| Message ID | 1409791248-27758-1-git-send-email-tim.gardner@canonical.com |
|---|---|
| State | New |
| Headers | show |
On 09/03/2014 05:40 PM, rtg.canonical@gmail.com wrote: > From: Tim Gardner <tim.gardner@canonical.com> > > BugLink: http://bugs.launchpad.net/bugs/1348670 > > Fix regression introduced in 3.2.60 by cherry-picking a post-3.14 patch that > depends on the set_acl methods being able to cope with a NULL ACL argument. > > Signed-off-by: Sergio Gelato <Sergio.Gelato@astro.su.se> > Signed-off-by: Tim Gardner <tim.gardner@canonical.com> > --- > fs/nfsd/vfs.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c > index 446dc01..fc208e4 100644 > --- a/fs/nfsd/vfs.c > +++ b/fs/nfsd/vfs.c > @@ -450,6 +450,9 @@ set_nfsv4_acl_one(struct dentry *dentry, struct posix_acl *pacl, char *key) > char *buf = NULL; > int error = 0; > > + if (!pacl) > + return vfs_setxattr(dentry, key, NULL, 0, 0); > + > buflen = posix_acl_xattr_size(pacl->a_count); > buf = kmalloc(buflen, GFP_KERNEL); > error = -ENOMEM; >
On Wed, Sep 03, 2014 at 05:40:48PM -0700, rtg.canonical@gmail.com wrote: > From: Tim Gardner <tim.gardner@canonical.com> > > BugLink: http://bugs.launchpad.net/bugs/1348670 > > Fix regression introduced in 3.2.60 by cherry-picking a post-3.14 patch that > depends on the set_acl methods being able to cope with a NULL ACL argument. > > Signed-off-by: Sergio Gelato <Sergio.Gelato@astro.su.se> > Signed-off-by: Tim Gardner <tim.gardner@canonical.com> > --- > fs/nfsd/vfs.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c > index 446dc01..fc208e4 100644 > --- a/fs/nfsd/vfs.c > +++ b/fs/nfsd/vfs.c > @@ -450,6 +450,9 @@ set_nfsv4_acl_one(struct dentry *dentry, struct posix_acl *pacl, char *key) > char *buf = NULL; > int error = 0; > > + if (!pacl) > + return vfs_setxattr(dentry, key, NULL, 0, 0); > + > buflen = posix_acl_xattr_size(pacl->a_count); > buf = kmalloc(buflen, GFP_KERNEL); > error = -ENOMEM; > -- > 1.9.1 > > > -- > kernel-team mailing list > kernel-team@lists.ubuntu.com > https://lists.ubuntu.com/mailman/listinfo/kernel-team I guess there was a typo on the 'Subject:' line: this patch is meant to be applied to Precise, not Lucid -- the bug refers to kernel version 3.2, not 2.6.32. Cheers, -- Luís
As pointed out by Luis, this seems to be a Precise SRU request. I have applied it there. Did you intend to include a Lucid one as well, or was this a title typo? Applied to Precise. -apw
* Luis Henriques [2014-09-04 09:52:45 +0100]: > I guess there was a typo on the 'Subject:' line: this patch is meant > to be applied to Precise, not Lucid -- the bug refers to kernel > version 3.2, not 2.6.32. The patch applies to 3.2 and 3.13 at least. If the change that caused the regression (NFSD: Call ->set_acl with a NULL ACL structure if no entries) has been applied to 2.6.32, then my patch (or something similar) may well be needed there too. The code being patched was refactored away in 3.14.
On Thu, Sep 04, 2014 at 10:49:23AM +0100, Andy Whitcroft wrote: > As pointed out by Luis, this seems to be a Precise SRU request. I have > applied it there. Did you intend to include a Lucid one as well, or was > this a title typo? > > Applied to Precise. On reviewing the bug, it also seems that this is likely applicable to T as well, the code being removed in v3.14 rendering U unaffected. -apw
On 09/04/2014 03:16 AM, Andy Whitcroft wrote: > On Thu, Sep 04, 2014 at 10:49:23AM +0100, Andy Whitcroft wrote: >> As pointed out by Luis, this seems to be a Precise SRU request. I have >> applied it there. Did you intend to include a Lucid one as well, or was >> this a title typo? >> >> Applied to Precise. > > On reviewing the bug, it also seems that this is likely applicable to T > as well, the code being removed in v3.14 rendering U unaffected. > > -apw > Doh! I should never send out patches after 5P. Indeed this does look applicable to 3.13. Shall I just apply it ? rtg
On Thu, Sep 04, 2014 at 07:35:40AM -0700, Tim Gardner wrote: > On 09/04/2014 03:16 AM, Andy Whitcroft wrote: > >On Thu, Sep 04, 2014 at 10:49:23AM +0100, Andy Whitcroft wrote: > >>As pointed out by Luis, this seems to be a Precise SRU request. I have > >>applied it there. Did you intend to include a Lucid one as well, or was > >>this a title typo? > >> > >>Applied to Precise. > > > >On reviewing the bug, it also seems that this is likely applicable to T > >as well, the code being removed in v3.14 rendering U unaffected. > > > >-apw > > > > Doh! I should never send out patches after 5P. Indeed this does look > applicable to 3.13. Shall I just apply it ? Works for me, if it is a cherry-pick there as well, and I can't see how it couldn't be: Acked-by: Andy Whitcroft <apw@canonical.com> -apw
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 446dc01..fc208e4 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -450,6 +450,9 @@ set_nfsv4_acl_one(struct dentry *dentry, struct posix_acl *pacl, char *key) char *buf = NULL; int error = 0; + if (!pacl) + return vfs_setxattr(dentry, key, NULL, 0, 0); + buflen = posix_acl_xattr_size(pacl->a_count); buf = kmalloc(buflen, GFP_KERNEL); error = -ENOMEM;