<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW --- - warn_unused_result noise" href="https://bugs.freedesktop.org/show_bug.cgi?id=58727#c8">Comment # 8</a> on <a class="bz_bug_link bz_status_NEW " title="NEW --- - warn_unused_result noise" href="https://bugs.freedesktop.org/show_bug.cgi?id=58727">bug 58727</a> from <a class="email" href="mailto:spitzak@gmail.com" title="Bill Spitzak <spitzak@gmail.com>"> Bill Spitzak</a> <pre>On 05/12/2014 11:28 PM, <a href="mailto:bugzilla-daemon@freedesktop.org">bugzilla-daemon@freedesktop.org</a> wrote: > *<a href="show_bug.cgi?id=58727#c7">Comment # 7</a> <<a class="bz_bug_link bz_status_NEW " title="NEW --- - warn_unused_result noise" href="show_bug.cgi?id=58727#c7">https://bugs.freedesktop.org/show_bug.cgi?id=58727#c7</a>> on > <a class="bz_bug_link bz_status_NEW " title="NEW --- - warn_unused_result noise" href="show_bug.cgi?id=58727">bug 58727</a> <<a class="bz_bug_link bz_status_NEW " title="NEW --- - warn_unused_result noise" href="show_bug.cgi?id=58727">https://bugs.freedesktop.org/show_bug.cgi?id=58727</a>> from Ravi > Nanjundappa <<a href="mailto:nravi.n@samsung.com">mailto:nravi.n@samsung.com</a>> * > > (In reply tocomment #6 <show_bug.cgi?id=58727#c6>) >> I think the complaint is about strtol(), which is being used apparently >> to find the other end of the numeric value (returned in the char** arg) >> while the number is ignored, thus causing this error. >> >> It does seem that nobody is going to "accidentally" ignore the result of >> strtol. The warning really should be on functions returning things you >> *must* do something with, for instance malloc() where if you don't free >> the returned value you have a memory leak. >> >> And they really should restore the (void)fn() work-around, which is >> readable. > > In some of the cases (for ex: _cairo_polygon_add_external_edge()), the return > values of the callee are used by the caller (for ex: ln. no. 63 and 177 in > cairo-path-fill.c file), while in other cases it is ignored for some reasons. > The warnings are thrown in these scenarios. In such a case, I fell like we can > use the approach provided by Sergey Shandar, @ > <a href="http://stackoverflow.com/questions/3614691/casting-to-void-doesnt-remove-warn-unused-result-error">http://stackoverflow.com/questions/3614691/casting-to-void-doesnt-remove-warn-unused-result-error</a> > > Bill, > Could you please share your opinion on this? My tests just confirm that the ridiculous workaround described in the link (of using a temporary and casting *that* to void) is required by gcc, instead of the more sensible workaround of casting the function call to void. I have no opinion on whether removing the warn-unused-result attribute from that function, or adding the workaround, or figuring out if the code should in fact should be paying attention to the result.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>