This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Improves __ieee754_exp() performance by greater than 5x on sparc/x86.
On Mon, 23 Oct 2017, Siddhesh Poyarekar wrote:
> On a related note, if we are comfortable dropping exp slow path, we
> should probably take a serious look at the log slow path too since IIRC
> I wasn't even able to trigger it after days of running the test and it's
> quite possible that nobody cares. We could drop it and see if anybody
> notices.
The basis for removing a slow path in an existing implementation should be
an error analysis of the remaining code that justifies that the slow path
is never needed to avoid large errors. That might be an error analysis
that assumes the correctness of the existing code and deduces an error
bound on the basis that larger errors would make a check for being
correctly rounded (probably the one that gates entering the slow path in
question) incorrect.
--
Joseph S. Myers
joseph@codesourcery.com