From: Thomas Monjalon <thomas.monjalon@6wind.com>
To: Santosh Shukla <sshukla@mvista.com>
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH] eal: map io resources for non x86 architectures
Date: Thu, 17 Dec 2015 11:14:38 +0100 [thread overview]
Message-ID: <2241331.HNmyzf8foi@xps13> (raw)
In-Reply-To: <CAAyOgsboGTcQ4CyksMpA400KkWDQSDbU490E9yQan4HMCzsxFg@mail.gmail.com>
Hi,
2015-12-17 15:37, Santosh Shukla:
> On Thu, Dec 17, 2015 at 3:32 PM, Santosh Shukla <sshukla@mvista.com> wrote:
> > On Thu, Dec 17, 2015 at 3:31 PM, Santosh Shukla <sshukla@mvista.com> wrote:
> >> On Thu, Dec 17, 2015 at 3:08 PM, Yuanhan Liu
> >> <yuanhan.liu@linux.intel.com> wrote:
> >>> On Wed, Dec 16, 2015 at 07:21:55PM +0530, Santosh Shukla wrote:
> >>>> On Wed, Dec 16, 2015 at 6:18 PM, Yuanhan Liu
> >>>> <yuanhan.liu@linux.intel.com> wrote:
> >>>> > On Wed, Dec 16, 2015 at 01:31:04PM +0100, David Marchand wrote:
> >>>> >> x86 requires a special set of instructions to access ioports, but other
> >>>> >> architectures let you remap io resources.
> >>>> >> So let eal remap io resources by accepting IORESOURCE_IO flag for
> >>>> >> architectures other than x86.
> >>>> >
> >>>> > One question: this patch could be a replacement of the igbuio_iomap patch
> >>>> > from Santosh? If so, I like it: It's more elegant.
> >>>> >
> >>>> > --yliu
> >>>> >
> >>>>
> >>>> I did tried similar in past but not in parse_sysfs (such that
> >>>> mem.resource_addr to accept IO_RESOURCE_IO types) and observed that
> >>>> pci_map_resource not able to map address hence segfault at tespmd
> >>>> initialization.
> >>>>
> >>>> i was getting these:
> >>>> EAL: pci_map_resource(): cannot mmap(19, 0x7fa5c00000, 0x20, 0x0):
> >>>> Invalid argument (0xffffffffffffffff)
> >>>
> >>> That's because ARM (at least the kernel) doesn't allow an IO map:
> >>>
> >>> arch/arm/kernel/bios32.c
> >>> ------------------------
> >>> 618 int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
> >>> 619 enum pci_mmap_state mmap_state, int write_combine)
> >>> 620 {
> >>> 621 if (mmap_state == pci_mmap_io)
> >>> 622 return -EINVAL;
> >>>
> >>> And with a quick glimpse of powerpc, I see no such limitation. Hence,
> >>> this peice of code may work only on Powerpc platform (and maybe a few
> >>> others we don't care).
> >>>
> >>> So, apparently, this will not work for ARM.
> >>>
> >>
> >> Right and I did shared detailed explanation on why it wont work on
> >> this link [1], infact this patch shouldn;t work for mips too.
> >>
> >> As I mentioned earlier I did tried similar approach and so to get
> >> everything working like iomem is currently in dpdk; we need to add
> >> something like pci_remap_iospace --> ioremap_page_range() but this api
> >> not really pci_mmap_page_range types. user need to write more code on
> >> top so to use this api efficiently, also this api looks like meant to
> >> use by arch file only in kernel space.
> >>
> >>
> > missed link;
> >
> > [1] http://dpdk.org/dev/patchwork/patch/9365/
> >
>
> IMO, it is worth keeping one special device file who could work across
> archs like arm/arm64/powerpc and others, who could map iopci bar to
> dpdk user-space. also this approach has no kernel version dependency
> too. BTW; I did mentioned in second approach in to add /dev/ioport
> interface in drivers/char/mem.c which could read more than byte in one
> single operation, but that has kernel dependency. However that
> approach too is arch agnostic.
Your first approach use an out-of-tree kernel module (igb_uio), so we cannot
really say there is no kernel dependency.
We should try to remove the need for any out-of-tree kernel module.
That's why the Linux upstream approach is a better solution.
next prev parent reply other threads:[~2015-12-17 10:15 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-14 13:00 [dpdk-dev] [ [PATCH v2] 00/13] Add virtio support in arm/arm64 Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 01/13] virtio: Introduce config RTE_VIRTIO_INC_VECTOR Santosh Shukla
2015-12-17 12:02 ` Santosh Shukla
2015-12-17 12:03 ` Thomas Monjalon
2015-12-17 12:18 ` Santosh Shukla
2015-12-17 23:24 ` Stephen Hemminger
2015-12-18 1:31 ` Yuanhan Liu
2015-12-18 9:52 ` Xie, Huawei
2015-12-18 10:41 ` Thomas Monjalon
2015-12-18 17:33 ` Stephen Hemminger
2015-12-18 18:11 ` Thomas Monjalon
2015-12-18 12:46 ` Santosh Shukla
2015-12-22 6:26 ` Yuanhan Liu
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 02/13] config: i686: set RTE_VIRTIO_INC_VECTOR=n Santosh Shukla
2015-12-17 12:03 ` Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 03/13] rte_io: armv7/v8: Introduce api to emulate x86-style of PCI/ISA ioport access Santosh Shukla
2015-12-14 14:25 ` Jerin Jacob
2015-12-14 16:29 ` Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 04/13] virtio_pci: use rte_io.h for non-x86 arch Santosh Shukla
2015-12-14 14:28 ` Jerin Jacob
2015-12-14 15:29 ` Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 05/13] virtio: change io_base datatype from uint32_t to uint64_type Santosh Shukla
2015-12-16 13:48 ` Yuanhan Liu
2015-12-16 14:01 ` Santosh Shukla
2015-12-16 14:23 ` Yuanhan Liu
2015-12-16 14:39 ` Santosh Shukla
2015-12-16 14:58 ` Yuanhan Liu
2015-12-16 15:05 ` Santosh Shukla
2015-12-17 7:19 ` Yuanhan Liu
2015-12-17 8:17 ` Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 06/13] config: armv7/v8: Enable RTE_LIBRTE_VIRTIO_PMD Santosh Shukla
2015-12-14 14:31 ` Jerin Jacob
2015-12-14 16:16 ` Santosh Shukla
2015-12-15 5:36 ` Jianbo Liu
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 07/13] linuxapp: eal: arm: Always return 0 for rte_eal_iopl_init() Santosh Shukla
2015-12-14 14:34 ` Jan Viktorin
2015-12-14 15:04 ` Santosh Shukla
2015-12-14 14:37 ` Jerin Jacob
2015-12-14 15:24 ` Santosh Shukla
2015-12-14 15:56 ` Jerin Jacob
2015-12-14 16:13 ` Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 08/13] rte_io: x86: Remove sys/io.h ifdef x86 clutter Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 09/13] igb_uio: ioport: map iopci region for armv7/v8 Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 10/13] include/exec-env: ioport: add rte_virt_ioport header file Santosh Shukla
2015-12-14 14:43 ` Jerin Jacob
2015-12-14 16:17 ` Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 11/13] virtio_ioport: armv7/v8: mmap virtio iopci bar region Santosh Shukla
2015-12-16 13:29 ` Yuanhan Liu
2015-12-16 14:20 ` Santosh Shukla
2015-12-16 14:37 ` Yuanhan Liu
2015-12-16 14:40 ` Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 12/13] virtio_ethdev: use virtio_ioport api at device init/close Santosh Shukla
2015-12-14 13:00 ` [dpdk-dev] [ [PATCH v2] 13/13] virtio_ethdev : fix format specifier error for 64bit addr case Santosh Shukla
2015-12-14 14:31 ` [dpdk-dev] [ [PATCH v2] 00/13] Add virtio support in arm/arm64 Jan Viktorin
2015-12-14 16:09 ` Santosh Shukla
2015-12-16 7:48 ` Santosh Shukla
2015-12-16 8:47 ` David Marchand
2015-12-16 11:43 ` Santosh Shukla
2015-12-16 12:31 ` [dpdk-dev] [PATCH] eal: map io resources for non x86 architectures David Marchand
2015-12-16 12:48 ` Yuanhan Liu
2015-12-16 13:34 ` David Marchand
2015-12-16 13:42 ` Yuanhan Liu
2015-12-16 13:51 ` Santosh Shukla
2015-12-17 9:38 ` Yuanhan Liu
2015-12-17 10:01 ` Santosh Shukla
2015-12-17 10:02 ` Santosh Shukla
2015-12-17 10:07 ` Santosh Shukla
2015-12-17 10:14 ` Thomas Monjalon [this message]
2015-12-17 10:21 ` Santosh Shukla
2015-12-17 10:33 ` Thomas Monjalon
2015-12-17 11:22 ` Santosh Shukla
2015-12-18 5:30 ` Yuanhan Liu
2015-12-18 6:34 ` Jerin Jacob
2015-12-18 7:55 ` Yuanhan Liu
2015-12-18 9:37 ` Thomas Monjalon
2015-12-18 7:54 ` Santosh Shukla
2015-12-18 8:21 ` Yuanhan Liu
2015-12-18 12:55 ` Santosh Shukla
2015-12-29 5:56 ` Santosh Shukla
2015-12-29 9:56 ` Burakov, Anatoly
2015-12-29 10:47 ` Santosh Shukla
2015-12-29 11:06 ` Burakov, Anatoly
2015-12-29 12:23 ` Santosh Shukla
2015-12-29 14:04 ` Alex Williamson
2015-12-29 14:51 ` Santosh Shukla
2015-12-31 14:27 ` Santosh Shukla
2015-12-16 13:15 ` Bruce Richardson
2015-12-16 13:29 ` David Marchand
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=2241331.HNmyzf8foi@xps13 \
--to=thomas.monjalon@6wind.com \
--cc=dev@dpdk.org \
--cc=sshukla@mvista.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).