[RFC PATCH net-next 1/6] filter: refactor BPF JIT for seccomp filters

From: Xi Wang
Date: Fri Apr 26 2013 - 03:54:37 EST


Currently, bpf_jit_compile() and bpf_jit_free() takes an sk_filter,
which seccomp filters cannot reuse.

Change bpf_jit_compile() to take a pointer to BPF instructions and
the length, and to return a JITted function.

Change bpf_jit_free() to take a JITted function.

Add JIT calls for seccomp filters.

Signed-off-by: Xi Wang <xi.wang@xxxxxxxxx>
---
include/linux/filter.h | 16 ++++++++++------
kernel/seccomp.c | 6 +++++-
net/core/filter.c | 6 ++----
3 files changed, 17 insertions(+), 11 deletions(-)

diff --git a/include/linux/filter.h b/include/linux/filter.h
index d1248f4..8743093 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -21,12 +21,14 @@ struct compat_sock_fprog {
struct sk_buff;
struct sock;

+typedef unsigned int (*bpf_func_t)(const struct sk_buff *skb,
+ const struct sock_filter *filter);
+
struct sk_filter
{
atomic_t refcnt;
unsigned int len; /* Number of filter blocks */
- unsigned int (*bpf_func)(const struct sk_buff *skb,
- const struct sock_filter *filter);
+ bpf_func_t bpf_func;
struct rcu_head rcu;
struct sock_filter insns[0];
};
@@ -48,11 +50,12 @@ extern int sk_chk_filter(struct sock_filter *filter, unsigned int flen);
extern int sk_get_filter(struct sock *sk, struct sock_filter __user *filter, unsigned len);

#ifdef CONFIG_BPF_JIT
+#include <stdarg.h>
#include <linux/linkage.h>
#include <linux/printk.h>

-extern void bpf_jit_compile(struct sk_filter *fp);
-extern void bpf_jit_free(struct sk_filter *fp);
+extern bpf_func_t bpf_jit_compile(struct sock_filter *filter, unsigned int flen);
+extern void bpf_jit_free(bpf_func_t bpf_func);

static inline void bpf_jit_dump(unsigned int flen, unsigned int proglen,
u32 pass, void *image)
@@ -65,10 +68,11 @@ static inline void bpf_jit_dump(unsigned int flen, unsigned int proglen,
}
#define SK_RUN_FILTER(FILTER, SKB) (*FILTER->bpf_func)(SKB, FILTER->insns)
#else
-static inline void bpf_jit_compile(struct sk_filter *fp)
+static inline bpf_func_t bpf_jit_compile(struct sock_filter *filter, unsigned int flen)
{
+ return sk_run_filter;
}
-static inline void bpf_jit_free(struct sk_filter *fp)
+static inline void bpf_jit_free(bpf_func_t bpf_func)
{
}
#define SK_RUN_FILTER(FILTER, SKB) sk_run_filter(SKB, FILTER->insns)
diff --git a/kernel/seccomp.c b/kernel/seccomp.c
index 5af44b5..f784feb 100644
--- a/kernel/seccomp.c
+++ b/kernel/seccomp.c
@@ -55,6 +55,7 @@ struct seccomp_filter {
atomic_t usage;
struct seccomp_filter *prev;
unsigned short len; /* Instruction count */
+ bpf_func_t bpf_func;
struct sock_filter insns[];
};

@@ -211,7 +212,7 @@ static u32 seccomp_run_filters(int syscall)
* value always takes priority (ignoring the DATA).
*/
for (f = current->seccomp.filter; f; f = f->prev) {
- u32 cur_ret = sk_run_filter(NULL, f->insns);
+ u32 cur_ret = SK_RUN_FILTER(f, NULL);
if ((cur_ret & SECCOMP_RET_ACTION) < (ret & SECCOMP_RET_ACTION))
ret = cur_ret;
}
@@ -273,6 +274,8 @@ static long seccomp_attach_filter(struct sock_fprog *fprog)
if (ret)
goto fail;

+ filter->bpf_func = bpf_jit_compile(filter->insns, filter->len);
+
/*
* If there is an existing filter, make it the prev and don't drop its
* task reference.
@@ -330,6 +333,7 @@ void put_seccomp_filter(struct task_struct *tsk)
while (orig && atomic_dec_and_test(&orig->usage)) {
struct seccomp_filter *freeme = orig;
orig = orig->prev;
+ bpf_jit_free(freeme->bpf_func);
kfree(freeme);
}
}
diff --git a/net/core/filter.c b/net/core/filter.c
index dad2a17..0a7900b 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -643,7 +643,7 @@ void sk_filter_release_rcu(struct rcu_head *rcu)
{
struct sk_filter *fp = container_of(rcu, struct sk_filter, rcu);

- bpf_jit_free(fp);
+ bpf_jit_free(fp->bpf_func);
kfree(fp);
}
EXPORT_SYMBOL(sk_filter_release_rcu);
@@ -652,13 +652,11 @@ static int __sk_prepare_filter(struct sk_filter *fp)
{
int err;

- fp->bpf_func = sk_run_filter;
-
err = sk_chk_filter(fp->insns, fp->len);
if (err)
return err;

- bpf_jit_compile(fp);
+ fp->bpf_func = bpf_jit_compile(fp->insns, fp->len);
return 0;
}

--
1.8.1.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/