From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41313) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1btXiC-0006P9-Ds for qemu-devel@nongnu.org; Mon, 10 Oct 2016 06:21:37 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1btXi8-0002EL-80 for qemu-devel@nongnu.org; Mon, 10 Oct 2016 06:21:35 -0400 Received: from 7.mo6.mail-out.ovh.net ([46.105.59.196]:51392) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1btXi8-0002E3-1f for qemu-devel@nongnu.org; Mon, 10 Oct 2016 06:21:32 -0400 Received: from player738.ha.ovh.net (b7.ovh.net [213.186.33.57]) by mo6.mail-out.ovh.net (Postfix) with ESMTP id EFF793C412 for ; Mon, 10 Oct 2016 12:21:30 +0200 (CEST) Date: Mon, 10 Oct 2016 12:21:22 +0200 From: Greg Kurz Message-ID: <20161010122122.71f57327@bahia> In-Reply-To: References: <648301670b2dc88aec68e48abd1bef07c1e201a3.1475990063.git.liqiang6-s@360.cn> <20161010110611.385ef476@bahia> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 2/2] 9pfs: fix memory leak about xattr value List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Li Qiang Cc: qemu-devel@nongnu.org, Li Qiang On Mon, 10 Oct 2016 17:15:10 +0800 Li Qiang wrote: > Oops! I have made this patch to dress the memory leak in v9fs_xattrcreate I guess you mean v9fs_xattrwalk(). There is no possible memory leak with v9fs_xattrwalk() because xattr_fidp comes from alloc_fid(). The leak can only occur when xattr_fidp comes from get_fid(). > but forget this and make another mistake. > > I mean we should call g_free for xattr.value first in xattr create. > > This avoid the memory leak. > > static void v9fs_xattrcreate(void *opaque) > { > ... > g_free(xattr_fidp->fs.xattr.value); > xattr_fidp->fs.xattr.value = g_malloc(size); > err = offset; > > } > Indeed we have a potential leak with v9fs_xattrcreate(). > Should I resend this path series? > No, you just need to send a patch that fixes the leak in v9fs_xattrcreate(). > 2016-10-10 17:06 GMT+08:00 Greg Kurz : > > > On Sat, 8 Oct 2016 22:27:08 -0700 > > Li Qiang wrote: > > > > > From: Li Qiang > > > > > > The 'fs.xattr.value' field in V9fsFidState object doesn't consider > > > the situation that this field has been allocated previously. Every > > > time, it will be allocated directly. This leads a host memory leak > > > issue. This patch fix this. > > > > > > > This cannot happen because v9fs_xattrwalk() always allocates this fid: > > > > xattr_fidp = alloc_fid(s, newfid); > > if (xattr_fidp == NULL) { > > err = -EINVAL; > > goto out; > > } > > > > > Signed-off-by: Li Qiang > > > --- > > > hw/9pfs/9p.c | 7 +++++-- > > > 1 file changed, 5 insertions(+), 2 deletions(-) > > > > > > diff --git a/hw/9pfs/9p.c b/hw/9pfs/9p.c > > > index 8751c19..3af23f9 100644 > > > --- a/hw/9pfs/9p.c > > > +++ b/hw/9pfs/9p.c > > > @@ -350,6 +350,7 @@ free_out: > > > v9fs_string_free(&fidp->fs.xattr.name); > > > free_value: > > > g_free(fidp->fs.xattr.value); > > > + fidp->fs.xattr.value = NULL; > > > return retval; > > > } > > > > > > @@ -3191,7 +3192,8 @@ static void v9fs_xattrwalk(void *opaque) > > > xattr_fidp->fid_type = P9_FID_XATTR; > > > xattr_fidp->fs.xattr.copied_len = -1; > > > if (size) { > > > - xattr_fidp->fs.xattr.value = g_malloc(size); > > > + xattr_fidp->fs.xattr.value = g_realloc( > > > + xattr_fidp->fs.xattr.value, size); > > > err = v9fs_co_llistxattr(pdu, &xattr_fidp->path, > > > xattr_fidp->fs.xattr.value, > > > xattr_fidp->fs.xattr.len); > > > @@ -3224,7 +3226,8 @@ static void v9fs_xattrwalk(void *opaque) > > > xattr_fidp->fid_type = P9_FID_XATTR; > > > xattr_fidp->fs.xattr.copied_len = -1; > > > if (size) { > > > - xattr_fidp->fs.xattr.value = g_malloc(size); > > > + xattr_fidp->fs.xattr.value = g_realloc( > > > + xattr_fidp->fs.xattr.value, size); > > > err = v9fs_co_lgetxattr(pdu, &xattr_fidp->path, > > > &name, xattr_fidp->fs.xattr.value, > > > xattr_fidp->fs.xattr.len); > > > >