diff mbox series

[1/3] selinux: fix cond_list corruption when changing booleans

Message ID 20210330131646.1401838-2-omosnace@redhat.com (mailing list archive)
State Changes Requested
Delegated to: Paul Moore
Headers show
Series selinux: fix changing booleans | expand

Commit Message

Ondrej Mosnacek March 30, 2021, 1:16 p.m. UTC
Currently, duplicate_policydb_cond_list() first copies the whole
conditional avtab and then tries to link to the correct entries in
cond_dup_av_list() using avtab_search(). However, since the conditional
avtab may contain multiple entries with the same key, this approach
often fails to find the right entry, potentially leading to wrong rules
being activated/deactivated when booleans are changed.

To fix this, instead start with an empty conditional avtab and add the
individual entries one-by-one while building the new av_lists. This
approach leads to the correct result, since each entry is present in the
av_lists exactly once.

The issue can be reproduced with Fedora policy as follows:

    # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
    allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
    allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True
    # setsebool ftpd_anon_write=off ftpd_connect_all_unreserved=off ftpd_connect_db=off ftpd_full_access=off

On fixed kernels, the sesearch output is the same after the setsebool
command:

    # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
    allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
    allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True

While on the broken kernels, it will be different:

    # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
    allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
    allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
    allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True

Fixes: c7c556f1e81b ("selinux: refactor changing booleans")
Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com>
---
 security/selinux/ss/avtab.c       | 90 +++++++++----------------------
 security/selinux/ss/avtab.h       |  2 +-
 security/selinux/ss/conditional.c | 12 ++---
 3 files changed, 33 insertions(+), 71 deletions(-)

Comments

Paul Moore March 31, 2021, 2:04 a.m. UTC | #1
On Tue, Mar 30, 2021 at 9:16 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
>
> Currently, duplicate_policydb_cond_list() first copies the whole
> conditional avtab and then tries to link to the correct entries in
> cond_dup_av_list() using avtab_search(). However, since the conditional
> avtab may contain multiple entries with the same key, this approach
> often fails to find the right entry, potentially leading to wrong rules
> being activated/deactivated when booleans are changed.
>
> To fix this, instead start with an empty conditional avtab and add the
> individual entries one-by-one while building the new av_lists. This
> approach leads to the correct result, since each entry is present in the
> av_lists exactly once.
>
> The issue can be reproduced with Fedora policy as follows:
>
>     # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
>     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
>     allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True
>     # setsebool ftpd_anon_write=off ftpd_connect_all_unreserved=off ftpd_connect_db=off ftpd_full_access=off
>
> On fixed kernels, the sesearch output is the same after the setsebool
> command:
>
>     # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
>     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
>     allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True
>
> While on the broken kernels, it will be different:
>
>     # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
>     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
>     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
>     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
>
> Fixes: c7c556f1e81b ("selinux: refactor changing booleans")
> Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com>
> ---
>  security/selinux/ss/avtab.c       | 90 +++++++++----------------------
>  security/selinux/ss/avtab.h       |  2 +-
>  security/selinux/ss/conditional.c | 12 ++---
>  3 files changed, 33 insertions(+), 71 deletions(-)
>
> diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c
> index 6dcb6aa4db7f..11f8f524de98 100644
> --- a/security/selinux/ss/avtab.c
> +++ b/security/selinux/ss/avtab.c
> @@ -305,86 +305,48 @@ void avtab_init(struct avtab *h)
>         h->nel = 0;
>  }
>
> -int avtab_alloc(struct avtab *h, u32 nrules)
> +static int avtab_alloc_common(struct avtab *h, u32 nslot)
>  {
> -       u32 mask = 0;
> -       u32 shift = 0;
> -       u32 work = nrules;
> -       u32 nslot = 0;
> -
> -       if (nrules == 0)
> -               goto avtab_alloc_out;
> -
> -       while (work) {
> -               work  = work >> 1;
> -               shift++;
> -       }
> -       if (shift > 2)
> -               shift = shift - 2;
> -       nslot = 1 << shift;
> -       if (nslot > MAX_AVTAB_HASH_BUCKETS)
> -               nslot = MAX_AVTAB_HASH_BUCKETS;
> -       mask = nslot - 1;
> -
>         h->htable = kvcalloc(nslot, sizeof(void *), GFP_KERNEL);
>         if (!h->htable)
>                 return -ENOMEM;

Hmmm, do we need to protect against 'nslot == 0'?  Unless I missed
something, a quick dive into kvcalloc() makes it look like it can
return non-NULL for zero length allocations, at least in the slab
case.

> - avtab_alloc_out:
>         h->nel = 0;
>         h->nslot = nslot;
> -       h->mask = mask;
> -       pr_debug("SELinux: %d avtab hash slots, %d rules.\n",
> -              h->nslot, nrules);
> +       h->mask = nslot - 1;

This is definitely not good if 'nslot <= 1';

>         return 0;
>  }
>
> -int avtab_duplicate(struct avtab *new, struct avtab *orig)
> +int avtab_alloc(struct avtab *h, u32 nrules)
>  {
> -       int i;
> -       struct avtab_node *node, *tmp, *tail;
> -
> -       memset(new, 0, sizeof(*new));
> +       int rc;
> +       u32 shift = 0;
> +       u32 work = nrules;
> +       u32 nslot = 0;
>
> -       new->htable = kvcalloc(orig->nslot, sizeof(void *), GFP_KERNEL);
> -       if (!new->htable)
> -               return -ENOMEM;
> -       new->nslot = orig->nslot;
> -       new->mask = orig->mask;
> -
> -       for (i = 0; i < orig->nslot; i++) {
> -               tail = NULL;
> -               for (node = orig->htable[i]; node; node = node->next) {
> -                       tmp = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL);
> -                       if (!tmp)
> -                               goto error;
> -                       tmp->key = node->key;
> -                       if (tmp->key.specified & AVTAB_XPERMS) {
> -                               tmp->datum.u.xperms =
> -                                       kmem_cache_zalloc(avtab_xperms_cachep,
> -                                                       GFP_KERNEL);
> -                               if (!tmp->datum.u.xperms) {
> -                                       kmem_cache_free(avtab_node_cachep, tmp);
> -                                       goto error;
> -                               }
> -                               tmp->datum.u.xperms = node->datum.u.xperms;
> -                       } else
> -                               tmp->datum.u.data = node->datum.u.data;
> -
> -                       if (tail)
> -                               tail->next = tmp;
> -                       else
> -                               new->htable[i] = tmp;
> -
> -                       tail = tmp;
> -                       new->nel++;
> +       if (nrules != 0) {
> +               while (work) {
> +                       work  = work >> 1;

Extra  horizontal  spaces  are  awkward  and  bad.

> +                       shift++;
>                 }
> +               if (shift > 2)
> +                       shift = shift - 2;

Since we are getting nit-picky with this code, why not make the
loop/if a bit more elegant?  How about something like below?

  u32 shift = 2;
  u32 work = nrules >> 4;
  while (work) {
    work >>= 1;
    shift++;
  }

> +               nslot = 1 << shift;
> +               if (nslot > MAX_AVTAB_HASH_BUCKETS)
> +                       nslot = MAX_AVTAB_HASH_BUCKETS;
>         }
>
> +       rc = avtab_alloc_common(h, nslot);
> +       if (rc)
> +               return rc;
> +
> +       pr_debug("SELinux: %d avtab hash slots, %d rules.\n", nslot, nrules);
>         return 0;
> -error:
> -       avtab_destroy(new);
> -       return -ENOMEM;
> +}
> +
> +int avtab_alloc_dup(struct avtab *new, const struct avtab *orig)
> +{
> +       return avtab_alloc_common(new, orig->nslot);
>  }
Ondrej Mosnacek March 31, 2021, 9:12 a.m. UTC | #2
On Wed, Mar 31, 2021 at 4:04 AM Paul Moore <paul@paul-moore.com> wrote:
> On Tue, Mar 30, 2021 at 9:16 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> >
> > Currently, duplicate_policydb_cond_list() first copies the whole
> > conditional avtab and then tries to link to the correct entries in
> > cond_dup_av_list() using avtab_search(). However, since the conditional
> > avtab may contain multiple entries with the same key, this approach
> > often fails to find the right entry, potentially leading to wrong rules
> > being activated/deactivated when booleans are changed.
> >
> > To fix this, instead start with an empty conditional avtab and add the
> > individual entries one-by-one while building the new av_lists. This
> > approach leads to the correct result, since each entry is present in the
> > av_lists exactly once.
> >
> > The issue can be reproduced with Fedora policy as follows:
> >
> >     # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
> >     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
> >     allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True
> >     # setsebool ftpd_anon_write=off ftpd_connect_all_unreserved=off ftpd_connect_db=off ftpd_full_access=off
> >
> > On fixed kernels, the sesearch output is the same after the setsebool
> > command:
> >
> >     # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
> >     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
> >     allow ftpd_t public_content_rw_t:dir { add_name create link remove_name rename reparent rmdir setattr unlink watch watch_reads write }; [ ftpd_anon_write ]:True
> >
> > While on the broken kernels, it will be different:
> >
> >     # sesearch -s ftpd_t -t public_content_rw_t -c dir -p create -A
> >     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
> >     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
> >     allow ftpd_t non_security_file_type:dir { add_name create getattr ioctl link lock open read remove_name rename reparent rmdir search setattr unlink watch watch_reads write }; [ ftpd_full_access ]:True
> >
> > Fixes: c7c556f1e81b ("selinux: refactor changing booleans")
> > Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com>
> > ---
> >  security/selinux/ss/avtab.c       | 90 +++++++++----------------------
> >  security/selinux/ss/avtab.h       |  2 +-
> >  security/selinux/ss/conditional.c | 12 ++---
> >  3 files changed, 33 insertions(+), 71 deletions(-)
> >
> > diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c
> > index 6dcb6aa4db7f..11f8f524de98 100644
> > --- a/security/selinux/ss/avtab.c
> > +++ b/security/selinux/ss/avtab.c
> > @@ -305,86 +305,48 @@ void avtab_init(struct avtab *h)
> >         h->nel = 0;
> >  }
> >
> > -int avtab_alloc(struct avtab *h, u32 nrules)
> > +static int avtab_alloc_common(struct avtab *h, u32 nslot)
> >  {
> > -       u32 mask = 0;
> > -       u32 shift = 0;
> > -       u32 work = nrules;
> > -       u32 nslot = 0;
> > -
> > -       if (nrules == 0)
> > -               goto avtab_alloc_out;
> > -
> > -       while (work) {
> > -               work  = work >> 1;
> > -               shift++;
> > -       }
> > -       if (shift > 2)
> > -               shift = shift - 2;
> > -       nslot = 1 << shift;
> > -       if (nslot > MAX_AVTAB_HASH_BUCKETS)
> > -               nslot = MAX_AVTAB_HASH_BUCKETS;
> > -       mask = nslot - 1;
> > -
> >         h->htable = kvcalloc(nslot, sizeof(void *), GFP_KERNEL);
> >         if (!h->htable)
> >                 return -ENOMEM;
>
> Hmmm, do we need to protect against 'nslot == 0'?  Unless I missed
> something, a quick dive into kvcalloc() makes it look like it can
> return non-NULL for zero length allocations, at least in the slab
> case.
>
> > - avtab_alloc_out:
> >         h->nel = 0;
> >         h->nslot = nslot;
> > -       h->mask = mask;
> > -       pr_debug("SELinux: %d avtab hash slots, %d rules.\n",
> > -              h->nslot, nrules);
> > +       h->mask = nslot - 1;
>
> This is definitely not good if 'nslot <= 1';

Yeah, in fact the current code doesn't seem to be safe against nslot
== 0 either... I'll insert another patch that fixes this.

>
> >         return 0;
> >  }
> >
> > -int avtab_duplicate(struct avtab *new, struct avtab *orig)
> > +int avtab_alloc(struct avtab *h, u32 nrules)
> >  {
> > -       int i;
> > -       struct avtab_node *node, *tmp, *tail;
> > -
> > -       memset(new, 0, sizeof(*new));
> > +       int rc;
> > +       u32 shift = 0;
> > +       u32 work = nrules;
> > +       u32 nslot = 0;
> >
> > -       new->htable = kvcalloc(orig->nslot, sizeof(void *), GFP_KERNEL);
> > -       if (!new->htable)
> > -               return -ENOMEM;
> > -       new->nslot = orig->nslot;
> > -       new->mask = orig->mask;
> > -
> > -       for (i = 0; i < orig->nslot; i++) {
> > -               tail = NULL;
> > -               for (node = orig->htable[i]; node; node = node->next) {
> > -                       tmp = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL);
> > -                       if (!tmp)
> > -                               goto error;
> > -                       tmp->key = node->key;
> > -                       if (tmp->key.specified & AVTAB_XPERMS) {
> > -                               tmp->datum.u.xperms =
> > -                                       kmem_cache_zalloc(avtab_xperms_cachep,
> > -                                                       GFP_KERNEL);
> > -                               if (!tmp->datum.u.xperms) {
> > -                                       kmem_cache_free(avtab_node_cachep, tmp);
> > -                                       goto error;
> > -                               }
> > -                               tmp->datum.u.xperms = node->datum.u.xperms;
> > -                       } else
> > -                               tmp->datum.u.data = node->datum.u.data;
> > -
> > -                       if (tail)
> > -                               tail->next = tmp;
> > -                       else
> > -                               new->htable[i] = tmp;
> > -
> > -                       tail = tmp;
> > -                       new->nel++;
> > +       if (nrules != 0) {
> > +               while (work) {
> > +                       work  = work >> 1;
>
> Extra  horizontal  spaces  are  awkward  and  bad.
>
> > +                       shift++;
> >                 }
> > +               if (shift > 2)
> > +                       shift = shift - 2;
>
> Since we are getting nit-picky with this code, why not make the
> loop/if a bit more elegant?  How about something like below?
>
>   u32 shift = 2;
>   u32 work = nrules >> 4;
>   while (work) {
>     work >>= 1;
>     shift++;
>   }

I think you meant:
u32 shift = 0;
u32 work = nrules >> 2;
while (work) {
    work >>= 1;
    shift++;
}

...which is equivalent to the current code and yes, I'll use that :)

>
> > +               nslot = 1 << shift;
> > +               if (nslot > MAX_AVTAB_HASH_BUCKETS)
> > +                       nslot = MAX_AVTAB_HASH_BUCKETS;
> >         }
> >
> > +       rc = avtab_alloc_common(h, nslot);
> > +       if (rc)
> > +               return rc;
> > +
> > +       pr_debug("SELinux: %d avtab hash slots, %d rules.\n", nslot, nrules);
> >         return 0;
> > -error:
> > -       avtab_destroy(new);
> > -       return -ENOMEM;
> > +}
> > +
> > +int avtab_alloc_dup(struct avtab *new, const struct avtab *orig)
> > +{
> > +       return avtab_alloc_common(new, orig->nslot);
> >  }
>
> --
> paul moore
> www.paul-moore.com
>
Paul Moore March 31, 2021, 11:19 p.m. UTC | #3
On Wed, Mar 31, 2021 at 5:12 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> On Wed, Mar 31, 2021 at 4:04 AM Paul Moore <paul@paul-moore.com> wrote:
> > On Tue, Mar 30, 2021 at 9:16 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:

...

> > > -int avtab_duplicate(struct avtab *new, struct avtab *orig)
> > > +int avtab_alloc(struct avtab *h, u32 nrules)
> > >  {
> > > -       int i;
> > > -       struct avtab_node *node, *tmp, *tail;
> > > -
> > > -       memset(new, 0, sizeof(*new));
> > > +       int rc;
> > > +       u32 shift = 0;
> > > +       u32 work = nrules;
> > > +       u32 nslot = 0;
> > >
> > > -       new->htable = kvcalloc(orig->nslot, sizeof(void *), GFP_KERNEL);
> > > -       if (!new->htable)
> > > -               return -ENOMEM;
> > > -       new->nslot = orig->nslot;
> > > -       new->mask = orig->mask;
> > > -
> > > -       for (i = 0; i < orig->nslot; i++) {
> > > -               tail = NULL;
> > > -               for (node = orig->htable[i]; node; node = node->next) {
> > > -                       tmp = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL);
> > > -                       if (!tmp)
> > > -                               goto error;
> > > -                       tmp->key = node->key;
> > > -                       if (tmp->key.specified & AVTAB_XPERMS) {
> > > -                               tmp->datum.u.xperms =
> > > -                                       kmem_cache_zalloc(avtab_xperms_cachep,
> > > -                                                       GFP_KERNEL);
> > > -                               if (!tmp->datum.u.xperms) {
> > > -                                       kmem_cache_free(avtab_node_cachep, tmp);
> > > -                                       goto error;
> > > -                               }
> > > -                               tmp->datum.u.xperms = node->datum.u.xperms;
> > > -                       } else
> > > -                               tmp->datum.u.data = node->datum.u.data;
> > > -
> > > -                       if (tail)
> > > -                               tail->next = tmp;
> > > -                       else
> > > -                               new->htable[i] = tmp;
> > > -
> > > -                       tail = tmp;
> > > -                       new->nel++;
> > > +       if (nrules != 0) {
> > > +               while (work) {
> > > +                       work  = work >> 1;
> >
> > Extra  horizontal  spaces  are  awkward  and  bad.
> >
> > > +                       shift++;
> > >                 }
> > > +               if (shift > 2)
> > > +                       shift = shift - 2;
> >
> > Since we are getting nit-picky with this code, why not make the
> > loop/if a bit more elegant?  How about something like below?
> >
> >   u32 shift = 2;
> >   u32 work = nrules >> 4;
> >   while (work) {
> >     work >>= 1;
> >     shift++;
> >   }
>
> I think you meant:
> u32 shift = 0;
> u32 work = nrules >> 2;
> while (work) {
>     work >>= 1;
>     shift++;
> }
>
> ...which is equivalent to the current code and yes, I'll use that :)

Well, no, not really, but that's okay as looking at it now we both got
it wrong :)

Basically I wanted to avoid the odd problem where the current code has
a dip in the number of slots/buckets when nrules is equal to 4, 5, 6,
or 7.  While the code I proposed yesterday did that, it inflated the
number of buckets beyond the current code; your suggestion had
problems with low numbers resulting in zero buckets.

I think what we really want is this:

  shift = 2;
  work = nrules >> 4;
  while (work) {
    work >>= 1;
    shift++;
  }

... it avoids any dips in the bucket count and it results in similar
bucket counts as the existing code for larger numbers of rules.
Ondrej Mosnacek April 1, 2021, 7:59 a.m. UTC | #4
On Thu, Apr 1, 2021 at 1:20 AM Paul Moore <paul@paul-moore.com> wrote:
> On Wed, Mar 31, 2021 at 5:12 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> > On Wed, Mar 31, 2021 at 4:04 AM Paul Moore <paul@paul-moore.com> wrote:
> > > On Tue, Mar 30, 2021 at 9:16 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
>
> ...
>
> > > > -int avtab_duplicate(struct avtab *new, struct avtab *orig)
> > > > +int avtab_alloc(struct avtab *h, u32 nrules)
> > > >  {
> > > > -       int i;
> > > > -       struct avtab_node *node, *tmp, *tail;
> > > > -
> > > > -       memset(new, 0, sizeof(*new));
> > > > +       int rc;
> > > > +       u32 shift = 0;
> > > > +       u32 work = nrules;
> > > > +       u32 nslot = 0;
> > > >
> > > > -       new->htable = kvcalloc(orig->nslot, sizeof(void *), GFP_KERNEL);
> > > > -       if (!new->htable)
> > > > -               return -ENOMEM;
> > > > -       new->nslot = orig->nslot;
> > > > -       new->mask = orig->mask;
> > > > -
> > > > -       for (i = 0; i < orig->nslot; i++) {
> > > > -               tail = NULL;
> > > > -               for (node = orig->htable[i]; node; node = node->next) {
> > > > -                       tmp = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL);
> > > > -                       if (!tmp)
> > > > -                               goto error;
> > > > -                       tmp->key = node->key;
> > > > -                       if (tmp->key.specified & AVTAB_XPERMS) {
> > > > -                               tmp->datum.u.xperms =
> > > > -                                       kmem_cache_zalloc(avtab_xperms_cachep,
> > > > -                                                       GFP_KERNEL);
> > > > -                               if (!tmp->datum.u.xperms) {
> > > > -                                       kmem_cache_free(avtab_node_cachep, tmp);
> > > > -                                       goto error;
> > > > -                               }
> > > > -                               tmp->datum.u.xperms = node->datum.u.xperms;
> > > > -                       } else
> > > > -                               tmp->datum.u.data = node->datum.u.data;
> > > > -
> > > > -                       if (tail)
> > > > -                               tail->next = tmp;
> > > > -                       else
> > > > -                               new->htable[i] = tmp;
> > > > -
> > > > -                       tail = tmp;
> > > > -                       new->nel++;
> > > > +       if (nrules != 0) {
> > > > +               while (work) {
> > > > +                       work  = work >> 1;
> > >
> > > Extra  horizontal  spaces  are  awkward  and  bad.
> > >
> > > > +                       shift++;
> > > >                 }
> > > > +               if (shift > 2)
> > > > +                       shift = shift - 2;
> > >
> > > Since we are getting nit-picky with this code, why not make the
> > > loop/if a bit more elegant?  How about something like below?
> > >
> > >   u32 shift = 2;
> > >   u32 work = nrules >> 4;
> > >   while (work) {
> > >     work >>= 1;
> > >     shift++;
> > >   }
> >
> > I think you meant:
> > u32 shift = 0;
> > u32 work = nrules >> 2;
> > while (work) {
> >     work >>= 1;
> >     shift++;
> > }
> >
> > ...which is equivalent to the current code and yes, I'll use that :)
>
> Well, no, not really, but that's okay as looking at it now we both got
> it wrong :)
>
> Basically I wanted to avoid the odd problem where the current code has
> a dip in the number of slots/buckets when nrules is equal to 4, 5, 6,
> or 7.  While the code I proposed yesterday did that, it inflated the
> number of buckets beyond the current code; your suggestion had
> problems with low numbers resulting in zero buckets.

Aah, I wrongly parsed the "if (shift > 2) shift -= 2;" statement...
Yeah, I guess even the original intent was different than what the
code does.

Anyway, my code doesn't result in zero buckets, at worst in 1 bucket
(nslot = 1 << shift), which I think is reasonable given that the
intent of the code seems to be to simply squeeze the number of slots
down to approx. 1/4 the number of rules.

To make it a bit more concrete: your code allocates 4 buckets for
nrules 1..15; my code allocates 1 bucket for nrules 1..3, 2 buckets
for nrules 4..7, and 4 buckets for 8..15. Both our solutions are
equivalent to the old code at nrules >= 8.

So I'll argue that my proposed solution is actually slightly better
(and avoids adding a new magic number).

>
> I think what we really want is this:
>
>   shift = 2;
>   work = nrules >> 4;
>   while (work) {
>     work >>= 1;
>     shift++;
>   }
>
> ... it avoids any dips in the bucket count and it results in similar
> bucket counts as the existing code for larger numbers of rules.

But that's exactly the same as your original suggestion :D

--
Ondrej Mosnacek
Software Engineer, Linux Security - SELinux kernel
Red Hat, Inc.
Paul Moore April 1, 2021, 3:36 p.m. UTC | #5
On Thu, Apr 1, 2021 at 3:59 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> On Thu, Apr 1, 2021 at 1:20 AM Paul Moore <paul@paul-moore.com> wrote:
> > On Wed, Mar 31, 2021 at 5:12 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> > > On Wed, Mar 31, 2021 at 4:04 AM Paul Moore <paul@paul-moore.com> wrote:
> > > > On Tue, Mar 30, 2021 at 9:16 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> >
> > ...
> >
> > > > > -int avtab_duplicate(struct avtab *new, struct avtab *orig)
> > > > > +int avtab_alloc(struct avtab *h, u32 nrules)
> > > > >  {
> > > > > -       int i;
> > > > > -       struct avtab_node *node, *tmp, *tail;
> > > > > -
> > > > > -       memset(new, 0, sizeof(*new));
> > > > > +       int rc;
> > > > > +       u32 shift = 0;
> > > > > +       u32 work = nrules;
> > > > > +       u32 nslot = 0;
> > > > >
> > > > > -       new->htable = kvcalloc(orig->nslot, sizeof(void *), GFP_KERNEL);
> > > > > -       if (!new->htable)
> > > > > -               return -ENOMEM;
> > > > > -       new->nslot = orig->nslot;
> > > > > -       new->mask = orig->mask;
> > > > > -
> > > > > -       for (i = 0; i < orig->nslot; i++) {
> > > > > -               tail = NULL;
> > > > > -               for (node = orig->htable[i]; node; node = node->next) {
> > > > > -                       tmp = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL);
> > > > > -                       if (!tmp)
> > > > > -                               goto error;
> > > > > -                       tmp->key = node->key;
> > > > > -                       if (tmp->key.specified & AVTAB_XPERMS) {
> > > > > -                               tmp->datum.u.xperms =
> > > > > -                                       kmem_cache_zalloc(avtab_xperms_cachep,
> > > > > -                                                       GFP_KERNEL);
> > > > > -                               if (!tmp->datum.u.xperms) {
> > > > > -                                       kmem_cache_free(avtab_node_cachep, tmp);
> > > > > -                                       goto error;
> > > > > -                               }
> > > > > -                               tmp->datum.u.xperms = node->datum.u.xperms;
> > > > > -                       } else
> > > > > -                               tmp->datum.u.data = node->datum.u.data;
> > > > > -
> > > > > -                       if (tail)
> > > > > -                               tail->next = tmp;
> > > > > -                       else
> > > > > -                               new->htable[i] = tmp;
> > > > > -
> > > > > -                       tail = tmp;
> > > > > -                       new->nel++;
> > > > > +       if (nrules != 0) {
> > > > > +               while (work) {
> > > > > +                       work  = work >> 1;
> > > >
> > > > Extra  horizontal  spaces  are  awkward  and  bad.
> > > >
> > > > > +                       shift++;
> > > > >                 }
> > > > > +               if (shift > 2)
> > > > > +                       shift = shift - 2;
> > > >
> > > > Since we are getting nit-picky with this code, why not make the
> > > > loop/if a bit more elegant?  How about something like below?
> > > >
> > > >   u32 shift = 2;
> > > >   u32 work = nrules >> 4;
> > > >   while (work) {
> > > >     work >>= 1;
> > > >     shift++;
> > > >   }
> > >
> > > I think you meant:
> > > u32 shift = 0;
> > > u32 work = nrules >> 2;
> > > while (work) {
> > >     work >>= 1;
> > >     shift++;
> > > }
> > >
> > > ...which is equivalent to the current code and yes, I'll use that :)
> >
> > Well, no, not really, but that's okay as looking at it now we both got
> > it wrong :)
> >
> > Basically I wanted to avoid the odd problem where the current code has
> > a dip in the number of slots/buckets when nrules is equal to 4, 5, 6,
> > or 7.  While the code I proposed yesterday did that, it inflated the
> > number of buckets beyond the current code; your suggestion had
> > problems with low numbers resulting in zero buckets.
>
> Aah, I wrongly parsed the "if (shift > 2) shift -= 2;" statement...
> Yeah, I guess even the original intent was different than what the
> code does.

I think we all mis-interpreted that bit of code at some point.  I
seriously doubt that dip was what the original author intended :)

> Anyway, my code doesn't result in zero buckets, at worst in 1 bucket
> (nslot = 1 << shift) ...

Sorry, yes, I mis-spoke (mis-typed?); I was talking about the
shift/exponent value.

> So I'll argue that my proposed solution is actually slightly better
> (and avoids adding a new magic number).

The magic number argument isn't really valid as both approaches use
them to some degree.  Creating a #define constant is overkill here,
but I guess a short comment wouldn't be a bad idea if you wanted to
add one; I'm not going to require it in this case.

Since we are at -rc5 I really want to wrap this up soon so I'm going
to make one final suggestion:

  shift = 1;
  work = nrules >> 3;
  while (work) {
    work >>= 1;
    shift++;
  }

... this preserves the original nslot count, minus the bump/dip seen
with low rule numbers.  The shift value starts at 1, increases to 2
when the rules reach 8, increases to 3 when the rules reach 16, and so
on.  Of the approaches we've discussed, I believe it is the most
faithful to original intent.
Ondrej Mosnacek April 1, 2021, 3:54 p.m. UTC | #6
On Thu, Apr 1, 2021 at 5:36 PM Paul Moore <paul@paul-moore.com> wrote:
> On Thu, Apr 1, 2021 at 3:59 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> > On Thu, Apr 1, 2021 at 1:20 AM Paul Moore <paul@paul-moore.com> wrote:
> > > On Wed, Mar 31, 2021 at 5:12 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> > > > On Wed, Mar 31, 2021 at 4:04 AM Paul Moore <paul@paul-moore.com> wrote:
> > > > > On Tue, Mar 30, 2021 at 9:16 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> > >
> > > ...
> > >
> > > > > > -int avtab_duplicate(struct avtab *new, struct avtab *orig)
> > > > > > +int avtab_alloc(struct avtab *h, u32 nrules)
> > > > > >  {
> > > > > > -       int i;
> > > > > > -       struct avtab_node *node, *tmp, *tail;
> > > > > > -
> > > > > > -       memset(new, 0, sizeof(*new));
> > > > > > +       int rc;
> > > > > > +       u32 shift = 0;
> > > > > > +       u32 work = nrules;
> > > > > > +       u32 nslot = 0;
> > > > > >
> > > > > > -       new->htable = kvcalloc(orig->nslot, sizeof(void *), GFP_KERNEL);
> > > > > > -       if (!new->htable)
> > > > > > -               return -ENOMEM;
> > > > > > -       new->nslot = orig->nslot;
> > > > > > -       new->mask = orig->mask;
> > > > > > -
> > > > > > -       for (i = 0; i < orig->nslot; i++) {
> > > > > > -               tail = NULL;
> > > > > > -               for (node = orig->htable[i]; node; node = node->next) {
> > > > > > -                       tmp = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL);
> > > > > > -                       if (!tmp)
> > > > > > -                               goto error;
> > > > > > -                       tmp->key = node->key;
> > > > > > -                       if (tmp->key.specified & AVTAB_XPERMS) {
> > > > > > -                               tmp->datum.u.xperms =
> > > > > > -                                       kmem_cache_zalloc(avtab_xperms_cachep,
> > > > > > -                                                       GFP_KERNEL);
> > > > > > -                               if (!tmp->datum.u.xperms) {
> > > > > > -                                       kmem_cache_free(avtab_node_cachep, tmp);
> > > > > > -                                       goto error;
> > > > > > -                               }
> > > > > > -                               tmp->datum.u.xperms = node->datum.u.xperms;
> > > > > > -                       } else
> > > > > > -                               tmp->datum.u.data = node->datum.u.data;
> > > > > > -
> > > > > > -                       if (tail)
> > > > > > -                               tail->next = tmp;
> > > > > > -                       else
> > > > > > -                               new->htable[i] = tmp;
> > > > > > -
> > > > > > -                       tail = tmp;
> > > > > > -                       new->nel++;
> > > > > > +       if (nrules != 0) {
> > > > > > +               while (work) {
> > > > > > +                       work  = work >> 1;
> > > > >
> > > > > Extra  horizontal  spaces  are  awkward  and  bad.
> > > > >
> > > > > > +                       shift++;
> > > > > >                 }
> > > > > > +               if (shift > 2)
> > > > > > +                       shift = shift - 2;
> > > > >
> > > > > Since we are getting nit-picky with this code, why not make the
> > > > > loop/if a bit more elegant?  How about something like below?
> > > > >
> > > > >   u32 shift = 2;
> > > > >   u32 work = nrules >> 4;
> > > > >   while (work) {
> > > > >     work >>= 1;
> > > > >     shift++;
> > > > >   }
> > > >
> > > > I think you meant:
> > > > u32 shift = 0;
> > > > u32 work = nrules >> 2;
> > > > while (work) {
> > > >     work >>= 1;
> > > >     shift++;
> > > > }
> > > >
> > > > ...which is equivalent to the current code and yes, I'll use that :)
> > >
> > > Well, no, not really, but that's okay as looking at it now we both got
> > > it wrong :)
> > >
> > > Basically I wanted to avoid the odd problem where the current code has
> > > a dip in the number of slots/buckets when nrules is equal to 4, 5, 6,
> > > or 7.  While the code I proposed yesterday did that, it inflated the
> > > number of buckets beyond the current code; your suggestion had
> > > problems with low numbers resulting in zero buckets.
> >
> > Aah, I wrongly parsed the "if (shift > 2) shift -= 2;" statement...
> > Yeah, I guess even the original intent was different than what the
> > code does.
>
> I think we all mis-interpreted that bit of code at some point.  I
> seriously doubt that dip was what the original author intended :)
>
> > Anyway, my code doesn't result in zero buckets, at worst in 1 bucket
> > (nslot = 1 << shift) ...
>
> Sorry, yes, I mis-spoke (mis-typed?); I was talking about the
> shift/exponent value.
>
> > So I'll argue that my proposed solution is actually slightly better
> > (and avoids adding a new magic number).
>
> The magic number argument isn't really valid as both approaches use
> them to some degree.  Creating a #define constant is overkill here,
> but I guess a short comment wouldn't be a bad idea if you wanted to
> add one; I'm not going to require it in this case.
>
> Since we are at -rc5 I really want to wrap this up soon so I'm going
> to make one final suggestion:
>
>   shift = 1;
>   work = nrules >> 3;
>   while (work) {
>     work >>= 1;
>     shift++;
>   }
>
> ... this preserves the original nslot count, minus the bump/dip seen
> with low rule numbers.  The shift value starts at 1, increases to 2
> when the rules reach 8, increases to 3 when the rules reach 16, and so
> on.  Of the approaches we've discussed, I believe it is the most
> faithful to original intent.

Ok, I agree it's not something worth obsessing over, so I'll just use
this last suggestion :) (One day maybe I'll try to simplify/optimize
it a bit, but that is for another patch...)
Paul Moore April 1, 2021, 3:56 p.m. UTC | #7
On Thu, Apr 1, 2021 at 11:54 AM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> Ok, I agree it's not something worth obsessing over, so I'll just use
> this last suggestion :) (One day maybe I'll try to simplify/optimize
> it a bit, but that is for another patch...)

Sounds good, thanks Ondrej.
diff mbox series

Patch

diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c
index 6dcb6aa4db7f..11f8f524de98 100644
--- a/security/selinux/ss/avtab.c
+++ b/security/selinux/ss/avtab.c
@@ -305,86 +305,48 @@  void avtab_init(struct avtab *h)
 	h->nel = 0;
 }
 
-int avtab_alloc(struct avtab *h, u32 nrules)
+static int avtab_alloc_common(struct avtab *h, u32 nslot)
 {
-	u32 mask = 0;
-	u32 shift = 0;
-	u32 work = nrules;
-	u32 nslot = 0;
-
-	if (nrules == 0)
-		goto avtab_alloc_out;
-
-	while (work) {
-		work  = work >> 1;
-		shift++;
-	}
-	if (shift > 2)
-		shift = shift - 2;
-	nslot = 1 << shift;
-	if (nslot > MAX_AVTAB_HASH_BUCKETS)
-		nslot = MAX_AVTAB_HASH_BUCKETS;
-	mask = nslot - 1;
-
 	h->htable = kvcalloc(nslot, sizeof(void *), GFP_KERNEL);
 	if (!h->htable)
 		return -ENOMEM;
 
- avtab_alloc_out:
 	h->nel = 0;
 	h->nslot = nslot;
-	h->mask = mask;
-	pr_debug("SELinux: %d avtab hash slots, %d rules.\n",
-	       h->nslot, nrules);
+	h->mask = nslot - 1;
 	return 0;
 }
 
-int avtab_duplicate(struct avtab *new, struct avtab *orig)
+int avtab_alloc(struct avtab *h, u32 nrules)
 {
-	int i;
-	struct avtab_node *node, *tmp, *tail;
-
-	memset(new, 0, sizeof(*new));
+	int rc;
+	u32 shift = 0;
+	u32 work = nrules;
+	u32 nslot = 0;
 
-	new->htable = kvcalloc(orig->nslot, sizeof(void *), GFP_KERNEL);
-	if (!new->htable)
-		return -ENOMEM;
-	new->nslot = orig->nslot;
-	new->mask = orig->mask;
-
-	for (i = 0; i < orig->nslot; i++) {
-		tail = NULL;
-		for (node = orig->htable[i]; node; node = node->next) {
-			tmp = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL);
-			if (!tmp)
-				goto error;
-			tmp->key = node->key;
-			if (tmp->key.specified & AVTAB_XPERMS) {
-				tmp->datum.u.xperms =
-					kmem_cache_zalloc(avtab_xperms_cachep,
-							GFP_KERNEL);
-				if (!tmp->datum.u.xperms) {
-					kmem_cache_free(avtab_node_cachep, tmp);
-					goto error;
-				}
-				tmp->datum.u.xperms = node->datum.u.xperms;
-			} else
-				tmp->datum.u.data = node->datum.u.data;
-
-			if (tail)
-				tail->next = tmp;
-			else
-				new->htable[i] = tmp;
-
-			tail = tmp;
-			new->nel++;
+	if (nrules != 0) {
+		while (work) {
+			work  = work >> 1;
+			shift++;
 		}
+		if (shift > 2)
+			shift = shift - 2;
+		nslot = 1 << shift;
+		if (nslot > MAX_AVTAB_HASH_BUCKETS)
+			nslot = MAX_AVTAB_HASH_BUCKETS;
 	}
 
+	rc = avtab_alloc_common(h, nslot);
+	if (rc)
+		return rc;
+
+	pr_debug("SELinux: %d avtab hash slots, %d rules.\n", nslot, nrules);
 	return 0;
-error:
-	avtab_destroy(new);
-	return -ENOMEM;
+}
+
+int avtab_alloc_dup(struct avtab *new, const struct avtab *orig)
+{
+	return avtab_alloc_common(new, orig->nslot);
 }
 
 void avtab_hash_eval(struct avtab *h, char *tag)
diff --git a/security/selinux/ss/avtab.h b/security/selinux/ss/avtab.h
index 4c4445ca9118..f2eeb36265d1 100644
--- a/security/selinux/ss/avtab.h
+++ b/security/selinux/ss/avtab.h
@@ -89,7 +89,7 @@  struct avtab {
 
 void avtab_init(struct avtab *h);
 int avtab_alloc(struct avtab *, u32);
-int avtab_duplicate(struct avtab *new, struct avtab *orig);
+int avtab_alloc_dup(struct avtab *new, const struct avtab *orig);
 struct avtab_datum *avtab_search(struct avtab *h, struct avtab_key *k);
 void avtab_destroy(struct avtab *h);
 void avtab_hash_eval(struct avtab *h, char *tag);
diff --git a/security/selinux/ss/conditional.c b/security/selinux/ss/conditional.c
index 0b32f3ab025e..1ef74c085f2b 100644
--- a/security/selinux/ss/conditional.c
+++ b/security/selinux/ss/conditional.c
@@ -605,7 +605,6 @@  static int cond_dup_av_list(struct cond_av_list *new,
 			struct cond_av_list *orig,
 			struct avtab *avtab)
 {
-	struct avtab_node *avnode;
 	u32 i;
 
 	memset(new, 0, sizeof(*new));
@@ -615,10 +614,11 @@  static int cond_dup_av_list(struct cond_av_list *new,
 		return -ENOMEM;
 
 	for (i = 0; i < orig->len; i++) {
-		avnode = avtab_search_node(avtab, &orig->nodes[i]->key);
-		if (WARN_ON(!avnode))
-			return -EINVAL;
-		new->nodes[i] = avnode;
+		new->nodes[i] = avtab_insert_nonunique(avtab,
+						       &orig->nodes[i]->key,
+						       &orig->nodes[i]->datum);
+		if (!new->nodes[i])
+			return -ENOMEM;
 		new->len++;
 	}
 
@@ -630,7 +630,7 @@  static int duplicate_policydb_cond_list(struct policydb *newp,
 {
 	int rc, i, j;
 
-	rc = avtab_duplicate(&newp->te_cond_avtab, &origp->te_cond_avtab);
+	rc = avtab_alloc_dup(&newp->te_cond_avtab, &origp->te_cond_avtab);
 	if (rc)
 		return rc;