Re: [PATCH] perf trace: Remove return value of trace__fprintf_tp_fields

[Benjamin Peterson]

On Thu, Jan 23, 2025 at 3:48 PM Namhyung Kim <namhyung@xxxxxxxxxx> wrote:
>
> On Wed, Jan 22, 2025 at 11:19:05AM -0800, Benjamin Peterson wrote:
> > On Wed, Jan 22, 2025 at 11:08 AM Namhyung Kim <namhyung@xxxxxxxxxx> wrote:
> > >
> > > Hello,
> > >
> > > On Tue, Jan 21, 2025 at 07:01:53PM -0800, Howard Chu wrote:
> > > > Hello fellow maintainers,
> > > >
> > > > Just a slight ping.
> > >
> > > Thanks for the reminder.
> >
> > Thanks for taking a look.
> >
> > >
> > > >
> > > > Thanks,
> > > > Howard
> > > >
> > > > On Fri, Dec 13, 2024 at 10:06 AM Howard Chu <howardchu95@xxxxxxxxx> wrote:
> > > > >
> > > > > Hello Benjamin,
> > > > >
> > > > > On Thu, Dec 12, 2024 at 7:56 PM Benjamin Peterson <benjamin@xxxxxxxxxxx> wrote:
> > > > > >
> > > > > > The return value of this function was meaningless and therefore ignored by
> > > > > > the caller. Remove it.
> > >
> > > In fact, I don't see the reason to merge this, as it's the convention to
> > > return the number of characters it prints.  Yes, it's not used but not
> > > much reason to remove it.
> >
> > The main reason to remove the return value is that it has no useful
> > interpretation and is therefore confusing. To wit, the function
> > returns two times the number of bytes it prints.
>
> Oh, ok.  Then please just fix it to return the correct number.

Alright, I've submitted v2 that simply propagates the return value of fprintf.