public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Richard Biener <richard.guenther@gmail.com>
To: marxin <mliska@suse.cz>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH 2/3] Smarter folding of __builtin_memchr
Date: Wed, 17 Aug 2016 08:41:00 -0000	[thread overview]
Message-ID: <CAFiYyc3QxgJgo4z0hFcz1U2i6FShOm33GH5j8EOrpL+t0nLGPg@mail.gmail.com> (raw)
In-Reply-To: <2f0f974b776ee917d094b78cd9b77b992988499c.1471416736.git.mliska@suse.cz>

On Tue, Aug 16, 2016 at 3:14 PM, marxin <mliska@suse.cz> wrote:
> gcc/ChangeLog:
>
> 2016-08-16  Martin Liska  <mliska@suse.cz>
>
>         * builtins.c (fold_builtin_memchr): Support following
>         transformations:
>         memchr (x, y, 0) -> NULL
>         memchr ("known_string", 'n', 5) -> &"known_string" + 1
>         memchr ("known_string", 'n', 1) -> NULL

Aww.  Can I convince you to move this and related foldings to gimple-fold.c?

Richard.

> ---
>  gcc/builtins.c | 20 ++++++++++++++++----
>  1 file changed, 16 insertions(+), 4 deletions(-)
>
> diff --git a/gcc/builtins.c b/gcc/builtins.c
> index 8f1c752..ac251f8 100644
> --- a/gcc/builtins.c
> +++ b/gcc/builtins.c
> @@ -7267,8 +7267,12 @@ fold_builtin_memchr (location_t loc, tree arg1, tree arg2, tree len, tree type)
>           || !tree_fits_uhwi_p (len))
>         return NULL_TREE;
>
> +      /* If the LEN parameter is zero, return zero.  */
> +      if (integer_zerop (len))
> +       return build_int_cst (TREE_TYPE (arg1), 0);
> +
>        p1 = c_getstr (arg1);
> -      if (p1 && compare_tree_int (len, strlen (p1) + 1) <= 0)
> +      if (p1)
>         {
>           char c;
>           const char *r;
> @@ -7281,9 +7285,17 @@ fold_builtin_memchr (location_t loc, tree arg1, tree arg2, tree len, tree type)
>
>           if (r == NULL)
>             return build_int_cst (TREE_TYPE (arg1), 0);
> -
> -         tem = fold_build_pointer_plus_hwi_loc (loc, arg1, r - p1);
> -         return fold_convert_loc (loc, type, tem);
> +         else

the else seems superfluous and just adds to indentation.

> +           {
> +             size_t offset = r - p1;
> +             if (compare_tree_int (len, offset) <= 0)
> +               return build_int_cst (TREE_TYPE (arg1), 0);
> +             else
> +               {
> +                 tem = fold_build_pointer_plus_hwi_loc (loc, arg1, offset);
> +                 return fold_convert_loc (loc, type, tem);
> +               }
> +           }
>         }
>        return NULL_TREE;
>      }
> --
> 2.9.2
>
>

  reply	other threads:[~2016-08-17  8:41 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-17  6:55 [PATCH 0/3] Better folding of 2 string builtin-ins marxin
2016-08-17  6:55 ` [PATCH 2/3] Smarter folding of __builtin_memchr marxin
2016-08-17  8:41   ` Richard Biener [this message]
2016-10-07  8:41     ` [PATCH 2/3] Fold __builtin_memchr (version 2) Martin Liška
2016-10-07 11:01       ` Richard Biener
2016-10-11  9:38         ` [PATCH] Fold __builtin_memchr (version 3) Martin Liška
2016-10-12  8:34           ` Richard Biener
2016-10-12  8:36           ` Richard Biener
2016-10-12 13:19             ` Martin Liška
2016-10-13 15:26               ` [PATCH] Fold __builtin_memchr (simplified version 4) Martin Liška
2016-10-14 11:57                 ` Richard Biener
2016-08-17  6:55 ` [PATCH 3/3] Test folding of strn{case}cmp and memchr marxin
2016-08-17  7:52   ` Martin Liška
2016-10-07  8:42     ` [PATCH 3/3] Test folding of str{n}{case}cmp and memchr (version 2) Martin Liška
2016-10-11  9:39       ` [PATCH] Test folding of str{n}{case}cmp and memchr (version 3) Martin Liška
2016-10-12  8:35         ` Richard Biener
2016-10-12 13:20           ` Martin Liška
2016-10-13 15:27             ` [PATCH] Test folding of str{n}{case}cmp and memchr (simplified version 4) Martin Liška
2016-10-14 11:58               ` Richard Biener
2016-08-17  6:55 ` [PATCH 1/3] Fold BUILT_IN_STRNCASECMP marxin
2016-08-17  7:10   ` Jakub Jelinek
2016-08-17  7:52     ` Martin Liška
2016-10-07  8:39   ` [PATCH 1/3] Fold __builtin_str{n}{case}cmp functions (version 2) Martin Liška
2016-10-07 10:50     ` Richard Biener
2016-10-11  9:26       ` Martin Liška
2016-10-11 10:27         ` Richard Biener
2016-10-11  9:28       ` [PATCH] Add a helper function: create_tmp Martin Liška
2016-10-11 10:30         ` Richard Biener
2016-10-11 10:31           ` Richard Biener
2016-10-12 10:50             ` Martin Liška
2016-10-11  9:28       ` [PATCH] Check \0-termination of string in c_getstr Martin Liška
2016-10-11 10:28         ` Richard Biener
2016-10-12 13:14           ` Martin Liška
2016-10-13 15:24             ` [PATCH] Check \0-termination of string in c_getstr (simplified version) Martin Liška
2016-10-14  9:38               ` Richard Biener
2016-10-14 11:10                 ` Martin Liška
2016-10-11  9:34       ` [PATCH] Fold __builtin_str{n}{case}cmp functions (version 3) Martin Liška
2016-10-12  8:30         ` Richard Biener
2016-10-12 13:17           ` Martin Liška
2016-10-13 15:25             ` [PATCH] Fold __builtin_str{n}{case}cmp functions (simplified version 4) Martin Liška
2016-10-14 11:55               ` Richard Biener
2016-10-12 13:48 ` [RFC] Possible folding opportunities for string built-ins Martin Liška
2016-10-12 15:55   ` Joseph Myers
2016-10-12 19:45     ` Jim Wilson
2016-10-13  8:38   ` Richard Biener

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=CAFiYyc3QxgJgo4z0hFcz1U2i6FShOm33GH5j8EOrpL+t0nLGPg@mail.gmail.com \
    --to=richard.guenther@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=mliska@suse.cz \
    /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).