On Mon, Nov 10, 2003 at 08:37:53AM +0200, Tal, Shachar wrote: > Just to give a few examples: > gcc/glibc's atoi() is 50% slower than VC6 runtime with Intel's massively > optimizing compiler.
Can you produce the generated assembly for both cases? which options were given to gcc? which glibc version exactly? > string switching (i.e. C switch on strings) is twice as fast in VC6/Intel > than gcc/glibc. likewise. > string tokenization is 20% slower on gcc/glibc than Intel. likewise. > AFAIK, TLS libraries, aside from requiring more actual code about stack > allocation, hold several stacks or one segmented stack (for several > threads), and will tend to use more space and hence perform worse on L1/L2 > caches. You aren't saying much here, except "threads can trash the cache", which is a given. > Overall, I wouldn't be surprised that a TLS implementation, on top of > optimizations, will amount to slower performance. I stopped being surprised at anything related to hardware or software a long time ago. What's interesting is figuring out why things behave as they do. -- Muli Ben-Yehuda http://www.mulix.org | http://mulix.livejournal.com/ "the nucleus of linux oscillates my world" - [EMAIL PROTECTED]
signature.asc
Description: Digital signature