Hello all.
Recently I've started to port and optimize our software for intel-based
macs.
I am using macmini (core duo 1.66 GHz) with OS X 10.4.6, XCode 2.2.1
and CHUD Tools 4.3.2
I've got several questions about Shark on intelmacs
1) The "Cycles" column sometime shows "N/A" for some instruction. Why
where are no information for instruction latency/throughoutput?
2) So far most of the warnings are only "LCP" warning/advice for the
code, recommending me to change loop counters from 16-bit to 32-bit
integers. I can not see any stalls, i was getting plenty of stall
warning for PowerPC code. Does this mean such info is not available
while profiling intel code?
3) The Advanced options only show two possible CPUs, Pentium 4 and
Pentium M. Which intel-based macs model has Pentium 4 and which has
Pentium M CPUs?
And another question about optimization.
I would like to compare how worse code would behave if compiler wont
use the SSE/SSE2 instruction for scalar floating point, and try to
compile code to use only x387 math. Which additional compiler flags I
should pass to gcc to prohibit usage of SSE/SSE2?
Thank you in advance.
--
Sincerely,
Rustam Muginov
