This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Patch] [BZ 15884] strcoll: improve performance by removing the cache


Ok, I'll do so. After sending this I noticed that there is a problem in json_attr_double. It's output is depended on the current locale - see the "mean" field below. What to do about it?

Am 14.10.2014 10:43, schrieb Siddhesh Poyarekar:
On Mon, Oct 13, 2014 at 02:19:48PM +0200, Leonhard Holz wrote:
please have a look at the attached JSON output of the strcoll benchmark.

Thanks.

   "strcoll": {
    "file list": {
     "directory": "..",
     "file_count": 39086,
     "locale": "en_US.UTF-8",
     "duration": 4.72532e+09,
     "iterations": 16,
     "mean": 2.95333e+08
    },

The "directory" and "file_count" directives are not necessary.  Also,
rename the "file list" key to "" (an empty string) to indicate that
this is the default case, i.e. plain text, which is how the default
benchtest output looks like.

    "word list": {
     "file": "strcoll-inputs/lorem_ipsum_vi_VN",
     "locale": "vi_VN.UTF-8",
     "duration": 4,86973e+07,
     "iterations": 16,
     "mean": 3,04358e+06
    },

Likewise, the "file" directive is not necessary since it is derived
from the locale name.  Instead of "word list", make the locale name as
the key so that you can get rid of the "locale" key.

With the above two changes, your "word list" and "file list" will end
up looking the same barring the extra 'locale' key in "plaintext":

"vi_VN.UTF-8": {
  "duration": 4,86973e+07,
  "iterations": 16,
  "mean": 3,04358e+06
},

"": {
  "locale": "en_US.UTF-8",
  "duration": 4.72532e+09,
  "iterations": 16,
  "mean": 2.95333e+08
},

Siddhesh



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]