This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Suspected bug lshift_value (expmed.c) and its patch
- From: Eric Botcazou <ebotcazou at adacore dot com>
- To: "Bingfeng Mei" <bmei at broadcom dot com>
- Cc: gcc-patches at gcc dot gnu dot org, "dave dot korn dot cygwin at googlemail dot com" <dave dot korn dot cygwin at googlemail dot com>
- Date: Thu, 19 Mar 2009 15:13:24 +0100
- Subject: Re: Suspected bug lshift_value (expmed.c) and its patch
- References: <7FB04A5C213E9943A72EE127DB74F0AD49F021D37D@SJEXCHCCR02.corp.ad.broadcom.com>
> For our target, HOST_WIDE_INT = 32.
That's a confusion between host and target. Porting the compiler really means
defining a new target only, you shouldn't need to port the host. However...
> The trouble is I cannot point out which target except our one (still
> private porting) has the same error, which has HOST_BITS_PER_WIDE_INT = 32
> and BITS_PER_WORD = 64.
... in practice 64-bit targets require a 64-bit hosts nowadays so you'd better
set need_64bit_hwint to yes for your target in config.gcc.
--
Eric Botcazou