for your time,
     Sergio Paracuellos

[0]: https://lore.kernel.org/linux-pci/20210807072409.9018-2-sergio.paracuellos@gmail.com/T/#m73c19c3b72fdf8c63e6d3fe8dc80aeee4e4adcaa

>
> >  #endif
> >  }
> >  EXPORT_SYMBOL(pci_remap_iospace);
> > --
> > 2.25.1
> >

  reply	other threads:[~2021-09-21 18:36 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-22 16:10 [PATCH v2 0/3] MIPS: ralink: properly handle pci IO resources Sergio Paracuellos
2021-08-22 16:10 ` [PATCH v2 1/3] MIPS: ralink: don't define PC_IOBASE but increase IO_SPACE_LIMIT Sergio Paracuellos
2021-09-02 11:08   ` Thomas Bogendoerfer
2021-08-22 16:10 ` [PATCH v2 2/3] PCI: fix 'pci_remap_iospace' for architectures with PCI_IOBASE not defined Sergio Paracuellos
2021-09-15 10:52   ` Sergio Paracuellos
2021-09-21 17:49   ` Bjorn Helgaas
2021-09-21 18:36     ` Sergio Paracuellos [this message]
2021-09-21 19:38       ` Bjorn Helgaas
2021-09-22  4:28         ` Sergio Paracuellos
2021-08-22 16:10 ` [PATCH v2 3/3] staging: mt7621-pci: set end limit for 'ioport_resource' Sergio Paracuellos
2021-08-27  9:01   ` Greg KH
2021-08-29 15:14     ` Sergio Paracuellos
2021-09-02  9:15       ` Greg KH
2021-09-02 10:15         ` Sergio Paracuellos
2021-09-02 11:08           ` Thomas Bogendoerfer
2021-09-02 11:19             ` Sergio Paracuellos
2021-09-21 15:28               ` Greg KH
2021-09-21 17:02                 ` Sergio Paracuellos
2021-09-21 17:08                   ` Greg KH

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='CAMhs-H90yvbnPN26C7P_431cYqD4=8yrCUrxUY3L5k1GdsCeiQ@mail.gmail.com' \
    --to=sergio.paracuellos@gmail.com \
    --cc=Liviu.Dudau@arm.com \
    --cc=arnd@arndb.de \
    --cc=bhelgaas@google.com \
    --cc=catalin.marinas@arm.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=helgaas@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-staging@lists.linux.dev \
    --cc=matthias.bgg@gmail.com \
    --cc=neil@brown.name \
    --cc=robh@kernel.org \
    --cc=tsbogend@alpha.franken.de \
    --subject='Re: [PATCH v2 2/3] PCI: fix '\''pci_remap_iospace'\'' for architectures with PCI_IOBASE not defined' \
    /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

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).