Thread (20 messages) 20 messages, 2 authors, 2021-12-03

Re: [RFC PATCH bpf-next v2 1/9] bpf: Introduce composable reg, ret and arg types.

From: Alexei Starovoitov <hidden>
Date: 2021-12-01 20:30:20

On Mon, Nov 29, 2021 at 05:29:40PM -0800, Hao Luo wrote:
quoted hunk ↗ jump to hunk
There are some common properties shared between bpf reg, ret and arg
values. For instance, a value may be a NULL pointer, or a pointer to
a read-only memory. Previously, to express these properties, enumeration
was used. For example, in order to test whether a reg value can be NULL,
reg_type_may_be_null() simply enumerates all types that are possibly
NULL. The problem of this approach is that it's not scalable and causes
a lot of duplication. These properties can be combined, for example, a
type could be either MAYBE_NULL or RDONLY, or both.

This patch series rewrites the layout of reg_type, arg_type and
ret_type, so that common properties can be extracted and represented as
composable flag. For example, one can write

 ARG_PTR_TO_MEM | PTR_MAYBE_NULL

which is equivalent to the previous

 ARG_PTR_TO_MEM_OR_NULL

The type ARG_PTR_TO_MEM are called "base types" in this patch. Base
types can be extended with flags. A flag occupies the higher bits while
base types sits in the lower bits.

This patch in particular sets up a set of macro for this purpose. The
followed patches rewrites arg_types, ret_types and reg_types
respectively.

Signed-off-by: Hao Luo <redacted>
---
 include/linux/bpf.h | 50 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 50 insertions(+)
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index cc7a0c36e7df..b592b3f7d223 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -297,6 +297,37 @@ bool bpf_map_meta_equal(const struct bpf_map *meta0,
 
 extern const struct bpf_map_ops bpf_map_offload_ops;
 
+/* bpf_type_flag contains a set of flags that are applicable to the values of
+ * arg_type, ret_type and reg_type. For example, a pointer value may be null,
+ * or a memory is read-only. We classify types into two categories: base types
+ * and extended types. Extended types are base types combined with a type flag.
+ *
+ * Currently there are no more than 32 base types in arg_type, ret_type and
+ * reg_types.
+ */
+#define BPF_BASE_TYPE_BITS	8
+
+enum bpf_type_flag {
+	/* PTR may be NULL. */
+	PTR_MAYBE_NULL		= BIT(0 + BPF_BASE_TYPE_BITS),
+
+	__BPF_TYPE_LAST_FLAG	= PTR_MAYBE_NULL,
+};
+
+#define BPF_BASE_TYPE_MASK	GENMASK(BPF_BASE_TYPE_BITS, 0)
+
+/* Max number of base types. */
+#define BPF_BASE_TYPE_LIMIT	(1UL << BPF_BASE_TYPE_BITS)
+
+/* Max number of all types. */
+#define BPF_TYPE_LIMIT		(__BPF_TYPE_LAST_FLAG | (__BPF_TYPE_LAST_FLAG - 1))
+
+/* extract base type. */
+#define BPF_BASE_TYPE(x)	((x) & BPF_BASE_TYPE_MASK)
+
+/* extract flags from an extended type. */
+#define BPF_TYPE_FLAG(x)	((enum bpf_type_flag)((x) & ~BPF_BASE_TYPE_MASK))
Overall I think it's really great.
The only suggestion is to use:
static inline u32 base_type(u32 x)
{
  return x & BPF_BASE_TYPE_MASK;
}
and
static inline u32 type_flag(u32 x) ..

The capital letter macros are too loud.

wdyt?
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help