Re: [PATCH] scsi: ufs: ufshcd-pltfrm: Signedness bug in ufshcd_parse_clock_info()

[Dan Carpenter]

On Fri, Aug 16, 2024 at 12:04:04PM +0530, Manivannan Sadhasivam wrote:
> On Fri, Aug 16, 2024 at 12:35:22AM +0300, Dan Carpenter wrote:
> > On Thu, Aug 15, 2024 at 10:47:30AM -0700, Bart Van Assche wrote:
> > > On 8/15/24 4:24 AM, Dan Carpenter wrote:
> > > > The "sz" variable needs to be a signed type for the error handling to
> > > > work as intended.
> > > 
> > > What error handling are you referring to? I haven't found any code that
> > > assigns a negative value to 'sz' in ufshcd_parse_clock_info(). Did I
> > > perhaps overlook something?
> > > 
> > 
> > Rob's patch in linux-next.
> > 
> 
> It would've been helpful if you added 'next' in the patch subject prefix.
> 

I guess that would helped in this case.  But most of the time when I see this
question it's because there are two different upstream maintainers modifying the
same code...  Anyway, sure, I can change my script to add "next" to the subject
when the FIXES_COMMIT isn't in Linus's tree.

if [ "$FIXES_COMMIT" != "" ] ; then
    if ! git merge-base --is-ancestor $FIXES_COMMIT origin/master ; then
        TREE=" next"
    fi
fi

regards,
dan carpenter