mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: John Garry <john.garry@huawei.com>
To: Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	<gregkh@linuxfoundation.org>, <jslaby@suse.com>,
	<p.zabel@pengutronix.de>, <heiko@sntech.de>,
	<ed.blake@sondrel.com>, <jhogan@kernel.org>
Cc: <linux-serial@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	<linuxarm@huawei.com>
Subject: Re: [RFC PATCH 0/2] serial: 8250_dw: IO space + polling mode support
Date: Mon, 26 Feb 2018 11:56:15 +0000	[thread overview]
Message-ID: <b5904a5a-66d8-4385-cc5b-a52529e9d28f@huawei.com> (raw)
In-Reply-To: <1519644517.10722.179.camel@linux.intel.com>

>>>>> Why you can't do properly in ACPI?
>>>
>>> No answer here either.
>>>
>>> Sorry, but with this level of communication it's no go for the
>>> series.
>>>
>>
>> Sorry if my answers did not tell you want you want to know.
>>
>> My point was that the 8250_pnp driver would be used for a pnp_device,
>> but we are creating a platform device for this UART slave so would
>> require a platform device driver, that which 8250_dw.c is. But I will
>> check on pnp device support.
>

Hi Andy,

> Perhaps it's not visible, though below is a description of the drivers
> we have:
>
> 8250_dw - OF/ACPI driver for Synopsys DW (+ DW DMA)
> 8250_lpss - PCI driver for Synopsys DW (+ DW DMA)
> 8250_of -  generic 8250 compatible driver for OF
> 8250_pci - generic 8250 compatible driver for PCI
> 8250_pnp - generic 8250 compatible driver for ACPI
>
> 8250_* (except core parts) - custom glue drivers per some IPs
>
> By description you gave your driver fits 8250_pnp if ACPI tables crafted
> properly.
>
> Share the ACPI excerpt and we can discuss further how to improve them.
>

For a bit of background, MFD support was discussed here initially:
https://lkml.org/lkml/2017/6/13/796

Here is the ACPI table:
Scope(_SB) {
   Device (LPC0) {
     Name (_HID, "HISI0191")  // HiSi LPC
     Name (_CRS, ResourceTemplate () {
       Memory32Fixed (ReadWrite, 0xa01b0000, 0x1000)
     })
   }

   Device (LPC0.CON0) {
     Name (_HID, "HISI1031")
     // Name (_CID, "PNP0501") // cannot support PNP
     Name (LORS, ResourceTemplate() {
       QWordIO (
         ResourceConsumer,
     MinNotFixed,     // _MIF
     MaxNotFixed,     // _MAF
     PosDecode,
     EntireRange,
     0x0,             // _GRA
     0x2F8,           // _MIN
     0x3fff,          // _MAX
     0x0,             // _TRA
     0x08,            // _LEN
     , ,
     IO02


The latest framework changes and host driver patchset are here:
https://lkml.org/lkml/2018/2/19/465

Here is how we probe the children:
static int hisi_lpc_acpi_set_io_res(struct device *child,
                     struct device *hostdev,
                     const struct resource **res,
                     int *num_res)
{

     [ ... In this part we just get the child resources ]


     /* translate the I/O resources */
     for (i = 0; i < count; i++) {
         int ret;

         if (!(resources[i].flags & IORESOURCE_IO))
             continue;
         ret = hisi_lpc_acpi_xlat_io_res(adev, host, &resources[i]);
         if (ret) {
             dev_err(child, "translate IO range failed(%d)\n", ret);
             return ret;
         }
     }
     *res = resources;
     *num_res = count;

     return 0;
}

static int hisi_lpc_acpi_probe(struct device *hostdev)
{
     struct acpi_device *adev = ACPI_COMPANION(hostdev);
     struct hisi_lpc_mfd_cell *hisi_lpc_mfd_cells;
     struct mfd_cell *mfd_cells;
     struct acpi_device *child;
     int size, ret, count = 0, cell_num = 0;

     list_for_each_entry(child, &adev->children, node)
         cell_num++;

     /* allocate the mfd cell and companion acpi info, one per child */
     size = sizeof(*mfd_cells) + sizeof(*hisi_lpc_mfd_cells);
     mfd_cells = devm_kcalloc(hostdev, cell_num, size, GFP_KERNEL);
     if (!mfd_cells)
         return -ENOMEM;

     hisi_lpc_mfd_cells = (struct hisi_lpc_mfd_cell *)
                     &mfd_cells[cell_num];
     /* Only consider the children of the host */
     list_for_each_entry(child, &adev->children, node) {
         struct mfd_cell *mfd_cell = &mfd_cells[count];
         struct hisi_lpc_mfd_cell *hisi_lpc_mfd_cell =
                     &hisi_lpc_mfd_cells[count];
         struct mfd_cell_acpi_match *acpi_match =
                     &hisi_lpc_mfd_cell->acpi_match;
         char *name = hisi_lpc_mfd_cell[count].name;
         char *pnpid = hisi_lpc_mfd_cell[count].pnpid;
         struct mfd_cell_acpi_match match = {
             .pnpid = pnpid,
         };

         snprintf(name, MFD_CHILD_NAME_LEN, MFD_CHILD_NAME_PREFIX"%s",
              acpi_device_hid(child));
         snprintf(pnpid, ACPI_ID_LEN, "%s", acpi_device_hid(child));

         memcpy(acpi_match, &match, sizeof(*acpi_match));
         mfd_cell->name = name;
         mfd_cell->acpi_match = acpi_match;

         ret = hisi_lpc_acpi_set_io_res(&child->dev, &adev->dev,
                            &mfd_cell->resources,
                            &mfd_cell->num_resources);
         if (ret) {
             dev_warn(&child->dev, "set resource fail(%d)\n", ret);
             return ret;
         }
         count++;
     }

     ret = mfd_add_devices(hostdev, PLATFORM_DEVID_NONE,
                   mfd_cells, cell_num, NULL, 0, NULL);
     if (ret) {
         dev_err(hostdev, "failed to add mfd cells (%d)\n", ret);
         return ret;
     }

     return 0;
}

As you know, this is not accepted upstream yet, but I really hope I'm 
close. Hence the RFC tag for the UART patchset.

Please let me know if you require more details.

Thanks,
John



  reply	other threads:[~2018-02-26 11:56 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-22 18:42 John Garry
2018-02-22 18:42 ` [RFC PATCH 1/2] serial: 8250_dw: add IO space support John Garry
2018-02-22 18:42 ` [RFC PATCH 2/2] serial: 8250_dw: support polling mode John Garry
2018-02-23 17:35   ` Andy Shevchenko
2018-02-26 10:54     ` John Garry
2018-02-23 10:30 ` [RFC PATCH 0/2] serial: 8250_dw: IO space + polling mode support Andy Shevchenko
2018-02-23 11:02   ` John Garry
2018-02-23 17:31     ` Andy Shevchenko
2018-02-26  9:33       ` John Garry
2018-02-26  9:53         ` Andy Shevchenko
2018-02-26 10:45           ` John Garry
2018-02-26 11:28             ` Andy Shevchenko
2018-02-26 11:56               ` John Garry [this message]
2018-02-26 12:21                 ` Andy Shevchenko
2018-02-26 12:27                   ` Andy Shevchenko
2018-02-26 13:15                     ` John Garry
2018-02-26 15:02                       ` Andy Shevchenko
2018-02-26 15:07                         ` John Garry
2018-04-12 16:31                           ` John Garry
2018-02-26 12:37                   ` John Garry

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=b5904a5a-66d8-4385-cc5b-a52529e9d28f@huawei.com \
    --to=john.garry@huawei.com \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=ed.blake@sondrel.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=heiko@sntech.de \
    --cc=jhogan@kernel.org \
    --cc=jslaby@suse.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-serial@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=p.zabel@pengutronix.de \
    /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

all inboxes | Powered by JetHome®