Movatterモバイル変換

This is the mail archive of thelibc-alpha@sourceware.orgmailing list for theglibc project.

Re: [PATCH 3/4] Add ILP32 support to aarch64

• From: Joseph Myers <joseph at codesourcery dot com>
• To: Steve Ellcey <sellcey at cavium dot com>
• Cc: libc-alpha <libc-alpha at sourceware dot org>
• Date: Thu, 3 Aug 2017 19:48:31 +0000
• Subject: Re: [PATCH 3/4] Add ILP32 support to aarch64
• Authentication-results: sourceware.org; auth=none
• References: <1501774579.3962.54.camel@cavium.com> <alpine.DEB.2.20.1708031744530.24143@digraph.polyomino.org.uk> <1501784546.3962.77.camel@cavium.com>

On Thu, 3 Aug 2017, Steve Ellcey wrote:> I hadn't considered that.  So maybe I could save the environment> (feholdexcept), do the calculation and see which exceptions, if any,> got raised.  Then restore the original environment (fesetenv) and raise> one or the other exceptions if needed.  Does that sound like a workable> solution?Yes, you could do that (see x86/x86_64 nearbyintl for example; that needs to preserve "invalid" for signaling NaNs, avoid "inexact" from frndint, but not clear any "inexact" that was already raised).  Whether doing so is optimal, versus using another implementation of the function for ILP32, may depend on the AArch64 performance characteristics of saving and restoring the environment.-- Joseph S. Myersjoseph@codesourcery.com

• References:
  • [PATCH 3/4] Add ILP32 support to aarch64
    • From: Steve Ellcey
  • Re: [PATCH 3/4] Add ILP32 support to aarch64
    • From: Joseph Myers
  • Re: [PATCH 3/4] Add ILP32 support to aarch64
    • From: Steve Ellcey