This is the mail archive of the
mailing list for the GCC project.
Re: [patch 7/8] Remove basic-block.h from cgraph.h
- From: Jeff Law <law at redhat dot com>
- To: Andrew MacLeod <amacleod at redhat dot com>, gcc-patches <gcc-patches at gcc dot gnu dot org>, Richard Biener <richard dot guenther at gmail dot com>
- Date: Fri, 18 Oct 2013 10:06:07 -0600
- Subject: Re: [patch 7/8] Remove basic-block.h from cgraph.h
- Authentication-results: sourceware.org; auth=none
- References: <5260856C dot 7070008 at redhat dot com> <52613B04 dot 10301 at redhat dot com>
On 10/18/13 07:43, Andrew MacLeod wrote:
I also happened to notice that basic-block.h was being included directly
by cgraph.h. The only thing cgraph.h and most of what use it need are
the typedefs for gcov_type:
typedef HOST_WIDEST_INT gcov_type;
typedef unsigned HOST_WIDEST_INT gcov_type_unsigned;
This patch moves gcov_type and gcov_type_unsigned to coretypes.h.
I will note this is the first time HWINT is introduced to coretypes, but
I believe system.h is suppose to always be included first, and that
includes HWINT. we could also include HWINT from coretypes... But Im
not sure what the standard practice for what goes into coretypes
In any case, by removing the basic-block.h include from cgraph.h,
varasm.c was the only .c file that required a new include.
Does this seem reasonable?
bootstraps on x86_64-unknown-linux-gnu with no new regressions. OK?
I'm not sure if we have a hard and fast rule, but I'm certainly OK with
making one that we have config.h, system.h, coretypes.h as the defined