Skip to contentSkip to conversation

seccomp: Make seccomp filter reusable

PATCH New Hengqi Chen 2 participants 5 messages Nov 29, 2023 – Dec 5, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

v3 cover seccomp: Make seccomp filter reusable

This patchset introduces a new operation and a new flag to split the SECCOMP_SET_MODE_FILTER process into two steps:
load filter and attach filter. Thus we can reuse the filter.

The new SECCOMP_LOAD_FILTER loads the filter and returns a fd which can be reused for different processes via UDS. With this new operation, we can eliminate a hot path of JITing BPF program (the filter) where we apply the same seccomp filter to thousands of micro VMs on a bare metal instance.

The new flag SECCOMP_FILTER_FLAG_FILTER_FD is used to attach a filter which is loaded via SECCOMP_LOAD_FILTER and represented by a seccomp filter fd.

Change logs:
v2 -> v3:
- Stick to cBPF, don't interfere with bpffs

v1 -> v2:
- Add a flag to attach filter from fd
- Update selftests

RFC -> v1:
- Addressed comments from Kees
- Reuse filter copy/create code (patch 1)
- Add a selftest (patch 4)

Hengqi Chen (3):
seccomp: Introduce SECCOMP_LOAD_FILTER operation
seccomp: Introduce new flag SECCOMP_FILTER_FLAG_FILTER_FD selftests/seccomp: Test seccomp filter load and attach

 include/linux/seccomp.h                       |  3 +-
 include/uapi/linux/seccomp.h                  |  3 +
 kernel/seccomp.c                              | 78 ++++++++++++++++++-
 tools/testing/selftests/seccomp/seccomp_bpf.c | 71 +++++++++++++++++
 4 files changed, 151 insertions(+), 4 deletions(-)

--
2.34.1

#1 lore raw

PATCH v3 1/3 seccomp: Introduce SECCOMP_LOAD_FILTER operation

This patch adds a new operation named SECCOMP_LOAD_FILTER. It accepts a sock_fprog the same as SECCOMP_SET_MODE_FILTER but only performs the loading process. If succeed, returns a new fd associated with the seccomp filter. The seccomp filter thus can be reused to attach to different processes.

---
 include/uapi/linux/seccomp.h |  1 +
 kernel/seccomp.c             | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 35 insertions(+)
include/uapi/linux/seccomp.h +1 -0
index dbfc9b37fcae..ee2c83697810 100644
@@ -16,6 +16,7 @@
 #define SECCOMP_SET_MODE_FILTER		1
 #define SECCOMP_GET_ACTION_AVAIL	2
 #define SECCOMP_GET_NOTIF_SIZES		3
+#define SECCOMP_LOAD_FILTER		4
 
 /* Valid flags for SECCOMP_SET_MODE_FILTER */
 #define SECCOMP_FILTER_FLAG_TSYNC		(1UL << 0)
kernel/seccomp.c +34 -0
index 255999ba9190..ef956c3d15c7 100644
@@ -1885,6 +1885,18 @@ static bool has_duplicate_listener(struct seccomp_filter *new_child)
 	return false;
 }
 
+static int seccomp_filter_put(struct inode *inode, struct file *file)
+{
+	struct seccomp_filter *filter = file->private_data;
+
+	__put_seccomp_filter(filter);
+	return 0;
+}
+
+static const struct file_operations seccomp_filter_fops = {
+	.release = seccomp_filter_put,
+};
+
 /**
  * seccomp_set_mode_filter: internal function for setting seccomp filter
  * @flags:  flags to change filter behavior
@@ -1996,12 +2008,29 @@ static long seccomp_set_mode_filter(unsigned int flags,
 	seccomp_filter_free(prepared);
 	return ret;
 }
+
+static long seccomp_load_filter(const char __user *filter)
+{
+	struct seccomp_filter *sfilter;
+
+	sfilter = seccomp_prepare_user_filter(filter);
+	if (IS_ERR(sfilter))
+		return PTR_ERR(sfilter);
+
+	return anon_inode_getfd("seccomp-filter", &seccomp_filter_fops,
+				sfilter, O_CLOEXEC);
+}
 #else
 static inline long seccomp_set_mode_filter(unsigned int flags,
 					   const char __user *filter)
 {
 	return -EINVAL;
 }
+
+static inline long seccomp_load_filter(const char __user *filter)
+{
+	return -EINVAL;
+}
 #endif
 
 static long seccomp_get_action_avail(const char __user *uaction)
@@ -2063,6 +2092,11 @@ static long do_seccomp(unsigned int op, unsigned int flags,
 			return -EINVAL;
 
 		return seccomp_get_notif_sizes(uargs);
+	case SECCOMP_LOAD_FILTER:
+		if (flags != 0)
+			return -EINVAL;
+
+		return seccomp_load_filter(uargs);
 	default:
 		return -EINVAL;
 	}
Signature · 1 line
2.34.1
#2 lore raw

PATCH v3 2/3 seccomp: Introduce new flag SECCOMP_FILTER_FLAG_FILTER_FD

In reply to Hengqi Chen

Add a new flag SECCOMP_FILTER_FLAG_FILTER_FD for SECCOMP_SET_MODE_FILTER. This indicates user supply a seccomp filter fd, not a sock_fprog. It allows us to attach the seccomp filter that is previously loaded via SECCOMP_LOAD_FILTER.

---
 include/linux/seccomp.h      |  3 ++-
 include/uapi/linux/seccomp.h |  2 ++
 kernel/seccomp.c             | 44 +++++++++++++++++++++++++++++++++---
 3 files changed, 45 insertions(+), 4 deletions(-)
include/linux/seccomp.h +2 -1
index 175079552f68..3257042c35fc 100644
@@ -9,7 +9,8 @@
 					 SECCOMP_FILTER_FLAG_SPEC_ALLOW | \
 					 SECCOMP_FILTER_FLAG_NEW_LISTENER | \
 					 SECCOMP_FILTER_FLAG_TSYNC_ESRCH | \
-					 SECCOMP_FILTER_FLAG_WAIT_KILLABLE_RECV)
+					 SECCOMP_FILTER_FLAG_WAIT_KILLABLE_RECV | \
+					 SECCOMP_FILTER_FLAG_FILTER_FD)
 
 /* sizeof() the first published struct seccomp_notif_addfd */
 #define SECCOMP_NOTIFY_ADDFD_SIZE_VER0 24
include/uapi/linux/seccomp.h +2 -0
index ee2c83697810..41e0ca56ef1c 100644
@@ -26,6 +26,8 @@
 #define SECCOMP_FILTER_FLAG_TSYNC_ESRCH		(1UL << 4)
 /* Received notifications wait in killable state (only respond to fatal signals) */
 #define SECCOMP_FILTER_FLAG_WAIT_KILLABLE_RECV	(1UL << 5)
+/* Indicates that the filter is a fd obtained from SECCOMP_LOAD_FILTER */
+#define SECCOMP_FILTER_FLAG_FILTER_FD		(1UL << 6)
 
 /*
  * All BPF programs must return a 32-bit value.
kernel/seccomp.c +41 -3

69 lines not on this page · raw message

Signature · 1 line
2.34.1
#3 lore raw

PATCH v3 3/3 selftests/seccomp: Test seccomp filter load and attach

In reply to Hengqi Chen

Add testcases to exercise the newly added seccomp filter load and attach functionalities.

---
 tools/testing/selftests/seccomp/seccomp_bpf.c | 71 +++++++++++++++++++
 1 file changed, 71 insertions(+)
tools/testing/selftests/seccomp/seccomp_bpf.c +71 -0
index 38f651469968..66eb72e6c1a3 100644
@@ -4735,6 +4735,77 @@ TEST(user_notification_wait_killable_fatal)
 	EXPECT_EQ(SIGTERM, WTERMSIG(status));
 }
 
+TEST(seccomp_filter_load_and_attach)
+{
+	struct sock_filter filter[] = {
+		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW),
+	};
+	struct sock_fprog prog = {
+		.len = (unsigned short)ARRAY_SIZE(filter),
+		.filter = filter,
+	};
+	int fd, ret, flags;
+
+	ret = prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
+	ASSERT_EQ(0, ret)
+	{
+		TH_LOG("Kernel does not support PR_SET_NO_NEW_PRIVS!");
+	}
+
+	flags = 0;
+	fd = seccomp(SECCOMP_LOAD_FILTER, flags, &prog);
+	ASSERT_GT(fd, -1);
+
+	flags = SECCOMP_FILTER_FLAG_FILTER_FD;
+	ret = seccomp(SECCOMP_SET_MODE_FILTER, flags, &fd);
+	ASSERT_EQ(ret, 0);
+
+	flags = SECCOMP_FILTER_FLAG_FILTER_FD;
+	ret = seccomp(SECCOMP_SET_MODE_FILTER, flags, &fd);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EEXIST);
+
+	flags = 0;
+	ret = seccomp(SECCOMP_SET_MODE_FILTER, flags, &prog);
+	ASSERT_EQ(ret, 0);
+
+	close(fd);
+}
+
+TEST(seccomp_attach_fd_failed)
+{
+	int fd, ret, flags;
+
+	ret = prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
+	ASSERT_EQ(0, ret)
+	{
+		TH_LOG("Kernel does not support PR_SET_NO_NEW_PRIVS!");
+	}
+
+	fd = socket(AF_UNIX, SOCK_STREAM, 0);
+	ASSERT_GT(fd, -1);
+
+	/* copy a sock_fprog from a fd */
+	flags = 0;
+	ret = seccomp(SECCOMP_SET_MODE_FILTER, flags, &fd);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EFAULT);
+
+	/* pass a non seccomp filter fd */
+	flags = SECCOMP_FILTER_FLAG_FILTER_FD;
+	ret = seccomp(SECCOMP_SET_MODE_FILTER, flags, &fd);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EINVAL);
+	close(fd);
+
+	/* pass a invalid fd */
+	fd = -1;
+	flags = SECCOMP_FILTER_FLAG_FILTER_FD;
+	ret = seccomp(SECCOMP_SET_MODE_FILTER, flags, &fd);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EBADF);
+}
+
 /*
  * TODO:
  * - expand NNP testing
Signature · 1 line
2.34.1
#4 lore raw

Re: [PATCH v3 3/3] selftests/seccomp: Test seccomp filter load and attach

kernel test robot noticed "kernel-selftests.seccomp.seccomp_bpf.fail" on:

Hello,

kernel test robot noticed "kernel-selftests.seccomp.seccomp_bpf.fail" on:

commit: 95084d9b2b5f0b593724819288f3cb4e2c585cb0 ("[PATCH v3 3/3] selftests/seccomp: Test seccomp filter load and attach")
url: https://github.com/intel-lab-lkp/linux/commits/Hengqi-Chen/seccomp-Introduce-SECCOMP_LOAD_FILTER-operation/20231129-134337
base: https://git.kernel.org/cgit/linux/kernel/git/kees/linux.git for-next/seccomp
patch link: https://lore.kernel.org/all/[email protected]/
patch subject: [PATCH v3 3/3] selftests/seccomp: Test seccomp filter load and attach

in testcase: kernel-selftests
version: kernel-selftests-x86_64-60acb023-1_20230329 with following parameters:

group: group-s

compiler: gcc-12
test machine: 36 threads 1 sockets Intel(R) Core(TM) i9-10980XE CPU @ 3.00GHz (Cascade Lake) with 32G memory

(please refer to attached dmesg/kmsg for entire log/backtrace)

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-lkp/[email protected]

we noticed one test added in this commit can pass, but another will fail.

# # RUN global.seccomp_filter_load_and_attach ...
# # OK global.seccomp_filter_load_and_attach

# ok 56 global.seccomp_filter_load_and_attach
# #  RUN           global.seccomp_attach_fd_failed ...
# # seccomp_bpf.c:4792:seccomp_attach_fd_failed:Expected errno (22) == EFAULT (14)
# # seccomp_attach_fd_failed: Test terminated by assertion
# #          FAIL  global.seccomp_attach_fd_failed
# not ok 57 global.seccomp_attach_fd_failed

...

# # FAILED: 97 / 98 tests passed.
# # Totals: pass:97 fail:1 xfail:0 xpass:0 skip:0 error:0 not ok 1 selftests: seccomp: seccomp_bpf # exit=1

we applied the patch set upon 31c65705a8cfa like below:

95084d9b2b5f0 (linux-review/Hengqi-Chen/seccomp-Introduce-SECCOMP_LOAD_FILTER-operation/20231129-134337) selftests/seccomp: Test seccomp filter load and attach 8fcda1c36e519 seccomp: Introduce new flag SECCOMP_FILTER_FLAG_FILTER_FD
bd86f21cfe1e0 seccomp: Introduce SECCOMP_LOAD_FILTER operation
31c65705a8cfa (kees/for-next/seccomp, kees/for-linus/seccomp) perf/benchmark: fix seccomp_unotify benchmark for 32-bit
ce9ecca0238b1 (tag: v6.6-rc2, hyperv/hyperv-next) Linux 6.6-rc2

not sure if this is the correct base, or is there any other dependency to run new test seccomp_attach_fd_failed?

The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20231205/[email protected]

Signature · 2 lines
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki