Comment #8 on issue 436 by [email protected]: Numeric results are wrong on
some systems due to x87 extended double precision
http://code.google.com/p/v8/issues/detail?id=436
On the 32-bit ia32 platform, even computers supporting SSE2 do not always
use SSE2
code, because the snapshot of precompiled native JS routines used at
startup needs to
work on any ia32 platform. Therefore, the generated code in the snapshot
uses FPU
instructions. There may be more of a possibility of eliminating
extended-precision
floating-point calculations on the x64 platform, since SSE2 is know to be
supported
on them.
Switching the FPU flags from extended precision to standard precision and
back is
probably not fast, because some library routines may already switch the
flags to set
the rounding mode differently, and use up the second set of fast flag
settings that
way already.
I think the resolution of this could be "will fix on X64, will not fix on
ia32"
--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
To unsubscribe from this group, send email to v8-dev+unsubscribegooglegroups.com or reply
to this email with the words "REMOVE ME" as the subject.
