This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [RFC][BZ #17943] Use long for int_fast8_t
- From: "Maciej W. Rozycki" <macro at linux-mips dot org>
- To: "H.J. Lu" <hjl dot tools at gmail dot com>
- Cc: Ondřej Bílka <neleai at seznam dot cz>, GNU C Library <libc-alpha at sourceware dot org>
- Date: Mon, 9 Feb 2015 14:21:45 +0000 (GMT)
- Subject: Re: [RFC][BZ #17943] Use long for int_fast8_t
- Authentication-results: sourceware.org; auth=none
- References: <20150208110426 dot GA28729 at domone> <alpine dot LFD dot 2 dot 11 dot 1502091225590 dot 22715 at eddie dot linux-mips dot org> <CAMe9rOrBJfr5YcijY-epvfig68Gc3Zm64zF2YXCTRDDo3J2CvQ at mail dot gmail dot com>
On Mon, 9 Feb 2015, H.J. Lu wrote:
> > [1] "Intel 64 and IA-32 Architectures Optimization Reference Manual",
> > Intel Corporation, Order Number: 248966-020, November 2009, Table 12-2
> > "Intel Atom Microarchitecture Instructions Latency Data", p. 12-21
>
> This manual is very old and Intel Atom Microarchitecture has been
> replaced by Silvermont Microarchitecture.
Maybe, I picked whatever I had handy. That doesn't change the fact this
silicon is out there. There's older yet x86 silicon in the field where
the divider has a similar characteristic.
As I say for a generic header you need to weigh the balance -- except
that the ABI has been set, so there's little if anything we can do, apart
from new architecture additions that can get their overrides right from
the beginning.
Maciej