[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: pkg/47941: devel/glib2 build failure on NetBSD/arm 6.1
The following reply was made to PR pkg/47941; it has been noted by GNATS.
From: Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>
Cc: joerg%britannica.bec.de@localhost, tsutsui%ceres.dti.ne.jp@localhost
Subject: Re: pkg/47941: devel/glib2 build failure on NetBSD/arm 6.1
Date: Thu, 20 Jun 2013 01:58:18 +0900
> > (1)inverted configure test for __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4
> Looks good.
> > (2)incorrect configure test for __sync_bool_compare_and_swap()
> This too.
> > (3)prototype mismatches of g_atomic functions in !G_ATOMIC_LOCK_FREE case
> This I haven't checked, but I assume you have tested.
During debug, first I fixed the problem (1) then got the following error:
>> gatomic.c:669:2: error: #error G_ATOMIC_LOCK_FREE defined, but incapable of
>> lock-free atomics.
due to the problem (2):
Next, I tracked and fixed the problem (2), I still got a error:
>> gatomic.c:686:2: error: conflicting types for 'g_atomic_int_get'
>> gatomic.h:34:25: note: previous declaration of 'g_atomic_int_get' was here
caused by the problem (3).
Fixes of the type of args caused another warning:
>> gatomic.c:803:28: warning: initialization discards qualifiers from pointer
>> target type
so patch-ce for gatomic.c also has the following lines:
>> +- volatile gpointer *ptr = atomic;
>> ++ const volatile gpointer *ptr = atomic;
Not sure why the previous i386 build (with the inverted test) didn't use
these glib internal g_atomic functions though.
> Please also bump
> the revision.
This is not a leaf but quite basic package so I hope proper
responsible persons take this problem.
(I guess the problem (1) (inverted test) is serious even for upstream)
Main Index |
Thread Index |