Intel C++ Compiler and Performance Library for QNX* Neutrino* RTOS

"FRAME_INTERPOLATION" definition in the IPP 7.0 H264 encoder

The "FRAME_INTERPOLATION" definition IS USED BY THE CURRENT ipp SAMPLE
CODE (H264 encoder) to switch the way how the half pel got interpolated.
But I am not clear about the difference if I turn it on or off. My unstanding
is :
(1) When the FRAME_INTERPOLATION turns on, all the half pel on
the frame will be interpolated once and saved for motion estimation
later;
(2) otherwise, the half pel interpolation is happen on-the-fly per block basis during the sub-pel ME.

Am I right? Thanks and Regards.

idb core dumps - idbc weird warnings - idbc cant catch my break points.

Hi,

I am opening idb and it coresafter start up. Also, idbc gives all of these warnings when I start it with and without any arguments. Furthermore,idbc fails to be able to catch break points in my code. It can catch break points in calls toglibc justfine butcannot provide abacktrace.I am compiling with g++ 4.1.2 20080704 (Red Hat 4.1.2-48) in RHEL 5.6.
I tried running idb under JRE 1.5.0_06-b05 and 1.6.0_16-b01.

Both are Version 12.0, Build [1.3842.2.129]

Thanks

IDB backtrace:

(gdb) backtrace full

idb core dumps - idbc weird warnings - idbc cant catch my break points.

Hi,

I am opening idb and it coresafter start up. Also, idbc gives all of these warnings when I start it with and without any arguments. Furthermore,idbc fails to be able to catch break points in my code. It can catch break points in calls toglibc justfine butcannot provide abacktrace.I am compiling with g++ 4.1.2 20080704 (Red Hat 4.1.2-48) in RHEL 5.6.
I tried running idb under JRE 1.5.0_06-b05 and 1.6.0_16-b01.

Both are Version 12.0, Build [1.3842.2.129]

Thanks

IDB backtrace:

(gdb) backtrace full

Is it possible yo use Cache as RAM?

Hello Everyone,
I am Vivek Chand, pursuing my BE in Dr.AIT at Bangalore.
I am designing an OS for MultiCore and ManyCore chips to fully exploit it which wolud be a modification to xv6 and JOS Kernel(MIT's exokernel)
I have the ideas for implementing an deterministic and race-free Operating System.
Now, to make a bit different from the existing OS design, i want to load my microkernel into L2 cache permanently.
i.e, I want my microkernel to be bootloaded into L2 Cache rather than into main memory.

Installation of Intel(R) C++ Compiler 11.1.009 with floating QNX license

Hi,
I have a QNX installation on a network share with a floating flexlm license (i.e. no local installation of the QNX Momentix environment). QNX momentix works properly (launching qcc from command line works fine) and QNX_CONIGURATION environment variable is set correctly.
Despite this, I have a message during Intel C++ Compiler installation saying QNX Momentix is not properly installed :

Error when using C++ Exception handling

An error in the Intel C++ Compiler for QNX configuration has been discovered. Below is a description of the problem and a simple workaround.

Problem: If a C++ application is using exception handling there will be an error in runtime when the exception is thrown.

Workaround: Make sure that you are using the option -fexceptions for the linking stage

订阅 Intel C++ Compiler and Performance Library for QNX* Neutrino* RTOS