public inbox for linux-acpi@vger.kernel.org
 help / color / mirror / Atom feed
From: Bjorn Helgaas <bjorn.helgaas-VXdhtT5mjnY@public.gmane.org>
To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Cc: Shaohua Li <shaohua.li-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
	Len <len.brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
	Dominik Brodowski <linux-JhLEnvuH02M@public.gmane.org>,
	Adam Belay <ambx1-IBH0VoN/3vPQT0dZR+AlfA@public.gmane.org>
Subject: Re: [RFC]convert ACPI driver model to Linux driver model - takes 2
Date: Thu, 20 Oct 2005 11:04:44 -0600	[thread overview]
Message-ID: <200510201104.44164.bjorn.helgaas@hp.com> (raw)
In-Reply-To: <1129185454.9074.15.camel-ECwVeV2eNyQD0+JXs3kMbRL4W9x8LtSr@public.gmane.org>

On Thursday 13 October 2005 12:37 am, Shaohua Li wrote:
> This is the updated patch for the topic. In this version, I completely
> convert devices in ACPI bus into Linux driver model. Now the device
> driver for ACPI bus device should be a 'struct acpi_phys_driver'. Old
> style driver (struct acpi_driver) now can't work for ACPI bus devices.

    - I don't like names that include "acpi_bus" because there
      really is no "bus".  If PCI, which really DOES have a bus,
      doesn't need it, ACPI surely doesn't.

    - Suspend/resume/probe/remove/shutdown are device-oriented
      interfaces and should be named with "device", not "bus".

    - We should not end up with two interfaces to register ACPI
      drivers.  ACPI is confusing enough as it is.  Having both
      acpi_phys_driver_register() and acpi_bus_register_driver()
      will be a disaster.  (Keeping both temporarily for during
      a transition is OK, but one should be explicitly deprecated.)

> +			length += scnprintf(scratch, buffer_size - length,
> +				"COMPTID=%s", cid_list->id[j].value);

This requires a sequence number (COMPTID%d) to support multiple
_CIDs, doesn't it?

> +static struct bus_type acpi_phys_bus_type = {
> +	.name = "acpi",
> +	.match = acpi_phys_bus_match,
> +	.hotplug = acpi_phys_bus_hotplug,

I think this should be just "acpi_hotplug".

> +	.suspend = acpi_phys_bus_suspend,
> +	.resume = acpi_phys_bus_resume,

And these should be "acpi_device_suspend" and "acpi_device_resume".

> +int acpi_phys_driver_register(struct acpi_phys_driver *drv)

The majority of driver registration interfaces use
"register_driver", not "driver_register".  So I think
this should be "acpi_register_driver", and you should
deprecate "acpi_bus_register_driver".  I still don't
see any fundamental reason for having two ways to
register ACPI drivers.

> +{
> +	int error;
> +	/*FIXME: add to a list */
> +	drv->phys_drv.name = drv->name;
> +	drv->phys_drv.bus = &acpi_phys_bus_type;
> +	drv->phys_drv.probe = acpi_phys_device_probe;
> +	drv->phys_drv.remove = acpi_phys_device_remove;
> +	drv->phys_drv.shutdown = acpi_phys_device_shutdown;

I think these should be "acpi_bus_type", "acpi_device_probe",
"acpi_device_remove", and "acpi_device_shutdown".


-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl

  parent reply	other threads:[~2005-10-20 17:04 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-10-13  6:37 [RFC]convert ACPI driver model to Linux driver model - takes 2 Shaohua Li
     [not found] ` <1129185454.9074.15.camel-ECwVeV2eNyQD0+JXs3kMbRL4W9x8LtSr@public.gmane.org>
2005-10-13 20:43   ` Bjorn Helgaas
     [not found]     ` <200510131443.47844.bjorn.helgaas-VXdhtT5mjnY@public.gmane.org>
2005-10-14  1:13       ` Shaohua Li
     [not found]         ` <1129252436.3948.12.camel-ECwVeV2eNyQD0+JXs3kMbRL4W9x8LtSr@public.gmane.org>
2005-10-14  3:21           ` Bjorn Helgaas
2005-10-20 17:04   ` Bjorn Helgaas [this message]
  -- strict thread matches above, loose matches on Subject: below --
2005-10-13  7:45 Yu, Luming
2005-10-13  8:20 ` Shaohua Li
2005-10-13  8:18 mika.penttila-9Aww8k/80nUxHbG02/KK1g
     [not found] ` <20051013081837.PTGN24578.fep01-app.kolumbus.fi-7B1qVitd7rZYRdgNIgFt/t8DI8KqlI9s@public.gmane.org>
2005-10-13  8:32   ` Shaohua Li
2005-10-14  9:35 Li, Shaohua
2005-10-14 12:47 ` Matthew Wilcox
     [not found]   ` <20051014124749.GD16113-6jwH94ZQLHl74goWV3ctuw@public.gmane.org>
2005-10-18  1:30     ` Shaohua Li
     [not found]       ` <1129599034.3965.6.camel-ECwVeV2eNyQD0+JXs3kMbRL4W9x8LtSr@public.gmane.org>
2005-10-18 17:32         ` Rajesh Shah
2005-10-14 15:54 ` Bjorn Helgaas
     [not found]   ` <200510140954.52206.bjorn.helgaas-VXdhtT5mjnY@public.gmane.org>
2005-10-18  3:06     ` Shaohua Li

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=200510201104.44164.bjorn.helgaas@hp.com \
    --to=bjorn.helgaas-vxdhtt5mjny@public.gmane.org \
    --cc=acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org \
    --cc=ambx1-IBH0VoN/3vPQT0dZR+AlfA@public.gmane.org \
    --cc=len.brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    --cc=linux-JhLEnvuH02M@public.gmane.org \
    --cc=shaohua.li-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    /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