On Wed, Mar 04, 2020 at 07:25:15AM -0800, Christoph Hellwig wrote:
> On Wed, Mar 04, 2020 at 06:22:59AM -0800, Matthew Wilcox wrote:
> > From: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
> >
> > page_offset() confusingly returns the number of bytes from the
> > beginning of the file and not the pgoff, which the tracepoint claims
> > to be returning. We're already returning the number of bytes from the
> > beginning of the file in the 'offset' parameter, so correct the pgoff
> > to be what was apparently intended.
> >
> > Fixes: 0b1b213fcf3a ("xfs: event tracing support")
> > Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
>
> I wonder if tracing the byte offset and just changing the name
> might be more useful. But I agree that we should fix it one way or
> another.
I covered that -- "We're already returning the number of bytes from the
beginning of the file in the 'offset' parameter, so correct the pgoff
to be what was apparently intended."
I mean, we could just delete the pgoff instead. Apparently nobody's
using it, or they would surely have noticed.
