From: Joel Sherrill <joel.sherrill@oarcorp.com>
To: Craig Howland <howland@LGSInnovations.com>,
"newlib@sourceware.org" <newlib@sourceware.org>
Subject: Re: Importing cacoshl.c and its dependent methos,
Date: Wed, 21 Jun 2017 15:09:00 -0000 [thread overview]
Message-ID: <356dfbd5-08e0-e8e9-6c7e-b9bb8c2b0cef@oarcorp.com> (raw)
In-Reply-To: <cf62efd3-73cb-ef54-659e-81c2e1b24d5b@LGSInnovations.com>
On 6/21/2017 10:02 AM, Craig Howland wrote:
> On 06/21/2017 08:04 AM, aditya upadhyay wrote:
>> So ,To not rely on stdbool.h from c99, do i have to port stdbool.h here also ?
>> Otherwise we can use
>> #ifndef __cplusplus
>> typedef enum { false, true } bool;
>> #endif
>>
>> Thanks a lot for your suggested modification. I am applying the
>> required the modification.
>>
>> Regards,
>> Aditya Upadhyay
>>
>> On Wed, Jun 21, 2017 at 1:58 PM, Corinna Vinschen <vinschen@redhat.com> wrote:
>>> On Jun 21 08:46, Sebastian Huber wrote:
>>>> On 21/06/17 04:24, aditya upadhyay wrote:
>>>>
>>>>> Hello Developers,
>>>>> In cacoshl.c, There is a dependency between cacoshl.c and csqrtl.c.
>>>>> In ported csqrtl.c from NetBSD, stdbool.h is there, But i saw there
>>>>> is no stdbool.h.
>>>>>
>>>>> So for bool data type, I have used enum. Is it a right way or not.
>>>>> please suggest or do i have to use stdbool.h from gcc library ?
>>>> <stdbool.h> is currently not used by Newlib. I am not sure if we should add
>>>> this dependency if its only needed by csqrtl.c (in overall NetBSD libm) and
>>>> not visible to the user.
>>>>
>>>> I would replace the #include <stdbool.h> with
>>>>
>>>> #ifndef __cplusplus
>>>> typedef enum { false, true } bool;
>>>> #endif
>>>>
>>>> and keep the rest of the code unchanged.
>>> Wait. Isn't stdbool.h available in gcc for ages? When building newlib,
>>> shouldn't we be able to rely on c99 being available?
>>>
>>>
>>> Corinna
>>>
> Yes, we can, and should. That's because complex.h and stdbool.h were added in
> C99, making C99 required for these functions. I did a quick check on some older
> cross-compiler installs that I have, and GCC 4.1.1 has stdbool.h (and 4.6.4 has
> complex.h).
Just to clarify, I did a test and added include <stdbool.h> to printf.c.
It worked and picked up the gcc version. Can we rely on this being present
or does newlib need to have its own?
I want to make sure the direction is quite clear. :)
--joel
> Craig
>
next prev parent reply other threads:[~2017-06-21 15:09 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-21 2:25 aditya upadhyay
2017-06-21 6:46 ` Sebastian Huber
2017-06-21 8:28 ` Corinna Vinschen
2017-06-21 12:05 ` aditya upadhyay
2017-06-21 15:02 ` Craig Howland
2017-06-21 15:09 ` Joel Sherrill [this message]
2017-06-21 15:18 ` Craig Howland
2017-06-26 6:05 ` Sebastian Huber
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=356dfbd5-08e0-e8e9-6c7e-b9bb8c2b0cef@oarcorp.com \
--to=joel.sherrill@oarcorp.com \
--cc=howland@LGSInnovations.com \
--cc=newlib@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).