This is the mail archive of the
mailing list for the newlib project.
Re: [PATCH libgloss]Using spec files to support two version of newlib library in one tool-chain release
- From: Joel Sherrill <joel dot sherrill at oarcorp dot com>
- To: Steve Ellcey <sellcey at mips dot com>, Bin Cheng <bin dot cheng at arm dot com>
- Cc: Craig Howland <howland at LGSInnovations dot com>, "newlib at sourceware dot org" <newlib at sourceware dot org>
- Date: Wed, 13 Aug 2014 10:41:19 -0500
- Subject: Re: [PATCH libgloss]Using spec files to support two version of newlib library in one tool-chain release
- Authentication-results: sourceware.org; auth=none
- References: <001e01cfb600$ce404f10$6ac0ed30$ at arm dot com> <53EA3269 dot 1080304 at LGSInnovations dot com> <1407866624 dot 2601 dot 77 dot camel at ubuntu-sellcey> <000001cfb6b9$b6fa3690$24eea3b0$ at arm dot com> <1407943957 dot 2601 dot 98 dot camel at ubuntu-sellcey>
On 8/13/2014 10:32 AM, Steve Ellcey wrote:
> On Wed, 2014-08-13 at 13:44 +0800, Bin Cheng wrote:
>> Thank both of you for the suggestion. I agree _s isn't appropriate here,
>> and searched and replaced it with _n in the attached patch. If anyone
>> thinks _n isn't clear enough I suggest we just use _nano here, since the
>> word is used elsewhere in both file name and configuration option name.
>> Is it OK for you guys?
> I am happy with _n or _nano.
I agree with this.
_p is profiling.
_g is sometimes used for debug versions.
_s (IMO) would likely be a normal build but with -Os variant.
Using _n or _nano and adding -Os would tweak
the build as far as it could be tweaked for size.
I assume that's the intent.
> Steve Ellcey
Joel Sherrill, Ph.D. Director of Research & Development
joel.sherrill@OARcorp.com On-Line Applications Research
Ask me about RTEMS: a free RTOS Huntsville AL 35805
Support Available (256) 722-9985