git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
To: "Carlo Marcelo Arenas Belón" <carenas@gmail.com>
Cc: gitster@pobox.com, git@vger.kernel.org
Subject: Re: [PATCH 2/3] grep: make PCRE2 aware of custom allocator
Date: Thu, 8 Aug 2019 15:56:54 +0200 (CEST)	[thread overview]
Message-ID: <nycvar.QRO.7.76.6.1908081556020.46@tvgsbejvaqbjf.bet> (raw)
In-Reply-To: <20190806085014.47776-3-carenas@gmail.com>

[-- Attachment #1: Type: text/plain, Size: 5228 bytes --]

Hi Carlo,

On Tue, 6 Aug 2019, Carlo Marcelo Arenas Belón wrote:

> Most of the code stolen from[1] to easy on comparison and including
> the deficiency of setting the global context even for patterns that
> won't need it.
>
> Ideally, the call from grep_init could be moved to a place where it
> could be set without needing a lock and at least with this approach
> we have a place to clear it (which is obviously missing more callers,
> but at least shows how it will look for the grep subcommand)
>
> I had also dropped most other users of the global context in a failed
> attempt to make the change smaller, but also to keep the current
> behaviour so that we could see the effect of enabling NED for PCRE2
> more clearly.
>
> Sadly, that will likely require a Windows box, as NED (at least our
> version) is horribly broken in macOS (maybe it wasn't 64 bit clean)
> and in Linux builds, but I can't reproduce your crasher and it is
> most likely slower than the system malloc.
>
> [1] https://public-inbox.org/git/pull.306.git.gitgitgadget@gmail.com/
>
> Suggested-by: Johannes Schindelin <johannes.schindelin@gmx.de>

Actually not so much suggested by me, as your patch still causes
crashes (mine didn't):

https://dev.azure.com/gitgitgadget/git/_build/results?buildId=13935&view=ms.vss-test-web.build-test-results-tab

Ciao,
Dscho

> ---
>  builtin/grep.c |  1 +
>  grep.c         | 31 +++++++++++++++++++++++++++++--
>  grep.h         |  1 +
>  3 files changed, 31 insertions(+), 2 deletions(-)
>
> diff --git a/builtin/grep.c b/builtin/grep.c
> index 560051784e..e49c20df60 100644
> --- a/builtin/grep.c
> +++ b/builtin/grep.c
> @@ -1145,5 +1145,6 @@ int cmd_grep(int argc, const char **argv, const char *prefix)
>  		run_pager(&opt, prefix);
>  	clear_pathspec(&pathspec);
>  	free_grep_patterns(&opt);
> +	grep_destroy();
>  	return !hit;
>  }
> diff --git a/grep.c b/grep.c
> index 0154998695..e748a6d68c 100644
> --- a/grep.c
> +++ b/grep.c
> @@ -16,6 +16,20 @@ static int grep_source_is_binary(struct grep_source *gs,
>
>  static struct grep_opt grep_defaults;
>
> +#ifdef USE_LIBPCRE2
> +static pcre2_general_context *pcre2_global_context;
> +
> +static void *pcre2_malloc(PCRE2_SIZE size, void *memory_data)
> +{
> +	return malloc(size);
> +}
> +
> +static void pcre2_free(void *pointer, void *memory_data)
> +{
> +	return free(pointer);
> +}
> +#endif
> +
>  static const char *color_grep_slots[] = {
>  	[GREP_COLOR_CONTEXT]	    = "context",
>  	[GREP_COLOR_FILENAME]	    = "filename",
> @@ -153,6 +167,7 @@ int grep_config(const char *var, const char *value, void *cb)
>   *
>   * If using PCRE make sure that the library is configured
>   * to use the right allocator (ex: NED)
> + * if any object is created it should be cleaned up in grep_destroy()
>   */
>  void grep_init(struct grep_opt *opt, struct repository *repo, const char *prefix)
>  {
> @@ -164,6 +179,10 @@ void grep_init(struct grep_opt *opt, struct repository *repo, const char *prefix
>  	pcre_malloc = malloc;
>  	pcre_free = free;
>  #endif
> +#ifdef USE_LIBPCRE2
> +	pcre2_global_context = pcre2_general_context_create(pcre2_malloc,
> +							pcre2_free, NULL);
> +#endif
>  #endif
>
>  	memset(opt, 0, sizeof(*opt));
> @@ -188,6 +207,13 @@ void grep_init(struct grep_opt *opt, struct repository *repo, const char *prefix
>  		color_set(opt->colors[i], def->colors[i]);
>  }
>
> +void grep_destroy(void)
> +{
> +#ifdef USE_LIBPCRE2
> +	pcre2_general_context_free(pcre2_global_context);
> +#endif
> +}
> +
>  static void grep_set_pattern_type_option(enum grep_pattern_type pattern_type, struct grep_opt *opt)
>  {
>  	/*
> @@ -509,7 +535,7 @@ static void compile_pcre2_pattern(struct grep_pat *p, const struct grep_opt *opt
>
>  	if (opt->ignore_case) {
>  		if (has_non_ascii(p->pattern)) {
> -			character_tables = pcre2_maketables(NULL);
> +			character_tables = pcre2_maketables(pcre2_global_context);
>  			p->pcre2_compile_context = pcre2_compile_context_create(NULL);
>  			pcre2_set_character_tables(p->pcre2_compile_context, character_tables);
>  		}
> @@ -560,7 +586,8 @@ static void compile_pcre2_pattern(struct grep_pat *p, const struct grep_opt *opt
>  			return;
>  		}
>
> -		p->pcre2_jit_stack = pcre2_jit_stack_create(1, 1024 * 1024, NULL);
> +		p->pcre2_jit_stack = pcre2_jit_stack_create(1, 1024 * 1024,
> +					pcre2_global_context);
>  		if (!p->pcre2_jit_stack)
>  			die("Couldn't allocate PCRE2 JIT stack");
>  		p->pcre2_match_context = pcre2_match_context_create(NULL);
> diff --git a/grep.h b/grep.h
> index 1875880f37..526c2db9ef 100644
> --- a/grep.h
> +++ b/grep.h
> @@ -189,6 +189,7 @@ struct grep_opt {
>  void init_grep_defaults(struct repository *);
>  int grep_config(const char *var, const char *value, void *);
>  void grep_init(struct grep_opt *, struct repository *repo, const char *prefix);
> +void grep_destroy(void);
>  void grep_commit_pattern_type(enum grep_pattern_type, struct grep_opt *opt);
>
>  void append_grep_pat(struct grep_opt *opt, const char *pat, size_t patlen, const char *origin, int no, enum grep_pat_token t);
> --
> 2.23.0.rc1
>
>

  reply	other threads:[~2019-08-08 13:57 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-05 11:51 [PATCH 0/1] Fix a problem with PCRE2 and nedmalloc, found via Azure Pipelines Johannes Schindelin via GitGitGadget
2019-08-05 11:51 ` [PATCH 1/1] pcre2: allow overriding the system allocator Johannes Schindelin via GitGitGadget
2019-08-05 16:19   ` Carlo Arenas
2019-08-05 16:27     ` Carlo Arenas
2019-08-05 19:32       ` Johannes Schindelin
2019-08-05 19:26     ` Johannes Schindelin
2019-08-05 21:53     ` Junio C Hamano
2019-08-06  6:24       ` Carlo Arenas
2019-08-06  8:50       ` [PATCH 0/3] grep: no leaks (WIP) Carlo Marcelo Arenas Belón
2019-08-06  8:50         ` [PATCH 1/3] grep: make PCRE1 aware of custom allocator Carlo Marcelo Arenas Belón
2019-08-08 13:54           ` Johannes Schindelin
2019-08-08 15:19             ` Carlo Arenas
2019-08-06  8:50         ` [PATCH 2/3] grep: make PCRE2 " Carlo Marcelo Arenas Belón
2019-08-08 13:56           ` Johannes Schindelin [this message]
2019-08-08 14:32             ` Carlo Arenas
2019-08-06  8:50         ` [PATCH 3/3] grep: avoid leak of chartables in PCRE2 Carlo Marcelo Arenas Belón
2019-08-06 16:36         ` [RFC PATCH v3 0/3] grep: no leaks or crashes (windows testing needed) Carlo Marcelo Arenas Belón
2019-08-06 16:36           ` [RFC PATCH v3 1/3] grep: make PCRE1 aware of custom allocator Carlo Marcelo Arenas Belón
2019-08-06 16:36           ` [RFC PATCH v3 2/3] grep: make PCRE2 " Carlo Marcelo Arenas Belón
2019-08-07  5:38             ` René Scharfe
2019-08-07  9:49               ` Carlo Arenas
2019-08-07 13:02                 ` René Scharfe
2019-08-07 13:08                   ` [PATCH 1/2] nedmalloc: do assignments only after the declaration section René Scharfe
2019-08-07 13:09                   ` [PATCH 2/2] nedmalloc: avoid compiler warning about unused value René Scharfe
2019-08-08  2:35                   ` [RFC PATCH v3 2/3] grep: make PCRE2 aware of custom allocator Carlo Arenas
2019-08-08  7:07                     ` René Scharfe
2019-08-08 12:38                       ` Carlo Arenas
2019-08-08 14:29                         ` René Scharfe
2019-08-08 20:18                           ` Johannes Schindelin
2019-08-07 18:15                 ` Junio C Hamano
2019-08-06 16:36           ` [RFC PATCH v3 3/3] grep: avoid leak of chartables in PCRE2 Carlo Marcelo Arenas Belón
2019-08-06 16:48           ` [RFC PATCH v3 0/3] grep: no leaks or crashes (windows testing needed) Junio C Hamano
2019-08-07 21:39           ` [RFC PATCH v4 " Carlo Marcelo Arenas Belón
2019-08-07 21:39             ` [RFC PATCH v4 1/3] grep: make PCRE1 aware of custom allocator Carlo Marcelo Arenas Belón
2019-08-07 21:39             ` [RFC PATCH v4 2/3] grep: make PCRE2 " Carlo Marcelo Arenas Belón
2019-08-07 22:28               ` Junio C Hamano
2019-08-07 21:39             ` [RFC PATCH v4 3/3] grep: avoid leak of chartables in PCRE2 Carlo Marcelo Arenas Belón
2019-08-09  3:02             ` [RFC PATCH v5 0/3] grep: almost no more leaks, hopefully no crashes Carlo Marcelo Arenas Belón
2019-08-09  3:02               ` [RFC PATCH v5 1/3] grep: make PCRE1 aware of custom allocator Carlo Marcelo Arenas Belón
2019-08-09  3:02               ` [RFC PATCH v5 2/3] grep: make PCRE2 " Carlo Marcelo Arenas Belón
2019-08-27  9:07                 ` Johannes Schindelin
2019-08-27 11:51                   ` Carlo Arenas
2019-10-03  5:02                     ` Junio C Hamano
2019-10-03  8:08                       ` Johannes Schindelin
2019-10-03 11:17                         ` Carlo Arenas
2019-10-03 18:23                           ` Johannes Schindelin
2019-10-03 22:57                           ` Junio C Hamano
2019-08-09  3:02               ` [RFC PATCH v5 3/3] grep: avoid leak of chartables in PCRE2 Carlo Marcelo Arenas Belón
2019-08-09 11:24               ` [RFC PATCH v5 0/3] grep: almost no more leaks, hopefully no crashes Carlo Arenas
2019-08-09 17:01                 ` René Scharfe
2019-08-09 17:46                   ` Junio C Hamano
2019-08-09 21:26                   ` Johannes Schindelin
2019-08-10  3:03                     ` [PATCH] SQUASH Carlo Marcelo Arenas Belón
2019-08-10  7:57                       ` René Scharfe
2019-08-10  8:42                         ` Carlo Arenas
2019-08-10 19:47                           ` René Scharfe
2019-08-12  7:35                             ` Carlo Arenas
2019-08-12 12:14                               ` René Scharfe
2019-08-12 12:28                                 ` Carlo Arenas
2019-08-10 13:57                       ` Johannes Schindelin
2019-08-10  3:05                     ` [RFC PATCH v5 0/3] grep: almost no more leaks, hopefully no crashes Carlo Arenas
2019-08-10  7:56                       ` René Scharfe
2019-08-10 12:40                         ` Carlo Arenas
2019-08-10 21:16                           ` René Scharfe
2019-08-08 20:21           ` [RFC PATCH v3 0/3] grep: no leaks or crashes (windows testing needed) Johannes Schindelin
2019-08-09  6:52             ` Carlo Arenas
2019-08-09 21:13               ` Johannes Schindelin
  -- strict thread matches above, loose matches on Subject: below --
2019-10-16 12:06 [PATCH 0/3] Revive 'pcre2-chartables-leakfix' Johannes Schindelin via GitGitGadget
2019-10-16 12:06 ` [PATCH 2/3] grep: make PCRE2 aware of custom allocator Carlo Marcelo Arenas Belón via GitGitGadget

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

  List information: http://vger.kernel.org/majordomo-info.html

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

  git send-email \
    --in-reply-to=nycvar.QRO.7.76.6.1908081556020.46@tvgsbejvaqbjf.bet \
    --to=johannes.schindelin@gmx.de \
    --cc=carenas@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    /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.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

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).