[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Compiling ATLAS with MinGW
From: |
Michael Goffioul |
Subject: |
Re: Compiling ATLAS with MinGW |
Date: |
Tue, 21 Feb 2012 22:15:27 +0000 |
On Tue, Feb 21, 2012 at 9:05 PM, Michael Goffioul
<address@hidden> wrote:
> PS: also I got some strange result when applying the same procedure on
> a Intel Atom N270. When compiling ATLAS, surprizingly it does not
> select SSE2 kernels for double-precision operation, but plain x87
> ones. OTOH, it does select SSE kernels for single-precision
> operations. The result is that the compiled ATLAS is almost 2.5x
> faster for single operations than double operations. Any hint on why
> x87 kernels appears faster than SSE2 kernels during ATLAS tuning are
> welcome :)
Wrong figure: single-precision is about 5x faster than double-precision.
Michael.
- Compiling ATLAS with MinGW, Michael Goffioul, 2012/02/18
- Re: Compiling ATLAS with MinGW, Tatsuro MATSUOKA, 2012/02/18
- Re: Compiling ATLAS with MinGW, Michael Goffioul, 2012/02/18
- Re: Compiling ATLAS with MinGW, Michael Goffioul, 2012/02/18
- Re: Compiling ATLAS with MinGW, Tatsuro MATSUOKA, 2012/02/18
- Re: Compiling ATLAS with MinGW, Michael Goffioul, 2012/02/19
- Re: Compiling ATLAS with MinGW, Tatsuro MATSUOKA, 2012/02/19
- Re: Compiling ATLAS with MinGW, Michael Goffioul, 2012/02/21
- Re: Compiling ATLAS with MinGW,
Michael Goffioul <=
- Re: Compiling ATLAS with MinGW, Tatsuro MATSUOKA, 2012/02/21
- Re: Compiling ATLAS with MinGW, Fredrik Lingvall, 2012/02/19
Re: Compiling ATLAS with MinGW, nitnit, 2012/02/19