Hi,
Side by side comparison with one session of CCS5 and CCS6.1 shows that V6 takes some 3 times of V5 in stepping operations.
Further trials on our machine suggest that the low speed is due to “variable” “disassembly” window updating. Closing these views in debugging facilitates the speed and could step/hit breakpoint as fast as normal.
However, such are the gist of the debugging perspective which should not be disabled.
JDK: latest 8u45, both 32 and 64 bit installed.
CCS is 6.1.0.00104.
the eclipse.ini is also already tuned when the 3x time stepping/breakpoint debugging performance difference is clearly felt:
-startup plugins/org.eclipse.equinox.launcher_1.3.0.v20140415-2008.jar --launcher.library plugins/org.eclipse.equinox.launcher.win32.win32.x86_1.1.200.v20140603-1326 -showsplash org.eclipse.platform --launcher.XXMaxPermSize 256m --launcher.defaultAction openFile --launcher.appendVmargs -vmargs -Xms384m -Xmx1280m originally (after install) -vmargs -Xms40m -Xmx512m
Eclipse.org says “The 3.x platform had many years to optimize the critical code paths, and these optimizations will take time to make on the 4.x stream over the coming releases.”
So what is TI’s stance on this? Basically it appears that the slowing for CCS6 is due to specific views. Can they be optimized/fixed quickly instead of waiting for the NOG like eclipse.org’s time-taking moving-ons?
Matt