[PATCH bpf-next] bpf, x86: Use emit_rsb_call() to call bpf_unwind_target()

[PATCH bpf-next] bpf, x86: Use emit_rsb_call() to call bpf_unwind_target()

[Alexei Starovoitov]

From: Alexei Starovoitov <ast@kernel.org>

clang inlines emit_unwind_ret_addr() and emit_patch() into do_jit() and
folds 'bpf_unwind_target - X86_PATCH_SIZE' into one relocation. It points
to a nop in the padding in front of the function, and objtool complains:

  vmlinux.o: warning: objtool: do_jit+0x1330: relocation to !ENDBR: .text+0xc7fab

It's only used to compute rel32 of a direct call, so nothing is wrong at
run time.

emit_rsb_call() does call depth accounting too and hides func from
the optimizer. Use it instead of open coding.

Fixes: cff94a3d78d8 ("bpf, x86: Support unwind_goto and unwind_exit")
Reported-by: Nathan Chancellor <nathan@kernel.org>
Closes: https://lore.kernel.org/bpf/20261021201456.GA3197963@ax162/
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 arch/x86/net/bpf_jit_comp.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 6a2fe3d0039d..71f7200dba00 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -2564,9 +2564,7 @@ static int emit_unwind_ret_addr(u8 **pprog, u8 *ip)
 
 	EMIT2(0x50, 0x56);			/* push rax; push rsi */
 	EMIT4(0x48, 0x8B, 0x7D, 0x08);		/* mov rdi, [rbp + 8] */
-	ip += prog - start;
-	ip += x86_call_depth_emit_accounting(&prog, bpf_unwind_target, ip);
-	if (emit_call(&prog, bpf_unwind_target, ip))
+	if (emit_rsb_call(&prog, bpf_unwind_target, ip + (prog - start)))
 		return -EINVAL;
 	EMIT4(0x48, 0x89, 0x45, 0x08);		/* mov [rbp + 8], rax */
 	EMIT2(0x5E, 0x58);			/* pop rsi; pop rax */

base-commit: b116ac6087541d3d98031e786adbe9c733cd6df3
-- 
2.55.0

Re: [PATCH bpf-next] bpf, x86: Use emit_rsb_call() to call bpf_unwind_target()

[Nathan Chancellor]

On Wed, Oct 21, 2026 at 10:08:44PM +0000, Alexei Starovoitov wrote:
> From: Alexei Starovoitov <ast@kernel.org>
> 
> clang inlines emit_unwind_ret_addr() and emit_patch() into do_jit() and
> folds 'bpf_unwind_target - X86_PATCH_SIZE' into one relocation. It points
> to a nop in the padding in front of the function, and objtool complains:
> 
>   vmlinux.o: warning: objtool: do_jit+0x1330: relocation to !ENDBR: .text+0xc7fab
> 
> It's only used to compute rel32 of a direct call, so nothing is wrong at
> run time.
> 
> emit_rsb_call() does call depth accounting too and hides func from
> the optimizer. Use it instead of open coding.
> 
> Fixes: cff94a3d78d8 ("bpf, x86: Support unwind_goto and unwind_exit")
> Reported-by: Nathan Chancellor <nathan@kernel.org>
> Closes: https://lore.kernel.org/bpf/20261021201456.GA3197963@ax162/
> Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Tested-by: Nathan Chancellor <nathan@kernel.org> # build

> ---
>  arch/x86/net/bpf_jit_comp.c | 4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
> index 6a2fe3d0039d..71f7200dba00 100644
> --- a/arch/x86/net/bpf_jit_comp.c
> +++ b/arch/x86/net/bpf_jit_comp.c
> @@ -2564,9 +2564,7 @@ static int emit_unwind_ret_addr(u8 **pprog, u8 *ip)
>  
>  	EMIT2(0x50, 0x56);			/* push rax; push rsi */
>  	EMIT4(0x48, 0x8B, 0x7D, 0x08);		/* mov rdi, [rbp + 8] */
> -	ip += prog - start;
> -	ip += x86_call_depth_emit_accounting(&prog, bpf_unwind_target, ip);
> -	if (emit_call(&prog, bpf_unwind_target, ip))
> +	if (emit_rsb_call(&prog, bpf_unwind_target, ip + (prog - start)))
>  		return -EINVAL;
>  	EMIT4(0x48, 0x89, 0x45, 0x08);		/* mov [rbp + 8], rax */
>  	EMIT2(0x5E, 0x58);			/* pop rsi; pop rax */
> 
> base-commit: b116ac6087541d3d98031e786adbe9c733cd6df3
> -- 
> 2.55.0
> 

-- 
Cheers,
Nathan

Re: [PATCH bpf-next] bpf, x86: Use emit_rsb_call() to call bpf_unwind_target()

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to bpf/bpf.git (master)
by Alexei Starovoitov <ast@kernel.org>:

On Wed, 21 Oct 2026 22:08:44 +0000 you wrote:
> From: Alexei Starovoitov <ast@kernel.org>
> 
> clang inlines emit_unwind_ret_addr() and emit_patch() into do_jit() and
> folds 'bpf_unwind_target - X86_PATCH_SIZE' into one relocation. It points
> to a nop in the padding in front of the function, and objtool complains:
> 
>   vmlinux.o: warning: objtool: do_jit+0x1330: relocation to !ENDBR: .text+0xc7fab
> 
> [...]

Here is the summary with links:
  - [bpf-next] bpf, x86: Use emit_rsb_call() to call bpf_unwind_target()
    https://git.kernel.org/bpf/bpf/c/02da62c2f896

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html