From: Junio C Hamano <gitster@pobox.com>
To: Johan Herland <johan@herland.net>
Cc: Johannes Sixt <j.sixt@viscovery.net>, git@vger.kernel.org
Subject: Re: [PATCH 2/2] Resolve double chmod() in move_temp_to_file()
Date: Fri, 27 Mar 2009 23:21:00 -0700 [thread overview]
Message-ID: <7viqlu1837.fsf@gitster.siamese.dyndns.org> (raw)
In-Reply-To: 200903261617.48362.johan@herland.net
Johan Herland <johan@herland.net> writes:
> diff --git a/sha1_file.c b/sha1_file.c
> index 87ac53b..05af3c5 100644
> --- a/sha1_file.c
> +++ b/sha1_file.c
> @@ -2252,7 +2252,7 @@ int move_temp_to_file(const char *tmpfile, const char *filename)
> /* FIXME!!! Collision check here ? */
> }
>
> - if (chmod(filename, 0444) || adjust_shared_perm(filename))
> + if (chmod(filename, get_shared_perm(0444)))
Your get_shared_perm() will end up feeding 0444 to S_ISDIR(), which would
most likely say "no" and cause real harm, but there is no guarantee that
we won't start checking S_ISREG() or other things in get_shared_perm()
later. I do not like this.
How about doing it this way instead?
One thing to note is that we seem to have been passing what we read from
st.st_mode, together with S_IFMT bits, to chmod(2); I do not think I've
seen any breakage reports on exotic systems (glibc on Linux seems to
ignore the higher bits), but from my reading of POSIX, I would not be
surprised if somebody's chmod(2) returned EINVAL.
-- >8 --
set_shared_perm(): sometimes we know what the final mode bits should look like
adjust_shared_perm() first obtains the mode bits from lstat(2), expecting
to find what the result of applying user's umask is, and then tweaked it
as necessary. When the file to be adjusted is created with mkstemp(3),
however, the mode thusly obtained does not have anything to do with usre's
umask, and we would need to start from 0444 in such a case and there is no
point running lstat(2) for such a path.
This introduces a new API set_shared_perm() to bypass the lstat(2) and
instead force setting the mode bits to the desired value directly.
adjust_shared_perm() becomes a thin wrapper to the function.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
cache.h | 3 ++-
path.c | 25 ++++++++++++++++---------
sha1_file.c | 4 ++--
3 files changed, 20 insertions(+), 12 deletions(-)
diff --git a/cache.h b/cache.h
index 189151d..e283bbe 100644
--- a/cache.h
+++ b/cache.h
@@ -613,7 +613,8 @@ enum sharedrepo {
PERM_EVERYBODY = 0664,
};
int git_config_perm(const char *var, const char *value);
-int adjust_shared_perm(const char *path);
+int set_shared_perm(const char *path, int mode);
+#define adjust_shared_perm(path) set_shared_perm((path), 0)
int safe_create_leading_directories(char *path);
int safe_create_leading_directories_const(const char *path);
char *enter_repo(char *path, int strict);
diff --git a/path.c b/path.c
index 42898e0..8a0a674 100644
--- a/path.c
+++ b/path.c
@@ -311,16 +311,23 @@ char *enter_repo(char *path, int strict)
return NULL;
}
-int adjust_shared_perm(const char *path)
+int set_shared_perm(const char *path, int mode)
{
struct stat st;
- int mode, tweak, shared;
+ int tweak, shared, orig_mode;
- if (!shared_repository)
+ if (!shared_repository) {
+ if (mode)
+ return chmod(path, mode & ~S_IFMT);
return 0;
- if (lstat(path, &st) < 0)
- return -1;
- mode = st.st_mode;
+ }
+ if (!mode) {
+ if (lstat(path, &st) < 0)
+ return -1;
+ mode = st.st_mode;
+ orig_mode = mode;
+ } else
+ orig_mode = 0;
if (shared_repository < 0)
shared = -shared_repository;
else
@@ -344,9 +351,9 @@ int adjust_shared_perm(const char *path)
}
if (((shared_repository < 0
- ? (st.st_mode & (FORCE_DIR_SET_GID | 0777))
- : (st.st_mode & mode)) != mode) &&
- chmod(path, mode) < 0)
+ ? (orig_mode & (FORCE_DIR_SET_GID | 0777))
+ : (orig_mode & mode)) != mode) &&
+ chmod(path, (mode & ~S_IFMT)) < 0)
return -2;
return 0;
}
diff --git a/sha1_file.c b/sha1_file.c
index 8869488..5bfc36c 100644
--- a/sha1_file.c
+++ b/sha1_file.c
@@ -2263,7 +2263,7 @@ int move_temp_to_file(const char *tmpfile, const char *filename)
*
* The same holds for FAT formatted media.
*
- * When this succeeds, we just return 0. We have nothing
+ * When this succeeds, we just return; we have nothing
* left to unlink.
*/
if (ret && ret != EEXIST) {
@@ -2280,7 +2280,7 @@ int move_temp_to_file(const char *tmpfile, const char *filename)
}
out:
- if (chmod(filename, 0444) || adjust_shared_perm(filename))
+ if (set_shared_perm(filename, (S_IFREG|0444)))
return error("unable to set permission to '%s'", filename);
return 0;
}
next prev parent reply other threads:[~2009-03-28 6:22 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-03-25 0:05 [BUG?] How to make a shared/restricted repo? Johan Herland
2009-03-25 0:26 ` Brandon Casey
2009-03-25 0:45 ` Johan Herland
2009-03-25 0:49 ` Junio C Hamano
2009-03-25 0:46 ` Junio C Hamano
2009-03-25 2:11 ` Johan Herland
2009-03-25 2:24 ` Junio C Hamano
2009-03-25 21:36 ` [PATCH/RFC 0/7] Restricting repository access (Was: [BUG?] How to make a shared/restricted repo?) Johan Herland
2009-03-25 21:37 ` [PATCH/RFC 1/7] Clarify documentation on permissions in shared repositories Johan Herland
2009-03-25 21:38 ` [PATCH/RFC 2/7] Cleanup: Remove unnecessary if-else clause Johan Herland
2009-03-25 21:39 ` [PATCH/RFC 3/7] Introduce core.restrictedRepository for restricting repository permissions Johan Herland
2009-03-25 21:39 ` [PATCH/RFC 4/7] git-init: Introduce --restricted for restricting repository access Johan Herland
2009-03-25 21:40 ` [PATCH/RFC 5/7] Add tests for "core.restrictedRepository" and "git init --restricted" Johan Herland
2009-03-25 21:41 ` [PATCH/RFC 6/7] git-init: Apply correct mode bits to template files in shared/restricted repo Johan Herland
2009-03-25 21:42 ` [PATCH/RFC 7/7] Apply restricted permissions to loose objects and pack files Johan Herland
2009-03-25 23:19 ` [BUG?] How to make a shared/restricted repo? Junio C Hamano
2009-03-26 0:22 ` Johan Herland
2009-03-26 7:23 ` Junio C Hamano
2009-03-26 8:29 ` Johan Herland
2009-03-26 8:41 ` Johannes Sixt
2009-03-26 9:44 ` Johan Herland
2009-03-26 9:58 ` Johannes Sixt
2009-03-26 15:02 ` [PATCH 0/2] chmod cleanup (Was: [BUG?] How to make a shared/restricted repo?) Johan Herland
2009-03-26 15:16 ` [PATCH 1/2] Move chmod(foo, 0444) into move_temp_to_file() Johan Herland
2009-03-28 6:14 ` Junio C Hamano
2009-03-28 10:48 ` Johan Herland
2009-03-26 15:17 ` [PATCH 2/2] Resolve double chmod() in move_temp_to_file() Johan Herland
2009-03-28 6:21 ` Junio C Hamano [this message]
2009-03-28 11:01 ` Johan Herland
2009-03-29 20:31 ` Junio C Hamano
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=7viqlu1837.fsf@gitster.siamese.dyndns.org \
--to=gitster@pobox.com \
--cc=git@vger.kernel.org \
--cc=j.sixt@viscovery.net \
--cc=johan@herland.net \
/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 NNTP newsgroup(s).