ts-7000
[Top] [All Lists]

[ts-7000] Crunch-compiled software runs on TS7260 but crashes under TS73

To:
Subject: [ts-7000] Crunch-compiled software runs on TS7260 but crashes under TS7300
From: "marcel.groothuis" <>
Date: Fri, 16 Nov 2007 16:03:11 -0000
Hello everybody,

We have both the TS7260 and the TS7300 boards and we want to use for
control software that uses floating point and runs under RTAI.

Both boards boot currently the EXACTLY the same kernel (2.4.26 TS-11)
and have the same NFS-root filesystem (gcc 3.3.6 uClibc 0.9.28 based).

Our test example (the Cirrus test_libm example for testing the
Maverick FPU) for soft float testing runs fine on both boards but the
crunch accelerated versions (both the library and the crunch compiled
version) run only succesful on the TS7260 board.

All software is identical, the CPU's are identical but the boards are
slightly different (different surrounding HW and different bootloaders).
As soon as we have 1 or more Maverick instructions in our userspace
binary, it will crash on the TS7300 (see below for the output)

Does somebody have any idea or suggestions why the same software runs
on the TS7260 and fails on the TS7300 (all other non-crunch software
runs perfectly)?

Grt,
Marcel Groothuis

University of Twente
Control Engineering

--
Output TS7260:
:~# ./crunchgcctest
speed test for :crunchtest.c, DOUBLE PRECISION
Each operation runs 360000 times

OPS: calculation took         16 ms
ADD: calculation took          1 ms,   test rv=2.2556635255738473e+06
SUB: calculation took          1 ms,   test rv=-2.2556635255738473e+06
MUL: calculation took         28 ms,   test rv=7.0863759618797886e+06
NRD: calculation took        696 ms,   test rv=1.9758917663375622e+06
DIV: calculation took       1018 ms,   test rv=1.9758917663375622e+06
SIN: calculation took       1443 ms,   test rv=-6.2828660229222333e+03
SQRT:calculation took       6430 ms,   test rv=9.0113199321607614e+05
TAN: calculation took       1934 ms,   test rv=-6.2838230795149946e+03
LOG: calculation took       2048 ms,   test rv=6.6063435248867202e+05
EXP: calculation took       2040 ms,   test rv=1.8944159438234556e+08
SINH:calculation took       3902 ms,   test rv=9.4720455132449400e+07
speed test for :test_libm.c, SINGLE PRECISION
Each operation runs 360000 times
 
OPS: calculation took         16 ms
ADD: calculation took         26 ms,   test rv=2.2519580000000000e+06
SUB: calculation took         26 ms,   test rv=-2.2519580000000000e+06
MUL: calculation took         19 ms,   test rv=7.0631995000000000e+06
DIV: calculation took       1020 ms,   test rv=1.9793940000000000e+06
SIN: calculation took       1518 ms,   test rv=-6.2849443359375000e+03
SQRT:calculation took       6062 ms,   test rv=9.0009625000000000e+05
TAN: calculation took       2012 ms,   test rv=-6.2849575195312500e+03
LOG: calculation took       2133 ms,   test rv=6.5972843750000000e+05
EXP: calculation took       2114 ms,   test rv=1.8995438400000000e+08
SINH:calculation took       3985 ms,   test rv=9.4977192000000000e+07

-----
Output TS7300:
:~# ./crunchgcctest
speed test for :crunchtest.c, DOUBLE PRECISION
Each operation runs 360000 times

OPS: calculation took         18 ms
pc : [<00000000>]    lr : [<00010eac>]    Not tainted
sp : 7ffffb54  ip : 00000000  fp : 7ffffd0c
r10: 0001c014  r9 : 0000000b  r8 : 7ffffc64
r7 : 00000065  r6 : 00000000  r5 : 00000000  r4 : 00000010
r3 : 00000000  r2 : 00000000  r1 : 00000000  r0 : 00000000
Flags: nZCv  IRQs on  FIQs on  Mode USER_32  Segment user
Control: C000317F  Table: E51D4000  DAC: 00000015
Segmentation fault





 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/ts-7000/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/ts-7000/join
    (Yahoo! ID required)

<*> To change settings via email:
     
    

<*> To unsubscribe from this group, send an email to:
    

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 

<Prev in Thread] Current Thread [Next in Thread>
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU