Am Donnerstag, 27. Januar 2022, 17:16:35 CET schrieb John David Anglin:
> On 2022-01-27 1:58 a.m., Helge Deller wrote:
> >>> diff --git a/drivers/parisc/sba_iommu.c b/drivers/parisc/sba_iommu.c
> >>> index e60690d38d67..374b9199878d 100644
> >>> --- a/drivers/parisc/sba_iommu.c
> >>> +++ b/drivers/parisc/sba_iommu.c
> >>> @@ -1047,7 +1047,7 @@ sba_unmap_sg(struct device *dev, struct
> >>> scatterlist
> >>> *sglist, int nents, spin_unlock_irqrestore(&ioc->res_lock, flags);
> >>>
> >>> #endif
> >>>
> >>> - while (sg_dma_len(sglist) && nents--) {
> >>> + while (nents && sg_dma_len(sglist)) {
> >>
> >> What about:
> >> for (; nents && sg_dma_len(sglist); nents--) {
> >
> > The way how Dave wrote it is more clean, IMHO.
>
> I'm going to leave the change to sba_iommu.c as proposed. While i'm sure
> the suggested for statement would be fine, I looked at how gcc handled the
> while loop. It is quite subtle. Except for an initial test and decrement,
> the iteration of nents is replaced by a calculation of the the final value
> for sglist.
>
> Regarding the newline, the file has several places where newlines precede
> #ifdef statements. I think the current style is okay and checkpatch.pl
> doesn't object to that format.
It was not about the #ifdef, it was the line immediately following the opening
brace of the loop.
EikeAttachment:
signature.asc
Description: This is a digitally signed message part.
