On Wed, 29 Aug 2018 18:07:46 +0200
Markus Armbruster <arm...@redhat.com> wrote:
> Cornelia Huck <coh...@redhat.com> writes:
>
> > Add two functions to print an error/warning report once depending
> > on a passed-in condition variable and flip it if printed. This is
> > useful if you want to print a message not once-globally, but e.g.
> > once-per-device.
> >
> > Inspired by warn_once() in hw/vfio/ccw.c.
>
> This leaves the reader wondering why you don't replace that function.
> You do, in PATCH 3. I'd either announce that here, or squash the two
> patches.
Squashing probably makes the most sense.
>
> >
> > Signed-off-by: Cornelia Huck <coh...@redhat.com>
> > ---
> > include/qemu/error-report.h | 5 +++++
> > util/qemu-error.c | 44
> > ++++++++++++++++++++++++++++++++++++++++++++
> > 2 files changed, 49 insertions(+)
> >
> > diff --git a/include/qemu/error-report.h b/include/qemu/error-report.h
> > index 72fab2b031..d2a6515e68 100644
> > --- a/include/qemu/error-report.h
> > +++ b/include/qemu/error-report.h
> > @@ -44,6 +44,11 @@ void error_report(const char *fmt, ...) GCC_FMT_ATTR(1,
> > 2);
> > void warn_report(const char *fmt, ...) GCC_FMT_ATTR(1, 2);
> > void info_report(const char *fmt, ...) GCC_FMT_ATTR(1, 2);
> >
> > +void error_report_once_cond(bool *printed, const char *fmt, ...)
> > + GCC_FMT_ATTR(2, 3);
> > +void warn_report_once_cond(bool *printed, const char *fmt, ...)
> > + GCC_FMT_ATTR(2, 3);
> > +
> > /*
> > * Similar to error_report(), except it prints the message just once.
> > * Return true when it prints, false otherwise.
> > diff --git a/util/qemu-error.c b/util/qemu-error.c
> > index a25d3b94c6..0894ab6995 100644
> > --- a/util/qemu-error.c
> > +++ b/util/qemu-error.c
> > @@ -310,3 +310,47 @@ void info_report(const char *fmt, ...)
> > vreport(REPORT_TYPE_INFO, fmt, ap);
> > va_end(ap);
> > }
> > +
> > +/*
> > + * If *printed is false, print an error message to current monitor if we
> > + * have one, else to stderr, and flip *printed to true.
> > + * If printed is NULL, do not print anything.
>
> Any particular reason for supporting null @printed?
>
> For what it's worth, warn_once() supports it, but its caller doesn't use
> it.
I can change this to asserting on NULL @printed. If you call this with
NULL, you probably have messed up somewhere already.
>
> > + * Format arguments like sprintf(). The resulting message should be
> > + * a single phrase, with no newline or trailing punctuation.
> > + * Prepend the current location and append a newline.
> > + * It's wrong to call this in a QMP monitor. Use error_setg() there.
> > + */
> > +void error_report_once_cond(bool *printed, const char *fmt, ...)
> > +{
> > + va_list ap;
> > +
> > + if (!printed || *printed) {
> > + return;
> > + }
> > + *printed = true;
> > + va_start(ap, fmt);
> > + vreport(REPORT_TYPE_ERROR, fmt, ap);
> > + va_end(ap);
> > +}
> > +
> > +/*
> > + * If *printed is false, print a warning message to current monitor if we
> > + * have one, else to stderr, and flip *printed to true.
> > + * If printed is NULL, do not print anything.
> > + * Format arguments like sprintf(). The resulting message should be
> > + * a single phrase, with no newline or trailing punctuation.
> > + * Prepend the current location and append a newline.
> > + * It's wrong to call this in a QMP monitor. Use error_setg() there.
> > + */
> > +void warn_report_once_cond(bool *printed, const char *fmt, ...)
> > +{
> > + va_list ap;
> > +
> > + if (!printed || *printed) {
> > + return;
> > + }
> > + *printed = true;
> > + va_start(ap, fmt);
> > + vreport(REPORT_TYPE_WARNING, fmt, ap);
> > + va_end(ap);
> > +}
