As a sidenote: it is insane that I am now referred to as "performance focused." If you'd told someone at RAD to "go talk to the performance guy", they would literally go to every other office _but_ mine. I was easily the least performance-focused person there!
-
-
You can even watch this happen if you'd like. Here's a stream where I input _incredibly_ simple code into LLVM and have it generate an absolute disastrous output:https://youtu.be/R5tBY9Zyw6o?t=5934 …
-
Well, did you tried to run it through Intel's compiler? Just to see the difference. What I'm saying - it is impossible for majority of programmers to write ASM code manually that takes into account speculative executions, pipelines and parallelization.
End of conversation
New conversation -
-
-
This is 100% false because you found some bug in the compiler? Give the impression to a beginner ASM programmer that he will always write better ASM than the compiler is a really bad advice. 1/2
-
Verify what the compiler is doing in hot part of the code is a good skill to have. Learn how to suggest the compiler how to avoid performance pitfalls is a better skill, IMHO. 2/2
- Show replies
New conversation -
-
-
Also insanely useful for debugging those horrible crashes, like the ones which only happen in release and shit
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
