BPF-HELPERS(7)

📦

version 6.20 · 9 versions

6.20 · arch rolling, nixpkgs unstable (latest)
6.06 · centos-stream 10
6.19 · alpine 3.24, arch rolling, nixos 26.05, ubuntu stonking
5.10 · ubuntu jammy
6.7 · ubuntu noble
6.13 · alpine 3.22, fedora 43, fedora 44, opensuse-leap 16.0
6.04 · centos-stream 9, debian bookworm
6.17 · alpine 3.23, ubuntu resolute
6.9.1 · debian trixie, fedora 43, ubuntu questing

💖 BPF-HELPERS(7) 💖

okay so hi bestie 🫶 this is BPF-HELPERS, and it lives in the Miscellaneous Information Manual (yes that's where the lil (7) comes from, iykyk). It is literally the master list of every eBPF helper function your program is allowed to call, and it is long, so buckle up. ✨

🌸 DESCRIPTION 🌸

okay so the whole extended Berkeley Packet Filter — aka eBPF — situation is this: you write little programs in a pseudo-assembly language, you attach them to one of the many kernel hooks, and they run whenever a specific event happens. 🎀

this framework is different from the older, "classic" BPF (or "cBPF") in a bunch of ways, and one of those ways is that you can actually call special functions — the "helpers" — from inside your program. these functions are not a free-for-all bestie, they're restricted to a white-list of helpers defined in the kernel. periodt. 💅

here's the vibe: eBPF programs use helpers to talk to the system, or to whatever context they're running in. like, they can print debugging messages, get the time since the system booted, mess with eBPF maps, or manipulate network packets. and because there are several eBPF program types and they do not all run in the same context, each program type can only call a subset of those helpers. 🙈

also due to eBPF conventions, a helper can't have more than five arguments. that's the limit, no negotiating. ✋

internally, eBPF programs call directly into the compiled helper functions without needing any foreign-function interface at all. so calling helpers introduces no overhead — excellent performance, we love to see it. 😌

this document is an attempt to list and document the helpers available to eBPF developers. they're sorted by chronological order — the oldest helpers in the kernel are at the top. ⭐

👑 HELPERS 👑

🦋 `bpf_map_lookup_elem`

void *bpf_map_lookup_elem(struct bpf_map *map, const void *key)

Description — do a lookup in map for an entry associated to key. 🔍

Return — the map value associated to key, or NULL if no entry was found. 💀

🦋 `bpf_map_update_elem`

long bpf_map_update_elem(struct bpf_map *map, const void *key, const void *value, u64 flags)

Description — add or update the value of the entry associated to key in map with value. and flags is one of these iconic three:

BPF_NOEXIST — the entry for key must not exist in the map.
BPF_EXIST — the entry for key must already exist in the map.
BPF_ANY — no condition at all on the existence of the entry for key. do whatever, bestie.

⚠️ heads up: the flag value BPF_NOEXIST cannot be used for maps of types BPF_MAP_TYPE_ARRAY or BPF_MAP_TYPE_PERCPU_ARRAY (because all elements always exist there), and the helper would return an error. no cap. 🚫

Return — 0 on success, or a negative error in case of failure. ✅

🦋 `bpf_map_delete_elem`

long bpf_map_delete_elem(struct bpf_map *map, const void *key)

Description — delete the entry with key from map. bye bestie 👋

Return — 0 on success, or a negative error in case of failure. ✅

🦋 `bpf_probe_read`

long bpf_probe_read(void *dst, u32 size, const void *unsafe_ptr)

Description — for tracing programs: safely attempt to read size bytes from the kernel space address unsafe_ptr, and store the data in dst. 🧚

generally though, use bpf_probe_read_user() or bpf_probe_read_kernel() instead. that's the move. 💅

Return — 0 on success, or a negative error in case of failure. ✅

🦋 `bpf_ktime_get_ns`

u64 bpf_ktime_get_ns(void)

Description — returns the time elapsed since system boot, in nanoseconds. it does not include the time the system was suspended. see: clock_gettime(CLOCK_MONOTONIC). ⏰

Return — the current ktime. ⭐

🦋 `bpf_trace_printk`

long bpf_trace_printk(const char *fmt, u32 fmt_size, ...)

Description — this helper is a "printk()-like" facility for debugging. it prints a message defined by the format fmt (of size fmt_size) to the file /sys/kernel/tracing/trace from TraceFS, if available. it can take up to three additional u64 arguments (because as an eBPF helper, the total number of arguments is limited to five, remember? 👆).

every single time the helper is called, it appends a line to the trace. 🫠 lines are discarded while /sys/kernel/tracing/trace is open, so use /sys/kernel/tracing/trace_pipe to avoid that. the format of the trace is customizable, and the exact output you get depends on the options set in /sys/kernel/tracing/trace_options (see also the README file under the same directory). however, it usually defaults to something like:

telnet-470   [001] .N.. 419421.045894: 0x00000001: <formatted msg>

in the above:

telnet is the name of the current task.
470 is the PID of the current task.
001 is the CPU number on which the task is running.
in .N.., each character refers to a set of options (whether irqs are enabled, scheduling options, whether hard/softirqs are running, level of preempt_disabled respectively). N means that TIF_NEED_RESCHED and PREEMPT_NEED_RESCHED are set.
419421.045894 is a timestamp.
0x00000001 is a fake value used by BPF for the instruction pointer register.
<formatted msg> is the message formatted with fmt.

the conversion specifiers supported by fmt are similar, but more limited than for printk(). they are: %d, %i, %u, %x, %ld, %li, %lu, %lx, %lld, %lli, %llu, %llx, %p, %s. no modifier (size of field, padding with zeroes, etc.) is available, and the helper will return -EINVAL (but print nothing 😶) if it encounters an unknown specifier.

also, note that bpf_trace_printk() is slow, and should only be used for debugging purposes. for this reason, a notice block (spanning several lines) is printed to kernel logs and states that the helper should not be used "for production use" the first time this helper is used (or more precisely, when trace_printk() buffers are allocated). for passing values to user space, perf events should be preferred. 💯

Return — the number of bytes written to the buffer, or a negative error in case of failure. ✅

🦋 `bpf_get_prandom_u32`

u32 bpf_get_prandom_u32(void)

Description — get a pseudo-random number. 🎲

from a security point of view, this helper uses its own pseudo-random internal state, and cannot be used to infer the seed of other random functions in the kernel. however — and this is the important part bestie — it is essential to note that the generator used by the helper is not cryptographically secure. 🙈

Return — a random 32-bit unsigned value. ⭐

🦋 `bpf_get_smp_processor_id`

u32 bpf_get_smp_processor_id(void)

Description — get the SMP (symmetric multiprocessing) processor id. note that all programs run with migration disabled, which means the SMP processor id is stable during the entire execution of the program. no surprises, we love stability. 🫶

Return — the SMP id of the processor running the program. ⭐

🦋 `bpf_skb_store_bytes`

long bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from, u32 len, u64 flags)

Description — store len bytes from address from into the packet associated to skb, at offset. the flags are a combination of the following values:

BPF_F_RECOMPUTE_CSUM — automatically update skb->csum after storing the bytes.
BPF_F_INVALIDATE_HASH — set skb->hash, skb->swhash and skb->l4hash to 0.

⚠️ a call to this helper is susceptible to change the underlying packet buffer. therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access. 💀

Return — 0 on success, or a negative error in case of failure. ✅

🦋 `bpf_l3_csum_replace`

long bpf_l3_csum_replace(struct sk_buff *skb, u32 offset, u64 from, u64 to, u64 size)

Description — recompute the layer 3 (e.g. IP) checksum for the packet associated to skb. the computation is incremental, so the helper has to know the former value of the header field that was modified (from), the new value of this field (to), and the number of bytes (2 or 4) for this field, stored in size. 🌸

alternatively, it's possible to store the difference between the previous and the new values of the header field in to, by setting from and size to 0. for both methods, offset indicates the location of the IP checksum within the packet. 📍

this helper works in combination with bpf_csum_diff(), which does not update the checksum in-place, but offers more flexibility and can handle sizes larger than 2 or 4 for the checksum to update. 💅

⚠️ same warning as before, and it's important: a call to this helper is susceptible to change the underlying packet buffer. therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access. 😭

Return — 0 on success, or a negative error in case of failure. ✅

🦋 `bpf_l4_csum_replace`

long bpf_l4_csum_replace(struct sk_buff *skb, u32 offset, u64 from, u64 to, u64 flags)

...and that's literally where this part stops, bestie — right in the middle of the next helper 💀🫶 more parts are coming, we are not done, do not go anywhere. ✨

so this helper is bpf_l4_csum_replace(), and bestie it is so useful, lemme tell you 💖

Description ✨

Recompute the layer 4 checksum (like, TCP, UDP or ICMP vibes) for the packet that's associated to skb. The computation is incremental, okay? So the helper literally needs to know:

the former value of the header field that got modified — that's from
the new value of that same field — that's to
and how many bytes that field is (2 or 4), which lives in the lowest four bits of flags

Alternatively 🫶 you can just store the difference between the previous and the new values of the header field in to, by setting from and the four lowest bits of flags to 0. Two paths, one slay.

For both methods, offset tells you where the IP checksum is living inside the packet.

On top of the field size, flags can have actual flags added in (bitwise OR, no cap):

with BPF_F_MARK_MANGLED_0, a null checksum is left untouched (unless BPF_F_MARK_ENFORCE is added as well) — and for updates that result in a null checksum, the value gets set to CSUM_MANGLED_0 instead 👑
BPF_F_PSEUDO_HDR says the modified header field is part of the pseudo-header
BPF_F_IPV6 should be set for IPv6 packets, duh 💅

This helper works as a duo with bpf_csum_diff(), which doesn't update the checksum in-place — but it's more flexible and can handle sizes larger than 2 or 4 for the checksum to update. Pick your fighter.

⚠️ A call to this helper is susceptible to change the underlying packet buffer. So at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access. Not me warning you, it's just facts 😭

Return 💫

0 on success, or a negative error in case of failure.

`long bpf_tail_call(void ctx, struct bpf_map prog_array_map, u32 index)`

Description 🌈

This one is a special helper and it's used to trigger a "tail call" — aka, to jump into another eBPF program. The same stack frame is used (but the values on the stack and in the registers for the caller are not accessible to the callee, sorry bestie 🙈).

This mechanism lets you do program chaining, either to raise the maximum number of available eBPF instructions, or to execute given programs in conditional blocks. For security reasons, there's an upper limit to how many successive tail calls can be performed. Rules are rules.

Upon call of this helper, the program attempts to jump into a program referenced at index index in prog_array_map — which is a special map of type BPF_MAP_TYPE_PROG_ARRAY — and passes ctx, a pointer to the context.

If the call succeeds, the kernel immediately runs the first instruction of the new program. This is not a function call and it never returns to the previous program, periodt. If the call fails, the helper just has no effect and the caller keeps running its subsequent instructions like nothing happened 🦋

A call can fail if:

the destination program for the jump doesn't exist (i.e. index is superior to the number of entries in prog_array_map), or
the maximum number of tail calls has been reached for this chain of programs — that limit is defined in the kernel by the macro MAX_TAIL_CALL_CNT (not accessible to user space) and it's currently set to 33 💀

Return 💫

0 on success, or a negative error in case of failure.

`long bpf_clone_redirect(struct sk_buff *skb, u32 ifindex, u64 flags)`

Description 🎀

Clone and redirect the packet associated to skb to another net device of index ifindex. Both ingress and egress interfaces can be used for redirection — the BPF_F_INGRESS value in flags is what makes the distinction (ingress path is selected if the flag is present, egress path otherwise). And that's the only flag supported for now, so don't go looking for more 👑

Compared to the bpf_redirect() helper, bpf_clone_redirect() comes with the associated cost of duplicating the packet buffer — but it can be executed out of the eBPF program. Meanwhile bpf_redirect() is more efficient, but it's handled through an action code where the redirection only happens after the eBPF program has returned. Efficiency vs. immediacy, the eternal struggle 😭

⚠️ A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access. Yeah, this warning shows up a lot in this manpage. Get used to it bestie 💅

Return 💫

0 on success, or a negative error in case of failure. A positive error indicates a potential drop or congestion in the target device. The particular positive error codes are not defined — so like, don't go assuming things.

`u64 bpf_get_current_pid_tgid(void)`

Description 🌸

Get the current pid and tgid. That's it, that's the whole vibe.

Return 💫

A 64-bit integer containing the current tgid and pid, and created as such: current_task->tgid << 32 | current_task->pid.

`u64 bpf_get_current_uid_gid(void)`

Description 🌸

Get the current uid and gid. Iconic, simple, no notes.

Return 💫

A 64-bit integer containing the current GID and UID, and created as such: current_gid << 32 | current_uid.

`long bpf_get_current_comm(void *buf, u32 size_of_buf)`

Description 🦋

Copy the comm attribute of the current task into buf of size_of_buf. The comm attribute contains the name of the executable (excluding the path) for the current task. The size_of_buf must be strictly positive, no zeroes allowed.

On success, the helper makes sure that the buf is NUL-terminated. On failure, it's filled with zeroes. Either way you're getting something 🫶

Return 💫

0 on success, or a negative error in case of failure.

`u32 bpf_get_cgroup_classid(struct sk_buff *skb)`

Description 🧚

Retrieve the classid for the current task — i.e., for the net_cls cgroup to which skb belongs.

This helper can be used on the TC egress path, but not on ingress. Read that again, it matters.

The net_cls cgroup provides an interface to tag network packets based on a user-provided identifier for all traffic coming from the tasks belonging to the related cgroup. See also the related kernel documentation, available from the Linux sources in file Documentation/admin-guide/cgroup-v1/net_cls.rst. Some light reading, bestie 📖

The Linux kernel has two versions for cgroups: there are cgroups v1 and cgroups v2. Both are available to users, who can use a mixture of them — but note that the net_cls cgroup is for cgroup v1 only. This makes it incompatible with BPF programs run on cgroups, which is a cgroup-v2-only feature (a socket can only hold data for one version of cgroups at a time). Not me breaking the news 💀

This helper is only available if the kernel was compiled with the CONFIG_CGROUP_NET_CLASSID configuration option set to "y" or to "m".

Return 💫

The classid, or 0 for the default unconfigured classid.

`long bpf_skb_vlan_push(struct sk_buff *skb, __be16 vlan_proto, u16 vlan_tci)`

Description 🎀

Push a vlan_tci (that's VLAN tag control information, iykyk) of protocol vlan_proto to the packet associated to skb, then update the checksum. Note that if vlan_proto is different from ETH_P_8021Q and ETH_P_8021AD, it is considered to be ETH_P_8021Q. Snatched either way 💅

⚠️ A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access. Yes, again. It's a whole thing.

Return 💫

0 on success, or a negative error in case of failure.

`long bpf_skb_vlan_pop(struct sk_buff *skb)`

Description 🦋

Pop a VLAN header from the packet associated to skb. It's giving removal energy, we love to see it.

⚠️ A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access.

Return 💫

0 on success, or a negative error in case of failure.

`long bpf_skb_get_tunnel_key(struct sk_buff skb, struct bpf_tunnel_key key, u32 size, u64 flags)`

Description 🌈

Get tunnel metadata ✨ This helper takes a pointer key to an empty struct bpf_tunnel_key of size, that will be filled with tunnel metadata for the packet associated to skb. The flags can be set to BPF_F_TUNINFO_IPV6, which indicates that the tunnel is based on IPv6 protocol instead of IPv4.

The struct bpf_tunnel_key is an object that generalizes the principal parameters used by various tunneling protocols into a single struct. This way, it can be used to easily make a decision based on the contents of the encapsulation header, "summarized" in this struct. So convenient, highkey obsessed 🫶

In particular, it holds:

the IP address of the remote end (IPv4 or IPv6, depending on the case) in key->remote_ipv4 or key->remote_ipv6
key->tunnel_id, which is generally mapped to a VNI (Virtual Network Identifier), making it programmable together with the bpf_skb_set_tunnel_key() helper

Let's imagine that the following code is part of a program attached to the TC ingress interface, on one end of a GRE tunnel, and is supposed to filter out all messages coming from remote ends with IPv4 address other than 10.0.0.1:

int ret;
struct bpf_tunnel_key key = {};

ret = bpf_skb_get_tunnel_key(skb, &key, sizeof(key), 0);
if (ret < 0)
        return TC_ACT_SHOT;     // drop packet

if (key.remote_ipv4 != 0x0a000001)
        return TC_ACT_SHOT;     // drop packet

return TC_ACT_OK;               // accept packet

okay so this interface ALSO works with literally every encapsulation device that can run in "collect metadata" mode 💅 and bestie the glow-up is real: instead of having one whole network device per specific configuration, the "collect metadata" mode only needs a single device, and you just pull the config out of this helper. chef's kiss.

And it plays nice with all the tunnels: VXLan, Geneve, GRE or IP in IP (IPIP). the whole squad 🫶

Return

0 on success, or a negative error in case of failure. simple as.

`long bpf_skb_set_tunnel_key(struct sk_buff skb, struct bpf_tunnel_key key, u32 size, u64 flags)`

Description

Populate the tunnel metadata for the packet hanging out with skb. The tunnel metadata gets set to the contents of key, of size. And the flags?? oh we can set those to a combo of the following, no cap:

BPF_F_TUNINFO_IPV6 — says the tunnel is IPv6-based instead of IPv4. it's giving protocol upgrade ✨
BPF_F_ZERO_CSUM_TX — for IPv4 packets, adds a flag to the tunnel metadata saying "hey, skip the checksum computation and just set the checksum to zeroes" 🎀
BPF_F_DONT_FRAGMENT — adds a flag to the tunnel metadata indicating the packet should not be fragmented. leave it whole, diva 💅
BPF_F_SEQ_NUMBER — adds a flag to the tunnel metadata indicating a sequence number should be added to the tunnel header before the packet goes out. This flag was added for GRE encapsulation, but might get used with other protocols too in the future 👀
BPF_F_NO_TUNNEL_KEY — adds a flag to the tunnel metadata indicating that no tunnel key should be set in the resulting tunnel header. sometimes less is more 🦋

Here's a typical usage on the transmit path:

struct bpf_tunnel_key key;
     populate key ...
bpf_skb_set_tunnel_key(skb, &key, sizeof(key), 0);
bpf_clone_redirect(skb, vxlan_dev_ifindex, 0);

also go read the description of the bpf_skb_get_tunnel_key() helper for extra info, they're besties 💖

Return

0 on success, or a negative error in case of failure.

`u64 bpf_perf_event_read(struct bpf_map *map, u64 flags)`

Description

Read the value of a perf event counter 🔥 This helper is lowkey dependent on a map of type BPF_MAP_TYPE_PERF_EVENT_ARRAY. The nature of the perf event counter gets picked when map is updated with perf event file descriptors. The map is an array whose size is the number of available CPUs, and each cell holds a value relative to one CPU. The value you're retrieving is signaled by flags, which contains the index of the CPU to look up, masked with BPF_F_INDEX_MASK. Orrr alternatively, flags can be set to BPF_F_CURRENT_CPU to say "yeah just gimme the value for the current CPU, thanks bestie" 💁‍♀️

📢 ngl, heads up: before Linux 4.13, only hardware perf events could be retrieved.

also!! be aware that the newer helper bpf_perf_event_read_value() is the recommended one over bpf_perf_event_read() in general. The latter has some ABI quirks where error and counter value are used as a return code (which is straight up wrong to do since ranges may overlap 😭). That issue is fixed with bpf_perf_event_read_value(), which at the same time provides more features than the bpf_perf_event_read() interface. So go read the description of bpf_perf_event_read_value() for the deets, iykyk.

Return

The value of the perf event counter read from the map, or a negative error code in case of failure.

`long bpf_redirect(u32 ifindex, u64 flags)`

Description

Redirect the packet to another net device of index ifindex. This helper is kinda similar to bpf_clone_redirect(), except the packet is not cloned, which gives you increased performance. no copy, just vibes ✨

Except for XDP, both ingress and egress interfaces can be used for redirection. The BPF_F_INGRESS value in flags is what makes the distinction (ingress path gets picked if the flag is present, egress path otherwise). RN though, XDP only supports redirection to the egress interface, and accepts no flag at all. rude but okay 💀

The same effect can also be attained with the more generic bpf_redirect_map(), which uses a BPF map to store the redirect target instead of handing it directly to the helper. iconic behaviour honestly 👑

Return

For XDP, the helper returns XDP_REDIRECT on success or XDP_ABORTED on error. For other program types, the values are TC_ACT_REDIRECT on success or TC_ACT_SHOT on error.

`u32 bpf_get_route_realm(struct sk_buff *skb)`

Description

Retrieve the realm or the route, that is to say the tclassid field of the destination for the skb. The identifier you get back is a user-provided tag, similar to the one used with the net_cls cgroup (see the description for the bpf_get_cgroup_classid() helper), but over here this tag is held by a route (a destination entry), not by a task. different holder, same energy 💅

Retrieving this identifier works with the clsact TC egress hook (see also tc-bpf(8)), or alternatively on conventional classful egress qdiscs, but not on the TC ingress path 🚫. In the case of the clsact TC egress hook, this has the advantage that, internally, the destination entry hasn't been dropped yet in the transmit path. So the destination entry does not need to be artificially held via netif_keep_dst() for a classful qdisc until the skb is freed. holding on for dear life? not required 💖

⚠️ this helper is available only if the kernel was compiled with the CONFIG_IP_ROUTE_CLASSID configuration option. no option, no helper, periodt.

Return

The realm of the route for the packet associated to skb, or 0 if none was found.

`long bpf_perf_event_output(void ctx, struct bpf_map map, u64 flags, void *data, u64 size)`

Description

Write a raw data blob into a special BPF perf event held by map of type BPF_MAP_TYPE_PERF_EVENT_ARRAY. This perf event must have the following attributes: PERF_SAMPLE_RAW as sample_type, PERF_TYPE_SOFTWARE as type, and PERF_COUNT_SW_BPF_OUTPUT as config. very specific, very demure 🎀

The flags are used to indicate the index in map for which the value must be put, masked with BPF_F_INDEX_MASK. Or alternatively, flags can be set to BPF_F_CURRENT_CPU to indicate that the index of the current CPU core should be used. yes queen, know your core ✨

The value to write, of size, is passed through the eBPF stack and pointed by data.

The context of the program, ctx, also needs to be passed to the helper. don't forget it, it'll be sad 😭

On user space, a program that wants to read the values needs to call perf_event_open() on the perf event (either for one or for all CPUs) and store the file descriptor into the map. This must be done before the eBPF program can send data into it. An example lives in the file samples/bpf/trace_output_user.c in the Linux kernel source tree (the eBPF program counterpart is in samples/bpf/trace_output.bpf.c). 🦋

bpf_perf_event_output() achieves better performance than bpf_trace_printk() for sharing data with user space, and is way better suited for streaming data out of eBPF programs. slay.

Also note this helper is not restricted to tracing use cases — it can be used with programs attached to TC or XDP as well, where it lets you pass data to user space listeners. And the data can be:

Only custom structs,
Only the packet payload, or
A combination of both.

pick your fighter 💜

Return

0 on success, or a negative error in case of failure.

`long bpf_skb_load_bytes(const void skb, u32 offset, void to, u32 len)`

Description

This helper was provided as an easy way to load data from a packet. You can use it to load len bytes from offset from the packet associated to skb, into the buffer pointed by to. easy peasy 🍓

Since Linux 4.7, usage of this helper has mostly been replaced by "direct packet access", which lets packet data be manipulated with skb->data and skb->data_end pointing respectively to the first byte of packet data and to the byte after the last byte of packet data. However, it remains useful if you want to read large quantities of data at once from a packet into the eBPF stack. still got a job, still getting paid 💅

Return

0 on success, or a negative error in case of failure.

`long bpf_get_stackid(void ctx, struct bpf_map map, u64 flags)`

Description

Walk a user or a kernel stack and return its id. To make this happen, the helper needs ctx, which is a pointer to the context the tracing program is executed on, and a pointer to a map of type BPF_MAP_TYPE_STACK_TRACE. say cheese bestie 📸

The last argument, flags, holds the number of stack frames to skip (from 0 to 255), masked with BPF_F_SKIP_FIELD_MASK. The next bits can be used to set a combo of the following flags:

BPF_F_USER_STACK — collect a user space stack instead of a kernel stack.
BPF_F_FAST_STACK_CMP — just compare stacks by their hash, that's it, that's the whole vibe 💅
BPF_F_REUSE_STACKID — if two totally different stacks happen to hash into the same stackid, the old one gets yeeted and discarded. bye bestie 🫶

The stack id you get back is a 32-bit long integer handle, and you can absolutely mix it with other stuff (including other stack ids) and use the whole combo as a key into maps. Super handy for cooking up all kinds of graphs — think flame graphs 🔥 or off-cpu graphs.

Okay so for actually walking a stack, this helper is a huge upgrade over bpf_probe_read(), which you can pair with unrolled loops, but ngl it's not efficient and it eats a ton of eBPF instructions. Instead, bpf_get_stackid() can collect up to PERF_MAX_STACK_DEPTH of both kernel and user frames. Heads up though: that limit is controlled with the sysctl program, and you should manually bump it up if you're profiling long user stacks (like stacks for Java programs fr fr). To do that, use:

# sysctl kernel.perf_event_max_stack=<new value>

Return — the positive (or null) stack id on success, or a negative error if it all goes sideways 😭

`s64 bpf_csum_diff(__be32 from, u32 from_size, __be32 to, u32 to_size, __wsum seed)`

Description

This one computes a checksum difference. It goes from the raw buffer pointed at by from, of length from_size (and lowkey that must be a multiple of 4), towards the raw buffer pointed at by to, of size to_size (same rule, girl, multiple of 4). Then there's an optional seed you can drop into the value — and yes it cascades, so the seed can come from a previous call to this same helper. ✨

It's flexible enough to be used a few different ways:

With from_size == 0, to_size > 0 and seed set to checksum, it can be used when pushing new data.
With from_size > 0, to_size == 0 and seed set to checksum, it can be used when removing data from a packet.
With from_size > 0, to_size > 0 and seed set to 0, it can be used to compute a diff. And note that from_size and to_size do not need to be equal. 💫

You can also team this helper up with bpf_l3_csum_replace() and bpf_l4_csum_replace(), feeding them the difference you computed with bpf_csum_diff(). It's giving teamwork.

Return — the checksum result, or a negative error code if something failed.

`long bpf_skb_get_tunnel_opt(struct sk_buff skb, void opt, u32 size)`

Description

Grabs the tunnel options metadata for the packet tied to skb, and stashes the raw tunnel option data into the buffer opt of size. 🎀

This helper works with encapsulation devices that can run in "collect metadata" mode (go peek at the related note in the description of bpf_skb_get_tunnel_key() for the tea on that). One iconic example: pairing it with the Geneve encapsulation protocol, where it lets you push (with the bpf_skb_get_tunnel_opt() helper) and retrieve arbitrary TLVs (Type-Length-Value headers) straight from the eBPF program. Full customization of those headers, no cap. 👑

Return — the size of the option data retrieved.

`long bpf_skb_set_tunnel_opt(struct sk_buff skb, void opt, u32 size)`

Description

Sets the tunnel options metadata for the packet tied to skb to the option data sitting in the raw buffer opt of size.

Also go check the description of the bpf_skb_get_tunnel_opt() helper for extra info. 🦋

Return — 0 on success, or a negative error if it fails.

`long bpf_skb_change_proto(struct sk_buff *skb, __be16 proto, u64 flags)`

Description

Changes the protocol of the skb to proto. Right now the supported moves are the glow-up from IPv4 to IPv6, and the reverse, IPv6 to IPv4. The helper handles all the groundwork for the transition, resize of the socket buffer included. Your eBPF program is expected to fill in the new headers (if there are any) via skb_store_bytes(), and to recompute the checksums with bpf_l3_csum_replace() and bpf_l4_csum_replace(). The main character use case here is doing NAT64 operations out of an eBPF program. 💜

Internally, the GSO type is marked as dodgy on purpose so that headers get checked and segments get recalculated by the GSO/GRO engine. The size for the GSO target is adapted too.

Warning bestie: all values for flags are reserved for future usage and must be left at zero. Do not touch. 🙈

A call to this helper is susceptible to changing the underlying packet buffer. So at load time, literally every check on pointers the verifier did earlier gets invalidated and must be redone — that matters if you're using this helper alongside direct packet access.

Return — 0 on success, or a negative error if it fails.

`long bpf_skb_change_type(struct sk_buff *skb, u32 type)`

Description

Changes the packet type for the packet tied to skb. Basically this = setting skb->pkt_type to type, except the eBPF program has no write access to skb->pkt_type other than through this helper. Using a helper here means errors get handled gracefully. 🌸

The big use case is changing incoming skbs to PACKET_HOST programmatically, instead of having to recirculate via redirect(..., BPF_F_INGRESS), for example.

Note that type only takes certain values. At this time, they are:

PACKET_HOST — packet is for us. 💖
PACKET_BROADCAST — send packet to all.
PACKET_MULTICAST — send packet to group.
PACKET_OTHERHOST — send packet to someone else.

Return — 0 on success, or a negative error if it fails.

`long bpf_skb_under_cgroup(struct sk_buff skb, struct bpf_map map, u32 index)`

Description

Checks whether skb is a descendant of the cgroup2 held by map of type BPF_MAP_TYPE_CGROUP_ARRAY, at index.

Return — the return value depends on how the test goes, and can be:

0, if the skb failed the cgroup2 descendant test.
1, if the skb succeeded the cgroup2 descendant test.
A negative error code, if an error occurred.

`u32 bpf_get_hash_recalc(struct sk_buff *skb)`

Description

Retrieves the hash of the packet, skb->hash. If it's not set — especially if the hash got cleared because of mangling — it recomputes that hash. After that, later accesses to the hash can just go directly through skb->hash. ✨

Calling bpf_set_hash_invalid(), changing a packet prototype with bpf_skb_change_proto(), or calling bpf_skb_store_bytes() with the BPF_F_INVALIDATE_HASH flag are all actions that are susceptible to clearing the hash and triggering a fresh computation on the next call to bpf_get_hash_recalc().

Return — the 32-bit hash.

`u64 bpf_get_current_task(void)`

Description

Gets the current task.

Return — a pointer to the current task struct.

`long bpf_probe_write_user(void dst, const void src, u32 len)`

Description

Attempts, in a safe way, to write len bytes from the buffer src to dst in memory. It only works for threads that are in user context, and dst must be a valid user space address. 🧚

This helper should not be used to implement any kind of security mechanism because of TOCTOU attacks — use it instead for debugging, diverting, and manipulating execution of semi-cooperative processes.

Also keep in mind: this feature is meant for experiments, and it has a risk of crashing the system and the running programs. 😭 So when an eBPF program using this helper gets attached, a warning including PID and process name is printed to kernel logs. You've been warned, bestie.

Return — 0 on success, or a negative error if it fails.

`long bpf_current_task_under_cgroup(struct bpf_map *map, u32 index)`

Description

Checks whether the probe being run is in the context of a given subset of the cgroup2 hierarchy. The cgroup2 to test is held by map of type BPF_MAP_TYPE_CGROUP_ARRAY, at index.

Return — the return value depends on how the test goes, and can be:

1, if the current task belongs to the cgroup2.
0, if the current task does not belong to the cgroup2.
A negative error code, if an error occurred.

`long bpf_skb_change_tail(struct sk_buff *skb, u32 len, u64 flags)`

Description

Resizes (trims or grows) the packet tied to skb to the new len. The flags are reserved for future usage and must be left at zero. 🫶

The basic idea: the helper does the work needed to change the size of the packet, then the eBPF program rewrites the rest via helpers like bpf_skb_store_bytes(), bpf_l3_csum_replace(), bpf_l3_csum_replace() and others. This helper is a slow path utility intended for replies with control messages. And because it's targeted at the slow path, the helper itself can afford to be slow: it implicitly linearizes, unclones and drops offloads from the skb. It's giving "no rush, I got time". 💅

A call to this helper is susceptible to change the underlying packet buffer. So at load time, every single check the verifier previously did on pointers gets invalidated, and it all has to be redone — that's the price of using this helper together with direct packet access, bestie. No skipping that step. 🙈

Return — 0 on success, or a negative error in case of failure. That's it, that's the whole vibe. 🎀

`long bpf_skb_pull_data(struct sk_buff *skb, u32 len)`

Description

Pulls in non-linear data when the skb is non-linear and not all of len are part of the linear section. Basically: make len bytes from skb readable and writable. And if you pass a zero for len? Then every byte in the linear part of skb gets made readable and writable. 💅

Bestie, listen, this helper is only needed for reading and writing with direct packet access. That's its entire job description.

So here's the tea on direct packet access: testing that the offsets you're accessing are within packet boundaries (i.e. testing on skb->data_end) is susceptible to fail if those offsets are invalid, or if the data you requested lives in non-linear parts of the skb. On failure the program can just bail out — or, in the case of a non-linear buffer, use a helper to make the data available. The bpf_skb_load_bytes() helper is the first solution for accessing the data. Another one is using bpf_skb_pull_data to pull in the non-linear parts once, then retest and eventually access the data. Two paths, pick your fighter. ✨

At the same time — and this is lowkey the cool part — this also makes sure the skb is uncloned, which is a necessary condition for direct write. But since it only needs to be an invariant for the write part, the verifier detects writes and adds a prologue that calls bpf_skb_pull_data() to effectively unclone the skb from the very beginning, in case it is indeed cloned. The kernel is doing the most so you don't have to. 👑

A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access.

Return — 0 on success, or a negative error in case of failure. 🌸

`s64 bpf_csum_update(struct sk_buff *skb, __wsum csum)`

Description

Adds the checksum csum into skb->csum — but only in the case where the driver has supplied a checksum for the entire packet into that field. Otherwise it returns an error. So check your driver first, babes. 💀

This helper is meant to be used together with bpf_csum_diff(), especially when the checksum needs updating after data has been written into the packet through direct packet access. They're a duo, a little bestie pair, iykyk. 🫶

Return — The checksum on success, or a negative error code in case of failure.

`void bpf_set_hash_invalid(struct sk_buff *skb)`

Description

Invalidate the current skb->hash. Use it after you've been mangling on headers through direct packet access, so you can signal that the hash is outdated and trigger a recalculation the next time the kernel tries to access this hash — or whenever the bpf_get_hash_recalc() helper gets called. Manifesting a fresh hash. ✨

Return — void. Nothing. Zero. Silch. (Not nothing as in "returns 0", nothing as in returns nothing at all, okay? Different things! 🙈)

`long bpf_get_numa_node_id(void)`

Description

Returns the id of the current NUMA node. The primary use case for this helper is picking sockets for the local NUMA node, when the program is attached to sockets using the SO_ATTACH_REUSEPORT_EBPF option (peep socket(7) for more on that). But ngl, the helper is also available to other eBPF program types, similarly to bpf_get_smp_processor_id(). Equal opportunity node-id-ing. 🌈

Return — The id of current NUMA node.

`long bpf_skb_change_head(struct sk_buff *skb, u32 len, u64 flags)`

Description

Grows the headroom of the packet associated to skb and adjusts the offset of the MAC header accordingly, adding len bytes of space. It automatically extends and reallocates memory as required. Very self-sufficient of it. 💅

This helper can be used on a layer 3 skb to push a MAC header, for redirection into a layer 2 device.

All values for flags are reserved for future usage, and must be left at zero. Don't get creative with it, bestie, it will not slay.

A call to this helper is susceptible to change the underlying packet buffer. So at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access.

Return — 0 on success, or a negative error in case of failure.

`long bpf_xdp_adjust_head(struct xdp_buff *xdp_md, int delta)`

Description

Adjust (read: move) xdp_md->data by delta bytes. And yes, bb, it is possible to use a negative value for delta. This helper can be used to prepare the packet for pushing or popping headers. Going both directions, we love a flexible icon. 🦋

Return — 0 on success, or a negative error in case of failure.

`long bpf_probe_read_str(void dst, u32 size, const void unsafe_ptr)`

Description

Copies a NUL terminated string from an unsafe kernel address unsafe_ptr to dst. Go read bpf_probe_read_kernel_str() for more details, it's the full lore drop. 📖

Generally though — use bpf_probe_read_user_str() or bpf_probe_read_kernel_str() instead. This one's giving legacy, and not in the cute way.

Return — On success, the strictly positive length of the string, including the trailing NUL character. On error, a negative value.

`u64 bpf_get_socket_cookie(struct sk_buff *skb)`

Description

If the struct sk_buff pointed by skb has a known socket, retrieve the cookie (generated by the kernel) of this socket. If no cookie has been set yet, generate a new one. And once generated, the socket cookie remains stable for the life of the socket — no take-backs, no glow-ups, no rebrands. 🌸

This helper is useful for monitoring per socket networking traffic statistics, since it provides a global socket identifier that can be assumed unique. It's the socket's social security number fr fr. 🎀

Return — A 8-byte long unique number on success, or 0 if the socket field is missing inside skb.

`u64 bpf_get_socket_cookie(struct bpf_sock_addr *ctx)`

Description

Equivalent to the bpf_get_socket_cookie() helper that accepts skb, but this one gets the socket from the struct bpf_sock_addr context. Same energy, different doorway. 🚪

Return — A 8-byte long unique number.

`u64 bpf_get_socket_cookie(struct bpf_sock_ops *ctx)`

Description

Equivalent to the bpf_get_socket_cookie() helper that accepts skb, but this one gets the socket from the struct bpf_sock_ops context. Yes bestie, it's the same helper name — the kernel is just built different. 💅

Return — A 8-byte long unique number.

`u64 bpf_get_socket_cookie(struct sock *sk)`

Description

Equivalent to the bpf_get_socket_cookie() helper that accepts sk, but this one gets the socket from a BTF struct sock. Oh, and bonus: this helper also works for sleepable programs. Naps included. 😴✨

Return — A 8-byte long unique number or 0 if sk is NULL.

`u32 bpf_get_socket_uid(struct sk_buff *skb)`

Description

Get the owner UID of the socket associated to skb. (Yes, the original spells it "socked", we don't talk about it. 🙈)

Return — The owner UID of the socket associated to skb. But heads up: if the socket is NULL, or if it is not a full socket (i.e. if it is a time-wait or a request socket instead), the overflowuid value is returned. And note that overflowuid might also be the actual UID value for the socket — so you genuinely cannot tell those two cases apart from the return value alone. Not confusing at all, right? 💀

`long bpf_set_hash(struct sk_buff *skb, u32 hash)`

Description

Set the full hash for skb — that means setting the field skb->hash — to the value hash. Done and dusted. 🎀

Return — 0

`long bpf_setsockopt(void bpf_socket, int level, int optname, void optval, int optlen)`

Description

Emulate a call to setsockopt() on the socket associated to bpf_socket, which must be a full socket. You have to specify the level at which the option resides and the name optname of the option — peep setsockopt(2) for more info. The option value of length optlen is pointed by optval. Very "same call, different font". ✨

bpf_socket should be one of the following:

struct bpf_sock_ops for BPF_PROG_TYPE_SOCK_OPS.
struct bpf_sock_addr for BPF_CGROUP_INET4_CONNECT, BPF_CGROUP_INET6_CONNECT and BPF_CGROUP_UNIX_CONNECT.

This helper actually implements a subset of setsockopt(). It supports the following levels:

SOL_SOCKET, which supports the following optnames: SO_RCVBUF, SO_SNDBUF, SO_MAX_PACING_RATE, SO_PRIORITY, SO_RCVLOWAT, SO_MARK, SO_BINDTODEVICE, SO_KEEPALIVE, SO_REUSEADDR, SO_REUSEPORT, SO_BINDTOIFINDEX, SO_TXREHASH.
IPPROTO_TCP, which supports the following optnames: TCP_CONGESTION, TCP_BPF_IW, TCP_BPF_SNDCWND_CLAMP, TCP_SAVE_SYN, TCP_KEEPIDLE, TCP_KEEPINTVL, TCP_KEEPCNT, TCP_SYNCNT, TCP_USER_TIMEOUT, TCP_NOTSENT_LOWAT, TCP_NODELAY, TCP_MAXSEG, TCP_WINDOW_CLAMP, TCP_THIN_LINEAR_TIMEOUTS, TCP_BPF_DELACK_MAX, TCP_BPF_RTO_MIN, TCP_BPF_SOCK_OPS_CB_FLAGS.
IPPROTO_IP, which supports optname IP_TOS.
IPPROTO_IPV6, which supports the following optnames: IPV6_TCLASS, IPV6_AUTOFLOWLABEL.

Return — 0 on success, or a negative error in case of failure.

`long bpf_skb_adjust_room(struct sk_buff *skb, s32 len_diff, u32 mode, u64 flags)`

Description

Grow or shrink the room for data in the packet associated to skb by len_diff, and according to the selected mode.

So, back to bpf_skb_adjust_room — by default, bestie, this helper is gonna reset any offloaded checksum indicator on the skb back to CHECKSUM_NONE. Lowkey rude, right? But you can dodge that with this one flag:

BPF_F_ADJ_ROOM_NO_CSUM_RESET — don't touch it. Do not reset the offloaded checksum data of the skb to CHECKSUM_NONE. Leave it be. 💅

There are two supported modes at this moment in time (yes, only two, gatekeeping is real):

BPF_ADJ_ROOM_MAC — adjusts room at the mac layer, meaning room space gets added or removed between the layer 2 and layer 3 headers.
BPF_ADJ_ROOM_NET — adjusts room at the network layer, so room space gets added or removed between the layer 3 and layer 4 headers.

And the flags that are supported rn, in order, no skips:

BPF_F_ADJ_ROOM_FIXED_GSO — do not adjust gso_size. And heads up: adjusting mss this way is not allowed for datagrams, periodt.
BPF_F_ADJ_ROOM_ENCAP_L3_IPV4, BPF_F_ADJ_ROOM_ENCAP_L3_IPV6 — any new space gets reserved to hold a tunnel header. Configure skb offsets and other fields accordingly.
BPF_F_ADJ_ROOM_ENCAP_L4_GRE, BPF_F_ADJ_ROOM_ENCAP_L4_UDP — use these with the ENCAP_L3 flags to further specify the tunnel type.
BPF_F_ADJ_ROOM_ENCAP_L2(len) — use with the ENCAP_L3/ENCAP_L4 flags to further specify the tunnel type; len is the length of the inner MAC header.
BPF_F_ADJ_ROOM_ENCAP_L2_ETH — use with the BPF_F_ADJ_ROOM_ENCAP_L2 flag to further specify the L2 type as Ethernet.
BPF_F_ADJ_ROOM_DECAP_L3_IPV4, BPF_F_ADJ_ROOM_DECAP_L3_IPV6 — these indicate the new IP header version after decapsulating the outer IP header. Used when the inner and outer IP versions are different. ✨ Important: these flags only trigger a protocol change without clearing any tunnel-specific GSO flags.
BPF_F_ADJ_ROOM_DECAP_L4_GRE — clear GRE tunnel GSO flags (SKB_GSO_GRE and SKB_GSO_GRE_CSUM) when decapsulating a GRE tunnel.
BPF_F_ADJ_ROOM_DECAP_L4_UDP — clear UDP tunnel GSO flags (SKB_GSO_UDP_TUNNEL and SKB_GSO_UDP_TUNNEL_CSUM) when decapsulating a UDP tunnel.
BPF_F_ADJ_ROOM_DECAP_IPXIP4 — clear the IPIP/SIT tunnel GSO flag (SKB_GSO_IPXIP4) when decapsulating a tunnel with an outer IPv4 header (IPv4-in-IPv4 or IPv6-in-IPv4).
BPF_F_ADJ_ROOM_DECAP_IPXIP6 — clear the IPv6 encapsulation tunnel GSO flag (SKB_GSO_IPXIP6) when decapsulating a tunnel with an outer IPv6 header (IPv6-in-IPv6 or IPv4-in-IPv6).

Okay so when you're using those decapsulation flags up there, the skb->encapsulation flag gets automatically cleared if all the tunnel-specific GSO flags (SKB_GSO_UDP_TUNNEL, SKB_GSO_UDP_TUNNEL_CSUM, SKB_GSO_GRE, SKB_GSO_GRE_CSUM, SKB_GSO_IPXIP4, SKB_GSO_IPXIP6) have been removed from the packet. That's how it handles the cases where every single tunnel layer has been decapsulated. 🧚

⚠️ psst — a call to this helper is susceptible to change the underlying packet buffer. So at load time, all checks on pointers that the verifier previously did get invalidated and have to be performed again, if you're using this helper in combination with direct packet access.

Returns — 0 on success, or a negative error in case of failure. 😭

long bpf_redirect_map(struct bpf_map *map, u64 key, u64 flags)

Description

Redirect that packet to the endpoint referenced by map at index key. Depending on its type, the map can hold references to net devices (for forwarding packets through other ports), or to CPUs (for redirecting XDP frames to another CPU — but as of this writing that's only implemented for native XDP, i.e. with driver support). 🦋

The lower two bits of flags get used as the return code if the map lookup fails. That's intentional, so the return value can be one of the XDP program return codes up to XDP_TX, as chosen by the caller. The higher bits of flags can be set to BPF_F_BROADCAST or BPF_F_EXCLUDE_INGRESS as defined below.

With BPF_F_BROADCAST the packet gets broadcasted to all the interfaces in the map, and with BPF_F_EXCLUDE_INGRESS the ingress interface gets excluded when doing that broadcasting.

See also bpf_redirect(), which only supports redirecting to an ifindex but doesn't require a map to do so. iykyk. 💜

Returns — XDP_REDIRECT on success, or the value of the two lower bits of the flags argument on error.

long bpf_sk_redirect_map(struct sk_buff *skb, struct bpf_map *map, u32 key, u64 flags)

Description

Redirects the packet to the socket referenced by map (which is of type BPF_MAP_TYPE_SOCKMAP) at index key. Both ingress and egress interfaces can be used for redirection, iconic. The BPF_F_INGRESS value in flags is what makes that distinction — the ingress path is selected if the flag is present, the egress path otherwise. And that's literally the only flag supported for now, no others, don't come for me. 🙈

Returns — SK_PASS on success, or SK_DROP on error.

long bpf_sock_map_update(struct bpf_sock_ops *skops, struct bpf_map *map, void *key, u64 flags)

Description

Adds an entry to, or updates, a map referencing sockets. The skops is used as the new value for the entry associated to key. And flags is one of these three, pick wisely:

BPF_NOEXIST — the entry for key must not exist in the map.
BPF_EXIST — the entry for key must already exist in the map.
BPF_ANY — no condition on the existence of the entry for key. Whatever, bestie. 🎀

If the map has eBPF programs (parser and verdict), those get inherited by the socket being added. But — if the socket is already attached to eBPF programs, this results in an error. Not the fun kind.

Returns — 0 on success, or a negative error in case of failure.

long bpf_xdp_adjust_meta(struct xdp_buff *xdp_md, int delta)

Description

Adjusts the address pointed by xdp_md->data_meta by delta (which can be positive or negative). 👑 Note that this operation modifies the address stored in xdp_md->data, so the latter must be loaded only after the helper has been called. Order matters, fr fr.

The use of xdp_md->data_meta is optional and programs are not required to use it. The rationale is that when the packet is processed with XDP (e.g. as a DoS filter), it's possible to push further meta data along with it before passing it to the stack, and to give the guarantee that an ingress eBPF program attached as a TC classifier on the same device can pick this up for further post-processing. Since TC works with socket buffers, it remains possible to set from XDP the mark or priority pointers, or other pointers for the socket buffer. Having this scratch space generic and programmable allows for more flexibility, because the user is free to store whatever meta data they need. Main character energy tbh. ✨

⚠️ also! A call to this helper is susceptible to change the underlying packet buffer. So at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access.

Returns — 0 on success, or a negative error in case of failure.

long bpf_perf_event_read_value(struct bpf_map *map, u64 flags, struct bpf_perf_event_value *buf, u32 buf_size)

Description

Reads the value of a perf event counter and stores it into buf of size buf_size. 🌸 This helper relies on a map of type BPF_MAP_TYPE_PERF_EVENT_ARRAY. The nature of the perf event counter is selected when map is updated with perf event file descriptors. The map is an array whose size is the number of available CPUs, and each cell contains a value relative to one CPU. The value to retrieve is indicated by flags, which contains the index of the CPU to look up, masked with BPF_F_INDEX_MASK. Alternatively, flags can be set to BPF_F_CURRENT_CPU to indicate that the value for the current CPU should be retrieved. 🫶

This helper behaves in a way close to the bpf_perf_event_read() helper, save that instead of just returning the value observed, it fills the buf structure. That allows for additional data to be retrieved: in particular, the enabled and running times (in buf->enabled and buf->running, respectively) get copied. In general, bpf_perf_event_read_value() is recommended over bpf_perf_event_read(), which has some ABI issues and provides fewer functionalities. 💅

These values are interesting because hardware PMU (Performance Monitoring Unit) counters are limited resources. When there are more PMU based perf events opened than there are available counters, the kernel will multiplex these events so each event gets a certain percentage (but not all) of the PMU time. In case that multiplexing happens, the number of samples or the counter value will not reflect the case compared to when no multiplexing occurs. And that makes comparison between different runs difficult. 😭 Typically, the counter value should be normalized before comparing to other experiments. The usual normalization is done as follows:

normalized_counter = counter * t_enabled / t_running

Where t_enabled is the time enabled for event and t_running is the time running for event since the last normalization. Both of those times get accumulated all the way back since the perf event was opened, so they just keep stacking. Now — if you wanna get a scaling factor between two separate invocations of an eBPF program? Bestie, use the CPU id as the key (which is literally the typical way the perf array usage model works anyway 🫶), stash the previous value there, and do the math right inside your eBPF program. Main character move.

Returns 0 on success, or a negative error if it flopped. 😭

`long bpf_perf_prog_read_value(struct bpf_perf_event_data ctx, struct bpf_perf_event_value buf, u32 buf_size)`

Description

So for an eBPF program that's attached to a perf event, this one grabs the value of the event counter tied to ctx and drops it into the structure that buf points at, size buf_size. ✨ Enabled and running times also get stored in that structure too — check the description of the helper bpf_perf_event_read_value() for the deets on that.

Return

0 on success, or a negative error in case of failure. 💅

`long bpf_getsockopt(void bpf_socket, int level, int optname, void optval, int optlen)`

Description

This one emulates a call to getsockopt() on the socket that bpf_socket is hooked up to — and it has to be a full socket, no cap. You gotta tell it the level where the option lives and the optname of the option, see getsockopt(2) for more. The retrieved value lands in the structure pointed to by opval, length optlen. 🌸

bpf_socket should be one of these, periodt:

struct bpf_sock_ops for BPF_PROG_TYPE_SOCK_OPS.
struct bpf_sock_addr for BPF_CGROUP_INET4_CONNECT, BPF_CGROUP_INET6_CONNECT and BPF_CGROUP_UNIX_CONNECT.

Heads up: this helper is genuinely only implementing a subset of getsockopt(). It supports the same set of optnames that bpf_setsockopt() supports. The exceptions — the drama — are that TCP_BPF_* is bpf_setsockopt() only, and TCP_SAVED_SYN is bpf_getsockopt() only. 💀

Return

0 on success, or a negative error in case of failure.

`long bpf_override_return(struct pt_regs *regs, u64 rc)`

Description

Error injection era 💉 — this helper uses kprobes to override the return value of the probed function and sets it to rc. The first argument is the context regs that the kprobe works on.

How does it work, you ask? It sets the PC (program counter) to an override function that runs in place of the original probed function. Which means — the probed function does not run at all. The replacement function just returns with the value you asked for. Iconic, slightly terrifying.

⚠️ This helper has security implications and is therefore subject to restrictions. It's only available if the kernel was compiled with the CONFIG_BPF_KPROBE_OVERRIDE configuration option, and even then it only works on functions tagged with ALLOW_ERROR_INJECTION in the kernel code.

Return

0

`long bpf_sock_ops_cb_flags_set(struct bpf_sock_ops *bpf_sock, int argval)`

Description

It attempts to set the value of the bpf_sock_ops_cb_flags field for the full TCP socket associated to bpf_sock_ops to argval. 🎀

The main purpose of this field is to decide whether there should be calls to eBPF programs of type BPF_PROG_TYPE_SOCK_OPS at various points in the TCP code. A program of the same type can change its value, per connection and whenever it needs to, once the connection is established. You can read this field directly, but you must use this helper for updates — that's how it can return an error if an eBPF program tries to set a callback that the current kernel doesn't support. No pretending bestie 💅

argval is a flag array, and it can combine these flags:

BPF_SOCK_OPS_RTO_CB_FLAG (retransmission time out)
BPF_SOCK_OPS_RETRANS_CB_FLAG (retransmission)
BPF_SOCK_OPS_STATE_CB_FLAG (TCP state change)
BPF_SOCK_OPS_RTT_CB_FLAG (every RTT)

So you can also use this function to clear a callback flag by setting the appropriate bit to zero. e.g. to disable the RTO callback:

bpf_sock_ops_cb_flags_set(bpf_sock,
       bpf_sock->bpf_sock_ops_cb_flags &
       ~BPF_SOCK_OPS_RTO_CB_FLAG)

And here are some examples of where you could call such an eBPF program: ⭐

When RTO fires.
When a packet is retransmitted.
When the connection terminates.
When a packet is sent.
When a packet is received.

Return

-EINVAL if the socket isn't a full TCP socket; otherwise, a positive number containing the bits that could not be set gets returned (which comes down to 0 if all bits were set as required — slay ✨).

`long bpf_msg_redirect_map(struct sk_msg_buff msg, struct bpf_map map, u32 key, u64 flags)`

Description

This helper lives in the world of programs implementing policies at the socket level. If the message msg is allowed to pass (i.e. if the verdict eBPF program returns SK_PASS ✅), redirect it to the socket referenced by map (of type BPF_MAP_TYPE_SOCKMAP) at index key. Both ingress and egress interfaces can be used for redirection. The BPF_F_INGRESS value in flags is what makes the distinction — ingress path gets selected if the flag is present, egress path otherwise. That's literally the only flag supported for now, bestie.

Return

SK_PASS on success, or SK_DROP on error. 😭

`long bpf_msg_apply_bytes(struct sk_msg_buff *msg, u32 bytes)`

Description

For socket policies: apply the verdict of the eBPF program to the next bytes (number of bytes) of message msg. 🌸

For example, this helper can be used in these cases:

A single sendmsg() or sendfile() system call contains multiple logical messages that the eBPF program is supposed to read and for which it should apply a verdict.
An eBPF program only cares to read the first bytes of a msg. If the message has a large payload, then setting up and calling the eBPF program repeatedly for all bytes — even though the verdict is already known — would just create unnecessary overhead.

When it's called from within an eBPF program, the helper sets a counter internal to the BPF infrastructure, and that counter is what applies the last verdict to the next bytes. Now watch this part carefully:

If bytes is smaller than the current data being processed from a sendmsg() or sendfile() system call, the first bytes will be sent and the eBPF program will be re-run with the pointer for start of data pointing to byte number bytes + 1.
If bytes is larger than the current data being processed, then the eBPF verdict will be applied to multiple sendmsg() or sendfile() calls until bytes are consumed.

Ngl, if a socket closes while the internal counter is holding a non-zero value, that's not a problem — because data isn't being buffered for bytes, it's sent as it is received. 🫶

Return

0

`long bpf_msg_cork_bytes(struct sk_msg_buff *msg, u32 bytes)`

Description

For socket policies: stop the execution of the verdict eBPF program for message msg until bytes (byte number) have been accumulated. 🛑

Use this when you need a specific number of bytes before a verdict can even be assigned, even if the data spans multiple sendmsg() or sendfile() calls. The extreme case? A user calling sendmsg() over and over with 1-byte long message segments. Obviously that's terrible for performance, but it is still valid. And if your eBPF program needs bytes bytes to validate a header, this helper is the one that stops the eBPF program from being called again until those bytes have been accumulated.

Return

0

`long bpf_msg_pull_data(struct sk_msg_buff *msg, u32 start, u32 end, u64 flags)`

Description

For socket policies: pull in non-linear data from user space for msg, and set the pointers msg->data and msg->data_end to the start and end byte offsets into msg, respectively. 🧚

Here's the lore. If a program of type BPF_PROG_TYPE_SK_MSG runs on a msg, it can only parse data that the (data, data_end) pointers have already consumed. For sendmsg() hooks this is likely the first scatterlist element. But for calls relying on the sendpage handler (e.g. sendfile()), this will be the range (0, 0) — because the data is shared with user space, and by default the whole objective is to avoid letting user space modify the data while (or after) the eBPF verdict is being decided. So this helper can be used to pull in data and to set the start and end pointer to given values. Data will be copied if necessary (i.e. if the data was not linear and if the start and end pointers don't point to the same chunk).

⚠️ A call to this helper is susceptible to changing the underlying packet buffer. So at load time, all the checks on pointers previously done by the verifier are invalidated and must be performed again — that's on you if you're using this helper in combination with direct packet access. 🙈

All values for flags are reserved for future usage, and must be left at zero. 👑

Return

0 on success, or a negative error in case of failure. ✨

🎀 `bpf_bind` — the "everyone shares ONE ip, bestie" helper 🎀

long bpf_bind(struct bpf_sock_addr *ctx, struct sockaddr *addr, int addr_len)

💖 Description

Bind the socket that belongs to ctx to the address that addr is pointing at, and the length of that address is addr_len. This is your glow-up for making an outgoing connection from the exact IP address you want — which is so useful when, like, every single process inside a cgroup is supposed to use one shared IP address on a host that has multiple IPs configured. One IP, one vibe, no confusion. 💅

It works for both IPv4 and IPv6, and for both TCP and UDP sockets. The domain (addr->sa_family) must be AF_INET (or AF_INET6). And here's the tea: it's advised that you pass a zero port (sin_port or sin6_port), which triggers IP_BIND_ADDRESS_NO_PORT-like behavior and lets the kernel efficiently pick up an unused port — as long as the 4-tuple is unique. Passing a non-zero port might lead to degraded performance. So like… don't. 🙈

✨ Return

0 on success, or a negative error in case of failure.

🌸 `bpf_xdp_adjust_tail` — move that `data_end`, periodt 🌸

long bpf_xdp_adjust_tail(struct xdp_buff *xdp_md, int delta)

💖 Description

Adjust (move) xdp_md->data_end by delta bytes. You can both shrink and grow the packet tail — shrinking is done by delta being a negative integer. Yes bestie, negatives are just shrinking with extra steps. 😌

⚠️ A call to this helper is susceptible to change the underlying packet buffer. So at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again if this helper is used in combination with direct packet access. Not me warning you twice — the verifier literally does that. 💀

✨ Return

0 on success, or a negative error in case of failure.

🦋 `bpf_skb_get_xfrm_state` — XFRM state, but make it accessible 🦋

long bpf_skb_get_xfrm_state(struct sk_buff *skb, u32 index, struct bpf_xfrm_state *xfrm_state, u32 size, u64 flags)

💖 Description

Retrieve the XFRM state (that's IP transform framework, btw — see also ip-xfrm(8)) at index in the XFRM "security path" for skb.

The retrieved value gets stored in the struct bpf_xfrm_state pointed to by xfrm_state, of length size.

All values for flags are reserved for future usage and must be left at zero. Not "basically zero", not "spiritually zero" — zero. 🫶

This helper is available only if the kernel was compiled with the CONFIG_XFRM configuration option.

✨ Return

0 on success, or a negative error in case of failure.

💜 `bpf_get_stack` — stack traces, but from inside the program 💜

long bpf_get_stack(void *ctx, void *buf, u32 size, u64 flags)

💖 Description

Return a user or a kernel stack into a buffer that your BPF program provides. To do that, the helper needs ctx, which is a pointer to the context on which the tracing program is executed. And to actually store the stacktrace, your BPF program hands over buf with a nonnegative size. Give it somewhere to live, bestie. 🏠

The last argument, flags, holds the number of stack frames to skip (from 0 to 255), masked with BPF_F_SKIP_FIELD_MASK. The next bits can be used to set these flags:

BPF_F_USER_STACK — collect a user space stack instead of a kernel stack.
BPF_F_USER_BUILD_ID — collect (build_id, file_offset) instead of ips for the user stack. Only valid if BPF_F_USER_STACK is also specified.
- file_offset is an offset relative to the beginning of the executable or shared object file backing the vma which the ip falls in. It is not an offset relative to that object's base address. So accordingly, it must be adjusted by adding (sh_addr - sh_offset), where sh_{addr,offset} correspond to the executable section containing file_offset in the object, for comparisons to symbols' st_value to be valid. Iykyk. 💅

bpf_get_stack() can collect up to PERF_MAX_STACK_DEPTH both kernel and user frames, subject to a sufficient large buffer size. Lowkey important: this limit can be controlled with the sysctl program, and it should be manually increased in order to profile long user stacks (such as stacks for Java programs). To do so, use:

# sysctl kernel.perf_event_max_stack=<new value>

✨ Return

The non-negative copied buf length equal to or less than size on success, or a negative error in case of failure.

🌈 `bpf_skb_load_bytes_relative` — load bytes, but pick your header first 🌈

long bpf_skb_load_bytes_relative(const void *skb, u32 offset, void *to, u32 len, u32 start_header)

💖 Description

This helper is similar to bpf_skb_load_bytes() in that it gives you an easy way to load len bytes from offset from the packet associated to skb, into the buffer pointed to by to. The difference from bpf_skb_load_bytes() is that a fifth argument, start_header, exists in order to select a base offset to start from. start_header can be one of:

BPF_HDR_START_MAC — the base offset to load data from is skb's mac header.
BPF_HDR_START_NET — the base offset to load data from is skb's network header.

In general, "direct packet access" is the preferred method to access packet data. However, this helper is in particular useful in socket filters where skb->data does not always point to the start of the mac header and where "direct packet access" is not available. Niche? Yes. Iconic? Also yes. 👑

✨ Return

0 on success, or a negative error in case of failure.

👑 `bpf_fib_lookup` — the big one, FIB lookup with main character energy 👑

long bpf_fib_lookup(void *ctx, struct bpf_fib_lookup *params, int plen, u32 flags)

💖 Description

Do a FIB lookup in the kernel tables using the parameters in params. If the lookup is successful and the result shows the packet is to be forwarded, the neighbor tables are searched for the nexthop. If that's successful too — meaning the FIB lookup shows forwarding and the nexthop is resolved — then the nexthop address is returned in ipv4_dst or ipv6_dst based on family, smac is set to the mac address of the egress device, dmac is set to the nexthop mac address, rt_metric is set to the metric from the route (IPv4/IPv6 only), and ifindex is set to the device index of the nexthop from the FIB lookup. It's giving full delivery. 📦

The plen argument is the size of the passed in struct. The flags argument can be a combination of one or more of the following values:

BPF_FIB_LOOKUP_DIRECT — do a direct table lookup vs full lookup using FIB rules.
BPF_FIB_LOOKUP_TBID — used with BPF_FIB_LOOKUP_DIRECT. Use the routing table ID present in params->tbid for the fib lookup.
BPF_FIB_LOOKUP_OUTPUT — perform lookup from an egress perspective (default is ingress).
BPF_FIB_LOOKUP_SKIP_NEIGH — skip the neighbour table lookup. params->dmac and params->smac will not be set as output. A common use case is to call bpf_redirect_neigh() after doing bpf_fib_lookup().
BPF_FIB_LOOKUP_SRC — derive and set the source IP addr in params->ipv{4,6}_src for the nexthop. If the src addr cannot be derived, BPF_FIB_LKUP_RET_NO_SRC_ADDR is returned. In this case, params->dmac and params->smac are not set either.
BPF_FIB_LOOKUP_MARK — use the mark present in params->mark for the fib lookup. This option should not be used with BPF_FIB_LOOKUP_DIRECT, as it only has meaning for full lookups.
BPF_FIB_LOOKUP_VLAN — if the fib lookup resolves to a VLAN device whose parent is a real (non-VLAN) device, set params->h_vlan_proto and params->h_vlan_TCI from the VLAN device and replace params->ifindex with the parent's ifindex. params->h_vlan_TCI carries the VID only, with PCP and DEI bits zero; a consumer wanting to set egress priority writes PCP itself. params->smac is the VLAN device's own address, which can differ from the parent's. Only the immediate parent is resolved; if it is itself a VLAN device (QinQ) or in another namespace, the egress cannot be reduced to a physical device plus one tag and the lookup returns BPF_FIB_LKUP_RET_VLAN_FAILURE with params->ifindex left at the input. To obtain the VLAN device's own ifindex, repeat the lookup without BPF_FIB_LOOKUP_VLAN, re-initializing params first: output fields overwrite the inputs they share storage with. The swap and the vlan fields are written only on success; other output fields keep the helper's existing behaviour, so a frag-needed result still reports the route mtu in params->mtu_result. This flag is only valid for XDP programs; tc programs receive -EINVAL since they can redirect to the VLAN device directly.

🫶 Genuine heads-up on that last one: re-initialize params before you repeat the lookup, because the output fields will happily overwrite the inputs they share storage with. She's confident, she's thorough, she will absolutely overwrite your inputs. 💀

BPF_FIB_LOOKUP_VLAN_INPUT — bestie this one is ✨fancy✨. It treats params->h_vlan_proto and params->h_vlan_TCI like an input VLAN tag and runs the lookup as if ingress had happened on the VLAN subinterface carrying that tag on params->ifindex. The VID is the low 12 bits of params->h_vlan_TCI, and params->h_vlan_proto must be ETH_P_8021Q or ETH_P_8021AD in network byte order — otherwise you get -EINVAL, no discussion 💀. If params->ifindex is itself a VLAN device, its inner (QinQ) subinterface is matched; for a bond or team, pass the master's ifindex. An unmatched tag, a down device, or one in another namespace returns BPF_FIB_LKUP_RET_NOT_FWDED, mirroring real ingress. A VID of 0 is looked up literally, so do not set this flag for priority-tagged frames. Cannot be combined with BPF_FIB_LOOKUP_TBID or BPF_FIB_LOOKUP_OUTPUT (returns -EINVAL).

ctx is either struct xdp_md for XDP programs or struct sk_buff tc cls_act programs.

Return 💫

< 0 if any input argument is invalid
0 on success (packet is forwarded, nexthop neighbor exists)
> 0 one of BPF_FIB_LKUP_RET_ codes explaining why the packet is not forwarded or needs assist from full stack

If lookup fails with BPF_FIB_LKUP_RET_FRAG_NEEDED, then the MTU was exceeded and output params->mtu_result contains the MTU. (Yeah, it hands you the MTU, slay.)

`long bpf_sock_hash_update(struct bpf_sock_ops skops, struct bpf_map map, void *key, u64 flags)`

Description

Add an entry to, or update a sockhash map referencing sockets. The skops is used as a new value for the entry associated to key. flags is one of:

BPF_NOEXIST — the entry for key must not exist in the map.
BPF_EXIST — the entry for key must already exist in the map.
BPF_ANY — no condition on the existence of the entry for key. ✨chaos mode✨

If the map has eBPF programs (parser and verdict), those will be inherited by the socket being added. If the socket is already attached to eBPF programs, this results in an error. 🙈

Return 0 on success, or a negative error in case of failure.

`long bpf_msg_redirect_hash(struct sk_msg_buff msg, struct bpf_map map, void *key, u64 flags)`

Description

This helper is used in programs implementing policies at the socket level. If the message msg is allowed to pass (i.e. if the verdict eBPF program returns SK_PASS), redirect it to the socket referenced by map (of type BPF_MAP_TYPE_SOCKHASH) using hash key. Both ingress and egress interfaces can be used for redirection. The BPF_F_INGRESS value in flags is used to make the distinction (ingress path is selected if the flag is present, egress path otherwise). This is the only flag supported for now. 🎀

Return SK_PASS on success, or SK_DROP on error.

`long bpf_sk_redirect_hash(struct sk_buff skb, struct bpf_map map, void *key, u64 flags)`

Description

This helper is used in programs implementing policies at the skb socket level. If the sk_buff skb is allowed to pass (i.e. if the verdict eBPF program returns SK_PASS), redirect it to the socket referenced by map (of type BPF_MAP_TYPE_SOCKHASH) using hash key. Both ingress and egress interfaces can be used for redirection. The BPF_F_INGRESS value in flags is used to make the distinction (ingress path is selected if the flag is present, egress otherwise). This is the only flag supported for now. 🦋

Return SK_PASS on success, or SK_DROP on error.

`long bpf_lwt_push_encap(struct sk_buff skb, u32 type, void hdr, u32 len)`

Description

Encapsulate the packet associated to skb within a Layer 3 protocol header. This header is provided in the buffer at address hdr, with len its size in bytes. type indicates the protocol of the header and can be one of:

BPF_LWT_ENCAP_SEG6 — IPv6 encapsulation with Segment Routing Header (struct ipv6_sr_hdr). hdr only contains the SRH, the IPv6 header is computed by the kernel. the kernel does the math for you 💅
BPF_LWT_ENCAP_SEG6_INLINE — only works if skb contains an IPv6 packet. Insert a Segment Routing Header (struct ipv6_sr_hdr) inside the IPv6 header.
BPF_LWT_ENCAP_IP — IP encapsulation (GRE/GUE/IPIP/etc). The outer header must be IPv4 or IPv6, followed by zero or more additional headers, up to LWT_BPF_MAX_HEADROOM total bytes in all prepended headers. Please note that if skb_is_gso(skb) is true, no more than two headers can be prepended, and the inner header, if present, should be either GRE or UDP/GUE. ⚠️ read that twice bestie

BPF_LWT_ENCAP_SEG6* types can be called by BPF programs of type BPF_PROG_TYPE_LWT_IN; BPF_LWT_ENCAP_IP type can be called by bpf programs of types BPF_PROG_TYPE_LWT_IN and BPF_PROG_TYPE_LWT_XMIT.

A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access. 😭 the verifier said "trust nothing"

Return 0 on success, or a negative error in case of failure.

`long bpf_lwt_seg6_store_bytes(struct sk_buff skb, u32 offset, const void from, u32 len)`

Description

Store len bytes from address from into the packet associated to skb, at offset. Only the flags, tag and TLVs inside the outermost IPv6 Segment Routing Header can be modified through this helper. 🚫 no touching anything else

A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access.

Return 0 on success, or a negative error in case of failure.

`long bpf_lwt_seg6_adjust_srh(struct sk_buff *skb, u32 offset, s32 delta)`

Description

Adjust the size allocated to TLVs in the outermost IPv6 Segment Routing Header contained in the packet associated to skb, at position offset by delta bytes. Only offsets after the segments are accepted. delta can be as well positive (growing) as negative (shrinking). ✨stretchy✨

A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access.

Return 0 on success, or a negative error in case of failure.

`long bpf_lwt_seg6_action(struct sk_buff skb, u32 action, void param, u32 param_len)`

Description

Apply an IPv6 Segment Routing action of type action to the packet associated to skb. Each action takes a parameter contained at address param, and of length param_len bytes. action can be one of:

SEG6_LOCAL_ACTION_END_X — End.X action: endpoint with Layer-3 cross-connect. Type of param: struct in6_addr.
SEG6_LOCAL_ACTION_END_T — End.T action: endpoint with specific IPv6 table lookup. Type of param: int.
SEG6_LOCAL_ACTION_END_B6 — End.B6 action: endpoint bound to an SRv6 policy. Type of param: struct ipv6_sr_hdr.
SEG6_LOCAL_ACTION_END_B6_ENCAP — End.B6.Encap action: endpoint bound to an SRv6 encapsulation policy. Type of param: struct ipv6_sr_hdr.

A call to this helper is susceptible to change the underlying packet buffer. Therefore, at load time, all checks on pointers previously done by the verifier are invalidated and must be performed again, if the helper is used in combination with direct packet access.

Return 0 on success, or a negative error in case of failure.

`long bpf_rc_repeat(void *ctx)`

Description

This helper is used in programs implementing IR decoding, to report a successfully decoded repeat key message. This delays the generation of a key up event for previously generated key down event. 🛋️ hold the button, bestie

Some IR protocols like NEC have a special IR message for repeating last button, for when a button is held down.

The ctx should point to the lirc sample as passed into the program.

This helper is only available is the kernel was compiled with the CONFIG_BPF_LIRC_MODE2 configuration option set to "y".

Return 0

`long bpf_rc_keydown(void *ctx, u32 protocol, u64 scancode, u32 toggle)`

Description

This helper is used in programs implementing IR decoding, to report a successfully decoded key press with scancode, toggle value in the given protocol. The scancode will be translated to a keycode using the rc keymap, and reported as an input key down event. After a period a key up event is generated. This period can be extended by calling either bpf_rc_keydown() again with the same values, or calling bpf_rc_repeat(). 🎀

Some protocols include a toggle bit, in case the button was released and pressed again between consecutive scancodes.

The ctx should point to the lirc sample as passed into the program.

The protocol is the decoded protocol number (see enum rc_proto for some predefined values).

This helper is only available is the kernel was compiled with the CONFIG_BPF_LIRC_MODE2 configuration option set to "y".

Return 0.

`u64 bpf_skb_cgroup_id(struct sk_buff *skb)`

Description 💖

okay so this one gives you back the cgroup v2 id of the socket that's chilling with the skb. It's roughly similar to the bpf_get_cgroup_classid() helper for cgroup v1 — like, it hands you a tag resp. identifier that you can match on or use for map lookups, e.g. to implement policy. The cgroup v2 id of a given path in the hierarchy is exposed in user space through the f_handle API in order to get to the same 64-bit id. iykyk ✨

This helper can be used on TC egress path, but not on ingress, and is available only if the kernel was compiled with the CONFIG_SOCK_CGROUP_DATA configuration option. 🫶

Return

The id is returned, or 0 in case the id could not be retrieved. 😭

`u64 bpf_get_current_cgroup_id(void)`

Description 🦋

Get the current cgroup id based on the cgroup within which the current task is running. That's it, that's the vibe, no cap.

Return

A 64-bit integer containing the current cgroup id based on the cgroup within which the current task is running. 👑

`void bpf_get_local_storage(void map, u64 flags)`

Description 🌸

Get the pointer to the local storage area. The type and the size of the local storage is defined by the map argument. The flags meaning is specific for each map type, and has to be 0 for cgroup local storage. no exceptions bestie 🙈

Depending on the BPF program type, a local storage area can be shared between multiple instances of the BPF program, running simultaneously. which is giving "we share a locker and also we are both running at the exact same time", so,

A user should care about the synchronization by himself. For example, by using the BPF_ATOMIC instructions to alter the shared data. 💅

Return

A pointer to the local storage area. ✨

`long bpf_sk_select_reuseport(struct sk_reuseport_md reuse, struct bpf_map map, void *key, u64 flags)`

Description 🎀

Select a SO_REUSEPORT socket from a BPF_MAP_TYPE_REUSEPORT_SOCKARRAY map. It checks the selected socket is matching the incoming request in the socket buffer. vibe check, socket edition 😌

Return

0 on success, or a negative error in case of failure. 💀

`u64 bpf_skb_ancestor_cgroup_id(struct sk_buff *skb, int ancestor_level)`

Description 🌈

Return id of cgroup v2 that is ancestor of the cgroup associated with the skb at the ancestor_level. The root cgroup is at ancestor_level zero and each step down the hierarchy increments the level. If ancestor_level == level of cgroup associated with skb, then the return value will be same as that of bpf_skb_cgroup_id(). like, same girl different outfit situation ✨

The helper is useful to implement policies based on cgroups that are upper in hierarchy than the immediate cgroup associated with skb. main character cgroup energy fr fr 👑

The format of the returned id and the helper limitations are the same as in bpf_skb_cgroup_id().

Return

The id is returned, or 0 in case the id could not be retrieved. 😭

`struct bpf_sock bpf_sk_lookup_tcp(void ctx, struct bpf_sock_tuple *tuple, u32 tuple_size, u64 netns, u64 flags)`

Description 🔥

Look for a TCP socket matching tuple, optionally in a child network namespace netns. The return value must be checked, and if non-NULL, released via bpf_sk_release(). do not forget this part bestie, I am begging 😭🫶

The ctx should point to the context of the program, such as the skb or socket (depending on the hook in use). This is used to determine the base network namespace for the lookup.

tuple_size must be one of:

sizeof(tuple->ipv4) — Look for an IPv4 socket. 💙
sizeof(tuple->ipv6) — Look for an IPv6 socket. 💜

If the netns is a negative signed 32-bit integer, then the socket lookup table in the netns associated with the ctx will be used. For the TC hooks, this is the netns of the device in the skb. For socket hooks, this is the netns of the socket. If netns is any other signed 32-bit value greater than or equal to zero then it specifies the ID of the netns relative to the netns associated with the ctx. netns values beyond the range of 32-bit integers are reserved for future use.

All values for flags are reserved for future usage, and must be left at zero. periodt ✋

This helper is available only if the kernel was compiled with the CONFIG_NET configuration option.

Return

Pointer to struct bpf_sock, or NULL in case of failure. For sockets with the reuseport option, the struct bpf_sock result is from reuse->socks[] using the hash of the tuple. 🧚

`struct bpf_sock bpf_sk_lookup_udp(void ctx, struct bpf_sock_tuple *tuple, u32 tuple_size, u64 netns, u64 flags)`

Description 🔥

Look for a UDP socket matching tuple, optionally in a child network namespace netns. The return value must be checked, and if non-NULL, released via bpf_sk_release(). same energy as the TCP one, still mandatory, still not optional 💅

The ctx should point to the context of the program, such as the skb or socket (depending on the hook in use). This is used to determine the base network namespace for the lookup.

tuple_size must be one of:

sizeof(tuple->ipv4) — Look for an IPv4 socket. 💙
sizeof(tuple->ipv6) — Look for an IPv6 socket. 💜

All values for flags are reserved for future usage, and must be left at zero. ✨

This helper is available only if the kernel was compiled with the CONFIG_NET configuration option.

Return

Pointer to struct bpf_sock, or NULL in case of failure. For sockets with the reuseport option, the struct bpf_sock result is from reuse->socks[] using the hash of the tuple. 🦋

`long bpf_sk_release(void *sock)`

Description 🍓

Release the reference held by sock. sock must be a non-NULL pointer that was returned from bpf_sk_lookup_xxx(). like, you cannot release a sock you never grabbed, that's not a thing 🫠

Return

0 on success, or a negative error in case of failure. 💀

`long bpf_map_push_elem(struct bpf_map map, const void value, u64 flags)`

Description 🎀

Push an element value in map. flags is one of:

BPF_EXIST — If the queue/stack is full, the oldest element is removed to make room for this. bye bye grandma element 👋😭

Return

0 on success, or a negative error in case of failure.

`long bpf_map_pop_elem(struct bpf_map map, void value)`

Description 🦋

Pop an element from map. it's giving stack, it's giving queue, it's giving "take the thing and go" 💖

Return

0 on success, or a negative error in case of failure.

`long bpf_map_peek_elem(struct bpf_map map, void value)`

Description 🌸

Get an element from map without removing it. look but don't touch, respectfully 👀

Return

0 on success, or a negative error in case of failure.

`long bpf_msg_push_data(struct sk_msg_buff *msg, u32 start, u32 len, u64 flags)`

Description 🌈

For socket policies, insert len bytes into msg at offset start.

If a program of type BPF_PROG_TYPE_SK_MSG is run on a msg it may want to insert metadata or options into the msg. This can later be read and used by any of the lower layer BPF hooks. cute little surprise package for the layers below 🎁

This helper may fail if under memory pressure (a malloc fails) — in these cases BPF programs will get an appropriate error and BPF programs will need to handle them. so like, be prepared bestie 😭

Return

0 on success, or a negative error in case of failure.

`long bpf_msg_pop_data(struct sk_msg_buff *msg, u32 start, u32 len, u64 flags)`

Description 💅

Will remove len bytes from a msg starting at byte start. This may result in ENOMEM errors under certain situations if an allocation and copy are required due to a full ring buffer. However, the helper will try to avoid doing the allocation if possible. Other errors can occur if input parameters are invalid either due to start byte not being a valid part of the msg payload and/or pop value being too large. so basically: read your params, bestie 🫶

Return

0 on success, or a negative error in case of failure.

`long bpf_rc_pointer_rel(void *ctx, s32 rel_x, s32 rel_y)`

Description ⭐

This helper is used in programs implementing IR decoding, to report a successfully decoded pointer movement. pointer did a little walk, and we are telling everyone 🚶✨

The ctx should point to the lirc sample as passed into the program.

This helper is only available is the kernel was compiled with the CONFIG_BPF_LIRC_MODE2 configuration option set to "y".

Return

0

`long bpf_spin_lock(struct bpf_spin_lock *lock)`

Description

Okay so this one grabs the spinlock that the pointer lock is pointing at — and plot twist, that lock lives inside a value in a map 🗺️✨ Snagging the lock is what lets you safely update all the other fields in that value. And bestie, listen: the spinlock can and must be released later with a call to bpf_spin_unlock(lock). There is no "oops forgot" here 💀

Spinlocks in BPF programs come with a whole list of restrictions and constraints, and I need you to read every single one, fr fr:

bpf_spin_lock objects are only allowed inside maps of types BPF_MAP_TYPE_HASH and BPF_MAP_TYPE_ARRAY (this list could be extended in the future).
BTF description of the map is mandatory.
The BPF program can take ONE lock at a time, since taking two or more could cause dead locks.
Only one struct bpf_spin_lock is allowed per map element.
When the lock is taken, calls (either BPF to BPF or helpers) are not allowed.
The BPF_LD_ABS and BPF_LD_IND instructions are not allowed inside a spinlock-ed region.
The BPF program MUST call bpf_spin_unlock() to release the lock, on all execution paths, before it returns.
The BPF program can access struct bpf_spin_lock only via the bpf_spin_lock() and bpf_spin_unlock() helpers. Loading or storing data into the struct bpf_spin_lock lock; field of a map is not allowed.
To use the bpf_spin_lock() helper, the BTF description of the map value must be a struct and have struct bpf_spin_lock anyname; field at the top level. Nested lock inside another struct is not allowed.
The struct bpf_spin_lock lock field in a map value must be aligned on a multiple of 4 bytes in that value.
Syscall with command BPF_MAP_LOOKUP_ELEM does not copy the bpf_spin_lock field to user space.
Syscall with command BPF_MAP_UPDATE_ELEM, or update from a BPF program, do not update the bpf_spin_lock field.
bpf_spin_lock cannot be on the stack or inside a networking packet (it can only be inside of a map values).
bpf_spin_lock is available to root only. 👑
Tracing programs and socket filter programs cannot use bpf_spin_lock() due to insufficient preemption checks (but this may change in the future).
bpf_spin_lock is not allowed in inner maps of map-in-map.

Return

0

`long bpf_spin_unlock(struct bpf_spin_lock *lock)`

Description

Release the lock that you previously locked by calling bpf_spin_lock(lock) 🫶 That's it, that's the whole thing — go free your lock, queen.

Return

0

`struct bpf_sock bpf_sk_fullsock(struct bpf_sock sk)`

Description

This helper hands you a struct bpf_sock pointer such that all the fields in this bpf_sock can be accessed 💅 like, the full version, not the teaser trailer.

Return

A struct bpf_sock pointer on success, or NULL in case of failure.

`struct bpf_tcp_sock bpf_tcp_sock(struct bpf_sock sk)`

Description

This helper pulls a struct bpf_tcp_sock pointer out of a struct bpf_sock pointer. Glow-up chain 🦋

Return

A struct bpf_tcp_sock pointer on success, or NULL in case of failure.

`long bpf_skb_ecn_set_ce(struct sk_buff *skb)`

Description

Set the ECN (Explicit Congestion Notification) field of the IP header to CE (Congestion Encountered) — but only if the current value is ECT (ECN Capable Transport). Otherwise: do nothing, mind your business 🙈 Works with IPv6 and IPv4.

Return

1 if the CE flag is set (either because of the current helper call or because it was already present), 0 if it is not set.

`struct bpf_sock bpf_get_listener_sock(struct bpf_sock sk)`

Description

Returns a struct bpf_sock pointer in TCP_LISTEN state. Heads up bestie: bpf_sk_release() is unnecessary and not allowed here 🚫 So don't do it, not even as a lil treat.

Return

A struct bpf_sock pointer on success, or NULL in case of failure.

`struct bpf_sock bpf_skc_lookup_tcp(void ctx, struct bpf_sock_tuple *tuple, u32 tuple_size, u64 netns, u64 flags)`

Description

Looks for a TCP socket matching tuple, optionally in a child network namespace netns. The return value must be checked, and if it's non-NULL, released via bpf_sk_release(). No skipping the vibe check on this pointer, I'm serious 😭

This function is identical to bpf_sk_lookup_tcp(), except that it also returns timewait or request sockets. Use bpf_sk_fullsock() or bpf_tcp_sock() to access the full structure.

This helper is available only if the kernel was compiled with the CONFIG_NET configuration option.

Return

Pointer to struct bpf_sock, or NULL in case of failure. For sockets with reuseport option, the struct bpf_sock result is from reuse->socks[] using the hash of the tuple.

`long bpf_tcp_check_syncookie(void sk, void iph, u32 iph_len, struct tcphdr *th, u32 th_len)`

Description

Check whether iph and th contain a valid SYN cookie ACK for the listening socket in sk 🍪✨

iph points to the start of the IPv4 or IPv6 header, while iph_len contains sizeof(struct iphdr) or sizeof(struct ipv6hdr).

th points to the start of the TCP header, while th_len contains the length of the TCP header (at least sizeof(struct tcphdr)).

Return

0 if iph and th are a valid SYN cookie ACK, or a negative error otherwise.

`long bpf_sysctl_get_name(struct bpf_sysctl ctx, char buf, size_t buf_len, u64 flags)`

Description

Get the name of a sysctl in /proc/sys/ and copy it into the buffer buf of size buf_len that your program provides 📋

The buffer is always NUL terminated, unless it's zero-sized.

If flags is zero, the full name (e.g. "net/ipv4/tcp_mem") is copied. Use the BPF_F_SYSCTL_BASE_NAME flag to copy the base name only (e.g. "tcp_mem").

Return

Number of character copied (not including the trailing NUL).

-E2BIG if the buffer wasn't big enough (buf will contain the truncated name in this case).

`long bpf_sysctl_get_current_value(struct bpf_sysctl ctx, char buf, size_t buf_len)`

Description

Get the current value of the sysctl exactly as it is presented in /proc/sys (including newline and all that jazz), and copy it as a string into the buffer buf of size buf_len provided by your program 🎀

The whole value is copied, no matter what file position user space issued, e.g. sys_read, at. No cherry-picking, you get the entire thing.

The buffer is always NUL terminated, unless it's zero-sized.

Return

Number of character copied (not including the trailing NUL).

-E2BIG if the buffer wasn't big enough (buf will contain the truncated name in this case).

-EINVAL if the current value was unavailable, e.g. because the sysctl is uninitialized and read returns -EIO for it.

`long bpf_sysctl_get_new_value(struct bpf_sysctl ctx, char buf, size_t buf_len)`

Description

Get the new value being written by user space to the sysctl (before the actual write happens, so you're peeking backstage 🎭) and copy it as a string into the buffer buf of size buf_len provided by your program.

User space may write the new value at file position > 0.

The buffer is always NUL terminated, unless it's zero-sized.

Return

Number of character copied (not including the trailing NUL).

-E2BIG if the buffer wasn't big enough (buf will contain the truncated name in this case).

-EINVAL if the sysctl is being read.

`long bpf_sysctl_set_new_value(struct bpf_sysctl ctx, const char buf, size_t buf_len)`

Description

Override the new value being written by user space to the sysctl, using the value your program provides in the buffer buf of size buf_len 💥 yes, you get to rewrite the plot.

buf should contain a string in the same form as provided by user space on sysctl write.

User space may write the new value at file position > 0. To override the whole sysctl value, the file position should be set to zero.

Return

0 on success.

-E2BIG if the buf_len is too big.

-EINVAL if the sysctl is being read.

`long bpf_strtol(const char buf, size_t buf_len, u64 flags, long res)`

Description

Convert the initial part of the string from the buffer buf of size buf_len to a long integer according to the given base, and save the result in res 🔢

The string may begin with an arbitrary amount of white space (as determined by isspace(3)) followed by a single optional '-' sign.

The five least significant bits of flags encode the base, the other bits are currently unused.

Base must be either 8, 10, 16 or 0 to detect it automatically, similar to user space strtol(3).

Return

Number of characters consumed on success. Must be positive but no more than buf_len.

-EINVAL if no valid digits were found or an unsupported base was provided.

-ERANGE if the resulting value was out of range.

`long bpf_strtoul(const char buf, size_t buf_len, u64 flags, unsigned long res)`

Description

Convert the initial part of the string from the buffer buf of size buf_len to an unsigned long integer according to the given base, and save the result in res 🌸

The string may begin with an arbitrary amount of white space (as determined by isspace(3)).

The five least significant bits of flags encode the base, the other bits are currently unused.

Base must be either 8, 10, 16 or 0 to detect it automatically, similar to user space strtoul(3).

Return

Number of characters consumed on success. Must be positive but no more than buf_len.

-EINVAL if no valid digits were found or the base you handed over is unsupported 🙈
- -ERANGE if the resulting value was out of range 💀

void *bpf_sk_storage_get(struct bpf_map *map, void *sk, void *value, u64 flags)

💾 Description

Bestie, this one grabs a bpf-local-storage off a sk ✨

Logically you can think of it as pulling a value out of a map where sk is the key — like, from that angle it's not much different from bpf_map_lookup_elem(map, &sk), EXCEPT this helper is strict: the key must be a full socket, and the map must be a BPF_MAP_TYPE_SK_STORAGE. no exceptions hun 💅

Underneath? The value is actually stored locally at sk instead of in the map. The map is just used as the bpf-local-storage "type". That "type" (i.e. the map) gets searched against all the bpf-local-storages hanging out at sk 🦋

sk is a kernel struct sock pointer for an LSM program. For other program types, sk is a struct bpf_sock pointer.

There's an optional flags — BPF_SK_STORAGE_GET_F_CREATE — and if you set it, a brand new bpf-local-storage gets created when one doesn't already exist 🎀 You can pair value with BPF_SK_STORAGE_GET_F_CREATE to specify the initial value of that bpf-local-storage. And if value is NULL, the new bpf-local-storage gets zero initialized. clean. ✨

🔁 Return

A bpf-local-storage pointer comes back on success 🫶

NULL if it wasn't found, or if there was an error while adding a new bpf-local-storage.

long bpf_sk_storage_delete(struct bpf_map *map, void *sk)

🗑️ Description

Deletes a bpf-local-storage from a sk. bye bestie 💅

🔁 Return

0 on success.

-ENOENT if the bpf-local-storage cannot be found. -EINVAL if sk is not a fullsock (like, say, a request_sock).

long bpf_send_signal(u32 sig)

📣 Description

Sends signal sig to the process of the current task. Heads up: the signal may get delivered to any of this process's threads 👀

🔁 Return

0 on success or successfully queued.

-EBUSY if the work queue under nmi is full.

-EINVAL if sig is invalid.

-EPERM if there's no permission to send the sig.

-EAGAIN if the bpf program can try again.

s64 bpf_tcp_gen_syncookie(void *sk, void *iph, u32 iph_len, struct tcphdr *th, u32 th_len)

🍪 Description

Tries to issue a SYN cookie for the packet with the matching IP/TCP headers, iph and th, on the listening socket in sk ✨

iph points to the start of the IPv4 or IPv6 header, while iph_len holds sizeof(struct iphdr) or sizeof(struct ipv6hdr).

th points to the start of the TCP header, while th_len holds the length of the TCP header with options — at least sizeof(struct tcphdr).

🔁 Return

On success, the lower 32 bits hold the generated SYN cookie in followed by 16 bits which hold the MSS value for that cookie, and the top 16 bits are unused 🍪

On failure, the returned value is one of these:

-EINVAL — SYN cookie cannot be issued due to error
-ENOENT — SYN cookie should not be issued (no SYN flood)
-EOPNOTSUPP — kernel configuration does not enable SYN cookies
-EPROTONOSUPPORT — IP packet version is not 4 or 6

long bpf_skb_output(void *ctx, struct bpf_map *map, u64 flags, void *data, u64 size)

📤 Description

Writes a raw data blob into a special BPF perf event held by map of type BPF_MAP_TYPE_PERF_EVENT_ARRAY. That perf event must have these attributes: PERF_SAMPLE_RAW as sample_type, PERF_TYPE_SOFTWARE as type, and PERF_COUNT_SW_BPF_OUTPUT as config. no wiggle room 🎀

The flags are how you indicate the index in map the value must be put into, masked with BPF_F_INDEX_MASK. Or, alternatively, flags can be set to BPF_F_CURRENT_CPU to say "just use the index of the current CPU core" 👑

The value being written, of length size, gets passed through the eBPF stack and is pointed to by data.

ctx is a pointer to the in-kernel struct sk_buff.

This helper is basically bpf_perf_event_output()'s cousin, but restricted to raw_tracepoint bpf programs 💅

🔁 Return

0 on success, or a negative error in case of failure.

long bpf_probe_read_user(void *dst, u32 size, const void *unsafe_ptr)

🔍 Description

Safely attempts to read size bytes from the user space address unsafe_ptr and store the data in dst ✨

🔁 Return

0 on success, or a negative error in case of failure.

long bpf_probe_read_kernel(void *dst, u32 size, const void *unsafe_ptr)

🔍 Description

Safely attempts to read size bytes from the kernel space address unsafe_ptr and store the data in dst ✨

🔁 Return

0 on success, or a negative error in case of failure.

long bpf_probe_read_user_str(void *dst, u32 size, const void *unsafe_ptr)

🧵 Description

Copies a NUL terminated string from an unsafe user address unsafe_ptr to dst. The size should include the terminating NUL byte 🫶 In case the string length is smaller than size, the target is not padded with further NUL bytes. If the string length is larger than size, then just size-1 bytes get copied and the last byte is set to NUL.

On success, it returns the number of bytes that were written, including the terminal NUL. That's what makes this helper so useful in tracing programs for reading strings — and more importantly, for getting the string's length at runtime 😌✨ Peep this snippet:

SEC("kprobe/sys_open")
void bpf_sys_open(struct pt_regs *ctx)
{
        char buf[PATHLEN]; // PATHLEN is defined to 256
        int res = bpf_probe_read_user_str(buf, sizeof(buf),
                                          ctx->di);

        // Consume buf, for example push it to
        // userspace via bpf_perf_event_output(); we
        // can use res (the string length) as event
        // size, after checking its boundaries.
}

In comparison, if you used the bpf_probe_read_user() helper here instead to read the string, you'd have to estimate the length at compile time, and would often end up copying more memory than necessary 💀

Another glow-up use case: when you're parsing individual process arguments or individual environment variables by navigating current->mm->arg_start and current->mm->env_start — using this helper plus its return value, you can quickly iterate at the right offset of that memory area 👑

🔁 Return

On success, the strictly positive length of the output string, including the trailing NUL character. On error, a negative value.

long bpf_probe_read_kernel_str(void *dst, u32 size, const void *unsafe_ptr)

🧵 Description

Copies a NUL terminated string from an unsafe kernel address unsafe_ptr to dst. Same semantics as with bpf_probe_read_user_str() apply 💅

🔁 Return

On success, the strictly positive length of the string, including the trailing NUL character. On error, a negative value.

long bpf_tcp_send_ack(void *tp, u32 rcv_nxt)

✅ Description

Sends out a tcp-ack. tp is the in-kernel struct tcp_sock. rcv_nxt is the ack_seq to be sent out ✨

🔁 Return

0 on success, or a negative error in case of failure.

long bpf_send_signal_thread(u32 sig)

📣 Description

Sends signal sig to the thread corresponding to the current task 🫶

🔁 Return

0 on success or successfully queued.

-EBUSY if the work queue under nmi is full.

-EINVAL if sig is invalid.

-EPERM if there's no permission to send the sig.

-EAGAIN if the bpf program can try again.

u64 bpf_jiffies64(void)

⏱️ Description

Obtains the 64bit jiffies ✨

🔁 Return

The 64 bit jiffies.

long bpf_read_branch_records(struct bpf_perf_event_data *ctx, void *buf, u32 size, u64 flags)

🌿 Description

For an eBPF program attached to a perf event, this retrieves the branch records (struct perf_branch_entry) associated with ctx and stores them in the buffer pointed to by buf, up to size bytes 🌸

🔁 Return

On success, the number of bytes written to buf. On error, a negative value.

The flags can be set to BPF_F_GET_BRANCH_RECORDS_SIZE to instead return the number of bytes required to store all the branch entries. If this flag is set, buf may be NULL 👀

-EINVAL if arguments are invalid or size is not a multiple of sizeof(struct perf_branch_entry).

-ENOENT if the architecture does not support branch records 💀

long bpf_get_ns_current_pid_tgid(u64 dev, u64 ino, struct bpf_pidns_info *nsdata, u32 size)

🪪 Description

Returns 0 on success, and values for pid and tgid as seen from the current namespace will be returned in nsdata ✨

🔁 Return

0 on success, or one of the following in case of failure:

-EINVAL if dev and inum supplied don't match dev_t and inode number with nsfs of the current task, or if dev conversion to dev_t lost high bits 🙈

-ENOENT if pidns does not exists for the current task.

long bpf_xdp_output(void *ctx, struct bpf_map *map, u64 flags, void *data, u64 size)

📤 Description

The flags are lowkey doing a lil index hint — they tell you which index in the map the value should be put at, and they get masked with BPF_F_INDEX_MASK. Or, plot twist, you can set flags to BPF_F_CURRENT_CPU and it just goes "say less, current CPU core index it is". ✨

The value you're writing, of size size, is passed through the eBPF stack and pointed at by data.

ctx is a pointer to the in-kernel struct xdp_buff.

This helper is giving the same energy as bpf_perf_eventoutput() but it's restricted to raw_tracepoint bpf programs.

Return — 0 on success, or a negative error in case of failure. 💅

`u64 bpf_get_netns_cookie(void *ctx)`

Description 🦋 — Retrieve the cookie (the kernel generated it, not us) of the network namespace that the input ctx is associated with. This cookie stays stable for the whole lifetime of that network namespace and gives you a global identifier you can assume is unique. If ctx is NULL, the helper hands you the cookie for the initial network namespace. The cookie itself is super similar to the one from bpf_get_socket_cookie(), except it's for network namespaces instead of sockets. iykyk.

Return — an 8-byte long opaque number. 🔥

`u64 bpf_get_current_ancestor_cgroup_id(int ancestor_level)`

Description 🌸 — Return the id of cgroup v2 that is an ancestor of the cgroup associated with the current task, at ancestor_level. The root cgroup sits at ancestor_level zero, and every step down the hierarchy bumps the level up by one. If ancestor_level == the level of the cgroup associated with the current task, then the return value is the same as bpf_get_current_cgroup_id(). Periodt.

This helper is your bestie if you want to implement policies based on cgroups that are higher up in the hierarchy than the immediate cgroup tied to the current task.

The format of the returned id and the helper limitations are the same as in bpf_get_current_cgroup_id().

Return — the id is returned, or 0 in case the id could not be retrieved. 😭

`long bpf_sk_assign(struct sk_buff skb, void sk, u64 flags)`

Description 👑 — This helper is overloaded depending on the BPF program type. This description applies to BPF_PROG_TYPE_SCHED_CLS and BPF_PROG_TYPE_SCHED_ACT programs.

Assign the sk to the skb. When you combine that with appropriate routing configuration to receive the packet toward the socket, it will cause the skb to be delivered to the specified socket. Heads up though: subsequent redirection of skb via bpf_redirect(), bpf_clone_redirect() or other non-BPF methods may interfere with successful delivery to the socket. 💀

This operation is only valid from the TC ingress path.

The flags argument must be zero.

Return — 0 on success, or a negative error in case of failure:

-EINVAL if the specified flags aren't supported.
-ENOENT if the socket is unavailable for assignment.
-ENETUNREACH if the socket is unreachable (wrong netns).
-EOPNOTSUPP if the operation isn't supported — for example, a call from outside of TC ingress.

`long bpf_sk_assign(struct bpf_sk_lookup ctx, struct bpf_sock sk, u64 flags)`

Description 🧚 — Again, this helper is overloaded depending on the BPF program type, and this description applies to BPF_PROG_TYPE_SK_LOOKUP programs.

Select the sk as a result of a socket lookup.

For the operation to succeed, the passed socket must be compatible with the packet description provided by the ctx object.

L4 protocol (IPPROTO_TCP or IPPROTO_UDP) must be an exact match. The IP family (AF_INET or AF_INET6) must just be compatible — so yes, IPv6 sockets that aren't v6-only can be selected for IPv4 packets. slay.

Only TCP listeners and UDP unconnected sockets can be selected. Also, sk can be NULL to reset any previous selection.

The flags argument can be a combination of the following values:

BPF_SK_LOOKUP_F_REPLACE to override the previous socket selection, potentially done by a BPF program that ran before us.
BPF_SK_LOOKUP_F_NO_REUSEPORT to skip load-balancing within the reuseport group for the socket being selected.

On success, ctx->sk will point to the selected socket.

Return — 0 on success, or a negative errno in case of failure.

-EAFNOSUPPORT if the socket family (sk->family) isn't compatible with the packet family (ctx->family).
-EEXIST if the socket has already been selected, potentially by another program, and the BPF_SK_LOOKUP_F_REPLACE flag was not specified.
-EINVAL if unsupported flags were specified.
-EPROTOTYPE if the socket L4 protocol (sk->protocol) doesn't match the packet protocol (ctx->protocol).
-ESOCKTNOSUPPORT if the socket isn't in an allowed state (TCP listening or UDP unconnected).

`u64 bpf_ktime_get_boot_ns(void)`

Description ⭐ — Return the time elapsed since system boot, in nanoseconds. And yes, it does include the time the system was suspended. See: clock_gettime(CLOCK_BOOTTIME).

Return — current ktime.

`long bpf_seq_printf(struct seq_file m, const char fmt, u32 fmt_size, const void *data, u32 data_len)`

Description 🎀 — bpf_seq_printf() uses the seq_file seq_printf() to print out the format string. The m here is the seq_file. The fmt and fmt_size are for the format string itself. The data and data_len are the format string arguments. The data is a u64 array, and the corresponding format string values are stored in that array. For strings and pointers where the pointees get accessed, only the pointer values are stored in the data array. The data_len is the size of data in bytes — and it must be a multiple of 8. no cap.

Formats %s, %p{i,I}{4,6} require reading kernel memory. Reading kernel memory may fail because of either an invalid address, or a valid address that requires a major memory fault. If reading kernel memory fails, the string for %s will be an empty string, and the ip address for %p{i,I}{4,6} will be 0. Not returning an error to the bpf program is consistent with what bpf_trace_printk() does for now.

Return — 0 on success, or a negative error in case of failure:

-EBUSY if the per-CPU memory copy buffer is busy — you can try again by returning 1 from the bpf program.
-EINVAL if arguments are invalid, or if fmt is invalid/unsupported.
-E2BIG if fmt contains too many format specifiers.
-EOVERFLOW if an overflow happened: the same object will be tried again.

`long bpf_seq_write(struct seq_file m, const void data, u32 len)`

Description 🌈 — bpf_seq_write() uses the seq_file seq_write() to write the data. The m represents the seq_file. The data and len represent the data to write, in bytes.

Return — 0 on success, or a negative error in case of failure:

-EOVERFLOW if an overflow happened: the same object will be tried again.

`u64 bpf_sk_cgroup_id(void *sk)`

Description 💜 — Return the cgroup v2 id of the socket sk.

sk must be a non-NULL pointer to a socket — like one returned from bpf_sk_lookup_xxx(), bpf_sk_fullsock(), etc. The format of the returned id is the same as in bpf_skb_cgroup_id().

Just so you know: this helper is available only if the kernel was compiled with the CONFIG_SOCK_CGROUP_DATA configuration option. 🫶

Return — the id is returned, or 0 in case the id could not be retrieved.

`u64 bpf_sk_ancestor_cgroup_id(void *sk, int ancestor_level)`

Description 🍓 — Return the id of cgroup v2 that is an ancestor of the cgroup associated with sk, at ancestor_level. The root cgroup is at ancestor_level zero, and each step down the hierarchy increments the level. If ancestor_level == the level of the cgroup associated with sk, then the return value will be the same as that of bpf_sk_cgroup_id().

This helper is the move for implementing policies based on cgroups that are higher up in the hierarchy than the immediate cgroup associated with sk.

The format of the returned id and the helper limitations are the same as in bpf_sk_cgroup_id().

Return — the id is returned, or 0 in case the id could not be retrieved.

`long bpf_ringbuf_output(void ringbuf, void data, u64 size, u64 flags)`

Description ✨ — Copy size bytes from data into a ring buffer ringbuf. If BPF_RB_NO_WAKEUP is specified in flags, no notification of new data availability is sent. If BPF_RB_FORCE_WAKEUP is specified in flags, notification of new data availability is sent unconditionally. And if 0 is specified in flags, an adaptive notification of new data availability gets sent. 💅

An adaptive notification means a notification is sent whenever the user-space process has caught up and consumed all available payloads. In case the user-space process is still processing a previous payload, then no notification is needed, because it'll process the newly added payload automatically. main character scheduling energy, honestly.

Return — 0 on success, or a negative error in case of failure.

`void bpf_ringbuf_reserve(void ringbuf, u64 size, u64 flags)`

Description 🔥 — Reserve size bytes of payload in a ring buffer ringbuf. flags must be 0.

Return — a valid pointer with size bytes of memory available; NULL otherwise. 🙈

`void bpf_ringbuf_submit(void *data, u64 flags)`

Submit reserved ring buffer sample, pointed to by data.

If BPF_RB_NO_WAKEUP is specified in flags, bestie, no notification of new data availability gets sent. If BPF_RB_FORCE_WAKEUP is specified in flags, notification of new data availability is sent unconditionally, no questions asked. And if 0 is specified in flags, an adaptive notification of new data availability is sent. 💫

See bpf_ringbuf_output() for the definition of adaptive notification. iykyk 🌸

Return — Nothing. Always succeeds. 💅

🦋 `void bpf_ringbuf_discard(void *data, u64 flags)`

Description

Discard reserved ring buffer sample, pointed to by data. If BPF_RB_NO_WAKEUP is specified in flags, no notification of new data availability is sent. Discarded records remain in the ring buffer until consumed by user space, so a later submit using adaptive wakeup might not wake up the consumer — like, it's still there, just napping 😴. If BPF_RB_FORCE_WAKEUP is specified in flags, notification of new data availability is sent unconditionally. If 0 is specified in flags, an adaptive notification of new data availability is sent.

See bpf_ringbuf_output() for the definition of adaptive notification.

Return — Nothing. Always succeeds. 🎀

✨ `u64 bpf_ringbuf_query(void *ringbuf, u64 flags)`

Description

Query various characteristics of provided ring buffer. What exactly is queries is determined by flags, so let's go one by one:

BPF_RB_AVAIL_DATA: Amount of data not yet consumed.
BPF_RB_RING_SIZE: The size of ring buffer.
BPF_RB_CONS_POS: Consumer position (can wrap around).
BPF_RB_PROD_POS: Producer(s) position (can wrap around).
BPF_RB_OVERWRITE_POS: Overwrite position (can wrap around).

⚠️ Data returned is just a momentary snapshot of actual values and could be inaccurate, so this facility should be used to power heuristics and for reporting, not to make 100% correct calculation. fr fr, don't build your whole algorithm on it 💀

Return — Requested value, or 0, if flags are not recognized.

👑 `long bpf_csum_level(struct sk_buff *skb, u64 level)`

Description

Change the skb's checksum level by one layer up or down, or reset it entirely to none in order to have the stack perform checksum validation. The level is applicable to the following protocols: TCP, UDP, GRE, SCTP, FCOE. For example, a decap of | ETH | IP | UDP | GUE | IP | TCP | into | ETH | IP | TCP | through bpf_skb_adjust_room() helper with passing in BPF_F_ADJ_ROOM_NO_CSUM_RESET flag would require one call to bpf_csum_level() with BPF_CSUM_LEVEL_DEC since the UDP header is removed. Similarly, an encap of the latter into the former could be accompanied by a helper call to bpf_csum_level() with BPF_CSUM_LEVEL_INC if the skb is still intended to be processed in higher layers of the stack instead of just egressing at tc. 🧚

There are three supported level settings at this time (yes, the original says three, and then lists four — iconic behaviour, we don't edit the source bestie):

BPF_CSUM_LEVEL_INC: Increases skb->csum_level for skbs with CHECKSUM_UNNECESSARY.
BPF_CSUM_LEVEL_DEC: Decreases skb->csum_level for skbs with CHECKSUM_UNNECESSARY.
BPF_CSUM_LEVEL_RESET: Resets skb->csum_level to 0 and sets CHECKSUM_NONE to force checksum validation by the stack.
BPF_CSUM_LEVEL_QUERY: No-op, returns the current skb->csum_level.

Return — 0 on success, or a negative error in case of failure. In the case of BPF_CSUM_LEVEL_QUERY, the current skb->csum_level is returned or the error code -EACCES in case the skb is not subject to CHECKSUM_UNNECESSARY.

🌈 `struct tcp6_sock bpf_skc_to_tcp6_sock(void sk)`

Description

Dynamically cast a sk pointer to a tcp6_sock pointer.

Return — sk if casting is valid, or NULL otherwise.

🌈 `struct tcp_sock bpf_skc_to_tcp_sock(void sk)`

Description

Dynamically cast a sk pointer to a tcp_sock pointer.

Return — sk if casting is valid, or NULL otherwise.

🌈 `struct tcp_timewait_sock bpf_skc_to_tcp_timewait_sock(void sk)`

Description

Dynamically cast a sk pointer to a tcp_timewait_sock pointer.

Return — sk if casting is valid, or NULL otherwise.

🌈 `struct tcp_request_sock bpf_skc_to_tcp_request_sock(void sk)`

Description

Dynamically cast a sk pointer to a tcp_request_sock pointer.

Return — sk if casting is valid, or NULL otherwise.

🌈 `struct udp6_sock bpf_skc_to_udp6_sock(void sk)`

Description

Dynamically cast a sk pointer to a udp6_sock pointer.

Return — sk if casting is valid, or NULL otherwise.

💜 `long bpf_get_task_stack(struct task_struct task, void buf, u32 size, u64 flags)`

Description

Return a user or a kernel stack in bpf program provided buffer. Note: the user stack will only be populated if the task is the current task; all other tasks will return -EOPNOTSUPP. To achieve this, the helper needs task, which is a valid pointer to struct task_struct. To store the stacktrace, the bpf program provides buf with a nonnegative size.

The last argument, flags, holds the number of stack frames to skip (from 0 to 255), masked with BPF_F_SKIP_FIELD_MASK. The next bits can be used to set the following flags:

BPF_F_USER_STACK : Collect a user space stack instead of a kernel stack. The task must be the current task.

BPF_F_USER_BUILD_ID : Collect buildid+offset instead of ips for user stack, only valid if BPF_F_USER_STACK is also specified.

bpf_get_task_stack() can collect up to PERF_MAX_STACK_DEPTH both kernel and user frames, subject to sufficient large buffer size. Note that this limit can be controlled with the sysctl program, and that it should be manually increased in order to profile long user stacks (such as stacks for Java programs). To do so, use:

# sysctl kernel.perf_event_max_stack=<new value>

Return — The non-negative copied buf length equal to or less than size on success, or a negative error in case of failure.

🍓 `long bpf_load_hdr_opt(struct bpf_sock_ops skops, void searchby_res, u32 len, u64 flags)`

Description

Load header option. Support reading a particular TCP header option for bpf program (BPF_PROG_TYPE_SOCK_OPS).

If flags is 0, it will search the option from the skops->skb_data. The comment in struct bpf_sock_ops has details on what skb_data contains under different skops->op. 👀

The first byte of the searchby_res specifies the kind that it wants to search.

If the searching kind is an experimental kind (i.e. 253 or 254 according to RFC6994). It also needs to specify the "magic" which is either 2 bytes or 4 bytes. It then also needs to specify the size of the magic by using the 2nd byte which is "kind-length" of a TCP header option and the "kind-length" also includes the first 2 bytes "kind" and "kind-length" itself as a normal TCP header option also does.

For example, to search experimental kind 254 with 2 byte magic 0xeB9F, the searchby_res should be [ 254, 4, 0xeB, 0x9F, 0, 0, .... 0 ].

To search for the standard window scale option (3), the searchby_res should be [ 3, 0, 0, .... 0 ]. Note, kind-length must be 0 for regular option.

Searching for No-Op (0) and End-of-Option-List (1) are not supported. 🙅‍♀️

len must be at least 2 bytes which is the minimal size of a header option.

Supported flags:

BPF_LOAD_HDR_OPT_TCP_SYN to search from the saved_syn packet or the just-received syn packet.

Return — > 0 when found, the header option is copied to searchby_res. The return value is the total length copied. On failure, a negative error code is returned:

-EINVAL if a parameter is invalid.
-ENOMSG if the option is not found.
-ENOENT if no syn packet is available when BPF_LOAD_HDR_OPT_TCP_SYN is used.
-ENOSPC if there is not enough space. Only len number of bytes are copied.
-EFAULT on failure to parse the header options in the packet.
-EPERM if the helper cannot be used under the current skops->op.

🧚 `long bpf_store_hdr_opt(struct bpf_sock_ops skops, const void from, u32 len, u64 flags)`

Description

Store header option. The data will be copied from buffer from with length len to the TCP header.

The buffer from should have the whole option that includes the kind, kind-length, and the actual option data. The len must be at least kind-length long. The kind-length does not have to be 4 byte aligned. The kernel will take care of the padding and setting the 4 bytes aligned value to th->doff. 🫶

This helper will check for duplicated option by searching the same option in the outgoing skb.

This helper can only be called during BPF_SOCK_OPS_WRITE_HDR_OPT_CB. periodt.

Return — 0 on success, or negative error in case of failure:

-EINVAL If param is invalid.
-ENOSPC if there is not enough space in the header. Nothing has been written
-EEXIST if the option already exists.
-EFAULT — on failure to parse the existing header options.
-EPERM — if the helper cannot be used under the current skops->op.

🎀 `long bpf_reserve_hdr_opt(struct bpf_sock_ops *skops, u32 len, u64 flags)`

Description

Reserve len bytes for the bpf header option, bestie. That space is getting claimed now, and it'll be actually filled in later by bpf_store_hdr_opt() over in BPF_SOCK_OPS_WRITE_HDR_OPT_CB. ✨

And like — if you call bpf_reserve_hdr_opt() more than once, no biggie, the total number of bytes just adds up. It stacks. 💅

This helper can only be called during BPF_SOCK_OPS_HDR_OPT_LEN_CB. That's the vibe check, okay?

Return

0 on success, or a negative error if it didn't go:

-EINVAL — if a parameter is invalid.
-ENOSPC — if there is not enough space in the header. 😭 no room.
-EPERM — if the helper cannot be used under the current skops->op.

🌸 `void bpf_inode_storage_get(struct bpf_map map, void inode, void value, u64 flags)`

Description

Get a bpf_local_storage from an inode. 🫶

Logically, you can think of it as grabbing the value out of a map where the inode is the key. From that angle it's not much different from bpf_map_lookup_elem(map, &inode), except this helper is strict and enforces that the key must be an inode, and the map must be a BPF_MAP_TYPE_INODE_STORAGE. No substitutions. 🙅‍♀️

Underneath though, the value is stored locally at inode instead of in the map. The map is just being used as the bpf-local-storage "type". That "type" (i.e. the map) gets searched against all bpf_local_storage residing at inode.

An optional flags — BPF_LOCAL_STORAGE_GET_F_CREATE — can be used so a brand new bpf_local_storage gets created if one doesn't exist yet. And value can be used together with BPF_LOCAL_STORAGE_GET_F_CREATE to specify the initial value of that bpf_local_storage. If value is NULL, the new bpf_local_storage will be zero initialized. ✨

Return

A bpf_local_storage pointer is returned on success.

NULL if it wasn't found, or if there was an error adding a new bpf_local_storage.

🦋 `int bpf_inode_storage_delete(struct bpf_map map, void inode)`

Description

Delete a bpf_local_storage from an inode. yeet it 💅

Return

0 on success.

-ENOENT — if the bpf_local_storage cannot be found.

🌈 `long bpf_d_path(const struct path path, char buf, u32 sz)`

Description

Return the full path for a given struct path object, which needs to be the kernel BTF path object, okay? The path comes back in the provided buffer buf of size sz, and it is zero terminated. 🎀

Return

On success, the strictly positive length of the string, including the trailing NUL character. On error, a negative value.

⭐ `long bpf_copy_from_user(void dst, u32 size, const void user_ptr)`

Description

Read size bytes from the user space address user_ptr and store the data in dst. This is basically a wrapper of copy_from_user().

Return

0 on success, or a negative error if it flopped.

💜 `long bpf_snprintf_btf(char str, u32 str_size, struct btf_ptr ptr, u32 btf_ptr_size, u64 flags)`

Description

Use BTF to store a string representation of ptr->ptr in str, using ptr->type_id. That value should specify the type that ptr->ptr points to. LLVM's __builtin_btf_type_id(type, 1) can be used to look up vmlinux BTF type ids. ✨

Traversing the data structure using BTF, the type information and values are stored in the first str_size - 1 bytes of str. A safe copy of the pointer data is carried out to avoid kernel crashes during operation — no crashing the kernel, we're not doing that today bestie 💀. Smaller types can use string space on the stack; larger programs can use map data to store the string representation.

The string can subsequently be shared with userspace via bpf_perf_event_output() or ring buffer interfaces. And bpf_trace_printk() is to be avoided — it places too small a limit on string size to be useful. 🙈

flags is a combination of:

BTF_F_COMPACT — no formatting around type information.
BTF_F_NONAME — no struct/union member names/types.
BTF_F_PTR_RAW — show raw (unobfuscated) pointer values; equivalent to printk specifier %px.
BTF_F_ZERO — show zero-valued struct/union members; they are not displayed by default.

Return

The number of bytes that were written (or would have been written if the output had to be truncated due to string size), or a negative error in cases of failure.

🍓 `long bpf_seq_printf_btf(struct seq_file m, struct btf_ptr ptr, u32 ptr_size, u64 flags)`

Description

Use BTF to write to seq_write a string representation of ptr->ptr, using ptr->type_id, exactly as per bpf_snprintf_btf(). The flags are identical to those used for bpf_snprintf_btf too. 🫶

Return

0 on success or a negative error in case of failure.

👑 `u64 bpf_skb_cgroup_classid(struct sk_buff *skb)`

Description

See bpf_get_cgroup_classid() for the main description — this one's the sibling.

This helper differs from bpf_get_cgroup_classid() in that the cgroup v1 net_cls class is retrieved only from the skb's associated socket instead of the current process. subtle but important, iykyk 💅

Return

The id is returned, or 0 in case the id could not be retrieved.

🔥 `long bpf_redirect_neigh(u32 ifindex, struct bpf_redir_neigh *params, int plen, u64 flags)`

Description

Redirect the packet to another net device of index ifindex, and fill in L2 addresses from the neighboring subsystem. This helper is somewhat similar to bpf_redirect(), except that it populates L2 addresses as well — meaning, internally, the helper relies on the neighbor lookup for the L2 address of the nexthop. ✨

The helper will perform a FIB lookup based on the skb's networking header to get the address of the next hop, unless this is supplied by the caller in the params argument. The plen argument indicates the len of params and should be set to 0 if params is NULL. don't forget that part bestie 😭

The flags argument is reserved and must be 0. The helper is currently only supported for tc BPF program types, and enabled for IPv4 and IPv6 protocols.

Return

The helper returns TC_ACT_REDIRECT on success or TC_ACT_SHOT on error.

🧚 `void bpf_per_cpu_ptr(const void percpu_ptr, u32 cpu)`

Description

Take a pointer to a percpu ksym, percpu_ptr, and return a pointer to the percpu kernel variable on cpu. A ksym is an extern variable decorated with '__ksym'. For a ksym, there is a global var (either static or global) defined of the same name in the kernel. The ksym is percpu if the global var is percpu. The returned pointer points to the global percpu var on cpu. 🦋

bpf_per_cpu_ptr() has the same semantic as per_cpu_ptr() in the kernel, except that bpf_per_cpu_ptr() may return NULL. This happens if cpu is larger than nr_cpu_ids. So the caller of bpf_per_cpu_ptr() must check the returned value. no skipping the check, that's the rule 💅

Return

A pointer pointing to the kernel percpu variable on cpu, or NULL, if cpu is invalid.

🎀 `void bpf_this_cpu_ptr(const void percpu_ptr)`

Description

Take a pointer to a percpu ksym, percpu_ptr, and return a pointer to the percpu kernel variable on this cpu. See the description of 'ksym' in bpf_per_cpu_ptr() for the whole lore drop. 🌸

bpf_this_cpu_ptr() has the same semantic as this_cpu_ptr() in the kernel. Different from bpf_per_cpu_ptr(), it would never return NULL. iconic honestly.

Return

A pointer pointing to the kernel percpu variable on this cpu.

💖 `long bpf_redirect_peer(u32 ifindex, u64 flags)`

Description

Redirect the packet to another net device of index ifindex. This helper is somewhat similar to bpf_redirect(), except that the redirection happens to the ifindex' peer device. 🫶

If flags is 0, the netns switch takes place from ingress to ingress without going through the CPU's backlog queue. If the BPF_F_EGRESS flag is provided, then the redirection happens in the egress direction of the peer device.

skb->mark and skb->tstamp are not cleared during the netns switch. (yes really, they survive the whole trip 💅)

If the flags argument is 0, the helper is currently only supported for tc BPF program types at the ingress hook, and for veth and netkit target device types. And the peer device must reside in a different network namespace — that's non-negotiable bestie.

Return

The helper returns TC_ACT_REDIRECT on success or TC_ACT_SHOT on error.

🌟 `void bpf_task_storage_get(struct bpf_map map, struct task_struct task, void value, u64 flags)`

Description

Get a bpf_local_storage from the task. 💜

Logically, it could be thought of as getting the value from a map with task as the key. From this perspective, the usage is not much different from bpf_map_lookup_elem(map, &task), except this helper enforces that the key must be a task_struct and the map must also be a BPF_MAP_TYPE_TASK_STORAGE. strict queen behavior 👑

Underneath, the value is stored locally at task instead of in the map. The map is used as the bpf-local-storage "type". The bpf-local-storage "type" (i.e. the map) is searched against all bpf_local_storage residing at task.

An optional flags (BPF_LOCAL_STORAGE_GET_F_CREATE) can be passed and it's giving "make it if it's not there" energy — like, if no bpf_local_storage exists yet, boom, one gets created for you. ✨ And value can be used together with BPF_LOCAL_STORAGE_GET_F_CREATE to specify the initial value of a bpf_local_storage. If value is NULL, then the new bpf_local_storage gets zero initialized. Bestie, that's the whole vibe. 🫶

Return — a bpf_local_storage pointer comes back on success. 💖

But NULL if it wasn't found, or if literally anything went wrong while adding a new bpf_local_storage. 😭

`long bpf_task_storage_delete(struct bpf_map map, struct task_struct task)`

Description

Delete a bpf_local_storage from a task. Bye bye, storage. 👋💅

Return — 0 on success, iconic, we love to see it.

-ENOENT if the bpf_local_storage cannot be found. Like, it's not there, girl, what do you want from me. 🙈

`struct task_struct *bpf_get_current_task_btf(void)`

Description

Returns a BTF pointer to the "current" task. 🌟 And plot twist — this pointer can also be handed to helpers that accept an ARG_PTR_TO_BTF_ID of type task_struct. So it's multitasking. Main character energy fr. 👑

Return — pointer to the current task. 💜

`long bpf_bprm_opts_set(struct linux_binprm *bprm, u64 flags)`

Description

Set or clear certain options on bprm, okay, pay attention bestie:

BPF_F_BPRM_SECUREEXEC — set the secureexec bit, which sets the AT_SECURE auxv for glibc. 🎀 And the bit gets cleared if the flag is not specified. So not passing it = unsetting it. No omission arc here, it's active.

Return — -EINVAL if invalid flags get passed, zero otherwise. Sorted. ✨

`u64 bpf_ktime_get_coarse_ns(void)`

Description

Returns a coarse-grained version of the time elapsed since system boot, in nanoseconds. ⏰ And no, it does not include time the system was suspended — that sleep time simply doesn't count, periodt.

See: clock_gettime(CLOCK_MONOTONIC_COARSE)

Return — current ktime. 🌸

`long bpf_ima_inode_hash(struct inode inode, void dst, u32 size)`

Description

Returns the stored IMA hash of the inode — if it's available, no promises. 🦋 And heads up: if the hash is larger than size, then only size bytes get copied to dst. So it truncates. Just so you know. 💀

Return — the hash_algo is returned on success, -EOPNOTSUPP if IMA is disabled, or -EINVAL if invalid arguments are passed. 🌈

`struct socket bpf_sock_from_file(struct file file)`

Description

If the given file represents a socket, it returns the associated socket. Glow-up reveal moment. ✨

Return — a pointer to a struct socket on success, or NULL if the file is not a socket. Like, wrong number, wrong vibe. 💅

`long bpf_check_mtu(void ctx, u32 ifindex, u32 mtu_len, s32 len_diff, u64 flags)`

Description

Check packet size against exceeding the MTU of a net device — based on ifindex. 📏 This helper will likely be used in combination with helpers that adjust/change the packet size. Teamwork makes the dream work.

The argument len_diff can be used for querying with a planned size change. 🧚 This lets you check MTU prior to changing packet ctx. And here's the cute lil edge case: providing a len_diff adjustment that is larger than the actual packet size — which results in a negative packet size — will in principle not exceed the MTU, which is why it is not considered a failure. 😌 Other BPF helpers are needed for performing the planned size change; therefore the responsibility for catching a negative packet size belongs in those helpers. Not ours. Not our problem. 🙈

Specifying ifindex zero means the MTU check is performed against the current net device. 🎀 This is practical if this isn't used prior to redirect.

On input mtu_len must be a valid pointer, else the verifier will reject your BPF program. No cap. 🛑 If the value mtu_len is initialized to zero, then the ctx packet size is used. When a value of mtu_len is provided as input, this specifies the L3 length that the MTU check is done against. ⭐ And remember: XDP and TC length operate at L2, but this value is L3, as this correlates to MTU and IP-header tot_len values which are L3 (similar behavior as bpf_fib_lookup). Keeping up? Good. 💖

The Linux kernel route table can configure MTUs on a more specific per-route level, which is not provided by this helper. 🚫 For route level MTU checks, use the bpf_fib_lookup() helper instead. Different helper, different job.

ctx is either struct xdp_md for XDP programs or struct sk_buff for tc cls_act programs. Two options, pick your fighter. 🥊

The flags argument can be a combination of one or more of the following values:

BPF_MTU_CHK_SEGS — this flag will only work for ctx struct sk_buff. If the packet context contains extra packet segment buffers (often known as GSO skb), then the MTU check is harder to check at this point, because in the transmit path it is possible for the skb packet to get re-segmented (depending on net device features). 😭 This could still be an MTU violation, so this flag enables performing an MTU check against segments, with a different violation return code to tell it apart. And no, the check cannot use len_diff. 🙅‍♀️

On return, the mtu_len pointer contains the MTU value of the net device. 🌟 Remember, the net device configured MTU is the L3 size, which is what's returned here, and XDP and TC length operate at L2. The helper takes this into account for you, but remember it when you use the MTU value in your BPF code. That's the PSA. 📢

Return

0 on success, and it populates the MTU value in the mtu_len pointer. ✨
< 0 if any input argument is invalid (mtu_len not updated). 💀

MTU violations return positive values, but also populate the MTU value in the mtu_len pointer, since this can be needed for implementing PMTU handling: 🫶

BPF_MTU_CHK_RET_FRAG_NEEDED
BPF_MTU_CHK_RET_SEGS_TOOBIG

`long bpf_for_each_map_elem(struct bpf_map map, void callback_fn, void *callback_ctx, u64 flags)`

Description

For each element in map, call the callback_fn function with map, callback_ctx and other map-specific parameters. 🔁 The callback_fn should be a static function and the callback_ctx should be a pointer to the stack. 📌 The flags is used to control certain aspects of the helper — and currently, the flags must be 0. That's it, that's the whole options menu. 💀

The following are a list of supported map types and their respective expected callback signatures: 👑

BPF_MAP_TYPE_HASH, BPF_MAP_TYPE_PERCPU_HASH, BPF_MAP_TYPE_LRU_HASH, BPF_MAP_TYPE_LRU_PERCPU_HASH, BPF_MAP_TYPE_ARRAY, BPF_MAP_TYPE_PERCPU_ARRAY

long (*callback_fn)(struct bpf_map *map, const void *key,
void *value, void *ctx);

For per_cpu maps, the map_value is the value on the CPU where the bpf_prog is running. 🌈 So yes, CPU-specific. Know that.

If callback_fn returns 0, the helper will continue to the next element. ➡️ If the return value is 1, the helper will skip the rest of the elements and return. 🛑 Other return values are not used now. Not now, bestie. Maybe later. 🙈

Return — the number of traversed map elements for success, -EINVAL for invalid flags. 💅

`long bpf_snprintf(char str, u32 str_size, const char fmt, u64 *data, u32 data_len)`

Description

Outputs a string into the str buffer of size str_size, based on a format string stored in a read-only map pointed by fmt. 🎀

Each format specifier in fmt corresponds to one u64 element in the data array. ⭐ For strings and pointers where pointees are accessed, only the pointer values are stored in the data array. The data_len is the size of data in bytes — and it must be a multiple of 8. Math check, do it. 🔢

Formats %s and %p{i,I}{4,6} require reading kernel memory. 🧠 And reading kernel memory may fail due to either an invalid address, or a valid address but requiring a major memory fault. If reading kernel memory fails, the string for %s will be an empty string, and the IP address for %p{i,I}{4,6} will be 0. 😭 And not returning an error to the bpf program is consistent with what bpf_trace_printk() does for now. That's the behavior, we're just reporting it.

Return — the strictly positive length of the formatted string, including the trailing zero character. ✨ If the return value is greater than str_size, then str contains a truncated string, guaranteed to be zero-terminated except when str_size is 0. 🫶

Or -EBUSY if the per-CPU memory copy buffer is busy. Busy girl, busy buffer. 💅

`long bpf_sys_bpf(u32 cmd, void *attr, u32 attr_size)`

Description

Execute the bpf syscall with the given arguments. 🖥️ That's the whole thing, she's simple like that.

Return — a syscall result. 📜

`long bpf_btf_find_by_name_kind(char *name, int name_sz, u32 kind, int flags)`

Description

Find a BTF type with the given name and kind in vmlinux BTF or in a module's BTFs. 🔍 Scavenger hunt but make it kernel.

Return — returns btf_id and btf_obj_fd in the lower and upper 32 bits. 🌸 Yes, two values, one long, they're roommates. iykyk.

`long bpf_sys_close(u32 fd)`

Description

Execute the close syscall for the given FD. 🚪 Shut the door on it.

Return — a syscall result. ✨

`long bpf_timer_init(struct bpf_timer timer, struct bpf_map map, u64 flags)`

Description

Initialize the timer. ⏲️ The first 4 bits of flags specify the clockid. Only CLOCK_MONOTONIC, CLOCK_REALTIME and CLOCK_BOOTTIME are allowed. 👑 All other bits of flags are reserved — hands off. And the verifier will reject the program if timer is not from the same map. Sorry bestie, matching pairs only. 💅

Return — 0 on success. -EBUSY if timer is already initialized. -EINVAL if invalid flags are passed. -EPERM if timer is in a map that doesn't have any user references. 🛑

And listen: user space should either hold a file descriptor to a map with timers or pin such map in bpffs. 📌 When the map is unpinned or the file descriptor is closed, all timers in the map will be cancelled and freed. Like, all of them. Gone. 🫠

`long bpf_timer_set_callback(struct bpf_timer timer, void callback_fn)`

Description

Configure the timer to call the callback_fn static function. 📞 Callback, you're up.

bpf_timer_start — ok bestie, this is THE moment

long bpf_timer_start(struct bpf_timer *timer, u64 nsecs, u64 flags)

Description — Sets the timer to go off N nanoseconds from like right now, iykyk ⏰. The callback you configured gets invoked in soft irq context on some cpu, and it does NOT repeat unless you call bpf_timer_start() all over again. And when you do re-arm it, the next invocation can legit migrate to a whole different cpu 💀. Since struct bpf_timer lives as a field inside a map element, the map owns the timer, periodt. bpf_timer_set_callback() bumps the refcnt of the BPF program to keep that callback_fn code valid, no cap 🫶. When the user space reference to a map drops to zero, every timer in the map gets cancelled and the corresponding program's refcnts get decremented. Why, you ask? To make sure that a Ctrl+C on a user process doesn't leave any timers running, literally so real of them. And if the map is pinned in bpffs, the callback_fn can re-arm itself indefinitely ✨. Also the bpf_map_update/delete_elem() helpers and user space sys_bpf commands cancel and free the timer in the given map element. The map can contain timers that invoke callback_fn-s from different programs, messy but iconic. The same callback_fn can serve different timers from different maps if the key/value layout matches across maps. Every bpf_timer_set_callback() can have its own different callback_fn.

flags can be one of these:

BPF_F_TIMER_ABS — start the timer in absolute expire value instead of the default relative one 💅
BPF_F_TIMER_CPU_PIN — timer gets pinned to the CPU of the caller, obsessed 🔥

Return — 0 on success bestie. -EINVAL if the timer wasn't initialized with bpf_timer_init() earlier, or if you passed invalid flags.

bpf_timer_cancel — nah we're ending this

long bpf_timer_cancel(struct bpf_timer *timer)

Description — Cancels the timer, and if callback_fn was already running, it waits for it to finish ⏳.

Return — 0 if the timer was not active. 1 if the timer was active. -EINVAL if the timer wasn't initialized with bpf_timer_init() earlier. -EDEADLK if callback_fn tried to call bpf_timer_cancel() on its own timer — which would've led to a deadlock otherwise 💀 so yeah, don't do that.

bpf_get_func_ip — where u at

u64 bpf_get_func_ip(void *ctx)

Description — Gets the address of the traced function (for tracing and kprobe programs). And when you call it for a kprobe program attached as a uprobe, it returns the probe address for both entry and return uprobe 👑

Return — Address of the traced function for kprobe. 0 for kprobes placed within the function (not at the entry). Address of the probe for uprobe and return uprobe.

bpf_get_attach_cookie — the little extra ✨

u64 bpf_get_attach_cookie(void *ctx)

Description — Gets the bpf_cookie value provided (optionally) when the program got attached. It might be different for each individual attachment, even if the BPF program itself is the same one. Expects the BPF program context ctx as the first argument.

Supported for these program types, bestie:

kprobe/uprobe;
tracepoint;
perf_event.

Return — The value you, the user, specified at BPF link creation/attachment time. Or 0, if you didn't specify anything 🙈.

bpf_task_pt_regs — giving registers

long bpf_task_pt_regs(struct task_struct *task)

Description — Gets the struct pt_regs associated with task.

Return — A pointer to struct pt_regs 💅

bpf_get_branch_snapshot — caught you in 4k

long bpf_get_branch_snapshot(void *entries, u32 size, u64 flags)

Description — Grabs the branch trace from hardware engines like Intel LBR 🔥. The hardware engine gets stopped shortly after the helper is called, so like... the user needs to filter branch entries based on the actual use case, ngl. To capture branch trace before the trigger point of the BPF program, you should call the helper at the beginning of the BPF program, iykyk.

The data is stored as struct perf_branch_entry into the output buffer entries. size is the size of entries in bytes. flags is reserved for now and must be zero.

Return — On success, the number of bytes written to buf. On error, a negative value.

-EINVAL if flags is not zero.
-ENOENT if the architecture doesn't support branch records 😭
bpf_trace_vprintk — vprintk but make it ✨format array✨

long bpf_trace_vprintk(const char *fmt, u32 fmt_size, const void *data,
u32 data_len)

Description — Behaves like the bpf_trace_printk() helper, but takes an array of u64 to format and can handle more format args as a result 👑. Arguments are to be used as in the bpf_seq_printf() helper.

Return — The number of bytes written to the buffer, or a negative error if it failed 💀

bpf_skc_to_unix_sock — glow up that cast

struct unix_sock *bpf_skc_to_unix_sock(void *sk)

Description — Dynamically casts an sk pointer to a unix_sock pointer. Slay.

Return — sk if the casting is valid, or NULL otherwise.

bpf_kallsyms_lookup_name — find that symbol bestie

long bpf_kallsyms_lookup_name(const char *name, int name_sz, int flags,
u64 *res)

Description — Gets the address of a kernel symbol, returned in res. res gets set to 0 if the symbol is not found, rip.

Return — On success, zero. On error, a negative value.

-EINVAL if flags is not zero.
-EINVAL if the string name is not the same size as name_sz.
-ENOENT if the symbol is not found.
-EPERM if the caller doesn't have permission to obtain the kernel address 🙈
bpf_find_vma — vma hunting szn

long bpf_find_vma(struct task_struct *task, u64 addr, void *callback_fn,
void *callback_ctx, u64 flags)

Description — Finds the vma of task that contains addr, then calls the callback_fn function with task, vma, and callback_ctx. The callback_fn should be a static function and the callback_ctx should be a pointer to the stack. The flags is used to control certain aspects of the helper. Currently, the flags must be 0.

The expected callback signature is

long (*callback_fn)(struct task_struct *task, struct
vm_area_struct *vma, void *callback_ctx);

Return — 0 on success. -ENOENT if task->mm is NULL, or no vma contains addr. -EBUSY if it failed to try lock mmap_lock. -EINVAL for invalid flags.

bpf_loop — we're looping, we're looping 💃

long bpf_loop(u32 nr_loops, void *callback_fn, void *callback_ctx, u64
flags)

Description — For nr_loops, call the callback_fn function with callback_ctx as the context parameter. The callback_fn should be a static function and the callback_ctx should be a pointer to the stack. The flags is used to control certain aspects of the helper. Currently, the flags must be 0. And rn nr_loops is limited to 1 << 23 (~8 million) loops, so like, know your limits bestie 💅

long (*callback_fn)(u64 index, void *ctx);

where index is the current index in the loop. The index is zero-indexed ✨

If callback_fn returns 0, the helper continues to the next loop. If the return value is 1, the helper will skip the rest of the loops and return. Other return values are not used right now, and they'll get rejected by the verifier — watch out 💀

Return — The number of loops performed, -EINVAL for invalid flags, -E2BIG if nr_loops exceeds the maximum number of loops.

bpf_strncmp — comparing, but safely bestie

long bpf_strncmp(const char *s1, u32 s1_sz, const char *s2)

Description — Does a strncmp() between s1 and s2. s1 doesn't need to be null-terminated, and s1_sz is the maximum storage size of s1. s2 must be a read-only string.

Return — An integer less than, equal to, or greater than zero if the first s1_sz bytes of s1 is found to be less than, to match, or be greater than s2.

bpf_get_func_arg — gimme that argument

long bpf_get_func_arg(void *ctx, u32 n, u64 *value)

Description — Gets the n-th argument register (zero based) of the traced function (for tracing programs), returned in value 👑

Return — 0 on success. -EINVAL if n >= argument register count of traced function.

bpf_get_func_ret — what did it return tho

long bpf_get_func_ret(void *ctx, u64 *value)

Description — Gets the return value of the traced function (for tracing programs) in value ✨

Return — 0 on success. -EOPNOTSUPP for tracing programs other than BPF_TRACE_FEXIT or BPF_MODIFY_RETURN 🙈

bpf_get_func_arg_cnt — counting the registers

long bpf_get_func_arg_cnt(void *ctx)

Description — Gets the number of registers of the traced function (for tracing programs) where the function arguments are stored in those registers.

Return — The number of argument registers of the traced function 💅

bpf_get_retval — peek the retval

int bpf_get_retval(void)

Description — Gets the BPF program's return value that will be returned to the upper layers.

This helper is currently supported by cgroup programs and only by the hooks where the BPF program's return value is returned to the userspace via errno, iykyk.

Return — The BPF program's return value 🫶

bpf_set_retval — set the retval, main character energy

int bpf_set_retval(int retval)

Description — Sets the BPF program's return value that will be returned to the upper layers.

This helper is currently supported by cgroup programs and only by the hooks where the BPF program's return value is returned to the userspace via errno.

Now bestie, pls note this corner case where the program exports an error via bpf_set_retval but signals success via 'return 1':

bpf_set_retval(-EPERM); return 1;

In this case, the BPF program's return value will use helper's -EPERM. This still holds true for cgroup/bind{4,6} which supports extra 'return 3' success case. 💅

Return — 0 on success, or a negative error in case of failure.

🌸 `u64 bpf_xdp_get_buff_len(struct xdp_buff *xdp_md)`

Description — Get the total size of a given xdp buff (linear and paged area). 🦋

Return — The total size of a given xdp buffer.

🎀 `long bpf_xdp_load_bytes(struct xdp_buff xdp_md, u32 offset, void buf, u32 len)`

Description — This helper is literally your easy-peasy way to load data from an xdp buffer, bestie. It can be used to load len bytes from offset from the frame associated to xdp_md, into the buffer pointed by buf. ✨

Return — 0 on success, or a negative error in case of failure.

🦋 `long bpf_xdp_store_bytes(struct xdp_buff xdp_md, u32 offset, void buf, u32 len)`

Description — Store len bytes from buffer buf into the frame associated to xdp_md, at offset. Putting things back, we love to see it. 💖

Return — 0 on success, or a negative error in case of failure.

🧚 `long bpf_copy_from_user_task(void dst, u32 size, const void user_ptr, struct task_struct *tsk, u64 flags)`

Description — Read size bytes from user space address user_ptr in tsk's address space, and stores the data in dst. flags is not used yet and is just sitting there for future extensibility, iconic of it. 🫶

Heads up: this helper can only be used by sleepable programs. No cap.

Return — 0 on success, or a negative error in case of failure. On error dst buffer is zeroed out.

💅 `long bpf_skb_set_tstamp(struct sk_buff *skb, u64 tstamp, u32 tstamp_type)`

Description — Change the __sk_buff->tstamp_type to tstamp_type and set tstamp to the __sk_buff->tstamp together. A two-for-one, serving looks. ✨

If there is no need to change the __sk_buff->tstamp_type, the tstamp value can be directly written to __sk_buff->tstamp instead. 🙂‍↕️

BPF_SKB_TSTAMP_DELIVERY_MONO is the only tstamp that will be kept during bpf_redirect_*(). A non zero tstamp must be used with the BPF_SKB_TSTAMP_DELIVERY_MONO tstamp_type.

A BPF_SKB_TSTAMP_UNSPEC tstamp_type can only be used with a zero tstamp.

Only IPv4 and IPv6 skb->protocol are supported. 🚫

This function is most useful when it needs to set a mono delivery time to __sk_buff->tstamp and then bpf_redirect_*() to the egress of an iface. For example, changing the (rcv) timestamp in __sk_buff->tstamp at ingress to a mono delivery time and then bpf_redirect_*() to <sch_fq@phy-dev>. The whole glow-up pipeline, periodt. 💅

Return — 0 on success. -EINVAL for invalid input, -EOPNOTSUPP for unsupported protocol.

🔥 `long bpf_ima_file_hash(struct file file, void dst, u32 size)`

Description — Returns a calculated IMA hash of the file. If the hash is larger than size, then only size bytes will be copied to dst. Trimming it down to fit, we respect it. 🎀

Return — The hash_algo is returned on success, -EOPNOTSUPP if the hash calculation failed or -EINVAL if invalid arguments are passed.

👑 `void bpf_kptr_xchg(void dst, void *ptr)`

Description — Exchange kptr at pointer dst with ptr, and return the old value. dst can be map value or local kptr. ptr can be NULL, otherwise it must be a referenced pointer which will be released when this helper is called. Swapping out, no drama. 💜

Return — The old value of kptr (which can be NULL). The returned pointer if not NULL, is a reference which must be released using its corresponding release function, or moved into a BPF map before program exit.

Do not lose track of that reference, bestie. The verifier is watching. 😭

🌈 `void bpf_map_lookup_percpu_elem(struct bpf_map map, const void *key, u32 cpu)`

Description — Perform a lookup in percpu map for an entry associated to key on cpu.

Return — Map value associated to key on cpu, or NULL if no entry was found or cpu is invalid.

🍓 `struct mptcp_sock bpf_skc_to_mptcp_sock(void sk)`

Description — Dynamically cast a sk pointer to a mptcp_sock pointer. ✨

Return — sk if casting is valid, or NULL otherwise.

🦋 `long bpf_dynptr_from_mem(void data, u64 size, u64 flags, struct bpf_dynptr ptr)`

Description — Get a dynptr to local memory data.

data must be a ptr to a map value. The maximum size supported is DYNPTR_MAX_SIZE. flags is currently unused.

Return — 0 on success, -E2BIG if the size exceeds DYNPTR_MAX_SIZE, -EINVAL if flags is not 0.

💖 `long bpf_ringbuf_reserve_dynptr(void ringbuf, u32 size, u64 flags, struct bpf_dynptr ptr)`

Description — Reserve size bytes of payload in a ring buffer ringbuf through the dynptr interface. flags must be 0.

Please note that a corresponding bpf_ringbuf_submit_dynptr or bpf_ringbuf_discard_dynptr must be called on ptr, even if the reservation fails. This is enforced by the verifier. You cannot skip the cleanup step, I don't make the rules. 😭

Return — 0 on success, or a negative error in case of failure.

🎀 `void bpf_ringbuf_submit_dynptr(struct bpf_dynptr *ptr, u64 flags)`

Description — Submit reserved ring buffer sample, pointed to by data, through the dynptr interface. This is a no-op if the dynptr is invalid/null. 🙂‍↕️

For more information on flags, please see 'bpf_ringbuf_submit'.

Return — Nothing. Always succeeds. Flawless victory. 👑

🌸 `void bpf_ringbuf_discard_dynptr(struct bpf_dynptr *ptr, u64 flags)`

Description — Discard reserved ring buffer sample through the dynptr interface. This is a no-op if the dynptr is invalid/null.

For more information on flags, please see 'bpf_ringbuf_discard'.

Return — Nothing. Always succeeds.

🧚 `long bpf_dynptr_read(void dst, u64 len, const struct bpf_dynptr src, u64 offset, u64 flags)`

Description — Read len bytes from src into dst, starting from offset into src. flags is currently unused.

Return — 0 on success, -E2BIG if offset + len exceeds the length of src's data, -EINVAL if src is an invalid dynptr or if flags is not 0.

💅 `long bpf_dynptr_write(const struct bpf_dynptr dst, u64 offset, void src, u64 len, u64 flags)`

Description — Write len bytes from src into dst, starting from offset into dst.

flags must be 0 except for skb-type dynptrs.

For skb-type dynptrs:

All data slices of the dynptr are automatically invalidated after bpf_dynptr_write(). This is because writing may pull the skb and change the underlying packet buffer. 💀
For flags, please see the flags accepted by bpf_skb_store_bytes().

Return — 0 on success, -E2BIG if offset + len exceeds the length of dst's data, -EINVAL if dst is an invalid dynptr or if dst is a read-only dynptr or if flags is not correct. For skb-type dynptrs, other errors correspond to errors returned by bpf_skb_store_bytes().

⭐ `void bpf_dynptr_data(const struct bpf_dynptr ptr, u64 offset, u64 len)`

Description — Get a pointer to the underlying dynptr data.

len must be a statically known value. The returned data slice is invalidated whenever the dynptr is invalidated.

skb and xdp type dynptrs may not use bpf_dynptr_data. They should instead use bpf_dynptr_slice and bpf_dynptr_slice_rdwr. Using the wrong one is giving "read the room" energy, and the verifier will clock it. 🙈

Return — Pointer to the underlying dynptr data, NULL if the dynptr is read-only, if the dynptr is invalid, or if the offset and length is out of bounds.

🔥 `s64 bpf_tcp_raw_gen_syncookie_ipv4(struct iphdr iph, struct tcphdr th, u32 th_len)`

Description — Try to issue a SYN cookie for the packet with corresponding IPv4/TCP headers, iph and th, without depending on a listening socket. Fully independent, main character energy fr. 👑

iph points to the IPv4 header.

th points to the start of the TCP header, while th_len contains the length of the TCP header (at least sizeof(struct tcphdr)).

Return — On success, lower 32 bits hold the generated SYN cookie in followed by 16 bits which hold the MSS value for that cookie, and the top 16 bits are unused.

On failure, the returned value is one of the following:

-EINVAL if th_len is invalid.

💜 `s64 bpf_tcp_raw_gen_syncookie_ipv6(struct ipv6hdr iph, struct tcphdr th, u32 th_len)`

Description — Try to issue a SYN cookie for the packet with corresponding IPv6/TCP headers, iph and th, without depending on a listening socket.

iph points to the IPv6 header.

th points to the start of the TCP header, while th_len contains the length of the TCP header (at least sizeof(struct tcphdr)).

Return — On success, lower 32 bits hold the generated SYN cookie in followed by 16 bits which hold the MSS value for that cookie, and the top 16 bits are unused.

On failure, the returned value is one of the following:

-EINVAL if th_len is invalid.
-EPROTONOSUPPORT if CONFIG_IPV6 is not builtin.

🍓 `long bpf_tcp_raw_check_syncookie_ipv4(struct iphdr iph, struct tcphdr th)`

Description — Check whether iph and th contain a valid SYN cookie ACK without depending on a listening socket.

iph points to the IPv4 header.

th points to the TCP header.

Return — 0 if iph and th are a valid SYN cookie ACK.

On failure, the returned value is one of the following:

-EACCES if the SYN cookie is not valid.

💖 `long bpf_tcp_raw_check_syncookie_ipv6(struct ipv6hdr iph, struct tcphdr th)`

alright bestie we are still inside the same helper from last time 💖 no new heading needed, she's still talking —

iph points to the IPv6 header, and th points to the TCP header, no listening socket required, she just checks the packets by themselves.

Return — 0 if iph and th are a valid SYN cookie ACK 🎀

And if it's giving failure, you get one of these instead:

-EACCES — the SYN cookie is not valid 😭
-EPROTONOSUPPORT — CONFIG_IPV6 is not builtin 💀

`u64 bpf_ktime_get_tai_ns(void)`

Description — a nonsettable system-wide clock derived from wall-clock time but ignoring leap seconds. The way this one doesn't experience discontinuities and backwards jumps caused by NTP inserting leap seconds like CLOCK_REALTIME does? iconic. no notes. 👑

See: clock_gettime(CLOCK_TAI)

Return — the current ktime. ✨

`long bpf_user_ringbuf_drain(struct bpf_map map, void callback_fn, void *ctx, u64 flags)`

Description — drain samples from the specified user ring buffer, and invoke the provided callback for each such sample:

long (*callback_fn)(const struct bpf_dynptr *dynptr, void *ctx);

so here's the vibe check on that callback: if callback_fn returns 0, the helper keeps going and tries to drain the next sample, up to a maximum of BPF_MAX_USER_RINGBUF_SAMPLES samples 🦋. If the return value is 1, the helper skips the rest of the samples and returns. Other return values are not used right now, and they'll be rejected by the verifier — so don't try it, bestie 🙈

Return — the number of drained samples if no error was encountered while draining samples, or 0 if no samples were present in the ring buffer. And here's the cute part: if a user-space producer was epoll-waiting on this map, and at least one sample was drained, they'll get an event notification telling them there's available space in the ring buffer 🌸. If the BPF_RB_NO_WAKEUP flag is passed to this function, no wakeup notification will be sent. If the BPF_RB_FORCE_WAKEUP flag is passed, a wakeup notification will be sent even if no sample was drained.

On failure, the returned value is one of the following:

-EBUSY — the ring buffer is contended, and another calling context was concurrently draining the ring buffer 😤
-EINVAL — user-space isn't properly tracking the ring buffer, because the producer position isn't aligned to 8 bytes, or a sample isn't aligned to 8 bytes, or the producer position doesn't match the advertised length of a sample 🫠
-E2BIG — user-space tried to publish a sample which is larger than the size of the ring buffer, or which cannot fit within a struct bpf_dynptr 💀

`void bpf_cgrp_storage_get(struct bpf_map map, struct cgroup cgroup, void value, u64 flags)`

Description — get a bpf_local_storage from the cgroup. 🎀

Logically, you could think of it as getting the value from a map with cgroup as the key. From this perspective, the usage is not much different from bpf_map_lookup_elem(map, &cgroup) — except this helper enforces that the key must be a cgroup struct and the map must also be a BPF_MAP_TYPE_CGRP_STORAGE. periodt. 💅

In reality though (plot twist), the local-storage value is embedded directly inside of the cgroup object itself, rather than being located in the BPF_MAP_TYPE_CGRP_STORAGE map. So when the local-storage value is queried for some map on a cgroup object, the kernel will perform an O(n) iteration over all of the live local-storage values for that cgroup object until the local-storage value for the map is found. that's the fine print bestie, don't sleep on it ⭐

An optional flags (BPF_LOCAL_STORAGE_GET_F_CREATE) can be used such that a new bpf_local_storage will be created if one does not exist. value can be used together with BPF_LOCAL_STORAGE_GET_F_CREATE to specify the initial value of a bpf_local_storage. If value is NULL, the new bpf_local_storage will be zero initialized. ✨

Return — a bpf_local_storage pointer is returned on success. 🫶

And NULL if not found or there was an error in adding a new bpf_local_storage. 😭

`long bpf_cgrp_storage_delete(struct bpf_map map, struct cgroup cgroup)`

Description — delete a bpf_local_storage from a cgroup. bye bestie 👋

Return — 0 on success. 🎉

-ENOENT if the bpf_local_storage cannot be found. 💀

🌈 EXAMPLES 🌈

Example usage for most of the eBPF helpers listed in this manual page are available within the Linux kernel sources, at the following locations:

samples/bpf/
tools/testing/selftests/bpf/

like, go read the source bestie, it's all there 📚✨

📜 LICENSE 📜

eBPF programs can have an associated license, passed along with the bytecode instructions to the kernel when the programs are loaded. The format for that string is identical to the one in use for kernel modules (Dual licenses, such as "Dual BSD/GPL", may be used). Some helper functions are only accessible to programs that are compatible with the GNU General Public License (GNU GPL). 👑

In order to use such helpers, the eBPF program must be loaded with the correct license string passed (via attr) to the bpf() system call, and this generally translates into the C source code of the program containing a line similar to the following:

char ____license[] __attribute__((section("license"), used)) = "GPL";

yes this actually matters, no you can't just vibe your way around the license check 🚫

🛠 IMPLEMENTATION 🛠

This manual page is an effort to document the existing eBPF helper functions. But as of this writing, the BPF sub-system is under heavy development — new eBPF program or map types are added, along with new helper functions, and some helpers are occasionally made available for additional program types. So in spite of the efforts of the community, this page might not be up-to-date 😭. If you want to check by yourself what helper functions exist in your kernel, or what types of programs they can support, here are some files among the kernel tree that you may be interested in:

include/uapi/linux/bpf.h is the main BPF header. It contains the full list of all helper functions, as well as many other BPF definitions including most of the flags, structs or constants used by the helpers.
net/core/filter.c contains the definition of most network-related helper functions, and the list of program types from which they can be used.
kernel/trace/bpf_trace.c is the equivalent for most tracing program-related helpers.
kernel/bpf/verifier.c contains the functions used to check that valid types of eBPF maps are used with a given helper function.
kernel/bpf/ directory contains other files in which additional helpers are defined (for cgroups, sockmaps, etc.).
The bpftool utility can be used to probe the availability of helper functions on the system (as well as supported program and map types, and a number of other parameters). To do so, run bpftool feature probe (see bpftool-feature(8) for details). Add the unprivileged keyword to list features available to unprivileged users. 🔍

Compatibility between helper functions and program types can generally be found in the files where helper functions are defined. Look for the struct bpf_func_proto objects and for functions returning them: these functions contain a list of helpers that a given program type can call. Note that the default: label of the switch ... case used to filter helpers can call other functions, themselves allowing access to additional helpers. The requirement for GPL license is also in those struct bpf_func_proto. 🧚

Compatibility between helper functions and map types can be found in the check_map_func_compatibility() function in file kernel/bpf/verifier.c.

Helper functions that invalidate the checks on data and data_end pointers for network processing are listed in function bpf_helper_changes_pkt_data() in file net/core/filter.c. ⚠️

and that's a wrap on the whole entire page bestie, we did it 🎉💖 this was BPF-HELPERS(7) from Linux v7.3, dated 2026-08-17 — go forth and write some unhinged-but-valid eBPF, you absolute icon 👑✨

📦 where this page comes from

arch rolling: man-pages 6.20-1 · BSD-2-Clause AND BSD-3-Clause AND BSD-4.3TAHOE AND BSD-4-Clause-UC AND GPL-1.0-or-later AND GPL-2.0-only AND GPL-2.0-or-later AND GPL-3.0-or-later AND LGPL-3.0-linking-exception AND LGPL-3.0-or-later AND LicenseRef-Public-Domain AND LicenseRef-UltraPermissive AND Linux-man-pages-1-para AND Linux-man-pages-copyleft AND Linux-man-pages-copyleft-2-para AND Linux-man-pages-copyleft-var AND MIT AND Spencer-94
nixpkgs unstable: man-pages 6.20
📜 the page's license: Linux-man-pages-copyleft
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash on 2026-11-03: a language model's rewrite of the original, under its terms

BPF-HELPERS(7)

💖 BPF-HELPERS(7) 💖

🌸 DESCRIPTION 🌸

👑 HELPERS 👑

🦋 bpf_map_lookup_elem

🦋 bpf_map_update_elem

🦋 bpf_map_delete_elem

🦋 bpf_probe_read

🦋 bpf_ktime_get_ns

🦋 bpf_trace_printk

🦋 bpf_get_prandom_u32

🦋 bpf_get_smp_processor_id

🦋 bpf_skb_store_bytes

🦋 bpf_l3_csum_replace

🦋 bpf_l4_csum_replace

Description ✨

Return 💫

long bpf_tail_call(void *ctx, struct bpf_map *prog_array_map, u32 index)

Description 🌈

Return 💫

long bpf_clone_redirect(struct sk_buff *skb, u32 ifindex, u64 flags)

Description 🎀

Return 💫

u64 bpf_get_current_pid_tgid(void)

Description 🌸

Return 💫

u64 bpf_get_current_uid_gid(void)

Description 🌸

Return 💫

long bpf_get_current_comm(void *buf, u32 size_of_buf)

Description 🦋

Return 💫

u32 bpf_get_cgroup_classid(struct sk_buff *skb)

Description 🧚

Return 💫

long bpf_skb_vlan_push(struct sk_buff *skb, __be16 vlan_proto, u16 vlan_tci)

Description 🎀

Return 💫

long bpf_skb_vlan_pop(struct sk_buff *skb)

Description 🦋

Return 💫

long bpf_skb_get_tunnel_key(struct sk_buff *skb, struct bpf_tunnel_key *key, u32 size, u64 flags)

Description 🌈

long bpf_skb_set_tunnel_key(struct sk_buff *skb, struct bpf_tunnel_key *key, u32 size, u64 flags)

u64 bpf_perf_event_read(struct bpf_map *map, u64 flags)

long bpf_redirect(u32 ifindex, u64 flags)

u32 bpf_get_route_realm(struct sk_buff *skb)

long bpf_perf_event_output(void *ctx, struct bpf_map *map, u64 flags, void *data, u64 size)

long bpf_skb_load_bytes(const void *skb, u32 offset, void *to, u32 len)

long bpf_get_stackid(void *ctx, struct bpf_map *map, u64 flags)

s64 bpf_csum_diff(__be32 *from, u32 from_size, __be32 *to, u32 to_size, __wsum seed)

long bpf_skb_get_tunnel_opt(struct sk_buff *skb, void *opt, u32 size)

long bpf_skb_set_tunnel_opt(struct sk_buff *skb, void *opt, u32 size)

long bpf_skb_change_proto(struct sk_buff *skb, __be16 proto, u64 flags)

long bpf_skb_change_type(struct sk_buff *skb, u32 type)

long bpf_skb_under_cgroup(struct sk_buff *skb, struct bpf_map *map, u32 index)

u32 bpf_get_hash_recalc(struct sk_buff *skb)

u64 bpf_get_current_task(void)

long bpf_probe_write_user(void *dst, const void *src, u32 len)

long bpf_current_task_under_cgroup(struct bpf_map *map, u32 index)

long bpf_skb_change_tail(struct sk_buff *skb, u32 len, u64 flags)

long bpf_skb_pull_data(struct sk_buff *skb, u32 len)

s64 bpf_csum_update(struct sk_buff *skb, __wsum csum)

void bpf_set_hash_invalid(struct sk_buff *skb)

long bpf_get_numa_node_id(void)

long bpf_skb_change_head(struct sk_buff *skb, u32 len, u64 flags)

long bpf_xdp_adjust_head(struct xdp_buff *xdp_md, int delta)

long bpf_probe_read_str(void *dst, u32 size, const void *unsafe_ptr)

u64 bpf_get_socket_cookie(struct sk_buff *skb)

u64 bpf_get_socket_cookie(struct bpf_sock_addr *ctx)

u64 bpf_get_socket_cookie(struct bpf_sock_ops *ctx)

u64 bpf_get_socket_cookie(struct sock *sk)

u32 bpf_get_socket_uid(struct sk_buff *skb)

long bpf_set_hash(struct sk_buff *skb, u32 hash)

long bpf_setsockopt(void *bpf_socket, int level, int optname, void *optval, int optlen)

long bpf_skb_adjust_room(struct sk_buff *skb, s32 len_diff, u32 mode, u64 flags)

Description

Description

Description

Description

🦋 `bpf_map_lookup_elem`

🦋 `bpf_map_update_elem`

🦋 `bpf_map_delete_elem`

🦋 `bpf_probe_read`

🦋 `bpf_ktime_get_ns`

🦋 `bpf_trace_printk`

🦋 `bpf_get_prandom_u32`

🦋 `bpf_get_smp_processor_id`

🦋 `bpf_skb_store_bytes`

🦋 `bpf_l3_csum_replace`

🦋 `bpf_l4_csum_replace`

`long bpf_tail_call(void ctx, struct bpf_map prog_array_map, u32 index)`

`long bpf_clone_redirect(struct sk_buff *skb, u32 ifindex, u64 flags)`

`u64 bpf_get_current_pid_tgid(void)`

`u64 bpf_get_current_uid_gid(void)`

`long bpf_get_current_comm(void *buf, u32 size_of_buf)`

`u32 bpf_get_cgroup_classid(struct sk_buff *skb)`

`long bpf_skb_vlan_push(struct sk_buff *skb, __be16 vlan_proto, u16 vlan_tci)`

`long bpf_skb_vlan_pop(struct sk_buff *skb)`

`long bpf_skb_get_tunnel_key(struct sk_buff skb, struct bpf_tunnel_key key, u32 size, u64 flags)`

`long bpf_skb_set_tunnel_key(struct sk_buff skb, struct bpf_tunnel_key key, u32 size, u64 flags)`

`u64 bpf_perf_event_read(struct bpf_map *map, u64 flags)`

`long bpf_redirect(u32 ifindex, u64 flags)`

`u32 bpf_get_route_realm(struct sk_buff *skb)`

`long bpf_perf_event_output(void ctx, struct bpf_map map, u64 flags, void *data, u64 size)`

`long bpf_skb_load_bytes(const void skb, u32 offset, void to, u32 len)`

`long bpf_get_stackid(void ctx, struct bpf_map map, u64 flags)`

`s64 bpf_csum_diff(__be32 from, u32 from_size, __be32 to, u32 to_size, __wsum seed)`

`long bpf_skb_get_tunnel_opt(struct sk_buff skb, void opt, u32 size)`

`long bpf_skb_set_tunnel_opt(struct sk_buff skb, void opt, u32 size)`

`long bpf_skb_change_proto(struct sk_buff *skb, __be16 proto, u64 flags)`

`long bpf_skb_change_type(struct sk_buff *skb, u32 type)`

`long bpf_skb_under_cgroup(struct sk_buff skb, struct bpf_map map, u32 index)`

`u32 bpf_get_hash_recalc(struct sk_buff *skb)`

`u64 bpf_get_current_task(void)`

`long bpf_probe_write_user(void dst, const void src, u32 len)`

`long bpf_current_task_under_cgroup(struct bpf_map *map, u32 index)`

`long bpf_skb_change_tail(struct sk_buff *skb, u32 len, u64 flags)`

`long bpf_skb_pull_data(struct sk_buff *skb, u32 len)`

`s64 bpf_csum_update(struct sk_buff *skb, __wsum csum)`

`void bpf_set_hash_invalid(struct sk_buff *skb)`

`long bpf_get_numa_node_id(void)`

`long bpf_skb_change_head(struct sk_buff *skb, u32 len, u64 flags)`

`long bpf_xdp_adjust_head(struct xdp_buff *xdp_md, int delta)`

`long bpf_probe_read_str(void dst, u32 size, const void unsafe_ptr)`

`u64 bpf_get_socket_cookie(struct sk_buff *skb)`

`u64 bpf_get_socket_cookie(struct bpf_sock_addr *ctx)`

`u64 bpf_get_socket_cookie(struct bpf_sock_ops *ctx)`

`u64 bpf_get_socket_cookie(struct sock *sk)`

`u32 bpf_get_socket_uid(struct sk_buff *skb)`

`long bpf_set_hash(struct sk_buff *skb, u32 hash)`

`long bpf_setsockopt(void bpf_socket, int level, int optname, void optval, int optlen)`

`long bpf_skb_adjust_room(struct sk_buff *skb, s32 len_diff, u32 mode, u64 flags)`

`long bpf_perf_prog_read_value(struct bpf_perf_event_data ctx, struct bpf_perf_event_value buf, u32 buf_size)`

`long bpf_getsockopt(void bpf_socket, int level, int optname, void optval, int optlen)`

`long bpf_override_return(struct pt_regs *regs, u64 rc)`

`long bpf_sock_ops_cb_flags_set(struct bpf_sock_ops *bpf_sock, int argval)`

`long bpf_msg_redirect_map(struct sk_msg_buff msg, struct bpf_map map, u32 key, u64 flags)`

`long bpf_msg_apply_bytes(struct sk_msg_buff *msg, u32 bytes)`

`long bpf_msg_cork_bytes(struct sk_msg_buff *msg, u32 bytes)`

`long bpf_msg_pull_data(struct sk_msg_buff *msg, u32 start, u32 end, u64 flags)`

🎀 `bpf_bind` — the "everyone shares ONE ip, bestie" helper 🎀

🌸 `bpf_xdp_adjust_tail` — move that `data_end`, periodt 🌸

🦋 `bpf_skb_get_xfrm_state` — XFRM state, but make it accessible 🦋

💜 `bpf_get_stack` — stack traces, but from inside the program 💜

🌈 `bpf_skb_load_bytes_relative` — load bytes, but pick your header first 🌈

👑 `bpf_fib_lookup` — the big one, FIB lookup with main character energy 👑

`long bpf_sock_hash_update(struct bpf_sock_ops skops, struct bpf_map map, void *key, u64 flags)`

`long bpf_msg_redirect_hash(struct sk_msg_buff msg, struct bpf_map map, void *key, u64 flags)`

`long bpf_sk_redirect_hash(struct sk_buff skb, struct bpf_map map, void *key, u64 flags)`

`long bpf_lwt_push_encap(struct sk_buff skb, u32 type, void hdr, u32 len)`

`long bpf_lwt_seg6_store_bytes(struct sk_buff skb, u32 offset, const void from, u32 len)`

`long bpf_lwt_seg6_adjust_srh(struct sk_buff *skb, u32 offset, s32 delta)`

`long bpf_lwt_seg6_action(struct sk_buff skb, u32 action, void param, u32 param_len)`

`long bpf_rc_repeat(void *ctx)`

`long bpf_rc_keydown(void *ctx, u32 protocol, u64 scancode, u32 toggle)`

`u64 bpf_skb_cgroup_id(struct sk_buff *skb)`

`u64 bpf_get_current_cgroup_id(void)`

`void bpf_get_local_storage(void map, u64 flags)`

`long bpf_sk_select_reuseport(struct sk_reuseport_md reuse, struct bpf_map map, void *key, u64 flags)`