1. Patchew
  2. linux
  3. [PATCH bpf-next v3 0/3] bpf, x86: inline bpf_get_current_task() for x86_64
  4. View patch

[PATCH bpf-next v3 1/3] bpf, x86: inline bpf_get_current_task() for x86_64

Menglong Dong posted 3 patches 1 month ago
There is a newer version of this series
[PATCH bpf-next v3 1/3] bpf, x86: inline bpf_get_current_task() for x86_64
Posted by Menglong Dong 1 month ago 
Inline bpf_get_current_task() and bpf_get_current_task_btf() for x86_64
to obtain better performance.

In !CONFIG_SMP case, the percpu variable is just a normal variable, and
we can read the current_task directly.

Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
---
v3:
- implement it in the verifier with BPF_MOV64_PERCPU_REG() instead of in
  x86_64 JIT.
---
 kernel/bpf/verifier.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 3d44c5d06623..520c413839ee 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -17688,6 +17688,8 @@ static bool verifier_inlines_helper_call(struct bpf_verifier_env *env, s32 imm)
 	switch (imm) {
 #ifdef CONFIG_X86_64
 	case BPF_FUNC_get_smp_processor_id:
+	case BPF_FUNC_get_current_task_btf:
+	case BPF_FUNC_get_current_task:
 		return env->prog->jit_requested && bpf_jit_supports_percpu_insn();
 #endif
 	default:
@@ -23273,6 +23275,24 @@ static int do_misc_fixups(struct bpf_verifier_env *env)
 			insn      = new_prog->insnsi + i + delta;
 			goto next_insn;
 		}
+
+		/* Implement bpf_get_current_task() and bpf_get_current_task_btf() inline. */
+		if ((insn->imm == BPF_FUNC_get_current_task || insn->imm == BPF_FUNC_get_current_task_btf) &&
+		    verifier_inlines_helper_call(env, insn->imm)) {
+			insn_buf[0] = BPF_MOV64_IMM(BPF_REG_0, (u32)(unsigned long)&current_task);
+			insn_buf[1] = BPF_MOV64_PERCPU_REG(BPF_REG_0, BPF_REG_0);
+			insn_buf[2] = BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0);
+			cnt = 3;
+
+			new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, cnt);
+			if (!new_prog)
+				return -ENOMEM;
+
+			delta    += cnt - 1;
+			env->prog = prog = new_prog;
+			insn      = new_prog->insnsi + i + delta;
+			goto next_insn;
+		}
 #endif
 		/* Implement bpf_get_func_arg inline. */
 		if (prog_type == BPF_PROG_TYPE_TRACING &&
-- 
2.52.0
Re: [PATCH bpf-next v3 1/3] bpf, x86: inline bpf_get_current_task() for x86_64
Posted by Menglong Dong 1 month ago 
On 2026/1/9 16:26, Menglong Dong wrote:
> Inline bpf_get_current_task() and bpf_get_current_task_btf() for x86_64
> to obtain better performance.
> 
> In !CONFIG_SMP case, the percpu variable is just a normal variable, and
> we can read the current_task directly.
> 
> Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
> ---
> v3:
> - implement it in the verifier with BPF_MOV64_PERCPU_REG() instead of in
>   x86_64 JIT.
> ---
>  kernel/bpf/verifier.c | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
> 
> diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> index 3d44c5d06623..520c413839ee 100644
> --- a/kernel/bpf/verifier.c
> +++ b/kernel/bpf/verifier.c
> @@ -17688,6 +17688,8 @@ static bool verifier_inlines_helper_call(struct bpf_verifier_env *env, s32 imm)
>  	switch (imm) {
>  #ifdef CONFIG_X86_64
>  	case BPF_FUNC_get_smp_processor_id:
> +	case BPF_FUNC_get_current_task_btf:
> +	case BPF_FUNC_get_current_task:
>  		return env->prog->jit_requested && bpf_jit_supports_percpu_insn();
>  #endif
>  	default:
> @@ -23273,6 +23275,24 @@ static int do_misc_fixups(struct bpf_verifier_env *env)
>  			insn      = new_prog->insnsi + i + delta;
>  			goto next_insn;
>  		}
> +
> +		/* Implement bpf_get_current_task() and bpf_get_current_task_btf() inline. */
> +		if ((insn->imm == BPF_FUNC_get_current_task || insn->imm == BPF_FUNC_get_current_task_btf) &&
> +		    verifier_inlines_helper_call(env, insn->imm)) {
> +			insn_buf[0] = BPF_MOV64_IMM(BPF_REG_0, (u32)(unsigned long)&current_task);

The !CONFIG_SMP case is still not handled properly here.
In the CONFIG_SMP case, I think &current_task is always
u32, so we can remove the casting. Therefore, this code
will be suitable for !CONFIG_SMP too.

> +			insn_buf[1] = BPF_MOV64_PERCPU_REG(BPF_REG_0, BPF_REG_0);

In !CONFIG_SMP case, &current_task is a normal variable
pointer, and the BPF_MOV64_PERCPU_REG() will be ignored,
which makes the whole logic right.

I'll do more testing on it.

> +			insn_buf[2] = BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0);
> +			cnt = 3;
> +
> +			new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, cnt);
> +			if (!new_prog)
> +				return -ENOMEM;
> +
> +			delta    += cnt - 1;
> +			env->prog = prog = new_prog;
> +			insn      = new_prog->insnsi + i + delta;
> +			goto next_insn;
> +		}
>  #endif
>  		/* Implement bpf_get_func_arg inline. */
>  		if (prog_type == BPF_PROG_TYPE_TRACING &&
>
Re: [PATCH bpf-next v3 1/3] bpf, x86: inline bpf_get_current_task() for x86_64
Posted by Menglong Dong 1 month ago 
On 2026/1/9 16:26 Menglong Dong <menglong8.dong@gmail.com> write:
> Inline bpf_get_current_task() and bpf_get_current_task_btf() for x86_64
> to obtain better performance.
> 
> In !CONFIG_SMP case, the percpu variable is just a normal variable, and
> we can read the current_task directly.

This sentence is redundant, and should be remove :/

> 
> Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
> ---
> v3:
> - implement it in the verifier with BPF_MOV64_PERCPU_REG() instead of in
>   x86_64 JIT.
> ---
>  kernel/bpf/verifier.c | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
> 
> diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> index 3d44c5d06623..520c413839ee 100644
> --- a/kernel/bpf/verifier.c
> +++ b/kernel/bpf/verifier.c
> @@ -17688,6 +17688,8 @@ static bool verifier_inlines_helper_call(struct bpf_verifier_env *env, s32 imm)
>  	switch (imm) {
>  #ifdef CONFIG_X86_64
>  	case BPF_FUNC_get_smp_processor_id:
> +	case BPF_FUNC_get_current_task_btf:
> +	case BPF_FUNC_get_current_task:
>  		return env->prog->jit_requested && bpf_jit_supports_percpu_insn();
>  #endif
>  	default:
> @@ -23273,6 +23275,24 @@ static int do_misc_fixups(struct bpf_verifier_env *env)
>  			insn      = new_prog->insnsi + i + delta;
>  			goto next_insn;
>  		}
> +
> +		/* Implement bpf_get_current_task() and bpf_get_current_task_btf() inline. */
> +		if ((insn->imm == BPF_FUNC_get_current_task || insn->imm == BPF_FUNC_get_current_task_btf) &&
> +		    verifier_inlines_helper_call(env, insn->imm)) {
> +			insn_buf[0] = BPF_MOV64_IMM(BPF_REG_0, (u32)(unsigned long)&current_task);
> +			insn_buf[1] = BPF_MOV64_PERCPU_REG(BPF_REG_0, BPF_REG_0);
> +			insn_buf[2] = BPF_LDX_MEM(BPF_DW, BPF_REG_0, BPF_REG_0, 0);
> +			cnt = 3;
> +
> +			new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, cnt);
> +			if (!new_prog)
> +				return -ENOMEM;
> +
> +			delta    += cnt - 1;
> +			env->prog = prog = new_prog;
> +			insn      = new_prog->insnsi + i + delta;
> +			goto next_insn;
> +		}
>  #endif
>  		/* Implement bpf_get_func_arg inline. */
>  		if (prog_type == BPF_PROG_TYPE_TRACING &&
> -- 
> 2.52.0
> 
> 
>

Patchew 2.1-devel-b67e4c2

© 2016 - 2026 Red Hat, Inc.