int t = (data[c] – 128) >> 31; The trick here is that if data[c] >= 128, then data[c] – 128 is nonnegative, otherwise it is negative. The highest bit in an int, the sign bit, is 1 if and only if that number is negative. >> is a shift that extends the sign bit, … Read more
Here are my thoughts on this after staring at it for a while. First of all, the issue is easily reproducible with -O2, so it’s better to use that as a reference, as it generates simple non-unrolled code that is easy to analyse. The problem with -O3 is essentially the same, it’s just a bit … Read more
Alignment optimisations 1. Use .p2align <abs-expr> <abs-expr> <abs-expr> instead of align. Grants fine-grained control using its 3 params param1 – Align to what boundary. param2 – Fill padding with what (zeroes or NOPs). param3 – Do NOT align if padding would exceed specified number of bytes. 2. Align the start of a frequently used code … Read more
The short answer: no, it is not. GCC does metrics ton of non trivial optimization and one of them is guessing branch probabilities judging by control flow graph. According to GCC manual: fno-guess-branch-probability Do not guess branch probabilities using heuristics. GCC uses heuristics to guess branch probabilities if they are not provided by profiling feedback … Read more
int t = (data[c] – 128) >> 31; The trick here is that if data[c] >= 128, then data[c] – 128 is nonnegative, otherwise it is negative. The highest bit in an int, the sign bit, is 1 if and only if that number is negative. >> is a shift that extends the sign bit, … Read more
Such a large difference is the product of two conditions. The first condition is related to the original code. The in-place merge is so efficient there would be difficulty devising anything significantly faster, even if coding manually at the assembly language level. The application of generics is straightforward, so the compiler ** produced the same … Read more
These instruction prefixes have no effect on modern processors (anything newer than Pentium 4). They just cost one byte of code space, and thus, not generating them is the right thing. For details, see Agner Fog’s optimization manuals, in particular 3. Microarchitecture: http://www.agner.org/optimize/ The “Intel® 64 and IA-32 Architectures Optimization Reference Manual” no longer mentions … Read more
AMD processor in the early-gigahertz era had a 40 cycle penalty every time you called a function Huh.. so large.. There is an “Indirect branch prediction” method, which helps to predict virtual function jump, IF there was the same indirect jump some time ago. There is still a penalty for first and mispredicted virt. function … Read more
people often … and state that programmers usually can predict where a branch could go (*) Experienced programmers often remind that human programmers are very bad at predicting that. 1- Is it possible to avoid branch mispredictions using some high level programming technique (i.e. no assembly)? Not in standard c++ or c. At least not … Read more
The canonical way to do static branch prediction is that if is predicted not-branched (i.e. every if clause is executed, not else), and loops and backward-gotos are taken. So, don’t put the common case in else if you expect static prediction to be significant. Getting around an untaken loop isn’t as easy; I’ve never tried … Read more