> -----Original Message-----
> From: Peter Maydell <peter.mayd...@linaro.org>
> Sent: Friday, May 20, 2022 8:21 AM
> To: Brian Cain <bc...@quicinc.com>
> Cc: Daniel P. Berrangé <berra...@redhat.com>; qemu-devel@nongnu.org;
> Alex Bennée <alex.ben...@linaro.org>
> Subject: Re: [PATCH] tests: Remove unused "wcount" from linux-test.c
>
> WARNING: This email originated from outside of Qualcomm. Please be wary of
> any links or attachments, and do not enable macros.
>
> On Fri, 20 May 2022 at 14:16, Brian Cain <bc...@quicinc.com> wrote:
> >
> > clang reports this variable as 'set but not used'.
> >
> > Signed-off-by: Brian Cain <bc...@quicinc.com>
> > ---
> > tests/tcg/multiarch/linux/linux-test.c | 4 +---
> > 1 file changed, 1 insertion(+), 3 deletions(-)
> >
> > diff --git a/tests/tcg/multiarch/linux/linux-test.c
> b/tests/tcg/multiarch/linux/linux-test.c
> > index 019d8175ca..d5781bb476 100644
> > --- a/tests/tcg/multiarch/linux/linux-test.c
> > +++ b/tests/tcg/multiarch/linux/linux-test.c
> > @@ -332,12 +332,11 @@ static void test_pipe(void)
> > fd_set rfds, wfds;
> > int fds[2], fd_max, ret;
> > uint8_t ch;
> > - int wcount, rcount;
> > + int rcount;
> >
> > chk_error(pipe(fds));
> > chk_error(fcntl(fds[0], F_SETFL, O_NONBLOCK));
> > chk_error(fcntl(fds[1], F_SETFL, O_NONBLOCK));
> > - wcount = 0;
> > rcount = 0;
> > for(;;) {
> > FD_ZERO(&rfds);
> > @@ -360,7 +359,6 @@ static void test_pipe(void)
> > if (FD_ISSET(fds[1], &wfds)) {
> > ch = 'a';
> > chk_error(write(fds[1], &ch, 1));
> > - wcount++;
> > }
> > }
> > }
>
> Another 'count' related oddity in this test code:
>
> if (rcount >= WCOUNT_MAX)
> break;
> why do we compare _r_count against _W_COUNT_MAX ?
I thought it was peculiar myself but *possibly* intentional. Or maybe wcount
should be compared with WCOUNT_MAX and rcount could be omitted?
-Brian
