| Message ID | 20200629221746.4033122-1-andriin@fb.com |
|---|---|
| State | Changes Requested |
| Delegated to: | BPF Maintainers |
| Headers | show |
| Series | [bpf] bpf: enforce BPF ringbuf size to be the power of 2 | expand |
On 6/29/20 3:17 PM, Andrii Nakryiko wrote: > BPF ringbuf assumes the size to be a multiple of page size and the power of > 2 value. The latter is important to avoid division while calculating position > inside the ring buffer and using (N-1) mask instead. This patch fixes omission > to enforce power-of-2 size rule. > > Fixes: 457f44363a88 ("bpf: Implement BPF ring buffer and verifier support for it") > Signed-off-by: Andrii Nakryiko <andriin@fb.com> Acked-by: Yonghong Song <yhs@fb.com>
On Mon, Jun 29, 2020 at 3:19 PM Andrii Nakryiko <andriin@fb.com> wrote: > > BPF ringbuf assumes the size to be a multiple of page size and the power of > 2 value. The latter is important to avoid division while calculating position > inside the ring buffer and using (N-1) mask instead. This patch fixes omission > to enforce power-of-2 size rule. > > Fixes: 457f44363a88 ("bpf: Implement BPF ring buffer and verifier support for it") > Signed-off-by: Andrii Nakryiko <andriin@fb.com> > --- > kernel/bpf/ringbuf.c | 5 +++-- > 1 file changed, 3 insertions(+), 2 deletions(-) > > diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c > index 180414bb0d3e..dcc8e8b9df10 100644 > --- a/kernel/bpf/ringbuf.c > +++ b/kernel/bpf/ringbuf.c > @@ -132,7 +132,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node) > { > struct bpf_ringbuf *rb; > > - if (!data_sz || !PAGE_ALIGNED(data_sz)) > + if (!is_power_of_2(data_sz) || !PAGE_ALIGNED(data_sz)) > return ERR_PTR(-EINVAL); What's the point checking the same value in two different places? The check below did that already. > #ifdef CONFIG_64BIT > @@ -166,7 +166,8 @@ static struct bpf_map *ringbuf_map_alloc(union bpf_attr *attr) > return ERR_PTR(-EINVAL); > > if (attr->key_size || attr->value_size || > - attr->max_entries == 0 || !PAGE_ALIGNED(attr->max_entries)) > + !is_power_of_2(attr->max_entries) || > + !PAGE_ALIGNED(attr->max_entries)) > return ERR_PTR(-EINVAL); > > rb_map = kzalloc(sizeof(*rb_map), GFP_USER); > -- > 2.24.1 >
On Mon, Jun 29, 2020 at 9:47 PM Alexei Starovoitov <alexei.starovoitov@gmail.com> wrote: > > On Mon, Jun 29, 2020 at 3:19 PM Andrii Nakryiko <andriin@fb.com> wrote: > > > > BPF ringbuf assumes the size to be a multiple of page size and the power of > > 2 value. The latter is important to avoid division while calculating position > > inside the ring buffer and using (N-1) mask instead. This patch fixes omission > > to enforce power-of-2 size rule. > > > > Fixes: 457f44363a88 ("bpf: Implement BPF ring buffer and verifier support for it") > > Signed-off-by: Andrii Nakryiko <andriin@fb.com> > > --- > > kernel/bpf/ringbuf.c | 5 +++-- > > 1 file changed, 3 insertions(+), 2 deletions(-) > > > > diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c > > index 180414bb0d3e..dcc8e8b9df10 100644 > > --- a/kernel/bpf/ringbuf.c > > +++ b/kernel/bpf/ringbuf.c > > @@ -132,7 +132,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node) > > { > > struct bpf_ringbuf *rb; > > > > - if (!data_sz || !PAGE_ALIGNED(data_sz)) > > + if (!is_power_of_2(data_sz) || !PAGE_ALIGNED(data_sz)) > > return ERR_PTR(-EINVAL); > > What's the point checking the same value in two different places? > The check below did that already. I was initially treating bpf_ringbuf_alloc() as a sort of internal API that some other code (outside of BPF map) might want to use. But I'll drop for now, it can always be added later. > > > #ifdef CONFIG_64BIT > > @@ -166,7 +166,8 @@ static struct bpf_map *ringbuf_map_alloc(union bpf_attr *attr) > > return ERR_PTR(-EINVAL); > > > > if (attr->key_size || attr->value_size || > > - attr->max_entries == 0 || !PAGE_ALIGNED(attr->max_entries)) > > + !is_power_of_2(attr->max_entries) || > > + !PAGE_ALIGNED(attr->max_entries)) > > return ERR_PTR(-EINVAL); > > > > rb_map = kzalloc(sizeof(*rb_map), GFP_USER); > > -- > > 2.24.1 > >
diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c index 180414bb0d3e..dcc8e8b9df10 100644 --- a/kernel/bpf/ringbuf.c +++ b/kernel/bpf/ringbuf.c @@ -132,7 +132,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node) { struct bpf_ringbuf *rb; - if (!data_sz || !PAGE_ALIGNED(data_sz)) + if (!is_power_of_2(data_sz) || !PAGE_ALIGNED(data_sz)) return ERR_PTR(-EINVAL); #ifdef CONFIG_64BIT @@ -166,7 +166,8 @@ static struct bpf_map *ringbuf_map_alloc(union bpf_attr *attr) return ERR_PTR(-EINVAL); if (attr->key_size || attr->value_size || - attr->max_entries == 0 || !PAGE_ALIGNED(attr->max_entries)) + !is_power_of_2(attr->max_entries) || + !PAGE_ALIGNED(attr->max_entries)) return ERR_PTR(-EINVAL); rb_map = kzalloc(sizeof(*rb_map), GFP_USER);
BPF ringbuf assumes the size to be a multiple of page size and the power of 2 value. The latter is important to avoid division while calculating position inside the ring buffer and using (N-1) mask instead. This patch fixes omission to enforce power-of-2 size rule. Fixes: 457f44363a88 ("bpf: Implement BPF ring buffer and verifier support for it") Signed-off-by: Andrii Nakryiko <andriin@fb.com> --- kernel/bpf/ringbuf.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)