I tried it, but it did not work at all (in Sweden).
It downloaded 50-100 Kbyte of data for every pan-operation. That is expensive.
There are better alternatives (5 times less data) at www.idevio.com// Magnus
Re:question for all you Java experts
on
Effective Java
·
· Score: 1
try ibm:s jdk.
it is 20 % faster than gcc-c-code on FIR-filter (float-perofrmance).
Almost ALL security exploits i have read about, are caused by humans making mistakes in their code. This is a human thing to do. This will always be like this. If a check is missing, someone will find that mistake.
How to fix this:
Use a stack-guarded compiler (c), or another language (java?). Then this cannot happen.
I get upset when i read about security vulnerabilities EVERY week in lwn.net.It is always the same mistake. Why not fix this for all ?
Just add that compiler-switch. Then the problem is no-more.
Some people think the application gets slower. This is not so. I cannot understand why redhat and microsoft don't do this ?
There have not been any breakins into any jsp-servers written under java, which performs range-checking on arguments (which is optimized away from the innen loops by the hotspot compiler).
if you think java is slow:
http://www.cs.vu.nl/manta/// magnus persson
Slashdot Mirror
Sorry. You have done a wrong calculation it should be 27 nanoseconds.
I tried it, but it did not work at all (in Sweden). It downloaded 50-100 Kbyte of data for every pan-operation. That is expensive. There are better alternatives (5 times less data) at www.idevio.com // Magnus
try ibm:s jdk. it is 20 % faster than gcc-c-code on FIR-filter (float-perofrmance).
Almost ALL security exploits i have read about, are caused by humans making mistakes in their code. This is a human thing to do. This will always be like this. If a check is missing, someone will find that mistake. How to fix this: Use a stack-guarded compiler (c), or another language (java?). Then this cannot happen. I get upset when i read about security vulnerabilities EVERY week in lwn.net .It is always the same mistake. Why not fix this for all ?
Just add that compiler-switch. Then the problem is no-more.
Some people think the application gets slower. This is not so. I cannot understand why redhat and microsoft don't do this ?
There have not been any breakins into any jsp-servers written under java, which performs range-checking on arguments (which is optimized away from the innen loops by the hotspot compiler).
if you think java is slow:
http://www.cs.vu.nl/manta/ // magnus persson