This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH 2/14] [x86_64] Vector math functions (added vector cos tests)
- From: Joseph Myers <joseph at codesourcery dot com>
- To: Andrew Senkevich <andrew dot n dot senkevich at gmail dot com>
- Cc: libc-alpha <libc-alpha at sourceware dot org>
- Date: Fri, 5 Jun 2015 12:14:11 +0000
- Subject: Re: [PATCH 2/14] [x86_64] Vector math functions (added vector cos tests)
- Authentication-results: sourceware.org; auth=none
- References: <CAMXFM3uL-z3r0Q2aSpnPQnGLkj1_SDmj+h=WUr4MaaMaRGGftg at mail dot gmail dot com> <alpine dot DEB dot 2 dot 10 dot 1505221547070 dot 16611 at digraph dot polyomino dot org dot uk> <CAMXFM3vdR5gEkHwkdfmN0cpHH8_Vzk-XcnYJdMJZBrv_P+nOGQ at mail dot gmail dot com> <alpine dot DEB dot 2 dot 10 dot 1505281726000 dot 16930 at digraph dot polyomino dot org dot uk> <CAMXFM3sPN1p7NRU19-NzT4S5P=-7gasbxLb1rs5q2CK_qnt6-Q at mail dot gmail dot com> <alpine dot DEB dot 2 dot 10 dot 1506041710150 dot 12011 at digraph dot polyomino dot org dot uk> <CAMXFM3sqmZLvBP+QWeL29y5+MJy1pOC_ei67N0ZtMH-xDVMV0Q at mail dot gmail dot com>
On Fri, 5 Jun 2015, Andrew Senkevich wrote:
> But current design was discussed and approved in december 2014, and
> already 5 patches with this current design were committed.
I can't find anything in December 2014 that looks to me like approval of
this aspect of the design of patch 9/N (as it was then), just discussion
of other issues with duplication in the code and incremental refactoring
of it.
I don't think my comments in any way conflict with the pieces that have
already been committed; they are specifically about the arrangement of
things that are new in this patch (namely, definitions and declarations
for wrapper functions used in testing, that are sufficiently mechanical
that they should not need repeating for every (function, vector format)
pair).
> Now I have series of 14 patches ready and such refactoring at this
> moment is enough significant effort. So I propose to add this
> refactoring as additional patch after this series. Is it ok?
No, I think the subsequent patches show up a deficiency in how each new
function is tested, and so that should be cleaned up to reduce duplication
before the second new function is added, rather than each patch adding
more avoidable duplication of boilerplate test code. It's unavoidable
that sometimes adding more functions shows up design issues that weren't
apparent with the addition of the first function.
--
Joseph S. Myers
joseph@codesourcery.com