Products       Learn       Buy       Support       Company
 
  Home > Products > Other Products > GoFast
   
 

Math Products


GoFast® for ARM / Cortex and CrossWorks


Fast Floating Point Library


Features

  • Fast
  • Reentrant
  • ROMable
  • Conforms to IEEE 754 Format
  • "Link and Go" compiler support for CrossWorks ARM
  • Includes complete source, test programs, project files, and startup code


Description

GoFast® for ARM / Cortex was carefully designed for high performance operation in embedded applications and ease of use including “link and go” compatibility with the CrossWorks C compiler. GoFast provides ROMable, reentrant IEEE and ANSI compatible ARM floating point support. It boosts the performance of an application’s math calculations or eliminates the need for a hardware floating-point coprocessor, in order to reduce product manufacturing cost. It is delivered with full assembly source code.

Currently GoFast is offered only for the ARM and Thumb-2 instruction sets, not Thumb.

Functionality

GoFast ARM / Cortex offers the following reentrant floating point routines, for both single and double precision:

  • intrinsic basic operations + - * /
  • intrinsic conversions
  • sqrt
  • sin, cos, tan
  • asin, acos, atan, atan2
  • sinh, cosh, atanh
  • log, log10, exp, pow
  • floor, ceil, fabs
  • modf, fmod, frexp, ldexp


Floating Point Technology

GoFast is based on “Architecture Independent Technology” (AIT) and proven floating point algorithms that were developed for over a decade. The algorithms have been thoroughly tested using automated methods.

Conformance and Testing

The accuracy of each GoFast Floating Point Library is within one (least significant) bit for arithmetic functions and two bits for transcendental functions, in most cases. The IEEE 754 Floating Point Format defines special representations for underflow, overflow, and invalid operation. The GoFast routines use these formats and adhere to the IEEE 754 error handling procedures in all applicable cases. Quality assurance and testing procedures have assured proper product operation. In addition, each delivery includes target specific test programs assuring confidence of product operation.

Timings

The following tables give the times for all floating point operations, for GoFast and the CrossWorks floating point library. The times, in microseconds, were measured using the indicated processor and evaluation board. The basic operations (add, subtract, multiply, divide, conversions, and comparisons) in the CrossWorks C library are hand-coded and faster than those in GoFast, so the CrossWorks versions are used instead. (If you only need these basic operations, you don’t need GoFast.) Thus, the routines linked are a mixture of both libraries, as indicated in bold below. GoFast provides the greatest benefit for the more complex operations, typically offering a 2x to 3x performance boost.
 

Microsecond Timings
 

ARM7:  AT91SAM7X256-EK, 48 MHz, RAM

 
Double Precision
Single Precision
Function
GoFast
CrossWorks
GoFast
CrossWorks
add
2.5
1.6
1.6
1.0
subtract
3.0
2.0
1.9
1.4
multiply
2.7
2.2
1.6
1.0
divide
16.5
9.2
8.3
2.3
sqrt
32.3
46.6
16.7
15.3
exp
13.1
46.6
4.3
32.4
log
27.2
60.2
11.3
28.9
log10
28.3
62.4
11.6
30.0
sin
11.6
41.9
4.0
21.0
cos
10.6
53.1
4.0
25.2
tan
28.2
59.9
11.3
28.5
asin
45.1
109.2
34.8
48.2
acos
43.8
107.6
42.2
46.8
atan
27.1
64.4
11.6
30.2
atan 2
42.8
75.5
19.2
34.6
pow
   40.5
  112.9
  16.7
  65.8
tanh
30.7
62.5
19.2
40.1
sinh
30.4
65.0
12.1
39.0
cosh
29.8
63.9
11.6
40.6
modf
1.4
1.9
1.1
1.3
fmod
1.6
16.5
1.0
8.0
fabs
0.5
0.5
0.3
0.3
floor
1.1
1.9
0.7
1.2
ceil
1.2
1.9
0.7
1.2
ldexp
2.2
2.3
1.4
1.6
frexp
1.3
1.8
0.8
1.3
cmp
1.7
1.0
1.2
0.8
fp to long
1.0
0.6
0.7
0.5
fp to ulong
1.0
0.5
0.7
0.5
long to fp
4.6
0.8
4.3
0.7
ulong to fp
4.5
0.7
4.2
0.7
sgl to dbl
0.9
0.5
dbl to sgl
1.1
0.6



GoFast is a registered trademark of Lantronix Inc. ARM is a registered trademark of ARM Ltd.

     back to top
 
  Register for More Info
 
 

 
Home       Sitemap       Contact