seccomp_attr_set(3)

💖 seccomp_attr_set(3) — libseccomp Documentation 💖

okay bestie, pull up a chair because we're talking about seccomp_attr_set and seccomp_attr_get from libseccomp, and honestly? they're the little knobs that control how your whole seccomp filter behaves. no cap.

---

✨ NAME ✨

seccomp_attr_set, seccomp_attr_get — manage the seccomp filter attributes 🎀

---

✨ SYNOPSIS ✨

#include <seccomp.h>

typedef void * scmp_filter_ctx;
enum scmp_filter_attr;

int seccomp_attr_set(scmp_filter_ctx ctx,
                     enum scmp_filter_attr attr, uint32_t value);
int seccomp_attr_get(scmp_filter_ctx ctx,
                     enum scmp_filter_attr attr, uint32_t *value);

Link with -lseccomp.

---

✨ DESCRIPTION ✨

so basically, seccomp_attr_set() sets the different seccomp filter attributes, and seccomp_attr_get() fetches them. these attributes are tunable values that affect how the library behaves when it's generating and loading the seccomp filter into the kernel. 💅

and here's the thing — the attributes get reset to their default values whenever the filter is initialized or reset via seccomp_init(3) or seccomp_reset(3). so don't come crying to me when your settings vanish, bestie, that's just the vibe. 😭

the filter context ctx is the value returned by the call to seccomp_init(3).

valid attr values are as follows, and yes we're going through all of them:

• SCMP_FLTATR_ACT_DEFAULT — the default filter action as specified in the call to seccomp_init(3) or seccomp_reset(3). this attribute is read-only. 👑
• SCMP_FLTATR_ACT_BADARCH — the filter action taken when the loaded filter does not match the architecture of the executing application. defaults to the SCMP_ACT_KILL action. 💀
• SCMP_FLTATR_CTL_NNP — a flag to specify if the NO_NEW_PRIVS functionality should be enabled before loading the seccomp filter into the kernel. setting this to off (value == 0) results in no action, meaning that loading the seccomp filter into the kernel will fail if CAP_SYS_ADMIN is missing and NO_NEW_PRIVS has not been externally set. defaults to on (value == 1).
• SCMP_FLTATR_CTL_TSYNC — a flag to specify if the kernel should attempt to synchronize the filters across all threads on seccomp_load(3). if the kernel is unable to synchronize all of the thread then the load operation will fail. this flag is only available on Linux Kernel 3.17 or greater; attempting to enable this flag on earlier kernels will result in an error being returned. defaults to off (value == 0).
• SCMP_FLTATR_API_TSKIP — a flag to specify if libseccomp should allow filter rules to be created for the -1 syscall. the -1 syscall value can be used by tracer programs to skip specific syscall invocations, see seccomp(2) for more information. defaults to off (value == 0).
• SCMP_FLTATR_CTL_LOG — a flag to specify if the kernel should log all filter actions taken except for the SCMP_ACT_ALLOW action. defaults to off (value == 0).
• SCMP_FLTATR_CTL_SSB — a flag to disable Speculative Store Bypass mitigations for this filter. defaults to off (value == 0).
• SCMP_FLTATR_CTL_OPTIMIZE — a flag to specify the optimization level of the seccomp filter. by default libseccomp generates a set of sequential if statements for each rule in the filter. seccomp_syscall_priority(3) can be used to prioritize the order for the default cause. the binary tree optimization sorts by syscall numbers and generates consistent O(log n) filter traversal for every rule in the filter. the binary tree may be advantageous for large filters. note that seccomp_syscall_priority(3) is ignored when SCMP_FLTATR_CTL_OPTIMIZE == 2. 🙈

the different optimization levels are described below:

• 0 — Reserved value, not currently used.
• 1 — Rules sorted by priority and complexity (DEFAULT).
• 2 — Binary tree sorted by syscall number.

• SCMP_FLTATR_API_SYSRAWRC — a flag to specify if libseccomp should pass system error codes back to the caller instead of the default -ECANCELED. defaults to off (value == 0).
• SCMP_FLTATR_CTL_WAITKILL — a flag to specify if libseccomp should request wait killable semantics when possible. defaults to off (value == 0).

---

✨ RETURN VALUE ✨

returns zero on success, or one of the following error codes on failure:

• -EACCES — Setting the attribute with the given value is not allowed.
• -EEXIST — The attribute does not exist.
• -EINVAL — Invalid input, either the context or architecture token is invalid.
• -EOPNOTSUPP — The library doesn't support the particular operation.

---

✨ EXAMPLES ✨

here's the moment you've been waiting for, a whole little program:

#include <seccomp.h>

int main(int argc, char *argv[])
{
     int rc = -1;
     scmp_filter_ctx ctx;

     ctx = seccomp_init(SCMP_ACT_ALLOW);
     if (ctx == NULL)
          goto out;

     /* ... */

     rc = seccomp_attr_set(ctx, SCMP_FLTATR_ACT_BADARCH, SCMP_ACT_TRAP);
     if (rc < 0)
          goto out;

     /* ... */

out:
     seccomp_release(ctx);
     return -rc;
}

---

✨ NOTES ✨

while the seccomp filter can be generated independent of the kernel, kernel support is required to load and enforce the seccomp filter generated by libseccomp. so like, you can build the thing all you want, but the kernel has to be in on the plan. 🫶

the libseccomp project site, with more information and the source code repository, can be found at https://github.com/seccomp/libseccomp. this tool, as well as the libseccomp library, is currently under development, please report any bugs at the project site or directly to the author. 🐛

---

✨ AUTHOR ✨

Paul Moore <paul@paul-moore.com>

---

✨ SEE ALSO ✨

seccomp_init(3), seccomp_reset(3), seccomp_load(3), seccomp(2)

---

and that's the whole page, bestie 🌈 — brought to you by Paul Moore <paul@paul-moore.com>, dated 21 September 2022, and living in the libseccomp Documentation for seccomp_attr_set(3). now go set some attributes and slay ✨💅🌸