public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Noah Goldstein <goldstein.w.n@gmail.com>
To: "H.J. Lu" <hjl.tools@gmail.com>
Cc: GNU C Library <libc-alpha@sourceware.org>,
	"Carlos O'Donell" <carlos@systemhalted.org>
Subject: Re: [PATCH v1 1/5] benchtests: Improve bench-strrchr
Date: Thu, 21 Apr 2022 17:07:51 -0500	[thread overview]
Message-ID: <CAFUsyfJzzBK_PhP5=jcKm=ivtEOgnuCNjPJ96=+U8g2r34fbeQ@mail.gmail.com> (raw)
In-Reply-To: <CAMe9rOqxES1eoHKOmYNLYSK-AKihPC060k=hOuwcWvz7_fkf5g@mail.gmail.com>

On Thu, Apr 21, 2022 at 3:13 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Wed, Apr 20, 2022 at 8:14 PM Noah Goldstein <goldstein.w.n@gmail.com> wrote:
> >
> > 1. Use json-lib for printing results.
> > 2. Expose all parameters (before pos, seek_char, and max_char where
> >    not printed).
> > 3. Add benchmarks that test multiple occurence of seek_char in the
> >    string.
> > ---
> >  benchtests/bench-strrchr.c | 126 ++++++++++++++++++++++++-------------
> >  1 file changed, 82 insertions(+), 44 deletions(-)
> >
> > diff --git a/benchtests/bench-strrchr.c b/benchtests/bench-strrchr.c
> > index abdae60c51..cceea77e1b 100644
> > --- a/benchtests/bench-strrchr.c
> > +++ b/benchtests/bench-strrchr.c
> > @@ -23,6 +23,7 @@
> >  # define TEST_NAME "strrchr"
> >  #endif
> >  #include "bench-string.h"
> > +#include "json-lib.h"
> >
> >  #define BIG_CHAR MAX_CHAR
> >
> > @@ -53,7 +54,8 @@ SIMPLE_STRRCHR (const CHAR *s, int c)
> >  }
> >
> >  static void
> > -do_one_test (impl_t *impl, const CHAR *s, int c, CHAR *exp_res)
> > +do_one_test (json_ctx_t *json_ctx, impl_t *impl, const CHAR *s, int c,
> > +            CHAR *exp_res)
> >  {
> >    CHAR *res = CALL (impl, s, c);
> >    size_t i, iters = INNER_LOOP_ITERS8;
> > @@ -61,8 +63,8 @@ do_one_test (impl_t *impl, const CHAR *s, int c, CHAR *exp_res)
> >
> >    if (res != exp_res)
> >      {
> > -      error (0, 0, "Wrong result in function %s %p %p", impl->name,
> > -            res, exp_res);
> > +      error (0, 0, "Wrong result in function %s %p %p", impl->name, res,
> > +            exp_res);
>
> These changes aren't needed.
>
> >        ret = 1;
> >        return;
> >      }
> > @@ -72,24 +74,28 @@ do_one_test (impl_t *impl, const CHAR *s, int c, CHAR *exp_res)
> >      {
> >        CALL (impl, s, c);
> >      }
> > -  TIMING_NOW (stop);
> >
> > +  TIMING_NOW (stop);
>
> Not needed.

Will fix in V2
>
> >    TIMING_DIFF (cur, start, stop);
> >
> > -  TIMING_PRINT_MEAN ((double) cur, (double) iters);
> > +  json_element_double (json_ctx, (double) cur / (double) iters);
> > +  return;
>
> Return isn't needed.

Will fix in V2.
>
> >  }
> >
> >  static void
> > -do_test (size_t align, size_t pos, size_t len, int seek_char, int max_char)
> > +do_test (json_ctx_t *json_ctx, size_t align, size_t pos, size_t len,
> > +        int seek_char, int max_char, size_t freq)
> >  /* For wcsrchr: align here means align not in bytes,
> >     but in wchar_ts, in bytes it will equal to align * (sizeof (wchar_t))
> >     len for wcschr here isn't in bytes but it's number of wchar_t symbols.  */
> >  {
> >    size_t i;
> > +  size_t pos_chunk_sz = freq ? (pos / freq) : pos;
> > +  size_t last_pos = len;
> >    CHAR *result;
> >    CHAR *buf = (CHAR *) buf1;
> >
> > -  align &= 7;
> > +  align &= (getpagesize () - 1);
>
> If we have such large alignments, the tests may be skipped.
> Should we change it to 127 instead?

There is logic around page cross cases in x86_64 versions so think
makes sense to support benchmarking it.

Also i think that would tend to give the previous version a bit of
an unfair disadvantage as the slow aligning case will never be
tested in the new version.

>
> >    if ((align + len) * sizeof (CHAR) >= page_size)
> >      return;
> >
> > @@ -103,6 +109,16 @@ do_test (size_t align, size_t pos, size_t len, int seek_char, int max_char)
> >        if ((i > pos || pos >= len) && buf[align + i] == seek_char)
> >         buf[align + i] = seek_char + 10 + (random () & 15);
> >      }
> > +
> > +  if (pos_chunk_sz == 0 && pos)
> > +    pos_chunk_sz = 1;
> > +
> > +  for (i = pos_chunk_sz; i < pos && i < len; i += pos_chunk_sz)
> > +    {
> > +      buf[align + i] = seek_char;
> > +      last_pos = i;
> > +    }
> > +
> >    buf[align + len] = 0;
> >
> >    if (pos < len)
> > @@ -110,66 +126,88 @@ do_test (size_t align, size_t pos, size_t len, int seek_char, int max_char)
> >        buf[align + pos] = seek_char;
> >        result = (CHAR *) (buf + align + pos);
> >      }
> > +  else if (last_pos < len)
> > +    result = (CHAR *) (buf + align + last_pos);
> >    else if (seek_char == 0)
> >      result = (CHAR *) (buf + align + len);
> >    else
> >      result = NULL;
> >
> > -  printf ("Length %4zd, alignment in bytes %2zd:", len, align * sizeof (CHAR));
> > +  json_element_object_begin (json_ctx);
> > +  json_attr_uint (json_ctx, "len", len);
> > +  json_attr_uint (json_ctx, "pos", pos);
> > +  json_attr_uint (json_ctx, "align", align);
> > +  json_attr_uint (json_ctx, "freq", freq);
> > +  json_attr_uint (json_ctx, "seek", seek_char);
> > +  json_attr_uint (json_ctx, "max_char", max_char);
> > +  json_array_begin (json_ctx, "timings");
> >
> >    FOR_EACH_IMPL (impl, 0)
> > -    do_one_test (impl, (CHAR *) (buf + align), seek_char, result);
> > +    do_one_test (json_ctx, impl, (CHAR *) (buf + align), seek_char, result);
> >
> > -  putchar ('\n');
> > +  json_array_end (json_ctx);
> > +  json_element_object_end (json_ctx);
> >  }
> >
> >  int
> >  test_main (void)
> >  {
> > -  size_t i;
> > +  json_ctx_t json_ctx;
> > +  size_t i, j;
> > +  int seek;
> >
> >    test_init ();
> > +  json_init (&json_ctx, 0, stdout);
> >
> > -  printf ("%20s", "");
> > -  FOR_EACH_IMPL (impl, 0)
> > -    printf ("\t%s", impl->name);
> > -  putchar ('\n');
> > +  json_document_begin (&json_ctx);
> > +  json_attr_string (&json_ctx, "timing_type", TIMING_TYPE);
> >
> > -  for (i = 1; i < 8; ++i)
> > -    {
> > -      do_test (0, 16 << i, 2048, 23, SMALL_CHAR);
> > -      do_test (i, 16 << i, 2048, 23, SMALL_CHAR);
> > -    }
> > +  json_attr_object_begin (&json_ctx, "functions");
> > +  json_attr_object_begin (&json_ctx, TEST_NAME);
> > +  json_attr_string (&json_ctx, "bench-variant", "");
> >
> > -  for (i = 1; i < 8; ++i)
> > -    {
> > -      do_test (i, 64, 256, 23, SMALL_CHAR);
> > -      do_test (i, 64, 256, 23, BIG_CHAR);
> > -    }
> > -
> > -  for (i = 0; i < 32; ++i)
> > -    {
> > -      do_test (0, i, i + 1, 23, SMALL_CHAR);
> > -      do_test (0, i, i + 1, 23, BIG_CHAR);
> > -    }
> > +  json_array_begin (&json_ctx, "ifuncs");
> > +  FOR_EACH_IMPL (impl, 0)
> > +    json_element_string (&json_ctx, impl->name);
> > +  json_array_end (&json_ctx);
> >
> > -  for (i = 1; i < 8; ++i)
> > -    {
> > -      do_test (0, 16 << i, 2048, 0, SMALL_CHAR);
> > -      do_test (i, 16 << i, 2048, 0, SMALL_CHAR);
> > -    }
> > +  json_array_begin (&json_ctx, "results");
> >
> > -  for (i = 1; i < 8; ++i)
> > +  for (seek = 0; seek <= 23; seek += 23)
> >      {
> > -      do_test (i, 64, 256, 0, SMALL_CHAR);
> > -      do_test (i, 64, 256, 0, BIG_CHAR);
> > +      for (j = 1; j < 32; j += j)
> > +       {
> > +         for (i = 1; i < 9; ++i)
> > +           {
> > +             do_test (&json_ctx, 0, 16 << i, 2048, seek, SMALL_CHAR, j);
> > +             do_test (&json_ctx, i, 16 << i, 2048, seek, SMALL_CHAR, j);
> > +           }
> > +
> > +         for (i = 1; i < 8; ++i)
> > +           {
> > +             do_test (&json_ctx, i, 64, 256, seek, SMALL_CHAR, j);
> > +             do_test (&json_ctx, i, 64, 256, seek, BIG_CHAR, j);
> > +
> > +             do_test (&json_ctx, i * 15, 64, 256, seek, SMALL_CHAR, j);
> > +             do_test (&json_ctx, i * 15, 64, 256, seek, BIG_CHAR, j);
> > +           }
> > +
> > +         for (i = 0; i < 32; ++i)
> > +           {
> > +             do_test (&json_ctx, 0, i, i + 1, seek, SMALL_CHAR, j);
> > +             do_test (&json_ctx, 0, i, i + 1, seek, BIG_CHAR, j);
> > +           }
> > +         if (seek == 0)
> > +           {
> > +             break;
> > +           }
> > +       }
> >      }
> >
> > -  for (i = 0; i < 32; ++i)
> > -    {
> > -      do_test (0, i, i + 1, 0, SMALL_CHAR);
> > -      do_test (0, i, i + 1, 0, BIG_CHAR);
> > -    }
> > +  json_array_end (&json_ctx);
> > +  json_attr_object_end (&json_ctx);
> > +  json_attr_object_end (&json_ctx);
> > +  json_document_end (&json_ctx);
> >
> >    return ret;
> >  }
> > --
> > 2.25.1
> >
>
>
> --
> H.J.

  reply	other threads:[~2022-04-21 22:08 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-21  3:14 Noah Goldstein
2022-04-21  3:14 ` [PATCH v1 2/5] x86: Optimize {str|wcs}rchr-sse2 Noah Goldstein
2022-04-21 20:26   ` H.J. Lu
2022-04-21 20:57     ` Noah Goldstein
2022-04-21 21:48       ` H.J. Lu
2022-04-21 22:23         ` Noah Goldstein
2022-04-21  3:14 ` [PATCH v1 3/5] x86: Add wcsrchr optimized with SSE4_1 in wcsrchr-sse4_1.S Noah Goldstein
2022-04-21  3:14 ` [PATCH v1 4/5] x86: Optimize {str|wcs}rchr-avx2 Noah Goldstein
2022-04-21  3:14 ` [PATCH v1 5/5] x86: Optimize {str|wcs}rchr-evex Noah Goldstein
2022-04-21 20:12 ` [PATCH v1 1/5] benchtests: Improve bench-strrchr H.J. Lu
2022-04-21 22:07   ` Noah Goldstein [this message]
2022-04-21 23:49     ` H.J. Lu
2022-04-22  1:11       ` Noah Goldstein
2022-04-21 22:22 ` [PATCH v2 1/4] " Noah Goldstein
2022-04-21 22:22   ` [PATCH v2 2/4] x86: Optimize {str|wcs}rchr-sse2 Noah Goldstein
2022-04-21 23:46     ` H.J. Lu
2022-04-22  1:54       ` Noah Goldstein
2022-04-21 22:22   ` [PATCH v2 3/4] x86: Optimize {str|wcs}rchr-avx2 Noah Goldstein
2022-04-21 22:22   ` [PATCH v2 4/4] x86: Optimize {str|wcs}rchr-evex Noah Goldstein
2022-04-21 23:59     ` H.J. Lu
2022-04-22  1:53       ` Noah Goldstein
2022-04-22  1:52 ` [PATCH v3 1/4] benchtests: Improve bench-strrchr Noah Goldstein
2022-04-22  1:52   ` [PATCH v3 2/4] x86: Optimize {str|wcs}rchr-sse2 Noah Goldstein
2022-04-22 19:06     ` H.J. Lu
2022-05-12 20:13       ` Sunil Pandey
2022-04-22  1:52   ` [PATCH v3 3/4] x86: Optimize {str|wcs}rchr-avx2 Noah Goldstein
2022-04-22 19:03     ` H.J. Lu
2022-05-12 20:14       ` Sunil Pandey
2022-07-20 15:33         ` Noah Goldstein
2022-04-22  1:52   ` [PATCH v3 4/4] x86: Optimize {str|wcs}rchr-evex Noah Goldstein
2022-04-22 19:04     ` H.J. Lu
2022-05-12 20:16       ` Sunil Pandey
2022-04-22 18:29   ` [PATCH v3 1/4] benchtests: Improve bench-strrchr H.J. Lu
2022-04-22 19:12     ` Noah Goldstein
2022-04-22 19:11 ` [PATCH v4 " Noah Goldstein
2022-04-23  1:53   ` H.J. Lu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAFUsyfJzzBK_PhP5=jcKm=ivtEOgnuCNjPJ96=+U8g2r34fbeQ@mail.gmail.com' \
    --to=goldstein.w.n@gmail.com \
    --cc=carlos@systemhalted.org \
    --cc=hjl.tools@gmail.com \
    --cc=libc-alpha@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).