RE: [PATCH v2] usb: gadget: update DECLARE_USB_FUNCTION(_INIT) macro

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> From: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> Sent: Monday, May 9, 2022 2:43 PM
> To: Linyu Yuan (QUIC) <quic_linyyuan@xxxxxxxxxxx>
> Cc: linux-usb@xxxxxxxxxxxxxxx; Jack Pham (QUIC) <quic_jackp@xxxxxxxxxxx>
> Subject: Re: [PATCH v2] usb: gadget: update
> DECLARE_USB_FUNCTION(_INIT) macro
> 
> On Mon, May 09, 2022 at 02:28:05PM +0800, Linyu Yuan wrote:
> > Take DECLARE_USB_FUNCTION_INIT(ffs, ffs_alloc_inst, ffs_alloc) as
> example,
> > it will generate function ffsmod_init/ffsmod_exit()
> > and variable ffsusb_func.
> 
> I do not think "as example" is needed here, right?
> 
> > Add possible character '_' in the macro which will generate
> > function/variable name in common format, ffs_mod_init/ffs_mod_exit()
> > and ffs_usb_func.
> 
> Ok, but why do this?  Why not add any other character?  What problem
> does this solve?
> 
> >
> > It will apply to all gadget functions which use this macro.
> 
> That is a given for any macro, and you do nto needs to state this.
Got it, thanks
> 
> >
> > Also do minor change accordingly to f_loopback.c and f_sourcesink.c.
> 
> Why "also"?  What minor change are you making and why?
> 

Will explain the detail change next version.

> When you have "also" in a changelog text, that's a huge hint it should
> be more than one commit, and I think this should be more than one commit
> (hint, the f_loopback.c change can go first).
> 
> > Signed-off-by: Linyu Yuan <quic_linyyuan@xxxxxxxxxxx>
> > ---
> > v2: fix issue report by kernel test robot <lkp@xxxxxxxxx>
> >
> >  drivers/usb/gadget/function/f_loopback.c   | 12 +-----------
> >  drivers/usb/gadget/function/f_sourcesink.c |  6 +++---
> >  include/linux/usb/composite.h              | 14 +++++++-------
> >  3 files changed, 11 insertions(+), 21 deletions(-)
> 
> As the first version showed, you didn't test-build this so I really do
> not understand why it is needed as you obviously are not using this
> change anywhere..  Why the extra churn for no real advantage?

No, on my local workspace, first I only choose f_fs.c to compile.
Sorry for that.

> 
> thanks,
> 
> greg k-h




[Index of Archives]     [Linux Media]     [Linux Input]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [Old Linux USB Devel Archive]

  Powered by Linux