On 04/29/2011 05:18 PM, Pádraig Brady wrote:
>>> I think the case for clearing the bits is a little
>>> stronger than the one for leaving them uninitialized, and would
>>> be even more in favor, it if only this initialization were portable:
>>>
>>>   struct sigaction action = {0,};

What would make it non-portable?  And should we raise a defect against
POSIX that requests that all types that allow extension fields should be
initializable via {0,} as a way to guarantee values in all extension fields?

> Proposed patch attached.
> 
> Subject: [PATCH] manywarnings: add -Wno-missing-field-initializers if needed
> 
> * m4/manywarnings.m4 (gl_MANYWARN_ALL_GCC): Add the above
> option if it's needed to allow initialization with { 0, },
> which is the case with GCC before version 4.7
> ---
>  ChangeLog          |    6 ++++++
>  m4/manywarnings.m4 |   51 ++++++++++++++++++++++++++++++++++++++++++++++++++-
>  2 files changed, 56 insertions(+), 1 deletions(-)

Looks nice from my reading of it.  Although for the cache variable
names, should we go with gl_cv_cc_nomfi_supported rather than
gl_cv_nomfi_supported (that is, add cc_ to each of the cache variable
names), to make it clear that the cache variables are all related to CC
characteristics?  (Compare how we use cv_func, cv_decl, and other such
namespace designators, and cv_nomfi isn't really a namespace).

-- 
Eric Blake   [email protected]    +1-801-349-2682
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to