[Bug math/22650] glibc causes build bustage on std::isfinite()

["aurelien at aurel32 dot net" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=22650

Aurelien Jarno <aurelien at aurel32 dot net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
                 CC|                            |aurelien at aurel32 dot net
         Resolution|---                         |INVALID

--- Comment #1 from Aurelien Jarno <aurelien at aurel32 dot net> ---
(In reply to Mark Straver from comment #0)
> Apparently, glibc 2.26 removed std::isfinite() from <cmath>? This causes
> build bustage as reported in e.g. Pale Moon.[1]

<cmath> is not provided by glibc, but by GCC.

> 15:18.98
> /var/cache/acbs/build/acbs.b1irsl91/palemoon/dom/canvas/
> CanvasRenderingContext2D.cpp: In function ‘bool
> mozilla::dom::ValidateRect(double&, double&, double&, double&)’:
> 15:18.98
> /var/cache/acbs/build/acbs.b1irsl91/palemoon/dom/canvas/
> CanvasRenderingContext2D.cpp:2420:8: error: ‘__builtin_isfinite’ is not a
> member of ‘std’
> 15:18.98    if (!std::isfinite((float)aX) | !std::isfinite((float)aY) |
> 15:18.98         ^
> 
> As far as I know, std::isfinite() should always be included in <cmath>.

This is a wrong assumption, std::isfinite function has been introduced in
C++11. Older standards like C++98 or C++03 only defined the isfinite macro.

-- 
You are receiving this mail because:
You are on the CC list for the bug.