This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Minimum floating-point requirements
- From: "Carlos O'Donell" <carlos at redhat dot com>
- To: David Edelsohn <dje dot gcc at gmail dot com>, "Joseph S. Myers" <joseph at codesourcery dot com>
- Cc: libc-alpha at sourceware dot org, Adhemerval Zanella <azanella at linux dot vnet dot ibm dot com>, Steve Munroe <sjmunroe at us dot ibm dot com>
- Date: Wed, 29 Jan 2014 23:33:25 -0500
- Subject: Re: Minimum floating-point requirements
- Authentication-results: sourceware.org; auth=none
- References: <CAGWvnym4yN=7rLrm0RRtNN++T=xwx8r3MUKJOfz4r+H=Z9zd7Q at mail dot gmail dot com>
On 01/29/2014 05:57 PM, David Edelsohn wrote:
> If GLIBC becomes overly strict and dictatorial under the guise of
> conformity and policy, it will drive new developers, new ports and new
> innovation away, ultimately undermining the community and project. One
> can push more of the responsibility and work to maintain
> non-conforming ports onto port-specific maintainers to discourage it,
> but it should not be the role of GLIBC to impose its will on
> ABI-specific decisions of a target.
Sure, in generic terms we all agree it's a bad thing for glibc
to force anyone, volunteer, paid developer, port maintainer,
FSF steward, to do something they don't want to do. These people,
particularly port maintainers, understand better than anyone the
needs of their users.
This isn't a generic discussion though, this is strictly about
floating point formats. If you have users whose needs are not
met by IEEE floating point formats, please speak up now.
Future ports should read glibc policy to understand why we want
IEEE format floating point. If after reading the policy, they
still believe they need a non-IEEE format floating point, they
should reach out to the community to talk about their user
requirements.
We will not unthinkingly follow policy if someone engages the
community to talk about their use requirements.
Cheers,
Carlos.