From: Mason <slash.tmp@free.fr>
To: linux-pci <linux-pci@vger.kernel.org>,
Linux ARM <linux-arm-kernel@lists.infradead.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Arnd Bergmann <arnd@arndb.de>, Rob Herring <robh@kernel.org>,
David Laight <david.laight@aculab.com>,
Phuong Nguyen <phuong_nguyen@sigmadesigns.com>,
Thibaud Cornic <thibaud_cornic@sigmadesigns.com>,
Marc Zyngier <marc.zyngier@arm.com>,
Shawn Lin <shawn.lin@rock-chips.com>,
Bjorn Helgaas <helgaas@kernel.org>
Subject: Re: Neophyte questions about PCIe
Date: Wed, 8 Mar 2017 14:39:09 +0100 [thread overview]
Message-ID: <0cd311b9-b621-432d-f95f-fafc5016bce0@free.fr> (raw)
In-Reply-To: <ced6a4bf-c94a-c116-1e94-52a8d675d132@free.fr>
On 07/03/2017 23:45, Mason wrote:
> 3) What happens if a device requires more than 256 MB of
> mem space? (Is that common? What kind of device? GPUs?)
> Our controller supports a remapping "facility" to add an
> offset to the bus address. Is such a feature supported
> by Linux at all? The problem is that this creates
> another race condition, as setting the offset register
> before an access may occur concurrently on two cores.
> Perhaps 256 MB is plenty on a 32-bit embedded device?
I was told that Linux does not support this kind of "dynamic remapping",
because access to PCI memory region is not handled through a call-back;
the driver just calls readl/writel directly on the pointer.
"We expect that any device we map into the address space is reachable
through static page table entries set up by ioremap() or pci_iomap()."
On a related subject, I asked about the max size of I/O space.
/**
* pci_remap_iospace - Remap the memory mapped I/O space
* @res: Resource describing the I/O space
* @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.
*/
int __weak pci_remap_iospace(const struct resource *res, phys_addr_t phys_addr)
{
#if defined(PCI_IOBASE) && defined(CONFIG_MMU)
unsigned long vaddr = (unsigned long)PCI_IOBASE + res->start;
if (!(res->flags & IORESOURCE_IO))
return -EINVAL;
if (res->end > IO_SPACE_LIMIT)
return -EINVAL;
return ioremap_page_range(vaddr, vaddr + resource_size(res), phys_addr,
pgprot_device(PAGE_KERNEL));
#else
/* this architecture does not have memory mapped I/O space,
so this function should never be called */
WARN_ONCE(1, "This architecture does not support memory mapped I/O\n");
return -ENODEV;
#endif
}
/* PCI fixed i/o mapping */
#define PCI_IO_VIRT_BASE 0xfee00000
#define PCI_IOBASE ((void __iomem *)PCI_IO_VIRT_BASE)
http://lxr.free-electrons.com/source/arch/arm/include/asm/io.h?v=4.9#L188
#ifdef CONFIG_NEED_MACH_IO_H
#include <mach/io.h>
#elif defined(CONFIG_PCI)
#define IO_SPACE_LIMIT ((resource_size_t)0xfffff)
#define __io(a) __typesafe_io(PCI_IO_VIRT_BASE + ((a) & IO_SPACE_LIMIT))
#else
#define __io(a) __typesafe_io((a) & IO_SPACE_LIMIT)
#endif
So the default seems to be 1 MB on arm32. But the platform seems allowed
to define a larger or a smaller space.
Regards.
next prev parent reply other threads:[~2017-03-08 13:39 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-07 22:45 Neophyte questions about PCIe Mason
2017-03-08 13:39 ` Mason [this message]
2017-03-08 13:54 ` David Laight
2017-03-08 14:17 ` Mason
2017-03-08 14:38 ` David Laight
2017-03-09 22:01 ` Jeremy Linton
2017-03-08 15:17 ` Bjorn Helgaas
2017-03-09 23:43 ` Mason
2017-03-10 13:15 ` Robin Murphy
2017-03-10 14:06 ` David Laight
2017-03-10 15:05 ` Mason
2017-03-10 15:14 ` David Laight
2017-03-10 15:33 ` Mason
2017-03-10 15:23 ` Robin Murphy
2017-03-10 15:35 ` David Laight
2017-03-10 16:00 ` Robin Murphy
2017-03-13 10:59 ` Mason
2017-03-13 11:56 ` Robin Murphy
2017-03-10 18:49 ` Bjorn Helgaas
2017-03-10 14:53 ` Mason
2017-03-10 16:45 ` Mason
2017-03-10 17:49 ` Mason
2017-03-11 10:57 ` Mason
2017-03-13 21:40 ` Bjorn Helgaas
2017-03-13 21:57 ` Mason
2017-03-13 22:46 ` Bjorn Helgaas
2017-03-14 10:23 ` David Laight
2017-03-14 12:05 ` Mason
2017-03-14 12:24 ` David Laight
2017-03-13 14:25 ` Mason
2017-03-14 14:00 ` Mason
2017-03-14 15:54 ` Mason
2017-03-14 21:46 ` 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=0cd311b9-b621-432d-f95f-fafc5016bce0@free.fr \
--to=slash.tmp@free.fr \
--cc=ard.biesheuvel@linaro.org \
--cc=arnd@arndb.de \
--cc=david.laight@aculab.com \
--cc=helgaas@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-pci@vger.kernel.org \
--cc=marc.zyngier@arm.com \
--cc=phuong_nguyen@sigmadesigns.com \
--cc=robh@kernel.org \
--cc=shawn.lin@rock-chips.com \
--cc=thibaud_cornic@sigmadesigns.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).