[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Do I want to debootstrap Debian armhf on my iMX53 board?



On Fri, Aug 19, 2011 at 01:29:38PM -0400, Lennart Sorensen wrote:
> On Fri, Aug 19, 2011 at 06:26:37PM +0100, David Given wrote:
> > Lennart Sorensen wrote:
> > [...]
> > > I have debootstrapped debian armhf unstable onto a sata drive which I
> > > have mounted under the microsd running ubuntu.  Certainly working on
> > > the sata is way faster than that crappy little microsd it came with.
> > 
> > Any chance you could run nbench
> > (http://www.tux.org/~mayer/linux/bmark.html) for a crude rule-of-toe
> > benchmark? FWIW, my SheevaPlug gets memory=3.3, integer=4.6, float=0.36.
> > 
> > Also, hdparm -tT on the SATA, SD and uSD would be interesting.
> 
> I think I can try that just as soon as libtool finished building.

# hdparm -tT /dev/sda /dev/mmcblk0

/dev/sda:
 Timing cached reads:   512 MB in  2.00 seconds = 255.59 MB/sec
 Timing buffered disk reads: 134 MB in  3.01 seconds =  44.49 MB/sec

/dev/mmcblk0:
 Timing cached reads:   490 MB in  2.00 seconds = 244.49 MB/sec
 Timing buffered disk reads:  36 MB in  3.09 seconds =  11.63 MB/sec

/dev/sda is a seagate momentus 5400.5 160GB laptop drive out of a broken
EEEpc 1008HA.

I will try giving the benchmark a try:

TEST                : Iterations/sec.  : Old Index   : New Index
                    :                  : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT        :          441.12  :      11.31  :       3.72
STRING SORT         :          55.715  :      24.90  :       3.85
BITFIELD            :      1.9778e+08  :      33.93  :       7.09
FP EMULATION        :          60.536  :      29.05  :       6.70
FOURIER             :          1788.9  :       2.03  :       1.14
ASSIGNMENT          :           7.787  :      29.63  :       7.69
IDEA                :          1481.3  :      22.66  :       6.73
HUFFMAN             :          693.62  :      19.23  :       6.14
NEURAL NET          :          2.1138  :       3.40  :       1.43
LU DECOMPOSITION    :          76.099  :       3.94  :       2.85
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX       : 23.169
FLOATING-POINT INDEX: 3.009
Baseline (MSDOS*)   : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU                 : 
L2 Cache            : 
OS                  : Linux 2.6.35.3-744-g27fdf7b
C compiler          : gcc version 4.6.1 (Debian 4.6.1-4) 
libc                : libc-2.13.so
MEMORY INDEX        : 5.943
INTEGER INDEX       : 5.664
FLOATING-POINT INDEX: 1.669
Baseline (LINUX)    : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.

I guess that's quite a bit faster than your sheevaplug then.  Especially
on floating point.  No idea if the default gcc settings the makefile
used are even ideal.

> I think I fixed the other place that was missed in the latest libtool fix.

So libtool seems to have built.  I sent a patch to bug 632614 for it.
Now I can actually apt-get build-dep vim, so I guess libtool is rather
essential.

-- 
Len Sorensen


Reply to: