On Sunday, 26 June 2016 at 16:40:08 UTC, Jay Norwood wrote:
> After watching Andre's sentinel thing, I'm playing with strlen on char strings with 4 terminating 0s instead of a single one.  Seems to work and is 4x faster compared to the runtime version.
>
> nothrow pure size_t strlen2(const(char)* c) {
>  if (c is null)
>    return 0;
>  size_t l=0;
>  while (*c){ c+=4; l+=4;}
>  while (*c==0){ c--; l--;}
>  return l+1;
> }
>
> This is the timing of my test case, which I can post if anyone is interested.
> strlen\Release>strlen
> 2738
> 681

Did you also compare to strlen from libc? I'd guess GNU libc uses a lot more tricks like vector instructions.

On Tuesday, 28 June 2016 at 01:53:22 UTC, deadalnix wrote:
> On Sunday, 26 June 2016 at 16:40:08 UTC, Jay Norwood wrote:
>> After watching Andre's sentinel thing, I'm playing with strlen on char strings with 4 terminating 0s instead of a single one.
>>  Seems to work and is 4x faster compared to the runtime version.
>>
>> nothrow pure size_t strlen2(const(char)* c) {
>>  if (c is null)
>>    return 0;
>>  size_t l=0;
>>  while (*c){ c+=4; l+=4;}
>>  while (*c==0){ c--; l--;}
>>  return l+1;
>> }
>>
>> This is the timing of my test case, which I can post if anyone is interested.
>> strlen\Release>strlen
>> 2738
>> 681
>
> If we were in interview, I'd ask you "what does this returns if you pass it an empty string ?"

Since no one is answering:

It depends on the memory right before c. But if there is at least one 0 right before it - which is quite likely - then you get some crazy big number returned.

On Tuesday, 28 June 2016 at 09:31:46 UTC, Sebastiaan Koppe wrote:
>> If we were in interview, I'd ask you "what does this returns if you pass it an empty string ?"
>
> Since no one is answering:
>
> It depends on the memory right before c. But if there is at least one 0 right before it - which is quite likely - then you get some crazy big number returned.

Yes, the test checked for 0 length but not with a preceding 0.  I posted the fix.

  if (c is null || *c==0)

On Tuesday, 28 June 2016 at 09:18:34 UTC, qznc wrote:
> Did you also compare to strlen from libc? I'd guess GNU libc uses a lot more tricks like vector instructions.

I did test with the libc strlen, although the D libraries did not have a strlen for dchar or wchar.  I'm currently using this for comparison, and playing around with shorter string lengths:

nothrow pure size_t strlen(const(char)* c) {
  if (c is null )
    return 0;
  const(char)* c_save = c;
  while (*c) {
    c++;
  }
  return c - c_save;
}

I'm also trying some tests on a PA device where I have tools to look at cache hits, misses, branches mispredicted.  Similar C code.