Re: Fixed point Vs Floating point
- From: JosephKK <quiettechblue@xxxxxxxxx>
- Date: Sat, 21 Jun 2008 17:56:40 -0700
On Sat, 21 Jun 2008 16:47:36 -0400, Phil Hobbs
<pcdhSpamMeSenseless@xxxxxxxxxxxxxxxxxx> wrote:
Phil Hobbs wrote:
JosephKK wrote:
Do you know of any processor that handles denormals in hardware,
other than by flushing to zero? I haven't found one, and I don't
think any exists. That's what I mean by "no hardware support".
Cheers,
Phil Hobbs
Gosh, just after i named multiple ones. Go study the databooks before
you respond.
None of the ones that you named do it--certainly not any Intel ones.
Support for denormals is provided by handling the EM_DENORMAL exception
in software, or by flushing to zero.
Nice bluff, though.
Cheers,
Phil Hobbs
I looked in the latest version of the Intel software optimization
manual, http://developer.intel.com/design/processor/manuals/248966.pdf,
and it turns out that the denormals are actually handled in firmware
rather than in the runtime library. So we're sort of both right--the
FPU handles it, but it's bog slow compared with real hardware support.
Cheers,
Phil Hobbs
I will gladly take a draw rather than drag this out.
.
- References:
- Re: Fixed point Vs Floating point
- From: skilambi
- Re: Fixed point Vs Floating point
- From: John Larkin
- Re: Fixed point Vs Floating point
- From: Phil Hobbs
- Re: Fixed point Vs Floating point
- From: John Larkin
- Re: Fixed point Vs Floating point
- From: Phil Hobbs
- Re: Fixed point Vs Floating point
- From: JosephKK
- Re: Fixed point Vs Floating point
- From: Phil Hobbs
- Re: Fixed point Vs Floating point
- From: JosephKK
- Re: Fixed point Vs Floating point
- From: Phil Hobbs
- Re: Fixed point Vs Floating point
- From: Phil Hobbs
- Re: Fixed point Vs Floating point
- Prev by Date: Re: Attorney generals trying to shut down usenet?
- Next by Date: Re: 8-core embedded processor
- Previous by thread: Re: Fixed point Vs Floating point
- Next by thread: Re: Fixed point Vs Floating point
- Index(es):
Relevant Pages
|
