Re: gcj's IO performance vs blackdown JDK

[Mohan Embar <gnustuff at thisiscool dot com>]

Hi Per,

>Well, first question is: how often do we get to the StringBuffer case,
>as opposed the new String(buffer, pos, i - pos) case?  I would guess
>that case is much more frequent, so we want to concentrate on that
>first.

>A optimization:  Use String.valueOf(char[], int, int) rather than
>new String(char[], int, int).  The former allocates one object; the
>latter allocates two objects.

I'll try this.

>On the other hand, the StringBuffer case is relatively frequent;
>otherwise your changes wouldn't have made such a difference.  It
>might be worthwhile trying to igure out why.

I think it's because Christopher's test case is contrived in that
it operates on a test data file of 100000 lines of 1000 columns
each, which means that we frequently "wrap around".

Which brings me to my next question: what if we made buffer
circular?

>One issue to consider is this is that we may be reading misaligned
>data.  If we could page-align the buffer we might do much better.
>There we need to concentrate on the underlying FileInputStream.
>It is important that this be properly buffered, and that the byte
>buffer be page aligned.  This means that the byte buffer has to
>be a raw page, rather than a Java array.

Like Bryce, I looked in FileInputStream, but I didn't look long enough
to catch the 100 byte buffer thing.

Thanks for the other tips.

-- Mohan
http://www.thisiscool.com/
http://www.animalsong.org/