All of lore.kernel.org
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: "Erik Elfström" <erik.elfstrom@gmail.com>
Cc: git@vger.kernel.org
Subject: Re: [PATCH v4 2/5] setup: sanity check file size in read_gitfile_gently
Date: Sat, 25 Apr 2015 09:47:51 -0700	[thread overview]
Message-ID: <xmqqa8xw17ew.fsf@gitster.dls.corp.google.com> (raw)
In-Reply-To: <1429952801-2646-3-git-send-email-erik.elfstrom@gmail.com> ("Erik =?utf-8?Q?Elfstr=C3=B6m=22's?= message of "Sat, 25 Apr 2015 11:06:38 +0200")

Erik Elfström <erik.elfstrom@gmail.com> writes:

> read_gitfile_gently will allocate a buffer to fit the entire file that
> should be read. Add a sanity check of the file size before opening to
> avoid allocating a potentially huge amount of memory if we come across
> a large file that someone happened to name ".git". The limit is set to
> a sufficiently unreasonable size that should never be exceeded by a
> genuine .git file.
>
> Signed-off-by: Erik Elfström <erik.elfstrom@gmail.com>
> ---
>
> I'm not sure about this one but it felt like the safe thing to do.
> This patch can be dropped if it is not desired.

I do not think it is wrong per-se, but the changes in this patch
shows why hardcoded values assigned to error_code without #define is
not a good idea, as these values are now exposed to the callers of
the new function.  After we gain a new caller that does care about
the exact error code (e.g. to react differently to the reason why we
failed to read by giving different error messages) if we decide to
revert this step, or if we decide to add a new error type, for
whatever reason, this organization forces the caller to be updated.

> I considered testing it using
>  "mkdir foo && truncate -s 200G foo/.git && git clean -f -d"
> but that feels like a pretty evil test that is likely to cause lots
> of problems and not fail in any good way.

Amen to that.

>
>  setup.c | 20 +++++++++++++-------
>  1 file changed, 13 insertions(+), 7 deletions(-)
>
> diff --git a/setup.c b/setup.c
> index e1897cc..ed87334 100644
> --- a/setup.c
> +++ b/setup.c
> @@ -364,22 +364,26 @@ const char *read_gitfile_gently(const char *path, int *return_error_code)
>  		error_code = 3;
>  		goto cleanup_return;
>  	}
> +	if (st.st_size > PATH_MAX * 4) {
> +		error_code = 4;
> +		goto cleanup_return;
> +	}
>  	buf = xmalloc(st.st_size + 1);
>  	len = read_in_full(fd, buf, st.st_size);
>  	close(fd);
>  	if (len != st.st_size) {
> -		error_code = 4;
> +		error_code = 5;
>  		goto cleanup_return;
>  	}
>  	buf[len] = '\0';
>  	if (!starts_with(buf, "gitdir: ")) {
> -		error_code = 5;
> +		error_code = 6;
>  		goto cleanup_return;
>  	}
>  	while (buf[len - 1] == '\n' || buf[len - 1] == '\r')
>  		len--;
>  	if (len < 9) {
> -		error_code = 6;
> +		error_code = 7;
>  		goto cleanup_return;
>  	}
>  	buf[len] = '\0';
> @@ -397,7 +401,7 @@ const char *read_gitfile_gently(const char *path, int *return_error_code)
>  	}
>  
>  	if (!is_git_directory(dir)) {
> -		error_code = 7;
> +		error_code = 8;
>  		goto cleanup_return;
>  	}
>  	path = real_path(dir);
> @@ -419,12 +423,14 @@ cleanup_return:
>  		case 3:
>  			die_errno("Error opening '%s'", path);
>  		case 4:
> -			die("Error reading %s", path);
> +			die("Too large to be a .git file: '%s'", path);
>  		case 5:
> -			die("Invalid gitfile format: %s", path);
> +			die("Error reading %s", path);
>  		case 6:
> -			die("No path in gitfile: %s", path);
> +			die("Invalid gitfile format: %s", path);
>  		case 7:
> +			die("No path in gitfile: %s", path);
> +		case 8:
>  			die("Not a git repository: %s", dir);
>  		default:
>  			assert(0);

  reply	other threads:[~2015-04-25 16:48 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-25  9:06 [PATCH v4 0/5] Improving performance of git clean Erik Elfström
2015-04-25  9:06 ` [PATCH v4 1/5] setup: add gentle version of read_gitfile Erik Elfström
2015-04-25 16:51   ` Junio C Hamano
2015-04-25 16:54     ` Junio C Hamano
2015-04-25  9:06 ` [PATCH v4 2/5] setup: sanity check file size in read_gitfile_gently Erik Elfström
2015-04-25 16:47   ` Junio C Hamano [this message]
2015-04-25 17:59     ` Erik Elfström
2015-04-26  4:29       ` Junio C Hamano
2015-04-26  6:49         ` [PATCH v5 0/5] Improving performance of git clean Erik Elfström
2015-04-26  6:49           ` [PATCH v5 1/5] setup: add gentle version of read_gitfile Erik Elfström
2015-04-28  6:17             ` Jeff King
2015-04-28 20:07               ` erik elfström
2015-04-28 20:19                 ` Jeff King
2015-04-28 20:34                   ` Jonathan Nieder
2015-04-28 20:36                     ` Jeff King
2015-04-28 20:42                       ` Jonathan Nieder
2015-04-28 20:48                         ` Jeff King
2015-04-28 21:06                           ` Jonathan Nieder
2015-04-28 23:34                           ` Junio C Hamano
2015-04-29 23:47             ` Stefan Beller
2015-04-30  1:35               ` Junio C Hamano
2015-04-26  6:49           ` [PATCH v5 2/5] setup: sanity check file size in read_gitfile_gently Erik Elfström
2015-04-28  6:02             ` Jeff King
2015-04-28  7:21               ` Windows path limites, was " Johannes Schindelin
2015-04-28 15:33                 ` Doug Kelly
2015-04-28 16:20                   ` Windows path limits, " Johannes Schindelin
2015-04-28 19:28               ` erik elfström
2015-04-29 15:42             ` Junio C Hamano
2015-04-26  6:49           ` [PATCH v5 3/5] t7300: add tests to document behavior of clean and nested git Erik Elfström
2015-04-26  6:49           ` [PATCH v5 4/5] p7300: add performance tests for clean Erik Elfström
2015-04-28  6:33             ` Jeff King
2015-04-28 19:36               ` erik elfström
2015-04-26  6:49           ` [PATCH v5 5/5] clean: improve performance when removing lots of directories Erik Elfström
2015-04-28  6:24             ` Jeff King
2015-04-28 20:31               ` erik elfström
2015-04-25  9:06 ` [PATCH v4 3/5] t7300: add tests to document behavior of clean and nested git Erik Elfström
2015-04-25  9:06 ` [PATCH v4 4/5] p7300: add performance tests for clean Erik Elfström
2015-04-25  9:06 ` [PATCH v4 5/5] clean: improve performance when removing lots of directories Erik Elfström

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=xmqqa8xw17ew.fsf@gitster.dls.corp.google.com \
    --to=gitster@pobox.com \
    --cc=erik.elfstrom@gmail.com \
    --cc=git@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.