Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1471604 > unrolled thread

Re: [PATCH v3 0/8] usb: ulpi: remove "dev" field from struct ulpi_ops

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v3 0/8] usb: ulpi: remove "dev" field from struct ulpi_ops Felipe Balbi <balbi@kernel.org> - 2016-08-29 10:00 +0200
    Re: [PATCH v3 0/8] usb: ulpi: remove "dev" field from struct ulpi_ops Greg KH <gregkh@linuxfoundation.org> - 2016-08-29 13:20 +0200

#1471604 — Re: [PATCH v3 0/8] usb: ulpi: remove "dev" field from struct ulpi_ops

[Multipart message — attachments visible in raw view] — view raw

Hi folks,

Tal Shorer <tal.shorer@gmail.com> writes:
> struct ulpi_ops is defined as follows:
>
> struct ulpi_ops {
>         struct device *dev;
>         int (*read)(struct ulpi_ops *ops, u8 addr);
>         int (*write)(struct ulpi_ops *ops, u8 addr, u8 val);
> };
>
> Upon calling ulpi_register_interface(), the struct device argument is
> put inside the struct ulpi_ops argument's dev field. Later, when
> calling the actual read()/write() operations, the struct ulpi_ops is
> passed to them and they use the stored device to access whatever
> private data they need.
>
> This means that if one wishes to reuse the same oprations for multiple
> interfaces (e.g if we have multiple instances of the same controller),
> any but the last interface registered will not operate properly (and
> the one that does work will be at the mercy of the others to not mess
> it up).
>
> I understand that barely any driver uses this bus right now, but I
> suppose it's there to be used at some point. We might as well fix the
> design here before we hit this bug.
>
> This series fixes this by passing the given struct device directly to
> the operation functions via ulpi->dev.parent in ulpi_read() and
> ulpi_write(). It also changes the operations struct to be constant
> since now nobody has a reason to modify it.
>
> Changes from v1:
>  * Split the actual api change into multiple patch as per Felipe Balbi's
>    suggestion. The series now first adds the new api, then migrates
>    everything to use and only then removes the old api.
>
> Changes from v2:
>  * Merge patches 2 and 3 (now patch 2)
>  * Merge patches 5 and 6 (now patch 4)
>  * Remove comment documenting the removed dev field in struct ulpi_ops
>
> Tal Shorer (8):
>   usb: ulpi: move setting of ulpi->dev parent up in ulpi_register()
>   usb: ulpi: add new api functions, {read|write}_dev()
>   usb: dwc3: ulpi: use new api
>   usb: ulpi: remove calls to old api callbacks
>   usb: ulpi: rename operations {read|write}_dev to simply {read|write}
>   usb: ulpi: remove "dev" field from struct ulpi_ops
>   usb: ulpi: make ops struct constant
>   usb: dwc3: ulpi: make dwc3_ulpi_ops constant

I have no idea how to apply these :-) Many of the patches touch dwc3,
but some touch common ulpi. Greg, is it okay if I take the entire series
and send it to you during my for-next pull request?

-- 
balbi

[toc] | [next] | [standalone]

#1471738

On Mon, Aug 29, 2016 at 10:58:25AM +0300, Felipe Balbi wrote:
> 
> Hi folks,
> 
> Tal Shorer <tal.shorer@gmail.com> writes:
> > struct ulpi_ops is defined as follows:
> >
> > struct ulpi_ops {
> >         struct device *dev;
> >         int (*read)(struct ulpi_ops *ops, u8 addr);
> >         int (*write)(struct ulpi_ops *ops, u8 addr, u8 val);
> > };
> >
> > Upon calling ulpi_register_interface(), the struct device argument is
> > put inside the struct ulpi_ops argument's dev field. Later, when
> > calling the actual read()/write() operations, the struct ulpi_ops is
> > passed to them and they use the stored device to access whatever
> > private data they need.
> >
> > This means that if one wishes to reuse the same oprations for multiple
> > interfaces (e.g if we have multiple instances of the same controller),
> > any but the last interface registered will not operate properly (and
> > the one that does work will be at the mercy of the others to not mess
> > it up).
> >
> > I understand that barely any driver uses this bus right now, but I
> > suppose it's there to be used at some point. We might as well fix the
> > design here before we hit this bug.
> >
> > This series fixes this by passing the given struct device directly to
> > the operation functions via ulpi->dev.parent in ulpi_read() and
> > ulpi_write(). It also changes the operations struct to be constant
> > since now nobody has a reason to modify it.
> >
> > Changes from v1:
> >  * Split the actual api change into multiple patch as per Felipe Balbi's
> >    suggestion. The series now first adds the new api, then migrates
> >    everything to use and only then removes the old api.
> >
> > Changes from v2:
> >  * Merge patches 2 and 3 (now patch 2)
> >  * Merge patches 5 and 6 (now patch 4)
> >  * Remove comment documenting the removed dev field in struct ulpi_ops
> >
> > Tal Shorer (8):
> >   usb: ulpi: move setting of ulpi->dev parent up in ulpi_register()
> >   usb: ulpi: add new api functions, {read|write}_dev()
> >   usb: dwc3: ulpi: use new api
> >   usb: ulpi: remove calls to old api callbacks
> >   usb: ulpi: rename operations {read|write}_dev to simply {read|write}
> >   usb: ulpi: remove "dev" field from struct ulpi_ops
> >   usb: ulpi: make ops struct constant
> >   usb: dwc3: ulpi: make dwc3_ulpi_ops constant
> 
> I have no idea how to apply these :-) Many of the patches touch dwc3,
> but some touch common ulpi. Greg, is it okay if I take the entire series
> and send it to you during my for-next pull request?

Sure, that works for me.

greg k-h

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web