Yep, it honestly never occurred to me that the name of the tool wasn't godbolt. I just went to the site and noticed "Compiler Explorer" for the first time, which seems so generic that I never thought of it as a name.
I host it locally for myself too and got the CLion plugin, it is amazing. Seeing actual codegen in projets with multiple files is really something and having the same scrolling with ASM or source too.
The floating point comment leaves out that one can use
#pragma omp simd reduction(+:res)
as a more precise way to achieve vectorization in the reduction (compile with -fopenmp-simd to only use it for SIMD without linking an OpenMP library): https://godbolt.org/z/17oTz1
Unfortunately, the pragma is not supported with the new-style class iterators in a released compiler, though it works in clang-trunk: https://godbolt.org/z/hbP11W
Note that Clang disables floating point contraction by default (so no vfmadd instructions), despite them being more accurate. One usually wants this globally (-ffp-contract=fast) except when trying to bitwise reproduce software compiled for pre-Haswell.
> I hope that some of these optimizations are a pleasant surprise and will factor in your decisions to write clear, intention-revealing code and leave it to the compiler to do the right thing.
This was my key takeaway from this article. Writing clear code that is easier to maintain will have good enough performance most of the time. I was particularly impressed with the devirtualization optimizations and will be less likely to shy away from using polymorphism in future due to performance concerns.
> Tail call removal. A recursive function that ends in a call to itself can often be rewritten as a loop, reducing call overhead and reducing the chance of stack overflow.
Most important: this optimization enables pipelined execution.
When people talk about a CPU executing an integer add instruction in ~1 cycle, what they actually mean is that the add has this latency when the CPU pipelines are full.
If you have an 11 stage pipeline... the add can often have a latency of ~11 cycles... if you write the _right_ code for it.
FWIW, at least on intel cpu call instructions are fully pipelined and effectively zero latency (although, like all jump instructions there is a limit of one call every other cycle).
Interestingly, I've seen infinite recursions get optimized to infinite tail calls. This makes debugging harder because instead of a stack overflow you just have an infinite loop and have to manually go kill the process and get a breakpoint.
Then, looking at the code it's not obvious where the infinite loop occurs.
Tail call removal is much broader than that. Compilers will remove even indirect tail calls, which is useful when building a fast interpreter. I tried this with a small example (using Godbolt, of course), now let’s see if it works for a full interpreter.
Nice try. You can’t escape being known as a verb now.
Everyone knows the tool as godbolt.