From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lino Sanfilippo Subject: Re: [PATCH 1/3] fanotify: only destroy mark when both mask and ignored_mask are cleared Date: Tue, 02 Dec 2014 03:11:01 +0100 Message-ID: <547D1FB5.80809@gmx.de> References: <1417304258-16838-1-git-send-email-LinoSanfilippo@gmx.de> <20141201090450.GB16185@quack.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Cc: eparis@redhat.com, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org To: Jan Kara Return-path: In-Reply-To: <20141201090450.GB16185@quack.suse.cz> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org Hi, On 01.12.2014 10:04, Jan Kara wrote: > On Sun 30-11-14 00:37:36, Lino Sanfilippo wrote: >> In fanotify_mark_remove_from_mask() a mark is destroyed if only one of both >> bitmasks (mask or ignored_mask) of a mark is cleared. However the other mask >> may still be set and contain information that should not be lost. Thus only >> destroy a mark if both masks are cleared. >> >> Signed-off-by: Lino Sanfilippo >> --- >> fs/notify/fanotify/fanotify_user.c | 6 +++++- >> 1 file changed, 5 insertions(+), 1 deletion(-) >> >> diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c >> index c991616..03a0dd1 100644 >> --- a/fs/notify/fanotify/fanotify_user.c >> +++ b/fs/notify/fanotify/fanotify_user.c >> @@ -488,6 +488,8 @@ static __u32 fanotify_mark_remove_from_mask(struct fsnotify_mark *fsn_mark, >> int *destroy) >> { >> __u32 oldmask; >> + __u32 new_mask; >> + __u32 new_ignored; >> >> spin_lock(&fsn_mark->lock); >> if (!(flags & FAN_MARK_IGNORED_MASK)) { >> @@ -497,9 +499,11 @@ static __u32 fanotify_mark_remove_from_mask(struct fsnotify_mark *fsn_mark, >> oldmask = fsn_mark->ignored_mask; >> fsnotify_set_mark_ignored_mask_locked(fsn_mark, (oldmask & ~mask)); >> } >> + new_mask = fsn_mark->mask; >> + new_ignored = fsn_mark->ignored_mask; >> spin_unlock(&fsn_mark->lock); >> >> - *destroy = !(oldmask & ~mask); >> + *destroy = !(new_mask | new_ignored); > There's no need for new variables, is there? You can just set *destroy > under the spinlock... > youre right, these variables are totally unneeded. I cant remember the reason why i introduced them (maybe leftovers from an earlier attempt). I will resend a cleaned up version of that patch. Regards, Lino