Re: [PATCH v5] usb-storage: Simplify protocol translation

[ShiHao <i.shihao.999@gmail.com>]

On Wed, Oct 08, 2025 at 09:31:59AM +0200, Greg KH wrote:
> On Wed, Oct 08, 2025 at 12:34:20PM +0530, Shi Hao wrote:
> > Simplify protocol translation for usb subclasses.
> > 
> > As suggested by Alan Stern  to remove the switch case labels  with
> > 'default'  or keep the  previous  if statements  instead of the switch
> > cases and advised  reasonable value when cmnd[0] >= 0xe0 .
> > 
> > keep  those usb subclasses to default and simplify logic removing
> > switch cases and set 6 as a reasonable value when cmnd[0] >= 0xe0
> > which fallbacks  to value 6 which is old cmd_len for those subclasses.
> > 
> > Signed-off-by: Shi Hao <i.shihao.999@gmail.com>
> > 
> > ---
> > 
> > changes v5:
> > - Reformulate as  v5  of original patch series
> > - No code changes ,clarified versioning history
> > 
> > changes v4:
> > - Simplify  protocol translation logic
> > - Set reasonable default when cmnd[0] >= 0xE0
> > - Remove switch-case statement
> > 
> > changes v3:
> > - Wrap  commit message lines properly
> > - Improve commit description
> > - fix  checkpatch.pl script errors
> > 
> > changes v2:
> > - Update subject commit message
> > - fix  checkpatch.pl warnings
> > - match from: and sign-off: name
> > ---
> >  drivers/usb/storage/transport.c | 21 +++++++--------------
> >  1 file changed, 7 insertions(+), 14 deletions(-)
> > 
> > diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
> > index 4d01f70f39ac..14cc608052d9 100644
> > --- a/drivers/usb/storage/transport.c
> > +++ b/drivers/usb/storage/transport.c
> > @@ -719,17 +719,9 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
> >  		scsi_eh_prep_cmnd(srb, &ses, NULL, 0, sense_size);
> >  
> >  		/* Protocol translation per scsi opcode group */
> > -		switch (us->subclass) {
> > -		case USB_SC_UFI:
> > -		case USB_SC_8020:
> > -		case USB_SC_8070:
> > -		case USB_SC_QIC:
> > -			srb->cmd_len = 12;
> > -			break;
> > -		/* Determine cmd_len based on scsi opcode group */
> > -		case USB_SC_RBC:
> > -		case USB_SC_SCSI:
> > -		case USB_SC_CYP_ATACB:
> > +		if (us->subclass == USB_SC_RBC || us->subclass == USB_SC_SCSI ||
> > +				us->subclass == USB_SC_CYP_ATACB) {
> > +		/* Determine cmd_len based on SCSI opcode group */
> 
> Always remember to run scripts/checkpatch.pl before sending changes out
> to others :(
> 
> thanks,
> 
> greg k-h


Hi greg ,

Thanks for the reminder. I ran scripts/checkpatch.pl --strict on
drivers/usb/storage/transport.c and all checkpatch.pl has been
resolved .If i need to make another patch for this new change
please let me know .Once again thanks Alan stern and greg k-h
for reviewing and giving your precious time .

Thanks shihao