This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: -mfpmath=sse -mno-80387 troubles

From: tbp <tbptbp at gmail dot com>
To: Uros Bizjak <uros at kss-loka dot si>
Cc: gcc at gcc dot gnu dot org
Date: Fri, 28 Jan 2005 17:23:37 +0100
Subject: Re: -mfpmath=sse -mno-80387 troubles
References: <41F8E9DA.9050504@kss-loka.si> <4fc48eb105012807541404ae35@mail.gmail.com>
Reply-to: tbp <tbptbp at gmail dot com>

Found the culprit.
Somewhere my rdtsc timer asks for a int64_t -> float64_t conversion (i
thought it would be straighter than uint64_t -> float64_t if a bit
wrong).

Without -mno-80387 there's a "fildll (%esp); fstpl  0x18(%esp)".
With -mno-80387 there's some pushing and then a call to floatdidf
which returns the result on the top of the fpu stack.

I'm sure that's not the only thing that goes wrong, but frankly i
wasn't expecting a cast to go through libc.
Live & learn i guess :)

Follow-Ups:
- Re: -mfpmath=sse -mno-80387 troubles
  - From: Richard Henderson

References:
- Re: -mfpmath=sse -mno-80387 troubles
  - From: Uros Bizjak
- Re: -mfpmath=sse -mno-80387 troubles
  - From: tbp

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]