Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

global percpu data v3 #8543

Draft
wants to merge 5 commits into
base: bpf-next_base
Choose a base branch
from
Draft

global percpu data v3 #8543

wants to merge 5 commits into from

Conversation

Asphaltt
Copy link
Contributor

No description provided.

Kernel Patches Daemon and others added 5 commits February 20, 2025 18:53
adding ci files
74bd319
@Asphaltt
bpf: Introduce global percpu data
c790626 
This patch introduces global percpu data, inspired by commit
6316f78 ("Merge branch 'support-global-data'"). It enables the
definition of global percpu variables in BPF, similar to the
DEFINE_PER_CPU() macro in the kernel[0].

For example, in BPF, it is able to define a global percpu variable like:

int data SEC(".percpu");

With this patch, tools like retsnoop[1] and bpflbr[2] can simplify their
BPF code for handling LBRs. The code can be updated from

static struct perf_branch_entry lbrs[1][MAX_LBR_ENTRIES] SEC(".data.lbrs");

to

static struct perf_branch_entry lbrs[MAX_LBR_ENTRIES] SEC(".percpu.lbrs");

This eliminates the need to retrieve the CPU ID using the
bpf_get_smp_processor_id() helper.

Additionally, by reusing global percpu data map, sharing information
between tail callers and callees or freplace callers and callees becomes
simpler compared to reusing percpu_array maps.

Links:
[0] https://github.com/torvalds/linux/blob/fbfd64d25c7af3b8695201ebc85efe90be28c5a3/include/linux/percpu-defs.h#L114
[1] https://github.com/anakryiko/retsnoop
[2] https://github.com/Asphaltt/bpflbr

Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
@Asphaltt
bpf, libbpf: Support global percpu data
1c2ab90 
This patch introduces support for global percpu data in libbpf by adding a
new ".percpu" section, similar to ".data". It enables efficient handling of
percpu global variables in bpf programs.

Key changes:

* Introduces the ".percpu" section in libbpf.
* Correct value size to __aligned(8) of ".percpu" map definition and btf.
* Creates internal maps for percpu data.
* Initializes and populates these maps accordingly.

This enhancement improves performance for workloads that benefit from
percpu storage.

Meanwhile, add bpf_map__is_internal_percpu() API to check whether the map
is an internal map used for global percpu variables.

Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
@Asphaltt
bpf, bpftool: Generate skeleton for global percpu data
ab1c059 
This patch enhances bpftool to generate skeletons that properly handle
global percpu variables. The generated skeleton now includes a dedicated
structure for percpu data, allowing users to initialize and access percpu
variables more efficiently.

Changes:

1. skeleton structure:

For global percpu variables, the skeleton now includes a nested
structure, e.g.:

struct test_global_percpu_data {
	struct bpf_object_skeleton *skeleton;
	struct bpf_object *obj;
	struct {
		struct bpf_map *percpu;
	} maps;
	// ...
	struct test_global_percpu_data__percpu {
		int data;
		int run;
		int data2;
	} __aligned(8) *percpu;

	// ...
};

  * The "struct test_global_percpu_data__percpu" points to initialized
    data, which is actually "maps.percpu->data".
  * Before loading the skeleton, updating the
    "struct test_global_percpu_data__percpu" modifies the initial value
    of the corresponding global percpu variables.
  * After loading the skeleton, accessing or updating this struct is not
    allowed because this struct pointer has been reset as NULL. Instead,
    users must interact with the global percpu variables via the
    "maps.percpu" map.

2. code changes:

  * Added support for ".percpu" sections in bpftool's map identification
    logic.
  * Modified skeleton generation to handle percpu data maps
    appropriately.
  * Updated libbpf to make "percpu" pointing to "maps.percpu->data".
  * Set ".percpu" struct points to NULL after loading the skeleton.

Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
@Asphaltt
selftests/bpf: Add cases to test global percpu data
1f2b8a8 
If the arch, like s390x, does not support percpu insn, this case won't
test global percpu data by checking -EOPNOTSUPP after loading prog.

The following APIs have been tested for global percpu data:
1. bpf_map__set_initial_value()
2. bpf_map__initial_value()
3. generated percpu struct pointer pointing to internal map's mmaped
4. bpf_map__lookup_elem() for global percpu data map
5. bpf_map__is_internal_percpu()

At the same time, the case is also tested with 'bpftool gen skeleton -L'.

125     global_percpu_data_init:OK
126     global_percpu_data_lskel:OK
Summary: 2/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
@kernel-patches-daemon-bpf kernel-patches-daemon-bpf bot force-pushed the bpf-next_base branch 12 times, most recently from df10b88 to 16566af Compare February 26, 2025 21:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@Asphaltt